Last week I was listening a 2009 BBC Radio ‘In our time' episode “discussing the epic feud between Sir Isaac Newton and Gottfried Leibniz over who invented an astonishingly powerful new mathematical tool - calculus.”
Coincidentally I was reading through the Chronicle of Higher Education Forum yesterday and came across this gem from 1994. “A Mathematical Model for the Determination of Total Area Under Glucose Tolerance and Other Metabolic Curves”, Mary M. Tai, Diabetes Care, 1994, 17, 152–154. The paper outlines a ‘new’ way of measuring the area under a curve by adding up areas of rectangles and triangles which the author calls Tai’s method. In fact the method described has been understood for centuries as is known as the trapezoid rule. (I use this method in statistics for humanities for calculating the Gini co-efficient.)
I’ve not heard about the paper before and the journal quickly published responses pointing out that this method was not new. However, the paper continues to be cited… and not only by people talking about it.
There is an interesting discussion of the paper on Stack Exchange about what should have happened to the paper. The first question might lie in wondering why no one out of the author, her colleagues, the reviewers or the journal editor had ever seen this before or something like it (I first remember coming across it in A-level Geography—pre 1994). On the other hand this demonstrates that it is perfectly possible for someone who has never seen as wheel to ‘invent’ the wheel, and the technique has been brought to a new audience, albeit about four centuries late.
This might be an extreme example, but I’m sure it’s not the only one of its kind. It begs the question though how much published research is genuinely innovative and how much is non-innovative stuff discovered independently.
Perhaps the most disturbing thing though is that twenty years later the paper is still being cited, and not just by people pointing out that this paper is nothing new. This paper from 2009 reads "Glucose and insulin areas were determined using Tai's model" (p.1046).