Rick Robinson

Dec 27th 2020

The Butterfly Effect Theory and the Limits of Predictability


In 1961, meteorologist Edward Lorenz, experimenting with computer simulations for weather prediction, made a slight tweak to one of the input variables for a simulation run. As MIT Technology Review recounts, he rounded off a value of 0.506127 to 0.506, restarted the program and stepped out to get a cup of coffee while the simulation ran.

When he got back and checked the results, he was startled to find that this tiny change to one input parameter had not just modified the resulting simulated weather, but completely transformed it. His discovery, later dubbed the butterfly effect theory after the nearly imperceptible results of a butterfly flapping its wings, would go on — in the best butterfly-effect manner — to completely transform scientists’ understanding of nature by giving rise to chaos theory.

So, what is the butterfly effect exactly?

What Is the Butterfly Effect?

In the 1960s, scientists’ conceptions of how nature worked were still deeply shaped by Sir Isaac Newton’s conception of an orderly, essentially predictable universe. As American Scientist reports, the philosopher Pierre-Simon Laplace had argued in the 18th century that if we knew the laws of nature and made sufficiently precise observations, “nothing would be uncertain and the future, as the past, would be present to [our] eyes.”

Lorenz was not initially concerned with the broad philosophical implications of his unexpected simulation result. For him, a little unpredictability in weather simulation was good news.

In previous tests, as MIT Technology Review explains, his simulated weather was too predictable, tending to repeat the same patterns over and over. Real weather, as we know, is full of surprises. To be useful, weather predictions needed to predict the unexpected rain showers, not just the ones obviously coming.

In fact, Lorenz’s discovery would build the foundation for modern computer weather simulations, which commonly employ an “ensemble” of simulation runs, with slightly varying input conditions, to test the range of likely weather possibilities.

Goodbye Seagulls, Hello Butterflies

Meanwhile, news of Lorenz’s work gradually spread from meteorologists to the broader scientific community. He had originally described what we now call the butterfly effect as “a seagull causing a storm.” In 1972, he gave a presentation to the American Association for the Advancement of Science, entitled “Predictability: Does the Flap of a Butterfly’s Wings in Brazil Set Off a Tornado in Texas?”

The change from seagulls to butterflies was suggested by another meteorologist, Philip Merilees. The butterfly effect theory now had a name that perhaps helped it capture the imaginations of both scientists and the public: What could seem more delicate and easy to miss than the flap of a butterfly’s wing?

Butterflies of Doom?

But there may have been more to butterflies than their image of delicacy. As Gizmodo notes, in 1952, nearly a decade before Lorenz’s surprising computer run, science fiction writer Ray Bradbury published a time travel story called “A Sound of Thunder.”

In the story, a traveler into the dinosaur-era past carelessly steps off a marked trail and crushes a butterfly. When the characters return to the present day, the world has been dreadfully altered by the cascading effects of that butterfly’s death. (No source, alas, explicitly tells us whether the theory’s name was inspired by Bradbury’s story.)

Butterflies of Uncertainty

Whatever the source of the name, the TV Tropes website explores how popular culture has taken up the butterfly effect theory with enthusiasm, though (not for the first or last time) the pop culture version of the theory is not always exactly what the scientists had in mind.

Thus, there is no way to use this information to trigger a vast result from a minute initial action. Instead, chaos theory, aka the butterfly effect theory, shows why we cannot fully predict the future.