The Newshoggers: Jeri Thompson - asset or albatross? [Part Two] However, thanks to a commenter at my place, I see by common definition "trophy wife" is indeed considered to be an insult.

All I can say is that I didn’t mean it as such. I define the term to simply mean a marriage where the wife is 20 years or more younger than the husband, with no other implications, no matter what circumstances led to the pairing. To the extent that I insulted the women in these marriages, I apologize, but frankly I think it’s silly not to take it as a compliment and rather dishonest not to admit it’s true. No matter what led to the marriage, or how well suited you are despite your age difference, or how happily married, a young wife is still a prize catch for an older man; one that he is rightly proud of and when you walk into a room together, you are a visual symbol of his success. I would take that as a compliment myself, but that’s just me.

This is an example of a very common phenomenon that, to my knowledge, has received no attention from linguists and cognitive scientists. Viz. persistent (long-term) personal misunderstanding of a word or phrase that has a particular codified restrictions set on integration or is  completely entrenched. The ‘trophy’ in ‘trophy wife’ can be interpreted with respect to the husband (she’s his trophy) or the wife (she’s a valuable trophy). The former is the entrenched integration (and a collocate with a specific semantic prosody) but here’s at least one prominent, literate (and female) person who has persistently interpreted the phrase using the latter integration (if we choose to believe the self-report, which I don’t see a problem with doing). Now, everyone I’ve talked to about it, has had the experience of a phrase or word that they have found out years later meant something other than what they’d thought. Very often, this is in funny contexts (something misheard as a child) but there are many more unspectacular instances of the same thing. What is interesting, is that if we assume (as I do following Langacker, Croft, Lakoff and others), that lexical items and phrases are constructions of essentially the same sort as what is traditionally known as syntactical rules (and morphonological rules, as well), then we would expect the same level of misunderstanding of general schematic constructions such as subject-verb agreement or object marking. Yet, these examples are rarely if ever reported. Do they not exist? The answer is that we simply do not know but I suspect that they do occur. They are unlikely to happen to constructions as generalized and commonly used as object marking but they certainly happen on the periphery of the system. For example, very few speakers distinguish between “*He came with Jane and I.” from He’s taller than Jane and I.” as instances of different generalized complement constructions. I would also imagine that most English speakers do not properly interpret the subjunctive in “She recommends that he do it.” This would be more pronounced in highly inflected languages (such as Slavonic languages) where object marking or subject-verb agreement are typically marked beyond doubt but it does happen when it is associated with certain lexical items. Clearly, more (or rather, at least, some) research is need.