User:Gilgamesh/Talk Archive 2

From Wikipedia, the free encyclopedia

Contents 1 Talk Archives 2 Request for assistance in a conflict between users regarding Canaanite and Hebrew linguistics articles 3 Apostrophe 4 Hebrew linguistics

[edit] Talk Archives

Page: 1 2 3 4 5 6 Active

[edit] Request for assistance in a conflict between users regarding Canaanite and Hebrew linguistics articles

I've moved the request/discussion from Wikipedia:Requests for mediation to Talk:Hebrew language and the issue has been listed at Wikipedia:Requests for comment, because according to Wikipedia:Dispute resolution, this should be done before mediation is attempted, and I do think that involving more people in the discussion will help move the issue forward.

Thanks, BCorr|Брайен, Co-chair of the Mediation Committee 12:52, 7 Jul 2004 (UTC)

Thank you. And I apologize for the procedural mistake; this is my first time and I didn't know quite what to do. - Gilgamesh 17:16, 7 July 2004 (UTC)

There's no need to apologize, and thank you. I very much appreciate the fact that you are working to resolve the dispute amicably. BCorr|Брайен 21:46, 7 Jul 2004 (UTC)

To be truthful, I didn't really have much hope that IZAK would soften his hard line. And the dispute only died down because so many other people disagreed with IZAK's science and methods. But I hoped that something positive might come out of resquesting assistance, and maybe better to get IZAK to co-request with me. Oh, and I didn't "threaten" IZAK as he said; I made a simple level-headed observation that people who abuse a system too much will often find it unavailable to them sometime in the future. - Gilgamesh 00:04, 8 July 2004 (UTC)

[edit] Apostrophe

Hi there. I'm asking both you and Fanger to share what you know about apostrophes in possessives ending in an s, specifically the possessive of Jesus, if it is indeed different. I was taught this both ways in school and have wondered what the truth was for years. I'm making a section on my talk page, if you want to discuss it there. Jdavidb 14:33, 7 Jul 2004 (UTC)

[edit] Hebrew linguistics

Gilgamesh, sorry for moaning again, but I find all those pronunciation guidelines in Hebrew very confusing and visually disturbing, e.g. your recent additions to Zephaniah. Are you sure there are no other ways of integrating this into Wikipedia? JFW | T@lk 21:13, 7 Jul 2004 (UTC)

Well perhaps...I don't know. I think they're all valid, as the Jewish diaspora had so many different liturgical dialects for centuries. As such, they really aren't different pronunciations or different spellings, but the same name in different forms of Hebrew that have been mutually separated since antiquity. After reading lots of material about how Sephardi and Mizrahi Jews feel that their distinctions are being ignored by Ashkenazi and Israeli Hebrew bias, and how much they value their liturgical dialects, I thought it would be worth including as many as I know. I mean, there are a couple million Jews to each of these major subdivisions. If there were a better way to integrate them into each article (say, with a side bar), and someone could show me how, then I'd do that instead. - Gilgamesh 21:20, 7 July 2004 (UTC)

There are too many dialects. I personally use the Dutch/German-Jewish dialect for liturgical purposes, which uses "[au]" for the cholem instead of the "[oi]" as favoured by Eastern-European Jews. Also, I pronounce the ayin as "[ng]", a custom introduced by Portugese Jews but taken over by the Dutch Ashkenazim.
My point is that you can't aim at giving the name in all dialects. It would be beautiful to see Moses transcribed as "[Moushei]", the way I would pronounce his name, but this is hardly useful in its context.
I suggest you take this to some page on Hebrew dialects, but please do not add this stuff to every page with a Hebrew word! :-~ JFW | T@lk 21:38, 7 Jul 2004 (UTC)

Understood...I suppose the lists would get monstrously large. ^^; But I think Tiberian, as an old uniting standard, should be on each page nevertheless. Fair? - Gilgamesh 21:44, 7 July 2004 (UTC)

I'm afraid I'm another one of those who thinks the pronounciation guidelines you have been adding are clunky and disturbing to the eye. They are also way over my head (as a native speaker of Hebrew, mind you) and for the most part archaic. The distinction between different dialects of Hebrew may have historical and liturgical significance, but for wikipedia purposes, it seems to me it is most useful to include only the most common pronounciations that people would be likely to encounter and search for. Sometimes that would be more than one pronounciation or spelling ("MAzel Tov" and "MaZAL Tov"), but very seldom are there more than two pronounciations of a Hebrew word still in use outside of prayer.

