everson at evertype.com
Mon Nov 22 11:42:22 GMT 2010
On 22 Nov 2010, at 11:29, j.mills at email.com wrote:
> I agree with Dan that Lhuyd is the only real source for establishing a minimal pair /ð/ and /θ/. Lhuyd (1707: 222) writes, “po kotha Lavar an Enez-ma” 'or the older speech of this island'. Lhuyd (1707: 250a) also writes “hui dho Kodha” 'that you fall'. So we have a minimal pair kotha ~ kodha. However, in order to accept this minimal pair, one also has to assume Lhuyd's testimony to be reliable.
I think we do, in initial and medial position, where there is a lack of inconsistency.
> If, on the one hand, one believes Lhuyd to be unreliable, then there is no reliable evidence for this phonemic dictinction.
Lhuyd is UNreliable where he writes "genev" which he could never have heard. It is quite plain that this is a parsing error based on his Welsh, generalized by him. He was not a trained phonetician as we are, and the error is neither surprising nor alarming.
> It follows then that we should write <th> everywhere. If, on the other hand, we accept Lhuyd's testimony as reliable, we adopt the minimal pair, /ð/ and /θ/, and write these <dh> and <th>. However one is disingenuous if one accepts Lhuyd's testimony only when it supports one's hypothesis, and rejects Lhuyd when his testimony doesn't suit.
We are obliged to explain his inconsistent use of -dh and -g in final unstressed syllables. In this context, I think that Lhuyd is not reliable, for the same reason that he is not reliable with -v in final unstressed syllables.
If this explanation does not convince you, Jon, please give us your own explanation for -ag/-ak and for -edh/-eth in Lhuyd.
Michael Everson * http://www.evertype.com/
More information about the Spellyans