ICU provides a mechanism to produce transliterations from one language
to the writing system of another. Here are the transliterators we add
to ICU in our engine. You can see how easy it would be to write an ICU
transliterator for your proposed system and then let the user chose to
turn on or off transliteration to their liking.
http://crosswire.org/svn/sword/trunk/icu/
Troy
Tyndale STEP Project wrote:
> Colin, transliteration is a big problem, mainly because the systems in
> use are so silly.
> You are right to distinguish between /scholarly /or /reading
> /transliteration.
> The /scholarly /systems try to express the Hebrew letters exactly in
> Latin characters,
> but the /reading /systems try to express what it would sound like.
>
> I regard the /scholarly /systems as a waste of time. They were only
> useful in the old days
> when typesetters weren't able to use Hebrew letters properly. Scholars
> who know Hebrew
> would prefer to have real Hebrew, and those who don't know Hebrew are
> just confused by all the accents.
>
> Personally I prefer the system used by rabbinic scholars, who use very
> simple transliteration,
> eg "loving kindness is /hesed.
> /This is fine for rabbinic scholars because they know what the
> underlying Hebrew looks like,
> and they don't need lots of clues about it from funny accents.
>
> The full-complex scholarly transliteration is difficult to produce and
> confusing to read,
> but the very-simplified rabbinic transliteration give the non-hebrew
> reader too few clues.
> For example, /hesed/ starts with a /chet/ not a /he/ so it is a harsh
> "h" (often indicated by a dot under the "h")
> and the middle letter is /Samek/ not /sin/ which both sound the same.
>
> Probably the best method is somewhere down the middle of the road.
> Let's see if we can do transliteration using only extended ASCII - which
> will help us and the readers.
>
> * Sin and Shin can be /s/ and /sh/ whereas normally they are "s"
> with a "v" on top or a "u" on top
> * Aleph and Ayin are silent and are normally represented by an
> closing single quote and an opening one,
> * This is silly and I prefer to use a single straight quote - ie '
> * However, it does help non-Hebrew readers if they can recognise the
> difference between the letters,
> * so I guess it is better to use the curly quotes. (this means we
> have left the extended ASCII already!)
> * Chet and He are the tough ones, because Chet is normally
> transliterated as "h" with a dot underneath it,
> * which is really difficult to do (there is no single Unicode glyph
> for this). What about /_h_/ and /h/ ?
> * Beth is pronounced "v" unless it is doubled, but it is better to
> use /b/ to help English readers and double it when the /dagesh/
> form occurs in the middle of a word.
> * They can easily take on board the rule that they should pronounce
> it "v" when a single "b" occurs in the middle of a word
> * It is probably best to ignore the doubled dagesh forms of Beth,
> Dalet, Kath, Pe and Tau.
> * Although they are theoretically different in sounds, in modern
> Hebrew they are mostly very similar,
> * or the difference is something which is difficult for us to
> reproduce,
> * and anyway, we don't really know how it was pronounced in ancient
> times,
> * so lets stick to /b, d, k, p, /and /t/
> * Dagesh forms in general indicate a doubled letter, which is easy
> to indicate in English, except at the start of a word or for "sh"
> /Shabbat /looks OK but /Bbabel /looks strange.
> * Tau and Tet are really difficult, because they sound exactly the
> same. In some systems Tau is /t/ and Tet is /+/,
> * but that is confusing. It is probably better to use /t /and /_t_/.
> * Sin and Samek are similarly identical in sound, so lets use /s/
> and /_s_/.
> * Tzade is another "s" sound which is often transliterated as "s"
> with a dot under. It is pronounced "tz" eg in /matzot./ We could
> use /_z_ /
> * Wav (or vav) is almost always pronounced "v", so it is less
> confusing to use /v./ But sometimes it is a vowel "o" (with a
> single dot over it) and sometimes "u" (with a single dot beside it).
