On the Refereeing in Statistical Science
The code proponents have made repeated use of the fact that one piece of their work has
appeared in a peer reviewed journal, namely Statistical Science. While the
referring process is normally totally confidential, the referee during part of the process
has made some statements about the process and I've both received his permission to pass
them on and I've confirmed them with the editor of Statitical Science at the time.
It is an important part of understanding the significance of the appearance of this paper
to know what was actually recommended.
First, you need to know that the paper was initially submitted to Proceedings of the
(U.S.) National Academy of Sciences (PNAS). The referees during that part of the
process included Persi Diaconis, at the time at Stanford and more recently at Harvard and
Cornell. Many, including me, regard him as the world's leading mathematical statistician.
He and the other referees decided the paper was not suitable for PNAS and it was rejected.
Diaconis recommended that the paper instead be considered for Statistical Science.
This journal is not a typical scientific journal with nothing other than dry, formal
mathematical presentations. It is a lively journal that includes historical reminisces and
Diaconis (who advised the editor but wasn't a formal referee on this iteration of the
process) did not recommend that the paper appear as a stand alone piece because he
regarded the paper as "a curious but not solid result." Rather, he proposed that
the paper appear as a discussion piece together with various rebuttals and replies to the
Rob Kass, the editor of the journal at the time, decided to publish it without
discussion although initially he agreed to weave some negative comments from Diaconis into
his editorial on the piece. In the end, Kass decided against even doing that. Kass
has posted further remarks on the
significance of the acceptance of the paper.
Kass is quoted in the New York Times as saying about papers that Statistical Science
accepts: "We hope that the material in them is correct but we also try to publish
pieces that are amusing to a wide variety of statisticians."
Aish HaTorah's website (http://www.discoveryseminar.org/bogus.htm)
says of the acceptance of this paper: "Professional scientific and mathematical
journals consult with a cadre of world class experts for the purpose of insuring that an
article containing a mathematical or scientific flaw is not published in their
journal." You decide whether you think that applies in this case given this
information about the process.
Note: This is additional commentary provided by Barry Simon and is not part of the Jewish