Skip to content


I’ve been thinking a lot lately about the problem with killing creatures in tabletop roleplaying games such as Dungeons and Dragons. I don’t think the problem is the violence itself. We play these sorts of games to escape into a different world and engage in heroic conquest in ways that we couldn’t (and/or wouldn’t want to) do in real life. As much as I loved Undertale, I’m not wishing that every roleplaying game was committed to making the case for pacifism. But I am thinking that those of us who run tabletop roleplaying games as the GM/DM/storyteller/etc. really need to rethink the way we set our players up for combat. I think this could apply to video games too, but the consumer is less often also involved in designing scenarios.

I have a confession to make: I ran two major campaigns in 4e D&D, and in each of them I railroaded the party into fighting a lot of things. The worst part is that I lead them into evil, “the termination of infinite play in unheard silence” (p. 32, Finite and Infinite Games). Evil, not just in the pitched battles with the “boss” creatures, but in the process of killing enemy mooks for experience to gain the levels they need to go toe-to-toe with higher-level enemies. Too many monsters or generic cultists were willing to fight to the death without having any purpose in the story except to be designated targets. If there is one thing I regret in my “A Trip to Castle Stirling” campaign, it was not setting a breaking point for all of the enemies (with appropriate punishments for good-aligned clerics and paladins who go for the kill after a surrender).

If I kill this, I will gain XP, which will make me better at killing other things, which may be guarding magical treasures I can use to kill progressively stronger things. I am living the murderhobo dream.
If I kill this, I will gain XP, which will make me better at killing other things, which may be guarding magical treasures I can use to kill progressively stronger things. I am living the murderhobo dream.

It makes sense, though. Take a look at your published modules for your favourite sword-and-sorcery tabletop game. See how they provide a wealth of statistics involving hit points, armour, movement, damage, attacks, etc. What don’t you see as often? How to defeat rather than fight monsters to the death. Will lizard people respond well to bribery? Dryads to flattery? Are demons actually quite cowardly? A skilled and determined DM would be able to improvise something, but the default course of action is to fight the minotaur until its HP reaches zero, it is dead, and the players gain XP and treasure. I did a little bit of experimenting with NPC-surrender with named NPCs and one group of snake-people in my last 4e campaign, but in retrospect it is kind of terrible that this was a variation rather than what usually happens. I can do better. We can do better.

I just finished playing in a tabletop campaign run under the Risus system. It’s better suited for short one-session games, but my regular tabletop group has found out that it is possible to keep it running as long as any D&D campaign. One of the most interesting ideas I came across in this system is that there is no such thing as HP. You run on cliché dice, and once you have been brought down to zero, you are out. Defeated. Not dead, unless it is explicitly a fight to the death, and the winner chooses the consequence to be death rather than some kind of last-minute mercy).

I ran a one-night Risus game a little while back, where the premise was that there is an annual scavenger hunt at a fictional university which everyone becomes irrationally obsessed with. Except one year when a mad scientist type became so obsessed with winning a broken version of the contest he ended up holing up in the abandoned areas of the basement and playing out a Phantom of the Opera sort of trope. The game ended when the party of player characters was able to outmatch him at his game. In the end, they defeated him more decisively than they could have if they killed him and made him the tragic protagonist of his own story. They responded to his outrageous villainy by handing him the old trophy and declaring him the winner of the impossible contest. He was left speechless, dumbfounded, and completely lost. The sense of pure and unadulterated defeat was palpable. It was wonderful, possibly one of the finest moments I have ever had as a GM. I want to do more of this, and less guiding players into the kill-XP-level-kill cycle. My future campaigns will have a Mercy button.

So, I propose that for the purposes of determining defeat conditions the DM/GM/storyteller/etc. should consider enemies in combat encounters as being part of one of three groups based on function:

  1. Robots: things that are designed or built with the specific function to fight and kill the players, includes not just mechanical robots but also reanimated skeletons, raised zombies, etc. They may be able to speak and understand language, but for them it is more like how computers understand input.
  2. Animals: creatures which are alive, likely sentient, but not capable of higher reasoning. This would include traditional animals, fantasy beasts, as well as anything else that runs primarily on instinct. Creatures such as zombies could also be this if they are created by a natural phenomenon rather than a person. They can communicate emotions but not ideas.
  3. People: sentient, intelligent beings capable of abstract thinking. These are not necessarily organic, bipedal, and humanoid. These are characters with agency, judgement, feelings, beliefs, values, and motives of their own. Sometimes we want to populate our fictional towns with people but end up putting a lot of human robots in there instead of human people. Where possible, people should be made to function as people.
Sans the skeleton and Flowey the flower are not people in form, but are people in function.
Sans the skeleton and Flowey the flower are not people in form, but are people in function.

For this purpose, the finer points of what constitutes the difference between an animal and a person at an ethical level are up to your group to decide. By all means, consider a cat to be a person, a cultist to be a robot, or a feral clockwork automaton to be an animal. I am not suggesting that this classification be used to determine whether or not it is ethical to hurt/kill/destroy the creature if it is attempting to coerce the players in some way. That is up to the players to decide. I am only drawing this distinction for the purpose of suggesting how each should be defeated other than a fight to the death:

  1. Robots should be able to be defeated by circumventing their programming or mechanics. Robots, if sufficiently provoked, will attempt to fight the player to the death unless specifically prevented from doing so. Of the three, it should be easier to justify destroying these than either of the other two types of creatures. By all means, smite the necromancer’s summoned undead without a second thought.
  2. Animals, when provoked, will try to fight to the death unless they become scared. Fight or flight. To provide a solution to avoid a fight to the death, the DM should back up one step and make the encounter with an animal not lead to a fight unless specific conditions are met. Additionally, the conditions under which the animal will submit or run away out of fear should be spelled out in your notes. The players should not be too harshly punished for killing the owlbear if they really felt like they had to.
  3. People should rarely try to fight to the death if given the chance to be spared, unless they are true believers in a cause. These are the hardest creatures to plan for, but that is what makes the tabletop game such a great medium. You can improvise a little at the table. People should always have a breaking point. Perhaps it is fear and self-preservation, perhaps it is greed, or perhaps it is an appeal to their conscience. Players may choose to execute defeated enemies who are people, but that should lead to more consequences than an experience point reward. Killing people should always require justification. Defeating them should be more satisfying than killing them.

Simply by considering what the function of the enemy is can help determine how we let players go about defeating them without a battle to the death. You can only go on so many super-hammy rants of to-the-death defiance as a DM before it starts losing its impact. By the time you hear it from the main villain of the campaign, it starts to sound the same as when the hobgoblin deckhand swore to end you at any cost because you wrecked his ship. If the hobgoblin is functioning as a person, let the players convince him to give up piracy instead. And don’t just bring him back to assist in the battle when you later fight the pirate lord to make it harder than if you had simply executed him earlier, either. That reduced him to a talking robot. The players should also have the option to stab him in the back on his way out of a holding cell. I am not calling for DMs to enforce morality, but rather to provide for the existence of a reasonably tenable morality that players can play with and around.

So, when you are designing encounters in a tabletop roleplaying game rather than a combat simulation game, consider making combat and death a part of the game but make it the means to an end rather than the goal itself. I think it makes the game more interesting whether the players have the choice of finding a non-violent solution, fighting only to submission, or choosing to go all the way. Giving players the choice of how to solve problems allows for a wider range of play and that fills me with determination.

Published inTabletop RPGVideo Games

Be First to Comment

Leave a Reply

Your email address will not be published. Required fields are marked *