Putting some science in prescriptivism!

Well, pseudo-science, of course, and it's almost about language change. Most people probably know Gresham's Law, originally an economic principle but pretty widely applied since. Informally, it's often put this way:

Bad money drives good money out of circulation.

I happened to get a link this morning to this piece of wretched journalism. The author isn't directly identified beyond "Mr. Kilpatrick", but it looks like it's our old friend James J. Kilpatrick, already honored among linguists (here). It starts with a big rant (his word) about the stupidity of the phrase "it remains to be seen", particularly offensive in a recent use by the NYT. (Don't ask.) But he's just warming up:

Consider another offense, i.e., the Times' addiction to "last" in a temporal context. A typical example appeared in an editorial five months ago:

"Over the last three decades, the number of overweight children in America has tripled to 16 percent ..."

This is the trouble with the adjective "last" : It is a coin worn down by Gresham's Law. At one time it meant only "final," i.e., the ultimate, the conclusion, the absolute end of something. The ninth inning is the last inning; the rites of burial are the last rites; Hamlet dies in the last act; Browning's last duchess was more than merely his most recent duchess; and when Sophie Tucker died in 1966, she was the last of the red-hot mamas.

The first meaning of "last" is still one of absolute finality, but usage has worn it down. In common usage, the last word no longer means the last word.

Just for the record, as far as I know, the word is an old superlative from late, and it's hard for me to get the point of using last only to refer to a closed set, rather than as the final or ultimate in a list so far. Anyway, I can't really think of using last week, as in the one just past, as an error at any level. (Merriam-Webster's Dictionary of English Usage gives some background, tracing it to 1927, and they endorse the obvious, sane view on this.)

But look at the appeal to Gresham: Last has been 'worn down by time', which sounds like the semantic bleaching so central to much grammatlicalization work. More directly, it has the ring of historical linguistics in the Romantic Age, where language change was deterioration from an earlier, purer state. Loss of morphological case marking and such clearly showed the decline of language over time. Prescriptivists have a wide streak of this running down their backs, of course, but I'm not used to seeing it underpinned with principles from economics. Change is bad enough, but now, the worst changes win, I suppose.