[biblitfonts] Re: Fwd: RE: meteg positioning

Bob,
Thanks for the helpful note.  I have forgotten so many of these details
since it has been so long ago that I taught the Hebrew Bible text critical
seminar.
Kent

Kent Richards
Executive Director
Professor Of Old Testament

kent.richards@xxxxxxxxxxxx
www.sbl-site.org
404.727.3038

The Luce Center
Emory University
825 Houston Mill Road  Suite 350
Atlanta, Georgia 30329

-----Original Message-----
From: biblitfonts-bounce@xxxxxxxxxxxxx
[mailto:biblitfonts-bounce@xxxxxxxxxxxxx]On Behalf Of Bob Buller
Sent: Friday, December 13, 2002 5:04 PM
To: biblitfonts@xxxxxxxxxxxxx
Subject: [biblitfonts] Re: Fwd: RE: meteg positioning

John,

Just to add a little more complexity to the historical background of the
metheg issue. I don't have a full copy of BHK, but the explanation you
reference here and quote in a later email makes sense. I do note that
BHS abandoned the addition of metheg, noting "The addition of Silluq,
which is occasionally lacking, and particularly of Metheg, which is
often omitted, has been discontinued, particularly as in L itself Metheg
is found both to the left and to the right of the vowel pointing, and
Silluq may also appear to the right" (xii).

According to Yeivin, "Ga'ya [the older term for what we know as metheg]
is generally written to the left of a vowel sign marked under the same
letter. In some MSS, such as A and L, this convention is carefully
maintained, with very few exceptions--and those usually due to
correction, or to lack of space in the regular position. In other MSS,
such as C, S, S1, ga'ya is often written to the right of the vowel sign,
without any particular reason. With shewa also, ga'ya is generally
written to the left of the shewa sign, but there are MSS in which it is
often written to the right. The same is generally true of hatef shewa
signs, but in some MSS, such as L and S1, the ga'ya sign may be written
between the two parts of the hatef sign; [example with hatef patah].
This does not occur in A and C" (§314).

I have no idea how this might affect what you're doing with the font
(it's way beyond me), just thought that we should be aware that metheg
will appear in various MSS that scholars might need to reproduce in all
three positions. In short, its placement to the right or in the middle
is not always artificial.

