The two cultures, revisited

I was having a lunch conversation today about an experience I had years ago, which I now realize touches upon what C.P. Snow referred to as “The Two Cultures”. Those cultures are, respectively, the cultures of scientific thought and of humanistic thought.

This particular experience dates from the very beginning of my career in computer graphics. I was part of an interdisciplinary team that was creating special effects for films and TV commercials. At some point Chris Wedge — a phenomenal animator, and later to become the founder of the great NY computer animation house Blue Sky Studios — asked me whether I could program a tool that would allow him to achieve a certain lighting effect.

All fired up by the task, and maybe too young and stupid to see beyond my own ambition, I stayed up all night and implemented a software tool that would allow Chris to do pretty much anything. It had all sorts of variables and parameters, complete with a cascading crescendo of calibrated components, creating a cornucopia of cool capabilities.

The next morning, flushed with pride, I showed Chris my creation. “Here,” I said, “you can use this to do all sorts of things. For example, this is the particular thing you wanted, if you just set these variables like so.”

He hated it. “I just wanted a tool that would let me do this“, he said, “I don’t care about all this other stuff.”

And that was the first time I got an inkling that there are two fundamentally different ways to look at the space of interesting problems. The ‘scientific’ approach looks for the most general solution, the one that will encompass as many answers as possible. THe ‘artistic’ approach doesn’t care about this vast space of all possible answers. Rather, it looks toward a particular human space of meaning, and is only interested in paths that lead to that space of meaning.

It’s not that one of these approaches is right and the other is wrong. They are both quite powerful, each in their own way. It’s more that these are two different languages — each better at approaching a different kind of truth.

When faced with the reality of these two different languages, maybe it’s best to be bilingual.

2 Responses to “The two cultures, revisited”

  1. Jonathan says:

    Hi Ken,

    One might argue that there are two types of people in the world: the ones who divide people into dichotomies and the ones who…

    That’s what makes the job of user interface or product design interesting- especially on 3d graphics programs. You’ve got one set of people who want to build things, and another group of people who want to tell stories with them.

    It reminds me of how my brother and I played with Legos as kids- I would always make things and he would whisk them away before I could dismantle them to return my creations to raw materials for my next idea. He would hoard the completed submarines and airplanes, castles and robots, so that he could tell stories with them. My outputs became his inputs for his creativity, which in hindsight is rather telling, considering that he’s finishing up an English degree, and I got an Industrial Design degree and then spent two and a half years selling and fixing 3d printers, laser cutters and other tech products for educational institutions.

    I see a similar dichotomy that has been more recently made a point of friction in society between two cultures- when people want to talk about “science vs. religion” it all gets quite silly, because one is the discipline of asking “how?” and the other is an attempt at explaining “why?”- building some kind of unifying narrative with all the messy bits to the universe, explaining our place in it and how we should go about being humans.

  2. Stephan Ahonen says:

    This is fairly prevalent in audio as well. Every digital audio workstation ships with a workhorse digital equalizer that offers more flexibility than anybody would have dreamed of even a few decades ago, yet a lot of folks nowadays will go for emulations of old hardware which feature far less overall control. Your workhorse stock digital eq offers controls for gain, bandwidth and center frequency, the three parameters which allow you to recreate the frequency curve of any second-order analog filter, but those really sought-after analog eqs offered just a gain knob, which had a complex relationship with frequency and bandwidth under the covers in a way that felt really natural when you turned the knob. A larger boost in gain would widen the bandwidth, or a cut would have a narrower bandwidth than a boost, or a shelving filter would move its center frequency downward as you applied more boost. Rather than searching through a three-dimensional space (gain/freq/bandwidth) for the sound you want, that old eq circuit only requires one dimension, how far the knob is turned, to get the sound you’re looking for.

    Of course, I tend to approach the craft from a very technical perspective, and tend to prefer the flexibility over simplicity. But I’m weird like that.

Leave a Reply