|
[spoiler alert - There is a spoiler for Enslaved in the indicated paragraph]
Do you want to be the good guy or the bad guy? It is a question quite a few games ask nowadays. Indecision does not pay off. Anybody who responds to an average game with any level of ethical nuance will end up being the Jack of All Trades, Master of None of Morality. This is not only a purposeful design element where “karma” is tracked based on how many “points” certain actions get, it is also implicit in almost all games: You fight the “bad guys”. Or, if you have a really original title on your hands, you might fight as a bad guy against the good guys!
It is binary ethics, and I will hardly claim it is unrealistic. Religious zealots and fervent nationalists base all their actions on the thought of fighting for good against the forces of evil. In turn, this is reflected in anything from fairy tales to the Ten Commandments. It is the idea that some acts are intrinsically good or bad – a concept that is quite easy to quantify and characterize in video games. Now the question is not only if this is good design. The question is also, is this a morality we want to support?
My suggestion is to base video game morality more on the reality of human nature, and less on a bland black-white dichotomy where no one wants to be caught dead wearing any shade of gray. Gray is the interesting color! It has more depth, reality and interesting choices. On top of that, morally black and white human beings do not exist. They are an abstract concept. We naturally use it to simplify our encounters with people that strongly disagree with us, and lose so much in the process: empathy and understanding. A truly relatable game will be made up of characters we can sympathize with. Enough designers put in effort to create a protagonist that is flawed but likeable. But what about the other side of the conflict? Seldomly have I seen games where the antagonist is a real human being. The tragedy of conflict, war and death is best shown by how futile it always is. There are only two features in human nature that cause us to fight with one another and create misery: Misunderstanding and Malfunction.
Misunderstanding is when people refuse to see, or are unaware of, other perspectives on the problem. They are so dominated by their feelings, or by their fellows, that they do not stop and think what they are doing to others. They do not realize the suffering they bring.
Malfunction is when people are mentally damaged. Psychopaths are incapable of empathy, and so are incapable of caring what damage they might do. Other mental disorders might make people blind and deaf to the suffering they cause.
[spoiler]
These are the “villains” you want to portray in a game. Think of the ending of Enslaved where the “bad guy” was only trying to do good in a very strange way. He did not understand very well what people needed, and so his intentions went wrong. Think of Heavy Rain, where the protagonist is trying to save his son, but has to do gruesome things to reach his goal. He is trying to do the right thing, but doing the right thing can be a very complex matter!
[/spoiler]
Design-wise it would be much more interesting if games would take an increasingly realistic approach to morality. Instead of giving us horns, halos and karma, give us interesting choices! Morality is defined by the joy and suffering we bring, so let that be the measure we show in games. Some games express this by letting companion characters or factions respond to your actions. It is a step in the right direction. Emotional resonance would be another important step. For instance, Trip taking control of Monkey in Enslaved is a “bad thing” in binary ethics. However, as soon as the player understands their predicament, it can hardly be seen as bad any longer. The realism of such an action creates rapport with the player. If such actions were cast into a meaningful player choice, just think of the response! Yet currently, such morally grey actions are often not rewarded in games. You cannot play as a true human being because only extremes are rewarded.
Yet the question of morality in games is not only one of design. I am one of the last people to say games are bad for us. For instance, I do not think violent games cause violent behavior in any significant way. Part of the reason is that we all understand that video game violence is not real. We are playing pretend and we all understand that. However, with morality, that line seems less clear. Too many people accept that morality is a matter of good against evil. It is only natural to them that it works like that in fiction too. I do not think video games make the issue worse. I do think that video games can make it better. Actually, books, movies, and educational institutions can do just as much or much more here. I realize this. At the same time, books and movies have a much greater repertoire of thought-provoking material than games do. On top of that, games can let people actively take part in the moral choices. It also partly engages a different audience that might not be reached by the movies and books in question.
Currently, technology and profit limit our options. To truly implement realistic ethics you would need the type of dynamic AI that does not exist yet. Scripting moral choices might allow a binary choice to get 3 or 4 branches instead, but it is essentially the same problem. Realistic morality is a subproblem of realistic human interaction in games. I realize there are many technical hurdles here. For instance, even if the game characters come to form interesting and dynamic responses to the player, how would they formulate them? Natural language is hardly their strength right now. At the same time, there are a lot steps we can take to improve morality in games based on the technology that we do have right now. It can be both part of the design, and the story.
