We resolve actions in RPGs by making checks, right? I’m not actually a master swordsman, but I can use my attack bonus to slay dragons. And I’m not actually a master thief, but I can use my Pick Locks skill to open a door. So even though I’m not as smart as my wizard with Intelligence 20, I should be able to make an Intelligence check to solve a riddle, right?
But, if so, why does that feel so unsatisfying?
Broadly speaking, it’s for the same reason that we don’t “solve” crosswords where the answers have already been penciled in.
We can also think of this in terms of the Czege Principle:
When one person is the author of both the character’s adversity and its resolution, play isn’t fun.
In the case of a riddle or puzzle, the resolution is, of course, figuring out the answer. If the interaction at the table is:
GM: Like the sun in a snowstorm, I must be broken before I can be used. What am I? Give me an Intelligence check.
Player: 18.
GM: The answer is, “An egg.”
The player has been excluded from participating. (And this largely remains true even if we muddy up the middle step a bit by, for example, requiring the player to say, “I’ll make an Intelligence check.”)
In The Art of Rulings, I propose three thresholds for making a ruling:
- Passive observation is automatically triggered.
- Player expertise activates character expertise.
- Player expertise can trump character expertise.
If you apply this metric to riddle-solving, you generally end up in a similar place: Player expertise activating character expertise means that “the characters don’t play themselves.” The players have to make some meaningful input in order to activate their character’s expertise (e.g. deciding to search a chest for traps in order to activate their character’s mechanical Search check), and in the case of a riddle or puzzle the only meaningful input is figuring out the solution. (Which, of course, obviates the need for the check.)
To put this a different way: The only meaningful part of solving a riddle is the LAST step. So if you reduce the solution to a mechanical check, you have taken all meaning away from the player.
Engage the players through their characters. If you’re ONLY engaging the characters, then the players are no longer playing the game.
BUT I WANT TO CHECK!
But let’s say that you (or your players) WANT to make the Intelligence check. This is generally due to one of two reasons:
First, the PCs are stuck and they need a solution to the riddle or puzzle in order to proceed.
Second, the players wants to play a character who is smarter than they are. Just like some players want to play a character who can win a heavyweight title bout (even though they absolutely cannot do that in real life), you’ll have players who want to solve riddles and puzzles that would be impossible for them in real life.
Fortunately, there are some techniques you can use without making riddles and puzzles meaningless.
NON-ESSENTIAL RIDDLES
The first thing you can do is make the riddle non-essential.
For example, consider the riddle of Moria’s door in The Fellowship of the Rings.
‘What does the writing say?’ asked Frodo, who was trying to decipher the inscription on the arch. ‘I thought I knew the elf-letters but I cannot read these.’
`The words are in the elven-tongue of the West of Middle-earth in the Elder Days,’ answered Gandalf. ‘But they do not say anything of importance to us. They say only: The Doors of Durin, Lord of Moria. Speak, friend, and enter. And underneath small and faint is written: I, Narvi, made them. Celebrimbor of Hollin drew these signs.’
`What does it mean by speak, friend, and enter?’ asked Merry.
‘That is plain enough,’ said Gimli. `If you are a friend, speak the password, and the doors will open, and you can enter.’
‘Yes,’ said Gandalf, ‘these doors are probably governed by words. Some dwarf-gates will open only at special times, or for particular persons; and some have locks and keys that are still needed when all necessary times and words are known. These doors have no key. In the days of Durin they were not secret. They usually stood open and doorwards sat here. But if they were shut, any who knew the opening word could speak it and pass in. At least so it is recorded, is it not, Gimli?’
‘It is,’ said the dwarf. `But what the word was is not remembered. Narvi and his craft and all his kindred have vanished from the earth.’
‘But do not you know the word, Gandalf?’ asked Boromir in surprise.
`No!’ said the wizard.
In order to open the door, they have to figure out the password! There’s only one solution!
… but that’s not the only way to open the door, is it? Particularly if it was a D&D group:
- They could break it down.
- They could trick the lake monster into breaking it for them.
- They could prepare and cast a knock spell.
It’s also not the only way into Moria. They could, for example, try to climb the mountain and enter through one of the windows or ventilation shafts.
Plus, they technically don’t need to get into Moria at all:
- They could go back and try to cross Caradhras again.
- They could go south through the Gap of Rohan.
- They could abandon their overland journey entirely, retreat to a western port, and sail to Gondor.
This is similar to the Three Clue Rule: If there are multiple paths to the goal, then a puzzle the players can’t figure out rendering one of them inaccessible is not a critical problem. So if the only reason you were making the check was because you felt compelled to force an answer on the players, making sure that the riddle or puzzle isn’t a single point of failure for the scenario (and being open to player suggestions for how they might route around it) sidesteps the problem.
ROLLING FOR CLUES
Speaking of the Three Clue Rule, let’s put a spin on our earlier example of unsatisfying play and consider a different type of puzzle:
GM: Lord Arthur D’armount has been murdered! Give me an Intelligence check.
Player: 18
GM: Bob did it.
That’s clearly absurd, right?
But nevertheless, a murder investigation scenario will almost always feature players using their character’s skills to search for clues and identify the ash as coming from a Trichinopoly cigar in a moment of Holmesian brilliance. Why does that work?
The difference is that these checks (or other mechanics) are delivering clues. It’s still the players who use those clues and take the rewarding final step of figuring out what they mean.
