Re: Tekem, the language (aka deriving verbs from nouns)

En réponse à Amanda Babcock <langs@...>:

> With all the help I've received on all-noun languages and deriving verbs
> from nouns, I figure it's only fair to show everybody what I've got so
> far
> with the language.
>
I was waiting for this :) .

>
> I suppose that means that it's the phonemics which is minimal, not the
> phonology after all.  I'll do phonemes first, and then get into the
> sound
> changes.
>
> The syllable structure is [C]V[S], where C is {p,t,k}, V is {a,e,o}
> (didn't want to go with a,i,u), and S is {s,l,m}.  This gives the
> following
> syllables:
>
>         a       e       o       pa      pe      po
>         as      es      os      pas     pes     pos
>         al      el      ol      pal     pel     pol
>         am      em      om      pam     pem     pom
>
>         ta      te      to      ka      ke      ko
>         tas     tes     tos     kas     kes     kos
>         tal     tel     tol     kal     kel     kol
>         tam     tem     tom     kam     kem     kom
>
Yeah, very minimal phonemics, only 9 phonemes :) (I think it holds the record
with Hawaian and Tahitian :) ).

> Sound changes occur: BETWEEN morphemes WITHIN words.  Bwahaha :)  If any
> human language actually does this, I'll be pleasantly surprised.
>
Well, in fact I don't see what's unnatural about it. To me it's plain normal
that if you put together a stem and an affix, some phenomena of assimilation,
dissimilation or whatever take place. Now, I'm not a specialist of sandhi, so I
may say stupid things...

> I am arbitrarily defining words as utterances which can stand on their
> own as answers to a question (though not necessarily as spontaneous
> utterances).  This includes things like nouns ("person"), nouns marked
> for a semantic role in a sentence ("language.FOC"), nouns with a copula
> stuck on the end, which can be said to act like verbs if one insists
> ("giving.FOC.be", to be a gift), nouns marked for a semantic role and
> renominalized ("giving.AGT.that", that which gives, giver; not sure if
> I'm keeping this as a separate category from noun clauses, but it's
> currently
> expressed differently than a real relative clause), and some clauses
> where
> the noun can incorporate into the verb derived from another noun
> (example
> further on).  Not all clauses can be expressed as single words, however.
>
Wow, that's a nice thing to define what's a word in a certain language :) . In
fact, I don't think we can come to a better definition of that word :) .

> Here are the sound changes.  Bear in mind that stops occur only
> syllable-
> initially, s, l, and m only occur syllable-finally, and the sound
> changes
> only occur at morpheme boundaries.
>
>         stops after vowels become voiced fricatives
>         /s/ before a vowel becomes [Z]
>         /s/ before a stop becomes [S]
>         [h] is inserted between two vowels
>         /l/ or /m/ before vowels are palatalized ([lj], [mj])
>         stops after /l/ or /m/ become voiced
>         /m/ assimilates to [n] before the dental (or dental-alveolar)
> stop
>         /m/ assimilates to [N] before the velar stop
>
> I'd like to have /l/ change into [r] in some context as well, but I
> don't
> know which.
>
In front of stops, so that al-ta -> arda? I don't think it's necessary but it
could be neat :) .

>
> Now for some grammar and vocabulary, pretty much in the order in which I
> came up with them.
>
>
[snip of derivational stuff]
Wow! Nice derivations. Add a little culture-oriented irregularities for words
whose derivations don't have an immediate sense and it becomes very nice! :)


> Now we come to the derivations that act as though they're deriving from
> verbs.
>
[snip]

Wow, I like it very much! Well, since I recognize a lot of my idea in it, I
should :) . I very much like the idea of renomalizing utterances with a role
affix to make derived nouns, as well as this copula affix, it's very much in
Kristian Jensen idea that trigger languages only have copula sentences :) . I
wanted to come up with something like that for my Itakian, but I didn't manage
unfortunately...

>
> Maybe one solution would have been to move the role marker off the
> "verb"
> and attach it to the trigger, moving the trigger in front of the "verb",
> and attaching "-tal" to it: "tekeNges katoNgoDal".  But I decided to do
> something more fun: noun incorporation!  I know almost nothing about
> noun
> incorporation, so this may very well be more alien syntax coming up.
>
[snip]

Well, your incorporation system looks quite correct to me, since you incorporate
only bare stems, which is usual (my incorporation system in Chasmäöcho is much
more alien, since it incorporates words along with their affixes :) ).

>
> And that's where it stands, as of last Friday actually (lost half the
> email message and had to recreate it).  Apologies for any messed-up
> terminology.  It's been 7 or 8 years since I took a ling class.
>
Looks nice to me. We must have the same messed-up terminology :) .

Christophe.

http://rainbow.conlang.free.fr

Reply