Re: Agglutinativity Index (was: Re: What's a good isolating language to look at)

On Fri, 9 Dec 2005 01:15:37 -0500, John Quijada <jq_ithkuil@...>
wrote:

>I'm not understanding something about this synthesis index.  Do zero-marked
>morpheme values get counted when determining the index?
No.

If you were calculating a fusion index (average number of meanings per
morpheme), a "zero-morpheme" could introduce a fraction with a zero
denominator and a nonzero numerator, making the whole average infinite.

Also, the problem could come up:
Lion = Lion-MASC-SING
Lioness = Lion-FEM-SING
Lions = Lion-MASC-PLUR
Lionesses = Lion-FEM-PLUR

Is "Lion" Lion-0 or Lion-0-0?  Is it 0, with MASC-SING fused, or 0-0, with
MASC and SING agglutinated?

Furthermore, for the synthesis index,

Lions = Lion-0-s because the 0-morph marks them as MASC
so "Lions" has three morphemes;

Lioness = Lion-ess-0 because the 0-morph marks her as SING
so "Lioness" has three morphemes;

Lion = Lion-0? or Lion-0-0? Does it have two morphemes? or three?

Anyway the average is between 2.75 and 3.00 morphemes per word, if we count
the zero morphemes.

So, it makes no sense to include the zero morphemes, either in the
synthesis index, or in the fusion index or agglutinativity index.

---

Read
http://scholar.google.com/scholar?q=author:"Pirkola"%
20intitle:"Morphological%20typology%20of%20languages%20for%20IR"
if you want;
I've printed it off, so I can read it.

Tom H.C. in MI

Reply