Re: Radical-Metathesis

Dirk Elzinga wrote:
> But try it out and see what happens. One question I have is how do you know
> which permutation you started from,
Well, if you know the semantics of 'maisali' then you will be able to calque
all the other forms, since the semantics is related to the permutation.

But if I told you that the set {l,s,m} is the root of "move", you wouldn't be
able to tell the order. If that's necessary (which it might be, to make the
permutations of individual words to a single cross-root pattern(?)) then I
might use a default order for all consonants, so that they always appear in
alphabetical order in a certain form.

> and what do the permutations mean?
The grammar is mainly influenced by Latejami, you could say my conlang
belongs to the same language family as Latejami and Saweli.

The radical metathesis represents a 3x2 matrix:

not defined: lbr, rbl
static:      rlb, blr
dynamic:     brl, lrb

(so 'laibari' is a hypernym to 'railabi' and 'bairali')

focused*:  rlb, brl, lbr
unfocused: blr, lrb, rbl

*Presence of the semantic role that isn't agent nor patient.


ROGER MILLS wrote:
> [snip]
Thanks, I'm unfortunately limited to Internet, but I will check it out.

> My newest conlang, Prevli, does use metathesis in the way you mention, I
> think: CVCVC is the base form, final -VC met. > CVCCV forms realis of
> verbs, and nouns, initial CV- met. > VCCVC forms irrealis. I'm still trying
> to puzzle it out...........;-(
What is a noun in irrealis?


John Vertical wrote:
> I'm not so sure about juggling *everything* around - I assume you have a
> lot of ablaut too, as per course for triconsonantal systems?
Yes. At least I excluded consonant mutation from the soup. :)

bairali 'learn'
birelo  'inform'

> but I think a system where one of the root consonants is a "class affix"
> (only an entirely morphonological class) that could syntactically pop up in
> a number of places would not be too far-fetched. I'm working on something
> similar currently.
Sketch 8.2.1 uses a CC root, with a third consonants indicating the main part
of the valency. Something like that? Or do you mean the third consonant only
appearing sometimes? Or anywhere in relation to the radicals?


caeruleancentaur wrote:
> Further interest would be added if a null consonant were allowed.
> Then one could have:
>
> C0C > 0CC, etc.
>
> I assume that the vowels are placed between the consonants and thus
> diphthongs, or at least double vowels, would be created.
>
> Hypothetically: ra0ot = raot
Hmm, I have considered to allow semi-vowels in the root, which creates a
similar effect, since /w/ and /u/ melts together, and /i/ and /j/...


-- Veoler

Replies