[b-hebrew] Zipf and B-Hebrew

biblical hebrew jcr.bhebrew at gmail.com
Sun Oct 14 06:28:10 EDT 2007

Hi Pere,

I think you have misunderstood my summary of Zipf's suggestion. He is not
merely suggesting that all languages have unambiguous words and ambiguous
words. He is stating that the combination of forces produced by the needs of
the hearer and the speaker produce the tendency for common words to have
multiple senses/meanings and for rare words to have specific unambiguous

He further suggests that there is a curve exhibited with common words were
common words have multiple meanings and more common words have more meanings
and even more common words have even more possible meanings. If you still
don't understand his suggestion please be more specific about which part/s
you didn't understand so that I can attempt to explain his suggestion

To illustrate the point let's take an example you suggested. You suggest
that hebrew )B has only one sense 'father'. Now while it is true to say that
)B in all of its instances can be translated with the single word 'father'
that does not prove that it has only one sense as the English word 'father'
can have many senses and the Hebrew word )B can have senses that are not
characteristic of English usage. e.g:

1) We can say that Abraham was Isaac's 'father'  or  Isaac's  )B
2) We can say that Abraham was Jacob's )B but not that he was his 'father'
3) We can say that Yhwh is our 'father' or )B - note that this is a
different sense to those above.
4) In a spiritual sense we can say that Abraham is the 'father' or )B of non
Isrealites who have entered into a friendly covenant relationship with Yhwh

These observations seem to be in harmony with Zipf's suggestion as )B in
Hebrew and 'father' in English are common words.

Contrast this with Hebrew 'tsohar' which is only attested once in the Hebrew
canon. Even though we are not entirely sure what a 'tsohar' is, we are
fairly sure that the word described a specific and unambiguous concept.

My gut feeling up to now is that Zipf may be right and what I have seen up
to now reinforces his ideas. What I was wondering was what the list members
opinions of his suggestions may be with respect to the Hebrew canon in light
of the wealth of experience with B Hebrew that the list members have. Do
people agree that the data uphold his suggestions? Can anyone think of any
concrete examples of low frequency words with multiple senses that would
contradict his suggestion? Conversely, can anyone think of any high
frequency words with only one specific sense that would contradict his



James Christian Read - BSc Computer Science

http://www.lamie.org/hebrew       Thesis 1 - Aleppo codex machine translation
http://www.lamie.org/lad-sim.doc  Thesis 2 - language acquisition simulation


On 10/14/07, pporta at oham.net <pporta at oham.net> wrote:
> > I was wondering what people's opinions about one of Zipf's  ideas  are.
> > Zipf, best  known  for  Zipf's law, suggests  that  there is a playoff
> > between ease of speaking and ease of understanding that forces the
> > distribution of words which have single unambiguous meanings and words
> > that
> > have multiple meanings discerned from context. That is to say that the
> > optimal situation for the speaker would be that language consisted of
> one
> > word which could express every possible meaning while the optimal
> > situation
> > for the hearer would be that every word had a unique meaning so that no
> > ambiguities need be resolved. He further suggests that these conflicting
> > interests create linguistic forces which cause any given language to, in
> a
> > manner of compromise between the needs of the speaker and of the hearer,
> > have a mix of unambiguous words and words with multiple meanings with
> the
> > rarest words typically having a very specific and unambiguous meaning
> and
> > the most common words having the most possible semantic meanings
> > disambiguated by context.
> >
> > I was wondering what the list members feelings about Zipf's proposal are
> > with respect to b-hebrew. Are his ideas supported by the evidence in the
> > b-hebrew corpus? Are there any examples which contradict his proposal?
> >
> > Thanks in advance for any help.
> >
> ______________
> I'm afraid you should be more precise if you would like to get some
> answers
> to your question. Every human language has words with a single meaning and
> words with two, three... meanings. I think the true is that this fact is
> quite independent from speaker's and hearer's will: that is why the
> context
> is so necessary to make the meaning of a given word clear and unambiguous.
> Let us consider
> 1. the noun )B:  it means "father" everywhere in the Bible.
> 2. the noun N"BEL: --it means "skin-bottle" in 1Sa 1:24 and --it means "a
> musical instrument" in 1Sa 10:5
> 3. the verb XFBAL: --it means "to bind" in Ez 18:16 and  --it means "to
> act
> corruptly" in Ne 1:7
> And so on.
> And this applies, I feel, to every human language: English, Russian,
> French... and Hebrew.
> Perhaps you should make Zipf's law more understandable so that list
> members
> may show their thoughts.
> Pere Porta
> Barcelona (Spain)

More information about the b-hebrew mailing list