Screwball reasons and gloriously simple distinctions

« previous post | next post »

In recent years, The New Yorker's coverage of the "descriptivist vs. prescriptivist" divide in English usage has been, shall we say, problematic. In 2012, we had Joan Acocella's "The English Wars," critiqued by Mark Liberman here and here. That was followed up by Ryan Bloom's Page-Turner piece, "Inescapably, You're Judged By Language," which I tackled in "The New Yorker vs. the descriptivist specter."

In Acocella's piece, Steven Pinker is set up as a descriptivist strawman on the basis of a wildly off-the-mark reading of an essay he contributed to the fifth edition of the American Heritage Dictionary. (Pinker serves as chair of the AHD Usage Panel.) He ably defended himself in a subsequent letter to the editor and at more length in a piece for Slate, "False Fronts in the Language Wars." Now another New Yorker critic, Nathan Heller, makes a mess of things in his review of Pinker's book The Sense of Style.

Heller accuses Pinker of "sentences [that] do not add up" and says that his usage suggestions "actually make the language more confused." But I found Heller's piece to be deeply confused, even while it purports to elevate clarity above all else. One source of confusion is a telling typo in a quote from Pinker:

Pinker did not write that "any prescriptive rules originated for screwball reasons." The correct quote is "many prescriptive rules originated for screwball reasons." (In a delicious irony, this sentence first appeared in Pinker's Slate piece on The New Yorker's previous misconstrual of his work.) As misquoted, the reader might think that Pinker, as one of those loosey-goosey descriptivists, is opposed to "any" prescriptive rule, since all that matters is "any way a lot of people use the language." Once again we see the false dichotomy outlined by Geoff Pullum in his post, "'Everything is correct' versus 'nothing is relevant'" — just the sort of dichotomy that Pinker's book seeks to transcend.

[Update: the "(m)any" typo has now been fixed in the online version.]

Heller's counterargument is that what counts as "correct English" (conflating style, usage, and grammar under one banner) is "correct" because it is somehow naturally better: more logical, more clear-thinking, more consistent:

It’s for grammatical consistency, not beauty or gentilesse, for example, that correct English has us say “It was he” instead of “It was him.” Pinker calls this offense “a schoolteacher rule” that is “a product of the usual three confusions: English with Latin, informal style with incorrect grammar, and syntax with semantics.” He’s done crucial research on language acquisition, and he offers an admirable account of syntax in his book, but it is unclear what he’s talking about here. As he knows, the nominative and accusative cases are the reason that we don’t say gibberish like “Her gave it to he and then sat by we here!” No idea is more basic to English syntax and grammar. In the phrase “It was he,” “it” and “he” are the same thing: they’re both the subject, and thus nominative. This is not “Latin.” (Our modern cases had their roots in tribal Germanic.)

But here's the thing. No native speaker would ever say "Her gave it to he and then sat by we here," but I'd wager that the vast majority of speakers (and even Heller himself when he's not monitoring himself for correct usage and grammatical consistency) would prefer "It was him" to "It was he." Accepting "It was him" as completely natural, idiomatic English does not require heading down a slippery slope where the nominative vs. accusative distinction is lost entirely. But if Heller would like to valorize "grammatical consistency" above all else, he's welcome to insist on "It was he," "It is I," etc., which, as Geoff Pullum noted, "is an extremely formal usage, encouraged by really old-fashioned prescriptivists but not seriously used these days by anyone except the unbearably affected." The argument that a pronoun-as-predicate must take the nominative case might be logically appealing, but as James Harbeck wrote on the topic, "language is not math."

Heller continues:

The same is true of “who” and “whom,” another nominative-accusative pair to which Pinker objects, sort of. He writes, “The best advice to writers is to calibrate their use of ‘whom’ to the complexity of the construction and the degree of formality they desire.” Yet who wants to undertake that calibration all the time? The glorious thing about the “who” and “whom” distinction is that it’s simple.

Ah yes, that gloriously simple distinction between "who" and "whom"! I guess all of those native (and non-native) speakers who feel anxious about the proper use of "who(m)" are just too thick-headed to appreciate something so simple. Let's just consider one example of how the selection of "who" vs. "whom" in standard English is far from straightforward. Here's a sentence I found in The New Yorker, written by, you guessed it, Nathan Heller:

But what about Sir Isaac Newton, whom some contend was autistic?
("Little Strangers," 11/19/12)

Simple, right? Just figure out whether the noun replaced by "who(m)" is a subject or object and choose accordingly. In this case, the "who(m)" fills the slot in the clause "Some contend (that) ___ was autistic." It's the subject of the embedded relative clause "___ was autistic," but it's tempting to use "whom" in sentences like this, if you're the type of person looking to use "whom" in the first place. (Whether Heller or his editors should be held responsible for the "whom" here, I cannot say.) Arnold Zwicky has written at length about the temptation to use "whom" in this sort of construction, e.g., "Whom shall I say [ ___ is calling ]?"

Appeals to rationalism have long been a hallmark of the prescriptivist mindset. But the messy details of actual language use will inevitably undermine that rationalism. For those who care to delve deep, it's gloriously complex, not gloriously simple.

[Update, 12/2: On his blog, Arnold Zwicky dives more deeply into the flaws of Heller's piece.]


  1. Sid Smith said,

    November 3, 2014 @ 7:26 pm

    Good. I actually thought about emailing LL with a plea for someone to pull apart this wretched NYer piece. (Where do you mail such things, tho?)

    I'd add that he's also wrong about 'conventions enforced only in American English (introducing restrictive clauses with “that” and nonrestrictive clauses with “which”)'.

