I have been mentally reasoning this out since, seriously, my sophomore year of undergrad.

Sep 03, 2010 19:34

(People with actual significant knowledge of syntax look away now.)

My sophomore year I took a syntax class where I sat next to thelionforreal and we wrote each other notes about popslash. We learned that syntax is like a machine for making chocolate chip cookies and yet I somehow got an A-.

My professor was a good descriptivist like most (probably all?) trained linguists, but all the same he assured me confidently that sentences ending in prepositions were ungrammatical--which meant, in his terms, not that they were incorrect according to some sort of style guide but that they would not be naturally generated by any speaker's internalized rules of language. They literally should not exist, and the fact that they did was evidently some sort of massive and ongoing series of errors in speech or writing.

I offered him an example, influenced by the construction that impeded my path back to my dorm that year. "That area is hard to get to." He appeared baffled and then dismissed the sentence, or possibly my argument, as either incomprehensible or imaginary.

Since then, I have thought a lot about the injustice of his position, and I have come to what is probably an obvious conclusion: there are a lot of sentences that appear to end in prepositions but really they don't. They end in words that look like prepositions that are really part of a verb or some other compound phrase.

Prepositions are words that indicate where something is in relation to something else: "the mouse is ______ the desk." Under, over, behind, below, before, beside, at, and so on. If I just said "The mouse is at" then you would presume I had trailed off in the middle of my thought. That would be a poorly formed sentence--something that I might say or write down for whatever reason, but which I would not mentally compose as a complete utterance. (My syntax professor was big on that point--few things people actually say or write are a good reflection of the syntactic process in their brains, he said, because of all the stammers and hesitations and half-repetitions and self-interruptions and so on.) "The mouse is at" would not arise naturally from my syntax as a complete thought--I would have to have thought "the mouse is at the desk" and then suffered some interruption or error en route to producing the actual words. That is an ungrammtical sentence in a linguistic sense. I would have to put a little asterisk next to it on my linguistics homework.

But if I say "the mouse is hard to get at" that is not what happened at all, because "get at" is a verb. It means something different from "get" + "at", and it is perfectly capable of occurring at the end of the sentence. You see what I'm getting at. For example. My suspicion is that the vast majority of sentences ending in "at" or "to" or "with" do so when those words are attached to verbs are perfectly coherent and thus, in linguistic terms, perfectly grammatical--and who gives a damn about Stunk & White and their attempt to make writing style in English conform to the structure of Latin anyway.

Uh. This pointless, pointless rant brought to you by one too many posts from Reasoning with Vampires, also available as
reasoningwithvamps_feed. It's pretty entertaining, if a little strident on certain rules of usage.

This entry was originally posted at http://dira.dreamwidth.org/548873.html. There are currently
comments there.

nerdery

Previous post Next post
Up