34
Last week's Legends & Lore post touched briefly on divination magic, sparking an ongoing conversation in the D&D Next community about the application of divination magic in games. This is the kind of conversation that applies across gaming of every kind - tabletop, LARP, video games, occasionally even board games. What we're really talking about is a means of asking questions of the universe (rather than a person); effects are generally divided according to what kinds of questions you can ask, what it costs to ask the question, and how clear of an answer you get; more rarely there are distinctions of chance to succeed in asking and other kinds of limitations. The central question is the breadth of divination powers, and how the flow of both story and game intersect with those powers.
I'm going to address specific viewpoints raised in the thread I linked above. I want to be clear that I'm not trying to get the last word by moving the conversation to my own blog - I just have enough to say on this topic that I don't feel like a G+ comment is a very good place to make these points. In future posts, I want to broaden the conversation to LARPing and other forms of gaming, because I can't count the number of conversations I've had about player-directed divination in the particular context of LARPs.
D&D, editions other than 4e
D&D provides a useful set of terms to use in talking about every kind of divination, even though I have years of scathing criticism for its specific handling of the matter. Divination is one of the eight classic schools of arcane spells in D&D. Its efficacy in a game varies wildly according to the style of the game - a fact that renders a whole lot of conversations about divination completely invalid because the two sides come from unrelated assumptions. On the one hand (and my preference), there are games in which mystery, intrigue, and lore are central to gameplay and the players' goals. In these games, even the humble detect evil is far more information than the DM wants to hand out, because it so readily resolves questions. (The fact that someone might usedetect evil to resolve questions of guilt and innocence is a whole different question - but I feel certain that if we had such powers in the real world, we'd use them in just that way.) Anything that amounts to detect lie is, if anything, even worse.
The other kind of game is, presumably, one in which mysteries aren't that big of a deal, and it doesn't matter if the PCs short-circuit them with a single spell. These two kinds of games have so little in common that one needs to be the system's default assumption, while the other is a rules module. Obviously, I'd like to see the style of gaming I identify with supported! I've heard a lot of arguments that all approaches to stories in games should account for divination in its full D&D implementation, though, and I just can't agree with that.
A good while back, Monte Cook, of Monte Cook fame, made a series of declarations about how to handle high-powered characters in adventures. (If you have a link to this post, I don't even know where to begin to look for it.) His advice was to avoid simply negating players' abilities - such as the dungeon that blocks all teleportation in or out, once your PCs have earned access to teleport spells and the like. He argued that simply negating their powers was dirty pool, and it was better to write the adventures in a way that accounted for their abilities and even required their use. This advice took root in the D&D gaming community, and it is often cited in conversations such as these. To quote another of the great philosophers of our time, though:
"Horse hockey!" - Col. Sherman Potter, M.A.S.H.
It's absolutely true that once a player has a power, you shouldn't take it away. It's better still to stop the problem before it's a problem by designing spells and effects with greater care. And you know what? That's exactly what Monte Cook did when he wrote his own variant D&D, Arcana Unearthed; it has far fewer divination spells of any great reach or scope, and teleportation is far more difficult to pull off. It drops the core D&D list of divinations in favor of things like object loresight,creature loresight, and the Akashic character class. The loresight effects use specific, limited questions, and lay out ways that those seeking to avoid detection can foil these spells. Akashics tap into the Akashic Memory to gain particular memories, which at least limits their knowledge to the range of things that people have ever known - anything unobserved, including the future, is off limits. Also, akashic memory requires a die roll to succeed, which is more restriction than most core D&D divinations have. The one big downside of the AU/AE implementation is that they take a long time to resolve, often involving a lot of off-the-cuff invention by the DM.
Contrast this with the open nature of divination - which, admittedly, still isn't that bad, because at least it leaves a lot of room for uncertainty - or commune, or discern location. Commune is a Reddit AMA with your deity of choice, though the entity is mostly restricted to yes/no questions. Scry is, of course, the first step of the three-part game-ruiner, scry-buff-teleport. Discern Location has some significant requirements, but it can be blocked "only by mind blank or the direct intervention of a deity." Admittedly, the one game I ran that continued to 15th level didn't have anyone casting this spell, but I have heard stories from other DMs about how this one spell brought a whole plotline to an immediate halt. Discern lies stops many forms of intrigue plotlines in their tracks - certainly anything that relies on the players to catch an antagonist in a lie through social interaction becomes quite easy.
The main problem with advising DMs to just write around the problems of extensive use of divination is that it ignores how hard that actually is. Spells that allow the caster to compose any question lead to the caster just repeating the process, gaining new information step-by-step, until the effect has been largely the same as if the problem had been solved in a single spell, but with still more of the session in which only one player participated. Further, the action of many sessions is not simply written prior to the session - it is something more emergent, the only logical result of the DM's foundation and the players' actions. Emergent plot is still improved with strong writing, as with the pruning of trees, but it's hardly fair to blame the DM for failing to account for divination in plotline written at such a remove. If the DM has to go out of his way to account for any one effect or group of effects, I think those effects have to justify their existence by a more stringent standard than normal. Legend lore is one of the most widely applicable divinations for getting some information, but the amount of information gained is based on your access to the object, and often takes huge amounts of time, so it has enough limitations and drawbacks that it doesn't get in the way of gameplay. In theory, its usage sets you off on adventures, rather than actually answering questions in full. Further, it is limited to things of legendary importance, though once you're powerful enough to cast the spell, this includes just about any likely antagonist. On the other hand, there are a lot of Greater Mysteries (tm) that are better solved with repeated castings of legend lore than any other means. That's kind of a problem, because if those Greater Mysteries are worth learning, that's a problem the wizard is solving on her own, without any contribution from the rest of the party and without leaving the safety of her tower. I realize, though, that not all tabletop campaigns are based on answering the deep mysteries of the setting rather than slaughtering bad guys and saving the world.
