This edition: May 27, 1999; First edition: September 19, 1997.
In [WRR2], Witztum, Rips and Rosenberg found a surprising correlation between famous rabbis and their dates of birth and death, as they appear as equidistant letter sequences in the Book of Genesis. We make a smaller or equal number of mistakes, and find the same phenomenon in Tolstoy's eternal creation "War and Peace".
Dror Bar-Natan Institute of Mathematics The Hebrew University Giv'at-Ram, Jerusalem 91904 Israel. drorbn@math.huji.ac.il |
Brendan McKay Department of Computer Science Australian National University Canberra, ACT, 0200 Australia. bdm@cs.anu.edu.au |
1. Introduction
1.1. Acknowledgement
2. Their choices, our choices
2.1. Modifications to specific appellations
2.2. Modifications to the list of personalities
2.3. Modifications to the list of dates
2.4. Our list of appellations
3. The Results
4. Bibliography
Our inspiration comes from reading the paper [WRR2] by Witztum, Rips and Rosenberg (WRR). The outline of the story in [WRR2] is as follows:
In [WRR2], WRR write:
The list of appellations for each personality was provided by Professor S.Z. Havlin, of the Department of Bibliography and Librarianship at Bar Ilan University, on the basis of a computer search of the "Responsa" database at that university.Contrary to what is suggested by the above quote, many of the appellations WRR use do not even appear in the Bar-Ilan Responsa database [Re]. Thus in addition to the Responsa database [Re] we will also refer to the Margalioth Encyclopedia [Marg], used by WRR to select the rabbis, and to the highly-regarded Encyclopedia Hebraica [Heb] used by WRR in several of their other investigations. We (like WRR) also use other sources as needed.
We note that there may be a case for removing the name imdy` altogether, for it is just the name of a town were the Yaabez was briefly a Rabbi, and not his last name. The Yaabez himself wrote in [Ya],
... The busy Yaakov Israel known as Yaabez good omen was never called Yaakov Emdyn (imdy`) (as has done the honourable in the address on this letter as is the deplorable habit in the tongues of the people). It is known I am not from the people of Emdyn, was not born there, do not expect to see it, but am looking after its welfare and benefit as in the former times when I was under it's load, [and] so also am I now diligent for it's remedy. ...Little did his plea help, and the Yaabez has several common appellations which are variants of the word imdy`. We keep them in our list. However, we can learn from this example that WRR didn't mind using names that were not used (or were even rejected) by their bearers. This lesson will become relevant in the case of Rabbi Shlomo Chelma below.
At this point we got the advice of two wise men. One suggested that we look at [Hei], a book on the Jewish laws in the state of Moravia, where Rabbi Menachem Mendel was the state's Rabbi. From the other wise man we learned to check the citations in the footnotes. One footnote, on page 111 of [Hei], he checked himself. It lead to an article [Marx], that contains a letter written by the son of a nephew of Rabbi Menachem Mendel in the late 17th century, only a few dozen years after Rabbi Menachem Mendel died in 1661. In that letter Rabbi Menachem Mendel's surname is given as qracmal. The following day (and a continent away), we checked the footnote on page 102. It lead us to two articles, [Hor] and [Ka], devoted to our Rabbi and his descendents. Both articles use the spelling qrakmal extensively and not just in passing, and the latter one even explains where the name comes from! It relates the name qrakmal to a certain Dayan, Rabbi Jonah Krochmals in the city of Cracow, where Rabbi Menachem Mendel was born. A transcript of the tombstone of Rabbi Jonah Krochmals is given in [Zu, page 180] and Krochmals is spelled qrakmalw there. Thus there is no doubt that the original spelling of the name Krochmal is qrakmal and thus it is fully justified to remove the appellation qrvkml from the list, an to put qrakmal instead.
We also note that once it is clear that qrakmal is an acceptable spelling for Krochmal, the usage of this spelling is mandatory according to the WRR rules, which state explicitly that Yiddish names are spelled as in the original Yiddish. Recall that qrakmal is a the Yiddish word for starch, and from [Ka] we learn that the first people to be called by the family name Krochmal were called so because they traded in starch.
However, in his "refutation" of a draft version of this article, Witztum argued against the use of the appellation qrakmal. We completely disagreed with his argument, but were amused to note that Witztum himself gives a reference for a source that uses the form qrvkmal. If, as Witztum indicates, qrvkmal is a valid form, then it must have been used along with qrvkml, by Witztum's own "mater-lectionis" rule. Thus we may feel free to use either qrakmal or qrvkmal instead of qrvkml. This time we chose qrvkmal.