And I'm curious about your comment that: Sephardi and Mizrahi Jews feel that their distinctions are being ignored by Ashkenazi and Israeli Hebrew bias given that the current pronounciation of modern Israeli Hebrew leans much more towards Sepharadi than Ashkenazi. Would you care to elaborate? --Woggly 09:16, 8 Jul 2004 (UTC)

Sorry, maybe my fault. :P I've been reading a lot of Loolwa. Anyway, as for commonness, I had a discussion with Mustafaa, and I think maybe it would be good to provide only two forms: Tiberian Hebrew (historic and the origin of the majority of liturgical dialects, and is very phonemically rich) and Modern Hebrew (standard language). Originally I was providing only the two, but I guess over time I became too obsessed with more and more detail, and got carried away. - Gilgamesh 9:30, 8 July 2004 (UTC)

Your response is very gracious. I've been reading a bit at the links you provided for Loolwa; while she makes several very valid points, most of them seem to be her own points - I can't think of a group for which she clearly speaks. Don't make the mistake of assuming that, just because she writes well and sensibly, her opinions are representative of commonly held opinions, even within the minorities on behalf of whom she writes. --Woggly 10:03, 8 Jul 2004 (UTC)

You appear to be going through articles and standardizing Hebrew (and Arabic) transliterations. This seems to me to be somewhat like trying to standardize all articles into either American spelling or US spelling without discussion. I have no objections to there being transliteration standards in Wikipedia, for Semitic words and names, for Sanskrit and Hindu words and names and so forth if consensus can be reached. If consensus cannot be reached, as with spelling differences and terminology differences, the differences should and must be tolerated.

Currently your actions are forcing the issue. Such an action is very reasonable way to force any issue: be bold, and if no-one objects then all is fine.

But I do object.

You are representing the pharangealized consonants by non-standard forms. They are almost universally represented by h, s, and t with underdot. I realize that on many (probably most) current systems and configurations the proper Unicode representation will not appear, whether the underdot is applied using precomposed characters or by the ASCII letters followed by a non-spacing underdot. (I expect the latter might be more successfull.)

However those interested can do some work to configure their systems and fonts to show such characters. Those uninterested can leave well-enough alone, content to see a blank box or other brower symbol or hex code for an otherwise unrepresentable character. And using a proper Unicoding encoding means that there will be no need for another change when font handling and browser capability improves to the point where the standard transliteration symbols are available.

Another question is whether one should bother to distinguish strong and weak prounciation of the bgdkpt letters, and if so how, whether they should be marked by a following h or by underscore. The standard in scholarly books is not to indicate weak pronunciation, probably because such indicatation is redundant as anyone who cares will know that if one of the bgdkpt letters follows a vowel and is not doubled then it is weakened. Accordingly indicating the weakening in transliteration is not the norm.

Yet another question is whether shewa's (and furtive pathah?) should be shown be suprscript letters or by a normal-sized letter with a breve. It is my impresion that superscript letters are far more common in scholarly texts than the breve convention. Also, should the simple vocal schwa be indicated by e or by ə?

Should alef and `ayin be indicated respectively by right and left single quotation marks or by the more scholarly and more unambiguous U+02BE and U+02BF (which is I think the right answer)? Currently however I use quote marks instead because they "good enough" and do properly display on 99.9% of current displays. Except they don't display properly in Arial at a normal resolution, they turn into straight quotation marks. :(

Again, I have no problem with you forcing this issue as you have. Transliteration conventions should have been brought up long before now and some kind of consensus reached and stated in the style guide ... even if that is a consensus not to force any standard. At the moment everything is in limbo. I suggest a discussion on the style sheet talk page or on a page linked to the style sheet talk page and a notice that such discussion is beginning placed on the Village Pump.

