Title Author Date

Error in review of Elishakoff Norowitz, Avi Nov 19, 2002

I have noticed an error in Amiel Rossow's review of Isaac Elishakoff's article concerning Torah Codes. The review states that WRR did not use a randomization of Genesis as a control text. However, I believe this is incorrect. In addition to randomizing the names and dates, WRR did test the correct names and dates against several control texts, including Genesis with randomized characters. According to WRR 1994, "The same calculations, using the same 999,999 random permutations, were performed for control texts. Our first control text, R, was obtained by permuting the letters of G randomly".

Related Articles: B'Tsel HaTorah

Title Author Date

Error in review of Elishakoff Rossow, Amiel Nov 19, 2002

Avi Norowitz is right when he states that WRR used as control texts a few
permuted version of Genesis. However WRR used as control texts also the text
of Isaiah and the Hebrew translation of Tolstoy's War and Peace. Norowitz is
wrong if he implies that these control texts were used in the same way as
the permuted lists of rabbis' dates/appellations. The statistical test used
by WRR was based on comparing the proximity within the pairs of semantically
related ELSs in the same (not permuted) text of Genesis, among one million
of permuted lists, and ranking the lists (rather than the permuted texts of
Genesis) in accordance with the values of that proximity. A similar
procedure was applied to the control texts. However, no statistical test was
performed based on comparing the control texts with the original Genesis.
The control texts played a role distinctively different from that of
permuted lists of dates and appellations. That this is so is evident, among
other things, from the use by WRR of Isaiah and War and Peace which have no
relation to Genesis. The permuted versions of Genesis served the same
purpose as Isaiah and War and Peace rather than as permuted lists of
dates/appellations. Therefore, Elishakoff's critique which implies that WRR
used permuted texts of Genesis for a statistical test, is irrelevant to the
actual situation with WRR's work, as was correctly noted in my review.
Norowitz's comment seems to be based on a misunderstanding. However, I'd
like to thank him for attention and taking time to write a comment.

Related Articles: B'Tsel HaTorah