Despite my being a staunch atheist, this seems like an appropriate day to ramble about my thoughts on death and resurrection in D&D.
There are some gamers for whom life is cheap - easy come, easy go. TPK? Just the way things go sometimes. Grab the dice, and let's make some more!
There are others who recoil in shock at the idea that a character could be removed from the campaign forever by death. All the time and effort they spend creating and developing a character, whether that means stats and legendary magic items or personality and status in the campaign, just can't be allowed to go down the drain!
There are elements of both styles that belong in D&D as I conceive of it. Adventuring can't be a cake walk of beating up on pushover enemies and seizing legendary hoards of treasure. Reward must be balanced with risk. Overcoming deadly dangers is a big part of the game's fun. But so is character and story development, and both the incentives and the opportunities for those are diminished if the cast of characters turns over with every adventure. I see magical resurrection, especially easily available forms such as the Raise Dead spell, as essentially an attempt to bridge the gap between exciting risk and story-protecting plot armor, to preserve the threat of death while allowing greater continuity. Unfortunately, I think it fails.
Nerfing death itself has some serious implications for role playing and player agency. Why would you ever run from an enemy, or surrender to one, if the worst that's likely to happen is that you'll lose some gold (and maybe a point of Constitution, depending on your edition) when your buddies drag you back to the local Jiffy-Raise Temple for repairs? What's heroic about a heroic sacrifice if all that's really being sacrificed is money? The fate of each individual character in a battle is nearly irrelevant;
so long as the group prevails, they'll all be back to do it again.
Death has no more significance than fouling out of a basketball game. You sit out the remainder of the game (or the scene, or the battle, or whatever,) and come back with a clean slate for the next one. You get to go on playing the character in whom you have so much invested, but that was never in doubt. His survival is ordained, and the significance of everything he does is blunted by that fact.
Sure, players will try to keep their characters alive, but not as hard as they would knowing that death is forever. There's literally nothing to be gained by chickening out but the reputation of a coward. The critically important distinction between living coward and dead hero is obliterated. The incentive to find non-combat resolutions to a major encounter, or even a climactic scene, is severely diminished. It's not a question of life or death, but of greater glory and a decisive conclusion or lesser glory and loose ends, and that's a no-brainer for most players.
None of this is to say that the gap between risk and continuity can't be bridged. The problem with meat-grinder games is often one of suppressed player agency. The party's options for completing the adventure and the means allowed them to do so are artificially limited, or the goal is chosen by default because there just aren't any subtler options than "kill monsters and gain treasure." When the only interesting and meaningful choices in the game are the ones made in the thick of combat, then it's going to be a very combat-oriented game, and that means a gauntlet of combat encounters and traps grinding down the party's hp and resources until something gives.
In a game that uses dice to resolve actions, there's always going to be an element of randomness to death. The key is to allow the players to make the choices that determine whether the dice are rolled in the first place, to allow them discretion in how much risk they take. Give them information that indicates what lies ahead, even if only subtly, so that they can make informed choices. Don't force them into combat, either overtly or implicitly by making it a condition of successfully completing the adventure. Don't let the players' expectations force them into it either. They've probably been conditioned to expect that the adventure must end with the villain's defeat in battle, but that's an unfortunate cliche. Death should always be a risk of combat, but combat should be a means of resolving conflict or advancing the story, not the means.
By giving players enough information to assess potential risks and rewards, and the freedom to choose which ones to take and the means to try to overcome them, you transfer the bulk of responsibility for character survival from the DM and the dice to the players themselves. Death still threatens, but it comes as a result of choices, not out of the blue. It becomes simply one possible danger of an adventuring career, not an eventual certainty that requires some means of reversal written into the rules. It regains all its impact and pathos, all the attendant heroism or tragedy or shock or horror, as it would in a good novel or in real life. That's how I think it should be done, anyway.
Cancel my subscription to the Resurrection.