Re: Squishing Lexemes together into one syllable

Thanks for the quick reply and welcome.

  >How to you construct the result?  Am I missing something?

  No, I just made those up to show what I meant.

  >
  And for a phonology that is that simple (say, 5 vowels + 20
  consonants), you'll exceed the available syllable when you define only
  six roots as prime numbers: 2, 3, 5, 7, 11, 13.  Here, 11 * 13 cannot
  be mapped uniquely to one of 100 syllables anymore.
  <

 Hmm, I didn't realize that. (Neither can 11 * 11 be done, either.) But
repeated roots are normally nonsense, so we can get rid of all composites with
repeated numbers in their prime factorizations (i.e. leaving only square-free
numbers). The ordering of different roots can also mean different things. That
opens up a lot of space for... for... bla, forget about prime numbers. My point
is that I think there still lots of room with a phonology as simple as that.
Don't forget diphthongs (languages without them (Ygyde) are few)!

  Leon

Henrik Theiling <theiling@...> wrote:  Hi!

Leon Lin writes:
> Hello! I'm new here.
Welcome to the list! :-)

>...
> According to Wikipedia: A synthetic language, in linguistic typology, is a
> language with a high morpheme-per-word ratio. How about high lexeme-per-word
> ratio languages (is there a name for this (polysynthetic?))? We already have
> those, like the oligosynthetic Ygyde. But Ygyde makes a whole new syllable
> for every lexeme added. I was thinking putting them into one syllable (or
> how ever many syllables the original lexemes had themselves):
>
> ba + bi = be?
> pi + ku = ty?
> maba + qubi = nobe?
> mama + papa = ??
How to you construct the result?  Am I missing something?  Or is this
already some mathematical structure that produces seemingly arbitrary
syllables that are not connected phonologically?  The above looks like
you are doing this phoneme-wise by some 'average' phoneme, however
that is defined.

>...
> Mathematically this problem is quite simple. We can use prime numbers for
> root lexemes and composite numbers can be the compounds of the prime numbers
> that multiply to it. Thanks to the Fundamental Theorem of Arithmetic, every
> number has a unique prime factorization.
Of course, the numbers grow quite quickly this way and so you need
longer words.  If you take Ygyde, which uses *all* its possible
syllables as roots IIRC, there is nothing left to use for composite
numbers.  And for a phonology that is that simple (say, 5 vowels + 20
consonants), you'll exceed the available syllable when you define only
six roots as prime numbers: 2, 3, 5, 7, 11, 13.  Here, 11 * 13 cannot
be mapped uniquely to one of 100 syllables anymore.

Do I misunderstand your idea?

Anyway, since human languages hardly work on numbers but on
phonological units, this is quite a theoretical construction not used
in natlangs.

> I am interested in such languages but so far I haven't found
> one. Maybe I'll be the first to make one.
That's what conlangs are for. :-)

**Henrik


 __________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com