Biblical Hebrew Forum

[Will Parsons <wbparsons AT alum.mit.edu>]

Hi Karl,

On Wed, 5 Dec 2012 16:15:15 -0800, K Randolph <kwrandolph AT gmail.com> wrote:
> Will:
> 
> On Wed, Dec 5, 2012 at 2:13 PM, Will Parsons <wbparsons AT alum.mit.edu> wrote:
>>
>> As far as the position of xi in the alphabet corresponding to samekh,
>> I think that *is* suggestive, but doesn't necessarily point to Hebrew
>> (or really Phoenician) samekh having a [ks] phoneme.  More likely, the
>> Greek sibilant phoneme(s) didn't match completely with the various
>> Phoenician sibilants, and the Phoenician samekh struck the Greek ear
>> as something whose nearest equivalent was like Greek /ks/.  (And
>> apparently Phoenician shin/sin was the best fit for Greek /s/.)
>>
>> "Artaxerxes" is interesting, but it seems to me it has to be
>> considered in conjunction with the even more famous Persian king named
>> "Xerxes".  It's hard to look at these two Greek forms without thinking
>> that "Artaxerxes" is an expanded form of "Xerxes", but the Persian
>> forms they're based on, "Artakhshaça" and "Khshayarsha", are not so
>> strikingly similar.  Of course the khsh (i.e., [xʃ]) sequence would be
>> represented in Greek by xi, but it's interesting to note that the
>> second [ʃ] in "Khshayarsha" was also represented by Greek xi.  I'm not
>> sure what a "good" Greek rendering of "Artakhshaça" would be, but I
>> can't help but think the actual form "Artaxerxes" has been assimilated
>> in form somewhat to the more famous "Xerxes".
> 
> We’re speculating here, and are we even sure of Persian period
> pronunciation? The Greek transliterations are suggestive, and may represent
> an older pronunciation than what has come down to us through Persian
> sources.

We can't be sure of the pronunciation, but I would certainly trust the
Persian transcriptions of their own names to be better than a
foreigner's renditions in another language.

>> > The Greeks got their alphabet from the Phoenicians, and admitted to that.
>> > The Phoenicians from the Hebrews, so I’m assuming that in the borrowing
>> > that the phonemes stayed the same, at least at first.
>>
>> I don't think that's a good assumption.
> 
> I think there’s a difference between adoption of an alphabet by a
> government appointed committee, and the adoption of one by the man on the
> street. And the Greek adoption of the alphabet mirrors more the adoption on
> the street variety.

I think your distinction is valid.  In ancient times, the adoption of
an alphabet was more likely to be ad hoc, though one can point to the
examples of Cyril (for Cyrillic, or possibly Glagolitic) and Wulfila
(for Gothic), where alphabets were carefully designed (though not by
committee), rather than creations of the "man on the street".

> The Greek language had fewer consonants, so the man on the street writing
> phonemically repurposed a few of the letters that had similar to vowel
> pronunciation as vowels (I suspect that those “ gutturals” didn’t have
> as guttural a pronunciation during Biblical times as later, when
> pronunciation changed under Aramaic influence), but that took probably a
> generation or three, and rather than changing the pronunciation of received
> letters, dropped consonants for which there were no Greek phonemes, and
> added consonants for consonantal phonemes not found in Hebrew or Phoenician.

Certainly the Phoenician alphabet underwent a considerable re-shaping
to make it a good vehicle for writing Greek, since Greek was quite
different from Phoenician.  But that's the point.  Just because xi
represents a cluster [ks] in Greek doesn't mean the Phoenician letter
from which it was borrowed represents a cluster.

(Incidentally, the creation of the Greek alphabet may not be entirely
due to "the man on the street".  It very well could be that some of
the adaptions, such as the remarkable repurposing of consonantal
letters as vowels, was the brain-child of an ancient equivelent of
Cyril or Wulfila, one whose name has been lost to history.)

> (Adoption by committee seems more willing to repurpose letters to fit the
> phonemes of the adopter language than as in ancient Greece to drop and add
> letters so that the phonemes represented by the letters remain the same or
> as close to the same as possible.)
> 
>>  I think it's safe to assume
>> that when the Greeks got their alphabet from the Phoenicians, they used
>> the Phoenician letters with values nearest to their own, but having
>> quite different phonologies, the correspondence between the sounds in
>> the two languages could be quite approximate.  (And of course, the
>> Greeks were quite creative in re-defining some Phoenician guttural
>> consonant letters as vowels.)
>>
>> Apart from what I've written above, I see as a more fundamental
>> problem with a consonant cluster like [ks] acting as a single phoneme
>> (and hence being represented by a single letter) in Hebrew (or other
>> Semitic languages).  If samekh *did* represent a cluster, then I would
>> expect to see at least some instances where samekh was used in words
>> where /k/ and /s/ as separate sounds happened to fall together, i.e.,
>> a parallel to Greek νυξ/nyx vs νυκτες/nyktes.
> 
> Why? I see no reason that would be the case. Just because it was found in
> Greek doesn’t mean that it should be found in other languages. I don’t know
> where that is found in any language other than Greek.

This doesn't really have anything to do with the Greek language, but
with Greek spelling.  There's nothing in Greek that requires xi (or
psi) to exist, and the fact that they are used is a peculiarity of
Greek orthography, without any deep significance.

For an example other than Greek, look at Coptic.  Coptic uses the
Greek alphabet supplemented with additional letters for sounds not
found in Greek, but Coptic phonology is different from Greek in many
particulars.  The Greek letters Φ/phi, Θ/theta, Χ/khi were originally
used in Greek for aspirated stop phonemes.  Coptic apparently did not
have aspirated stops, but it did have consonant clusters that could
include an /h/ phoneme.  Not surprisingly, phi, &c. were used in the
numerous Greek loanwords in Coptic, but also, perhaps more
surprisingly, in native Coptic words to represent a consonant
cluster.  For example:

/p/   = masculine definite article
/ho/  = face
/pho/ = "the face", spelled ΦΟ, with a phi.

This is parallel to the Greek use of xi and psi, but (and this is my
point) we do not see anything similar in Phoenician or Hebrew for
samekh.


> Do you find that with Tsada, another “consonant cluster”?

As a matter of fact, I do, though with some hesitancy.

I am somewhat hesitant because:

1)  Although I am unaware of any language (including Greek) where a
    consonantal cluster like [ks] functions a single phoneme, there
    are many languages where [ts] (or similar) function as single
    phonemes, including modern pronunciations of Hebrew.  The
    difference between [ts] and [ks] is whether the two components
    have the same point of articulation, which allows the two elements
    of [ts] to be more closely bound together as a unit than in the
    case of [ks].

2)  Apparently, many scholars reconstruct affricate phonemes such as
    [ts] for Proto-Semitic.  On one hand, I am reluctant to dismiss
    conclusions of those who are probably more knowledgeable than I
    am, and whose reasons for coming to these conclusions I am
    ignorant of.  On the other hand, I am also reluctant to simply
    accept these without understanding the basis for those
    conclusions.  (Of course, conclusions regarding Proto-Semitic
    would not necessarily apply to Hebrew in any case.)

My own reasons for doubting that (t)sadhe represented an affricate in
the ancient pronunciation of Hebrew (despite my qualms) are indeed
similar to why I see samekh as representing [ks] unlikely.  In
particular, if (t)sadhe represented something like [ts], then I would
expect to see (e.g.) hithpa`el forms where /hit+s.../ become
/hi[ts].../ (with a צ) rather than /hist.../ (with metathesis of ת and
a sibilant).

-- 
Will Parsons
μη φαινεσθαι, αλλ' ειναι.