Re[2]: [tied] *kW- "?"

From: Brian M. Scott
Message: 40034
Date: 2005-09-16

At 8:21:18 PM on Thursday, September 15, 2005, Patrick Ryan
wrote:

> From: "glen gordon" <glengordon01@...>

>> This 'unmotivated' palatalization of /k/ shouldn't be
>> shocking. Some dialects of English do this whereby "car"
>> is pronounced with palatalization of "c" before "a".

Which of course also happened in the development of Latin
<carrum>, <cause> to French <char>, <chose>; Latin /k-/
followed by /a/ or /au/ was already [t'] (palatalized [t])
in Gallo-Romance.

>> Notice also the variation of the word "Tuesday" or
>> "tune". Historically we know that it is /t-/, not /tj-/.

<Tuesday> is a bad example: in ME it had /Iu/, which
regularly developed into /ju/ after /h/ and /k/ and
fluctuates between /ju/ and /u/ after /t, d, l, n, s/. Here
the palatalization has an obvious source.

>> Spontaneous palatalization strikes again, bwahahaha!

> Nothing but nothing is unmotivated in language development
> or in any other historical process.

This looks like an article of faith. Certainly it would be
difficult to support on any other basis. It isn't even
really clear that 'motivated' can be usefully defined. (On
this general topic I recommend the discussion in Chapter 7
of Roger Lass, Historical Linguistics and Language Change.)

> It is supremely important to retain the palatalized
> dorsals where we can identify them because they allow us
> to know that the pre-PIE vowel in that position was /e/.

This is an argument for retaining the distinction between
*k^ and *k; it has nothing to do with their phonetic values.

> 'Markedness' is a useless concept. If it had any
> legitimacy, Khoisan could not exist with its very "marked"
> clicks.

This is an absurd straw man.

Brian