Print

Print


On Tue, 8 Nov 2016 20:39:47 -0500, Alex Fink <[log in to unmask]> wrote:

>On Tue, 8 Nov 2016 18:52:38 -0500, Anthony Miles <[log in to unmask]> wrote:
>
>>The South American language Tariana has a phenomenon known as H metathesis,  in which a glottal fricative is metathesized to the  initial consonant of the root  regardless of how many syllables that may be in distance. Is this long-distance metathesis attested elsewhere  or is it unique to this language?  
>
>Basque has a rule of identical description: 

The Basque rule, I should have said, allows /h/ to be superimposed on a voiceless plosive, producing the complex phonemes /p_h t_h k_h/ _ph th kh_.  In fact it seems to prefer sitting on a voiceless plosive to sitting by itself, and this is stronger than its preference for initial position: so words like _ikhatz_ 'charcoal', _ikhusi_ 'see' don't have variants in _hik-_.  When there are two voiceless plosives, however, aspiration categorically moves to the earlier one: thus the word for 'bosom', which has an initial consonant of vacillating voicedness, takes the forms _golkho_ and _kholko_ but never *_kolkho_.  

(Also, all this is only for the northern dialects of Basque.  The southern ones have lost /h/ outright.)

I write again because I've just learned that Quechua and Aymara have a comparable rule to thàt, about co-occurrence restrictions on aspirated and unaspirated stops.  It's a lot like the left-right reflection of good old Grassmann's law in Greek and Sanskrit (and remember that the Paninian school of grammarians stated Grassmann's law in a form affected by rule inversion, whereby aspiration is *moved leftward* in a word if its original home becomes unsuitable!)  Here's a handy review by Mike Cahill of Margaret MacEachern's _Laryngeal Cooccurrence Restrictions_, which excerpts some of its data tables about which patterns of such restrictions are attested:
  https://linguistlist.org/issues/12/12-2990.html

Alex