But I know a g bent’ may sound obsurd
I’m sure you’ve had the conversation, at some point in your life, where you’ve discussed the fact that some words have prefixes or suffixes that indicate an antonym, but no antonym exists. For example, we say someone or some act is uncouth, but not couth; our hair can be unkempt, but not kempt; you can be disgruntled, but not gruntled. Those examples all depend on simple prefix removal, but such words need not be constrained in that way. Consider, for example, feckless, which might be matched with a term like feckful (but isn’t). These terms are frequently referred to, unimaginatively, as unpaired words.
A related issue is the cranberry morpheme, one of my favorite terms in linguistics. (…) Cranberry morphemes don’t have a meaning separate from their particular bound morphemes. You know what pre- means even if I don’t type out the rest of a word. Contrast that with the most common cranberry morphemes, such as -ceive in perceive, receive, or conceive. -ceive is not a suffix; it has no meaning outside of the context of the morphemes it attaches to. (…) So with cran-: unlike, say, the black in blackberry (an unbound morpheme), cran- has no intrinsic semantic definition.
artwork { Robert Indiana, Source I, 1959 }
