Volume 11 Number 34
Produced: Fri Jan 21 0:16:39 1994
Subjects Discussed In This Issue:
Codes Research
[Hayim Hendeles]
Hidden Codes
[Mike Gerver]
----------------------------------------------------------------------
From: Hayim Hendeles <hayim@...>
Date: Tue, 4 Jan 94 10:04:07 -0800
Subject: Codes Research
Someone commented in a recent post:
A few weeks ago I posted a question to the practitioners of the
codes. I noted that advanced codes had been found in other
literatures and not only in the Torah (By the way, I am sure
the same codes found in the forah could be found in Nach). I
also raised the most important issue, namely, ... in my
opinion, even though the codes are a fraud I do not have any
problem using them in Discovery like seminars. This is in
opposition to the opinion of my friend Prof. [name deleted] who
told me that the codes are dangerous. He also told me that some
missionaries have found codes saying Jewus [sic] is God.
...
fundamentally affected. The computers would have come up with
just as many other codes that agreed with this form.)
And in another post, someone else comments:
Yes, the codes are still playing a central role in the
Discovery seminar, at least in the way it's advertised. In a
flyer about a seminar to be presented on Long Island on Sunday
Dec 19, there are two prominent topics: (1) Codes research (2)
"FAILSAFE -- a series of workshops which employ the techniques
of the Israeli Mossad to explore the Torah's origin." Does this
suggest a carnival atmosphere? Does anybody have more
information about this (new?) subject, FAILSAFE? Is there any
doubt that the "Israeli Mossad's techniques" will demonstrate
the divine origin of the Torah?
...
* Ask if the codes are supposed to convince someone to return
to Mitzvot and you are told something like "Of course not,
it's just interesting material."
* If you read the Discovery brochures, however, they seem to
be saying that they will convince you if you are not already
enlightened.
First of all, let me preface my remarks by mentioning the fact that I
happen to be one of the Codes lecturers for Discovery. I have a copy of
the paper on the Codes, and while I am not a mathematician, I have some
math background and am familiar with the math. Furthermore, I have
reviewed the paper with several mathematicians, who themselves have
reviewed it/discussed it with their colleagues. It was only after I
received their approval, that I even began considering giving the codes
lectures.
In addition, several of my colleagues have consulted prominent Rabbanim
and Gedolim, who have approved lecturing about the codes PROVIDED
that every word said is 100% accurate. Thus, when I give the lectures,
I am very careful to point out exactly what parts of the code are
known to us by Tradition, and what parts appear to be a fascinating
mathematical phenomenon - which is the topic of the paper and
to the best of my knowledge, are not part of our Tradition.
I have hesitated to respond until now, as have others who have worked
on the subject, because of the controversial nature of the codes which
will undoubtedly generate much flames and hate mail (which I will not
respond to), and numerous questions which I *may not* have the time to
respond to.
I am only responding in this forum publically, to protect the honor of
the Discovery seminars. These are conducted by individuals who
sincerely wish to bring back those people who are lost to their faith.
As the previous poster alleges, I will not deny that the purpose
of these Seminars is to provide evidence that the Torah is indeed of
divine Origin. Unfortunately, in these sophisticated times, this is
considered to be a crime by many whose only god is that of Science. It
is our intent to demonstrate that belief in a Divine origin of the
Torah is not merely an unsubstantiated belief of religious fanatics,
but also a reasonable conclusion based on the available evidence. (No
intent is made to provide a mathematical proof of anything.)
The goal is not necessarily to make the audience into observant Jews -
but to bring them one step closer to G-d and the Torah. As I once heard
from a great Rabbi, who said in the name of Rabbi Yisroel Salanter
zt"l (I believe), that it is worthwhile to lecture someone for 24 hours
straight so that instead of smoking 40 cigarettes on the Shabbos, he
should only smoke 39 cigarettes. To that extent, I can tell you
definitely that the success we have had is outstanding.
For those who are critical of Discovery and their methods, if you
believe you have a better approach to attract our estranged brothers,
it will behoove all of us and Kllal Yisroel, for you to post your
ideas. If you have ideas on how we can improve our program, we
would love to hear it. Otherwise, I suggest you keep your comments
to yourself, as it does no one any good. Slander has caused Kllal Yisroel
enough grief in the past 2000 years - we do not need any more.
Despite the allegations of the above poster that:
... in my opinion, even though the codes are a fraud ...
I am assuming the author intended to use the word "error". "Fraud"
implies a deliberate attempt to mislead the public, knowing fully well
the material to be false. If the author truly believes otherwise,
then he should post his evidence to that effect, as we would all
like to see it - myself included.
Several more points, before responding to the questions raised above.
There are several lecturers on the Codes topics, as there are different
audiences. Each lecturer presents the material at a different level.
