Bathrobe sent me this Lingua Franca post by Ben Yagoda about one of the more controversial grammatical subjects around, the use of whomever:
My suggested test for choosing between whoever and whomever is to put that hole in the sentence aside for a moment and look at the rest of the clause. If takes an object at the end (“I was involved with … her”), use whomever; if it takes a subject at the beginning (“She … abandoned these chickens”), use whoever. Another method is to see if you can swap in he who (use whoever) or him who (whomever.)
The New York Times recently published this sentence: “Whomever Mr. Trump nominates will inherit that investigation.” True to form, I mocked the Times on Twitter for falling into the whomever trap.
And I got blowback.
The redoubtable Jan Freeman, former language columnist for the Boston Globe, tweeted, “I too would both say & write ‘whoever,’ but ‘whomever’ is grammatically correct here, no?” I know enough not to dismiss anything Jan says, but after some consideration I answered, “No.” I felt bolstered in my opinion after her subsequent tweet, “As in (the oft-misquoted) ‘let him who is without sin cast the first stone.’”
(Let me take a moment to mock Yagoda for misplacing the period at the end of the “Another method is to see…” sentence.) Yagoda winds up saying “I now concede that [the Times] is right and I was wrong,” which is noble of him; me, I’m in favor of doing away with the antiquated and confusing whomever altogether, but as a bomb-throwing descriptivist, I would be, wouldn’t I? At any rate, Bathrobe says:
There are two points where I think he’s wrong.
1. He incorrectly applies his own test at Whomever Mr. Trump nominates will inherit that investigation.
2. He treats “he who” as a fused relative.
I thought I’d throw it open to discussion from the floor; what say you?
Addendum. I forgot when I posted this that we had discussed “whomever” back in 2007.
Yagoda’s second method seems wrong. “He who” versus “him who”? No. Shouldn’t it be “he who” (or “him who”) versus “he whom” (or “him whom”)?
He who abandoned these chickens has a lot to answer for.
Whoever abandoned these chickens has a lot to answer for.
They are looking for him who abandoned these chickens.
They are looking for whoever abandoned these chickens.
He whom I was involved with is gone.
Whomever I was involved with is gone.
Things went south with him whom I was involved with.
Things went south with whomever I was involved with.
i wonder if all of your readers are familiar with Thurber’s priceless advice about “whom”?
“Take the common expression, ‘Whom are you, anyways?’ That is of course, strictly speaking, correct — and yet how formal, how stilted!”
http://downwithtyranny.blogspot.com/2009/12/thurber-tonight-ladies-and-gentlemens.html
Based on what I’ve read, at least, Jan Freeman is right about both sentences being “correct” from a traditional prescriptivist perspective. I can understand why it’s not obvious to Yagoda–figuring out the “whomever” sentence took me a minute, and I used to think “let he who…” was “correct” up until learning otherwise from a member of an online grammar discussion forum–but I think someone who is planning to mock someone else for incorrect grammar or to write a blog post about grammar should take more time to do research.
I had an argument about “let him who” with professional editors I worked for 20 years ago who couldn’t be convinced that “let he who” was not correct.
As Ø says, Ben Yagoda’s methods are contradictory. The first method corresponds to “relative clause wins” and the second to “matrix clause wins”.
FWIW, I’ll mention that Danish doesn’t have a commonly used whoever/whomever fused relative, but it has similar nominative/oblique case pronoun confusion to English. In (at least my) colloquial speech, oblique case pronouns are allowed in subjects unless they are the complete unmodified subject. But as a hypercorrection of this, I often see people using nominative case pronouns in non-subjects when they don’t stand on their own (corresponding to English “to he and I”, “to he who” etc.).
> misplacing the period at the end
I’ve given up understanding the rules for punctuation at the end of quotations and parentheses, so I just put them where they belong structurally.
Parentheses shouldn’t be a problem. If the whole sentence is parenthesized, then its punctuation goes inside too. If it’s only part of a sentence, then the ending punctuation goes outside with the rest of the sentence. It’s only quotation marks that some styles use illogical order for.