    Er, no. The that/which convention is enforced in, for example, all of the four UK national papers where I've worked as a subeditor, including The Times, where I've been subbing for the last 14 years.

  2. Gregory Kusnick said,

    November 3, 2014 @ 7:26 pm

    Perhaps I'm being naïve, but is it really Heller's position that the following two sentences do not have identical structure?

    "It resembled him."
    "It was him."

    Is there some (non-schoolteacherish) rule that prevents was from functioning as a transitive verb in sentences like these?

  3. Sid Smith said,

    November 3, 2014 @ 7:50 pm

    Love this bit:

    '[Pinker] writes, “The best advice to writers is to calibrate their use of ‘whom’ to the complexity of the construction and the degree of formality they desire.” Yet who wants to undertake that calibration all the time?'

    Well, maybe the same people who calibrate their language when talking to their spouse or their boss or their aged aunt.

  4. Anschel Schaffer-Cohen said,

    November 3, 2014 @ 7:51 pm

    The logic thing is especially silly when you consider that (modern) prescriptivist types never actually propose modifying the standard language to fix its logical flaws. In this very instance, for example, the whole fuss about "It is I" seems to gloss over the fact that, if "I" is the subject of the sentence, the verb should be "am". Why are we allowing this blatant inconsistency to persist in a supposedly logical language? Surely we should follow the Spanish model and say "Am I" ("Soy yo")!

    Also, I'm reminded of a joke: A professor asks her students if anyone thinks English is Ergative-Absolutive. A wisecracker in the back row raises his hand and shouts "Me!"

  5. Sid Smith said,

    November 3, 2014 @ 8:21 pm

    'correct English has us say “It was he” instead of “It was him.”'

    Does the NYer writer say "It is I"? If not, why not?

    'The glorious thing about the “who” and “whom” distinction'

    Does the NYer writer say "Whom were you with last night?" If not, why not?

  6. Ralph Hickok said,

    November 3, 2014 @ 8:31 pm

    @Gregory Kusnick:
    The so-called "to be verbs" are neither transitive nor intransitive.

  7. Chris C. said,

    November 3, 2014 @ 9:18 pm

    Here's something I've never understood. Why, in a sentence like "It was he/him," should that last word be nominative rather than accusative? I can understand it being nominative if it stands at the head of a clause, but by itself? It seems to me a kind of insistence on a pigheaded consistency to lay down a rule that it must be nominative.

    But I could be wrong about that, which is why I'm asking.

    I've also wondered recently if this sort of distinction in English serves any function at all other than to distinguish registers.

  8. Ben w said,

    November 3, 2014 @ 9:19 pm

    While conservatives think it exists and is procreation, liberal linguists agree that the copula has no object.

  9. Michael Dockray said,

    November 3, 2014 @ 11:12 pm

    Would Heller say, "It was we"?

  10. Alexander said,

    November 3, 2014 @ 11:22 pm

    " they’re both the subject, and thus nominative. "

    This comment is confused in many substantive ways. But even on the face of it, the singular definite "the subject" is odd. Perhaps the author senses that saying "subjects" instead would get him into trouble, and that the trouble would subvert his position?

  11. Ken said,

    November 3, 2014 @ 11:31 pm

    Update: the "(m)any" typo

    You are kind. When a mistake is that convenient for the writer, I tend to assume it is not a typo.

  12. JR said,

    November 3, 2014 @ 11:31 pm

    @Anschel: Indeed. As the "tribal Germanics," upon which English is based, say: "Ich bin's!"

  13. Jerry Friedman said,

    November 4, 2014 @ 12:31 am

    I see no reason to doubt that Heller would say "It was we" and "It is I" (or "It's I"). Some people do—some only when they remember to, as Ben Zimmer notes.

    Gregory Kusnick: Yes, that's really his position. I don't know what he'd say about "resemble", but he says that in "It was he", "it" and "he" are the same thing, which is not true of "It resembled him."

    (I also don't know what he'd say about a sentence such as "And the remaining angles also coincide with the remaining angles and equal them, the angle ABC equals the angle DEF…", from a translation of Euclid.)

    Maybe two things on the side of "It is he" are worth noting. One is that it seems to be a century or two older than "It is him", to judge by the OED. Its first citation for "he" in this position was published in 1500, perhaps written c. 1450, and its first for "him" in this position is from 1697 (but compare Macbeth, "And damn'd be him that first cries hold, enough"). The first one for "I" (not counting the earlier "it am I", as Anschel Schaffer-Cohen suggested) is from perhaps 1475, and the first one for "me" is from 1592. Apparently the Anglo-Saxons didn't need these constructions.

    The other is that when I looked a bit for other languages where the complement of the copula isn't nominative, I had trouble finding examples. Russian is a prominent one. (I don't think French really counts, since moi etc. can be nominative in other constructions.) But I got the impression that in most languages where you can tell, the complement of a copula is nominative. Of course, this doesn't mean there's any kind of law of nature that English has to work the same way, but it does suggest a tendency.

    So nobody misunderstands—I use object or accusative pronouns, never subject or nominative ones, in this situation.

    (If anybody wants to argue with any of the terminology I've used here, I'm not attached to it.)

