Re: Phaleran number

Tom Wier wrote:

>(/a:/ is phonemic in Phaleran).
I take it then that the circumflex signals a long vowel?

>   Reduplication is somewhat more
>complicated.
I love reduplication. </random comment>

>    No longer a productive process, it was massively
>productive several centuries ago, and still affects lexical classes
>like body parts, most family relations, and most animate nouns
>(excluding lower orders of life like sea-slugs, most plants, arthropods,
>etc.).  However, the template that results in the reduplicant operates
>under stricter phonological rules than does the rest of the phonology:
>typologically marked phonemes like glottalic consonants and long
>vowels are simplified, as are consonant clusters:
So, the reduplicant template is a CV- prefix, with some kind of Emergence
of the Unmarked effect.


>     t'olka    'arm'                        tot'olka    'arms'
>     nemes    'sister'                     nenemes   'sisters'
>     yamna    'son-in-law'            yayamna    'sons-in-law'
>     fû           'domesticated         fufû           'dom. owls'
>                    owl'
>     tlîna        'black bear'            titlîna         'black bears'
>     thuran    'imperial janissary'   tuthuran     'imp. janissaries'
You mention a bimoraic minimum in the segment of the email I snipped. This
makes me wonder about your forms for 'domesticated owl': fû ~ fufû. Is the
vowel underlyingly long here? Or does the lengthening requirement disregard
the reduplicant?

I like the way you simplify your cluster in 'black bear'. It is more common
to see reduplicants contain strings that are also contiguous strings in the
base, but your system is well attested. Another possibility is to copy the
consonant immediately before the copied vowel; so tlîna would be litlîna or
tlilîna. If the later form looks wierd, that is exactly what Pima/Papago do
with clusters: tloogi 'truck' > tlologi. (or sometimes tlolgi; Pima deletes
the stem vowel if that will not result in an impermissible cluster).



Marcus Smith

Unfortunately, or luckily,
no language is tyrannically consistent.
All grammars leak.
                -- Edward Sapir

Reply