At the same time we note that WRR's use of variants like zkvta and zkvtv sets a precedent which we are allowed to follow elsewhere. Hence for Rabbi II-28, Rabbi Moshe Margalith, we add the variant mrglyvt of his last name. This variant appears in the titles of his entries in [Az, Vin, St, Ju, Fri], in the entry for his book in [B-Y] (p. 487) and also in the index of [Marg].
More than that, one expert we consulted said that there is a possibility that even this single appearance is no more than a typo in the original addendum. The reason for this suggestion is that a different permutation of these letters, ahic"r rph, appears with its expansion, any hxiyr imnval cy ryqy (me the young, Immanuel Hai Ricchi), and an explanation of the rph part, several times in several of this rabbi's books. We note that in Hebrew the latter expansion makes much more sense than the former, since the former taken literally means "I'm alive, the young Immanuel Ricchi". Indeed, the acronym ahic"r is listed in the dictionary of acronyms [AY], while a"c hi"r is not. (By the way, since ahic"r rph is also a non-pronounced signature it may be considered wrong to include it as well, if one wishes to take WRR's commitment to pronounced appellations more seriously than they did.) Be that as it may, the reasons brought before are more than enough to show clearly that including a"c hi"r is extremely wrong. We remove it.
On the other hand, we found two other signatures (or self-references) used by this rabbi, that unlike ahic"r rph and achi"r rph, are definitely pronounced, and therefore we add them. These are: hiwy"r - appears in [Ri1] page 4. It is a permuted acronym of hxiyr wwmy imnval ryqy yx"v, and also serves as a word-play on the name of the book "Hon-Ashir". avhb vr"i - appears in [Ri2] - part 2, page 53b. means "loving friend" and also is an acronym of "Ricchi Immanuel".
Thus we wish to add the appellations clma, cilma, wlmh clma, and wlmh cilma. In practice we only add cilma and wlmh clma, for the other two appellations do not fit within 5-8 letters. We will add that it appears that this Rabbi didn't use these names himself, but was only called so by others, however as WRR's precedent with the name Emdin (that was mentioned above) shows, this needn't make any difference, and these names can be included.
The old and new lists, a comparative table | ||||
---|---|---|---|---|
# | Original entry | we remove | we add | new entry |
1 | hawkvl, hrab"d, hrab"y, hrb ab"d, rby abrho |
hawkvl, hrab"d, hrab"y, hrb ab"d, rby abrho |
||
2 | zri abrho, yxcqy, rby abrho | zri abrho, yxcqy, rby abrho | ||
3 | hmlaj, rby abrho | hmlaj | rby abrho | |
4 | abrho sbi, xrvr hmr, rby abrho | --- completely removed --- | ||
5 | rby ahr` | rby ahr` | ||
6 | miwy hwo, miwy y/h/v/h | bil miwy h', miwy h' |
bil miwy h', miwy h', miwy hwo, miwy y/h/v/h |
|
7 | avpnhyo, rby dvd | avpnhyo, rby dvd | ||
8 | dvd hngyd, rby dvd | --- completely removed --- | ||
9 | dvd nyev, rby dvd | dvd nyev, rby dvd | ||
10 | rby cyyo | hmhrc"a, mhrc"a | hmhrc"a, mhrc"a, rby cyyo | |
11 | bnbnwt, rby cyyo | bnbnwt | bnbnwty, hrb hcby"b, hrb cby"b, rb cby"b |
bnbnwty, hrb hcby"b, hrb cby"b, rb cby"b, rby cyyo |
12 | bil hns, bil ns, kpvsy, rby cyyo |
bil hns, bil ns, kpvsy |
kapvsy | kapvsy, rby cyyo |
13 | hmhrc"w, cyyo wbty, mhrc"w, rby cyyo |
hmhrc"w, cyyo wbty, mhrc"w, rby cyyo |
||
14 | cvt yayr | cvt yayr | ||
15 | rby yhvdh | hr"y csyd, yhvdh sg"l |
hr"y csyd, yhvdh sg"l, rby yhvdh |
|
16 | mhr"y iyaw, rby yhvdh | iayaw | mhr"y iyaw, iayaw, rby yhvdh | |
17 | rby yhvsf | rby yhvsf | ||
18 | mgny wlmh, rby yhvwi | mgny wlmh, rby yhvwi | ||
19 | hmhry"e, hmhrym"e, erany, yvsf erny, mhry"e, mhrym"e, merany, merny, rby yvsf |
hr"y erany, hr"y erny, r"y erany, r"y erny |
