Reflections after analyzing data from a failed experiment

The feeling is sickening. Your stomach sinks immediately, plunging thickly into darkness. Then, a sense of vertigo, a chilling Hitchcockian background-moving-but-character-standing-still effect where your eyes are forced to continually adjust to your computer screen.

You’re looking at nothing. Well, not nothing, exactly. Just not what you expect, which feels like nothing. Feels worse than nothing, because this so-called “something”, this shadow-something, is a hollow, almost mocking facsimile of a something.

You think: What went wrong? You think: Did I make some catastrophic mistake? You think: You shouldn’t have allowed yourself to hope.

In science, hope is a dangerous thing. Hope is not objective. Hope is not convinced by negative data or swayed by repeated failure. Hope endures; it perseveres. “Hope is a thing with feathers,” wrote Emily Dickinson, “that…sings the tune without the words, and never stops at all.” Well, sorry, Emily, but in graduate school, your feathered hope is an endangered species, and science is slowly encroaching, licking its lips.

You’ll try again, of course. Repeat the experiment, perhaps. Or try something different. It doesn’t matter. You’ll move on. The only way to succeed is to maintain forward momentum.

You close your eyes, breathe in, breathe out. Science is horrible, you think. It’s grueling, demanding, indifferent to struggle and emotion. You remember that every medical professional program at OHSU has a timeline for graduation — four years for medical school, four years for dental school, etc. — except for one. In graduate school, your graduation time is contingent on a complex, unpredictable amalgamation of skill and luck. Some people graduate early, some people graduate late. Science is capricious like that.

I’ve always likened science to art. Like artists, we go through a long mentoring program where we learn unique styles and approaches. If we are successful, we both get paid by a major benefactor to practice what we’ve learned. We both engage in highly creative work. There is a cult of personality built around the titans of our fields. To some extent, we both strongly admire those who maintain their integrity and eschew corporate influence. We both are constantly evaluated by our peers, taking inspiration from the works of others to improve our own. We both are only as good as our latest creation.

And yet they’re fundamentally different. In art, truth is subjective and individualistic; in science, truth is repeatable and verifiable. In art, your work is uniquely yours, bearing your own fingerprint; in science, your work can be scooped, simply repackaged and published by someone else.

Here’s the most devastating difference: Art, at its highest ideal, is creative socialism, giving everyone an opportunity for their ideas to flourish and succeed. However, science, at its highest ideal, is creative capitalism, allowing only the flourishing of work with the largest weight of evidence and the highest judgment of your peers and the greatest success of your experiments.

Ultimately, that’s good. We need science to be judgmental. Science wouldn’t be science without it. But remember: This characteristic is good mainly for the beneficiaries of science, not the practitioners. We are the ones stuck dealing with the negative data, the ones struggling for months to extricate a small, floating signal from an ocean of noise.

In these moments, we need to remember that science isn’t inevitable. It’s not a foregone conclusion. Rather, science is a constant, thankless war, and we are its soldiers. We are fighting for the truth. We are fighting against the natural forces of entropy. We are fighting against hundreds of possible confounding variables, too many to consider. And we are fighting even against ourselves, against our own reptilian, pattern-obsessed brains hungry for any explanation.

Even though I know I’m extremely privileged—privileged to be a graduate student, privileged to experience this middle-class struggle, privileged even to do scientific research in the first place—damn it, sometimes war is hell.

And I’m done fighting today. May tomorrow be better.

One response to “Reflections after analyzing data from a failed experiment

  1. David, you are extremely adept at weaving your words and descriptions into a compelling story — thank you for a peek into the world of a grad school scientist!

Comments are closed.