  14. Ran Ari-Gur said,

    November 4, 2014 @ 1:32 am

    @Jerry Friedman: I think that English "it's me" and French "c'est moi" are actually quite comparable. It's true that French sometimes allows bare "moi" to be used as a subject (when it's appropriately stressed and so on) while English does not, but this hardly makes "moi" nominative. Rather, the common factor is that English "me" and French "moi" are the unmarked forms (the nominatives "I" and "je" being very restricted in their distribution), whereas in the familiar related languages, it is the nominative form that is unmarked. So perhaps the real relevant tendency is for the complement of a copula to use the unmarked form.

    (The same explanation probably accounts for the rule that Anschel Schaffer-Cohen alludes to above: to "Who did this?" we can reply either "Me" or "I did", but the one-word reply "I" would be rather stilted.)

  15. mike said,

    November 4, 2014 @ 1:33 am

    Regarding who/whom, in Garner's Modern American Usage, Bryan Garner has a couple of interesting sections on "The Objective 'who'" and on "The Nominative 'whom'" (p 861ff in the 3rd edition) that each contain about ten examples from print media in which writers got confused and used "who" or "whom" against case. (As per Ben's example from Heller's previous writing.) For an example of correctly used objective "whom" he cites the horoscope for 7/9/97 in the "Washington Times" and says "If a horoscope writer … can get it right, then you'd think that other journalists would as well." (I think there's an implication here that horoscope writers must be particularly unsophisticated writers compared to, what, normal journalists–?) Regarding nominative "whom," Garner says "Among the toughest contexts in which to get the pronouns right are those involving linking verbs."

    Both sets of examples and Garner's commentary seem like they almost willfully ignore the rampant confusion in vernacular English about the governing grammar for "whom" especially, and by extension, that function simply doesn't require traditional case marking.

    Heller: "The glorious thing about the “who” and “whom” distinction is that it’s simple." Yeah, not so much.

  16. Andrew McKenzie said,

    November 4, 2014 @ 1:34 am

    The standard answer about "It is he" from (generative) syntactic theory is this: Nominative case (and accusative case) are not meaningful to begin with. They are purely structural cases, whose presence is licensed and governed by the syntactic structure. We see this in language after language. For English, it works like this: In the predicate position, the pronoun does not receive nominative case in English. But it must bear *some* case. So it gets a 'default' case.

    In English, that default case is the `objective' case. Hence, "It is him." This is also why, in stand-alone utterances, pronouns get objective case. "Who did it? Him." That's why, in non-finite clauses, subjects get objective case "Him running the show would be a disaster" "For him to run the show would be a disaster", etc.

    There's no particular reason why the default would be objective case in English. In some languages it's the nominative, and in others, it's some other case altogether. In the past it may well have been the nominative; in that case "It was he" would have been just fine.

    Now, not every linguist will agree with this particular implementation of case, due to differences in theory that non-linguists would find excruciatingly arcane. But linguists will agree on this: Linguistic theories attempt to understand languages based on the language's own logic, while prescriptivists attempt to shape languages according to what the prescriptivist feels is logical to them.

  17. Keith said,

    November 4, 2014 @ 4:36 am

    @Jerry and Ran,

    The forms "moi", "lui", "eux" in French , where the pronoun is the subject, are known as the "disjunctive" forms; I don't see why we can't use that term for English pronouns, too.

    In my own everyday usage, I tend to use these disjunctive forms when the pronoun is not directly followed by a verb, and the other (conjunctive?) form when there is a following verb.

    So for instance, I'll say "It was me, I did it.", never *"it was me that did it", although I've heard that construction many times.

    That works for comparisons, too. "I'm older than him" seems fine to me, but I'll just as readily say "I'm older than he is", but I don't think I'd ever say "I'm older than he".

    This is slightly similar to the comparison construction in Russian, where you can use either the conjunction чем + nominative pronoun or the genitive pronoun alone, so both "Он старше, чем Я" and "Он старше меня" are correct.

  18. Barrie England said,

    November 4, 2014 @ 4:38 am

    ‘No native speaker would ever say "Her gave it to he and then sat by we here".’ In some dialects ‘her’, ‘he’ and ‘we’ are used in both subject and object (and preposition complement) position. Although that actual sentence is perhaps unlikely, would it not be more accurate to say ‘No native speaker of Standard English . . .’?

  19. Marek said,

    November 4, 2014 @ 5:54 am

    @ Jerry Friedman:
    >The other is that when I looked a bit for other languages where the complement of the copula isn't nominative, I had trouble finding examples.

    Polish provides another example, as it requires the instrumental case:

    (On) jest złodziejem / !złodziej.
    (3.SG.NOM) is thief.INS. / !thief.NOM
    "He's a thief."

    The nominative form also exists, but it's colloquial and prescriptively villified – the exact opposite of what Heller suggests for English! That's despite both languages sharing "no idea more basic" than the (dubious) nominative subject 'rule'.

    As already noted by others, the fact that contemporary English permits "It is him" might be linked to the fact that it also has the copula agree with the expletive 'it' in the first place, rather than the predicate (as in historical "It am I"). Another basic idea horribly violated, yet somehow overlooked by prescriptivists.

    @Main post:
    >but it's tempting to use "whom" in sentences like this, if you're the type of person looking to use "whom" in the first place.

    Indeed. In fact, disagreements over who/whom are one of the most frequent grammar edit wars on wikis. And I, for one, am much more annoyed by seeing people insert 'whom' where it doesn't belong than not making the distinction.

  20. The Hickory Wind said,

    November 4, 2014 @ 6:08 am

    No native speaker would ever say "Her gave it to he and then sat by we here,"

    Clearly you've never been to Suffolk.

