Historically, D&D has been both too indulgent and too heavy-handed in its approach to being wounded in combat. On the one hand, magical healing is cheap and easy, and fixes everything. There is very little that can happen to you in a typical D&D combat that can’t be fixed by the first-level healing abilities of a cleric, druid, or paladin (from 2e onwards), or even a bard (starting in 3e). Sure, cure light wounds can’t fix being poisoned, or blinded, but, then again, how often do those things come up in combat? There isn’t much restriction in saying that magical healing can’t regrow a lost limb if there’s no way to actually lose the limb in combat in the first place ... and there isn’t, in most editions of the game. You can be at zero hit points—
On the other hand, if by some mischance you do manage to die, then you’re just boned. Sure, there’s raise dead and similar spells to allow characters to ignore even the worst possible outcome of being wounded, but those are very high-level spells, with the net effect that, by the time you can cast them, you probably don’t need them any more. Your biggest chance of dying is when you’re low-level, when neither you nor anyone in your party is even close to being able to cast raise dead (or resurrection, or restoration, or regeneration, or even reincarnation, which is about as lame a death-defying “R” spell as there is). Of course, you could hire some high-level cleric to cast it for you, but that requires a lot of gold ... which, again, at low levels you’re unlikely to possess. And, even if your party has the cash, they’ve still got to stop what they’re doing, perhaps right in the middle of fighting their way through the dungeon to the ultimate boss fight, battle their way back out to the surface (carrying your lifeless corpse), and then trek back to town, lay out a huge wad of gold pieces, and finally start all over again. And historically it has been ridiculously easy to die in D&D: Gygaxian lore is full of stories of instant death for characters and potentially apocryphal quotes like “I can’t tell you how you died, because your next character might enter this room too.”
So, overall, traditional D&D has had no consequences for getting hurt, until you’re dead, at which point the consequences are overwhelming. And recent editions haven’t improved the situation. Oh, sure: they’ve attempted to address problem #2 by making it harder to die. Nowadays, instead of being dead as soon as you get to 0 hit points, you’re only dying at that point (to steal a phrase, you’re only mostly dead), and, depending on which edition we’re talking about, it can be anywhere from trivial to convoluted to slip over to the other side. But this doesn’t really address problem #2: it only postpones it. It makes death a bit less likely, but it’s still exactly as much of a pain in the ass when it finally does happen. And their solution to problem #1 is to put their fingers in their ears and repeat “there is no problem!” over and over until we almost believe it.
What it all comes down to is consequences. If you play a roleplaying game where there is no possibility of dying, there is nothing at stake. The risks are not real, and you have no motivation to play it smart, to avoid rushing into danger, to occasionally decide to back down and live to fight another day. Because you know you’ll live today. I had a game once where I (as the GM) described hordes of goblins guarding an objective, and my players said, “okay, we’ll just go in there and kill them all.” Because they knew they could, and they knew it might take more rounds than the typical combat, but so what? Eventually they would prevail, because the goblins couldn’t possibly kill them. They’re goblins, after all. What’s the worst they could do? “They’d have to crit me just to hit me,” one of my mathier players pointed out. “So one in 20 will do a little damage. There’s, what? a hundred of them? So about 5 of ’em will do a little damage. I’ve got dozens of hit points and the cleric could heal me if I needed it, which I won’t. In fact, let me save you a bit of trouble and I’ll work out exactly how many of them I can kill every round, so that way you’ll know exactly how many rounds before they’re all dead.” Back of the envelope calculations does not an epic battle make, and I wasn’t even looking for an epic battle. I just wanted them to solve a problem without slaughtering all the natives for once.
So there must be consequences. Sometimes you’ll read articles on the Internet about how D&D was actually better when you could die at the drop of a hat. These are good articles, by knowledgeable, erudite players. But I think they miss the point. What they’re really saying is that the game is no good if there aren’t consequences. And I don’t disagree with that at all.
But is death the only possible consequence? the only consequence of consequence? You see, there’s a pretty big problem with death: creating a new character is a huge investment of time and effort. And I personally, as a GM, have already talked about why I play D&D and my philosophy that an RPG is a shared story, and, as in any story, character is king. So I not only want my players to put the normal amount of effort into creating their characters—
So there has to be something better. There have to be consequences, but death is too much. So the pledge I make my players is, I will not kill your character, unless you agree to it. We are after all telling a story, and sometimes the characters in a story die, and that’s right, and proper. If you think your character should die to advance the story, or you just want to try a new character and want to have your existing character go out in a blaze of glory, I’m all for that. We’ll give them an absolutely glorious death. But, barring that, I promise you that I won’t kill your character.
But I also promise you that, if you are reckless, or careless, or sometimes just because the dice gods are cruel, there will be consequences. Every time you hit 0 hit points, there will be a lasting repercussion, and it will not go away just because the party cleric tossed a few healing spells your way. Maybe it will go away on its own (eventually), or maybe you can undergo a quest to get it sorted, or maybe you’ll just be stuck with it forever ... disabilities make for fantastic roleplaying opportunities, after all. What sort of consequences are we talking about? Oh, the possibilities are inifinite. You could lose a finger. Or a hand. Or an arm. Or a leg, or an eye, or a spleen. You could be blinded, deafened, lamed, or paralyzed from the waist down. You could be scarred or burned horribly and have your charisma impacted. You could suffer a concussion and have your intelligence lowered. You could go into a coma and wake up days, weeks, or months later ... possibly with a crippling new phobia. You could be driven partially or completely insane in very creative ways. You could stub your toe, or you could have to be carried around on a travois by your companions and fed soft foods through a straw. Don’t ever imagine that there are not fates worse than death. (For more great ideas on terrible things a GM can do to players, check out this article from Dice of Doom.)
Which is not to say that I want all my players to be living in constant fear. But a little bit of fear is healthy. When I tell you “there’s a hundred goblins surrounding the encampment; how will you get in?”, you do not want to tell me that you’re going to hack your way through. Trust me: this does not end well for you. And, hey: if you get your hand lopped off, you can always jam on a stylish hook. Also, an eyepatch makes a lovely fashion accessory. Or, hell: jam a magical gem into that empty eye socket and scare the living daylights out of any low-level foes you happen to meet. The possibilities are endless. So, to the question at hand, I say forget death: I choose consequences.
Insert evil laugh here.