Fact-checking commas

« previous post | next post »

The opening of John McPhee's article on fact-checking in the current New Yorker (Checkpoints, Feb 9 & 16, 2009) suggests that checking the facts means checking each word for its factuality. Quoting a legendary fact-checker there, he writes:

Each word in the piece that has even a shred of fact clinging to it is scrutinized, and, if passed, given the checker's imprimatur, which consists of a tiny pencil tick.

This is revealed later on to be a metaphor and/or a record-keeping device; I think all involved know that literally checking at the word-level would be mostly pretty vacuous, and would miss a lot of assertions. My favorite non-word-level anecdote in the article:

Penn's daughter Margaret fished in the Delaware, and wrote home to a brother asking him to "buy for me a four joynted, strong fishing Rod and Real with strong good Lines …"

The problem was not with the rod or the real but with William Penn's offspring. Should there be commas around Margaret or no commas around Margaret? The presence of absence of commas would, in effect, say whether Penn had one daughter or more than one. The commas—there or missing there—were not just commas; they were facts, neither more nor less factual than the kegs of Bud or the colors of Santa's suit.

This is more or less perfect for anyone who has ever published a book on commas (not in an Eats-Shoots-and-Leaves way, but rather with the goal of using them as a window into certain intonational and semantic effects). The issue is whether Margaret functions as a restrictive modifier of Penn's daughter. If it does, then the way is paved for Penn to have more than one daughter:

(1) Penn's daughter Margaret is taller than his daughter Betsy.

If Margaret is merely appositional, then we'd do well to wrap it in commas, and then (1) becomes strange (indicated with #), because the first occurrence of Penn's daughter, lacking additional restrictive modifiers, presupposes that Penn has only one daughter:

(2) #Penn's daughter, Margaret, is taller than his daughter, Betsy.

McPhee has it right with the slight "in effect" hedge: failing to include the commas would not assert that Penn has more than one daughter, it would just suggest this pragmatically. We tend to expect that each restrictive modifier we encounter will do some work, usually by removing some objects from consideration and focusing us on others. Thus, "Penn's daughter Margaret" creates the expectation that Penn has at least one non-Margaret daughter.

The phrase "kegs of Bud" in the above brings us back to an earlier anecdote. That one also turns on the distinction between misleading claims and false ones. In a submitted manuscript, McPhee had simply guessed that a certain Budweiser factory produced 13,000 kegs a day, knowing that the fact-checkers would get the right number. The right number turned out to be 18,000. Again, I think McPhee's claim would not have been false, just misleading (and not only because it was bullshit). At the start of a semester of pragmatics, I often tell my students that I will try, later in the course, to convince them that it is true, albeit misleading, to say that I have seven fingers and four toes.


  1. Nicholas Waller said,

    February 6, 2009 @ 10:07 am

    "it is true, albeit misleading, to say that I have seven fingers and four toes." It is to avoid this sort of thing that we (in the UK anyway) have to promise to tell The Truth, The Whole Truth and Nothing But The Truth.

  2. Chris Potts said,

    February 6, 2009 @ 10:41 am

    Nicholas Waller!

    It is to avoid this sort of thing that we (in the UK anyway) have to promise to tell The Truth, The Whole Truth and Nothing But The Truth.

    The whole truth? Absolutely all of it? Or just what the discourse participants take to be the relevant truth (with all the uncertainty that this brings with it)? I think it must be the whole truth in this less absolute sense. Telling the whole truth would be confusing, it would take too long, and it would create misleading pragmatic meanings of its own (as we struggled to make it all relevant somehow to the issues at hand).

  3. Nick Lamb said,

    February 6, 2009 @ 10:42 am

    Nicholas, you can find an earlier Language Log speaking of the matter of whether these could be three separate things. If the information you've linked is correct (which it probably isn't since it is from a newspaper) then it helps to answer some of the questions in that thread.

  4. marie-lucie said,

    February 6, 2009 @ 10:43 am

    It is very common to hear North American men introducing their wives with "This is my wife Cindy" (no pause or intonation change), and they do not mean to suggest that they have another wife by a different name. Similarly "This is my daughter Margaret" does not imply anything about the existence of other daughters, although "This is my daughter, Margaret" would be unambiguous in indicating that she is the only daughter.

  5. Chris Potts said,

    February 6, 2009 @ 10:49 am


    It might be useful to think about your cases in terms of what is relevant and what range of possibilities the discourse participants are likely to consider as live ones. For your "wife" example, we should ask whether the discourse participants even consider it realistic for a single person to have more than one wife. For your "daughter" case, we should ask whether there is any need here to wonder about the full set of offspring. I would expect our answers to these questions to condition where, and how reliably, we deployed these commas (intonational effects).

  6. Benjamin Zimmer said,

    February 6, 2009 @ 11:05 am

    Nicholas/Nick: Mark Liberman's "Types of truth" post is here.

  7. Tim McCormack said,

    February 6, 2009 @ 11:08 am

    @marie-lucie: Analogously, polyamorous folk can say "This is my girlfriend Jane" without meaning that there's only one.

