This may not be a sensible question, and even if it is it may not have a sensible answer, but is there any discernible rationale to the Welsh mutation system? I don’t mean in the fact of having mutations at all, though that is in itself interesting (and I believe there is no evidence for mutation in the very early continental Celtic languages), but in the particular rules that have come to apply. For example, why should only feminine singular nouns mutate after the definite article? Why should there be mutation after short-form verbs but not after the verb noun? Why should some prepositions cause mutation but not others? Why should there be mutation after ei (his) but not eu? If asked these questions would a native Welsh speaker say ‘Well, that’s just the way the language is’ or would he or she perceive definite reasons for the mutation system to work the way it does?
I’m not complaining – I’m just curious as how these things evolve and become the subject of a common consensus. No doubt queries could be raised about quirks of English grammar which, since I am inside the language as it were, just don’t occur to me, though I might sometimes wonder about our orthography: why so many ways to pronounce the –ough combination, what exactly is the point of the p in words like ptarmigan and so on.