I might be persuaded to your transliteration, even though not standard anywhere else (so far as I know). But placing non-standard transliteration on article after article does need some discussion and consensus.

Jallan 15:18, 12 Jul 2004 (UTC)

I've had long talks on these issues with Mustafaa, a Canaanite languages buff. We both agreed that proper academic standard (both in Ancient Hebrew and Standard Hebrew) supercede common social norms for the sake of academic work. Social norms are worthy of mention, however. As for the Hebrew linguistics, they are relevant to a wide range of religious and linguistic studies, not purely for Jewish or Israeli issues. The simple fact of the matter is, Tiberian Hebrew is the oldest variety of Hebrew to have its pronunciation standardized (8th century C.E.), and most liturgical dialects (including Sephardic and Ashkenazic) were based on it. The bh/gh/dh/kh/ph/th spellings are an intentional substitute for the underscored spellings not previously practical. The reason both Mustafaa and I use alternative lettering is that they are available in the vast majority of web fonts, and most of the characters are in the widely-displayed 0x20-0x17F Unicode character range.

As for the modern form, Standard Hebrew is intentionally conservative in its transliteration, as shown by the official names of most Israeli cities.

As for Wikipedia using a tiny Arial typeface, that issue is still under dispute; some people don't like it precisely for the reason that it makes some characters so small as to not be distinguishable. But it is the typeface that is at fault, not the article's content. (Ask Wikipedia to fix the typeface. Don't "fix" the articles instead.) Maybe someday when it becomes more practical, we can migrate all the alternative letters to the most academic ones, including underscores, dot-belows, etc. We could even do that today or tomorrow if everyone thinks it's the best thing to do. (I've noticed that Windows XP's updated Tahoma typeface now supports all special Hebrew and Arabic transliteration letters except for the special aleph/ayin half-circles.)

As for issues of Ashkenazi discrimination against Sephardi and Mizrahi Jews, once again read up at Loolwa Khazzoom's website, The Jewish MultiCultural Project, and Jews Indigenous to the Middle East and North Africa. Loolwa in particular has written many published articles and essays on her experiences of discrimination and cultural marginalization in Ashkenazi-dominated Jewish communities.

- Gilgamesh 23:31, 12 July 2004 (UTC)

I think you have misunderstood my objections. I agree with you fully and absolutely on the need for scholarly transliterations of Semitic languages and other languages in Wikipedia. I also am aware of Ashkenazi discrimation against other dialects of Hebrew.

I follow standard Tiberian pointing myself for transliteration (when not sometimes transliterating the consonants only).

What I am objecting to is non-standard transliteration appearing throughout Wikipedia without discussion. Is the system you are using here for pharygneals ever been used by anyone else on the web or in any published material? A newly invented transliteration system does not become acceptable because two people have decided to use it. That comes close to being "original research". Similarly Wikipedia should not be the place to introduce novel terminology or a novel classification sytem.

Underscored spellings are quite practical, e.g. Negeb rather than Neghebh. However generally scholars simply spell Negeb which is good enough. That is the standard in academic books. Latin-letter transliterations of Tiberian in Wikipedia should follow standards actually used. In academic writing I've generally only seen marking of softened vowels in transliteration tables or ocassionally when someone is making a particular point about Masoretic pronunciation. Otherwise, the softening is not marked.

I would guess that it has become conventional not to mark softening in normal transliteration because context alone makes it clear without the ugliness of h and h and more h and without numerous fussy underline diacritics. Also, the softening was late development and not found in the pronuncation of Biblical Hebrew which is another reason why it might be felt better not to mark it. (Of course vowel sounds of true Biblical Hebrew would have been different also, for example Tiberian melek would probably have been pronounced more like 'malk' in Biblical Hebrew.)

But what I would like to see is an open discussion by as many as are interested as to whether we should adopt your system or adopt fully the academic system in common use (which I think far preferable) or some mixture of the two. I am willing to abide by consensus if there is consensus among those interested in using your system. But I am not willing to accept a novel and idiosyncratic transliteration system in Wikipeida in place of a standard one being applied without consensus. Why should I or anyone use a system which so far as I know has been previously used by only two other people in the entire world?

As you mentioned, difficulties in displaying the standard accepted characters are fast disappearing.

On current Macintosh systems, if any requested character is not found in the current font, the system searches through all fonts until the character is found and displays that version of the character. It will not match the characters about it exactly in style, but at least the proper character appears. Accordingly Macintosh users have only to load a single font with a missing character to view the character regardless of their current font settings.

Mozilla on Windows and Un*x machines follows this same procedure. So does Netscape under Windows. Other browsers may also do so by now though Internet Explorer does not. It has been a long time since I have played around with numerous browsers and compared their capabilities. I normally use Mozilla.

So ... yes ... we can probably now start using the proper characters. I have been doing so at need in various articles where Semitic names and words appeared and have no wish to see my use or anyone's use of the proper Unicode characters for the normal Latin letter transliterations of pharyngeals replaced by non-standard kludge symbols ... unless there is consensus that this should be done. Without consensus it seems to me I could with some reason call any such changes ... replacing a character in standard use in Semitic transcription with one never before so used ... as vandalism, just as though someone who favors simplified spelling were to go through article after article and replace all occurrences of though by tho and through by thru. (In fact I myself would be happy to do such a thing as I favor such simplified spellings ... but Wikipedia style is for good reason by its rules conservative in such matters. I would also like to throw away the Latinized Greek forms used for ancient Greek names and use pure Greek transliteration, that is Kirkê instead of Circe and so on. But I don't. One must bow to convention in a project like this.)

If you and Mustafaa without general discussion have a right to insert idiosyncratic non-standard transliterations into Wikipedia and change current and correct transliterations to fit, then I and anyone else else has a similar right to insert standard transliterations and change your idiosyncratic transliterations to fit, indeed, probably far more right. That is why discussion and consensus is needed.

The best answer is, I think, to begin using the proper characters and to create a page accessible from the Coummunity Portal (and able to be linked to from any article) which would be called something like Wikipedia: Viewing undisplayable characters. The page would contain a short discussion about various browsers (recommending browsers for non-Mac systems that implement automatic substitution of missing characers). It would also have an external link to Unicode fonts for Windows Computers which includes links to many free and inexpensive fonts (also usable on Un*x platforms) which contain a fuller than usual complement of characters or which contain characters needed for particular languages. Any article making much use of characters that are not found in most fonts might also link to that page with a standard note: "If you cannot view all characters in this article properly, please look at Wikipedia: Viewing undisplayable characters.

In short, I currently strongly favor using the standard characters over kludges and in guiding people to upgrade their systems to be able to see the standard characters. I also don't want people like you and I who agree that one should use characters outside of the Latin-1 range whenever they are useful not to be able to produce a united front against those who disagree. But if, for example, you use şûr instead of the standard transliteration where it would be coded as ṣûr > ṣûr or as ̣ûr > ṣûr, it can reasonably be argued (and I do argue) that s.ûr is just as adequate as a kludge and far easier to enter, or cûr or tzûr or tsûr, all of them more standard because found in more places than is your innovative and novel usage. Those other kludges at least are not new ones. They are ones used elsewhere and already familiar to some readers, some of them familiar to many readers. The world and Wikipedia should not be burdened with yet another new kludge method to transliterate Hebrew into Latin characters. If you do support a kludge method, then support one already in use rather than inventing a new one. If there is to be one transliteration convention recommended on Wikipedia for Semitic languages, then consensus is needed as to what that convention should be. Otherwise, I am obviously quite as much at liberty as you are to use any convention I like and to change your material, especially when I prefer the standard that is mostly used.

I agree so much with what you are doing in including Tiberian transliterations in articles, as I also have done, but not with attempting to force an idiosyncratic non-standard form of transliteration without discussion. If indeed, the method you are using is one that has become standard among some group on the web or elsewhere, please indicate where it has been used and my objections to it will greatly decrease insofar as it actually is in use outside of Wikipedia.

I also, as you do, agree that Wikipedia should cease attempting to force a particular font on users. I and many others said so in a poll on meta-Wiki some time back. It seems this was in vain for the moment. The unnecessary POV of forcing of a particular style continues unabated.

Jallan 02:23, 13 Jul 2004 (UTC)

Alright... You are very rational and logical. You raised points I had not before considered. Since most of the work is mine, I will fix the Unicode myself. I've always been able to, so I'll take up the task. BTW, I did not invent the convention of using breves for certain hateph vowels. From the academic texts I've seen, schwa is used for syllabic shewa, breved vowels are used for sylabbic hateph vowels, but superscript miniature vowels are used for non-syllabic hateph vowels. Example of a syllabic hateph: ʾĔḏôm. Example of a non-syllabic hateph: ʾAh^arōn. - Gilgamesh 02:36, 13 July 2004 (UTC)

Don't get me wrong here... I totally approve of having Tiberian Hebrew transcriptions, but don't much care what transcription is used for Hebrew, as long as it's one-to-one - and as long as I'm not the one who gets to do all te work of converting it! :) But I think Jallan's arguments are good; the more standard the better, all other things being equal. In Arabic articles, I usually consider the Arabic letters sufficient indication, but I probably ought to start using a full transcription system sometime. - Mustafaa 04:40, 13 Jul 2004 (UTC)

I've worked all day and night. Barring perhaps a few oversights, I have converted as many of my Hebrew transliterations as possible to Tiberian and Standard transliterations using the proper Unicode characters. The exceptions in this case are the superscript vowels, for the reason that there seems to be no distinct Unicode character for in-text superscript "e". For these vowels, I used HTML tags: ^{<small>e</small>} - Gilgamesh 10:51, 14 July 2004 (UTC)

I can support this, with one exception which I will get to much later, as reasonably representing methods in common use. Normally I think either the superscript method is used throughout or the breve method. I generally like the superscript for technical linguistic use as it makes the shewas stand out. But it looks hideous when a proper name begins with alef or `ayin followed by a shewa and the shewa is capitalized. I suppose it is because the breve method is less uncouth throughout that some prefer it. Mixing the two as you are doing makes visual sense. though I doubt that in fact a vocal shewa following an alef or `ayin is any less syllabic or more syllabic than a normal shewa following a resh or beth and I don't recall ever seeing them mixed. (But normally one reads for meaning and gets used to somewhat different methods and tends to be blind to the actual method being used). Still, Ă as the capital of ^a is visually so superior at the beginning of a name I cannot do anything but commend it. But I know I've never encountered a distinction between syllabic and non-syllabic hatef-shewas in any Hebrew grammar. All shewas are non-syllabic in Masoretic understanding, unless you wish to reckon the normal consonant-followed-by-vocal-shewa as a syllable.

As to superscript letters: there is U+1E43 MODIFIER LETTER SMALL A, U+1D4A MODIFIER LETTER SMALL SCHWA, U+1E49 MODIFIER LETTER SMALL I and U+1D52 MODIFIER LETTER SMALL O. See Unicode: Phonetic Extensions. All these have compatibility decompositions to superscript style applied to the corresponding regular letters which means Unicode considers that users should consider the superscript letters as generally indistinguishable from a superscript style applied to the corresponding normal letters. The value in using the hard superscripts would be that the letters should better match the weight of other characters in the font (in whatever few fonts actually currently support these) and the formatting is not lost when transferred to a plain text environment. On the other hand the value in applying superscript style to normal characters is that the results are far more likely to be visible in current browers. Similarly in the few cases where I've used an underline I've done it by html style for the same reason. However, if no-one is complaining about your efforts, then I will probably also switch to using hard-underscored characters, at least for cases where the transliteration letter actually represents a distinct letter in the character set rather than, as in Hebrew, a variant pronunciation of a letter.

Of course there are slightly different conventions for transliteration, even using the same character system, for example I prefer ʾÎyôḇ to ʾIyyôḇ and Śārâ instead of Śārāh and would probably use the latter in the body of any article where I was transliterating those names for some reason. But both methods are used and I don't complain about your use for heading purposes. Similarly I would, as mentioned in my previous comments, generally ignore softening in transliteration as most do, but quite agree with the use of the softening bar indication for use in the heading paragraphs. I have also done so on occasion in heading paragraphs.

Now for my one serious complaint. The use of ə seems to me non-standard. That I do have a problem with.