This article is meant as thought-provoking. What do you think right and wrong really are? Does it matter how we model it in games? Are black and white more fun than grey? Can we develop games that have a profound impact on how people view the world? If so, how would we do that? Or have we already done that with some games?
I realize I might have skipped some points. For instance, mythical creatures are quite prevalent in games, and it is only “natural” that some of them are pure evil or pure good. Also, some games might already tick some of the boxes I mention here. If you know any, I would be grateful to hear about it, as well as any other thoughts you might have on the topic. In the worst case, we can always fight about whose morality is better 
[Reposted from Think Feel Play]
|
Perhaps the only way to handle a morality system is to ignore the question of "morality" outright? After all, there's no way for a game to objectively define "good" and "evil" without revealing the personal beliefs of the designers, and, as you mentioned, allowing the player to choose between well-defined norms (such as being Mother Teresa or Hitler) is boring. Maybe it's better to simply present interesting choices and let players decide for themselves whether they are behaving in a way consistent with good or evil.
There are ways to hide these limitations though. I think Bioware has recently become excellent in this regard, making the traditional binary moral choices feel larger and more influential to the story then they actually are. Also, they do a good job in creating choices that are eons beyond the simplistic tripe that was the little sister "moral conundrum" of Bioshock. Both of Bioware's options usually have an up and downside, and are much more often about political or philosophical worldview than they are about mustache twirling or angelic halos.
As far as a cheap way to provide feedback for player choice morality, I think that designers could be doing a lot more with NPC characters. As humans, one of our most basic needs is the approval of others. Existing player choice could be greatly enhanced without too much extra design cost, by allowing the loyalties of NPCs to vary based on your previous moral choices. Imagine an NPC who became so fed up with your ignominious choices that she deserted the party or even joined with the antagonists! As a counterpoint to this, another NPC might be so impressed by your level head under pressure that she chooses to put herself between you and a bullet. These types of scenes have a huge potential to be emotionally griping because they are a direct response to an emotionally charged player action . . . and you still only need to design two of them for each moral choice.
Another shortcut to moral grayitude might be to weave it into the existing design constraints. Just because your release date only allows you the time to design for two player choices, doesn't mean that those choices have to fall at opposite moral poles. Heroic characters in literature often have to chose between two equally bad options and are made all the more interesting because of it.
As Shoshannah says, one of the reasons we run into problems with good/evil in games is because of simplistic ideas about what those things are. The "evil" options in games are often quite boring and unimaginative, and they pretty much always boil down to either killing or intimidating through violence. But when we think about evil in the real world, the people and actions we think of aren't usually people who run around killing anyone they don't like, because that kind of person is extremely rare. For example, over the past couple of years there has been a lot of rage directed at investment bankers and their actions (as an aside, I think this is one of the reasons that Shinra in Final Fantasy 7 is such an interesting antagonist - it's primary "evil" attribute is not physical violence, but over-extraction of resources).
I'll give an example here of how I think we could handle this better. Let's say there's a character who has a key that you need in order to enter a fortress, and that character never lets the key out of their sight. Normally the "evil" option to get the key will be to kill them and take it off their body. What if, instead, the evil option was to trick the character into thinking that it was in their best interests to follow your party, then gaining that character's trust over time such that they paid less attention to the safety of the key, then one night once their guard is down you take the key while they're sleeping and then sneak away, abandoning the character to explain to their leader how they lost the key to the fortress? That's a fairly "evil" way of treating someone, but it's far more interesting both narratively *and* mechanically.
I'm going to add in something that I wrote for an old blog entry of my own as well, because I think it's pretty relevant to this topic:
. . . what if instead of good/bad/neutral morality, we let players choose left/centre/right politics . . . there are plenty of other ways dialogues could be set up: upper/middle/lower class, atheist/agnostic/devout religious views, humanistic/scientific/religious world views, etc. There are many ways to provide players interesting options.
Ultimately, whether a moral system is black/white or gray is not the real issue. Aasimov's rules for robots seem absolute, but the way in which these rules can be subverted and the dilemmas they give rise to that are more fascinating that a thousand bland "there is no right or wrong only meaningless choices" stories. Execution is more important than orientation.
A better dilemma using the same building blocks would look something like this - you're in the castle of a king who has information that you desperately need in order to continue your quest. As you're walking through the castle, you notice that the prince has had some soldiers kidnap a local woman and that he is about to rape her. Your first instinct is to step in and stop the prince, but you *really* need the information that the king has and you're not sure you can risk losing his trust by attacking his son. So do you step in, or not step in? What if the information that the king has could save 1000 lives? Now we're at a point where we're dealing with grey areas, but the choice is meaningful both narratively and mechanically.
