Rules for Creativity

So far, this has all been preliminaries. From this post, I want to start working out how to make creativity part of a game. It makes sense to start with what I think is the hardest problem: rules for coming up with ideas. If I can solve this problem, the rest should be relatively easy.

First, I think I do need rules for this. It’s supposed to be central to the game, so the mechanics should reflect that. One fundamental reason why they should reflect it is that it should be possible to play characters who are more creative than you are. Obviously, most role-players are so creative that this will rarely be necessary, but it’s good design to take account of the extreme cases. Another reason is that it’s quite likely that the game will be fantasy, so the player just won’t have enough background knowledge to come up with ideas that work in the context of the game setting.

The simplest rule would be roll something, and have the quality of the idea measured by how high the result is. However, I don’t think that’s a good mechanic, because there are no interesting choices for the player or character to make.

We can start to make it interesting by letting the player decide what to roll, but, by itself, that isn’t much help. Players should always choose to roll whatever they have the best rating in. There might be some interesting cases where the player has a large pool but a smaller number to keep for one option, and a smaller pool keeping a larger number for another, but even then the dice mechanics are such that keeping more dice is almost always better. In any case, given that players, quite sensibly, often put some effort into optimising their characters, there will be too many cases where one option has the largest pool and the largest number of dice to keep. (I need a term for “the number of dice to keep”, as well. Maybe one will come to me as I go along.)

At this point, it’s worth remembering that coming up with the idea is not the end of the process; the characters will then have to make the idea real, whatever that involves. The created idea should have game statistics that feed into that process.

So, as the first step, I’m going to give an idea multiple statistics. This also helps with the mechanics for coming up with the idea, because the different statistics can be in tension with each other. Methods and approaches that help one might harm another, so you might get bonuses and penalties to rolls depending on your choices.

The nature of the statistics will vary depending on the sort of idea you’re trying to create. For example, for an artistic concept “originality” is an important statistic, but it’s completely irrelevant to a scientific theory. On the other hand, the idea behind a scientific theory needs to be fruitful (I’ve done far too much history and philosophy of science to say “true”), but that doesn’t matter for art.

There is one obvious statistic that is generally applicable: the ease with which an idea can be realised. That even has an obvious application to the next stage of the game process, and is a good place to have the players and characters make trade-offs. They can go for a better idea that’s harder to implement, or one that’s easier to implement but not so good. I do this in real life all the time, frequently going for the easier idea because there’s a deadline looming, so it’s a decision that can be made in character.

One possibility would be to have two statistics, for quality and ease, but I think that’s still a bit too simple for something that’s supposed to be central to the game. I’d like to have at least two quality statistics, so that direct comparison of two ideas is not always easy. For art, originality would be one, and the other could be emotional impact. These are naturally in tension, because artworks that draw on lots of standard images tend to have more emotional impact than those that have to create audience involvement from nothing. For science, they might be fruitfulness and elegance. A rather complicated theory might be very fruitful (for example, the standard model of particle physics, or most theory in molecular biology), while an elegant theory often has limits in how much it can actually predict (string theory, or neutral theory in ecology, might be examples).

We might also go for two ease statistics, but that will depend on what happens with the mechanics for realising the ideas. For now, I’ll assume that there’s only one.

This creates a pattern of two quality statistics and one (or possibly two) ease statistics. I think three or four statistics is enough for an idea; they are central to the game, but, at least at the moment, I think they are individual challenges, not continuing characters. Once this pattern is set, however, all types of idea have to follow it. The two quality statistics will differ depending on what kind of idea it is, and the ease statistics might, but the structure should be the same in every case. This is because using different mechanical structures makes a game hard to remember and use, particularly if you are using different structures for essentially the same task.

The astute reader may have noticed that we still don’t have mechanics for creating the ideas. We do, however, now have something for the mechanics to aim at. However, before the mechanics can be created, something else is necessary. There needs to be a reason why the players can’t just keep rolling until they get the result they want. Most role-playing games focus on combat, where this reason is obvious: your opponent is trying to kill you. However, it’s much less clear why you can’t just keep trying to come up with an idea.

One limit is time. It should take a certain amount of game time to make each roll, probably measured in days, and if there’s a time limit, that will effectively restrict the number of rolls allowed. However, I don’t want characters in this game to be restricted to freelancers with deadlines, so I need some other restrictions.

The other one that occurs to me is what is referred to as path dependence. The idea is that the result you get depends on the route you take to get there, and that taking actions in a different order would not get you to the same place. Thus, every decision the characters take effectively locks them into a particular path, for this idea, and they can’t backtrack.

There’s an aspect of psychology that is helpful here. Once you have started looking at a problem in a certain way, it’s difficult to start looking at it in a completely different way. Even if you start over from the beginning, your ideas are still influenced by the way you tackled things last time. Of course, some people are better at avoiding this influence than others, so we can have game statistics that determine how good you are at starting over. Still, it serves to make path dependence plausible.

Thanks to path dependence, players cannot simply repeat an action if they get a bad roll, even if they have unlimited time. Thus, bad rolls have an impact, and the decisions the players make about strategy have an impact.

So, the mechanics for coming up with ideas should be path dependent, take time, and aim at (at least) three statistics that describe the final idea. The next post will look at what those mechanics should actually be.