Mostly, indeed almost universally I think, normal vocal shewa is represented either by superscript ^e or breved ĕ. I like using the genuine IPA shewa (though its use seems to me unusual and if called on it I would be willing to draw back). I believe I've only seen it used somewhat rarely with the superscript method as ^ə. I don't think I've ever seen non-superscript ə in Latin-letter Hebrew transcription. The use of ^e or ĕ rather than ^ə may have arisen originally for typographical reasons, most printers not having sufficient superscript ^ə in the days of hard type. But it was also unambiguous as anyone who knows Tiberian Hebrew pointing knows that ^e or ĕ in transliteration must stand for hatef-segol following alef, `ayin, heh, and het and must stand for normal vocal shewa anywhere else.

But it feels very wrong to me to apply superscripting to hatef-shewas and not to do the same with normal vocal shewa, whether one represents normal vocal shewa by ^e or by the less normal ^ə. If indeed you can point out sources that use your method, I withdraw any objection other than that of personal preference for indicating clearly the shortness of vocal shewa and for following more common usage. If you cannot find such sources, then a more normal representation should be adopted. The more normal your transliteration, the less it can be validly objected to.

Since you have taken all this on yourself without seeking consensus, I am still in part waiting to see what objections might be raised before madly going off to fix things in articles I've largely written or revised to at least use the half-ring characters in preference to the quote symbols I'd been using and to introduce proper underscore characters in Ugaritic forms. Perhaps the era of Unicode characters is finally comiming into its own. But perhaps not.

Jallan 03:43, 18 Jul 2004 (UTC)

Well, in this regard I've relied on academic webpages and textbooks, and I know that the linguistics department at Brigham Young University (one of the biggest institutions of linguistic study on earth) uses plain ə ĕ ă ŏ when they constitute a syllable, but use superscript for final pre-pharyngeal ^a. But come to think of it, while I could have told you before that I was absolutely certain that the superscripts were used after א ה ח ע before other consonants, now I think I'm not so sure. They could simply be using breve forms, but I also remember being told that these hateph vowels represent a transition between consonants in the same way as shewa, but that the guttural consonants can never take an ordinary shewa and must take a hateph instead. As such, in these cases, the superscript vowels represent the briefest of hatephs where an equivilent shewa would usually never be transliterated at all. - Gilgamesh 04:06, 18 July 2004 (UTC)

A problem is, I think, that both of us have seen so many different transcription systems for Hebrew and for other scripts, and often read for meaning rather than worrying about any anomalies or oddities. It is hard to recall exactly what is or is not standard.

But as far as I know, and I just checked good old Gesenius/Kautch on this, vocal shewas indicate what was a shortening of formerly full-length short vowels in open syllables. The vowel coloration was also lost, except when preceded by a "gutteral", whence the special hatef-shewa Tiberian notation. Furtive pathah is another beast entirely, perhaps a faint pathah to replace somewhat in length and timbre a final gutteral that was already often not pronounced.

Generally scholarly Hebrew transription systems either use superscripts for all vocal shewas or breved vowels for all vocal shewas. The latter system sometimes still uses a superscript for furtive pathah. But there is normally no mixing of superscipt vocal shewas with breved vocal shewas within a single system. The Brigham Young system you describe seems to be a modification of the breved vocal shewa system in which non-hatef vocal shewas are indicated by the IPA shewa symbol instead of by ĕ. Breve may have been omitted for tyopgraphical reasons. But of course breve is redundant on ǝ in any case since in a Hebrew transcription it must be represent a half-length vowel. The Brigham Young system seems to be the standard breve system with addition of IPA shewa as an extra flourish.

Generally a system using superscripts does not use any casing, even on proper names. After all, it is transliterating a script which uses no casing. A form like ʾ^Aḇîḡáyil looks silly, though I think I've occasionally seen such. The breve system is far more likely to optionally use casing for readibility and is better if one wishes the same forms of proper names that appear in strict transcription of Hebrew to also appear in non-Hebrew Latin text.

