At 8:58 PM -0500 01/30/02, Doug Ball wrote:
>Etona a Dirk (Tirik for monolingual Skerre):
>>  I thought I'd introduce some features of a new project that is
>>  underway. This is something that I don't usually do; while I enjoy
>>  the discussions here, I tend to view conlanging as a solitary
>>  activity, and I have been reluctant to share volatile features of
>>  projects in progress. So I'm new at this :-). That said, here are
>>  some things I find interesting in my latest project, tentatively
>>  called Ustekkli ["ust@kli] (nominalized plural of the perfective
>>  _ust_ of _usta_ 'speak', hence 'words'. I know, I know, but I gotta
>>  call it something).
>Based on the recent discussions of what names conlangs begin with, you
>seemed to have picked a good name, since few conlangs start with letter "U."
>(Uatakassí is the only one I can think of).

:-) The name will change eventually.

>  > Inventories
>>  consonants
>>  p       t               kj      k
>>  b       d               gj      g
>>  f       th      s       sj              h
>  > m               n
>  > r,l
>  > w                       j
>>  The consonants are presented in the orthography with one exception;
>>  <th> is written with <thorn>. <kj, gj, sj> are alveopalatal
>>  consonants rather than clusters; <kj> and <gj> are voiceless and
>>  voiced affricates, respectively, <sj> is [S]. <j> is jod.
>Why not have /j/ be written as {y} as to avoid any possible confusion with
>the {kj}, {gj}, and {sj} digraphs?

I think there is an etymological reason for this mapping; the <kj>,
etc were in fact clusters of /k/ and /j/ at one time. I'm not sure
how much this will matter since I only have one word so far with <sj>
-- meisje ["me:S@] 'be still, quiet; schweigen' and none with either
of the affricates.

Another consonant inventory note; I think that there should also be
voiceless liquids alongside the voiced; orthographically these are
<hr> and <hl>.

>(snip Vowels and some phonology)
>  > syllable contact
>  >
>  > Here is how syllable contact plays out. In Ustekkli, stressed
>>  syllables must be heavy; this is accomplished by gemination or vowel
>>  lengthening. The choice between the two depends on if there is close
>  > contact or loose contact between the vowel of the stressed syllable
>  > and the following consonant. If there is close contact, the consonant
>  > is geminated; this creates a heavy syllable:
>  >
>>  If there is loose contact between the vowel of the stressed syllable
>  > and the following consonant, the vowel is lengthened, which also
>  > creates a heavy syllable:
>  >
>  > nikk-r  /nikr/  ["nIk.k=r]  (close contact)
>  > nik-r   /nikr/  ["ni:.k=r]  (loose contact)
>>  oatt-n  /oatn/  ["wat.t=n]  (close contact)
>>  oat-n   /oatn/  ["[log in to unmask]]  (loose contact)
>Can you illustrate step-by-step how these are derived?  I'm still a bit
>unsure how close contact and loose contact work in action.

Well, it's not so much a matter of derivation as it is of phonetic
realization. There is an underlying distinction between close and
loose contact, but this distinction is only realized in stressed
syllables. For unstressed syllables, there is no phonetic realization
of contact. If the syllable is characterized by close contact, it
will be realized as gemination when stressed; if the syllable is
characterized by loose contact, there will be vowel lengthening when

It's also a problem I'm working on in Norwegian phonology (I'm taking
a break from Shoshoni for a while); there is a requirement in
Norwegian (at least Eastern Urban Norwegian) that stressed syllables
are heavy, but the implementation of this stress-to-weight
requirement varies lexically. Some stressed syllables have a long
vowel, others are closed by a consonant. This mutual dependency of
stress and vowel length/gemination seems to have provided quite a
puzzle for researchers looking at the phenomenon, but to my
knowledge, no one has proposed syllable contact as a possible
analytical strategy. In that sense, Ustekkli is partly a workshop
where I can try out various analyses of the Norwegian pattern.

A question you didn't ask (but could have), is how syllable contact
is to be represented in underlying forms. I haven't worked this out

>(snip nouns--unfortunate use of the term "absolutive" in Uto-Aztecan--do the
>Ustekklists use a different term to avoid the confusion?)

This use of the term 'absolutive' is traditional in Uto-Aztecan. I
think the term works here, and I don't plan to change it right away.
When a native Ustekkli grammatical tradition emerges, we'll have to
see what term will be used to describe these suffixes.

>  > (Object prefixes and truncation are shamelessly stolen from O'odham,
>>  a Uto-Aztecan language of central and southern Arizona and
>>  neighboring Mexico.) There is a derivational process of reduplication
>>  which carries an iterative meaning; so from _hipi_ 'drink' there is
>>  also _hippi_ 'sip' (some opaque phonological alternations going on
>>  there), and from _bieme_ 'jump' you get _bibieme_ 'hop'.
>What about reduplication of 'usta'--how does Ustekkli reduplicate
>vowel-initials? (couldn't resist this question :-))

I don't know. I suspect though that since this reduplication is
irregular (_hippi_ from _hipi_ alongside _bibieme_ from _bieme_) it
is also sporadic and won't apply to all verbs even when it might be
semantically sensible to do so.

>The syntax seems to be vaguely like O'odham, too, although with some
>differences--Ustekkli keeps initial determiners, but if I remember things
>correctly from my experience with O'odham, O'odham doesn't.

Well, the similarities to O'odham syntax are more than vague; I
lifted it entirely. I'm expecting to see differences emerge as I work
on it (you've already noticed one), but the goal for Ustekkli wasn't
to wow the world with  exotic grammar that I made up All By Myself,
but to establish a working language that will be part of a dimly
imagined Northern European culture. Perhaps Ustekkli is the modern
descendant of the language of the Pre-IE Europeans ...

>  > Not too exciting right now, but there are some promising features, I think.
>Depends on what you consider exciting, but I think Ustekkli is pretty
>interesting already--I've never seen any language manipulate syllable
>contact, so that's interesting, and it's always nice to see a slightly
>less-than-configurational AUX2 language.  So, please explain more, as you
>develop it and have a chance to either post here or put something up a on

Thanks for your interest.

Dirk Elzinga                                            [log in to unmask]

"Speech is human, silence is divine, yet also brutish and dead;
therefore we must learn both arts."
- Thomas Carlyle