Linguistic commentary from a guy who takes things too literally

Who Was Looking for Whom?

Posted by Neal on November 7, 2006

At about suppertime last night, the phone rang. The caller ID showed it was Doug’s friend James calling. I picked up the phone, preparing to tell James that Doug couldn’t play because he was about to have supper. “Hello?”

“Hi, Neal, it’s Rita.” Oh, not James. James’s mom. “Can you tell James it’s time to come home?”

“No, I can’t,” I told her, “because he’s not here.”

So Rita signed off to go find James and I went back to getting Doug and Adam’s supper ready. My wife, who had heard my half of the conversation from the next room, asked:

Who was looking for whom?

Yeah, that’s right, whom. She knows how to use the word correctly, and isn’t afraid to do so. But what I found more interesting about her question was that it had two wh- words, and yet I was able to answer with just, “Rita was looking for James.”

So what? you ask. Well, usually multiple-wh interrogatives (as they’re known) expect a pair-list answer. In this case, a pair-list answer would be a list of seeker/sought pairs, such as, “Rita was looking for James; John was looking for Marsha; and Sue was looking for his pa.”

I know what you’re thinking. You’re thinking, “But Neal, in the context it was obvious that your wife was expecting just a single-pair answer. Clearly, these questions can be answered with a single pair or a pair list, just depending on the context.” OK, then, what about this often-cited example (originally due to one Krystyna Wachowicz, I believe), for which real-world knowledge would seem to point toward a single-pair answer?

Who killed Robert Kennedy when?

For most English speakers, this question doesn’t go over so well. (It doesn’t for me. Does it for you?) It seems to imply that RFK was killed on multiple occasions. I find myself imagining SF stories with characters hopping among alternate timelines that differ in when and where (and even whether) RFK was killed. If I wanted to ask this question with an expectation of a single-pair answer, I’d have to say something like:

Who killed Robert Kennedy, and when (did they do it)?

So if a single-pair context is not enough to make the RFK sentence felicitous, why would it work for Who was looking for whom??

There is a known exception to the rule that multiple-wh questions expect pair-list answers: disjunctive resolution contexts. That is, if you already know the identity of the participants in the event you’re asking about, and you just need to know which participant filled which role. For example…

Rico went a bit too far,
Tony sailed across the bar,
And then the punches flew and chairs were smashed in two.
There was blood and a single gun shot
But just who shot who?
(Barry Manilow, “Copacabana”)

The only answer needed is either “Rico shot Tony” or “Tony shot Rico”. (Hey, I’m not gonna spoil it!) But when my wife asked who was looking for whom, it wasn’t a case of wondering whether Rita was looking for James or vice versa, so I’m still looking for an explanation.

In fact, this isn’t the first multiple-wh question I’ve heard recently that expected a single-pair answer. There was also this one that I heard on the BBC World Service on October 2, 2006, when the correspondent asked someone who had become addicted to online gambling:

How much did you lose in what length of time?

Again, it’s only the context that tells you a single-pair answer is appropriate, but why doesn’t the question have to be phrased:

How much did you lose, and in what length of time (did you lose it)?

8 Responses to “Who Was Looking for Whom?”

  1. Michael said

    the “Who killed Robert Kennedy when?” actually sounds okay to me. I guess I may be giving it a different force when I imagine the most likely scenario where I would expect to hear that construction. Somebody walks in on a conversation on the morning of June 5th 1968 and hears a mumbled conversation about somebody shooting RFK. The hearer isn’t sure if this is an old story being told as a curiosity (“his brother shot him with a bb-gun…”) or a recent story that made the news.

    So the emphasis would probably rise on the “when.”

    What about a company in which rumors of downsizing are floating around and one person says “i’ve been talking to people and I think I have an idea about what’s going on.” A nearby co-worker asks “well who told you what?” Is it a different question?

    The RFK question is assuming that each wh- has a single correct answer. The second question is assuming that there are various answers for each.

  2. Neal said

    What you’ve described is known as a REF question. I don’t remember exactly how it got its name, but they request that a speaker repeat an earlier utterance because something wasn’t understood. The original utterance can be a statement (like your context), or a question, for example, “Who shot RFK on June 5, 1968?” with the date mumbled. You’re right; REF questions are another case in which multiple-wh questions expect a pair-list answer, but my wife’s question was not a REF question.

    Your “Who told you what?” is a good example of a multiple-wh question expecting a pair-list answer. But the context alone doesn’t cause this expectation. if I said I’d only spoken to one person about the rumors, then I’d find it strange for someone to ask, “Who told you what?” I’d expect something more like, “Who did you talk to, and what did he/she/they say?”

  3. Ran said

    I think that context is part of the explanation, but that different contexts are differently strong; with “Who killed Robert Kennedy when?”, I think we apply the pair-list-answer interpretation before we really think about how nonsensical that is, and then we realize how nonsensical that is and don’t adjust our interpretation. (I can’t give an explanation for why, though; ordinarily we do re-interpret things that don’t make sense, until we find an explanation that does make sense, and ordinarily we don’t notice the re-interpretation unless it’s a particularly difficult one. Seeing as single-pair is kind of a subset of pair-list, it seems like it should be a particularly easy one.) In your wife’s case, the context was very immediate, as she was talking about a phone call that you were still thinking about, so everything got processed at once and you applied the right interpretation.

  4. Laura said

    I agree with Ron. I think that this is something like an echo question and so a single-pair answer is expected. What is more interesting for me is, that in this case, I suppose, you cannot make a coordination of the wh-words as would be expected to happen in case single-pair meaning. Or is it possible? (I’m not a native speaker) Can you say something like: Who and Whom was [he] looking for? or Who and for whom was he looking? That sounds like nonsense for me… (In Czech it’s quite OK)

  5. Neal said

    Michael: I made a mistake. What Michael described is actually called an echo question. (A REF question is what you ask when you don’t know which things someone is referring to. For example, you roommate says, “She did it!” and you don’t know who she refers to or what it is. So you ask, “Who did what?” expecting a single-pair answer.)

    Ran: Maybe. It’s as reasonable as anything I’ve heard so far on the subject.

    Laura: No, my wife’s question definitely wasn’t an echo question, because she wasn’t asking me to repeat or clarify anything I’d said. It was just a straight-up request for information. You’re right: argument-argument coordinations are robustly ungrammatical in English, whether in declaratives or interrogatives. Adjunct-argument coordinations vary in acceptability depending on a number of things.

    As it happens, multiple- and coordinated-wh interrogatives are a research interest for me, too, and I have a list of 24 such questions that someone translated into Czech for me. Would you be interested in looking at them and giving some semantic judgments?

  6. Laura said

    It will be a pleasure 🙂

  7. Laura said

    And actually, I have just read an article by Vera Gribanova, where she says that multiple wh-interrogatives are actually ambiguous between the single-pair and multi-pair reading, which would be also my intuition for Czech. Unfirtunately I cannot give any reliable intuitions about English, but your wife’s reply seems to approve this behavior for English as well.

  8. […] “Hey, neat! Another multiple-wh question with a single-pair answer!” […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: