Bible
codes, also known as Torah codes, are words, phrases and clusters
of words and phrases that some people believe are meaningful and exist
intentionally in coded form in the text of the Bible. These codes were
made famous by the book The Bible Code, which claims that these codes
can predict the future. All of these claims are strongly denied by
skeptics and many religious groups.
Bible Code Overview
The
primary method by which purportedly meaningful messages are
extracted is the Equidistant Letter Sequence (ELS). To obtain an ELS
from a text, choose a starting point (any letter) and a skip (a number,
possibly negative). Then, beginning at the starting point, select
letters from the text at equal spacing as given by the skip. For
example, the bold letters in this sentence form an ELS for the word
SAFEST. (The skip is -4. Spaces and punctuation are ignored.)
Often more than
one ELS related to some topic can be displayed
simultaneously in an ELS letter array. This is produced by writing out
the text in a regular grid, with exactly the same number of letters in
each line, then cutting out a rectangle. In the example below, we show
part of the King James edition of Genesis (26:5-10) with 33 letters per
line. ELSs for BIBLE and CODE are shown. Normally only a smaller
rectangle would be displayed, such as the rectangle drawn in the
figure. In that case there would be letters missing between adjacent
lines in the picture, but it is essential that the number of missing
letters be the same for each pair of adjacent lines.
Arrange the
letters from Genesis (26:5-10) in a 33 column grid and you
get a word search with "Bible" and "code". Myriad other arrangements
can yield other words. Although we have shown examples in English
texts, Bible codes proponents usually use a Hebrew Bible text. For
religious reasons, most Jewish proponents use only the Torah
(Genesis-Deuteronomy).
Bible Code History
As far as is known, the 13th century
Spanish
rabbi Bachya ben Asher was
the first to describe an ELS in the Bible. His 4-letter example related
to the traditional zero-point of the Jewish calendar. Over the
following centuries there are some hints that the ELS technique was
known, but few definite examples have been found from before the middle
of the 20th century. At this point many examples were found by the
Slovakian rabbi Michael Ber Weissmandl and published by his students
after his death in 1957. Nevertheless, the practice remained known only
to a few until the early 1980s, when some discoveries of an Israeli
school teacher Avraham Oren came to the attention of the mathematician
Eliyahu Rips at the Hebrew University of Jerusalem. Rips then took up
the study together with his religious studies partner Doron Witztum and
several others.
Rips and
Witztum invented the ELS letter array and used a computer to
find many examples. About 1985 they decided to carry out a formal test
and the great rabbis experiment was born. This experiment tested the
hypothesis that ELSs for the names of famous rabbis could be found
closer to ELSs of their dates of birth and death than chance alone
could explain. The definition of "close" was very complicated, but
roughly two ELSs are close if they can be displayed together in a small
rectangle. The experimental result suggested very strongly that the
Bible codes phenomenon was real.
The great rabbis
experiment went through several iterations but was
eventually (1994) published in the peer-reviewed journal Statistical
Science. Although neither the Editor nor the referees were convinced by
it, neither could they find much wrong with it, so the paper was
published as a "challenging puzzle".
Witztum and Rips
also performed other experiments, most of them
successful, though none of them were published in journals. Another
successful experiment, in which the names of the famous rabbis were
matched against the places (rather than dates) of the famous rabbis,
was conducted by Harold Gans, an employee of the United States National
Security Agency.
The Bible codes
became known to the public primarily due to the
American journalist Michael Drosnin, whose book The Bible Code (Simon
and Schuster, 1997) was a best-seller in many countries. Drosnin's most
famous success was to predict the assassination of Israeli Prime
Minister, Yitzhak Rabin, using (so he claimed) the Bible code.
Opponents claim that in the political atmosphere of the time,
predicting the fact that Rabin would be assassinated with no additional
details is hardly impressive. In 2003, Drosnin published a second book
on the same subject.
Practice of the
Bible codes also spread into certain Christian circles,
especially in the United States. The main early proponents were Yaakov
Rambsel, a "Messianic Jew" (or a Jew that believes Jesus was the
Messiah) and Grant Jeffrey. By 2000, most books and most web sites
devoted to the codes were produced by Christians.
Bible Code
Criticism
The
primary objection advanced against Bible codes of the Drosnin
variety is that similar patterns can be found in books other than the
Bible. Although the probability of an ELS in a random place being a
meaningful word is low, there are so many possible starting points and
skips that many such words are expected to appear. Responding to an
explicit challenge from Drosnin, who claimed that only the Bible could
yield ELS, Australian mathematician Brendan McKay found many ELS letter
arrays in Moby Dick that contain ELSs related to modern events. Other
people, such as US physicist Dave Thomas, found other examples in many
texts. In addition, Drosnin had used the flexibility of Hebrew
orthography to his advantage, freely mixing classic (no vowels, Y and W
strictly consonant) and modern (Y and W used to indicate i and u
vowels) modes, as well as variances in spelling of K and T, to wrench
out the desired meaning.
Code proponents
respond by claiming that the ELS letter arrays
appearing in the Bible are better in some way than those appearing in
other books. They also like to hypothesize and investigate new types of
codes to stay ahead of criticism. However, in the absence of an
objective measure of quality and an objective way to select test
subjects, it is not possible to positively determine whether any
particular observation is significant or not. For that reason, most of
the serious effort of the skeptics has been focused on the "scientific"
claims of Witztum, Rips and Gans.
In 1999, McKay,
together with mathematicians Dror Bar-Natan and Gil
Kalai, and psychologist Maya Bar-Hillel, published a paper in
Statistical Science which they claim provides an adequate refutation of
the earlier paper of Witztum and Rips. Their main points were:
The data used by Witztum and Rips was a list of
Rabbi names in Hebrew. The Hebrew language is somewhat flexible as far
as name spelling goes, and each Rabbi has several different
appellations (aliases and nicknames), so special care should be taken
as to how to choose the particular names searched for. So their result
could be explained by claiming the data was not collected properly.
From the paper: "...the data was very far from [being] tightly defined
by the rules of their experiment. Rather, there was enormous "wiggle
room" available, especially in the choice of names for the famous
rabbis".
There is indirect evidence that the data were
not, in fact, collected properly; that is, the choice of names and
spellings was somehow biased towards those supporting the codes
hypothesis.
Attempts at replicating the experiment failed
to achieve the same result. From the paper: "A technical problem that
gave us some difficulty is that WWR have been unable to provide us with
their original computer programs. Neither the two programs distributed
by WWR, nor our own independent implementations of the algorithm as
described in WWR's papers, consistently produce the exact distances
listed [by WWR]".
Needless to say,
there has been a continuing debate on these claims.
The experiment of
Gans has also received critical attention. Several
attempts at replicating it, designed by mathematician Barry Simon, gave
negative results. Finally, a committee at the Hebrew University,
comprising both codes proponents and skeptics, ran two replications
using outside experts to compile the data. Both replications failed to
find the phenomenon that Gans' original experiment claimed to find.
As of 2003, there
are still a few university scientists who support the
codes. The main two are Eliyahu Rips (see above) and Robert Haralick
(an electrical engineer at the City University of New York). However,
there is no doubt that the overwhelming majority of scientists who have
looked at the claims reject them. Whether that remains the consensus of
scientific opinion, or whether new evidence will cause them to
reexamine their views, remains to be seen.
