On 27/02/2018 00:12, Fog Flowerzone wrote
>> Yet Frankish _þwalja_ became_toaille_ in Old French with initial 
>> [t]. Why?  Had continental Germanic already begun th-stopping as 
>> early as the 5th/6th centuries AD?
> I would think this is due to the fact that [θ] is an allophone that 
> only occurs word-finally rather than a phoneme, correct?

If I knew, I would not have asked.

It seems to me odd that /h/ was reintroduced in Germanic loan words,
when, as far as we know, [h] did not occur as phone in Old French; yet
apparently a phone that did occur cannot become a phoneme through borrowing.

In fact English demonstrates quite the reverse.  In Old English [f] and
[v] were both positional variants of /f/.  Yet it did not prevent the
initial [v] in French borrowings such as _vase, voice, vow_ being
adopted and thus leading to /v/ becoming a separate phoneme in Middle
English beside /f/.  So no, I do not think the fact that [θ] was only an
allophone /ð/ would per_se have prevented its becoming a phoneme.

I think David's reply yesterday pointed in a more plausible direction, 
as I observed in my reply to his email.