  21. Mark Liberman said,

    November 4, 2014 @ 6:19 am

    The New Yorker's coverage of the whom issue has decayed substantially since 1929, when James Thurber wrote

    The number of people who use "whom" and "who" wrongly is appalling. The problem is a difficult one and it is complicated by the importance of tone, or taste. Take the common expression, "Whom are you, anyways?" That is of course, strictly speaking, correct – and yet how formal, how stilted! The usage to be preferred in ordinary speech and writing is "Who are you, anyways?" "Whom" should be used in the nominative case only when a note of dignity or austerity is desired. For example, if a writer is dealing with a meeting of, say, the British Cabinet, it would be better to have the Premier greet a new arrival, such as an under-secretary, with a "Whom are you, anyways?" rather than a "Who are you, anyways?" – always granted that the Premier is sincerely unaware of the man's identity. To address a person one knows by a "Whom are you?" is a mark either of incredible lapse of memory or inexcusable arrogance. "How are you?" is a much kindlier salutation.

    For the rest, see "Thurber on 'who' and 'whom'", 5/13/2012.

  22. Luke said,

    November 4, 2014 @ 6:34 am

    Michael Dockray

    "Would Heller say, "It was we"?"

    Possibly if he comes from Somerset, home to the well known toy shop "Toys be we", though more likely he'd say "it were we".

  23. Gunnar H said,

    November 4, 2014 @ 6:45 am

    @Andrew McKenzie:

    Thanks for that explanation!

    @ Jerry Friedman:
    The situation in Norwegian is somewhat similar to that in English: Traditionally, pronouns in this position would be in nominative form, but the accusative is now commonly used.

    The official language "Academy" officially approves of both forms, without preference. (One example given is "It must have been they/them.") Perhaps more surprisingly, so does Riksmålsforbundet, which prescribes a more conservative form of Norwegian.

  24. Michael Dockray said,

    November 4, 2014 @ 8:11 am

    Perhaps in Somerset one might hear "It was we.". But I seriously doubt that Heller would say it.

    Who was it at the door?
    It was we.

    No, I don't think so.

  25. Alexander said,

    November 4, 2014 @ 8:27 am

    While we're at it, it should be noted how odd it is to claim (as the NYer author vaguely implies) that in "It is John", that "it" and "John" refer to the same thing, and hence that "it" refers to John, rather than being expletive. This view has been defended, even by serious linguists. But it imposes the burden of explaining why it would generally be odd to say "It is tall and blond", in those same circumstances where one might say "It is John" or "It is a young man" – a fact which, on the face of it, relates to the contrast between "*It is tall and blond (who is) at the door" and "It is a young man (who is) at the door."

  26. mollymooly said,

    November 4, 2014 @ 8:29 am

    Eugene Field would say 'Tis only I, though his young son said It's only me.

  27. Marek said,

    November 4, 2014 @ 8:51 am

    IMHO, this entire claim is fundamentally wrong:

    >No idea is more basic to English syntax and grammar. In the phrase “It was he,” “it” and “he” are the same thing: they’re both the subject, and thus nominative. This is not “Latin.”

    They're neither the same thing (do any linguists seriously claim that?) nor is "he" the subject in this sentence. It's a predicative complement. "It" is the only subject in the sentence.

    Historically, the complement was indeed a target of verbal agreement in this construction (as in "It am I"), thus licensing the nominative case. But since the agreement later shifted to the actual subject (the expletive "it"), there's nothing else to govern the nominative case for the complement. The case assignment is strictly idiosyncratic and construction-specific, and it just so happens English pronouns with no agreement source tend to end up in the oblique case.

    In other words, the case of the complement in "it [be] [NP]" construction does not, in any way, follow from basic English syntax and grammar – rendering the entire point moot.

  28. AB said,

    November 4, 2014 @ 10:08 am

    Never mind the details, isn't it plain weird that a lay reviewer is offering technical corrections to a (very) distinguished expert?
    Would this happen in any other field?

    [(myl) Climate change? Vaccine side-effects? GMO foods?]

  29. Robert Coren said,

    November 4, 2014 @ 10:47 am

    @Keith: 'So for instance, I'll say "It was me, I did it.", never *"it was me that did it"'

    Of course not. You'd say "It was me what did it", wouldn't you?

  30. AB said,

    November 4, 2014 @ 10:59 am

    When Zimmer met Heller:

  31. J. W. Brewer said,

    November 4, 2014 @ 11:28 am

    There is a certain irony in AB accusing the reviewer of lese-majeste. Who are you to criticize the self-proclaimed experts is a standard prescriptivist riff. There is, as far as I can tell, nothing in Pinker's scholarly background that makes the views he sets forth in this particular book beyond criticism, or even statistically likely to be worthwhile. The search for a middle ground in usage discussions between bogus prescriptivism and nonjudgmental scientific descriptivism (A Is Right and B Is Wrong, Because I Say So v. A and B Are Both Grammatical And Everything Beyond That Is Subjective And Unscientific) is a worthwhile one, but coming up with sensible advice as to how to approach stylistic choices between alternatives that are all grammatical without just falling back on Because I Say So is tough work, and there's no a priori reason to believe Pinker will be better at it than anyone else you might pull off the street who doesn't start with the burden of affirmatively stupid ideas about grammar.

  32. David B said,

    November 4, 2014 @ 11:40 am

    @AB: Would this happen in any other field?

    My wife's a transportation engineer focusing on traffic and transit operations, and it happens to her and her colleagues all. the. time.

    We suspect that it's a similar phenomenon: Everybody talks, so therefore they're automatically experts in linguistics. Everyone drives, so therefore they're automatically experts in transportation.

