I had just finished playing some trashcan basketball in the lab today (I won 4-3 with a tremendous 3-game run at the end) when I decided to give myself a quick post-game pep talk before resuming work. What I wanted to say was something along the lines of “All right brain, time to re-apply yourself to the work the work at hand”, but who’s going to be inspired by such bland talk? It’s the down-home idioms and dialectal speech that coaches use that inspire their teams. Everyone knows that.

So I attempted to rephrase the pep talk with an idiom involving knuckling, but I couldn’t figure out whether it was time to knuckle down to some work or to knuckle under to some work. This is not a preposition to take lightly, as it turns out: knuckling down is focusing and applying oneself (akin to buckling down) whereas knuckling under is to give in or submit. And I am not the sort to supplicate myself to my work.

What’s weird about it, though, is that the OED claims that knuckle down can have either of these two dissimilar meanings. The basis for knuckle down as “submit” appears to be based on a single attestation in the Times (I presume London’s) in 1888, where knuckle down was (mistakenly?) used instead of knuckle under.

This stands at the periphery of one of the questions that I think about a lot, since I’m big on the idea that linguistic theory and grammar ought to based on actual usage rather than pie-in-the-sky beliefs about “proper” usage. How can we tell whether this usage of knuckle down was a mistake, an intentional usage that is peculiar to this writer/editor, or an intentional usage that reflects a general acceptance of knuckle down to mean ‘submit’? Regrettably, I don’t think we can tell in this situation, because there’s just not enough data for this idiom at that time.

But suppose for a second that we could tell.  If it’s a mistake, it seems pretty clear to me that it shouldn’t be considered a real meaning for knuckle down; at the very least, it ought not to be codified in the august OED.  If it’s an instance of a generally accepted meaning at that time period, then obviously it ought be considered a real meaning for knuckle down.  But what if it represents a meaning used only by a small group of speakers?  If one person can use knuckle down to mean “submit”, does our grammar need to be equipped to handle that?  Should we say something like “knuckle down can mean submit, only it doesn’t to most people”?  Basically, this is the case that presents itself with slang: some people agree that the slang term has some meaning, but not everybody.  Should we treat these minority meanings as valid but of limited familiarity, or should they be brushed aside in favor of the standard meanings, and only admitted to the fold when they become too popular to ignore?

You probably know my answer by now: generally the former, occasionally the latter.  I have somewhere on the order of thirteen distinct definitions of cheese in my vocabulary (cheese it!, cheesed off, etc.), each of which I consider a quite valid, if not universally recognized, English word.  On the other hand, I will likely never accept bootylicious as a valid word of English.  The way I look at it is that, frankly, a lot of the currently “valid” words of English are understood by fewer people than most slang.  Take, for instance, the word senary.  The OED says it’s a valid word, meaning “pertaining to the number six”.  But I’ll bet fewer people know what senary means than know what cheesed off means.

I’ve wandered off-topic, so let me wrap things up.  The point of this post was simple; I wanted to prevent other people from confusing knuckle down (“focus”) and knuckle under (“submit”) like I almost did.  But as it turns out, maybe the OED’s right and you can’t actually confuse them because they’re already the same.