Wednesday, February 06, 2008

Not so grating

Minnesota Public Radio puts out a show called Grammar Grater: with Luke Taylor. It is done weekly.

Each show is a short piece about five minutes long that addresses a simple issue of grammar with radio skits the occasional short interview and a quick sermonette. The arguments aren't always prescriptive. I don't always disagree with them. At times the show even suggests that complainers should just get used to the fact that language is always changing and it's not a bad thing.

Good for them.

But they do stumble. Sometimes egregiously. Sometimes more reasonably. My quibble for the day: the treatment of the Minnesota dialect sentence final '[verb] with' form.

Buffy uses this all the time.

'I think I wanna go with.'
'Are you going to come with?'
'No I just rode with.' (as a response to a question like 'Did you go in the store with them?')

Episode 20: Unfinished Business (listen here]) begins with an analogy using the legend of Constanza Mozart waking up her husband by playing an unfinished C major scale.

"Wolfgang couldn't stand it. His ear was begging for some kind of fulfillment, so he'd leap out of bed, rush to the piano and bang out the final note, much to his relief."

From this little tale the topic switches to language forms that do the same. This is where the '[verb] with' form comes in.

Some sentences have the Constanza Mozart...effect, though—if you can resolve it with a direct object, it's probably best to do so:

'Are you coming with?' sounds better when it's 'Are you coming with us?'

'Mr. Posthumous was planning to take over' is more specific when it becomes 'Mr. Posthumous was planning to take over the balloon factory!'

Let's take a quick look at some of the gaps. First of all the phrase "to take over" is not simply "to take over [X]" with a null object. Take over can be either intransitive or transitive but it's not merely because of the null or expressed object. Each sentence uses the phrase differently. Let's do a quick substitution test to see if take over = take over

Using assume control for take over we have

He is planning to assume control.
*He is planning to assume control the factory.

Using lead for take over we have

He is planning to lead.
He is planning to lead the factory.

What's going on here? The word/phrase forms take over and lead are ambiguously transitive/intransitive. But the form assume control does not show that same ambiguity. So it's not fair to say that the underlying forms of leadT and leadI is the same in each sentence simply with a null complement in one and an expressed complement in the other. Is the intransitive use less specific? Maybe. Not really tho. But specificity isn't really a grammatical issue.

Let's look at the other pair of sentences. Is "coming with" simply ellipsis from "coming with us"? Is it basically "Are you coming with ___?"

This is a common analysis of this sentence type. It's used in an almost identical sense. Come with us is almost always an appropriate substitution for come with. But my favorite recent analysis comes from John Spartz here at Purdue who has been investigating the non-complement form as a particle that attaches to verbs of movement.

Just today John was telling me that some analyses suggest a null complement that is always first person. But when I asked Buffy to use a with-final phrase, she came up with the above "rode with" sentence that would have to take a 3rd person complement.

One way to think of the "with" in these sentences is as along. This might not be the perfect substitution. I wonder if John has found any "with with" sentences. John?

With all this in mind: The suggestion that one sentence "sounds better when it's [X]" and another sentence "is more specific when it becomes [Y]" puts too much faith in the premise that the alternate forms are underlyingly identical. But I say they're not. So it's like saying 'Meatloaf tastes better when it's lasagna' or 'Wine is sweeter when it becomes Cola.' Even if the premise is true the judgment might not be.

Now the episode does end with a decent bit of advice: don't let the prescription against ending a sentence with a preposition force you into awkward or "clunky" constructions.

In the name of effective language Taylor actually says "it's okay to have sentences ending in prepositions."

It's a start.


  1. What's the official position on analyses that posit that a construction actually involves ellipsis? Like the nominal "come with [us]" structure. I've heard similar explanations for constructs like "He is taller than I" instead of the much more common "He is taller than me," with the explanation that the supposed real structure is "He is taller than I [am]". Etc.

    It seems suspect to me to offer an explanation that purports to "correct" a common usage (specifically, that offers a usage that people only use if they've been school-trained in its use) by pointing to phantom elements that are supposedly elided.

    Any thoughts?

  2. "Come with" and its kin always struck me as a variant of "come along" rather than as "missing" or "eliding" their objects.

  3. I agree that the come along comparison works pretty well. I think Spartz has done some work to show that with doesn't work with verbs stasis or non-movement verbs. Buffy doesn't judge "stay with" or "sit with" as well-formed.

    regarding ellipsis:
    There's of course no official word. But there's plenty of evidence that there are segments that function in a sentence even if they're not uttered.

    But I wouldn't say that a linguistic analysis that proposes a null form is the same as a prescriptive argument that seeks to change usage by using a null segment as evidence of what should be said.

    I'd expect a linguistic analysis to use the absence or presence of a null form in order to describe why people differ in they're judgments and usage -- never to argue that a speaker should change use.

  4. Robert Burchfield once mused about a construction like Me, I always go to bed at 11:00 as part of a discussion of the surprisingly sparse use of nominative pronouns in English (see also: Me and my friend went to the movies). An argument could be made, maybe-maybe, that nominative is used only when a pronoun a) is non-compound and b) immediately precedes the verb. Move the pronoun or add a buddy, and all bets are off. (Not however explaining between you and I.)

    A prescriptively-oriented friend of mine proposed that me in Me, I ... "obviously" meant [As for] me, I .... This is also a fellow who maintains that "of course" people didn't use Me and my friend ... and that this latter construction was an uneducated anomaly. I'm not so sure, of course. That was the discussion that first got me suspicious about positing elements what just warn't there.

    I kinda like the idea that really the objective form of a pronoun is the base form, and under certain limited circumstances, we deign to decline it nominatively. :-) IOW, nominative isn't about case, it's more about sentence position.

    Ok, now I'll up-shut.

  5. Is it possible that "with..." is not a less-eliptical form than simply omitting it might be?

    That is, when I say, "Are you coming?", is it possible I mean, "Are you coming (with me)?"


Thanks for reaching out.

You can also contact me at wishydig[at]gmail[d0t]com.