For Joel Kingsolver, the music of chance was the buzz of a hungry yellow jacket.
Kingsolver, a zoologist at the University of Washington, and his assistant, Kristina Williams, released a bunch of cabbage white caterpillars this summer on a collard patch at the university's Urban Horticulture Center to study their survival. It was part of research into the influence of different environments and temperatures on caterpillars.
Just two days after their release, however, most of the caterpillars in the collard patch had disappeared.
So what did the real world do to these caterpillars? After a little observation, the answer came buzzing by.
"What we found was that western yellow jackets were swooping in and carrying off all of our caterpillars, mostly within a few hours," Kingsolver said.
"They'd come in and locate a caterpillar, jump on top of it, bite it and try to carry it away. If it was too big, they would bite it in half and carry half away."
After some library research, Williams found that there wasn't a lot of information about yellow jackets and their nasty relationship with caterpillars -- and thus there was the potential for new research. So what could have been a disaster turned instead into a new branch of the caterpillar experiment, about the feeding habits of the yellow jackets and their possible use as pest control for caterpillars.
"They sure were effective at taking our caterpillars out of our collard garden," Kingsolver said.
Such can be the way of science, according to scientists and science historians. Despite every effort to reduce the role of chance in an experiment, it always sneaks in like a bug. And once in a long while, as in Kingsolver's case, that element of chance can lead to discoveries, some of them brilliant.
"There's a lot more creativity and chance in science than is normally credited," said Keith Benson, a professor of medical history at the University of Washington and executive secretary of the History of Science Society.
"Hunches and serendipity play as large or a greater role than a design direction."
The history of science is rife with instances of accidental discoveries. Modern chemistry originated during the medieval pursuit of alchemy. Alchemists worked to turn lead into gold. Instead, they discovered elements, relationships between compounds, and many of the basic principles of chemistry. Lead, to this day, remains lead.
As Royston Roberts details in his book "Serendipity: Accidental Discoveries in Science," many of the modern world's greatest discoveries came from chance inspiration. For example:
Penicillin was discovered in 1928 when a mold accidentally sneaked into a petri dish from an unrelated bacteria experiment. Alexander Fleming, the scientist doing the experiment, noticed that the mold had kept the other bacteria from growing, and he decided to isolate and study the mold.
Nylon fibers were discovered in 1935 when scientists at DuPont noticed that they could make strong fibers by drawing a stirrer from a container full of a formerly useless experimental compound. The discovery led to strong fishing lines, silk substitutes, guitar strings, panty hose and countless other applications.
Modern rubber was invented in 1839 when Charles Goodyear accidentally dropped a piece of natural latex mixed with sulfur onto a hot stove. The cooked rubber was strong, flexible and could handle changes in temperature, unlike other treated latex. The process became known as vulcanization.
Bruce Hevly, a science historian at the University of Washington, said much of modern science revolves around the principle of giving the scientist the freedom to take chances, to pursue research for its own sake. This opens the door to serendipity.
The end of the last century, for example, yielded a whole batch of inventions as scientists began to tinker with new technology and machinery that in the past had been dominated by craftsmen. The craftsmen had focused on creating a product; the scientists were focused on interesting knowledge. The result was the second industrial revolution with its electric engines, new chemicals and other important discoveries.
Hevly said the culture of allowing scientific freedom to experiment expanded after World War II, when governments realized the value of funding untargeted research as a way to produce new technologies. Even inside the Soviet Union, scientists were allowed more freedoms than others because politicians realized those freedoms were needed to encourage creative research.
Few scientists see serendipity as something mystical or supernatural, Hevly said. Rather, they see it as the product of a trained mind based on clean, solid research.
Hevly also argues that many accidental discoveries are really intentional, and that scientists only see them as accidents afterward.
"When people point to chance or to accidents, it's a way to retrospectively recognize that one's ideas have changed substantially," Hevly said. "After the fact, it's hard to think that you didn't understand something."
Others, like Geoff Loftus, a professor of psychology at the University of Washington, say that serendipity is nice when it happens, but most science doesn't revolve around it.
"It's quite unusual that some unexpected circumstance would lead to some interesting, valuable consequence, Loftus said. "Chance is something that you usually try to avoid."
Loftus said much of the scientific method is about reducing the bad kind of chance, the "statistical noise" that can ruin an experiment. That's because the purpose of an experiment is to find whether something causes something else to happen. Error and variations outside the controls of the experiment will skew results and make it difficult to determine a cause.
"Random events don't usually provide interesting consequences," Loftus said. Still, he said, scientists should keep their eyes open when conducting research for the rare moment of genius that comes from chance.