What is (e.g.) "danged --> hanged" called?

[Laurence Horn]

At 11:49 AM -0500 9/25/05, Cohen, Gerald Leonard wrote:
>         Original message from Joel S. Berson, Sept. 25, 2005:
>>  If it were the case that the expression "I'll be hanged if I know
>>why" arose from "I'll be danged if I know why" (or vice versa, I
>>suppose), what would that process be called by grammarians?
>>
>* * *
>
>           It's vice versa. "I'll be danged" probably derives from a
>blending of "I'll be damned" and "I'll be hanged," and the
>linguistic term for this feature is "a blend." Also, linguists
>distinguish between "lexical blends" (e.g. "splotch" from "blotch"
>and "spot" and "syntactic blends" (e.g. "time and again" (from "time
>after time" and "again and again").
>
>                 "Danged" looks like a lexical blend.
>
>         Gerald Cohen
>         P.S. For syntactic blends there are two more terms:
>"contamination" and "anacoluthon."  I've done a lot of work on
>syntactic blends but offhand can't recall "contamination" and
>"anacoluthon" being applied to lexical blends too.

I wonder whether we need to appeal to blending in the case of
"danged"; it could be a straightforward euphemism.  The earlier form
appears to be "dang it" (see HDAS), which again looks to be in the
family of "dagnab it", "goshdarn", "durn it", "dadblamed", and so on.
Are we sure that "hang" is involved in the derivation at all?

If "danged" (as a euphemism for "damned") lost its transparency for
speakers and they rationalized it to "hanged", I'd think that would
be a garden-variety folk etymology.

Larry