Creativity as courage

Thomas Heatherwick, the famed and sometimes controversial architect, gave an interview in Wired years ago. I read it as a young influenceable designer, unsure what to make of an industry dominated by ego. He came across as excitable and unbothered by the cynicism of the real world. Deeply relatable to me at the time. The New Victorian masterbuilder, as he calls himself, devoting himself to just building great things. Entire neighbourhoods, iconic buses, Olympic cauldrons.

In this interview, he likened his creative process to that of a detective: asking questions over questions and eliminating ways in which a problem cannot be solved. In theory, this leaves you with the single correct answer.

What I failed to understand then, and am just beginning to understand now, is the personal steer in this. What sounded like inevitability is in fact applied individuality. Which questions to ask, how to weigh them, and what to eliminate. Common sense filtered through personal experience.

What I also know now: It’s much harder than that. It’s easy for the accomplished designer to talk about a deepdive creative process as if anybody can simply reproduce it. Theory swiftly put into practice, no friction attached. But it takes years of experience, making bad decisions and asking terrible questions to get to be close to getting it right. More importantly, it takes years of asking bad questions and listening poorly to understand what good questions are and how to properly listen.

There’s still a lot of truth in this. Not just in where to end up, but in how to get there.

Every time we find ourselves at a point that is not familiar, or easy, or emotionally accessible we have a choice. We can turn something down and deem it uninteresting / silly / irrelevant / plain. Or we can meet it with curiosity, and try to learn more. And more. And then some more. Everything we don’t know gives us a chance to test our current assumptions against it, and to find out where it hurts.

For me, the key is to ask questions like a child: Unwavering curiosity with zero sense of etiquette. Be cute enough to get away to be annoying.

This shouldn’t need to be said but the whole point of asking questions is to be curious for answers. There is a difference between taking things in and to learn, and letting soundwaves wash over you while your mind drifts off to some far distant land. It is ok to lose track, or to get distracted. It is also ok to ask again, or ask differently, as long as you are willing and ready to let the answers take you somewhere new.

The principle here is simple: You start with a lot of ideas, and each question eliminates more and more of them until you have whittled it down to a single one. The practice is harder: You need to have a set of assumptions and be open to being entirely wrong. Strong opinions loosely held.

So each elimination of ideas / directions / routes is a step forward toward making a unique choice. The uniqueness of the choice matters, since we are hoping that this process helps us make work that is different (and hopefully better) than what anybody with a different point of view could make.

The problem is this: Each choice is actually a series of choices wearing a trenchcoat.

The opportunity is not just in eliminating routes not to take, but in finding routes that seemed unreasonable before. And that’s what creative endeavours often come to (and what I believe creativity often comes down to). Connecting things that are relevant but not obvious. Pulling on the unexpected through a related thread. Finding a fresh view on something previously attempted. Pointing to where it hurts.

So this entire process is never as clear cut as Tommy H makes it out to be. It is a messy blend of rigour and inspiration, serendipity and discipline. But at the core is almost always a good sense of what to pursue and what to let go. Making a series of choices to build something new.

But what happens when we give up this power? We have now entered a time where we are no longer building tools or instruments to do something, but agents that can act as an extension of us, be it brains or hands. ChatGPT has taken over as our default surface that we use to interact with the online world. Surfing the web has become deep sea diving employing an autonomous underwater drone.

There is a lot to love about this and there is a lot to hate. This is not the point I want to drag this into. The part that bothers me is the willingness to hand over our ability to make a decision, and therefore to have a voice.

There are a few fundamental places that feel like we are stepping backwards in our cultural responsibility to become better at making.

In my experience, a lot of interactions with Chat-based AI interfaces go something like this:

1. Ask the model to be somebody who could help

2. Use your language to describe an outcome

3. Be presented with options

4. Choose the most palatable option, possibly debate a bit with your model of choice

5. Settle for something that feels good enough

Obviously, this is not how everybody uses it. And it is not how every model works. I am simplifying to make a point: We take the struggle out of a process that rewards struggle.

We replace the individuality of our perspective with a shortcut to something that feels good enough. And this at a time when we have the tools where we should thrive for hyper-fucking-exciting.

However, this is not an AI-only problem. I have seen this creep into the way people act across the process of making many times. We make variations over variations, and it can be a good idea to go broad. But instead of committing and moving forward, we push a few of them a bit, and never put a stake in the ground.

And I understand why: A stake in the ground can be seen across the battlefield. It makes you vulnerable and open to critique, and even criticism. But it also makes you real, and visible, and allows others to see what you stand for.

I set out to write this blog (are blogs back?) in the spirit of writing in a way that is helpful and optimistic. So let’s apply this lense.

Modern tooling as come a long way, and largely that is a good thing. It has been a long time since we have had such a rich opportunity to make.

Making any artefact has never been easier. We have the power and reach of one hundred hands holding one hundred tools. The potential impact and speed to make and present it to people is unparalleled. (Of course, has has also never been easier to make something that looks like a delightfully stuffed steak pasty just to turn out to be a mouth full of sawdust).

But also, it has never been easier to make more tools. Obviously, making better tools is not a new idea. The first piece of work a carpenter makes is the workbench. Artisans working with metal create moulds, jigs and machinery to improve their practice. And in our studio we have build various generative digital tools to help our clients produce brand expressions they couldn’t access otherwise. To do this, we have employed toolkits like p5.js – a tool to build tools. And I cannot wait to get ClaudeAI to help me build more tools.

Beneath this, there is a much wider opportunity though. “I saw the angel in the marble, and carved until I set him free”, Michelangelo is famed to say. Born 550 years ago, he worked and excelled across many disciplines, but might be most famous for his sculptures. Sculpture, a classic practice of moving forward, blow by blow, decision by decision. A steady process of commitment, choices hammered into stone.

It seems like we are forgetting what we knew all those centuries ago. It is not just about asking questions and eliminating paths, but the moment of deciding where to go. Not just traversing down a decision tree but adding branches and skipping stops where it feels right to us. Moving forward while choosing our own adventure. Each step filtered through our own experience, knowledge and individual point of view. To reveal what was already in the marble, and chisel away the superflous material.

At this time of technological excitement, we should be making the best work of our lives, all of us. Obviously, in this context “best” means “most uniquely you”. Because anybody can prompt a GPT to sound like a go-to-market expert, but nobody sees what you see just the way you see it.

— Lazers,
Michael