While chatting about walking the other day, as we do, Fiona mentioned something she'd read in Shane O'Mara's excellent book In Praise Of Walking. Humans are, apparently, incapable of walking in straight lines when blindfolded or otherwise prevented from seeing landmarks.
Participants were asked to walk either in a large and dense forest or in the Sahara desert. Their task was simple: to walk in a straight line for a minimum set period, usually a few hours. Some walked in the day, others at night. All wore GPS tracking devices. While walking without reliable visual cues in the fog, or with heavy cloud-cover, the subjects regularly veered left or right, and eventually crossed the path they had been on. In clear daylight, they sometimes veered from a straight path but neither systematically walked in circles nor repeatedly crossed their own path. The result was the same in moonlight.O'Mara, p84
Here's another account of the phenomena, from Robert Krulwich and NPR.
The idea that walkers might drift a bit is fine, but that you might go in actual circles seems ridiculous. It's evidently true though, so why this reaction?
Did our ancestors find this phenomena weird? Did nomadic tribes just abandon the very notion of going for a walk in the fog? Does this happen with people who have lived for generations in deserts and forests or just with interlopers?
Do we find it weird because in an urban environment there are always visual cues to guide us? We might not know exactly where we are but we know what direction we're going in, most of the time.
Maybe straight walking is an anomaly, a relatively recent invention of the Romans? In Watling Street, John Higgs talks about how, before the Romans, British roads meandered and weaved around the physicality of the landscape. Maybe we didn't need the ability to navigate because the land would get us there eventually. It was only when we imposed our lines and made the land a problem to be solved that we started getting lost without them.
I don't know, and that makes it really interesting.