It's a Mad, Mad, Mad, Mad WorldTLDR Ditch the DMG's clunky insanity rules and use the Exhaustion mechanic instead.
Ever since I first started reading Edgar Allan Poe in Middle School, I”ve been fascinated with insanity as a literary device. The Cthulu tales, with their stories of narrators driven mad by incomprehensible cosmic forces, appealed to me as well. But, then or now, I never got around to playing any Cthulu style horror games that featured an insanity mechanic as a core element. As intrigued as i wqs by it, i stuck to trasitional heroic fantasy games with the occasional forays into sci-fi and superhero genres. Fifth Edition of Dungeons & Dragons has optional rules for insanity (DMG pages 258-260) and, since the release of the Curse of Strahd campaign book, I’ve seen a lot of gamemasters consider applying those optional rules or houseruling their own. I find the insanity rules given, and many of the ones suggested by the community, a bit problematic and so decided to offer an alternative.
So what do I find problematic? D&D is primarily a heroic fantasy game. Unlike in a Chthulu style horror game, player characters are not meant to be representative of ordinary people. In a horror game, players generally play characters who are regular folk caught up in extraordinary events that threaten to fracture their comprehension of reality. A heroic fantasy character is just that, a hero, and, by definition, not bound to the limits of normalcy. Just by virtue of being first level, they have already proven themselves to be extraordinary individuals. On a daily basis, they undertake adventures that, frankly, in their most mundane forms, would probably be enough to separate an ordinary person from their sanity.
Horror games tend to riff on the real world in the sense that the every day experience of the characters shares the narrow band of mundane reality that we, as players, experience, until they are caught up in the supernatural events. The dissonance between the regular world of expectation and supernatural worlds, along with the nature of that other world, is the mental threat that must be, at all times, resisted. In a fantasy game, however, the world of expectation is already supernatural. The band of reality that is mundane to the character is almost impossibly vast. From the cavalcade of different species that inhabit just the civilized spaces oto the daily contact with all manner of forces divine and infernal, what would it take to shatter the mind of even a low level Warlock, for example, who, on top of whatever things she sees on a daily adventuring basis, is already, by choice, reaching out into dark voids to convince alien entities to fuel her arcane powers?
Of course, any game world can set the bar however they’d like so let’s assume that even these hardened adventurers have their thresholds and Strahd, or something else, is going to push them right to those edges and, perhaps, beyond. Bring on the mechanics.
The DMG does a couple of things. First, it mentions either using a dedicated Sanity attribute or, alternatively, using Wisdom or Charisma for madness related saving throws. It then lays out three different levels of insanity, separated by their severity: the first lasts minutes; the second hours; and the third is permanent. The effects range from reflexive reactions, in the short term, to various compulsions, delusions, and physical ailments.
For example: Evilynne the Warlock is subjected to visions of horror unimaginable even to one such as her. She fails her sanity check. As the GM, I decide that the experience is so powerful it leaves her permanently changed. I roll a d100 and the result is 78. She is now convinced that powerful enemies are hunting her, and their agents are everywhere she goes. she is sure they’re watching her all the time.
First things first, whether you go with using a dedicated Sanity attribute or not should come down to how fluid you want sanity to be. If you want to model the crumbling resolve of beings forced too far outside their mental comfort zones for too long, the Sanity attribute will model that. If you want to reflect the effects of jarring shocks to otherwise steady, hardened minds, using the existing attributes works fine. No particular problems here. The problems, for me, arises with the various effects and their durations.
Let’s start with the durations first. Even the shortest term condition lasts a minute so, if it occurs during combat, that character is, essentially, out of action or, if not, then their choice of actions are extremely curtailed. In any event, there is no additional save so, unless cured by external means (spells, basically), until the duration expires, they’re stuck. The next tier lasts from ten hours, at minimum, to a hair over four days. Given the strangeness of the time flow in the average game, this could either last a ridiculously short amount of real time, for the party that manages to take long rests after every encounter or has the luxury of regular downtime, or almost forever for the party for whom an hour or two of in-game time can take a whole session to resolve. Again, there’s no second chance to save out. If a spellcaster of the right stripe isnt available, the character is stuck in their psychosis until the timer unes out. For the last tier, they’re just stuck for good. Depending on your style, this might be less an issue than a feature. Given the nature of the effects, however, I land on the problematic side of the spectrum.
Almost all of the effects given in the DMG‘s madness tables are roleplaying-based, that is they express the effects in terms of the fiction as opposed to the mechanics. The mechanical impact of these effects must be adjudicated based on the fictional actions. Confused? Maybe it would help if I step back for a second.
Most often in a game like 5E, the fiction, that is what a character does or does not do or what does or doesn’t happen to that character arise from the mechanics, as expressed by the rules and results of the die rolls arising from those rules. For example: John plays the character of Evilynne. John decides that Evilynne will attack a cultist. He makes his rolls and the results give him a critical hit that drops the cultist’s hit points to zero. So far, this is all mechanics. What does it mean to what happened within the world? Perhaps the GM would allow John to describe how Evilynne impales the cultist on the end of her rapier. That description is the fiction. The attack and damage rolls, the subtraction of hit points, are all the mechanics. With this in mind, let’s take another look at the madness effect I used in an earlier example:
I am convinced that powerful enemies are hunting me, and their agents are everywhere I go. I am sure they’re watching me all the time.