And when someone playing a super-genius character like Sherlock Holmes or Reed Richards or Wile E. Coyote wants to make an IQ check to solve a riddle, we can do the same thing: Instead of giving them the solution, we give them a clue.
When we’re talking about a murder mystery, this distinction between clue and conclusion can feel fairly obvious. If we’re talking about Myst-like puzzles or Gollum-style riddle battles, on the other hand, it can be a little harder to figure out clues that aren’t just the solution.
This is often more art than science and will be heavily dependent on the specific riddle or puzzle, but a trick I frequently find useful is to break the riddle or puzzle apart conceptually and then give clues that only make one part of the riddle or puzzle more explicit. For example, let’s look at the simple riddle we used at the beginning of this essay:
Like the sun in a snowstorm, I must be broken before I can be used.
We could look at the first part of the riddle (“like a sun in a snowstorm”) and say, “You’re pretty sure the ‘sun’ and ‘snowstorm’ are referring to colors.”
Something else we can learn from mysteries is that you can also deliver clues to riddles or puzzles diegetically. Like Henry Jones, Sr.’s journal in Indiana Jones and the Last Crusade, the PCs can piece together lore and rumor, and perhaps investigate the area around the puzzle, for hints to the puzzle’s solution. If there are a bunch of large stone pillars, each etched with a strange rune, for example:
- Researching the runes in a library might be useful in identifying which runes are related to each other.
- Searching around the pillars might discover scrape marks on the floor, indicating that they’ve been moved around.
- A giant-slayer’s journal might describe the relevant rules of Brobdingnagian chess.
And some of these, of course, might also be things that a successful Arcana or Giant Lore check would recall.
There’s also making rolls concerning the context in which the riddle appears. It could reveal things such as a History check to know that during the time the door with the password was made, people were trusting enough that the password would probably not be very secure. Or an Arcana check to reveal that from all the character knows about sphinxes, they generally don’t eat groups of people for huddling up and quietly discussing answer ideas amongst themselves before actually answering.
Thank you for this one – very interesting. I’ve always felt riddles and puzzles should be part of fantasy RPGs because they are part of the source material, but struggle with the dissociated mechanics that snap the players out of the game whilst they figure it out (or not, which happens far too often). Absolutely, I agree that puzzles shouldn’t be a block to continuing, and there should be a way around – climbing over the obstacle or breaking it down, for example, instead of working out the puzzle built into the lock.
When it comes to clues, however…
I regularly tell and solve “lateral thinking” puzzles. These are basically Twenty Questions, only about a situation, not an object. And the questions, appropriately asked by players, provide clues.
The classic is, “A man walks into a bar and asks for water. The bartender points a gun at him instead. The man thanks him, and leaves. Explain.”
This I can tell in an RPG session by your description of how riddles should work, because the solution comes after players ask yes-and-no questions to narrow down the solutions. If they get stumped, but pass an Int check, the DM can suggest, “The man isn’t thirsty. He got what he wanted. The bartender was trying to help him.” and so on. As long as I don’t just give them everything, the end solution (“The man had hiccups.”) can still be satisfying to work out.
A more tactile sort of puzzle I’ve used is the matchstick puzzle, where you take matchsticks (or in my case, pencils) and must arrange them into a given shape in a certain number of moves. The clue can be in the form of demonstrating one of several moves, or stating, “This is one of the ones you have to move.”
And a third is actually giving them a physical puzzle to assemble. (such as a message in an obvious code or a torn-up letter.) The problem there is that often one person will assemble it while the others ignore them and ask questions.
While I understand the desire for someone playing a clever (High Int) character to want to have an advantage in these kind of situations, one of the lessons we should take from the LOTR reference above is that it was a hobbit (I can’t quite remember if it was Merry or Pippin) who stumbled upon the solution, not the age old wizard or the culture savvy dwarf. Herein lies potential for humour and surprise and maybe a break from stereotype that is not only fun but actually as realistic as the “smart” person always coming up with the answers.
One thing that I would suggest on the idea of giving players clues is that you still might want to let them resolve the riddle with a check. Maybe they just can’t figure it out, that’s fine, their Investigator can. But we can still make the gathering of clues the relevant mechanic, similar to how we would tactically “gather” enemy HP to resolve combat.
So say we start with a resolution that’s nearly impossible to resolve – they’d need a 40 and they only have a +10 on a d20 system. They can’t make checks endlessly, they have a limited number of guesses so each attempt has a cost. And that’s the other thing missing here – cost. In combat, every decision has a cost. If I swing my sword, I can’t fire my bow. If I cast a spell, I can’t run away (or at least, I run away more slowly). So these decisions must have a cost.
If I’m researching the runes on the pillar, maybe that takes four hours – letting the BBEG get further ahead of me. If I’m searching around the pillars, maybe there’s another trap I accidentally spring. Maybe I contact my old professor to research one of the runes, and he charges me by the hour. But each one not only gives me, the player, a clue towards the resolution of the riddle but it also gives me a +5 on the check. If I can find the scrap of paper from the last dungeon delver (“sun in a snowstorm – yellow on white?”) and research three of the runes (“bird, birth, and cooking… wut?”) each for an additional +5, now I only need a 10 to solve it mechanically. That still feels satisfying, I spent eight hours, stepped in a beartrap, and took out another student loan to solve that riddle! But just because I’m not good at puzzles doesn’t mean my character can’t be.
Matryoshka riddles!