since both parfit and perfect have an 'r' in them I don't quite get your point there?
Sorry, I wasn't referring to the r in parfit, but where the c /k/ came from. It's not there in parfit, though it was in the earlier Latin perfectus, and that's why the letter was inserted, silently at first, and subsequently modifying the pronunciation. There are two ways words come into a language: either they are inherited from the earlier version of the language, or they are borrowed from a different language. In the former case, sound changes tend mostly to be regular, but in the latter, anything may happen, and often does.
I was in a hurry to miss the brunt of commuter traffic this morning, and was a bit abrupt in my posting.
To try to get a handle on what happened, we'd probably have to do the following: (1) see what any earlier researchers had disclosed; (2) track down the early citations to determine the actual MS spellings (and not the sometimes corrected spelling of diplomatic texts); (3) determine whether it only came in through one source (sometimes these words come in through several, unrelated borrows); and (4) compare the sound change with other, similar, and contemporary borrowings. That would be a beginning anyway.
