LINGUIST List 2.439

Mon 26 Aug 1991

Disc: Wh-

Editor for this issue: <>

Directory

1. Tom Lai, 'th' in English
2. "Michael Kac", Re: Why 'wh-'
3. Jason Johnston, Re: Why 'wh-'

Bruce Samuelson writes:
>Another example of phonological-semantic congruence is 'th' words in English:
>phonological semantic phonological semantic
>'th' is voiced functor word 'th' is voiceless content word
>the, this, that, these, then, think, thistle, thermal, thaw, three,
>there, their, therefore, thus, theory, thumb, throw, threw...
>thee, thou... through (a counter example)
The voiced functor words given above have heterogenous etymology.
'the', 'this', 'that', 'these' 'then' and 'there'(?) have German cognates
all beginning with 'd'.
'thee' and 'thou' do have German counterparts, but they also correspond
to second person pronouns (singular) in other IE languages (cf. Latin 'tu').
As for the voiceless 'th', 'thermal' and 'theory' are late borrowings while
'think', 'three' and 'through' again correspond to German words
beginning with (voiced) 'd'.
 Tom Lai

In regard to the nascent discussion as to why sound change occurs, I wonder
if the more perplexing question is why it DOESN'T occur. Why, for example,
are there still languages (English, for one) in which you can get velars be-
fore front vowels?
Michael Kac

Bruce Samuelson asks how the congruence "voiced th- starts function word,
voiceless th- starts content word" got started. Old English had both voiced
and voiceless spirants, but they were positional variants. Subsequently,
voicing (which is a form of weakening or lenition) was generalized in
precisely those forms that are typically *unstressed*, i.e. function as
opposed to content words. In a similar way, "of" and "off" are both
reflexes of a single word (cognate with German "ab"), again with the voiced
fricative appearing in the typically unstressed member of the pair. Of
course now the distinction has become phonemic (minimal pairs "thy-thigh",
"of-off") and is maintained regardless of stress.
All this is the standard textbook explanation. Still, I'm not sure why the
same process hasn't been generalized to voice the initial consonant of
words like "for", "so" and "such", or the finals of "if" and "this". I
imagine the functional/semantic coherence of the voiced th- words does have
something to do with it.
Jason Johnston
Dept of Linguistics, F12
University of Sydney, NSW 2006