Biblical Hebrew Forum

[ianyoung AT plato.arts.usyd.edu.au]

On Sun, 29 Nov 1998 19:47:01 -0500 (EST) rhendel AT mcs.drexel.edu (Russell
Hendel) wrote:

>I am responding to Ian Youngs discussion of his study on AM in the Bible.
>I am a mathematician by nature and hold an actuarial degree--therefore
>I FOCUS ON THE PURELY STATISTICAL ISSUES RAISED or ASSUMED BY PROFESSOR
>YOUNG.
>
>1st) Rather than bore you with mathematical jargon let me give a CLEAR
>and LUCID analogy: Suppose 39 professors have a silver dollar. Some
>professors flip it 10-30 times while other professors flip it a few
>100 times. We then tabulate the number of heads and compute the percentages.
>
>We expect on the average to get 50% heads. But it would not be surprising
>or contradict any law of statisitics if there was wide variation. In
>fact the following statistical laws may surprise you
>---if you flip the coin 10,000 times then you should expect only 2%
>variation from the average of 50 (that is to say if 100 people flipped
>the coin, each person flipping it 10,000 times then 95% of these people
>would have between 4900 and 5100 heads (a deviation of 100/5000 or 2 %)
>
>---if however you flip the coin only 100 times then it is very reasonable
>to expect a 20% variation (This is a mathematical theorem).
>
>---if tiy flip the coin only 16 times it is reasonable to expect a 25%
>variation.
>
>2nd) To clinch the analogy we let correspond
>       39 professors = 39 books of the bible
>       Coin flips    = Occurences of the word AM
>       Head / Tail   = Singular Plural
>
>Note that while we expect 50% heads I don't know what the average
>is for singular plural
>
>The POINT I AM MAKING is that a 20% variation in plurality is PERFECTLY
>consistent with ONE UNIFORM LAW if the number of cases per book is
>under 100.
>
>3rd) I would further question whether Professor Young had the assistance
>of a statistician--it is not uncommon for people to make simple inferences
>without at least consulting one in many fields (e.g. Social studies,
>psyhcology etc)_
>
>4th) Professor Young's assertion that conclusions cannot be reached because
>of this variation is simply incorrect. Again let us go back to the head
>flipping analogy. Suppose while our 39 professors are flipping coins their
>wives flip say quarters. Further suppose that we have less than 100 flips
>(so we expect 20% variation).
>
>But suppose their wives achieved heads ONLY when their husbands did and
>their wives achieved tails only when their husbands did. Then we would
>have discovered (a most peculiar) law--we could be certain that the
>wives coins and husbands coins were linked.
>
>In mathematical terminology professor young should not be studying the
>variation in a single variable (plurality) but the CORRELATION between
>2 variables--say Plurality and meaning.
>
>I am NOT saying I know for a fact that there is a law...but e.g. the
>explanation I gave this morning based on other work I have seen that
>       * Singular AM denotes ACTING AS ONE
>       * Plural AM denotes ACTING IN FACTIONS
>may (or may not) correlate with usage (As professor young would quickly
>point out it may not be clear in each case of AM whether they are acting
>as a unit or not).
>
>I assure the other readers of this list that the above analysis could
>probably be carried out on many similar works. I am therefore very interested
>if other members of the list could volunteer similar conclusions that
>they have reached and give details.
>
>I am not belittling in any way the work already done by professor young.
>In fact, with a little more data and some easily performed tests we
>might be able to reach some interesting conclusions--some of which might
>be new.
>
>There is a tendency on many professions to avoid mathematics..I do hope
>the readers of this list are not that way..I believe alot could be
>accomplished and I would be willing to shed light on what to look for
>and the types of results/tests needed
>
>I look forward to some interesting responses
>
>Russell Jay Hendel; pHD ASA RHEndel @ mcs drexel edu


I would like to thank Prof Hendel for his comments above, and separately
regarding Judges 9:36-37 regarding collective nouns.  In response, I would
like to clarify a few points.

(1)     I don't believe we are dealing with enough data for statistics to play
an important role in the discussion, and I don't wish to give the idea that my
article claims anything on the basis of statistics.  In fact, I criticise
others for their inappropriate use of statistics

(2)     Once again, let me emphasise that my main reservation in this sort of
study is that the attempt to explain the behaviour of `am in ancient Hebrew is
done on the sole basis of the Masoretic Text.  I believe there is a serious
question whether the current distribution of singular/plural verbs with `am in
the Masoretic Text reflects adequately the "original" distribution of those
forms deriving from the authors the biblical books.  The figure of 20% in my
original communication related to the proportion of variations from the
Masoretic Text's distribution of singular/plural verbs with `am in the textual
evidence (by which I mean Qumran etc), not to the variations in the
proportions of singular/plural verbs in various books of the MT itself - here
there can be 100% variation ie. some books have only singular, others only
plural.  Nevertheless, I would agree, with such a relatively small amount of
data (no more than 76 in any book), we cannot argue definite conclusions even
from these distribution statistics.

The textual evidence, in any case, means (to use Prof Hendel's image) that we
see quite a few of the coins flipped simultaneously both heads and tails ie
the very accuracy of the basic data is called into question.

(3)     I would like Prof Hendel to clarify his own position in regard to the
phenomena in the Masoretic Text.  Do you consider all verbal behaviour with a
noun like `am to be predictable and subject to the strict rules?  Or is it
merely the toss of coin?  Even if the decisive factor is whether the `am is
conceived as a unity or in parts (an entirely reasonable suggestion) is this a
matter of strict rules or does a measure of artistic choice come into the
matter?

My main interest in that particular section of my article was to demonstrate
that we cannot explain the distribution in the Masoretic Text of the Hebrew
Bible of singular/plural verbs solely as the product of mechanical rules.
Would you agree with this proposition?  

(4)     I must say that it is in my nature to feel a little sceptical about
the confidence expressed in explaining phenomena like the parallel sentences
in Judges 9:36-37.  It seems too much to me like working back from the answer
to the question ie. we know why singular/plural nouns are each chosen to be
with `am, therefore this must be the reason for the variation here.

This isn't to say the suggestion isn't right - just unprovable.



Ian Young
Semitic Studies
University of Sydney