Should we teach how to cope with uncertainty and incomplete data?

Scott Linneman, Geology, Western Washington University

For many years I have introduced plate tectonic concepts to students using the Discovering Plate Boundaries exercise that Dale Sawyer developed (and many of us have amended). Despite a fair amount of simplification, the exercise still requires students to deal with real geologic data sets with real ambiguities and a lot of uncertainty. More 'novice' than any other, my pre-service elementary teachers find themselves baffled when they encounter gaps in the ocean floor age map or when they figure out that we don't know about all the mid-ocean ridge eruptions. How does the novice-to-expert transition happen that allows us to not be paralyzed by incomplete or messy data?

Maybe this comfort develops as we discover the historical nature of science. The fossil record is notoriously incomplete and this idea should be an outcome of every historical geology course. Or perhaps there are critical field trips that we all experience. The first personal encounter with a major unconformity seems to be an epiphany for most geo-students. "You mean we don't know what happened between the times that the schist was metamorphosed and when it was uplifted and exposed at the surface and when the sandstone was deposited?" I suggest that a personal encounter may be required, because I've seen the most elegant and well-presented lectures on the unconformities in the Grand Canyon not accomplish this conceptual leap. Putting your hand across an unconformity seems to evoke an emotional reaction about what we don't know about the deep history of the planet.

Ours is not a forward experimentation kind of science. We cannot construct planet scale experiments or even small scale experiments that take millennia to run (unless you want to alter the climate or biodiversity of ecosystems). The size, complexity and age of the earth systems mean we must rely on modeling. And models demand simplification.

Supporting Attachments