Biblical Hebrew Forum

[Peter Kirk <peter AT qaya.org>]

On 09/02/2006 07:46, Yitzhak Sapir wrote:

> ...  Perhaps the words you cite remained the same, but at least
> "colour"'s spelling/pronunciation may be related to the GVS -- so that the
> variant spelling convention does indicate an attempt to modernize after a
> sound change.
>
>   
I deliberately chose words whose Shakespearean, modern British English, 
and American English pronunciations are very similar. The change from 
"colour" to "color" promoted by Noah Webster etc for American English 
was unrelated to pronunciation, motivated only by a desire for 
simplification and to make American English distinctive
> > there is indeed a possibility of a pronunciation difference (but then there is
> > that even if the spelling is unchanged) but also a very good chance that
> > the pronunciation was unchanged.
> >     
>
> How do you measure this "good chance"?  Or is it just a possibility that you
> would like to think is a good chance?
>   

My point is that although there may have been a pronunciation change 
there is no evidence for this in the spelling. This whole thread is 
about my insistence on evidence for your hypothesis of major 
pronunciation changes between inscriptional and Torah Hebrew. Yes, we 
agree that it may have happened, even that it was quite probable. But so 
far no evidence for this has been produced which cannot be explained 
simply as a change of spelling convention.
>   
> ...
> Since sound changes happen all the time, ...

Yes.

> ... and the spelling discloses
> that there was most likely a difference in time between the spelling
> convention of pre-exilic inscriptions and of the Bible in its current
> consonantal spelling, we could similarly theorize that the
> pronunciation changed, simply because it always does over time
> (to varying degrees, of course). ...

OK, but these pronunciation changes may be quite independent of the 
purely conventional spelling changes.

> ... We have variant spellings of "day"
> in the inscriptions: ym and ywm.  Under such circumstances, we
> might expect that final -o or -aw, even if written with an -h, might
> sometimes be written as an -w by a less skilled scribe.  But it's
> pretty widely agreed that it isn't. ...

Well, not agreed by Karl, and I thought he had quoted some scholars who 
hold that the -w in the Siloam inscription is a singular possessive.

> ... So if you want to say that spelling
> is simply not a good indicator of pronunciation ever, and we can't
> learn much about pronunciation from spelling, because the spelling
> may hide innovations, you will be left with the simple conclusion
> that there probably were sound changes between the stage
> represented by pre-exilic inscriptions and the stage represented by
> the Massoretic Text, but we simply can't figure out what they were.
>   

Yes. Of course this begs a lot of questions about what you mean by "the 
stage represented by the Massoretic Text". I presume we are not talking 
about the Masoretic pronunciation here, for that is known to be rather 
different from the pronunciation in say the Persian period more than 
1000 years earlier, and some of the intervening changes (e.g. a > i in 
closed first syllables, loss of distinction between the two ayins and 
the two hets) can be determined e.g. from transliterations in LXX. But 
if we are talking about changes in the little more than a century from 
Josiah to Ezra, while we can agree that very likely there were some, we 
have very little evidence for what they were.
>   
> ...
> Or that they were not spelled with any letter in the end.  That is, that
> -o was simply not written in the pre-exilic inscriptions.  In any case,
> beyond the common cases of masculine possessive forms, what
> other cases can you think of?  I tried to think of other examples but
> right now, they don't come to mind, and you say that they are very
> common.
>   

Well, masculine possessives are very common, but of course they are the 
main point under discussion. Other words ending in -o are not so common, 
but we do have some names like Shilo and Shlomo which we have mentioned, 
and in fact quite a number of other proper names in the Hebrew Bible.

It certainly seems to have become the practice in at least later 
inscriptional Hebrew not to leave word final vowels unmarked, so it is 
hardly likely that word final -o was an exception. But it is I suppose 
possible that in the corpus of inscriptions there are no examples of 
word final -o, apart from the masculine possessives under discussion.
>   
> > > I don't know how well we know that -h can represent -o^, and
> > > whether it is not based on the assumption that -h in pre-exilic
> > > inscriptions was pronounced -o^ rather than as a consonant.
> > > Sarfatti specifically mentions that -h came to represent -o^
> > > because the above change (-ahu > -au > -o^) was not reflected
> > > in the orthography.
> > >       
> > This seems improbable to me unless some explanation can be given for how
> > -h came to represent -a and -e. The simplest hypothesis is that -h was
> > first written as a mater lectionis for any final vowel, or at least for
> > -a, -e and -o. The suggestion that its use for -o came about by a quite
> > different mechanism from its use for -a and -e is improbable by Occam's
> > Razor.
> >     
>
> My impression, based on the similar use of the variant "he"-based sign for
> feminine words in Arabic, is that -ah was originally consonantal as well, ...

Well, maybe, but given the remarkable similarity between Hebrew and 
Arabic conventions of pronouncing -a in the absolute and -at in the 
construct or possessed case, this common feature probably goes back to 
proto-NW Semitic, which is well before the surviving inscriptional 
Hebrew. Also the -t in the construct and in the Hebrew -et variant 
forms, plus the use of -t as a feminine marker even in languages as 
distant as Egyptian, strongly suggests that the original consonantal 
sound, if there was one, was -t rather than -h.

> ... or
> else why would Arabic not base its variant on the Alef? ...

I am not aware of the history of Arabic use of matres lectionis, but the 
use of alef for long a and he (to which two dots were later added) for 
final -a suggests to me that the Arabic conventions are borrowed from 
post-biblical Hebrew and perhaps wider use in the region, in Hebrew, 
Aramaic and Syriac, during the early Christian era. I note also that in 
Arabic the final -a in feminine nouns is short, and so would not be 
written with alef which in Arabic is a marker of a long -a.

