Horror and Consequences (Traditional and Video Gaming)

I've been playing a lot of video games recently and I spent a few hours in Dying Light, before dropping back in time to Dead Island: Riptide to see how Techland evolved their formula.

Both games are pretty similar, first-person melee-combat action games with looting, and some 'RPG' elements like a skill tree. What interested me is that both games seemed surprisingly close to being full-scale 'true' zombie-themed horror games, in gameplay at least if not in tone. You're scrabbling around in containers to scavenge and repair improvised weapons, managing a limited pool of health that enemies can drain really fast, and your combat options are limited by a small 'stamina' pool where your character needs to take a break after swinging around their weapons a bunch.

---

What separates them from being genuinely 'horrifying' is their lack of any sort of permanent consequences for failure. If you die in DI or DL, you lose some money and/or experience points and get back up at a checkpoint, with all objectives still cleared and all defeated enemies staying dead. Thing is, impermanent consequences are kind of a staple in video games. Even many 'true horror' games like Silent Hill or Amnesia simply restore the player to a save point or quick-save on death. Video games often struggle to make these failures have any impact.

Limited saves or restricted checkpoints are one approach, for example Resident Evil makes you use consumable 'ink ribbons' at pre-defined save points. My issue here is that this approach doesn't give your mistakes 'permanent consequences', but rather just forces you to re-play segments of a game you've already cleared in order to take a second try at a difficult part. In games that have repetitive or tedious mechanics (such as Resident Evil's notoriously awkward controls), this can kill the tension even faster than a consequence-free death would.

Another approach to solving this is permadeath, seen in a lot of roguelike-inspired games and also integrated as a difficulty mode in games including Dead Space. Personally I can't recommend either approach. In the roguelike approach, you procedurally-generate your levels so each layout is unique, but that unfortunately means that you can't have a talented designer hand-placing encounters and designing a consistent and well-paced experience. The full-permadeath mode, on the other hand, has the problems of the 'limited saves' system above cranked up to 11. Now if you die, you have to replay the entire game! Better hope the cut-scenes can be skipped (which they often can't be) and that the game doesn't have any annoying bugs that might cause you to unfairly die (which it often will).

---

'Respawn' and 'checkpoint' mechanics are fairly rare in traditional RPGs, a few notable exceptions being resurrection rituals in high-level D&D, or Paranoia's backup clones. In theory this means you have more permanent consequences, but there's a catch. Making a new PC is usually time consuming, and becomes far more difficult if you want to give them a unique personality. This is compounded if you're playing a more complex system like Pathfinder at say, level 10, where suddenly you need to work out 10 levels worth of classes, prestige classes, feats, skills, attributes and racial bonuses before you even get into questions like "what is their name?" or "what are their ideals?"

Some video games, notably ZombiU have this sort of approach, where a dead player character stays dead and is simply replaced by a new PC. They inherit the same drawbacks too; why make your character unique or invest time into their design if they could die at a moment's notice and need to be replaced? This is where you get all those apocryphal tales of the Dwarf paladin with six identical paladin brothers, from players who just want to play a dwarf paladin and can't be bothered to completely change their approach to the campaign when they inevitably die.

---

I don't have a clear-cut solution for this 'problem' and I'm sure to a lot of people this isn't a problem at all. I just figured it was worth saying something about. For the sake of a satisfying conclusion, I'll drop some potential solutions here.

1. Replace most 'deaths' with a different consequence. The excellent Skyrim mod known as Frostfall has PCs be 'rescued' by-default if they collapse from exposure and hypothermia, with the player waking up in a strange place with stat penalties and missing items. Running out of fuel or food in Sunless Sea doesn't just end the game, but triggers random events and skill checks where your crew desperately try to burn cargo as fuel, or draw lots to see which of them gets used as replacement rations. Maybe the Medusa doesn't kill the whole party, just petrifies them for 80 years and the campaign resumes from there? If a party member would 'die,' houserule instead that they've been 'critically injured' and need something like an expensive operation or expert surgery or a kick-ass elvish plant-based prosthetic cyberlimb.
2. Defer death with a wounds or debilities system. As PCs get closer to dying, they rack up penalties such as concussions, burns, or missing body parts. Check out the OSR wounds table on Cavegirlgames for a great example. Ideally, the players will slow down when they realize they're at risk of permanent damage, or have to make the conscious choice to keep putting themselves at greater risk. This is good if you can give really unpleasant and graphic descriptions of the injuries, so it doesn't work as well if your players are gamist-types and just view those injuries as just stat penalties. It's also not a good option if your players are particularly squeamish or have bad responses to gore; even horror games are played for entertainment and not to deliberately dredge up everyone's past traumas.
3. Have a pool of available PCs to replace dead ones. This needs more investment from the players, since they'll be making maybe three or more characters even if they only use one. Decide for yourself if the pool is shared or individual. This has the same multiple-character problems I listed above, but having players create a pool in advance minimizes downtime while they think of a replacement for that dwarf paladin. Also, if players get invested in their characters they could be genuinely excited to play as someone else from the pool even if they miss their old PC. Finally, you can let players swap characters in and out from the pool in downtime or between sessions, which might encourage people to experiment with new builds or even do 'crazy' old school stuff like 3d6-in-order.
4. Have resurrection be an option. No, I don't mean 'cart the body back to the village chapel,' I mean real-shit 'go on a spirit-quest into the underworld to drag their soul out.' Cheap and simple resurrection undoes the permanence of death, whereas a potentially fatal quest into the realm of the damned is both a fun concept for a short arc and difficult enough that it won't ever be the first option to solve the party's problems. Having the party play Frankenstein and assemble a new body for the thief's disembodied brain could also be great fun, or having them run errands for the mob to pay off a replacement chassis for the team's pilot AI. Just remember that the key immersion-breaking question of resurrection in any setting is "why doesn't everyone just do this instead of dying?"
5. All of the above, at once. Overkill? Maybe. But the point here is consequences. The Bard's bad choice to seduce the dragon could lose them an arm, and then they start to bleed out, so the party needs to get them to a surgeon but they pissed off the surgeon earlier by stealing from them so the bard dies, so the party tracks down a powerful Paladin of Debauchery (from the character pool) to aid them on a noble quest: to enter the realm of the dragon god and free the Bard from their clutches in the name of Eros & Bacchus!