So was I correct in ending my second paragraph with “.).”? According to your rule, the period should be outside the ending parenthesis, but the “etc” also needs a dot. Usually, I think you’d leave out one of the periods if “etc” comes at the end of a sentence, but this time there’s an ending parenthesis between the two periods.
Is ‘whomever’ a particularly AmE problem? It just isn’t in my (BrE) vocabulary.
The confusing part here is that English uses one relative pronoun in such cases, even though “logic” would require two, one accusative and one nominative. And indeed, such sentences as Whomever Mr. Trump nominates will inherit that investigation. can’t be translated with a single relative pronoun into, say, German: Wen [ACC] auch immer er nominiert, diese Person [NOM] wird diese Untersuchung erben. (Colloquially you’d replace diese Person or whatever by der or die, but that requires assuming the nominee’s gender. …So you’d go for der, because wer is secretly treated as masculine.)
Assuming masculine:
They are looking for him who abandoned these chickens. [two pronouns]
They are looking for whoever abandoned these chickens. [one pronoun]
Sie suchen den [ACC], der [NOM] diese Hühner im Stich gelassen hat. [two pronouns]
Sie suchen wen [ACC] auch immer, der [NOM] diese Hühner im Stich gelassen hat. [two pronouns; quite unidiomatic because auch immer is so cumbersome]
That is not correct and never was.
And that’s for purely cosmetic reasons: quotation marks tend to create a pretty large space between the last letter and the punctuation, and putting the punctuation right under the quotation mark isn’t an option.
That’s how I do it.
I’m unhappy with the rule of reducing .. to . , too, so I don’t apply it privately. I routinely stumble when a sentence ends with an abbreviation and the next sentence begins with a word that would be capitalized anyway.
The Yagoda post just seems a muddle to me. Like LH, I wouldn’t object if whomever went the way of the dodo, but for those who insist on using it, this page succinctly and clearly explains when whomever is possible and when it is not: https://www.grammarbook.com/grammar/whoever.asp
There’s nothing descriptive about that! I think you need to describe the confusion and treasure it as part of the rich tapestry of human existence. 🙂
Yer, well worrabaht ‘whomsoever’….WHOMSOEVER RULES. OK?
David, the Thurber piece is a joke, a parody.
So was I correct in ending my second paragraph with “.).”?
Perfectly correct, and I award you the Copyeditor’s Medal of Excellence!
That is not correct and never was.
As Ø says, it’s a joke, son. The passage continues:
>There’s nothing descriptive about that!
Exactly. IMHO, one shouldn’t be able to evade the title of prescriptivist just because what one prescribes is the true idiomatic grammar of the majority, or something that some other prescriptivist proscribes. You can’t describe stuff with bombs.
Hat knows that very well; he’s parodying anticipated objections from those who want to enforce “whomever”.
Thanks to David Marjanović for the German grammar explanation. That is indeed more logical and English should steal it. Do we have any idea how far back that two-pronoun construction goes? Was it in the common ancestor but lost from English, or did German innovate it?
Looks like young’uns don’t know about Thurber.
Danish also doesn’t have anything like whoever, and I actually find it hard to grasp the semantic distinction between that and she/he who. Is it that he/she who leaves open the possibility that we actually know who that is?
(We do have the choice between two relativizers som and der, which differ in that der can only be the subject of the relative clause). Also I put the period after the parentheses because I like it that way, irregardless of grammar.
“.).”!? Really?
Danish also doesn’t have anything like whoever, and I actually find it hard to grasp the semantic distinction between that and she/he who. Is it that he/she who leaves open the possibility that we actually know who that is?
There is overlap, but “s/he who” is generally only used to express an axiomatic truth (“He who laughs last, laughs best”) and is formal register whereas “whoever” expresses the idea that “who,” exactly, is inconsequential or beside the point. When in doubt, use whoever.
Three things
1) Are there styles that will let me put the question mark after a non-sentence quote within the period???? Please tell me there are.
2) Friend – “let he who is without sin cast the first stone.”
Me – “Sure. Let he.”
This usually wins the argument.
3) > Yagoda’s second method seems wrong. “He who” versus “him who”? No. Shouldn’t it be “he who” (or “him who”) versus “he whom” (or “him whom”)?
I think you’re misunderstanding Yagoda’s second method. His point is that because “he” is the appropriate replacement for the entire clause starting with whomever, you should use whoever.