  8. language hat said,

    February 6, 2009 @ 11:25 am

    We are talking about printed matter, not conversation. As a copyeditor myself, I can assure you that the rule for commas is as stated by McPhee, and I enforce it frequently, changing, e.g., "Author's book, Title" to "Author's book Title" if the author has written more than one book (a very frequent situation in which this comes up).

  9. Robert Coren said,

    February 6, 2009 @ 11:25 am

    That business with the commas caught my attention too, and I was hoping it would come up here. My take is that "Penn's daughter, Margaret,…" definitely implies that she was his only daughter, but the absence of commas only leave open the possibility that she isn't.

  10. Robert Coren said,

    February 6, 2009 @ 11:28 am

    As to "the whole truth": In George Bernard Shaw's play Captain Brassbound's Conversion, Lady Cicely Wayneflete replies to the suggestion that she has not told "the whole truth" in her testimony by saying "As if anyone ever knew the whole truth about anything!"

  11. Morten Jonsson said,

    February 6, 2009 @ 11:29 am

    That ambiguity between restrictive and appositional is the point of the joke in the Newhart show: "This is my brother Darryl, and this is my other brother Darryl."

  12. marie-lucie said,

    February 6, 2009 @ 12:09 pm

    Chris Potts, I am aware that in the North American cultural context listeners are not really pragmatically confused about whether "my wife Cindy" is part of a monogamous or polygamous union, but the lack of an intonational marker seems to be expanded to the ambiguous situation where "my daughter Margaret" or "my son Mark" can refer either to one of a set of children or to a single one fitting the description. It seems to me (but I have not carried out a survey) that in speech the type of noun phrase without an intonational marker is more common than the one with such as marker. In published writing, one tends to see a comma for the single type of person ("my wife, Cindy, and I …") but one cannot tell whether this reflects the usage of the author or the editor.

  13. Reinhold Aman said,

    February 6, 2009 @ 2:04 pm

    Take my wife Cindy, please! (Wife's name is "Cindy")
    Take my wife, Cindy, please! (Ambiguous; addressing Cindy?)

  14. Ray Girvan said,

    February 6, 2009 @ 2:30 pm

    Chris Potts: Telling the whole truth would be confusing, it would take too long…

    Just so, as in Airplane II:

    Steve McCroskey: Jacobs, I want to know absolutely everything that's happened up till now.
    Jacobs: Well, let's see. First the earth cooled. And then the dinosaurs came, but they got too big and fat, so they all died and they turned into oil. And then the Arabs came and they bought Mercedes Benzes. And Prince Charles started wearing all of Lady Di's clothes. I couldn't believe it.

  15. Robert E. Harris said,

    February 6, 2009 @ 2:39 pm

    I find volume of beer production stated in barrels, not kegs. The usual keg (as used for fraternity "keggers") is half a barrel. A U. S. beer barrel is 31 U. S. gallons. (An oil barrel is 42 U.S. gallons.)

    I recall an article I read (somewhere) years ago that had a title like "There Are XXX Trees in Russia." The article was about the fact finders and checkers at Time Magazine. The authors put in XXX where they needed an unknown fact or figure. Maybe The New Yorker needs some of these checkers.


  16. Mark Liberman said,

    February 6, 2009 @ 2:51 pm

    In this post, Chris is acting as a theory checker.

  17. Doug Cutting said,

    February 6, 2009 @ 3:33 pm

    Robert E. Harris: I find volume of beer production stated in barrels, not kegs.

    In the original article, it was the sound of the word 'keg' contained in place names near the brewery that led McPhee to wonder about Budweiser's production.

  18. Dave said,

    February 6, 2009 @ 3:52 pm

    " I have seven fingers and four toes"

    It might be "logically" true but it's true in a useless way. What useful information would be conveyed with such a statement? By convention, this statement implies that you have no more than seven fingers and 4 toes (which could be false).

    I'd say it's true (if you have no more than the indicated fingers/toes) and false otherwise (because the intent of saying it is to be misleading).

  19. Nathan Myers said,

    February 6, 2009 @ 5:03 pm

    If fact-checkers are inserting commas when it turns out the daughter I mention is my only one, they're wasting their time and their employers' money unless it matters to the story whether she's the only one.

  20. JanetK said,

    February 7, 2009 @ 6:17 am

    Wouldn't it be a lovely world if we could just stop using the comma as an element of grammar and use it as simply an indication of a small pause.

  21. mollymooly said,

    February 7, 2009 @ 12:29 pm

    literally checking at the word-level would be mostly pretty vacuous

    I might as well be the one to quote Mary McCarthy on Lillian Hellman:

    Every word she writes is a lie, including 'and' and 'the'.

  22. Robert F said,

    February 7, 2009 @ 10:31 pm

    Implicature is actually a rather good trick. Essentially it encodes the more common meanings as shorter sentences, decreasing the average length of an utterance. This is similar to how things like Huffman encoding for compression work.

RSS feed for comments on this post