Print

Print


staving Alex Fink:
> Just as it says on the tin: I've just realised I have no idea how the
> English rule that initial<x->  is /z-/ might've come about.
>
> There's nothing about /ks-/ motivating voicing; for that matter, Greek had
> initial /ps-/ too, and we're perfectly happy to leave that voiceless in
> borrowings.  There's no earlier-established context I can think of in which
> <x>  can be /z/ that it might have spread from;<x>  can be /gz/
> intervocalically in Latinate words, but is that really enough?  I thought it
> might have something to do with the initial loans passing through German,
> where of course original initial *s- has become /z-/ and the pairing<s->
> /z-/ is productive, but as far as I can make out (correct me,
> German-speakers) initial<x->  borrowings get /ks-/ there...
>
> Alex

Initial <x> is only found in loanwords or learned coinings in English. 
Neither /ks/ nor /gz/ is permissible word initially in English 
phonotactics. I imagine that the rule is that impermissible initial 
clusters simplify to their last element, and that <x> is interpreted as 
/gz/, simplifying to [z], because of the following vowel.

Note that French seems to be more tolerant of odd initial clusters in 
Greek borrowings (eg /psikOloZi:/). How is "xylophone" pronounced in French?

Pete