Accounting for overspecification and indifference to visual accuracy in manuscript diagrams: a tentative explanation based on transmission (Q1672043)

Extant medieval manuscripts of a Greek mathematical or astronomical work -- such as those of Archimedes, Aristarchus, or Euclid -- present a peculiar tendency to represent a higher regularity among the geometric objects than required by the argument, and often are inaccurate depictions of the objects discussed in the text. While the standard view is that these tendencies go back to the original manuscripts, the author proposes a different explanation, namely that the characteristics in question are always the result of successive copies of a text, and thus are due to the chain of transmission and not to the original. To prove his point, the author has asked ``many groups of students to play the role of copyists\dots In each group, around 20 students were placed in a circle. I gave a folder to one of them, containing a drawing of the diagram as I presume the Greek author drew it.'' The results of the experiment conclusively show the tendency of successive copyists to create the more regular diagram and to eventually move significantly away from the original.