  33. AB said,

    November 4, 2014 @ 12:07 pm

    I wasn't questioning his qualification to review the book, or to disagree on Pinker's judgements about what counts as good style. I meant the specific, detailed, false claims about nominative and accusative, I and he, and so on. Why make this stuff up?

    "The search for a middle ground in usage discussions between bogus prescriptivism and nonjudgmental scientific descriptivism…is a useful on."

    Up to a point. "Usage discussion" covers many different activities. Part of the problem with journalistic pieces like this one is they often imply that prescriptivism and descriptivism are simply different approaches to the same activity.
    I don't expect my dictionary, for example, to strike a "middle ground". The OED would not benefit from being a little bit more prescriptive.
    I am grateful to editors, teachers, poets and professors of cognitive linguistics who offer presciptions for stylish writing. But I would expect their arguments to be literary, political, aesthetic and sociological more often than strictly linguistic.
    Nine times out of ten, when the argument runs "because the logic of English grammar demands it", you are reading a bad argument in favour of bad advice.

  34. J. W. Brewer said,

    November 4, 2014 @ 1:21 pm

    The reviewer isn't making it up. He's repeating the conventional wisdom he's been previously exposed to (and that the editor or "fact-checker" of the piece has likewise been exposed to). I certainly agree that arguments from "logic" in this area are usually a red flag. I suspect Pinker thinks that his sort of cognitive-science approach can show that opinions of the form "construction A is easier to understand than equally-grammatical alternative construction B" can to some extent be evaluated scientifically and are not a pure matter of de gustibus non est blah blah. Well, maybe. Or maybe he will just turn out to have inchoate aesthetic intuitions about style that seem in search of a more objective-sounding justification.

  35. Bloix said,

    November 4, 2014 @ 2:55 pm

    “It looks as if my date is here!”
    This is parody, right?

  36. hector said,

    November 4, 2014 @ 4:15 pm

    As to expertise: Heller probably thinks he is an expert, a professional writer doing front-line work in the trenches of communicating with the public, as opposed to all the pointy-headed intellectuals in academe.

    When I read his article (last night? the night before?) I was mildly appalled at how poorly argued it was. In the magazine utopia of one's dreams one expects better from the New Yorker.

    One can easily forget that publishing is an industry. Product must be created. Deadlines must be met. Written works must be edited within a time frame. Authoritarian usage rules speed up the process since they can't be argued against. Usage rules become part of the culture of the publishing world. Usage rules for the language itself come to be seen as logical by extension: the logic of the publishing process mandates them, therefore they are logical for the language itself.

    This is, of course, a logical leap, and hence illogical. Or, as Ben Zimmer puts it, "deeply confused."

  37. Tom V said,

    November 4, 2014 @ 5:30 pm

    I remember an old Saturday Evening Post cartoon of a woman writing a check in a shop and asking "Who shall I make it toom?".

  38. David Morris said,

    November 4, 2014 @ 9:48 pm

    Yesterday's lesson included two pages on tag questions. One exercise gave a statement with a choice of three tags. One example was: "They mustn't leave until they hand in their papers, ____ must them, must they, mustn't they)?'. Two or three of the students chose 'must them'. Another exercise was to correct the error in the sentence. One was 'Boris and Katia divorced last year, did they? (which for me is grammatical in the context of great surprise). Similarly, some students (possibly the same ones) 'corrected' it to 'didn't them'.
    I am not aware that any native speakers use object pronouns in question tags. I would be interested to know if native speakers in Suffolk do.

  39. David O said,

    November 5, 2014 @ 1:03 am

    I stopped reading this piece earlier today after reading the assertion that the rule against "which" with restrictive clauses aids clarity. But that paragraph about the predicative nominative is just risible. The error he's committing is already pointed out by Pinker: the confusion of syntax with semantics. (Though Nathan does admit he found it "unclear what he's talking about here".) He and it may be the same thing, but "he" and "it" are not. Thanks to the use-mention distinction, we should all be able to understand this. By the same logic, wouldn't we say "Young Werther killed he" rather than "Young Werther killed himself"? And shouldn't the French cut out all those illogical reflexive pronouns? Should Luc tell his dentist "Je je brossé les dents chaque jour"?

    I have no qualifications in this field, and I found the article embarrassing. So it makes me wonder how someone like that thinks they have a better understanding of this area than a Harvard professor with an academic specialization in psycholinguistics.

    I'm a fan of Pinker and it seems like when he says something he's pretty sure he's right. The magazine critiques you read of his books are frequently cat-melodeon. There were some similar intellectually flyweight dismissals of The Better Angels of Our Nature too.

  40. David O said,

    November 5, 2014 @ 1:16 am

    I guess that should be "brosse". The sentence was originally in another tense and I neglected to remove the accent.

  41. Nathan Myers said,

    November 5, 2014 @ 2:08 am

    "But what about Sir Isaac Newton, whom was autistic, some contend?"

    Is hypercorrection a marker of the bounder? The climber? The pathologically insecure? All three? Just asking.

    I have been known to peeve. Peevery is a classic vice: It doesn't matter whether you're right, only whether it brings you as much satisfaction as other, cheaper vices. But if being called a "Grammar Nanny" instead of a "Grammar Nazi" would make it less appealing, what does that say about you? And, is it OK to start a sentence with a conjunction, anymore? Or three in a row?

  42. Nathan Myers said,

    November 5, 2014 @ 3:26 am

    It is perhaps worth noting that the New Yorker fixed their "any"/"many" typo, if that's what it was, in their online edition. We will have to wait and see if they publish a correction and apology in print on, you know, paper. No apology leapt out online, so maybe we had better not hold our collective breath hoping.

