Biblical Hebrew Forum

[Peter Kirk <peterkirk AT qaya.org>]

On 17/01/2004 19:58, Trevor Peterson wrote:

> I keep coming back to the issue of transliteration, mostly because I'm not 
> happy with any system I've come across so far. Some of the things that bother 
> me are
>
> 1) the problem of representing matres lectionis in a responsible manner
> 2) how to represent the vowels (according to a seven-vowel system)
> 3) balancing representation of the written signs with readability
>
> I like the MCW system for its precision in representing the written text and its 
> utility in a plain-text environment. On the other hand, it's ugly to read. I like what 
> the SBL manual calls the "general purpose" system--basically what's popular 
> in Jewish writing--for its simplicity and readability; but it sacrifices precision.
>
> Anyway, I'm interested in feedback on what I've come up with so far:
>
> 1) For the consonants, I basically follow the system used in JBL, CBQ, etc. 
> One quirk is that I prefer dotted-k to q for qof, simply because it's a more 
> consistent representation of the emphatics.
> 2) For the vowels, I do not represent length. I use seven distinct signs and 
> three subscripted vowels for the half-vowels. For the full vowels, I use e 
> for tsere, o for holem, u for kubbutz, i for hireq, and a for patah. I use 
> the combined ae symbol for segol and a with a circle overtop for qametz 
> (regardless of whether it comes from original a or u). This is basically the 
> system used in Saenz-Badillos. I represent half-vowels with subscripted a, e, 
> and o. (It might be more precise to use ae and circle-a, but that seems 
> needlessly complicated.) I don't represent shva at all, regardless of whether 
> it's so-called vocal shva or silent.
>  

How do you distinguish <C, sheva, C> from <C, dagesh>? e.g. how do you 
write what is HAL:LW. YFH. in M-C encoding? Would you write halluw yahh 
(with superscipted w)? There is something missing here, surely. (And do 
you write he with mappiq as double h, or as single h distinguished from 
the mater h by a not being superscripted?) (Also I am assuming u with 
superscript w for vav shruqa).

Surely open e or e grave would be phonetically more appropriate (and 
more compact) than ae diphthong for segol.

> 3) I represent doubling by writing the consonant doubled. Where necessary, I 
> represent spirantization of bgdkft letters with a horizontal stroke over or 
> under the letter. This is standard practice.
> 4) I represent matres lectionis by writing the vowel according to the 
> equivalences given in #2 and following it with a superscript of the mater 
> consonant. This shows the reader both the consonantal writing and the 
> Masoretic vocalization while still facilitating readability. It has the 
> advantage over the SBL and CBQ system that it distinguishes between, for 
> instance, tsere-yod, tsere-he, and segol-he.
>  
What do you do about word final alef in biblical Aramaic (since you have 
just argued on the Aramaic list that this was silent and so a mater long 
before the Tiberian Masoretic period)?

> 5) I'm not sure what to do with quiescent alef and furtive patah. Right now, 
> my inclination is simply to write them as ' and a, respectively. Anyone who 
> knows Hebrew should be able to recognize where alef is quiescent without 
> needing it pointed out; since consonants are always written, the only way to 
> have a after another vowel is if the vowel is a furtive patah. ...
Not quite. There are a few other cases, mostly Yerushala(y)im and one 
word in Exodus 20:4, where two vowel points occur with one base character.

> ... I don't want to write quiescent alef as a superscript, because it is 
> actually the opposite of a mater. I don't want to put either one in 
> parenthesis, because I think that looks ugly and is probably unnecessary.
>
>  
Quiescent alef has no pointing, and so is distinguished (but only 
non-finally) from non-quiescent alef which is pointed at least with 
sheva. Your system would lost that distinction. There are just 51 cases 
in the WLC e-text of alef with sheva, all following a vowel and followed 
by a consonant, which in this system would not be distinguished from 
quiescent alef (GEN 4:23; 46:29; EXO 14:6; 15:6,11; LEV 5:5; 8:7; NUM 
30:3; JOS 8:9; JDG 9:35; 21:22; 2SA 7:16; 1KI 20:14; 2KI 9:21; 1CH 7:38; 
8:35; 2CH 13:3,13,13; 19:10,10; JOB 12:18; 31:40; PSA 10:8; 34:22,23; 
41:3; 58:5; 65:7; 69:16; 105:22; 149:8; ISA 1:18; 24:6; 32:9; 34:3; 
42:21; JER 1:17; 2:3,31; 50:7; EZK 6:6; 25:12; HOS 4:15; 5:15; 10:2; 
13:1; 14:1; JOL 1:18; 2:20; ZEC 11:5).

> I think that's the gist of the system. Obviously, it doesn't work very well 
> in a plain-text environment, although it can all be represented in a 
> plain-text source file for LaTeX. And that's fine for now. I've just about 
> given up on finding a system that I like for both plain-text e-mail and 
> typeset media.
>
>  
You can in fact do almost all of this with Unicode plain text, but as 
there is no full set of suffixed vowels you will have to use 
superscripted ones or (more traditionally) normal size ones with breve. 
You can find superscripted h, w and y under "spacing modifier letters" 
(02B0, 02B7, 02B8) and superscripted a, e and o (and schwa, if you want 
it for vocal sheva) under "phonetic extensions" (1D43, 1D49, 1D52, 1D4A).


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