Saturday, October 4, 2014

The Name Problem

I think I just had a name epiphany. Names are a special POS in Ceqli, behaving like pronouns, so it's a good idea for them to have their own phonemic shape. For some time now, the rule has been that you can convert a morpheme to a name with the suffix -zo.

Jan - know, to know
Janzo - John

And to make names more compact, I decided to allow the sort-of suffix -t to do the same thing:

Janzo = Jant

Bad idea.  For one thing, it messes the phonetics up, requiring a big pause after the -t so as not to blend with any following sounds. For another, it doesn't fit well after q or m, and has to be come -et in such situations, making everything that much more complicated.

Simplifed:

The -zo ending works just fine, and can also accommodate long names with the be- reversal particle trick, thus.

Da bekyam bezo Franklin Delano Rozevelt zo.

Let's say that -zo, in most cases, can be replaced with -z.

Jan - to know
Janzo - John
Janz - John

All that is needed now is a rule that in the morpheme shape nCnF (one or more leading phonemes followed by one or more following phonemes), Z cannot be followed by another C. That means one of my favorite words, zban, has to be dropped. There will be no zb, zd, or zg combos.  That means when you come to a -z, and the next sound is not a following phoneme, you have the end of a word with -z, and that word is a name.

This also has the upside of resembling the Loglan use of -s in the same way, and also the English tendency to make nicknames from sort of a pseudo-plural, like "pickles" or "freckles." I figure that choosing between Janzo and Janz will be a matter of usage.

To replace zban, I propose kel, from Gaelic.  Kel = spouse, keli = wife, kelo = husband.