How Character Death Lands D&D in a Tug-of-War Between Game and Story

Character death has always torn Dungeons & Dragons between game and story. The threat of death makes the game exciting, but actual death brings a character’s story to an unsatisfying end that can disrupt a campaign.

Without characters facing the potential of dying, the game lacks tension and stakes. Decisions lose consequences because however characters act, they emerge unharmed—stronger thanks to experience. At his crankiest, D&D co-creator Gary Gygax railed against players who wanted campaigns that never threatened characters. “How I detest namby-pamby whiners that expect to play a real RPG without threat of character death or loss of a level, stat points, or even choice magic items! Without such possibilities, what it the purpose of play, a race to see which character can have the greatest level, highest stats, and largest horde of treasure? That is just too flaccid for words.” (See The Game-Design Trends That Turned D&D Into a Game Gary Gygax Disliked.)

The D&D Open moments before my character (lower left) died to a Marilith

My favorite D&D sessions came in the D&D Open, tournaments where long odds made casualties almost inevitable. (See Why the Awesome Dungeons & Dragons Championship Should Return.) The peril adds a weight and urgency to play that made ordinary game sessions pale. In D&D campaigns, the threat of death makes the characters feel like heroes because they brave real risks that can end their imaginary lives.

But while the threat of character death adds excitement and vitality to D&D, actual character deaths usually sour the game.

In those tournament games, dying meant dropping from the session. One player’s fun ended. In campaign games, death proves much worse. We players invest time in our characters. Creating one takes a half hour or more of effort to build, and then to invent a history, personality, and so on. Playing a character to higher levels demands hundreds of hours. A dead character seems to waste that investment. (Funny how we simply stop playing characters without feeling the same loss.) We become emotionally attached to our characters. We imagine story arcs for them where they grow and change and reach a meaningful conclusion to their journey. Death stifles those plans. Losing a character hurts.

Character deaths can also hurt campaigns. Today’s roleplayers foster connections between characters in a party. Dungeon masters strive to link a campaign world and its conflicts to characters. A death cuts all those connections and stalls story threads without resolution. The group faces the problem of introducing a new character into play and possibly reviving broken plot threads.

Through the years, D&D’s designers coped with the problem of death by making characters harder to kill. Originally, in the Blackmoor campaign that led to D&D, a single hit killed a character. “It didn’t take too long for players to get attached to their characters,” D&D co-creator Dave Arneson said. He introduced hit points. (See The Tangled Origins of D&D’s Armor Class, Hit Points, and Twenty-Sided Die Rolls To-Hit.)  In the 1974 D&D game, a character reduced to 0 hit points dropped dead, beyond healing. By the release of Advanced Dungeons & Dragons (1979) characters fell unconscious at 0 hp, and only died at -10 hp.

Even in original D&D, spells like raise dead often made death into a brief setback rather than a permanent loss. In the Dungeon Master’s Guide, Gary wrote, “Death due to combat is no great matter in most cases, for the character can often be brought back by means of a clerical spell.”

Resurrection magic spares characters from dropping out of play, but to hold some stakes, death must keep enough sting to make it feared. The original raise dead forced characters into two weeks of rest. In busy campaigns like Gary’s, that sidelined a character and forced players to bring one from the bench. (See Dungeons & Dragons and the Dream of the Grand Campaign.) Still, a two week break rates as a trivial setback, especially when few campaigns keep a calendar. AD&D added a risk of failed resurrections and a limit to the number of resurrections. Third edition features the best version of raise dead: Revived characters lost a level. This consequence adds enough fear of death to lend tension to the game, while limiting the chance of removing a character from the game.

AD&D’s reincarnation spell rates as the worst example of resurrection magic. Players would have characters kill themselves until they reincarnated as an ogre mage. Clearly, those campaigns featured very little role playing.

Fifth edition suffers from D&D’s most namby-pamby version of raise dead. After a long weekend, resurrected characters shake all the effects of death. Further, most characters can afford resurrections without a loan. (See Dungeons & Dragons Stopped Giving XP for Gold, but the Insane Economy Remains.) This lack of risk can drain the game of excitement. No wonder the designers of the Tomb of Annihilation adventure felt a need to make death more threatening. The circumstances of the adventure temporarily blocked raising the dead and imposed a risk of permanent death. Gary would be proud.