    Has the New Yorker fallen so far, in recent decades, or have I risen? George Packer is unabashed over his mad drumbeating for war in Iraq. Joan Acocella is repeatedly indulged in her confused ignorance. Malcolm Gladwell turns out to be a brilliantly well-spoken but utterly unprincipled shill for the crustiest of plutocrats. Who next? Is there reason to suppose integrity in any of the rest? May I please continue to admire

  43. Sid Smith said,

    November 5, 2014 @ 3:27 am

    "I am not aware that any native speakers use object pronouns in question tags. I would be interested to know if native speakers in Suffolk do."

    In Devon you can certainly say "didn't her" and "didn't us".

  44. Nathan Myers said,

    November 5, 2014 @ 3:28 am

    … Atul Gawande, and hope he finds more respectable company?

  45. John Walden said,

    November 5, 2014 @ 3:50 am

    @ Keith. The comparison between "me" and "moi" has been made by more than one astute and wise poster to Language Log. Search it for "disjunctive" and there are a few. One example is here:

    and a particularly fine one can be found here:

    The use of the term "disjunctive" would indeed save a lot of hassle. "Me and Keith agree" would be acceptable, but rude of me, while "Keith and I agree" would be "better" English but not better manners than "Keith and me agree"

  46. John Walden said,

    November 5, 2014 @ 3:56 am


    More on the politeness of "me" first, here:

  47. maidhc said,

    November 5, 2014 @ 4:33 am

    Echoing some previous posts, “Her gave it to he and then sat by we here!” would be perfectly acceptable in some dialects of British English. Us do be speaking suchlike when there be no linguists listenin to we.

    The New Yorker has been strange even going back to the 1920s. It's a sort of amiable eccentricity, but it gets more and more out of date. It makes one wish they could track along being precisely 50 years behind the times, rather than from time to time dredging up usage controversies found in the orations of President Garfield.

  48. Michael Watts said,

    November 5, 2014 @ 7:45 am

    I got the impression that in most languages where you can tell, the complement of a copula is nominative.

    The rule in Latin is different — it's accurately described by Heller, actually.

    In Latin, the phrases linked by a copula appear in the same case, no matter what that case is. For example (I can't vouch for the authenticity of the word-ordering here, but it's syntactically if not stylistically correct):

    Dixit eam esse sacerdotem.

    ( be.pres.inf

    "(subject unspecified) said that she was a priest" — the equivalents of "she" and "priest" both appear in accusative case.

  49. Rodger C said,

    November 5, 2014 @ 7:48 am

    @Bloix: "It looks as if my date is here" sounds like perfectly normal vernacular English to me.

    @Nathan Myers: I prefer "Grammar Ninny."

  50. Jerry Friedman said,

    November 5, 2014 @ 10:26 am

    Gunnar H and Marek: Thanks for the additional examples of copula complements—and prescriptivism about them.

    Michael Watts: Actually, that's a common prescriptive rule in English, too. "They thought him to be me." I was oversimplifying.

    Michael Dockray: The relevant examples of "It is we" at COCA are "It is we who" and things like "It is we Americans who". You could argue that in vernacular American English, the "who" changes the rule. (Unexamined COCA results: "is we who": 31; "is us who": 3.) Nevertheless, I think it's possible that Heller, if he's ever in a situation where he has to choose between "It was we" and "It was us", chooses "we". Another possibility is that he avoids both.

    Andrew McKenzie: Thanks for the explanation. I'm sure generative linguists have a field day describing the variations in non-standard American English: "Me and him were talking", "for she and I", "for her and I", "his running the show", "it is we who", and anything I've left out.

  51. mollymooly said,

    November 5, 2014 @ 11:55 am

    @Michael Watts: Searching Google Books for "me to be him" finds grammarians who argue for a similar rule in English.

  52. Richard Hershberger said,

    November 5, 2014 @ 12:31 pm

    The article discussed here is an example of why I don't subscribe to the New Yorker. Demographically, I am in the center of their target market. Many of my friends read it. I often myself near a copy, and I often thumb through it, and consider subscribing. Then something like this comes along: an utterly asinine piece in an area I know a little about. If it were a one-off, I would shrug it off. But their coverage of language is so consistently asinine that it must be an editorial policy. At which point I worry about articles on subjects I don't know enough about to spot asininities. This is an occupational hazard of reading, but it is simply self-preservation to avoid exposing myself systematically.

  53. David Morris said,

    November 5, 2014 @ 2:30 pm

    @ Sid Smith: Thanks for the information.

  54. Sid Smith said,

    November 5, 2014 @ 5:04 pm

    @ David Morris

    Just to expand the point: you can say "didn't her" and "didn't us" in the Devonshire dialect because you can say "Her didn't" and "Us didn't". For some reason you can't, I think, say "Him/Me didn't".

  55. ThomasH said,

    November 5, 2014 @ 5:49 pm

    I'd like to see a discussion concerning the prescriptiveness/descriptiveness of the following

    Personally it seems both futile — without more actual language transactions between the two countries — and pointless, with bonus points for the complaint about English loan words being part of the "problem.".

  56. Seth said,

    November 5, 2014 @ 7:37 pm

    Ok, a couple of things:

    1) ANYONE who says "Well of course it has to be 'I am he' because…" knows nothing about linguistic analysis. Moreover, they don't even understand how science works: you don't bend the facts based on theory (which is itself based on a superficial understanding of syntax.) You first establish what the facts are and THEN build theories to account for that. And native speakers of (American English, at least) know that "I am he" gets a "??" at the very least. This is so basic I'd wager Pinker's critic has no linguistic training at all (or either forgot everything he ever learned.)

