Making distinctions 1
« previous post | next post »
I send daily cards (by snail mail) to a small number of friends. Mostly I just write about what I've been doing, which these days means a kind of log of my postings (Language Log, my blog, ADS-L, mostly). I realized a few weeks ago that I sometimes said
I posted yesterday to X
(where X is the place the posting appeared, not the topic of the posting), sometimes
I posted yesterday on X
and sometimes
I posted yesterday in X
and that my choices seemed essentially whimsical. There were to cards and in cards and on cards, but I was certainly not making some distinction in truth-functional meaning; in my cards I wrote the three variants pretty much interchangeably (though I tended to be consistent on any particular card).
Call this Case 1: post(ing) to/on/in
(There will be further cases in future postings.)
First note: I am not saying that the three prepositions are generally interchangeable. That would be silly; there are many contexts in which they're clearly not, contexts in which they have clearly distinct meanings. This is about one specific context.
Second note: I am not saying that everyone does things the way I do, much less that everyone should do things the way I do. I'm merely reporting on my own usage. I'm sure that other people have different systems.
Third note: other prepositions are possible in the posting context, as in
I posted yesterday for X
but it seems to me that such uses are in fact semantically distinct from to/on/in.
What's at issue here is the No Distinction Without a Difference principle (which has a long history), in particular in its strong form, as enunciated by Dwight Bolinger ("Entailment and the meaning of structures", Glossa 2.2.119-27 (1968)):
A difference in syntactic form always spells a difference in meaning.
I don't go all the way down this road with Bolinger, opting instead for a hedged formulation, along the following lines:
Lexical and syntactic variation is generally unfree; variants usually have (subtly) different meanings or discourse functions, which can be observed in certain contexts (though these differences might not be of consequence in many contexts).
It follows that analysts should take seriously the possibility that variants differ semantically or pragmatically, at least in some circumstances. For and because, if and whether, and many other pairs turn out to be subtly but significantly different.
But there are lots of ways of being different in use. Some linguistic differences index differences in variety or stylistic level; some correspond to differences in ease of production or perception; some have to do with explicitness/clarity vs. brevity; and so on.
The case of the prepositions to/on/in in combination with the verb post or the noun post(ing) is of yet another type. These are three different metaphorizations of the posting scenario:
one-dimensional to, expressing point location of the goal;
two-dimensional on, expressing location on a surface;
three-dimensional in, expressing location within a (bounded) space.
(What I'm saying here isn't even slightly original; there's a huge literature on English prepositions and their uses. And the account above is incomplete; it's merely suggestive.)
That is, in this context the three prepositions aren't distinct in truth-functional meaning, but they are different in the way they present this meaning, from three different points of view. Bolinger would, I'm pretty sure, have considered this to count as a "meaning difference"; he had a generous view of what counted as "meaning".
But my larger point is that it would be silly for someone to tell me that in any particular example, only one of the three choices of preposition is "correct" (One Right Way), or that there's a meaning distinction here that should be preserved.