I always thought that "cw" in Old English represented /kw/, and the same for modern English "qu", and that the change from one to the other was purely orthographic, since the "qu" digraph was more familiar to Anglo-Norman scribes. But on an English Stack Exchange question, I came across the following comment with 6 upvotes:
You should really explain that the second group are almost entirely due to the Normans being unable to pronounce the Celtic/Anglo-Saxon "cw" and so substituting the nearest Latin equivalent, "qu".
"qu" was not pronounced exactly as /kw/ by the Normans, any more than it is so by modern Frenchmen: English took several hundred years to smooth the two together. There is an interesting question about exactly which phonemes the Normans distinguished that the Saxons did not and vice versa
What does this mean? In modern French, I know that "qu" is generally pronounced as a simple /k/. But that doesn't exactly seem to be what this is saying. Does anyone know if there is evidence for some systematic distinction between a "Celtic/Anglo-Saxon 'cw'" and a Latinate "qu" during the relevant time period? If they were distinct, what was the phonetic difference?