On Fri, 16 Feb 2001, Tommaso R. Donnarumma wrote:

> This is very plausible, I think, but rather boring, so I
> decided to assign a three-way distinction to the proto-
> language:
>    voiceless       voiceless  voiced
>    glottalised     aspirated  voiced
>    aspirated       aspirated  voiceless
> Problem is, glottalised > aspirated doesn't seem very
> plausible to me.  My question is, how does it look like
> to you?  Is there any known example of this kind of
> change?

Someone already commented on the sound-production aspect.

However, the phonological analysis evident in the creation of the Korean
alphabet had the following "hierarchy" from sounds considered strongest
to sounds considered weakest (for stops, kind of--the way they looked at
phonology was a bit different):

t'ak: tensified (glottalized?)
ch'ach'eong: aspirated
ch'eong: voiceless
pulch'eong, pult'ak: (voiced) nasals

("The Phonological Analysis Reflected in the Korean Writing System," p.
165, Young-Key Kim Renaud; in _The Korean Alphabet_, ed. by same)

So if the proto-lang people "looked" at their language in this way, the
sound-change from glottalized to aspirated as a lenition-of-sorts.

Sorry to refer to that obscure language, Korean; but the only other book
I have on language-examples is _Pacific Languages_ and I don't remember
seeing much aspiration/glottalization in the phonologies discussed.  Hope
this is of some use.