> ... Note also the use
> of final Alef in various names of Biblical persons from pre-exilic times that
> are transliterated in foreign inscriptions with an -a. ...

If we are now talking about Hebrew transliterated into Akkadian, in the 
Hebrew of that period alef represented a clearly pronounced glottal 
stop, but this is not clearly represented in Akkadian, or in English, 
because these languages lack the glottal stop. (Well, in fact glottal 
stops are very common in the colloquial English of my area, but that's 
another matter.) So there are two distinct classes of Hebrew words which 
should not be confused: those ending in alef, a glottal stop, and those 
ending in he, mostly representing a word final vowel. Alef was used for 
word final -a in Aramaic and in DSS Hebrew, but not I think in biblical 
Hebrew.

> ... This leaves us with -h
> for -e only, and while I'm not sure right now how widely this is used, it
> cannot be used to claim that -h was the default "final vowel" indicator.
>
>   
It is very clear that final he was used in inscriptional Hebrew also for 
words with final -a, at least for words which are pronounced with final 
-a both in later Hebrew and in Arabic and Aramaic indicating that this 
was also the proto-NW Semitic pronunciation. So, while we can't be sure 
of the pre-exilic pronunciation, it is reasonable to assume that at 
least some final he's indicate a final -a pronunciation. Also, surely 
there are Hebrew names ending in he and now pronounced -a, such as 
Josiah, which are represented in Akkadian inscriptions; are these 
written with -e? Or are they always written with transcriptions of full 
-yahu endings?
> > I would not call a change in spelling conventions a translation at all,
> > for there is no change in the spoken form.
> >     
>
> Read again the article in the Cambridge History.
>   

Do I have to spell out yet again what should have been obvious? There is 
no change in the spoken form of very many of the words whose spelling 
has changed. The pronunciation changes for which there is evidence are 
largely independent of the spelling changes and not reflected in modern 
English spelling. That is why long vowels in English are pronounced 
quite differently from in almost every other Latin script based language.

> ...
> It's one more indication of the passage of time.  That the different
> spelling convention is later and is part of a whole bunch of changes
> that reflected that passage in time.  One of those changes would
> be pronunciation.
>   

In fact the changes of spelling convention in Shakespeare I referred to 
are not later but earlier. "Liue" became "live" and "Iuliet" became 
"Juliet" already in the 17th century, whereas many of the pronunciation 
changes mentioned in your Cambridge History article are from the 18th 
century. This is simply not a coherent package of linked changes, rather 
the spelling changes are independent of the pronunciation changes.
>   
> ...
> > My point about the loan words was not that they are a parallel with the
> > Hebrew but to correct your error that Shakespeare would have pronounced
> > "Iuliet" with an initial I or Y sound, rather than J.
> >     
>
> When did I ever say that?
>   

In reply to my
> Yet it
> > seems to me that by your method of reasoning the change of spelling
> > implies that "Iuliet" must in fact have been pronounced "eye-oo-li-et".
you wrote
> It is a fact that the pronunciation was different:
> http://www.bartleby.com/224/1502.html
Perhaps you meant to say that the general pronunciation of English was 
different, as shown in this link but irrelevantly to my particular 
point. But what I thought you were actually claiming was that 
Shakespeare's pronunciation of "Iuliet" was different from the modern 
pronunciation of "Juliet", and commenting positively on my 
"eye-oo-li-et". But his pronunciation was not different, or only very 
subtly so.

>   
> > That's not what I was saying. What I was suggesting was that pre-exilic
> > Hebrew, like almost any language, was not monolithic in its lexicon or
> > orthography, and so one would expect some variation between different
> > sources and different registers of language without interpreting this
> > variation as necessary evidence that they were composed at different times.
> >     
>
> What it means, though, is that you accept that the epigraphic evidence
> suggests 1) different spelling convention and 2) probably a slightly
> different lexicon.  In essence, that the evidence does imply what I am
> saying.  All you are saying is that perhaps there are reasons why we
> should ignore this evidence -- for example, dialectical differences which
> would require us to look at each inscription separately from all others.
> Even if we found a whole library of late pre-exilic documents, this
> wouldn't prove anything because perhaps there were dialectical
> differences between this library and the Hebrew Bible that were in
> existence already in this time.  Well, yes, perhaps, but now your
> position is one of saying that the evidence says one thing but you
> say another, and that's not a very good position on which to stand.
>
>   
Well, your original claim was that the differences of spelling 
convention, to which we have added some subtle lexical differences, 
proves that there was a time difference between inscriptional and 
biblical Hebrew. But such differences can easily by synchronic i.e. 
dialectal rather than diachronic. If you want to demonstrate that the 
differences are diachronic, you need to provide some evidence, rather 
than assert that a difference implies a time gap.

The kind of evidence which you might be able to provide is variant forms 
of the same text from before and after the Exile. Yes, they exist - 
there is a pre-exilic version of the priestly blessing - or has this 
been rejected as a forgery? And then of course there are the parallel 
passages between the originally probably pre-exilic Samuel and Kings and 
the post-exilic Chronicles. There are of course a number of complex 
analytical issues here, but it should be possible to demonstrate from 
these to some extent how the language changed from before to after the 
Exile.

--
Peter Kirk
peter AT qaya.org (personal)
peterkirk AT qaya.org (work)
http://www.qaya.org/