Re: Arrghhhh!!!! and a Summary of What has Gone Before Saul Epstein Wed, 26 Nov 1997 23:36:12 -0600 From: Rob Zook Date: Wednesday, November 26, 1997 6:57 PM > Now that the mindless emotive noise is out of the way..,I wonder if I'm > the only one impatient at the pace of things around here. Gee, the last time I thought about it -- which, admittedly, was over a week ago -- I remember being almost giddy at the progress made in the preceding month... > Anyway, also I wonder now what we should do next. We need to flesh out > the vocabulary and the grammar. Naturally fleshing out the vocabular > seems a necessary first step before we can flesh out the grammar. Actually, we could "finish" the grammar with as much vocabulary as we now have. Which is not to say that we should wait on the vocabulary, only that we can expand the grammar along the way, without having to wait. > We can > create words by random manual construction, by random automated > construction, or by a statistical automated construction (based on the > words we have). > > I assume since I've mentioned the automated construction several times > that no one has any objection to doing that? I'll assume that's what we > want to do from now on unless someone voices some objections. I'd really > like to hear Marketa's opinion on this idea though (hint hint). Yes. She didn't say what she was going to hospital for... I hope everything's OK. Personally, I think a judicious combination of a statistical automated construction method with an intuitive manual construction method will serve us very well. > Anyway as a prelude to expanding the vocabulary we needed to analyze the > bits we do have, which Marketa and her father have constructed and culled > from various sources (canon and not). Saul has firstly come up with an > alternative transcription system for writing Vulcan with English letters: > > Vowels: > > Zvelebil Epstein > Notation Notation > i i > u u > e e > o o > ^ ^ > a a > i: ii > u: uu > e: ee > o: oo > ^: ^^ > a~ a~ > e~ e~ > o~ o~ Pretty much. I'm still not sure about those nasalized vowels. I don't remember if we asked Marketa to ask about them. If we can't get a statement one way or the other from the expert, we'll just have to decide and personally I'm for making them allophones of the other vowels. Or keeping them and eliminated the phonemicity of the "length" distinction. > Consonants: [snip] > I only have one afterthought to Saul's system. I would rather see > as , which makes it consistent with and gets rid of another > instance of that silly little ~ thingie. Well, it's a perfectly dignified thingie, except when it occurs by itself in a character-space. It looks like an operator, or a scripty hyphen. The only reason I kept it in this case is that the tilde-superscript n is the recognized symbol for a palatal nasal stop. That's different from "ny," which could be a palatalized dental nasal or a full-blown nasal+approximant. I'd like, if at all possible, to avoid situations in which we use a sequence of letters for a distinct sound if each letter in the sequence also represents a distinct sound. "ng" bothers me for the same reason, but I don't know what to do about that... > Also, Saul has indirectly > proposed the adoption of general linguistic terminology as much as > possible. Which among other things means letters surrounded by <> refer > to spelling conventions, letters surrounded by [] refer to phonetic > spelling, and letters surrounded by // refer to phonemic spellings. Ah, uh, hmm. Yes, that's all correct. And I'll use them a lot, so it will help others if they know them. But I honestly don't look to have them adopted by others, though it could help us understand what others are saying. While a degree of technicality is extremely useful at times, it can also be frustrating and off-putting. In short: don't feel bad if you can't figure out what "allophone" means. Once we get the phonology worked out, it won't really matter. > Next we extracted what we think are at least *some* of the allowed word > initial, and word ending consonant clusters: > > Word Initial Consonant Clusters: [snip] > [approximate]+[tap]: > lr yr wr Yikes. How did those sneak in? > Word Ending Consonant Clusters: [snip] > [approximate]+[lateral] > yl wl ywhl wwhl > > [fricative]+[lateral] > fl vl sl zl cl jl xl hl > fwhl vwhl swhl zwhl cwhl jwhl xwhl hwhl And these? I'm not looking forward to trying to pronounce any cluster with /whl/, in any position. > [aspirated stop]+[stop] > khp khb kht khd khk khg khq > th thb tht thd thk thg thq > dh dhb dht dhd dhk dhg dhq These on the other hand, look better every time I see them. I mean, written they still give off this initial sense of impossibility, but spoken they sound very good, very Vulcan. We may want to impose a same-voicing rule on them, as for the fricative+stops. And I think they're probably permitted at the beginnings of words too. > Now far as I can tell, we only need to do two things before we can > start creating words the easy way. We need to know how to form > syllables, and we need to finalize a frequency distribution for > each of the following: > > 1. phonemes > 2. consonant clusters > 3. syllable forms and diphthongs, which we haven't touched. > I'm going to suggest that we leave the frequencies for the phonemes > pretty much as they are (except for increasing the values on the > ones which do not appear in the dictionary to say 10%). I think in addition we should boost each of the "long" vowels, especially aa. I think in a lot of cases the coiners of the terms in our core vocabulary probably meant our /aa/ when they wrote . > Then letting items two and three have unary frequency distributions > now, and then doing some trial runs to see what kind of goofy > words appear and then adjust the frequencies for them as we try and > eliminate goofy looking words. Goofy of course meaning something > highly unpleasant looking as defined subjectively by anyone who > complains about any particular word ;-) > > We can keep the words which crop up in these preliminary runs which > we like and assign them some meaning either manually or at random. Sounds very good. > Now, that I've burst out with all I needed to say, for Vulcan's sake > SOMEBODY ELSE SAY SOMETHING DAMMIT! Ha. First I was waiting for Marketa to come back. Then I took apart my home computer to replace a part and had an interesting time getting it to operate once I had it put back together. Once that was taken care of, I got back to work on the web version of my morphemic analyses, which is very nearly done. You can check it out at http://www,uit,net/liberty/vulcan but you might want to wait till it really is finished. My purpose with it is to give a consistent presentation of a group of sentences and phrases, along with translations presented in a consistent fashion, and English paraphrases. So that we can really start to get a feel for how the language works on the word and sentence level. Once it's done, I'll be curious to know how clear and/or helpful people find it. -- from Saul Epstein liberty*uit,net www,johnco,cc,ks,us/~sepstein "Surak ow'phaaper thes'hi thes'tca'; thes'phaadjar thes'hi suraketca'." -- K'dvarin Urswhl'at