Oh dear, game morality. I wrote a lengthy piece on that topic a while back, mainly because I got vastly annoyed at how games handle morality, partly because I wanted to see whether there was any interest on a game 'developer' (meaning, lots of enthusiasts, a few actual dev's) forum for fleshing out morality. It turned into a nice debate, and one thing is, people might not actually want to be bothered by moral dilemma's in their games. Anyway, I don't want to cramp the forums style here so, if anyone wants to read some stuff on morality, click the spoiler. I just fished it out of the .doc file that I wrote it in, perhaps a year ago or so, but it might still be interesting.
(oh you can ignore the alternative I give; at the time I kinda felt obliged to try and deliver something 'better' or otherwise I'd only be complaining. I'm quite happy to only complain in fact.)
[spoiler:7808e9cbe9]On the Genealogy of Game Morality – A call for more than Good and Evil
(again, this topic is far too long, I apologize in advance)
Some games have far too long stuck to a system of dualism between good and evil.
1.1. Locating the phenomenon
Games that bother with morality usually fulfil this function with a dualistic system of good and evil. On the one side stands a bastion of selflessness, helpful, kind, and generally the poor bastard who gets screwed over when it comes to cool looking skills in the end. And on the other side stands the pure baby-eating evil, egoistic, lying, deceitful, and backstabbing shady character clad in black and red. For some games, this is entirely enough. For other games, the implications of this system inherently create limitations that have far too long restricted their social aspects to an unnecessary shallow level.
1.2. Good or evil is all you get, but that's fine
Bioshock is a good example of a game that doesn't really need more refined morality. Yes, the limitation to only harvest or save little sisters can be deplored, but there isn't much that can be done with this anyway in my opinion. It is a linear game and a shooter, these generally suffice with a clear-cut distinction between good and evil. Other examples include the recent Infamous, or games that don't give you any choice as to what you do, and whatever other game which doesn't have much text or doesn't require you to interact with other NPC's. Star Wars games, because of their lore, also require duality.
1.3. I judge thee and thy actions, none shall change my verdict
Baldur's Gate type games, Neverwinter Nights, Fallout series, Mass Effect: these kinds of games would greatly benefit from a more refined morality system. Although they all have their own touches to it, and they give morality a different name each, all of them go with the basic dichotomy between good and evil. Baldur's Gate and NWN call it reputation, Fallout calls it karma, and Mass Effect has Paragon and Renegade. The most important common aspect between these games is that they make morality known to the player; they judge him by a statistic.
1.4. Non-dualistic ways of solving morality in games
The dualistic version is certainly not the only one available in game morality. A very often used way to judge the player is through character interaction, so that the game-world judges the character in a more refined way and dependant on the specific person you interact with. Deus Ex judged the player for his ruthlessness (in the first part at least), Gothic had you form alliances and thus create enemies; The Witcher is another good example of a game that doesn't rely on black and white choices and judgements. Some of the games under 1.3, such as Mass Effect, also use this method in combination with a dualistic system to create a more fleshed out world. The problem of this non-dualistic approach lies in its complexity and thus limitation to a relatively small world and a few main characters. Also, this approach usually isn't very flexible and is limited to only a few choices or 2-3 ways of playing a game. Its biggest downfall is simply that it isn't a system nor has been systematically used.
2. Implications and inherent limitations of dualistic morality
Being the nature of dualism, all you'll end up with will be good and evil actions. Neutrality, while often mentioned, basically has no importance in a system governed by dualism. The deeper implications of dualism are much more interesting than neutrality anyway, so let us move on to that.
By communicating the moral value of the player's actions to the player, the game judges the actions of the player as if there was an omniscient God-like authority, deciding whether your actions were good, evil, or not significant enough to register on the moral-o-meter. Morality thus is no longer in the players' hands. He cannot decide that an action was good, evil, or a necessary evil or anything else. He'll be deemed evil or good regardless of his own judgement, and because of the general usage of the morality statistic for other mechanics, the entire world will make it noticeable to the player, for example through bartering, conversations, questlines, and often skills. By giving this morality statistic such high importance, it is pretty much required for such a game to communicate the God-like judgement to the player. And so as not to confuse the player as to whether he is doing right or wrong with his potential doings, most actions of dual morality have little depth to them, and one cannot surprise a player by having an unintended evil be the outcome of an at first good action. The route to good or evil is crystal clear. As a consequence, perhaps unintended, the way people play such a game often becomes governed by this statistic, and they play pure good or pure evil.
True, you can ignore the statistic and play through such a game in any way. But the constant reminder by the entire world, the often glaring red or blue numbers of your moral-o-meter, and the remarks of NPC's in conversations will constantly remember you what you are according to the game. Story, quests, other NPC's and companions; all are morally impoverished since duality requires them to visibly go for either side. At most, they can switch sides. What you end up with is a simplified and impoverished moral and social world, dependent on a morality system clearly visible to the player.
3. The Dilemma of Alternatives and unpleasant surprises
The inherent problem behind systemizing morality is the fact that the importance attributed to such a system by the rest of the game pretty much makes it necessary to communicate judgement to the player so that he knows what effects his future choices will have.
If you want to have an actual system of morality, not just isolated events of interaction as the games under 1.4 do, you are basically required to communicate the judgement statistics to the player. And once again the entire circle starts. You give names to certain ways of playing, you communicate this to the player, and then the player aims to achieve one of these ways of playing fully, yet again impoverishing many elements of a game. A system of morality basically shoots itself in the foot by disabling challenging moral situations, because ultimately, morally grey situations are decided by people and not by a mechanic. Neither can you really surprise a player with an outcome of his actions. He'd be frustrated by suddenly being deemed treacherous, and he'd reload and get the intended outcome. This is also why games usually never have consequences that happen far in the future. The invested time makes it unattractive to load a possibly not even existing save-game from before the event in question. Consequences have to be limited, direct and clear, and there can be usually only two or at most three ways of solving a problem.
4.1. An alternative: People and Types of Morality
Therefore, instead of focusing on morality, I would like to vouch for a better simulation of people themselves and their ways of judging other people. This is already being used partly in many of the games under 1.3., but has never been developed into an actual mechanic governing a morality system. Ideally, judgement of any kind would not have to be communicated to the player through statistics but only through the world itself, while underlying mechanics and statistics can hum away in the background doing their job. But I also believe that, even in communicating any statistics, the following system would still be a one-uppance on the traditional dualism.
What I would like to see realised would be a system wherein each NPC in the world is attributed degrees of a few defining characteristics that decide their outlook on the world and the actions of other people, mainly you. The old D&D system already did something in the manner through lawful and chaotic combined with dualistic morality, but this is simply too...simple.
In contrast, imagine a character who is defined as Law-abiding (20/100), Pious (80/100), Greedy (80/100), Helpful (40/100), Utilitarian (70/100), Trustworthy (30/100) and you'd have the background statistics for a crooked money-grubbing bishop who values his faith as much as his coin, who is somewhat reluctant to help and who believes that the ends justify the means in most cases. As long as you donate to him, or the church, or wherever donations end up with this character, it wouldn't really matter much to him what crimes you commit. As long as you pay him well, you might find him to be trustworthy, but this relationship depends entirely on coin. Obviously, you should be weary of blackmail.
These statistics all revolve around dichotomies themselves, but the combination of all of them make up for a far more precise approximation than traditional systems I believe. The specific statistics of main importance still have to be worked out closer and the following list is of a highly preliminary character.
Law-abiding - Law-breaking
Honest - Deceitful
Generous/selfless - Greedy/Egoistic
Pious - Impious (dependent on specific faiths of course)
Helpful - Unhelpful
Trustworthy - Treacherous
Deontologist (actions themselves have to be 'good') - Consequential/Utilitarian (the ends justify the means)
Grateful - Unappreciative
As said, this is preliminary and more a suggestion than a set system. The biggest problem of these characteristics is that they don't exist on their own and are in fact interconnected. For example, to determine whether a character would betray you, multiple of these statistics would have to be used. If this character had previously been rescued by you and he was grateful, this would be a strong deterrent. On the other hand, if he doesn't care about such things, perhaps money can satisfy his loyalty. Percentages should be weighted off against one another to decide whether the final outcome ends in betrayal.
Now, for the player character, the standard way of starting out would be with all statistics at 50/100. Another way of determining starting statistics could be through a common set of choices regarding hypothetical situations, a system which in Morrowind was used to decide your character class. Any choice the player makes would change these statistics incrementally. The problem which remains now is how the world perceives the player character. As with your own view of people in the world, it would be unrealistic if everyone else had a satellite connection to your centre of morality, making you an obvious give-away.
[/spoiler:7808e9cbe9]