Of course writers mix systems in the same text. I'm looking now at Speiser's Anchor Bible Genesis which follows the general scholarly transliteration for Classical Hebrew text, except macron is also used where more commonly circumflex would be used and sere is represented by e with an underdot. Occasionally Rabbinical forms are transcribed in a more modern transcription using f. Speiser also uses apostrophe forms rather than the half circles. In Koehler and Baumgartner's The Hebrew and Aramaic Lexicon of the Old Testament, vocal shewas are superscript but the breve also appears occasionally to mark the normal short vowels, presumably when there presence is unusual or pertinent to a paricular discussion. They also use macrons where more precise transcription would use circumflex.

But mixing superscript shewas with breve shewas in the same word I find unusual and bothersome, except for initial capitals (that is capitals following alef or `ayin). My own choice would be to use the superscripts, either with no capitlization or with transformation of a superscript that requires capitalization into a breved capital. But all breves is fine with me also and has its own virtues. If you decide to go that route I won't mind changing things that I have done to fit.

Jallan 16:28, 18 Jul 2004 (UTC)

Well, I would prefer to use all breves. However, I should still use superscript "a" for before final guttural consonants. Also, by using all breves, even with only this one exception for superscript "a", I could eliminate ^{<small>a</small>} altogether, and just go with ª (Unicode 0xAA). I prefer this route, personally. I never liked mixing HTML with Wikipedia, but did it because there is no Unicode character for superscript "e" (so I superscripted "a" and "o" for consistency). - Gilgamesh 04:58, 19 July 2004 (UTC)

Actually, I found another source that uses ə for shewa and breved vowels for hatephs for Tiberian transliteration. Dictionary.com uses it in etymologies of Hebrew words. Personally, I think it has academic credit. [1] [2] [3] [4] - Gilgamesh 05:12, 19 July 2004 (UTC)

That suggestion is fine with me. The breve method looks less technical and there is some value in explicitly distinguishing furtive pathah from hatef-pathah (though I doubt there are any cases where one could be confused with the other if indicated by the same character as is also done).

As to using U+00AA, both this and U+1D43 are given compatibility decomposition to "<super> 0061 a" (though not cross-referenced to each other).

You seem to have missed my note above:

As to superscript letters: there is U+1E43 MODIFIER LETTER SMALL A, U+1D4A MODIFIER LETTER SMALL SCHWA, U+1E49 MODIFIER LETTER SMALL I and U+1D52 MODIFIER LETTER SMALL O. See Unicode: Phonetic Extensions.

Most standard Latin letters do now have superscripts in Unicode including all those needed for representation of shewa if one wishes to go that route.

The respective properties of the two superscript Unicode a forms (as found at Unicode: Data.txt) are:

00AA;FEMININE ORDINAL INDICATOR;Ll;0;L;<super> 0061;;;;N;;;;;

1D43;MODIFIER LETTER SMALL A;Lm;0;L;<super> 0061;;;;N;;;;;

The only difference is that U+00AA is UCD (Unicode General Category) Ll (Letter, lowercase) while U+1D43 is UCD Lm (Letter, modifier). U+1D43 is the more correct one to use, as all the superscript Latin letters characters intended for phonetic use have UCD Lm in Unicode, though the description modifier is somewhat inaccurate for most uses of such characters. Also, the semantic meaning "feminine ordinal indicator" (abbreviated in the HTML entity name &ordf;) is incorrect and occasionally in some fonts the character appears with an underscore. That said, I've no objection to using U+00AA rather than U+1D43. U+1D43 and other characters in that section of Unicode are new additions, and until there is far more font support for U+1D43 than at present, I feel U+00AA will do excellently.

Of course, within an article can use anyone can use any other reasonable Hebrew transcriptions that they prefer (whether for Classical or for modern Hebrew). My concern was that if a single style transcription is to be generally used and imposed in the headings of the articles, that the transcriptions should match general scholarly usage where such usage is at least 90% in agreement and follow one of the common usages where standard transcription usages differ in minor ways. It seems to me that the usage you now advocate reasonably accomplishes that goal. No-one familiar with scholarly Hebrew transcription would find anything puzzling or unusual or odd in the transcription, even if not exactly their own preferred style.

Jallan 18:22, 20 Jul 2004 (UTC)