Non-technical audiences have a difficult time following the lectures
presented by mathematicians - but at the same time one with a technical
background will get very little out of the lectures given by a Rabbi.
Unfortunately, many technically-oriented people have very little
patience or respect for the non-technical average man in the street who
cannot appreciate a highly technical presentation. I have
received numerous criticisms for my lectures, which are "too technical"
and "contain "too much math" for Mr. Joe Average. Eliminating the
math for the benefit of Mr. Joe Average, will earn me condemnation
from the technically oriented people.
In general, it is the technical people with little tolerance or respect
for Mr. Joe Average, who criticize those lectures intended for a
non-technically oriented audience.
Obviously, there is no ideal (practical) solution to this problem -
other then to recognize the different speakers and audiences.
And finally, the answers to some of the questions raised in earlier posts.
As far as those who claim that "codes" have been discovered on other
documents as well, this topic is usually discussed at the seminars.
Unfortunately, the use of the generic term "code" to refer to so many
different things, leads to much confusion. Many of these pseudo-codes
are discussed at the lecture, and quickly discarded. In general,
for a code to be of interest, several questions need to be asked:
#1) Are these "codes" significant? In any random document,
statistically you would expect to find a certain number of codes. Are
the codes found within the statistical norm, or are they vastly
different then one would expect at random. As a related question, what
is used for a control text? How do verify your statistical predictions?
#2) Has the data/texts been fudged to achieve the desired results?
Some of the known codes are found to be a result of various manipulations
of the texts, in order to achieve these results.
#3) In line with #1 above, can these codes be shown to be apriori?
#4) Are the results repeatable?
#5) And finally, even if the "codes" you find are real, what does it
prove? E.g. our Shabbos Zemiros are replete with "codes", which
can only be used to show the author deliberately encoded his name
in the Zimra. And therefore, what? Such "codes" have obviously, absolutely
nothing to do with the type of codes reported in the Discovery seminars.
(As an aside, an earlier post referred to "codes" found in
Shakespeare. I am not familiar with this at all, but am very intrigued
by it - and would love to hear more about it. Assuming this is true,
the work required such genius (and I don't think anyone will argue with
that statement), that I wouldn't be surprised at all, if some sort of
"code" was found in his work. And assuming they would pass all the
statistical tests (e.g. those outlined above) you might conclude that
these "codes" are deliberate. But what does that prove? Only that the
author was quite smart! The Codes' lectures refer to similiar "codes" in
the Torah which can only be used to prove *at best* that the author was
very clever. Obviously, this is of little interest to anyone.)
Unfortunately, the author above fails to specify any details about
the supposed "advanced codes" found in other literatures as well.
Assuming this represents serious work and is not just another smokescreen,
if these codes pass all the statistical tests outlined above, I would
be happy to pass this information along to one of the mathematicians
doing the research on the topic. For the past year, in order to
satisfy some of the reviewers, they have been trying numerous different
control texts, and to date every single variation tried for the
control texts have all produced the statistically expected results.
Undoubtedly, this information will be quite germane to their research.
The original author also comments:
By the way, I am sure the same codes found in the Torah
could be found in Nach.
Thank you very much for your opinion. The bulk of the work done to date
has only been done on the book of Genesis. Thus, the paper on the codes
is entitled "Equidistant Letter Sequences in the book of Genesis". But
if you have more information on other books, I am sure we would all be
interested in hearing it.
Can the codes be dangerous? Perhaps. But then again, so are knives.
In the hands of a surgeon, it is a lifesaver, but in the hands of
a mugger, it is a deadly weapon. In the words of the Posuk,
"Ki yesharim darchei Hashem, tzaddikim yelchu Bam, Uposhim
yikashlu bam"
Sincerely,
Hayim Hendeles
----------------------------------------------------------------------
From: <GERVER@...> (Mike Gerver)
Date: Sun, 16 Jan 1994 2:30:31 -0500 (EST)
Subject: Hidden Codes
Some of the responses to my earlier "hidden codes" postings, in particular
those of Marc Shapiro in v10n87, and Rick Turkel in v10n96, indicate that
different people mean different things when they talk about "hidden codes,"
and that this ambiguity is causing confusion in the discussion. Let me
distiguish between two different claims:
I. The claim that there exist equidistant letter sequences in the Torah
which spell out various words and names, without any reference to whether
or not these letter sequences would be likely to occur by chance.
II. The claim that some subset of such letter sequences would be extremely
unlikely to occur by chance, so unlikely that any reasonable person, even
a fair-minded atheist, would admit that some other explanation is needed.
