Exercise 10 of the Playing With Purpose series
The Exercise
Take a problem you’re currently stuck on, something real, something you’ve been circling without resolution. Write it down as a question. The question you’ve been asking yourself about it.
Now look at that question and ask: is this the real question, or is this the first question?
Rewrite it five times, each time going one layer deeper. Not a different angle on the same surface, but a deeper version. What is the question behind the question? What assumption is baked into the original phrasing that you haven’t examined? What would you need to believe for that question to be the right one?
Then answer these:
The reflection questions:
- How different is the fifth version from the first?
- Which version feels more uncomfortable? That discomfort is usually pointing at the real question.
- Have you been working hard on an answer to a question nobody actually needed you to answer?
- What changes in your approach to the problem if you use the deeper question instead?
- Who in your life asks you questions that make you think differently? What do they do that you don’t?
What I Learned When I Did This
There is a version of Einstein’s thinking that gets passed around, probably not precisely accurate as a quote but true as a principle, which says something like: if you had an hour to solve a problem and your life depended on it, spend fifty-five minutes figuring out the right question and five minutes on the answer. I’ve thought about that idea for years, and the more I watch people work, including myself, the more I think it points at the most expensive mistake in creative and intellectual life, which is working very hard on the wrong question.
Here is how I first understood the depth of this problem, not in the abstract but through direct experience. I had been using AI tools for a while, doing what most people do with them, asking them things in the most natural way, the way you’d ask a quick question of a knowledgeable friend, and getting results that were broadly correct and deeply unsatisfying. And at some point I started paying attention to why the answers kept landing just short of useful, and what I found was that the problem was almost never the AI’s knowledge. The problem was my question.
The way large language models work is essentially the same as the way our educational system trained us to answer questions: take the input, find the most statistically likely correct response, deliver it as fast as possible. If the question is vague, the answer will be the average of every similar question ever asked, which means the answer will be competent and generic and built for no one in particular. The AI is doing exactly what we trained it to do. It’s doing exactly what we trained our children to do. The question is the constraint. The quality of the answer is bounded by the quality of the question.
I started writing longer prompts. Then much longer. A paragraph, then two paragraphs, then sometimes sixteen. My students watch me do this and they’re always a little startled, because the cultural expectation around AI is that you ask it something quick and it gives you something. But the more specific the question, the narrower the search space, and the narrower the search space, the closer the answer gets to something that’s actually built for the specific thing you’re trying to do. Instead of the average of a hundred billion interactions, you’re working with the average of a hundred million, then a million, then a thousand, and at some point you’re getting something that feels like it was made for your exact situation because it almost was.
What this practice revealed about my own thinking is uncomfortable and worth saying out loud: I had been asking first questions for most of my life and calling them real questions. The quick formulation, the obvious framing, the way the problem presents itself on the surface before you start digging, I had been accepting those as the actual problems and then working very hard to solve them. Some of the frustration I’ve felt over years of creative work, the sense of effort without corresponding progress, was the frustration of a person doing excellent work on the wrong problem.
Einstein, or whoever actually said it, was right. The question is the work. The answer is almost always downstream of getting the question right, and getting the question right takes most of the time. What I’d put in front of anyone doing this exercise is: the question that makes you uncomfortable when you write it, the one that feels a little too direct or a little too honest or a little too close to something you’d rather not look at, that one is probably the real question. The first five were practice.
Playing With Purpose is a series of creativity exercises for people who want to understand and reclaim their creative capacity. You don’t need to be an artist. You need to be curious.
Further Reading
- Stage 3 in Design Thinking: Ideate (IxDF) — Ideation methods and creative exercises
- What Is Creativity? (IxDF) — Foundations of creative thinking
- What is Gamestorming? (IxDF) — Game-based ideation and creativity
