GM: "You see a humanoid figure in the mist ahead of you, wearing some kind of ceremonial clothing"
P1: "What kind of clothing? Any symbols I recognise?"
P2: "I'm drawing my .45 revolver"
GM: "The robes are done in green and blue, with strange glyphs and undersea creatures on them. The design is mildly repulsive. OK, he's pointing his finger at one of you guys. Both of you roll a d10."
GM: "Heh heh. (PC2) attempts to turn inside out as his body is gripped by some blasphemous, horrible power. He explodes, splattering everyone with shredded scraps of internal organs and skin. A thick blood-mist hangs in the air. Sanity roll."
Does combat need to work like this?
So, how do you work mortality into a rôle-playing game?
If your players are good rôle-players, they shouldn't exploit the effective immunity from death you have given them. If some players do, the real rôle-players will eventually lose out. Over time, this can lead to munchkinism, as the good rôle-players either leave the group, or stop taking the game as seriously.
The rules get to be exactly the same for all of the characters in the game, players or not. No more of the "bad guys never use armour-piercing ammo" fudges that keep the PCs alive in many Cyberpunk games. The PCs will die, usually by bad luck. Rôle-playing anything other than stereotypes won't be worth it, because there is no time to develop characters properly.
Player characters are given one or more chances to escape a lethal situation during their game life. This can be handled from within the game mechanics (such as Warhammer Fantasy Roleplay's fate points and Torg's possibilities), or as something more abstract or informal, which might be better for rôle-playing.
A convenient equivalent in Cyberpunk games is the loss of a limb or two. The PC survives, but has to take time out to get a cyberlimb. Cyberware has its own special disadvantages, and the onset of cyberpsychosis should limit the number of replacements a PC can obtain (this is also known as the "Black Knight" approach).
If a character gets into big trouble, especially with large organisations, the death of a friend, contact or dependent could be substituted. This provides a good opportunity for rôle-playing, but might be difficult for the gamemaster to justify.
A combination of the above may be used to good effect: by playing the first few sessions of a campaign with lethal combat, the players can develop a respect for the scenario, and the gamemaster can ease off later on, maybe using the "nine lives" rules or similar.
In RPGs which rely on combat for much of their thrill, how should it be handled? Abstract systems allow quick resolution, but limit each character's options and can even deprive combat-centred characters of rôle-playing opportunities. Detailed and more realistic systems require a lot of arithmetic, and if they are automated or computerised, may become rigid and distanced from the player.
It's best to set a percentage of time you want to devote to combat, and pick a detailed system if there isn't going to be much of it, or a simple one if there's a lot of combat. This ties in with your choice on mortality:
Low-mortality (#1) combat means that your PCs probably won't be afraid of getting into fights, and there will be a lot of combat. A simple, quick resolution system should be used.
Lethal (#2) or disabling (#3) combat will not turn up very often if your players are smart, so the best resolution system will be a more complicated one. Detail and realism allow better rôle-playing in combat, with a larger number of reasonable alternatives to choose from.
The example at the beginning of this article is an extreme example of bad combat resolution. There are two major flaws here:
The mismatch of lethality with a simple system kills PCs quickly, and leaves players with few reasons for their character's death. Dying on account of a single die roll, especially when the character's abilities have nothing to do with it, is very bad.
The Gamemaster also used fiat (planned outcome) to kill a PC. In addition to deciding that a player character should die in that encounter, the GM has made it clear (by the rolls) that he doesn't care which one it is. Not particularly conducive to rôle-playing.
The amount of danger the PCs face during the game is the GM's own choice. There is no right or wrong level of mortality for the many rôle-playing games, but some games rely on risk-taking by the PCs for excitement.
A few players might make a low-mortality game unsuitable for rôle-playing by exploiting risk-free combat. The GM can correct this by increasing combat mortality, at the risk of upsetting players by killing or maiming their characters.
Starting with high-mortality combat, and then decreasing the mortality seems better. Even if a PC does die, he or she will have been in the game for only a short time, and the player will have made a smaller rôle-playing commitment to that character.
When using low-mortality combat, keep the resolution system simple. There will usually be a lot of combat, and a simple system saves time. High-mortality combat works best with a detailed system, because there is less combat, and what there is, is much more important.