Most definitely, it is very important to distinguish between bet and vet, pe and fe.
I think it would be best to do this without any vowel-points at all, but unfortunately there aren't enough letters. So we should do it with al few as possible. Your vowels at the end of the word look ugly.
Here's my proposal:
- same as roman orthography
- same as roman orthography
- same as roman orthogrophy
- bet with dagesh
- fe without a dagesh
- vav with holam
- pe with dagesh
- vav with a dot in the middle (I forgot what it's called).
- vet without dagesh or double vav
- khaf (without a dagesh)
The above sample isn't at all meant too seriously - and quite a hip-shot. And for sure, you're right that the ending vowels look ugly.
BTW, here's some more info one could choose from: http://www.fa-kuan.muc.de/GENESIS.RXML
A question to both you and Craig (Estrangela): what about final and initial variants of consonants? Does your transliteration of Lojban use them? Why/why not?
Yes of course it does. Hebrew would look quite strange without finals. The only possible problem would be a final "p", but since this is extremely rare in Lojban it doesn't cause much of a problem. When it does occur, you could either put a dagesh in the final fe, or just use a regular pe like in Modern Hebrew. — Adam
No problem with Yiddish transliteration:
''Yes, I think Yiddish-style is far and away the best method, but
what to do about the vowel pairs? --John Cowan''
This seems to be quite easy! Here's a sample (following the Book):
Since in "ii", "iu" etc. and "ua", "ui" etc. the first character has a *consonant* value, there's no need for an initial "shtumer alef" (which has to be used e.g. for ".i"!).
Because giving "uu" with "tsvey vovn", I'd propose to have "vet" for "v" (with the further need having "dagesh-bet" for "b").
The Yiddish "ay"/"ey"/"oy" convention should not be used. ('I've changed my opinion - see below ;-)')
In Yiddish there exists no "schwa", so the "schwa-alef" is a new creation for Lojbanic needs. (One could use "schwa-ayin" as well) - .aulun.
Strangely enough, I think I would prefer the double-stroke "gershayim" to the single stroke geresh for the Lojban '. In Hebrew orthography, the single-stroke geresh/apostrophe tends to indicate a modification of a single letter (as in Modern Hebrew Z' for the sound of lojban j or G' for dj or Tz' for tc), or else an abbreviation that's one letter long (a usage which stretches back further in time, as in A', B', etc. for 1, 2,... or 1000, 2000,... (so a number like 3231 would be written G'RL"A)), while the double-stroke gershayim is used for abbreviations longer than one letter long or multi-letter numbers (placed always before the last letter). Somehow, as a mid-word symbol, " seems to look better than ' in Hebrew, though it may seem a little counterintuitive. Other than that, Yiddish-style is definitely the best choice (though one could find stuff to quibble about, probably). I've long felt that the closest analog to the Lojban period is the Yiddish "silent aleph", at least at the beginnings of vowel-initial words. --mi'e mark.
To be honest, I don't like single stroke geresh very much - yet for the reasons mentioned above (i.e. the specific use of gershayim in modern Hebrew, and Yiddish as well) double-stroke geresh doesn't seem a good solution either. What prevents us to simply use Hebrew "he'/hey" with the restriction that it is only allowed between vowels? I'm accustomed to ' in Latin script, but why stick to it in other transcriptions?
I've put a sample here: http://www.fa-kuan.muc.de/LOJHEBR.RXML
For differentiating "oi" and "ui", the latter could be written with vav-dagesh. — mi'e .aulun.
Remember, I'm the one who likes h for apostrophe in Latin letters, so I won't squawk too loudly about using heh in Hebrew. Though h, being an ascending letter, does stand out rather better than Hebrew heh. Not really a good reason to oppose heh, just something to recall. I think I still like the double-stroke... *shrug*. Hmm, I thought you were using kamatz-aleph for o, hence kamatz-aleph-yod vs. erm... I guess it would be shtumer-aleph-vav-yod? Or just vav-yod? I think I do like the idea of the full shuruq, now that I think about it. mi'e markI see in your sample that you're using full Yiddish conventions, including double-yod-kamatz for ai. Any reason for the change of heart?
Sorry, I feel like I should come to an end with my series of hip-shots ;-( But yes, I now think that it's best keeping as close as possible to only one convention that fits best (which for sure is the Yiddish one). So I'll be going to redo the above samples: 'ei' -> shtumer alef-tsvey yudn; 'ai' -> shtumer alef- komets tsvey yudn; 'oi' -> shtumer alef-vov-yud; 'ui' -> melupn vov-yud - no shtumer alef because of the 'u' being consonental! (Maybe due to this difference in spelling of 'oi' and 'ui' the melupn might not be necessary.)
It's for this general reason also that I'm still hesitating to replace tsvey vovn by veys/veth which I'd prefer - since the "double-u" seems to be created after the German/Polish 'w' (= v!) hence not too appropriate with regard to international conventions.
The problem with 'uu' still exists (maybe: vov-melupn vov?) — mi'e .aulun.
Mmm. A lot to think about. I also noticed you were going whole-hog with Yiddish, to the extent that j -> zayin-shin. That's probably better than the chet that was suggested earlier, but it breaks the one-letter/one-phoneme reading. As does some of the vowel stuff, I guess. I caught myself musing about something possibly interesting for j: dagesh-zayin (or maybe dagesh-shin). It could work! And incidentally, I'm sort of in favor of dagesh-peh for p and peh-rafe for f. The more clues, the better. Similarly, a melupm wherever possibly helpful would not go amiss. But I have to ponder more about the rest of the vowel stuff. On the whole, writing it in perfect Yiddish transliteration is not hard to read and looks pretty good, but it feels like breaking some of the elegance of Lojban. This may or may not be vital: it's only a writing convention. --mi'e mark
Uncial was also only a writing convention, as is Tengwar. Go for the elegance. While I do not read Hebrew letters, I agree with mark on everything he said after he stopped mentioning letter names and I started understanding him. Maintain all elegance that is availible! - mi'e. kreig.daniyl.
