Take A Step Back

Feb 03, 2017

BY JIM SAWYER – CHIEF SCIENTIST

“I don’t want to paint things like Picasso’s women and Matisse’s odalisques lying on couches with pillows. I don’t want to paint people. I want to paint something I have never seen before. I don’t want to make what I’m looking at. I want the fragments.”
– Ellsworth Kelly

Perhaps my favorite contemporary artist of all time, Ellsworth Kelly, passed away late in 2015. Kelly is best known for vivid, minimalist, large-scale paintings that fill the walls of art galleries with intense, provocative swaths of color. A bit of a rebellious thinker, even among the giants in contemporary art, Kelly’s work calls attention to the relationship between the piece itself and the perception of the person viewing it.

Specifically, through the conjunction of powerfully striking colors not found in nature, Kelly implicitly invites you to focus on your own perceptions of the objects being perceived. He also calls on you to question the form of a “painting” itself by presenting the canvas as hard-edged, non-traditional geometric shapes like triangles, half-circles, or rhomboids. It’s intentional, thought provoking, and to my eye, beautiful.

Some of Kelly’s work is on permanent display at my local art museum, the extraordinary High Museum of Art here in Atlanta.[1]If you’re at all into that art stuff, you really should come to Atlanta and check it out sometime. I’ll buy you a glass of wine or two. And I’ve observed that far too often, unknowing visitors, walking right next to the piece, perhaps pausing ever so briefly, simply give it a passing glance and shrug their shoulders. “Oh, those are really bright colors.” “That’s just two squares with nothing on them!” “I don’t get it.”

And these well-intentioned folks are likely bona fide modern art appreciators, or they wouldn’t even be there in the first place. They’re looking for Warhol’s rainbow-colored “Marilyn” silkscreens or Rauschenberg’s incorporation of commercial imagery within abstract expressionist paintings or Oldenburg’s grossly oversized still life as physically imposing objet d’art[2]Yeah, I just went all French on you. That was for my Canadian friends. or a dozen other things.

The thing is, they were viewing Kelly’s piece from just a couple of feet away—the way we often do as we explore museums. We pause in front of the painting, lean in, and scrutinize the details in order to appreciate them. But with Kelly’s work, we can’t see it all up close. It does us no good to stand right next to it. We simply can’t perceive the meaning, the impact, the beauty—what Kelly wants us to reflect upon—at that narrow distance.

We need to step back. Way back.

“I’m interested in the space between the viewer and the surface of the painting – the forms and the way they work in their surroundings. I’m interested in how they react to a room.”
– Ellsworth Kelly

At a distance—from across the room—your eyes can start to perceive the totality of Kelly’s vision. Is it a painting? Is it a sculpture? Have I ever seen that exact color before? Have I ever seen that color juxtaposed with those other colors in precisely that way? If I turn off any preconceived associations with meaning, and focus on what I am perceiving, can I begin to tap into my relationship with the space between me and the painting itself? Can I start to gain a comprehensive appreciation of what Kelly’s art is hoping to evoke?

Sometimes it’s the same thing in data science. We get caught up in the minutiae, by necessity. It’s our job to ensure that every single nugget of data in every data source we are working with is well-defined, clean, correct, complete, explainable, and most of all, understood. It’s our job to test every assumption about the nature of that data, to ensure that the modeling approaches we use are appropriate for the data we have. It’s our job to produce a comprehensive volume of high-quality, reproducible analysis that provides thorough, defensible, quantitative answers to the problem we have chosen (or were asked) to study. It’s our job to closely and exhaustively scrutinize the details of our data in order to fully solve the problem. We lean in.

But what is the story that is emerging after weeks and weeks of heads-down, detail-oriented scientific exploration? What is the beautiful evidence[3]As the incomparable Edward Tufte teaches us. You’ve heard of him, yes? we’ve discovered now that we’ve unlocked the secrets hidden within the data itself? What is the vision that this rich, meaningful set of data—the overall totality of the data we have, together, in front of us, brilliantly exposed—unlocks in us now that we’ve understood and revealed it? What do we want our clients to perceive—to appreciate—about it and about us as a result?

Sometimes, we need to step back. Way back.

Footnotes   [ + ]