Best,
Bob

John Hudson wrote:
>
> FYYFI
>
> >X-Sieve: cmu-sieve 2.0
> >Subject: RE: meteg positioning
> >Date: Thu, 12 Dec 2002 09:55:32 -0800
> >X-MS-Has-Attach:
> >X-MS-TNEF-Correlator:
> >Thread-Topic: meteg positioning
> >Thread-Index: AcKh+Y4GLma/4fkWR8uoTSuw6WuFRgACJV9Q
> >From: "Paul Nelson" <paulnel@xxxxxxxxxxxxxxxxxxx>
> >To: "John Hudson" <tiro@xxxxxxxx>,
> >         <Peter_Constable@xxxxxxx>
> >Cc: <Joan_Wardell@xxxxxxx>,
> >         <eevans@xxxxxxxxxxxx>
> >X-OriginalArrivalTime: 12 Dec 2002 17:55:32.0373 (UTC)
> >FILETIME=[AA32EC50:01C2A207]
> >X-Scanner: OK. Scanned in 0.02 seconds.
> >
> >I am looking at the following as the best solution after talking with a
> >typesetter in Jerusalem.
> >
> >Logical stream -> displayed
> >Meteg + (05B1..05B3) -> meteg on right of nikud
> >
> >(05B1..05B3) + Meteg -> meteg on left of nikud
> >
> >(05B1..05B3) + ZWJ + Meteg -> form ligated glyph
> >
> >Now, if you have data in NFC form, you will only be able to get the last
> >two on the list because meteg *always* comes after (05B1..05B3) when NFC
> >is applied to the data.
> >
> >I understand by default that the meteg should always form on the left of
> >the nikud. The joined form is found in a few places of the original text
> >and should not be default. Therefore it is best that the ZWJ be used to
> >force the ligating behavior as defined by Unicode.
> >
> >There are some very interesting things that are making me to loosen the
> >rules and recommending the use of 'mkmk' in some cases to correct
> >positioning.
> >
> >My goal is not to limit the amazing ambiguous combining of vowels, but
> >to allow a person who has an enormous amount of time and willingness to
> >deal with all of the inconsistencies to make a font that will be
> >acceptable.
> >
> >Somehow font vendors need to have a clear understanding of what to
> >expect from text being input so they can make fonts that work without
> >having to have hundreds of exception tables.
> >
> >I am working on a font specification that will be the basis for the
> >Microsoft fonts to do a reasonable attempt at rendering Biblical text. I
> >know that it will not be 100% correct, but it will be reasonable.
> >
> >Regards,
> >
> >Paul
> >
> >
> >
> >-----Original Message-----
> >From: John Hudson [mailto:tiro@xxxxxxxx]
> >Sent: Thursday, December 12, 2002 8:14 AM
> >To: Peter_Constable@xxxxxxx
> >Cc: Paul Nelson; Joan_Wardell@xxxxxxx; eevans@xxxxxxxxxxxx
> >Subject: Re: meteg positioning
> >
> >
> >Dear Peter,
> >
> >I'm cc'ing this to Eli Evans at Libronix, since he and I were discussing
> >
> >this only a couple of days ago.
> >
> > >I'm wondering how you guys are dealing or planning to deal with
> > >alternate positioning of meteg.
> > >
> > >Here's the issue: In the BHS text, meteg can occur in different
> > >relative positions with vowel points. For the hataf vowels
> > >(05B1..05B3), which have two parts, the meteg can occur in all three
> > >positions (L, R and between the two parts).
> >
> >My understanding from Eli is that this is an idiosyncratic typographical
> >
> >convention of BHK/BHS, the purpose of which was to provide implied meteg
> >
> >'missing' from original manuscripts, while indicating that they are
> >editorial editions. So when the meteg on one side -- I can't remember
> >which
> >side -- it means the meteg is original, and when it appears on the other
> >
> >side it means it is an editorial edition. [Eli, please corrent me if I
> >have
> >this wrong.] So the first thing to realise is that this is not, per se,
> >a
> >Hebrew processing issue: it is a German typographical issue.
> >
> >I'm now wondering if this textual peculiarity could be addressed by
> >insertion ZWNJ between meteg and the vowel? Obviously some contextual
> >GPOS
> >gymnastics will be required to get the desired results, but we should
> >remember that this is an issue that only needs to be addressed in fonts
> >for
> >BHS text. [It would be worthwhile to ask the GBS if they intend to
> >employ
> >the same technique in the BHQ text.] The big issue then becomes whether
> >this will produce problems for the new mark re-ordering implemented in
> >Uniscribe. Paul, what would happen in the following circumstances:
> >
> >          Consonant + hataf segol + ZWNJ + meteg
> >          Consonant + meteg + ZWNJ + hataf segol
> >
> >Is there any way that this kind of approach could be used to force
> >display
> >of meteg to the right or left of the vowel?
> >
> >Regards, John
> >
> >
> > >For other, single-part vowels, it can occur on either
> > >side. Some examples:
> > >
> > >- 2 Chron 6:8:12 < 05D4  05B1, 05BD ... > = he hataf segol + medial
> > >meteg
> > >
> > >- Ps 50:21:5 < 05D4  05B1, 05BD ... > = he hataf segol + left meteg
> > >
> > >- Ps 85:7:1 < 05D4  05B2, 05BD ... > = he hataf patah + right meteg
> > >
> > >Some positions are far more common than others: the most common
> > >positionings are left of non hataf vowels, and medial with hataf
> > >vowels. Here's the results of a bit of research Joan did:
> > >
> > >Vowel        L(75) R(95) medial(35) (CCAT encoding number in parens)
> > >-----------------------------
> > >sheva         615    11     0
> > >hataf segol     2     0    28
> > >hataf patah     2     0     0
> > >hataf qamats    2     2    50
> > >hiriq        7318    27   N/A
> > >tsere        4377    38   N/A
> > >segol        4741    59   N/A
> > >patah        3933   412   N/A
> > >qamats     11,427   184   N/A
> > >holam        6581     9   N/A
> > >qubuts        132     1   N/A
> > >just dagesh   118   159   N/A    This is shureq (vav dagesh) or any
> > >consonant-dagesh preceding meteg.
> > >consonant    1003     7   N/A
> > >
> > >Included above is:
> > >between 2
> > >vowels       144      0     0     These are all "LFIM" combinations of
> > >Jerusalem (qamats-meteg-hiriq)
> > >
> > >For the single-part vowels, alternate positioning might in principle
> > >have been handled in terms of encoding order. Because these things are
> > >in different Unicode combining classes, though, the different relative
> > >ordering of encoded characters are canonically equivalent and cannot be
> >
> > >used to make a semantic or presentational distinction. Even if that
> > >weren't the case, only two relative orders are possible but three
> > >positions are involved with the hataf vowels.
> > >
> > >Have either of you given any thought to this particular problem?
> > >
> > >
> > >
> > >Peter
> >
> >
> >Tiro Typeworks          www.tiro.com
> >Vancouver, BC           tiro@xxxxxxxx
> >
> >A book is a visitor whose visits may be rare,
> >or frequent, or so continual that it haunts you
> >like your shadow and becomes a part of you.
> >                         - al-Jahiz, The Book of Animals
>
> Tiro Typeworks          www.tiro.com
> Vancouver, BC           tiro@xxxxxxxx
>
> A book is a visitor whose visits may be rare,
> or frequent, or so continual that it haunts you
> like your shadow and becomes a part of you.
>                         - al-Jahiz, The Book of Animals


Other related posts: