Re: Phonologically redundant vocabulary

Hi Jim & Henrik,

On Thu, 13 Apr 2006, Jim Henry wrote:
[snip]
> > > > > http://www.pobox.com/~jimhenry/conlang/redundancy.htm
> > >
> > > > Nice stuff. :-)
> > > >
> > > > At the time of the thread, I was also thinking of a engelang with a
> > > > self-regregating morphology plus redundant word building.  These
> > > > are nice tools for implementing a beast like that.
[snip]
> [on higher redundancy requirements for longer words:]
> > For simplification of the complexity of the problem, how about using a
> > two-level algorithm: first compute a set of syllables with redundancy,
> > and then use these syllables as the alphabet for making a word with
> > redundancy on the syllables.  You then get a two-level redundancy
> > phonology: each pair of syllables is different in the minimum amount
> > of redundancy you choose, and each word is redundant wrt. syllables
> > used.
> >
> > E.g. a very simple example:
> >
> > Step 1:
> >   Slot 1: ptk
> >   Slot 2: aiu
> >   =>
> >   pa
> >   ti
> >   ku
> >
> > Step 2:
> >   Slot 1: pa ti ku
> >   Slot 2: pa ti ku
> >   =>
> >   pati
> >   tiku
> >   kupa
> >
> > Of cause, this is not yet self-segregating, but you could change the
> > slots to achieve that.
Yes, it seems sensible to have two levels of analysis
in your algorithm - otherwise translating constraints
on syllables to constraints on characters may become
painfully opaque.


> Yes, this is good... It reminds me of an early version of this engelang
> that I had almost forgotten about.  I think in that one I had a simple
> phonology that generated 36 redundant syllables; 24 of them were
> monosyllabic words, 6 were dedicated for use as the first syllable
> of disyllabic words, and 6 more as the second syllable of disyllabic
> words.  (That still didn't leave room for more than 60 words in the
> whole language -- more parsimonious than even Toki Pona --
> but in retrospect I reckon the same system could allow arbitrarily
> long words if you use any number of syllables from the initial set
> followed by one syllable from the final set, or vice versa.  Or maybe
> the particular syllable chosen from the initial set tells you how many
> syllables to expect to follow?)
The obvious way to do that is to have every
word begin with the count of syllables that
follow.  Eg, assuming only CV syllables, and if:
0 = ko
1 = sa-sa
2 = sa-du
3 = sa-ti
...
9 = sa-ba
10 = du-sa-ko
11 = du-sa-sa
...
100 = ti-sa-ko-ko
...
123 = ti-sa-du-ti
etc,

the word whose meaning is given by the 12
syllables:
ra-fi-to-pu-na-wa-mi-ha-xi-pu-ta-ra

is :
du-sa-du-ra-fi-to-pu-na-wa-mi-ha-xi-pu-ta-ra.


> I think it was something like:
>
> general syllable shape C(S)V(N)
>
> monosyllabic words: CV, CSVN
> initial syllables: CSV
> final syllables: CVN
>
> I may well go back and try more permutations of
> this scheme, now that I have better scripts.
A good tool is always cool ...

Regards,
Yahya


--
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.385 / Virus Database: 268.4.1/311 - Release Date: 13/4/06