[NGL] Translation: Ku Valaksiya à Weksaforda
Jack Durst
> >Perfectly acceptable, but there is an older derivation {ocodir} refering
> >to medicine in that sense as well as any perscribed medical device.
> I considered it, but for aesthetic reasons I wanted to exclude the
> possible sense of "non-chemical medical device".
Cool. It's your piece and you're certainly entitled to your own aestetic
judgement.
> [...]
> >I second {edč}, but believe that {hekai'} is unnessicarily redundant as
> >the use of {edč} would already make the distinction in most cases where it
> >would be used. However, I have no objection if someone else seconds.
> Well, I think that {edč} really needs {hekai'}, or else VXT vector
> length words, to go with it for some circumstances. For example, let
> us coin:
I'd have gone with a vector, myself, and that was *exactly* what I was
thinking of when I made the comment, though Ogden {korti} is still
perfectly acceptable as a compromise between making and not making the
distinction.
> *geruh V - scrape, shave
Second.
> {Ta edčm be geruh.}
> Which means roughly "he spent (some) time shaving." It doesn't mean
> much, unless you're making a catalog of where Buddy's time went. Then
> take:
Actually, the very vagueness could be usefull sometimes, but agreed that
it didn't convey much information.
{Ta edčm 45 sudu be geruh} conveys all the information anyone could ever
want with no need for a general word, of course, but the absence of a less
drastic option is felt.
> ...which has some clear narrative application. The use without {edč}
> is possibly like this:
> {"South Park" je sine inhčkai', newiy ta _naesite_ hekaig.}
One problem with your example: The double-letter rule seems to be applied
differently in your dialect. Tokcir would treat {'} as a letter for
purposes of the double letter rule, producing {inhčkai'i, hekai'ig}.
It would not be pronounced in the basic form *except* when the word
preceded a word beginning in a vouel or an H; it might be swallowed if
derivation put in a consonant after it ({hekaije} for example), but before
a vouel {'} would be required and pronounced, as in {edčhekai'ar}. As
Tokcir treats double vouels as long, the difference does exist in that
dialect, and {hekaig} would be just as much 3 vouels in a row (well,
actually 4 vouels as the I in -ig is long, too) as {edčhekaiar} is.
> >> The following are more on the nature of explaining an idiom I used
> >> than making a proposal:
> >> ge sasevodit - to drown oneself
> >This one is strange, since derivation already provides a means for saying
> >it better {sasevodic} but it is perfectly understandable and applicable.
> The definition listed for the modifier is "verb is made AP based".
> What does that mean in simple terms?
It means that the subject performs the action on itself. Non-directional
{tibe} is like this, too. {Tibeom.} means "I moved (myself)"
So, {limpaic} is equivilant to {limpafe xes} or {ge limpait} but shorter.
> [...]
> >> be sie'fe q vamaksiya
> >It's spelled {si'e} not {sie'}
> So noted. Is "sky" {sie'}?
{ci'e}, IIRC.
> >> & di incąxa mec inąco zupo.
> >I don't understand the need for a {di} in this line, {te incąxa} would
> >seem to cover the meaning just fine.
> It's because the doctor in this stanza is advising the old woman on
> what she needs to do to make the old man blind. In my view the mood is
> plainly advisory, so if the doctor will choose any modal at all, he
> will probably choose {di}; it's just as easy to say as {te} and fits
> to mood of what he's saying better.
I'd probably have gone with a single instance of mood/tense/aspect on the
first verb and just let the rest be generic and inherit their tenses
unless rythim required a repitition.
Sincerely,
Jack Durst
Sp...@sierra.net
[this posting written in Net English]
Jack Durst
> From: Stephen DeGrace <c72...@morgan.ucs.mun.ca>
>
> On Tue, 5 Oct 1999 00:42:04 -0700 (PDT), Jack Durst <sp...@sierra.net>
> wrote:
>
> [...]
> As I said, on an experimental basis, I wanted to try and get away from
> the natural language bias of conflating time and space and using the
> same words to describe both. I'm curious to see if it might pay some
> small dividend in pointing out different ways we can think about time
> - potentially valuable, because think about time is something a
> modern, post-industrial society does to the point of obsession.