When Marc Shapiro talks about "hidden codes," he apparently means Hidden
Codes I. Then I agree with him that it is easy to verify that hidden
codes exist in the Torah, that they also exist in other literature (the
Koran, Shakespeare), and that it doesn't do any harm to look for them in
the Torah, use them for drashot, etc. I would add the caveat, however, that
it could do harm to show them to a mathematically naive audience, who might
assume that they were unlikely to occur by chance, and who might then
erroneously attach more significance to them than they deserve. To see
whether you are dealing with a mathematically naive audience, tell them
that if there are 24 people in a room, there is a better than 50% chance
that two of them have the same birthday. If your audience is amazed by
this fact, then they are mathematically naive, and caution should be
exercised when showing them examples of Hidden Codes I, to avoid violating
the mitzvah of "lifnei iver, lo titen mikhshol" [do not put a stumbling
block before the blind].
Rick Turkel also seems to think that Witztum et al are talking about Hidden
Codes I, or that their claim of statistical significance is superficial.
He asks, for example, whether they might have analyzed the data in a lot
of different ways, or chosen their data according to a lot of different
rules, until they found a set of equidistant letter sequences that they
could claim was unlikely to occur by chance. But in fact, Witztum et al
are definitely talking about Hidden Codes II, and not in a superficial way.
Their analysis may very well be wrong, in fact I rather suspect it is, but
they are not complete idiots. Their paper claims to deal with the issues
that Rick brings up. The list of names and yahrzeit dates, for example, is
chosen from an encyclopedia according to an objective, and not very
arbitrary, set of rules. Similarly, their method of measuring how close
the letter sequences for names are to the letter sequences for dates, is
reasonable sounding, not very arbitrary. There may be 1000 different ways
they could have chosen the list of names and dates, and the method of
measuring how close the letter sequences are, that would sound equally
reasonable. But they claim that there is only a chance of 1.e-17 that
they would obtain the results they do by chance, and you would not expect
this to occur if they tried only 1.e+3 different ways of choosing lists
and defining closeness.
This is not to say that Witztum et al's analysis is correct. One possibility
is that their list of names and dates is not chosen according to _any_
objective rule, but that (contrary to what is stated in the manuscript)
they picked and chose whom to put on the list, according to how close the
letter sequence for the name was to the letter sequence for the date.
This isn't necessarily intentionally fraudulent, they could be fooling
themselves with wishful thinking. For example, they could be including only
some last names but not all last names, and they could be including only
certain titles of books (by which the authors are known). Another
possibility is that their statistical analysis is in error, more likely
due to a software error than a mathematical error I would think. Another
possibility is that, due to some subtle correlation in names, dates, and
the letter sequences of Hebrew, the probability of the results occuring
is actually much greater than 1.e-17. I'm not sure quite how this could
occur, but I note that there could be correlations in the lists that are
far from obvious. For example, I originally found it very convincing
that when they shifted over the lists of names and dates, associating each
name with the yahrzeit date of the next person on the list, then all of
the correlations disappeared. That would be a convincing test if there
were no natural reason for the names and yahrzeit dates to be correlated
with each other. But in fact it is entirely natural that there should be
some correlation between name and dates. For the vast majority of people
listed in the encyclopedia they used, there was no known yahrzeit date.
For the earlier people, especially, the yahzeit date was more likely to be
remembered if there was something memorable about it, for example if it
was Rosh Chodesh, or Erev Yom Kippur. And different names were popular
for the earlier people (mostly Sephardic) than for the later people (mostly
Ashkenazic). So some correlation should exist between names and dates. I
have no idea why that would affect their analysis, offhand it seems that
it wouldn't, but it illustrates that there could be subtleties in
calculating the probability that they overlooked.
I am currently looking into all of these possibilities and will report if
I come up with anything. While I agree with Marc Shapiro that fooling
around with Hidden Codes I is harmless, I think that making false claims
of Hidden Codes II is dangerous.
By the way, in answer to the questions that have been posed about the effect
that different texts (Yemenite vs. Ashkenazic and Sephardic) would have on
the codes, it seems that they would have little effect on Witztum et al's
conclusions (if they are correct), although they would affect some individual
correlations between names and dates. Of the differences in the texts that
Marc lists in v10n99, only three of them occur in Genesis, and all of these
are in the first 9 chapters. Since the letter sequences Witztum et al deal
with all have between 5 and 8 letters, a large fraction of them probably do
not include any letters in the first 9 chapters, and would be unaffected.
The correlations between letter sequences for individual names and dates
are not statistically significant, only the entire set of them (about 300
pairs) is, and this would probably not be affected by the use of the Yemenite
text. (Witztum et al do look at the Samaritan text, which has a much larger
number of differences from our text, and do not find a statistically
significant correlation.)
Mike Gerver, <gerver@...>
----------------------------------------------------------------------
End of Volume 11 Issue 34
