Re: [XTalk] "Conclusion" to Jesus Quest
- Statistics are irrelevant in this case, as the text under discussion does
not use the word about which the article speculates.
>I would like to apologize for my tone in the recent
>discussion. I see now that the reason Listmembers
>have become so frustrated is that they have been
>arguing past each other and failing to address
>the central issue in my paper, that is to say,
>my statistical argument. This statistical method
>needs to be understood clearly before the paper can
>be assessed. If anyone would like to discuss this
>method with me, please feel free to contact me
>on list or off, or, alternatively, call me at
>home at (212) 744-9450.
- Pursuant to my earlier message of today, I would
like to add that technically Bob Schacht is correct:
the Zipf distribution function can indeed be
used in my calculation instead of the random
approach that I used in my study. However,
having just gotten off the phone with Dr. Bob
Gorman, let me say that Dr. Gorman feels this
would not change the outcome significantly. (That
is why he did not suggest using it in the first
place.) If you want to go ahead and do the
calculation anyway, you will see this.
Actually, there is no specific reference to the
Zipf distribution function in statistical literature,
because the term did not originate within the
profession of statistics. It originated, I believe,
with a philologist. All it is is a multinomial
distribution. In order for such a distribution to
affect my calculation, it would have to be strongly
skewed toward "stirps," the metaphor in question.
That is to say, a high percentage of words from
the pen of Severus or the author of fragment 2
would have to have been "stirps." Since it is
obvious that this wasn't the case, there is no
point really in using a multinomial distribution here,
and it wouldn't affect the outcome much anyway.
Still, if you feel it would, do the math and present
it to us.
PO Box 286510
New York, NY 10128
Tel. (212) 744-9450
Get free email and a permanent address at http://www.netaddress.com/?N=1
- At 01:57 PM 12/2/00 -0500, you wrote:
>Pursuant to my earlier message of today, I wouldNot only can, but it would be highly appropriate. Zipf distributions have
>like to add that technically Bob Schacht is correct:
>the Zipf distribution function can indeed be
>used in my calculation instead of the random
>approach that I used in my study.
been shown to characterize use of words in a natural language (like
English) and the popularity of library books, so typically
* a language has a few words ("the", "and", etc.) that are used
extremely often, and a library has a few books that everybody wants to
borrow (current bestsellers)
* a language has quite a lot of words ("dog", "house", etc.) that
are used relatively much, and a library has a good number of books that
many people want to borrow (crime novels and such)
* a language has an abundance of words ("Zipf",
"double-logarithmic", etc.) that are almost never used, and a library
has piles and piles of books that are only checked out every few years
(reference manuals for Apple II word processors, etc.)
>However, having just gotten off the phone with Dr. BobHe "feels" that?
>Gorman, let me say that Dr. Gorman feels this
>would not change the outcome significantly.
>(That is why he did not suggest using it in the firstI'm not the one trying to prove that the "Christiani" were zealots.
>place.) If you want to go ahead and do the
>calculation anyway, you will see this.
>Actually, there is no specific reference to theThis is baloney.
>Zipf distribution function in statistical literature,
>because the term did not originate within theOh? Did Dr. Gorman tell you that? Zipf's law, named after the Harvard
>profession of statistics. It originated, I believe,
>with a philologist. All it is is a multinomial
linguistic professor George Kingsley Zipf (1902-1950), is the observation
that frequency of occurrence of some event ( P ), as a function of the rank
( i) when the rank is determined by the above frequency of occurrence, is a
power-law function P(i) ~ 1/i**a with the exponent (shown by **) "a" close
to unity. Note that the tilde ~ indicates "approximately equal to". The
most famous example of Zipf's law is the frequency of English words. For an
example, see the distribution at
suffers (as here) from an attempt to show a formula involving exponents in
a text medium that does not show exponents clearly. Nevertheless, the site
shows word distributions in an English text sample.
Some bibliographic references:
Fedorowicz, J., "The Theoretical Foundation of Zipf"s Law and Its
the Bibliographic Database Environment", Journal of the American Society for
Information Science, Volume 33, Number 5, 1982
Wyllys, R.E., "Empirical and theoretical bases of Zipf"s law", Library Trends,
Volume 3, Number 1, 1981
Zipf, G. K. (1949), Human Behaviour and the Principle of Least Effort,
Addison-Wesley Publishing Company: Cambridge, Massachusetts.
Instructions on how to apply Zipf's law to word frequencies may be found at
You might also find useful:
G. R. Turner (1997) Relationship Between Vocabulary, Text Length and Zipf's
Law. Available online at:
I think your case would be considerably strengthened if you had available a
frequency analysis of latin words in the extant words of Tacitus, and used
the Zipf distribution to test your hypothesis. Then maybe more people would
find your case persuasive.
[Non-text portions of this message have been removed]