    2) The whole argument can be dismissed with pretty quickly, as follows. Suppose that it's "I am he" and "It is I" and that the reason is copula requiring nominative case blah, blah, blah. Let's apply it to some other cases. You've just found an old picture of yourself with friends. Imagine saying, "Look! There's I. And there's I again by the sea!"

  57. Mark F. said,

    November 5, 2014 @ 7:52 pm

    So what governs pronoun case in the Devonshire dialect? Is it just free? (Well, I guess not, based on Sid's last comment.) Does the choice have some semantic content? Has it been studied?

  58. Fred Trof said,

    November 5, 2014 @ 8:35 pm

    'for who the bell tolls'

  59. David Morris said,

    November 5, 2014 @ 9:49 pm

    My class just watched an episode of a comedy. One character received an unexpected phone call and said to another "It's him!". The second character asked "Who's him … er … Who's he?". I think we could informally say "Who's '*him*'?'.

  60. Jerry Friedman said,

    November 6, 2014 @ 12:59 am

    Seth: Anyone who says "Well of course it has to be 'I am he' because…" is not trying to do linguistic analysis. They're trying to tell people what the right way to talk or speak is. I don't think there's any sound reason to tell people (men or boys, anyway) not to say "I am him", but you're criticizing prescriptivists for doing something improperly that they're not even attempting.

    "There's me!" isn't natural for me, I think. I'd be more likely to say "That's me!" Our prescriptivist might say "There I am!" or any of a number of other possibilities rather than "There's I!"

    (I might say "There's my sister and brother and me." Or even "There are…")

  61. Sid Smith said,

    November 6, 2014 @ 2:55 am

    "So what governs pronoun case in the Devonshire dialect?" Mark F.

    'Tis pirate talk, me dear. Don't ee argue with pirates!

    [Yes, "ee" can be "you".]

    I'm not a linguist, I'm afraid; I just lived in Devon for 6 years. Maybe it's labour-saving: it's easier to say "her" (actually, "er") than "she", whereas "him" ("im") and "he" ("ee") are fine.

    The Grammar section of this wikipedia entry seems pretty good. Apparently there are similarities to some North American dialects:

  62. RP said,

    November 6, 2014 @ 4:45 am

    This dissertation looks highly relevant, esp. p101 onwards ( ). The maps on pp308-9 are worth a look, too.

  63. Robert Coren said,

    November 6, 2014 @ 10:21 am

    I recently saw a reference to a quote from Sarah Palin to the effect that the election results were "a victory for We the People". I've seen similar constructions elsewhere, from speakers who surely wouldn't say "a victory for we". There seems to be an idea floating out there that "We the People" is a noun-like phrase (I'm sure there's a more accurate technical term for what I'm talking about) that doesn't get modified by case.

    I note that the original source of this phrase in American discourse, namely the Preamble to the United States Constitution, has a comma after "we".

  64. Keith said,

    November 6, 2014 @ 10:40 am

    @ Robert Coren

    @Keith: 'So for instance, I'll say "It was me, I did it.", never *"it was me that did it"'

    Of course not. You'd say "It was me what did it", wouldn't you?

    Or maybe "it were me what done it".

  65. J. W. Brewer said,

    November 6, 2014 @ 10:50 am

    Note the inconsistency of pronoun-case-following-copula in the famous text "I am he as you are he as you are me / And we are all together." But it doesn't, in context, make it seem particularly stilted or wrong.

    A rather high percentage of living Americans (from the generational cohort that includes myself and Ms. Palin down to the generational cohort that includes my fifth grader) learned in childhood the text of the Preamble to the Constitution as sung to the melody popularized back in the '70's by Schoolhouse Rock. That setting of the text treats "We the People" as a single unit, with prosody/intonation that is inconsistent with a comma after the first word. But I suspect this was not an innovation at the time and was consistent with a preexisting oral tradition of schoolchildren being taught to declaim the text from memory.

  66. Guy said,

    November 6, 2014 @ 11:46 am

    Obviously, "he hurt himself" should be "he hurt he", because they are both the same thing and both the subject. Likewise "he went to the store and the clerk helped him" should be "he went to the store and the clerk helped he" because they are both the subject (of "go").


  67. Brett said,

    November 6, 2014 @ 11:47 am

    @J. W. Brewer: To anyone who remembers what the physical text of the constitution looks like, it's hard not to think of "We the People" as a unit.

    @Robert Coren: There's no comma that I can see after "We."

  68. Guy said,

    November 6, 2014 @ 11:57 am

    @Robert Coren

    I don't know if actual linguists would agree with me, but I would consider that usage of "we the people" to be an example of metalinguistic usage (or at least I wouldn't bat an eye at the expression being put in quotes in the usages you cite). Such usages usually stop being amenable to ordinary inflectional processes. For example we say "I think it was more big than enormous" when using "big" metalinguistically (with the meaning that it wasn't quite big enough to justify using the word enormous), whereas "I think it was bigger than enormous" has a very different meaning – that it's so big that enormous is not sufficient to adequately convey how large it is – and "more big" is usually not acceptable outside of metalinguistic uses.