While raise dead can benefit the game, resurrection magic must give D&D novelists fits. How do you place characters in peril when rules-savvy readers know how easy resurrection should be? I’m told most D&D writers just ignore such easy miracles in their tales. Who can blame them for looking away and whistling past the empty graveyard?

Fifth edition’s forgiving rules for character death make characters above 4th level just about impossible to kill. Because fallen characters usually survive if anyone stands to revive them, total party kills have become more common than individual deaths.

In modern D&D, only new characters typically die. The designers embrace this quirk. To them, a new character represents a small enough time and emotional investment to feel disposable. Still, new players typically play new characters, so the design risks distressing potential converts by dealing abrupt deaths. (See Sly Flourish on Building 1st Level Combat Encounters. Hint: Have a kindly priest cast aid on your players’ 1st-level party.)

While the risk of character death has shrunk, the threat of death has always remained part of D&D. Despite ranting against players who failed to risk characters’ lives, Gary saw the tension between needing death in the game and the pain of actually losing characters. In the Dungeon Master’s Guide, he writes, “It is very demoralizing to the players to lose a cared-for player character when they have played well,” and then he suggests ways to avoid killing a character. But in the same paragraph, he warns DMs to never tamper with the system shock roll that a character must make to be successfully raised from the dead. “There MUST be some final death or immortality will take over and the game will become boring because the player characters will have 9+ lives each.”

Next: Death, stakes, and bargaining. How letting characters opt out of death might improve a D&D game and its story.

This entry was posted in Role-playing game design, Role-playing game history and tagged , , , , , , . Bookmark the permalink.

10 Responses to How Character Death Lands D&D in a Tug-of-War Between Game and Story

  1. PK says:

    One of the greatest oddities of 5E is that, due to enemy HP scaling, it is a pain in the arse to win a battle against powerful enemies, and yet it is very hard to actually *lose* a fight thanks to how easy it is to get characters back on their feet. This combination of low stakes and tedious busywork has made 5E combat incredibly dry.

    I much prefer B/X’s “combat is incredibly lethal, you have maybe 3,4 hp on average, so avoid combat at all costs” playstyle. There’s actual threat, and when we succeed, we know we earned it. And if we die, it takes 10 minutes max to whip up a new character. Often less.

  2. Jacob Blalock says:

    I tend to find that with 5e the limitations of Revivify and Raise Dead tend to make enemies that disintegrate (e.g. Zombie Beholder) or destroy vital body parts (e.g. Mindflayer) responsible for a disproportionate number of player-character permadeaths.

    Outside of games with a shortage of healing the only other deaths come from moments when PCs are distant from the healer(s) or when an enemy that has swallowed or grappled a dying PC flys or burrows away with their midnight snack.

    This makes me feel as the DM like I have a lot of control over player death, which I like. It does have the downside though that 5e players sometimes feel personally attacked when you decide to up the stakes.

    After Level 13 permadeath is pretty much non-existent (like with any fictional superheroes), but you can give a baddie a “Nine Lives Stealer” and make off with a PC’s soul to create an extended quest around character death and resurrection.

    • DMDAD says:

      “This makes me feel as the DM like I have a lot of control over player death, which I like.”

      Its funny, this is the thing about 5E as a GM or a player that I really don’t like. Killing a downed player is pretty easy by having the creature crit hit stab a character causing 2 death save failures per hit. However as the DM that means I’m making that choice to specifically kill the character, which kind of makes the DM a jerk. I prefer the dice to be in charge of death like the previous editions where dice determined whether a PC went to -10.

      For me 5E plays a bit like a game of chess where no one is allowed to take your queen. Its like chess, plays like chess, can feel fun like chess, but unless you really make a major mistake the outcome is pre-ordained. As a player it is fun to win, but it lacks a certain quality of drama and meaning. Much like the video game fantasy genre that 5E leans into so heavily, ie. you put your time in and you eventually finish the story – which of course is big fun for a lot of people, just not as much for me.