> * Yod is difficult because sometimes it is the consonant "y" and
> sometimes the vowel "i" (after the single-dot /hireq/), and
> sometimes the vowel é (after the double-dot /sere/). It is best to
> keep this distinction.
> * although there is a distinction between very short "e" (shewa) and
> short "e" (/seghol/) and long "e" (/sere/), we could ignore it. Or
> we could use e, ê (as in fête) and é (as in paté)
> * similarly with short & long "a", we could use /a and â/
> * with "i" and "o" and "u", the long forms include a letter (a
> /yod/, /vav/ and /vav /respectively) which we could mark by
> underline: /_a_ _i_ _o_ _u_ /
> * /yod/ following a /sere/ simply sounds like /é /but to indicate
> there is a letter involved, we could underline it as /_é_/ - like
> the other vowels which involve a /yod /or /vav/
> * there is a tricky problem with a few words where the Hebrew vowel
> for a long "a" (the "T" shaped /qames/) is actually pronounced
> "o". The rule is that when it occurs without an accent (ie a "<"
> sign over the letter) AND it is followed by a letter with no
> audible vowel (ie it has no accent or it has a /shewa/ or it has a
> /dagesh/) then it is "o". To make it even trickier, most Hebrew
> Bibles don't have the accents (the "<" signs) - ugg! For this
> special "o" we could use /ô/
>
> So "king" is /melek/ and "loving kindness is /_h_e_s_ed /and "light" is
> /'_o_r/ and "eye" is /'ayin
> /The alphabet is /' b g d h (v /or/ o /or /u) z _h_ _t_ (y /or /i /or
> /é) k l m n _s_ ' p q r (s /or /sh) t
> / and vowels are /a â e ê é _é_ i _i_ o ô _o_ u _u_
> / where the general rule is that an accented vowel is stronger, and an
> underlined vowel is a combination of a letter and pointing
>
>
> This is what Gen.1.1 looks like in traditional transliterations:
> Emacs!
> This is what Gen.1.1 looks like in the proposed simplified transliteration:
> Genesis 1:1 /Beré'sh_i_t bara' 'eloh_i_m 'êt hashâmayim ve'ét ha'are_z_.
>
> /This simplified method doesn't lose any of the information conveyed by
> the complex method,
> and it is far easier to read, and when you do read it, the result is
> phonetically correct in a
> fairly intuitive way.
>
> What do you think? Are we being too maverik, or are we making a strike
> for common sense?
> It is certainly easier to encode!
> I can fairly easily produce a whole Bible in this transliteration.
>
> David IB
>
>
> At 11:30 11/11/2009, Tyndale STEP Project wrote:
>> Do we have a preferred method for transliterating Hebrew? The
>> dictionaries I'm drawing on seem to vary slightly and we want to be
>> consistent. Possibly Tyndale House has a preferred "house style"?
>>
>> One issue is whether we want to go for a "scholarly" or a "reading"
>> transliteration, i.e., whether we want to distinguish long and short
>> vowels, samedh and sin etc, so that the original Hebrew can be
>> more-or-less reconstructed or provide something that enables the reader
>> to pronounce the Hebrew word. I'm more in favour of the latter, on the
>> grounds if the reader is sufficiently interested to care about the
>> difference, they probably know enough Hebrew to be able to read it, but
>> happy to go either way.
>>
>> (Fortunately this issue doesn't arise in Greek - there seems to be a
>> single standard.)
>>
>> Colin
>>
>> --
>> Posted By Tyndale STEP Project to Tyndale STEP - History
>> <http://tyndalestep-hist.blogspot.com/2009/11/hebrew-transliteration.html>
>> on 11/11/2009 03:30:00 AM
>
>
> --
> Posted By Tyndale STEP Project to Tyndale STEP - Programming
> <http://tyndalestep-prog.blogspot.com/2009/11/hebrew-transliteration.html>
> on 11/11/2009 09:40:00 AM
No comments:
Post a Comment