hmhry"e, hmhrym"e, hr"y erany, hr"y erny, erany, yvsf erny, mhry"e, mhrym"e, merany, merny, rby yvsf, r"y erany, r"y erny |
|
20 | pry mgdyo, rby yvsf, tavmyo | --- completely removed --- | ||
21 | hryb"r, yiqb byrb, mhr"y byrb, rby yiqb |
hryb"r, yiqb byrb, mhr"y byrb, rby yiqb |
||
22 | bil hlq"e, cagyz | cagyz | mhr"y cgyz, r"y cgyz |
bil hlq"e, mhr"y cgyz, r"y cgyz |
23 | hmhry"l, yiqb hlvy, yiqb sg"l, mhr"y hlvy, mhry"l, mhr"y sg"l, mvly`, rby yiqb |
hmhry"l, yiqb hlvy, yiqb sg"l, mhr"y hlvy, mhry"l, mhr"y sg"l, mvly`, rby yiqb |
||
24 | hyib"u, hryib"u, hr"y imd`, hr"y imdy`, imdy` |
hr"y imd` | r"y imdy` | hyib"u, hryib"u, hr"y imdy`, imdy`, r"y imdy` |
25 | hvrvvyu, yxcq hlvy, rby yxcq |
hvrvvyu | hvrvbyu | hvrvbyu, yxcq hlvy, rby yxcq |
26 | xmc xdq, qrvkml, rby mnco, rby mindl |
qrvkml | qrvkmal | xmc xdq, qrvkmal, rby mnco, rby mindl |
27 | hmhrm"z, hmzl"`, zkvta, zkvtv, mhrm"z, mhr"o zkvt, mwh zkvt, mwh zkvta, mwh zkvtv, qvl hrm"z, rby mwh |
zkvta, zkvtv, mwh zkvta, mwh zkvtv |
hmhrm"z, hmzl"`, mhrm"z, mhr"o zkvt, mwh zkvt, qvl hrm"z, rby mwh |
|
28 | mrglyt, pny mwh, rby mwh | mrglyvt | mrglyvt, mrglyt, pny mwh, rby mwh |
|
29 | rby izryh | rby izryh | ||
30 | a"c hi"r, ywr lbb | a"c hi"r, ywr lbb |
avhb vr"i, hv` iwyr, hiwy"r |
avhb vr"i, hv` iwyr, hiwy"r |
31 | hmhrw"w, mhrw"w, mzrcy, rby wlvo, wriby, wr wlvo |
hmhrw"w, mhrw"w, mzrcy, wr wlvo |
rby wlvo, wriby | |
32 | rby wlmh | cilma, wlmh clma | cilma, rby wlmh, wlmh clma | |
33 | --- new Rabbi --- | ayznweae, ayznwee, mhr"o a"w, rby mayr |
We have shown our list of appellations (as it appeared in a draft version of this paper) to Professor Menachem Cohen, of the Department of Bible at Bar-Ilan University, Israel. In reference to our list and to the original WRR-Havlin list he wrote in [Co2]:
... I see no essential difference between the two lists for the purpose of using them for skip experiments in any text.The present list is even more similar to the WRR-Havlin list than the one Cohen based his judgement upon.
P1 | P2 | P3 | P4 | |
our list on War and Peace | 20,887 | 1,091 | 831 | 57 |
our list on Genesis | 14,179,712 | 1,299,355 | 16,825,659 | 724,260 |
the WRR list on Genesis ([WRR2], rescaled) | 45,300 | 500 | 57,000 | 400 |
Comparing the first and the last two rows, we see that our list does as well on War and Peace as the WRR list does on Genesis. Tolstoy would have enjoyed knowing that. Some of the numbers in the second row are "middle of the way" - smallish but not very small. They are "smallish" because our list is highly correlated with the original WRR list, on which it was based. The fact that they are not very small needs to be explained by WRR, not by us. Why is it that an equally valid list of appellations (our list) does so much worse than their list on Genesis?
Comments: The computations of the significance levels for
our list was carried out using a program functionally equivalent to a
program WRR gave us, els2.c
. The significance levels for
the WRR list were taken from [WRR2], except for a
rescaling by a factor of 100 to account for the the fact that they used
only 106 permutations in their computations. Using our own
program, which is equivalent to a program WRR produced when their lists
were already present and hence it is more susceptible to bias, we get
somewhat better results for the WRR list on Genesis, but these results
are still weaker than our results on War and Peace. Unfortunately, we
were not able to obtain from WRR the exact programs they used to
compute the results in [WRR2].
The text we used was given to us by WRR. It consists of the first 78,064 letters (the length of Genesis) of a Hebrew translation of Tolstoy's War and Peace.
© Copyright (1997-9) Dror Bar-Natan, drorbn@math.huji.ac.il and Brendan McKay, bdm@cs.anu.edu.au.