Find the path is justly known as ruin the adventure. Any trap that can be avoided, no matter how fiendish, even traps that require secret passwords, are foiled by this spell. Both Arcana Evolved and Pathfinder got rid of this spell without hesitation, because this is the kind of divination that spoils even games that aren't about mysteries and secrets.
In the aforementioned G+ conversation, a lot of the responses pointed out that the DM doesn't have to give out more information than he wants. This is a more complicated discussion than I think they're acknowledging. See, most DMs that start a campaign at low levels are thinking about how to solve the problems of low-level play, not the problems of a year or more of semi-monthly play. These are not DMs who are warning their players up-front that the parameters of scry are changing to remove S-B-T tactics, or ones who are carefully shaping expectations about the school of Divination to warn players that what they see in the book is not how things are going to work in the game. Absent such a warning, I think players are justly vexed when they finally learn that new spell, only to find that it does nothing like what it says on the tin. "Sorry, in this campaign detect evil only works 50% of the time, and often registers false positives" is about like saying, "I know you've been working toward fireball for four levels now, but it does 1d6 damage in this campaign, not 1d6 per caster level." The spell's description is usually the player's only basis for shaping expectations, and there are a lot of pitfalls involved in making a PC's hard-earned powers fall short of their expectations.
What do you do about this? Well, including a note in the spell description that you get what you get and divination isn't 100% reliable is a good start. Removing all mystical means of detecting lies or compelling truth is a very good idea (or tightly restricting access to such spells). Ultimately, any single spell or set of spells that let a spellcaster resolve a major plotline without any input from other characters is probably bad for the game. If the spells instead carry cost, risk, or challenge to the whole party, that's quite a different matter. It's altogether too easy for divination to become the only investigative tool you need, sidelining things like Gather Information, Search, and so on - and that's just one spell slot out of the wizard's whole arsenal for the day. Even if it were a larger commitment, that only encourages the wizard to stop the party's adventure and hole up in a tower for awhile longer. I think there are really only two cases in which divination magic should be the best tool for the job.
When solving problems that begin and end in magic. Do you have a purely-mystical problem, such as discord of the spheres or the like? I think it's totally cool for magic to be the best answer for magic.
Trying to gain information about magical phenomena, for example: what does this magic item do? What is the effect of the ritual we are about to disrupt? And so on. Spells to gather clues are inherently better for the game's health than cutting straight to the answer.
When every other reasonable path has failed. If you've exhausted all sources of rumors and come up empty (possibly through bad dice-luck, if that's how the DM runs things) or you have given it a sincere effort but can't figure out what comes next, divination is good because it keeps the story moving when otherwise it would grind to a halt.
Using magic to scout ahead in a dungeon is... technically divination, sure, but these tend to be written so similarly to conjurations that the difference is arbitrary (other than, of course, find the path). What I'm saying is, I don't have any problem with spending a resource that matters at the time (a 5th level slot) to do the scouting; I think it's a much bigger deal when the wizard is working from home. I've always wanted to see a few more divination spells that were directly combat-useful, whether it's cursing someone with too much insight or creating images that telegraph where the target will move next, making the attacker's work easier.
There are also arguments that Divination magic isn't overpowered because it's just another area of arms-race. It's certainly possible to structure a metagame along these lines, with both sides of a conflict pursuing divination and divination-blocking spells. On the other hand, the core rules of D&D provide only a few defenses against divination, and most of those provide flat immunity. Since there's no evident counter to that immunity, the rules don't present much of an arms race - and every single villain needs to be equipped with the same spells. Since I don't think we want every story to be about the divination arms race (though it's fine and good for one story arc or campaign), I'm pretty sure that nipping it in the bud and writing in tight limitations on player-directed divination works better in the greater span of D&D.
For tabletop gaming in general, I want to add one quick note about Mage: the Awakening - if everyone is a spellcaster, it's much more okay for information-gathering magic to be your first choice. On its face, I really like the fact that every Realm has its own set of sensing spells, but my players certainly felt like they were not all created equal. Were I to run Mage again, I'd probably review the spells described in the book and look for ways to emphasize the usefulness of each of them. Beyond those initial sensing spells, most Realms have one or two primary ways to gather information - sending a ghost or spirit to scout, looking forward or backward in time, comprehending the way the fate of one thing is connected to the fate of another, and so on. The one that was hardest for me to manage, as ST, was always Sense Consciousness, since I needed to give the player a count of all thinking minds in the area with a spell that was nearly-always active. Good luck sneaking up on that mage!