Taking your example: Your on a heroic quest to slay some demon (the usual). You are known far and wide for all the good deeds you have done.
You enter the kindom you describe, and stop the soliders. Guess what? You may be the embodiment of good, but within that one kingdom, you are now seen as the evil one (you disobyed the king, after all).
Instead of a global good/bad meter, I think we should be moving more toward how specific entities (be it people, kingdoms, ect) view the player and his/her actions.
From a role-playing perspective, I also found it interesting that with the correct build, choices, and some creativity, players can make it through the game without killing any human characters.
Too bad there was a right answer and a wrong answer. For fear of falling behind with my moral alignment, I chose to be good and to keep the research data, thereby sentencing every other race that is not Krogan to enslavement/ a horrible bloody death when they rally up their numbers and attack again.
I wasn't as satisfied with the resolution, but I loved the dilemma and think this type of choice is the direction moral alignment-based games should go.
*If you didn't play Mass Effect 2, the "Krogans" are a race of aliens that are big, brutish, and hard to kill. They also breed fast and recently tried to take over the universe. "Genophage" is a biological weapon used against them that effectively makes reproduction very unlikely. It was designed to keep their species in check, but is actually driving them to extinction.
On Ultima Online: Siege Perilous, way back in the old school Player Killer Wars, your reputation was determined by the number of 'innocents' you had killed. 5 kills, and you were designated as a murderer, and were subject to being instakilled if you tried to go into town, among other limitations. Groups formed consisting of player killers (PKs) and anti player killers (Anti-PKs), and they were ever at war with each other.
The picture was complicated by the fact that frequently, players unwilling to become murderers themselves would ally with the PKs, presenting the Anti-PKs with a dilema. How do you kill a PK ally without becoming a murderer yourself? If you didn't attack the 'innocent' PK allies, you presented the PKs with a significant tactical advantage.
In the end, the planned system of UO morality became irrelevant. It was a players actions that represented the true morality system, at least on UO:Siege Perilous. If you truly defended the innocent, you were one of the Good Guys, regardless of your actual game status. If you were one who preyed on the innocent, either directly or by supporting the actual murderers, you were one of the Bad Guys, and players on Siege Perilous knew the difference based on Guild Tags. My guild, a well known group of Anti-PKs were cheered and supported where ever we went, even though we had many members who, under the official game system, were 'murderers'.
You need a fairly open game environment for this to work, however. It isn't feasible in a faction based game like WoW, since you can't attack a member of your own faction, no matter how richly they may deserve it.
If you want offer gamers more interesting moral choices, begin by providing them freedom of choice. By their choices shall you know them.
So this time I picked a few general points I would like to respond to.
1. Everyone gave great examples of games where morality is treated differently. Fallout: New Vegas I find especially interesting. I'm currently in my first playthrough and this game was half the inspiration for this article. It has the faction-based "realistic" and rewarding morality, while at the same time it gives you bad karma for stealing while no one saw you do it!
2. Given the current technical limitations, I think Adam's input is very valuable: If you can only have two options, then make both of them different shades of gray. Jake's example of the Krogan Genophage in Mass Effecy illustrates this very well, eventhough the designers will still award paragon/renegade points in a binary fashion.
3. I know the AI Technology to make these types of interesting choices is very very far away. At the same time, I'm currently doing a Master's degree in Technical Artificial Intelligence. I'm looking for inspiration for tie-ins of academic research into game AI that is not path-finding or other procedural matters. I hope to get a topic approved (next year somewhere) that relates to making game AI "more human/interesting". My point is, we shouldn't shy away from looking at how future technologies could help us just because they are not here right now. Pushing those limits is what inspires the research to develop that future technology :)
I will keep doing my best to respond to individual comments in the future. It's the least I can do for all the valuable insights I gain here at Gamasutra. Thanks, guys!
I suggested that for an MMO perhaps it's best to not define morality for the player. Let the players define their own system as an emergent behavior. However, as UO also demonstrated, if you don't allow the players an effective way to enforce player justice, or whatever standard of morality they want to define, it's very easy for the game to degenerate into a universal Grand Theft Auto mode of player interactions. Actions should have consequences. When they don't, chaos rules.
There is a fine line to be walked there.