It's your proposal, and, as I said, I have no objection to it on the
formal level. It is an interesting distinction and I support your right
to make your own astetic judgements; I simply disagree and so do not
second.
> _feeling_ of the *particular* speaker, without the need for anyone
> ever sitting down and saying, "Okay, this is a fair interval to
> roughly characterise as a "short" vector with regards to time in this
> particular frame of reference, this is a fair interval to characterise
> as a "medium" vector with regards to time in this particular frame of
> reference..." and so on. {hekai'} is _completely_ subjective - all
> that matters is what the _speaker_ _thinks_ about the interval. He or
> she, by employing it, is communicating purely a feeling or an
> *opinion* about a length (volume?) of time.
Perhaps it's because I'm the least experienced with vectors, but I've
never felt an objectivity of this sort in the length vectors. Perhaps
Jerry could explain it better... I had always thought that imprecise time
vectors were strictly from the referent's perspective as to the degree of
inspecificity; just as the difference between a near and far past in my
system is entirely based on the narrator's perception of events with only
very general guidelines.
> >> ...which has some clear narrative application. The use without {edč}
> >> is possibly like this:
> >> {"South Park" je sine inhčkai', newiy ta _naesite_ hekaig.}
> >One problem with your example: The double-letter rule seems to be applied
> >differently in your dialect. Tokcir would treat {'} as a letter for
> >purposes of the double letter rule, producing {inhčkai'i, hekai'ig}.
> Hmm. I have to confess that what is occuring is an automatic
> generalisation of something that used to be a noted exception. {Dia'},
> for example, is specificaly defined so that the ' gets dropped if
> possible in derivation. What it comes down to is the unspoken
> existence for me of two classes of words ending in ' - those for which
Interesting. For me, {Dia'} is the only such word, making it a simple
irregularity in the morpheme. I don't know for sure how I feel about the
existance of two such classes. I'd like to keep the orthography as
phonemic as possible, so keeping it to a simple irregularity in some
morphemes plus a spelling rule seems like the option I'd like most.
Though, looking at my own writings, I notice that there really is no
consistant rule as to weather the {'} should be swallowed in derivation,
even before a consonant. For example {u'těbe} is written with the final
{'} intact, while with a number of other words, the {'} is readily
swallowed before consonants, as in {emaci} "attraction". It's rare enough
that the corpus really dosn't reveal a pattern.
The only discernable rules in Tokcir at the moment are:
1. the ' is always included before a vouel or an H
2. the ' is never included when it's a derivational by-product and the
next derivation in line is valid without it
(ie. sangei' -> sangeig)
3. it is always included when it is part of a one-sylable word
Perhaps another rule to be added could be that when it follows a single
vouel at the end of a morpheme it's always both pronounced and written,
since, in that case, it could only logically exist as a real letter in the
word. My own writings don't support it, but it makes a good proposal.
> Long Timers" :-) ). This is the source of the difference you have
> noted, a sort of unstated rule operating half-consciously on an
> undeclared class in my dialect. It's a bad habit, really, although I
> do hasten to add that I think that it should be possible to specify
> irregularities in a word's derivation, as we have done with
> {-(s/t/st)(o/a/i)r} for example, that could potentially include
> specification of collapse of the glottal stop. Even so, things should
> be in general done consistently, and we shouldn't encourage, in my
> view, _too_ much proliferation of exceptions and special cases.
I certainly agree that it is proper to include irregularities where it
benefits the sound of the language to do so, and that it shouldn't be
encouraged without a higher standard of proof.
> Myself, I tend to make my {-ig}s short, {hekaig} winds up getting
> pronounced with the a and i almost separated with only a bit of
> blending in the middle, and I think it works out to only two vowels
Actually, that would be a long vouel in my dialect, as {-ai-} is a
dypthong in Tokcir (pronounced {-ay-}) unless one of the two vouels is
long.
> when I say it. But it makes me wonder if I could save some trouble by
> modifying my proposal such that {heka} becomes the adverbial form from
> which the adjectival {hekai'} is derived. That would make sense since
> a ig part of the proposed use of the morpheme is adverbial. I'll keep
> my proposal afloat for the timebeing, but I modify it to:
> *heka Av - much#little (time)
That's certainly better, since the existing rule would then apply in force
as its being the product of derivation. Chooding adverb as the bese still
has its problems, though, there are very few naturally adverbail morphemes
in the language for a reason: the derivational system is not as usefull
for them. Nominal or verbal use would be more derivationally productive.
Jack Durst
One point, though, the division of spacetime is by no means universal; its
non-division in many Native American languages was the very thing which
prompted Benjamin Whorf to consider the Sapir-Whorf hypothesis.
In Navaho the past-future division is non-existant on the grammatical
level, they effectively speak in what would be NGL generic tenses. In Old
English the distinction was twofold past/non-past with no aspect marking.
In biblical Hebrew the distinction is purely on aspect.
Likewise, time and space are not always seperated, some South American
languages use something like an underdeveloped VXT system wherin space and
time are treated by the same inflections. One might mark
past/present/future, but just as easily mark up/down/back/forward/
or left/right instead, with no distinction being made between tenses and
the other three dimensions except which is more important for the currant
narrative.
Modern English combines tense and aspect to the extent that they are
indistinguishable except in combination. Dakota goes so far as to include
"with a knife" as a modal. And, of course, there are nearly a hundred
languages which have *no* grammatical marking of aspect, tense or
direction, with those functions being marked by adverbs.
As with anything non-universal, the varriation of the world's languages is
stunning.
Jack Durst
> Well NGL is very lucky to have you as a linguist major among its
> founders. I bought a copy of The Languages of the World, I guess I'd
> better start reading it. I'm pretty sure though that loglan did say that
> the threefold division of space was universal. I guess they were wrong
> about that. Thanks much for the mini-lesson.
There are some universals in the area, but they more have to do with the
mechanics of the language than they do with the particular concepts.
There aren't always three dimensions and 6 directions, some languages have
up to 15 cardinal directions. Some languages include time as a dimension,
making 8 cardinal directions and 4 dimensions. One language I studied had
only two directions (something like clockwise and counterclockwise) and
assigned spacial orientations as an intersection of two circles (but I
don't remember if that one was a natlang or not)
> >past/present/future, but just as easily mark up/down/back/forward/
> >or left/right instead, with no distinction being made between tenses and
> >the other three dimensions except which is more important for the currant
> >narrative.
> Very interesting. This is very broadly similar to the structure of
> loglan. I didn't just clone loglan, however; I remapped space and time
> to vectors so that the quantified tense is so much more natural to deal
> with; in addition we have the assurance that if it works in math, it
> must work in language. Due to Stephen's input on this thread VXT now
> also distinguishes time intervals from space intervals when desired.
Frankly, I think the VXT system is vastly improved over what I've studied
of the loglan tense system. The idea of a mathematical vector is
extremely well suited to the description of space and makes a very
interesting way to describe time.
As for "if it works in math it must work in language" I can only tell you
I disagree there. Languages have a strange sort of logic all their own
which is sometimes at odds with other systems. In math, two negatives
make a positive, in language two negatives make an emphatic negative or
maybe a moderated positive, depending on the language. Examples like this
abound.
> >As with anything non-universal, the varriation of the world's languages is
> >stunning.
> Jack, what do you see as language universals, if any? (If you have time
> for this question)
There are a lot of universals in the world's languages, but,
unofrtunately, most of them are implications of other items or totally
useless. Some examples:
All natural languages have a set of consonants and a set of vouels, this
isn't strictly nessicary, as it is theoretically possible to have a
language consisting solely of vouels. Not only that, but all languages
have at least two of each, and almost all languages have more consonant
phonemes than vouel phonemes.
All natural languages have a formal division between verbs (which often
include adjectives and adverbs) and nouns.
All languages which admit a fixed word order without case marking have
some conformance to certain parsing principles. Likewise, all languages
which do not have a fixed word order have some sort of grammatical role or
case marking scheme.
Then of course there are statistical univarsals, things which *most*
languages conform to, like the fact that if a language has SVO word order,
it is likely to have prepositions while with SOV order it is likely to
have postpositions or locative cases.