>>Unless your understanding of "pidgin" is different
>> from mine,
>> you are asserting that proto-Romance uses Celtic vocabulary in
>> Latin-based
>> structures, or Latin vocabulary in Celtic-based structures, or
>> some mix.
>
> I don't get it. Tok Pisin doesn't live up to your criteria, as far
> as I know.
Tok Pisin is an English-lexifier lingua franca in New Guinea, learned as a
first language by some speakers, and as a second language by others. It
exists in several forms, creolised, extended and simple. I quote Ishtla
Singh "Pidgins and Creoles":
"Creole Tok Pisin should be (and, in fact, is) more linguistically
complex than extended pidgin Tok Pisin"
And of course an extended pidgin is more complex than a pidgin. But the
characteristics are still there even in creolised Tok Pisin:
(a) lexicon from one language, structures not from that language (The
lexifier in this case is English)
(b) transparent vocabulary (How much more transparent do you want than
"pikini bilong mrs queen" for Prince Charles?)
(c) no complex morphology.
(d) no surface grammatical complexity (e.g. no plural markers)
Peter