Actually th'other undergoes fortition all the way to the stop, at least in my usage

Neither one not t'other.

----Original message----
From : [log in to unmask]
Date : 20/09/2017 - 13:47 (BST)
To : [log in to unmask]
Subject : Re: YAEPT word-initial eth

I don't suppose the OED mentioned what group has initial θ in "thither"?
That could be assimilation, I suppose, but it feels like a
spelling-influenced example (two th's in the same word must obviously spell
the same sound, right?). As I said, my intuition went the other way for

On Wed, Sep 20, 2017 at 05:32 And Rosta <[log in to unmask]> wrote:

> On 19 Sep 2017 20:17, "Mike S." <[log in to unmask]> wrote:
> On Mon, Sep 18, 2017 at 7:08 PM, And Rosta <[log in to unmask]> wrote:
> >
> > In PDE, there is -- I would contend -- no syntactically coherent way of
> > capturing word classes containing eth-initial words but excluding
> > prepositions. Whether that holds also for the th-voicing stage of the
> > language, I don't know, but a reasonable default hypothesis is that it
> > does.
> >
> It seems to me all the affected words are +definite in meaning: the
> definite article, the demonstrative pronouns, adjectives, and adverbs, and
> even "thou" and kin.  IMHO what we have is not a syntactic condition on a
> phonological change, but a semantic one.  The few words -- like the
> conjunction "than" -- that don't seem to fit the +def constraint are bound
> up with words that do.  "Than" is a twin of "then", and the primary sense
> of "then" is "at that time", which is +def.
> Under this view, "through", "three", "thousand", etc. were never going to
> change simply because they are not +def in meaning, nor related to words
> that are.
> As Peter Bleakley suggested, the ultimate origin of this odd rule is
> perhaps related to the regular use of some of the th-function words,
> especially "the", in unstressed positions.  The new pronunciation became
> default for these words, became associated with definiteness, and spread by
> analogy to related forms.
> It's an ingenious idea, but the obvious problem cases are _than_ and
> _though_ and perhaps clausal _that_ (/ðət/). I don't see in what
> (pertinent) sense _than_ is a twin of _then_.
> On Tue, Sep 19, 2017 at 9:31 AM, And Rosta <[log in to unmask]> wrote:
> >
> > /θɹti/ and /ðɪðɹ/ are the pronunciations I had in mind. But I discover,
> > from checking OED, that there is indeed a US /θɪðɹ/, which then joins
> > _thirty_ as an example of initial simplex /θ/ in closed-class words).
> >
> >
> I do not recall hearing /θɪðɹ/, but surely this anomalous pronunciation
> occurs only because "thither" has dropped out of colloquial use and
> reappears under the influence of old writings.  Sometimes, it reappears not
> properly "reintegrated" into the paradigm of demonstrative pronouns and
> such.  That's certainly the way I would interpret that pronunciation --
> it's not a holdout from the ancient change but merely the result of
> speakers tripping over an out-of-date word.
> Yes, I wouldn't consider it a holdout. But it's a significant datum. It
> confirms my contention that there is no prohibition against initial þ in
> function words. It also might suggest that for þither speakers (who, being
> American, lack _thicky_ and _thonder_), initial ð is restricted to
> monosyllables and hence initial þ and ð are in complementary distribution
> (i.e. ð- in monosyllabic function words and þ- elsewhere).
> --And.