Feb 6, 2012

Abolishing Alignment

I can honestly tell you that I have no idea what the alignment of any PC in my Emern game is, nor do I have much interest in what they are. Alignment is pretty much the worst "pseudo-mechanic" in D&D, and one that can be gotten rid of without much work. I recommend doing so.

Alignment is bad because it's either useless or overly restricting. If it's a simple description of behaviour without prescriptive ideals, then it's useless. I had this problem in an Iron Heroes game I played in (that had alignment in it), where my polite, pacifist, vegetarian doctor was "lawful evil" simply because he didn't fundamentally care about other people except as means to his own ends. He considered violence to be a failure of will and intellect, and he ate a vegetarian diet for health reasons. I considered the character selfish to the point of sociopathy, but his actual actions weren't particularly bad, and his intentions were more arrogant than intentionally cruel. If anything, the character saw himself as the one good man in a world run by killers and monsters, and it was incumbent on him to seize control from them for the good of everyone. I don't really consider lawful evil to be a useful description of that kind of behaviour or worldview, and it got me thinking about how inadequate the alignment system is.

If it's not descriptive, then it's prescriptive, which is even worse in actual play, since the alignments are incoherent mishmashes, and you're artificially pushed into situations where two conflicting principles each dictate different, irreconcilable behaviours. This is considered "dramatic conflict", but it's really closer to metaphysical horror in the vein of Kafka. Having specialised in moral philosophy in university, this kind of false moral dichotomy makes me see red, as it totally differs from anything resembling real life moral decisions.

One option that's been attempted recently is to rationalise and dehumanise alignment, so it becomes a cosmic jersey indicating which team, angels or elder gods, one supports. This usually coincides with simplifications from the nine-point alignment chart into a binary split with "wishy-washy" as the middle choice. This can lead to more rational outcomes, but it doesn't really resolve the underlying problems. Alignment stops being a useful description of character behaviour, and it also leads to either a manichean worldview or the question why one doesn't simply use some sort of allegiance system?

The purpose of moral systems is to provide people with guidance in navigating the world, and the purpose of alignment is theoretically to categorise those moral systems in useful ways. Over the years I've experimented with creating different scales or categories split along more useful and relevant moral features (like how intense one's akrasia is, etc.), but I've abandoned each of these in turn. Ultimately, they either measure things that serve no purpose to measure, or they become so complex that they almost block out and predetermine roleplaying. I think the best way to capture morality in games is not to lean back on mechanics for them, but to push them completely into the domain of roleplaying. This captures the contingency and specificity that operate in true moral reasoning far more accurately than "lawful good" ever will.

Having made this decision, I was surprised at how easy it was to change the spells and abilities that interact with the alignment system. There tends to be only a few of these in any given version of D&D. One method I use is to strip Negative and Positive Energy of their moral qualities and simply change all spells that "detect evil" or "harm only evil creatures" to affect only creatures made of or charged with Negative Energy, and for "detect good" etc. to affect only Positive Energy. This still has some of the cosmic jersey approach, and one can stop there if one wishes, but I've been moving toward collapsing this into simply "divine energy". Undead, angels, demons, clerics, are all charged with divine energy, perhaps with slightly different flavours to distinguish the source of each, but it's all the same energy ultimately. Paladins who detect evil detect the presence or absence of this energy, and maybe whether it's their associated god's flavour or not, but that's it. This also keeps "detect evil" from being as constant an annoyance as it can be otherwise.

And of course, most people aren't charged with the stuff at all, no matter how good or evil they are. Spells that would previously affect only some alignment now either affect only creatures infused with divine energy or not infused with divine energy as appropriate. This makes spells like Holy Word a bit more like AoEs that you have to be careful about deploying, instead of IFF-equipped hunter-killer god missiles.

In Emern, this works because there is a single god of magic, the Hollow, and there are some suggestions in game that all gods are essentially instantiations or aspects of the Hollow, despite having their own clerical powers, churches, etc. I recommend trying it out in your game. If you still want to deal with moral issues and considerations (and I think people should), then leave it for roleplaying.


  1. Completely agreed. For my purposes (which usually run toward a Gothic take on fantasy) "Detect Evil" is a spell that can ruin a good, old-fashioned mystery. I want no part of that.

    1. Yeah, it's yet another problem with Detect Evil working on people who cheat on their taxes.

    2. Solution: original sin. 'Detect Evil' defaults to 'Detect Human.'

      I like the idea of using it as a 'Detect Covenant' spell: Muslims would be able to determine whether or not they're Muslim, Dhimmi or infidels. Christians would we able to tell whether their own political affiliation within the church (based on who tonsured or ordained them) counts (Detect Excommunicate). Punks would cast Detect Sellout. Jungians would cast Detect Projection of the Shadow.

  2. A well written piece. I stopped using alignment by gradually forgetting I was supposed to use it in my early teens. I do however feel Im not getting some D&D subtlety when someone whose D&D insights are sound still uses it. Evil and Goodness are driving forces of the old Romances but I find it hard to imagine *any* D&D adventurers are roleplayed like those characters and the extreme plot coaxing improbabilities and coincidences are anathema to the modern hip D&Der. In short I dont get alignment (for adult gamers).

  3. I do not like alignment as a psuedo-mechanic, but I do like it as a roleplaying tool. In fact, the second paragraph of this post illustrates very well why I like it; there is nothing either in your descriptions of Wayland here or back then that is unreasonable to me for a "Lawful Evil" character. I am not suggesting that the characterization was in any way dependant on alignment, but I find the system (pre-4th edition, anyway) to be both flexible and inspiring enough as a "seed" - especially for new players! - that it is worth having around and even writing at the top of a character sheet.

    As for 'detect evil' and such, I am in agreement.

    1. It might be of value for new players to help shape their conception of what is heroic and not, I'll agree there, though I'd rather there were more useful tools provided to D&D for that purpose.

      Still, one of the best campaigns for characterisation that you and I ever played was the Shadowrun game where Rob was Shane Amarillo and I played my grandson, and we didn't have alignment for that. Nor am I sure that categorising Amarillo as "chaotic good" or Jack Liberty as "chaotic evil" (which would be my categorisations) would have done more than deflated any chance they had of being friends.

      As someone who plays chaotic evil characters fairly frequently yourself, I'm sure you know how frustrating it can be when you tell people out of game that you're CE and they start jockeying in character to deal with it, as if you had it stamped on your forehead in game. The lack of rigid categories in the Shadowrun game really allowed us to explore that relationship without preconceptions about how they should relate to one another.