Towards better alignment
Can we all agree that D&D alignment is… Weird? It’s, at the same time, part of the cultural landscape (with its own Wikipedia article and popular memes) and something rarely used in 5th Edition. Most monsters have an alignment, but players seem to not really care about this aspect of the game.
Nowadays, if a DM doesn’t deliberately bring attention to alignment, then most people won’t even think about it. I’ve seen a few discussions online about whether Chaotic paladins make sense, but this is probably the full extent of alignment’s impact in contemporary D&D games.
But why is this the case? Alignment is supposed to help new players think about their characters and DMs figure out the best way to create drama; how can we make it do that? Before we come up with answers, let’s first talk about the problems with alignment.
Problems
In my opinion, there are two main issues with the alignment system: lots of players see it as prescriptive instead of descriptive and some alignments “sound bad”.
Saying people consider alignment prescriptive means that they think the alignment chosen at the beginning of character creation should guide their actions during the game. This is a significant problem because the system becomes stifling. Matt Mercer hits the nail on the head in this tweet from 2018:
[…] Alignment shouldn’t inform a character’s actions, actions should inform alignment. Or…just ignore it entirely! — Matt Mercer
A friend of mine’s first character was a paladin and, following the PHB’s “charity and justice” suggestion, they made him Lawful Good… And were miserable for it! My friend just wanted to play D&D and smite some malevolent goblins, not role play Mother Teresa every week. Ideally, they should have chosen an alignment that described their nascent character.
The second problem has to do with Evil. You can spend your whole life debating this concept (as philosophers have done since forever), but the fact is that people don’t want to be called evil even if their actions fit the description1. As Matt Colville put it in a 2016 video:
People want to do things like randomly murder people, but not be perceived as evil. — Matt Colville.
Colville goes on to say that many players use Chaotic as an excuse to create characters that don’t care about the consequences of their actions; in other words, they don’t want to say that their characters are Evil because the word just sounds really bad. This is actually supported by data: according to D&D Beyond, in 2019 almost 50% of characters created on the website were either Chaotic Good or Chaotic Neutral.
This is also an issue because we want alignment to be actually descriptive, not something the player writes for their character’s Tinder profile. As a DM, I want to know the PCs’ alignment in order to know how to better create drama and, if everyone describes their characters as Chaotic Good/Neutral, then I don’t have anything to go off of.
Answers
The first step in making alignment relevant is to clarify (especially for new players) that it is descriptive and not prescriptive. I suggest going as far as presenting it to players in the form of nouns, something their characters believe in instead of something they are. For example, Lawful Good would become Law & Good.
The second step is to change the words themselves to something that players are not afraid to assign to a character. Paraphrasing Colville, Evil is better interpreted as “every man for himself” and Chaos as “people are best when left to their own devices”. Why not instead choose words that better convey these meanings?
My suggestion is Order vs. Freedom and Unity vs. Autonomy. For starters, all of these words have positive connotations2 and, therefore, we can expect players not to reject any alignment based purely on how good it sounds. Additionally, we have room in each axis for morally good and bad people; this means that it is a flexible system that is more likely to reflect characters' actual personal philosophies. If all your characters are Freedom & Autonomy (formerly CE), you can have a Order & Unity (formerly LG) villain that still makes sense.
To finish off, here is my version of the alignment guide found in the PHB; note that, instead of Neutral, I simply omit the axis in which the character is unaligned. Here are my nine alignments and a fictional quote for each:
-
Order & Unity (formerly LG): “We must protect our people and our land. We are the only ones standing between civilization and chaos.”
-
Unity (formerly NG): “We’re all in this together. I don’t care about what happens next as long as that remains true.”
-
Freedom & Unity (formerly CG): “Death to the king! Long live the people!”
-
Order (formerly LN): “You can be with me or against me, I’ll uphold my oath regardless.”
-
Neutral (formerly N): “The world is already too complicated, I don’t like taking sides.”
-
Freedom (formerly CN): “As long as I don’t have serve to anyone, I’ll be fine.”
-
Order & Autonomy (formerly LE): “The first rule is: never steal from another thief. Second rule is: never talk to me again.”
-
Autonomy (formerly NE): “I’ll do whatever it takes to save him. Please do not get in my way.”
-
Freedom & Autonomy (formerly CE): “I know our interests align only temporarily, but I’m glad it’s to chop the head of the king off.”
If you liked this system and think it can really solve the problems outlined in the past section, I suggest reading my post called Worldbuilding pt. 01: cosmology, where I create a whole universe based on these four concepts. See you next time!
P.S.: I’m sure more people have already thought of this specific alternative to alignment (like Matt Colville), but I’ve never seen it written down in this way. If I ever find any other articles that predate mine, I’ll make sure to link them here.
-
According to Mohammad et al. (2013), “evil” and “chaos” are both associated with fear and sadness, while “law” and “good” are associated respectively with trust and joy. ↩︎
-
Again according to Mohammad et al. (2013), “unity”, “freedom”, “independence”, and “orderly” are all associated with positivity. “Autonomy” and “order” weren’t in the dataset, so I used their closest counterparts. ↩︎