  3. EvilDan says:

    Having killed and been killed in all editions except the last, character death was everpresent and led to the best character stories and usually a scar. Having a resurrected wizard come back and not be able to grow hair from a fireball gone wrong, did nothing but give my player a reason to buy and paint a new fig. We kept # of times able to come back = CON score but without being able to die, there would be no tension. And if you total party killed, you made new NPCs to PCs to retrieve the corpses. Campaigns are built on rise and fall. To lose that is the short comings of an inexperienced GM. We can all make it better.

  4. Keith S says:

    The idea that “a death cuts all those connections and stalls story threads without resolution” ignores the impact of character death on the rest of the party, and suggests that the story depends on one character. That viewpoint trivializes the efforts of the party and its surviving members.

    As a DM I try to keep an eye on where the PCs are danger-wise. Not to bail them out, but I may drop a hint to the party if they’re unaware of the peril and the player hasn’t indicated it. Since 2008 my four play groups (about 20 players) have experienced three character deaths. They’ve all been important milestones in the stories, not endpoints for the story. Most of my players have vivid memories of the events. Two of three players whose characters have died have told me they appreciated how the deaths were handled and what they meant to the story. The third is a casual player, who was content to roll up another character similar to his first and just play on.

  5. alphastream says:

    2E and earlier editions simply had fragile PCs. You could die easily, end of story.

    3E had monsters that were absolutely brutal at all tiers, plus some really exploitable loopholes (such as non-associated class levels) that created sky-high challenges. This all meant that characters could easily die, even when they had full hit points, if the DM knew how to craft monsters. Unfortunately, it was incredibly taxing to modify monsters.

    4E gave PCs too much of a safety net between HPs and healing surges, though the edition also had some amazing challenges (especially after the devs went back and corrected the monster design math).

    5E… 5E on paper looks more fragile than 4E, but it has not been in play. Characters are very resilient and have a lot of hit points compared to monster damage. Monsters are often given special abilities and to balance that they do less damage… but the abilities don’t actually threaten PCs with death. This problem is even worse at high tiers of play, where monster damage is absolutely shameful. Most monsters have no chance. If they hit 100% of the time they still could not drop all the PCs to 0 hit points. And when that isn’t the case, there is no way for the PCs to be defeated in most fights. To me, the 5E solution is pretty simple: add damage.

  6. HDA says:

    Character death only derails your story if you have one planned in advance. At that point, it’s not much of a game, is it? The disconnect has been revealed by a character’s death, but in that case it was always there.

  7. Abelhawk says:

    I have a couple of house rules that make death a bit more dangerous and limiting:
    1. When a character is brought to 0 hit points, they gain a level of exhaustion. Levels of exhaustion gained in this way go away after a short rest, or if the character is brought to half their hit point maximum.
    2. When a character dies and is brought back to life, they receive one permanent death saving throw failure. A character with three permanent death saving throw failures cannot be brought back to life by any means.

    • Ilbranteloth says:

      I like #2, that’s an interesting idea (and a scaled back version of the AD&D limitations on how many times a character could be resurrected.

      However, in the last 30+ years, resurrection magic has basically been unavailable in my campaign. Combat is deadly, and PCs tend to avoid it as much as possible, or at least until they can plan ahead to get the best advantage. Having said that, there are still plenty of times where they don’t have a choice, and so they fight smart and use a lot of tactics. Our combat rules are a bit different too, but even when playing core rules it has an impact.

      But I think the biggest difference, and what David might be alluding to (I’m interested in the next post), is that we let the players decide if it’s the right time for the PC to die.

      More often than not they choose to just let them die, since it’s an ongoing open campaign, they have multiple characters, and we aren’t focused on super-heroic characters in an epic pre-authored story arc. But sometimes it’s just not the right time, for whatever reason. In that case, it may be simply a dramatic scene, or they might choose to suffer some significant, sometimes permanent, consequence, etc.

      We far prefer to override the dice and prevent a death than to go through the resurrection revolving door. And there really isn’t anybody better to decide whether the character has reached their end than the player. Of course, there can be input from the DM/table. And I’ve found that most of the time, the players don’t want to seem like they are “cheating” or whatever. They are usually harder on their PCs than I would be.

  8. Pingback: Can a DM Have Monsters Kill Fallen Characters Without Bringing Hurt Feelings? | DMDavid

Leave a Reply