Literal-Minded

Linguistic commentary from a guy who takes things too literally

Who May I Ask Is Calling? Part II

Posted by Neal on January 28, 2009

So as I was saying, the voicemail system asked me

Who may I ask is calling?

a question that is syntactically ill-formed, and impossible to assemble a coherent meaning for if you’re going just by the ordinary rules of English syntax and semantics. My two hypotheses for the origin of this phrasing were:

  1. It’s a blending of Who may I say is calling? and May I ask who is calling? I’ll refer to this structure as the embedded structure, since we have one clause (who is calling) embedded inside the main one (may I ask/say).
  2. It’s Who, may I ask, is calling? with the parenthetical intonation worn down. I’ll refer to this structure as the parenthetical structure.

If only there were some way of knowing for a sentence like this whether the who at the front is the subject of an embedded verb or the main verb.

In fact, some speakers do make such a distinction. They are the users of hypercorrect whom, speakers who will say things like Whom do you think is the best choice? Standard grammars frown upon sentences like this because the objective form whom is being used for the subject of is the best choice. They point out that you’d never say *I think him is the best choice or *I think her is the best choice; you’d use he or she in these sentences. Accordingly, the question should use the subjective who form. Hypercorrecters, however, follow a rule that calls for whom when the animate wh-pronoun is anything other than the subject of the main verb — whether an object (e.g. Whom did you talk to?) or an embedded subject (Whom may I say is calling?). So if we find attestations of Whom may I ask is calling?, that would point to the embedded structure as a source, with the say accidentally swapped for an ask. Right?

Well, not necessarily. I’ve just described what I’ll call Hypercorrect Whom Rule A. There are other ways of using hypercorrect whom. It’s possible that some speakers use whom when the personal wh-pronoun is anything other than a subject of the main verb that is near the main verb. These speakers would say Whom may I say is calling?, like those who have Hypercorrect Whom Rule A, but for a different reason: because who(m) is separated from is calling by the body of the main clause. And inconveniently, they would also say Whom, may I ask, is calling?, because the who(m) is separated from the main verb by a parenthetical phrase. I’ll call this Hypercorrect Whom Rule B. So attestations of Whom may I ask is calling? do not clearly point to the embedded structure as an origin, after all.

Moreover, let us note that both Rule A and Rule B are only in effect if the speaker is adopting a more formal register of speech or writing. It’s well known that still other speakers have a rule for hypercorrect whom that states heightened formality all by itself is enough to justify the use of whom. These speakers would say Whom may I say is calling?, like followers of Rules A and B; and Whom, may I ask, is calling?, like followers of Rule B; and might even say Whom is calling, please? just to sound formal and correct, even though this whom is the subject of the adjacent verb is calling. I’ll call this Hypercorrect Whom Rule C. (For a more extensive discussion of factors influencing use of whom, see this 2007 Language Log post by Arnold Zwicky.)

But we don’t have to give up yet. Let’s say we have a set of examples of {Who/Whom} may I ask is calling? that we’ve collected. If this construction is an embedded structure, then the speakers who have any combination of Rules A, B, and C will use whom. But if this construction is a parenthetical structure, then only the speakers who have Rule B or C will use whom. So if Who(m) may I ask is calling? is an embedded structure, the percentage of attestations with whom will be greater than it will be if Who(m) may I ask is calling? is a parenthetical structure. What we need to do, then, is find out what this percentage is, and see if it is closer to the percentage we would expect for an embedded structure or for a parenthetical structure.

But what percentages would we expect for an embedded structure or a parenthetical structure? To answer the first question, let’s take a sample sentence that could have whom because of any of Rules A, B, or C. Who(m) may I say is calling? will do nicely. I did a search for both Who may I say is calling? and Whom may I say is calling?” in CoCA, as well as in Google Groups, Google Blogs, and Google Books. The results:

Hypercorrect whom in embedded structures
Test sentence: Who may I say is calling? / Whom may I say is calling?

  • CoCA: 0 / 0 (%age for whom undefined)
  • Google Groups: 35 / 17 (33% for whom)
  • Google Blogs: 44 / 17 (28% for whom)
  • Google Books (2000-2009): 200 / 60 (23% for whom)
  • Google Books (1990-1999): 70 / 21 (24% for whom)
  • Google Books (1980-1989): 56 / 17 (23% for whom)
  • Google Books (1940-1979): 15 / 0 (0% for whom)

Overall percentage for whom: 24%

To get a more accurate picture, we’d need to do the same tests for structurally similar sentences, but this’ll do for now.

To answer the second question, let’s take a sample sentence that could have whom only because of Rules B or C. Who(m) may I ask …? will work, provided we filter out Who(m) may I ask is calling?, and the throw out hits where who(m) functions as an object or embedded subject. I’m ignoring the commas because I’ve seen that in many attestations where may I ask is clearly parenthetical, writers will leave out the commas. Also, because of the sheer number of hits to inspect in Google Books, even taken a decade at a time, I just took the years 2007-2009, which gave me a manageable sample. The results:

Hypercorrect whom in parenthetical structures
Test sentence: Who may I ask [VERB]? / Whom may I ask [VERB]?
  • CoCA: 2 / 0 (0% for whom)
  • Google Groups: 112 / 21 (16% for whom)
  • Google Blogs: 142 / 9 (6% for whom)
  • Google Books (2007-2009): 114 / 57 (33% for whom)

Overall percentage for whom: 19%

The data bear out my prediction that the percentage of hypercorrect whom should be greater in embedded structures than parenthetical ones. Even so, the numbers don’t paint that clear a picture. The difference between 19% and 24% could be attributable to some invalid simplifications I made (for example, in using just one sample sentence). It’s also inconvenient that I got a result of 33% for each sample sentence in one of the searches I did (Google Groups in the first one, Google Books in the second). Oh, well, we’ve come this far, so let’s see what percentage of Who(m) may I ask is calling? uses whom, and make our comparison. The results:

Hypercorrect whom in Who may I say is calling? / Whom may I say is calling?

