Lawful Good or Neutral Evil? Open Palm or Closed Fist? Such questions were inconceivable in gaming’s early days. Rogue didn’t care if you styled yourself a noble hero, and the classic Gold Box games wouldn’t let you be much else. The first major North American computer roleplaying game to seriously consider any form of player alignment, Ultima IV, demanded the player conform to the game’s eight virtues. When your choices are limited to honesty, compassion, humility and the like, it’s hard to be, well, evil.

But just how far have we come in the past two decades?

Ultima‘s Virtues, Fallout‘s Karma and Reputation, Fable‘s point system, Jade Empire‘s Paths and other player alignment systems owe an enormous debt to the original alignment system – the one introduced by Gary Gygax and Dave Arneson in Dungeons & Dragons. Originally just a single axis with three points (Law, Neutrality, Chaos), Advanced Dungeons & Dragons expanded it into the two-axis, nine-point Good-Evil/Law-Chaos framework used most famously in the digital realm by BioWare and Black Isle.

On the surface, the “Gygax model” seems to offer more flexibility for player alignment systems than its most common CRPG alternative – single-axis Good/Evil models. Both typically associate Good with altruism and Evil with self-interest, but the Gygax model offers the additional axis of Law and Chaos, corresponding roughly to authoritarianism and libertarianism. This allows games implementing the Gygax model to assign alignment labels to actions that wouldn’t be clear-cut in a single-axis Good/Evil model.

For example, is rescuing your convicted friend from prison a Good act if she is actually guilty? This is awkward to address in a single-axis framework, but in a game implementing the Gygax model it would shift the player’s alignment in the Chaotic direction. Regardless of motive, it is a clear defiance of law. What about a player who advocates changing the law, then? Civil disobedience is certainly Chaotic, but legitimate legal protest doesn’t exactly conform to the Gygax model’s ideal of Lawful, which entails conforming to the will of a sovereign.

The Gygax model’s weaknesses stem from its origin in D&D, which began life as a glorified tabletop miniature wargame. Players in D&D needed to be able to choose from balanced classes with defined roles, and the player group needed a common goal. The alignment system restricted the actions of players choosing to play certain character classes, which compensated for the special abilities of those classes. The prominence of altruism and the overall emphasis on heroism at the Good end of the alignment spectrum helped foster player cooperation. To make it work, Good and Evil were precisely defined, tossing aside more than two millennia of philosophy, which made it difficult for any game implementing the model to ask meaningful questions about the nature of Good or Evil.

Typically, the Gygax model is codified in a CRPG by assigning alignment weights to various actions and dialogue choices. A player wishing to be Good should choose the most polite dialogue options and make sure to commit acts of kindness on a regular basis; a player wishing to be Evil can feel free to fling all manner of insults at non-player characters and then rob them blind. Neutrality is awkward – rather than remaining Neutral by striving to chart a course along the middle ground between Good and Evil and/or Law and Chaos, players remain Neutral by perpetrating an equal weighting of Good and Evil acts. There’s no way to differentiate a shift toward Neutrality from a shift toward Good, Evil, Law or Chaos in the Gygax model.

This way, though, the player’s conscience is salved preemptively. When the inevitable slaughter of hordes of enemies begins, a Good player has already rationalized killing them as the inevitable consequence of their Evil nature – and their experience point value. If the player has the audacity to choose Evil, well, what conscience is there to salve? Indiscriminate slaughter is Evil’s watchword.

Alignment weights can also be abused more directly. If you pickpocket a farmer but rescue the blacksmith’s daughter from her kidnappers, the net result is a shift in your alignment toward Good. Trying this approach in real life will make you pretty unpopular and maybe even land you in jail, but somehow it works just fine in BioWare’s Neverwinter Nights. In fact, a Paladin in NWN can go to an inn, loot someone else’s wardrobe and remain a Paladin; in tabletop D&D, the Paladin would immediately lose all unique Paladin abilities and have to undertake an atonement quest to regain them.

The outcome of these problems is twofold: Most CRPGs that implement an alignment system only assign weights to a small subset of actions – hence NWN‘s looting Paladins – and also severely restrict player choice in terms of both dialogue and course of action, in order to focus the game’s narrative. In other words, the principal effects of the Gygax model on North American CRPGs have been less open, more linear gameplay, clich

Obsidian Entertainment, formed by a group of Black Isle alumni after Interplay closed Black Isle, crafted the sequel to KotOR, adding an Influence value to the alignment system, to track the extent to which NPCs were open to persuasion by the player. This is a natural extension of Fallout‘s Reputation – a single number that determines whether word of the player’s altruistic or self-interested actions (represented by the Karma value) has reached a given NPC. It contrasts starkly with PST, where a sufficiently charismatic player can persuade an NPC to virtually any course of action. Such NPC-specific player attributes, combined with branchy dialogue trees, can create richer player-NPC interactions and allow a developer to pose more interesting ethical dilemmas to the player.

Still, the strength of Gygax-model CRPGs like PST suggest it may not be the inherent restrictiveness of the Gygax model holding back North American CRPGs, but rather CRPG developers using it as a crutch to simplify characterization and storytelling. Because Good and Evil are defined so clearly by the Gygax model, it’s extremely easy to create cookie-cutter characters and railroad players along a couple of paths toward the all-too-common “Good ending” and “Evil ending” – that is, if an “Evil ending” is even included.

Settings, graphics and physics are becoming increasingly complex in CRPGs, and player alignment models need to grow along with them. The kinds of deficiencies revealed by NWN‘s looting Paladins have to be resolved, but there is much further to go. Obsidian has explored Evil player alignments more deeply than most, and they’ve made an effort to chart some of the ethical murk between the Gygax model’s well-defined Good and Evil, but even they have yet to stray too far. We need an ambitious developer to take that first complete step outside the long shadow of the Gygax model, abandon Good and Evil, and treat in-game ethics with the detail and nuance they deserve.

Raja Doake is a would-be writer in engineer’s clothing who lives in Ontario, Canada, and spends way too much time thinking about this stuff.

You may also like