DRAFT. This edition: April 7, 1998. First edition: March 3, 1998.

We study the Witztum-Rips-Rosenberg (WRR) sample of nations and find clear evidence that their results were obtained by selective data manipulation and are therefore invalid. Our tool is the study of variations - we vary the sample of nations in many ways, and find that the variations are almost always "worse" than the original. We argue that the only way this can be possible is if the original was "tuned" in one way or another. Finally, we show that "tuning" is a sufficiently strong process that can by itself produce results similar to WRR's.

The original paper of WRR is unpublished, but widely distributed in preprint form. Since it could be unfair to publish a critique of a paper that is unavailable to many readers for comparison, we are also providing a HTML edition of WRR's paper. We have attempted to make the copy exact, but will of course correct any errors that we are notified about.

Our paper is in English, but there are many Hebrew words and phrases within it. There are several ways of encoding Hebrew. Click on the link below that looks the most understandable to you.

• Hebrew: מדגם העמים של ויצטום ריפס ורוזנברג.
• Reversed Hebrew: .גרבנזורו ספיר םוטציו לש םימעה םגדמ
• Hebrew coded in lower-case latin: mdgo himyo wl vyxevo ryps vrvznbrg.
• Michigan-Clairmont Hebrew coding: MDGM H(MYM $L WYC+WM RYPS WRWZNBRG.
• On some systems, Hebrew printing is next to impossible. Here's a printer-ready PostScript version, with all the necessary fonts built in:
  (sorry, figures are not included. The PostScript version of the only figure directly relevant to the text is here).

More information on Web Hebrew:

Back to the Torah Codes page
Back to the Mathematical Miracles page

© Copyright (1998) Dror Bar-Natan drorbn@math.huji.ac.il, Brendan McKay bdm@cs.anu.edu.au, and Shlomo Sternberg shlomo@math.harvard.edu.