Who thinks the Earth is flat?

This post was first published in 2006 on the Alphapsy blog.

Less people than you think. Most people have the representation of Columbus valiantly fighting against the authorities and finally convincing these obscurantist scholars coming right from the middle-ages (actually it was the middle-ages) that the Earth is round and not flat. It turns out that this is not true at all: medieval scholars have always known that the Earth is round.

Children then? Surely children fall for the flat Earth? Studies by Stella Vosniadou and her colleagues in the 90’s were conforting this idea: in some experiments, children tended to draw flat Earths. However, these results have also been disputed: using a simpler methodology, Gavin Nobes and his colleagues have shown that children prefer round Earths.

The Flammarion woodcut



I'm a bit ashamed to admit that until a few days ago, I thought that the flat Earth was a common misconception in the middle-ages. Happily, I was corrected when I read Edward Grant's God and Reason in the Middle-Ages. After making a very interesting case regarding the very important role played by reason in the middle-ages (I found a lot of common points between scholastic and modern formal logic and formal semantics – both in the strengths and in the weaknesses), Grant tries to explain why misconceptions about medieval scholars are so widespread. He spends some time explaining that they never believed that the Earth is flat: Aristotle (their main source when it comes to natural philosophy) had made a very convincing argument regarding the roundness of the Earth, and they followed him on this point.



It seems that the idea that Columbus had to fight to establish that the Earth is not flat has been set up in the XIXth century in a book named The Life and Voyages of Christopher Columbus. The history of the scam is nicely described in Jeffrey Russell's Inventing the Flat Earth. Columbus and Modern Historians. I won't dwell on that here since wikipedia does such a nice job describing the history of flat Earth theories.

What of the children then? People have been studying children's naive representation of the Earth for some thirty years now. Until last year, the widely accepted conclusion was that children's representation of the Earth is quite wrong. To quote the Nobes et al. paper:

Children tend to produce very intriguing pictures and responses, including the flat Earth, which is a disk or rectangle with people living only on top; the hollow Earth, with people living inside; the flattened sphere, with people on top; and the dual Earth, with one Earth being flat and the other round and in the sky.

However, the authors point to a potential problem with this account: it is based on children's drawings, and it is well known that production is a hard task (for example, children understand most words before they are able to use them appropriately). Perhaps the demands of the task were too high for young children. In this case, they should be better if the demands were lower. To test this hypothesis, the authors asked children to order pictures of the Earth from the most accurate description to the least accurate. In this case, children (as young as 5 years old) display a view of the Earth much closer to the scientific one: they tend to prefer a spherical Earth over all other shapes (like a flattened or a hollow sphere), and they barely ever choose a disk (presumably, the parents of the ones who do have been reading too much of Pratchett's Discworld novels); they prefer pictures with the sky around it instead of on top of it; similarly, they prefer people to be depicted around instead of on top.

This is another instance in the long list of results in developmental psychology that have to be revised because the task demands were too high: when the task is well tailored for children's abilities, they get much better. We should not discount the first findings however: the fact that children draw flat (or otherwise misshaped) Earth is very interesting (and cute). And it poses an interesting question: if children have this nearly 'scientific' representation of the Earth, why do they fail to draw it right? A possibility is that they are just like adults who think they know something but in fact don't. Frank Keil and his colleagues have carried out experiments showing that when it comes to causal systems of average complexity (e.g. the rotor of an helicopter), we think we know how it works when in fact we only have the slightest idea. But, as the children in Nobes et al. experiments, perhaps adults would be able to pick up the correct description. Anyway since we don't quite know why adults display this gap it might not shed much light on the phenomenon in children, but the analogy is interesting anyway.

Some comments on this post:

Alberto Masala wrote: Actually, there is something misterious about middle-ages theories of earth: is is not flatness, but the absurd and still consensual belief (from late antiquity on) that the equator was strictly impossible to reach and cross because of a torrid climate incompatible with life. The first Portugueses sailors crossing equator were paralized by terror… has anyone any clue to explain this mysteriously strong belief?

Olivier Morin wrote: Well, I think the belief that Ecuador is a fiery belt around the Earth is a perfectly rational one from a middle-age perspective; you can easily see that the climate gets colder and colder as you move to the North, and you may have heard from Roman tales that in the extreme north, everything is frozen to the point of making life impossible. Since climate gets warmer as you move southwards, it is reasonable to extrapolate and explain this tendency at the same time, by positing that a fiery belt produces the heat that warms up the air (the influence of the Belt is weaker as you move northwards, which is as it should be).

No comments yet