Meanwhile, I've altered my sample above, and I think that 'uu' should no longer be a problem (writing it with vov-melupm vov).
With regard to elegance, I first thought of going for something like Tengwar in Hebrew (with the vowels consequently written in punctuation only). Yet, like .adam. has pointed out clearly, this doesn't look elegantly in Hebrew. Writing something similar to Hebrew's "Lesestützen" ('immoth-hakkriah; matres lectionis) style doesn't seem appropriate for Lojban, or at least pretty sophisticated. Also, one cannot achieve Hebrew's frail elegance since e.g. Lojban doesn't have finals 'n, m, ts, x ...' except in cmene, so the endings wouldn't look very pleasant anyway. So, repentantly, I came to mame-loshn script that indeed looks somewhat fleshy-heavy compared to Hebrew, but it fits best of all. (BTW, it's also for its heavy ductus that I still keep rejecting Tengwar 'Beleriand' mode with all the vowels written out) — mi'e .aulun.
Don't be so hard on yourself. Personally, I think a Yiddish-style orthography is really quite nice and "right" for Lojban. Vowels can be relegated to pure diacriticals in languages where they carry the semantic weight differently, not in Lojban. I actually find the Yiddish-style rather nice on the page and not terribly hard to read. (it may actually help that I don't know Yiddish very much, so I don't have native-language expectations getting in the way). --mi'e mark.
I've been thinking more about this now. Using all of the orthographic conventions of Yiddish (kamatz-aleph for o, zayin-shin for j, double-yod with shtumer-aleph but ayin without, etc) will give you something that's not unfamiliar-looking to Yiddish-reading eyes and will be probably pretty decent... but is it Lojban? It's basically like proposing an "English alphabet orthography" for Lojban in which we write "ee mee nelshee loy shmavo zhoy geesmoo geeheh tahdnee wee la lozhban." I mean... yeah, it's okay, it fits the conventions of English... but it isn't really Lojban. On the other hand, go too far off and it looks weird in other ways. Mustn't be blinded by pretty familiarity and similarity to something else, I guess. Hmm. I still think full-letter vowels are a necessity: vowel-points are okay for Hebrew, where vowels are not on the same level as consonants when it comes to determining meaning, but not for Lojban (though one might argue that Devanagari proves that's wrong). Well, think about it. --mi'e mark.
I don't see the whole issue too seriously, since IMO why at all be looking for a non-latin script for Lojban! And still again: for some special purposes, it's nice to have it more "romantic" and use other scripts - and just play around a bit adjusting or inventing those scripts. But, if there really were to choose a different font like Hebrew, Tengwar or even Hiragana, it wouldn't be anything else than just a matter of habitude getting familiar with one or another convention and after some time, nobody would be asking if it really were "Lojban" or not - even with regard to your English example
In the eyes of someone unfamiliar with e.g. the Hebrew script, even a Lojban text would be nothing but Hebrew etc., a Mongolian text written in cyrillic characters seems to be Russian to someone who doesn't read Russian, and even people reading Chinese Hanzi will hesitate for a while when coming across with old Vietnamese texts written in 13th century "chu Nom" (an adaptation of Chinese "chu Han" to fit with Vietnamese phonology).
On the other hand, nobody familiar with latin script will ever mistake a, say, Turkish or Finnish or German text for Latin or English etc.
Also, the matter of scripts and orthographic convention is not a static one: each language is developing and their scripts are comparable to suits which - after some time - no longer would fit. Others were taylored badly from the beginning (like the very beautiful scripts for the Japanese language). Still others, although custom-made like Latin scripts for Rumanian or Turkish, are no longer up to date in our world of internet compared e.g. to the Roman Uighur script which is pure ASCII. — mi'e .aulun.
-- -- ----
Here's my proposal (mi'e filip.), based on my proposed Arabic Orthography:
(Apologies for the half-rendered vowel points; the version of IE I used to display this apparently doesn't put them underneath consonants which have dagesh or rafe attached.)
It seems to be pretty similar to the scheme .aulun. used right at the top, with vowels as points (though sheva is strictly y and isn't used beneath consonants which have no associated vowel). It uses qamats for o on the model of Yiddish, rather than holam, so that all vowel points are beneath the consonants. Like the Arabic scheme, it uses `ayin as a null carrier for diphthongs. It uses gershayim for comma and 'alef for dot.
I'm not sure whether to use final-form consonants, especially since Lojban words tend to be vowel-final so the consonant isn't the last thing in the word... and consonant-final cmene are followed by a pause, which if written with 'alef looks rather funny since you'd have a final-form consonant followed by another letter.
I couldn't find anything useful for j so I picked tsadi, more or less at random. Oh, and I use khaf (with rafe) rather than het for x, for symmetry with k and the b/v and p/f pairs.
OK, I'll bite the question no-one's asked yet>> Why do these proposals use tsere (which in Sefardic Hebrew is a sound like lojbanic "ei") instead of segol, which is exactly lojbanic "e"? (c)het seems less confusing than khaf for x, and qof for k. — xwaver