I discovered an example of Stigler’s Law

from Wikipedia:

Stigler’s Law of Eponymy is a process proposed by University of Chicago statistics professor Stephen Stigler in his 1980 publication “Stigler’s law of eponymy”.[1] In its simplest and strongest form it says: “No scientific discovery is named after its original discoverer.” Stigler attributes its discovery to sociologist Robert K. Merton (which makes the law self-referencing).

So what is the example I discovered?*  I’ve often hear someone mention one, but never more than one, of the following: Campbell’s Law (1976), Goodhart’s Law (1975), and the Lucas Critique (1976).  Just recently I came across a reference to “Steve Kerr’s classic 1975 article, [On the Folly of Rewarding A While Hoping for B]”.

OK, so they aren’t identical (the Lucas critique probably has the most unique content), but they are all closely related.  I looked over each original article (except the one for Goodhart’s Law which doesn’t appear to be online) and none of them cite the other.  I’m sure that with some research one would find that any one of these alone would be an example of Stigler’s Law, that is, we could find very similar insights from someone like Adam Smith or J.S. Mill.

A couple questions:

1. Had the authors been exposed to the others’ similar ideas at the time they published theirs?  Did they make a conscious decision not to cite?  (Note that while the publications have dates, we have no idea when each author first had their idea.)

2. Let’s assume that none of them got their idea directly from one of the others.  What does it say that they each published their idea in 1975-76?  Is it unusual that multiple people had similar ideas, later recognized to be important, at the same time?  Or is it, as I suspect, merely unusual that all four different people are recognized?  I wish I could be chatting with Robert Merton about this.

* While I’m a little proud of the fact I noticed these similarities, and that I edited Wikipedia to link the first three ideas, I hope its obvious that my claim to having “discovered” this example is intentionally ironic.

4 Responses to I discovered an example of Stigler’s Law

  1. MishaTeplitskiy says:

    Derek Del Solla Price has an interesting discussion about this issue in his book? from the 1960s?. A real simple model of scientific discovery that he uses predicts approximately the right distribution of multiple discoveries per 1000 “problems waiting to be discovered”. Fun stuff. I’ve been thinking about fooling around with his model this summer in case, ya know, Europe gets boring.

  2. Michael Bishop says:

    Thanks for the suggestion. http://en.wikipedia.org/wiki/Derek_J._de_Solla_Price sounds like an interesting thinker. I haven’t looked at the model you mention, but I wonder… what is the data used to test the model? In many/most cases there may be no record of multiple discovery despite the fact that it was indeed discovered by more than one person.

    • MishaTeplitskiy says:

      The data came from … Merton, of course! (Cited in Merton, Proceedings of the American Philosophical Society, 105 [October, 1961], 483)

  3. MishaTeplitskiy says:

    Don’t know what discoveries were tallied up to make the table, but here it is:

    Number of simultaneous discoverers | Cases
    2 | 179
    3 | 51
    4 | 17
    5 | 6
    6 | 8

Leave a comment