Easy, right? It generates a clear roleplaying perogative for the afflicted character. But how does this condition affect the mechanics of how the character plays? Firstly, in many if not all games, powerful enemies are hunting the character and their agents are everywhere and watching all the time. A player could argue, probably with some success, that this condition is barely more than a slight increase on the paranoia their character already feels. On top of that, as the GM, I have to now decide what mechanical penalties result from this condition, when to apply them, and then discuss, cajole, or, in some cases I’m sure, argue about why I am applying penalty X in situation Y. Indeed, I can see some players asking for mechanical advantages due expressly to this condition: I think Evilynne should get advantage on her Insight check because she’s so suspicious that she has a better chance seeing through any ruses. Maybe but that’s not really the point, is it? While a few of the effects do either include mechanics or are entirely mechanical, the vast majority, like this paranoia effect, are not. So what are our options?
First, let’s take a look at creating a mechanical madness effect and let that determine roleplay instead of working the other way around. I’m going to make it dead simple and just call it disadvantage on ability checks. It’s broad and it’s vague but it’s mechanically clear as crystal. Everyone knows immediately what it means and how it works. There is zero ambiguity. If you’ve got the condition, you suffer the disadvantage, full stop. But what does it mean in the fiction? This is the easy part, it can mean anything the GM or, better yet, the player wants it to be. Let’s revisit and revise the sad plight of Evilynne the Warlock.
So Evilynne was attempting to decipher an eldritch scroll written in the language of a foul outer plane when it’s alien potency was too much for her mortal mind to bear. As a GM, I inform John, Evilynne‘s player, that she now is at disadvantage on ability checks. I then ask him, what did Evilynne experience when reading thay scroll that fractured her psyche? John can then tell me what ails the warlock. Obviously if, as the GM, I don’t find it suitable, we can suss it out in a discussion, but, by default, the player can pick something that appeals to him to play. If the player doesnt like the concept of paranoia, he need not choose it. Whatever he does choose, he’s more likely to actually play up because he chose it.
Now, we could get more granular with the effects but I’m not sure it would be beneficial to do so. The more broad and vague the condition, the more open it is to all manner of fictional interpretations. It’s also both a serious effect but not completely disabling, ammounting to a steady -2 penalty to rolls. It can hurt but it can also be overcome. Contrast this with a condition that dictates the character is frightened and now must expend every action to run away. Not a lot of leeway for player agency there.
Ultimately, I think the system I am proposing here is less about sanity being lost in dribs and drabs than it is the fight against insanity being lost in dribs and drabs. The hypothetical effect I proposed does not inflict insanity on the character as much as it describes a loss of composure– ground being ceded to impulses that the character can no longer fully contain. Only at the last, is the character’s sanity truly and potentially irrovocably stripped away.
There are no guidelines listed in the DMG about which severity is appropriate for what kind of experience. Ok, fine, so I can think of it myself. Sure. But often this becomes a source of contention: how do you stay consistent? what’s fair? There’s also no simple escalation method. If a character is afflicted by one short term insanity effect and they gain another, do those effects combine into a medium term effect? Do you just double-up? Something else? It feels a bit half-baked.
Instead, let’s use a simple scaling up: start at zero, gain one level every time an insanity check is failed. Effects for each level gained are cumulative with the ones before. Easy peasy. Easy for the player to keep track of, easy for the GM to scan quickly and see which effects are in place, and easy math. Find some restorative peace? Lose a level. Hear the mad whispers of a demonic sludge? Gain a level. No mystery. Everyone can understand and predict the process.
Exhausting the Possibilities
Often when I’m thinking of a new mechanic, I want to take stock of the mechanics already in the game, see what I can hijack or piggyback on, before I go ahead and start grafting something new. In this case, I thought of a mechanic that perfectly fits the idea that I’ve been building for handling insanity: Exhaustion.
Exhaustion really is a perfect fit for what I’m aiming for. It’s got ascending levels of severity and cumulative effects that are mechnical, rather than fictional. Even the name makes sense: mental exhaustion is a thing, as much as physical exhaustion.
|1||Disadvantage on ability checks|
|3||Disadvantage on attack rolls and saving throws|
|4||Hit point maximum halved|
|5||Speed reduced to 0|
Looking over this list of effects, I can easily justify any of those in the fiction as a result of mental deterioration. As a player, if I’m not comfortable roleplaying a mental illness or delusion, I’m not forced to. What’s more, as a GM, I can use the two together, treating either physical exhaustion or mental exhaustion as one variable (after all, they often go hand in hand, as in the sailor going mad from exposure and lack of water), or I can keep them separate if that makes more sense. In either case, I can harness the existing mechanics and mechanisms. No muss, no fuss.
Let me know on Twitter if you find this useful (or not) or if you have any suggestions or ideas.