  • CoCA: 1 / 0 (0% for whom)
  • Google Groups: 20 / 4 (17% for whom)
  • Google Blogs: 45 / 10 (18% for whom)
  • Google Books (2000-2009): 55 / 6 (10% for whom)
  • Google Books (1990-1999): 10 / 2 (17% for whom)
  • Google Books (1980-1989): 3 / 0 (0% for whom)
  • Google Books (1940-1979): 2 / 0 (0% for whom)

Overall percentage for whom: 14%

The overall percentage here (14%) is closer to the 19% for the parenthetical structure than the 24% for the embedded structure, so the evidence so far seems to point (albeit weakly) to Who, may I ask, is calling? rather than a blend of May I ask who is calling and Who may I say is calling? as an origin for Who may I ask is calling.

Even so, the recorded message I heard reminded me more of Who may I say is calling? than Who, may I ask, is calling? On pragmatic grounds, given that a parenthetical may I ask usually has a confrontational edge to it (check out these examples to see what I mean), I wouldn’t expect it to be something that an office manager would want an employee saying to potential customers.

Finally, there’s one possibility that I’ve implicitly ignored all this time: Maybe it isn’t one or the other. It could be that some speakers produced Who may I ask is calling? by blending Who may I say is calling? and May I ask who is calling?, and their phrasing was then misinterpreted as Who, may I ask, is calling? by other speakers. Or maybe some speakers were in the habit of saying, Who, may I ask, is calling?, which was analyzed as a variant of Who may I say is calling? by other speakers. Or maybe both things happened.

add to del.icio.us : Bookmark Post in Technorati : Add to Blinkslist : add to furl : Digg it : add to ma.gnolia : Stumble It! : add to simpy : seed the vine : : : post to facebook : Bookmark on Google

9 Responses to “Who May I Ask Is Calling? Part II”

  1. Gordon P. Hemsley said

    The response to Whom do you think is the best choice? would be I think the best choice is him.. There is no usage conflict here.

  2. Ran said

    @Gordon P. Hemsley: I think the response to “Who do you think is the best choice” would be “I think he is the best choice” (subject “who” → subject “he”). And in traditional grammar, which arguably is relevant if we’re talking about words like “whom”, the response to “Who do you think the best choice is” would be “I think the best choice is he” (predicate nominative “who” → predicate nominative “he”).

  3. The Ridger said

    Only when “he” becomes predicative does it turn into “him” for most speakers (for some it never will), and for many of those speakers it stays “he” in writing.

    I think he is the best choice.
    He, I think, is the best choice.
    He is, I think, the best choice.
    He is the best choice, I think.

    For many speakers the nominative (subject) case form of all the pronouns is cliticized. These people would, I think, follow Rule B above.

    And I am reminded of an exchange between Sheriff Gillespie and Bubba (I think he was Lt. Skinner at the time).

    Gillespie: He had to have help.
    Bubba: Yeah, he got help. And I know who from, too.
    Gillespie: From whom, Bubba?
    Bubba: From Tom Jackson, that’s who from.

  4. viola said

    @ The Ridger: Just confuse me more by bringing in Bubba and Gillespie! :o)
    Whom/who? A couple simple rules in layman’s terms (Laymen terms? Yet another question.) would assist me a great deal on the proper usage of these words.

  5. The Ridger said

    There’s really only one rule for formal use: “whom” where you would use “him” – it’s the object case.

    So NOT “whom, may I ask, is calling” because not “May ask is HIM calling”

    But most colloquial speakers only use “whom” after prepositions these days.

  6. The Ridger said

    Maybe I should add: you need to make sure you have “who/whom” assigned to the right clause. Even if it’s the ‘subject’ of an embedded clause’s verb, if there’s object raising happening, it should still be “whom”.

    “I told him to come = I said he should come” but note the the ‘subject’ of come is raised to be the object of “tell” in the first construction. So “Whom did you tell to come?” but “Who did you say should come?” It’s that latter which gets most people in trouble.

    Again, though: in colloquial English almost always the fronted question word “who” is just left “who” even when it ‘should’ be “whom” – “Who do you trust?” asked Clinton, and even Safire said “Whom do you trust?” would be wrong though correct. Of course, Safire advised us to reword the sentence (“Which person do you trust?” or “Which person did you see?”), which is (a) cowardly and (b) ineffective, given that “who” can be answered by “four firefighters and a cat in a tree” while “which person” cannot.

  7. Ellen K. said

    Seems to me the easiest and simplest guideline is when in doubt, use “who”.

  8. viola said

    @Ellen
    Yes…it seems like the additional “m” could potentially mean trouble, especially for the doubting individuals.

  9. The Ridger said

    Yep, linguists have been predicting the death of “whom” for a long time – a hundred years ago Sapir thought “whom” would shortly join “to-day” as a quaint relic – but it hangs on.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

 
%d bloggers like this: