Commentary / World

Why scientists solve harder problems first

by Faye Flam

Bloomberg

It’s a paradox of science: How is it that researchers keep achieving the impossible while leaving seemingly simple tasks incomplete? As astronomer Martin Rees put it in a recent essay in The Atlantic, scientists can detect two black holes colliding a billion light-years away, and yet they’ve learned very little about how to treat the common cold.

Rees offers that it’s partly a matter of scale: Phenomena on the astronomical scale of stars and the tiny scale of atoms unfold in predictable ways, while the complexity of the world in between keeps us guessing as to what will happen next. This has historically been the case. Long before people understood that germs existed, they could predict the motions of the stars and planets and use them for navigation.

But that’s just part of the story. Common misunderstandings can also distort our ability to differentiate the easy tasks from difficult or impossible ones.

In many cases, people underestimate the difficulty of controlling our environment. Scientists observed the black hole collision through ripples in space that followed. They didn’t control the event. And while the common cold is a mundane phenomenon, and the cold viruses are well studied, controlling them is far from trivial.

It’s hard to control much of anything in this world. While archaeologists argue over how many hundreds of thousands of years ago our ancestors “tamed” or “controlled” fire, we read that wildfires are “out of control” in California. Sometimes fire still shows us who is boss.

I posed the colds-versus-black-hole puzzle to Edward Tenner, a historian at Princeton University, whose upcoming book is titled “The Efficiency Paradox: What Big Data Can’t Do.”

He pointed out that people often underestimate the brilliance of nature — and wrongly assume that common natural phenomena can be easily duplicated with technology.

In artificial intelligence research, for example, it’s been possible to create computers with superhuman abilities in chess and other games, such as go, but very hard to reproduce something little children do naturally — deal with the ambiguity of language.

While computers are getting smarter all the time, he said one way he thinks we could still distinguish them from humans is to ask them to interpret aphorisms. Take “A rolling stone gathers no moss.” As a child, he said, he wondered whether moss was good, maybe a metaphor for money. Or was moss bad — an encumbrance that made it better to keep rolling along? That kind of ambiguity makes it possible for little kids to appreciate humor and poetry. Robots might be able to produce something like poetry through trial and error, but they may never come to appreciate it the way we do.

This tendency to underestimate the creativity and power of the natural world may help explain why people think colds should be easy to wipe out. If something as mindless as the immune system can kill colds, how could it be out of reach of technology? But that line of thinking doesn’t give proper credit to the intricacy of the immune system.

There’s another factor in the black-hole-versus-common-cold comparison, as I learned from Joshua Plotkin, who studies viruses as a professor of computational biology at the University of Pennsylvania.

For him, the difference is that the black hole detection was a project with a well-defined path, while there’s no obvious way to approach curing the common cold. People knew that colliding black holes and certain other energy-producing phenomena would release a specific kind of wave they could pick up, if they could build a sensitive enough detector. And physicists could measure their progress over the years as their detectors got more and more sensitive.

As Plotkin puts it, the black holes presented a rational problem.

In contrast, some of the greatest advances in biomedical research happened through tinkering and serendipity. That’s how we got antibiotics. Or to give a more recent example, the powerful technology known as CRISPR, or gene editing, started with people studying how yogurt-making bacteria protect themselves from bacteria-invading viruses.

What does this mean for the future? Rational projects with clear paths forward often succeed in achieving the loftiest of goals, whether it’s reading all 3 billion chemical code letters in a human genome, detecting the Higgs Boson, or getting close-up pictures of Pluto.

But part of value and the fun of science is in the unpredictable part. Not every open-ended, exploratory project will turn up antibiotics, or gene editing, but having lots of such projects going guarantees that 2018 will bring at least a few surprises.

Faye Flam is a Bloomberg View columnist. She has written for The Economist, The New York Times, The Washington Post, Psychology Today, Science, New Scientist and other publications. She has a degree in geophysics from the California Institute of Technology and has been a Knight-Wallace fellow at the University of Michigan.