In grammar we talk about irregular verbs (and sometimes nouns: child/children, e.g.). Steven Pinker points out that children learn the past tense of irregular verbs very early, before they learn the rules for forming past tenses. They know that the past tense of "go" is "went" in just the same way that they know that the past tense of "talk" is "talked." When they get a little older, though, they twig to the rule, and start applying it to irregular verbs: "I goed to the playground." They use irregular verbs correctly before we use them incorrectly, and then we have to relearn the irregularities.
Here's how I would describe the phenomenology of this piece of grammatical knowledge: we learn past tense (or aspect) and fully internalize it. "Went" is as fully the past tense of "go" as "talked" is of "talk." Or it would actually be better to reverse the order: "Talked" bears the same, completely transparent, relation to "talk" as "went" does to "go." First we learn transparency.
Then, scientists in the nursery, we develop the rules of this transparency. Not that we specify them for ourselves. The rules are transparent too. They too are natural, and unspecified (cf. Witggenstein on rule following and Hannah Ginsborg on "primitive normativity"). We go for what "sounds right." But then we misapply the natural and give up on an earlier natural, internalized, disposition to say "went." Two natural and transparent rules come into some conflict, and we go with the one that's more recent and has a wider application: that one too sounds right, at least for a while.
But our elders correct us. We relearn the irregular verbs, and return to a more archaic natural transparency in their use. How much this is a return, and how much a newly internalized rule, can probably not be quite determined, but I suspect that the irregular verbs hang on, like all very basic pieces of language (e.g. the near-universal preservation of the proto-Indo-European word for hand in Indo-European languages), because they occur so early in our own individual experiences of language. I think that we return with a sort of archaic relief to the familiar irregularities of our early childhood.
It doesn't matter very much whether that's true or not, but if it is, it may shed some light on what will follow in my next post - where I will propose adding a new term to the lexicon of rhyme.
This is the first of a series of short posts about rhyme's relation to meter.