Collecting and Displaying Playthings
Posted by Neal on June 25, 2010
Last weekend we went to see Toy Story 3, and I recommend it to everyone. What I like about the Pixar movies is that even though they start with an unrealistic premise (toys that live, monsters that use closet doors as teleportation devices, superheroes), the movies follow, delight in, and exploit their internal logic. In this respect, TS3 was just as clever and funny as the first two, although I understand some weirdos cried while watching it.
Anyway, the linguistically interesting item came not from the movie itself (though I welcome your observations in the comments), but Nick Chordas’s review in The Columbus Dispatch. At one point he compares it to Toy Story 2, which, he wrote, showed how
…collecting and displaying pristine playthings wreak havoc on their fragile psyches.
That plural verb wreak brought me up short. At first I thought, “Ah, it looks like that plural direct object playthings confused Mr. Chordas into making a plural verb.” The real subject, of course, would be the gerund that takes playthings as a direct object. I looked back in the sentence to find it again: displaying. But then I saw that there were two gerunds, joined by an and, both laying claim to this one direct object: collecting and displaying. So maybe Chordas was right to make the verb plural. In that case, why did the plural form sound so strange to me?
For a while I was thinking it was because of the shared direct object. The structure would look like the one in the diagram below. In this diagram, I use the category NP/NP to indicate a noun phrase (NP) that is missing an NP. The reasoning goes like this: collecting playthings is a good NP; therefore, plain old collecting is an NP missing an NP (in this case, playthings). Both gerunds have the category NP/NP, and so the coordination of the two has this category.
But although the two gerunds are coordinated, by the time we’ve arrived at the entire phrase collecting and displaying playthings, it’s not a coordinate structure anymore. It’s a phrase that contains a coordinate structure down inside it. Therefore, I reasoned, the phrase should count as a singular.
However, that logic doesn’t hold up. I mean, look at these sentences:
The old man and woman were sitting at the table.
A picture and a recording of my grandfather are kept in this box.
In the first sentence, we have a singular man and the singular woman conjoined by and; in the second it’s a picture and a recording. (It could also be a picture and a recording of my grandfather, but I’m looking at the parse in which of my grandfather modifies both picture and recording.) But the coordinate structures man and woman and a picture and a recording aren’t the subjects of the sentences. They’re just elements within the subjects the old man and woman and a picture and a recording of my grandfather. So by the kind of reasoning I used for collecting and displaying playthings, these sentences should be
*The old man and woman was sitting at the table.
*A picture and a recording of my grandfather is kept in this box.
So if the factored-out direct object of collecting and displaying isn’t a reason to expect a singular verb, what was leading me to expect one? My next hypothesis was on semantic grounds. Maybe I had been expecting some notional agreement: That is, maybe I had been considering the collecting and displaying of playthings to be components of a single activity, in the same way that you might think of ham and eggs as a single dish and say Ham and eggs isn’t what I ordered.
So does a shared direct object for coordinated gerunds trigger notional agreement? In the next post, we’ll see…