  69. bevrowe said,

    November 6, 2014 @ 12:22 pm

    I went last night to see a play called "Who do we think we are?" and with this discussion echoing in my head I turned this into "whom do we …". I suppose that "who" is "correct" because the whole clause "Who are we?" is the object of "think". But "whom" feels a bit possible and does occur (see e.g and

  70. Adrian said,

    November 6, 2014 @ 12:43 pm

    Speaking as a rationalist, I object to Heller's piece being described as an appeal to rationalism. Reason and logic are not the same thing.

  71. Guy said,

    November 6, 2014 @ 1:41 pm


    There are two unsettled issues at play there (I'm going to ignore the third issue of "whom" being mostly absent from informal style and assume we are discussing the case in which we are being sufficiently formal to observe the who/whom alternation).

    First, there is the issue of case in relative and interrogative clauses in which the preposed interrogative phrase is linked to a gap wich would call for nominative (usually, but not in this case, because it has subject function in the clause which it is an immediate constituent of), but which is not the subject of the relative clause itself. For example: "who(m) did they say was responsible?" There is no established rule telling us which case is correct because there are two competing rules of roughly equal usage. "Who(m)" is not subject of the "do" clause but it is linked to the gap which is subject of the "be" clause. The two competing rules insist on different cases here.

    Second – and this is only relevant if we adopt the "whom" favoring rule above – is the issue of whether predicative complements inherit their case from their predicands. This is issue being discussed in this post. If we "de-prepose" the clause you are thinking about, leaving supportive "do" in the construction but putting it in its "natural" declarative position, we get "we do think we are who(m)".

    In this form, we see that who(m) is predicative complement to specifying be with "we" as its predicand. If we adopt the rule that such predicative complements have their case controlled by their predicands, it should be "who", if we adopt the rule that predicative complements are accusative, it should be "whom".

    We can isolate this second issue from the first by looking at "I don't know who(m) you are".

    TL;DR: Both are probably grammatical. "Whom" is appropriate if a) writing in formal style and b) EITHER i) adopting the rule that case is governed by function directly in the relative clause OR ii) adopting the competing rule that case is governed by the function of the gap in the clause that the gap is an immediate constituent of and rejecting the rule that the case of predicative complements is governed by their predicands. Otherwise, the appropriate word is "who".

  72. Seth said,

    November 6, 2014 @ 1:48 pm


    No, I am criticizing them for the method they use to determine what is right and wrong. Excusing them for not doing analysis is similar to saying "Well, you can't blame this tribe of people for what they say about the universe because they aren't doing science." Prescriptvists have no non-question begging principles, and yet they seem to think they are talking about what is and isn't grammatical in the same sense we talk about correct and incorrect solutions in mathematics. Actually, they're free to privilege whichever manner of talking they like as long as they realize the "right and wrong" they're talking about are scientifically meaningless.

    Ok, so we agree that "That's me!" is perfectly grammatical, and "That's I" get's at least a "*". That's already sufficient to destroy the line of argument of the original author. Also, "There's me by the sea" or "That one is me by the sea" are perfectly grammatical; that prescriptivists like to squirm out of the situation by rewording things is cheating. Finally, I love the example of my instructor of syntax: where exactly does the prescriptivist think "me" is receiving case in the "Mad" Magazine example of "What me worry?"

  73. J. W. Brewer said,

    November 6, 2014 @ 1:51 pm

    Well, there might perhaps be better and worse kinds of rationalism. (Consider how it's good to be "rational" but "rationalistic" perhaps tends to be a pejorative.) I had no trouble understanding BZ's pejorative use of "rationalism" in context, but one might I suppose amend it to more clearly specify the bad kind, perhaps approximately what Hayek called "constructivist rationalism" — the kind that cannot understand how complex yet orderly human institutions (a natural language, for example) can evolve and sustain themselves without a bunch of official-type people in charge enacting rules and issuing orders.

  74. Fritz Keppler said,

    November 6, 2014 @ 2:52 pm

    "My class just watched an episode of a comedy. One character received an unexpected phone call and said to another "It's him!". The second character asked "Who's him … er … Who's he?". I think we could informally say "Who's '*him*'?'."

    In Noo Awlins we say "Who dat?"

  75. Marek said,

    November 6, 2014 @ 4:08 pm

    This makes me wonder:

    What's the equivalent of "it's him/he" in dialects which drop the copula altogether (like AAVE)?

  76. Robert Coren said,

    November 7, 2014 @ 11:46 am

    @Brett: I guess I just assumed there was a comma, because that's how it would be written in formal style nowadays. I stand corrected.

  77. Bob Davis said,

    November 7, 2014 @ 1:26 pm

    Radio show from the 1940's(?)
    Woman's voice: Henry! Henry Aldrich!
    Henry's voice: Are you calling I, Mother?

  78. David J. Littleboy said,

    November 8, 2014 @ 12:32 am

    Tom Rush used to sing at least one verse of Bo Diddley's Who do you love as "Whom do you love" saying "This is for all you Harvard students out there"…

    He doesn't do it here, though. Being NYC and not Boston, I guess New Yorker editors didn't hassle him in those days.

  79. Robert Coren said,

    November 10, 2014 @ 10:38 am

    Then there's today's cautionary Rhymes With Orange.

  80. Bloix said,

    November 10, 2014 @ 5:47 pm

    Rodger C –
    Okay, the sentence invokes 1950's or early '60's teenage life – it's a scene right out of Happy Days. The speaker has to be a girl, right? because the boy picks the girl up at her house, not the other way around. So there's a 17-year old girl looking out the upstairs window, and according to Heller, she says, not
    "It looks like my date is here!"
    "It looks as if my date is here!"

    Don't you find the whole scenario amusing?

RSS feed for comments on this post