I have a colleague who collects maps of Africa that demonstrate a specific phenomenon: the developed world's unlearning of African geography. Across the centuries, the maps seem to show that mapmakers know less about the geography of the African continent—particularly the internal parts—than previously was the case. Rivers change direction; mountain ranges disappear. This unlearning, my colleague argues, comes from notions about the acceptability of sources of information previously used. These maps show the social nature of “learning,” the idea that while in many cases there may be actual answers (after all, African geography exists), what information you look for, and from whom, determines how you will view the information you get, and ultimately what you will do with it.