He would change his mind and accept “Whomever Mr. Trump nominates, the press will tear into, because if you reverse the sentence and replace the clause, you get “the press will tear into HIM.”
In other words, he wants who(m)ever to take on the grammatical case dictated by the clause’s place in the sentence, not by with the pronoun’s place in the clause.
“He who knows not; and knows not that he knows not is a fool, shun him. He who knows not; and knows that he knows not is a child, teach him. He who knows; and knows not that he knows is asleep, wake him. He who knows; and knows that he knows is wise, follow him.” This is said to be Persian, but I have no idea if it really is.
“Who are you, ANYWAYS”?!!@&$^&O)@
Once we’re having a debate like Victorian grammarians, it’s fair to mention that in Latin a relative pronoun agrees in gender and number with its antecedent, but takes the case required by its function in the relative clause. I haven’t thought of this in the last twenty years, but I’m still going to venture: Quemcumque Praeses Trump nominaverit inquisitionem illam hereditate accipiet.
On these classical grounds, whomever is regular while whoever is an irregular instance of attraction into the case of the implied antecedent. American editors’ enduring willingness to apply Latin grammar to English is puzzling, but it has helped me out before.
Excellent; my old Latin teacher Brother Auger would look upon you with benevolence.
@ryan: How do you understand Yagoda’s first test? I think his first test yields “Mr Trump nominates him”, therefore “whomever”.
Ah. I like that much better. 🙂
Also, ich bin mit meinem Latein am Ende.
In Classical Greek, the relative pronoun is often “attracted” from its “proper” case into the case of its antecedent; or to put it in echt descriptive terms, both constructions are correct. Smyth’s excellent grammar cites Milton for an English parallel: “Vengeance is his, or whose he sole appoints.”
It happens in Latin too, but much less often: still, Gildersleeve and Lodge has Cicero saying: Hoc confirmamus illo augurio quo diximus “we confirm this by the augury which we mentioned.”
More evidence that “whom” is no longer part of the English parole, when even trained linguists botch its traditional usage.
It is somewhat reminiscent of the current Swedish practice of spelling the third person plural pronounced “dom” as either “de” or “dem” depending on the case it is supposed to be in,* but the difference is that “whom” is still pronounced with a final ‘m’ when it is said, almost like a spelling pronunciation.
* I’m told that in Skåne, they still distinguish the subjective and objective forms.
@David Eddyshaw: but a good pedant never lets Latin irregularities creep into his English, or else he should admit that Let he who is without sin is also correct as a case of attractio inversa. Surely we cannot have that, can we?
Having found Plautus: sed istum quem quaeris ego sum, I’m now contemplating the wondrous pedantry of But it is I whom you are looking for. I should however confess, if only to Brother Auger, I often find it instinctive to carry my native it is I into English.
There is nothing actually wrong with it is I, it’s not like saying c’est je or something.
Classical Greek attraction follows a hierarchy, e.g. from the nominative to the genitive, but not the other way around.
@Giacomo Ponzetto: I often find it instinctive to carry my native it is I into English.
Based on your name, wouldn’t the English calque be “I’m I” (sono io)?
He who knows; and knows that he knows, is a know-it-all; avoid him at all costs.
That’s been my experience, anyway.
@John Cowan and @Stephen Carlson, you’re both right, but somehow I’ve learned English enough not to even think of saying “I’m I,” but my English hasn’t become sufficiently idiomatic to switch to “it’s me” instinctively. However, I sometimes consciously choose it against my instincts, because I have the impression that “it is I” comes across as emphatic, pedantic, or pompous. Then again, it’s possibly not the only reason why I may come across that way, regardless of language.
irregardless
O tempora o mores!
Carry right on!
Stephen C. Carlson: It is somewhat reminiscent of the current Swedish practice of spelling the third person plural pronounced “dom” as either “de” or “dem” depending on the case it is supposed to be in,* but the difference is that “whom” is still pronounced with a final ‘m’ when it is said, almost like a spelling pronunciation.
* I’m told that in Skåne, they still distinguish the subjective and objective forms.
Norwegian dialects have merged the two, generally to nominative dei in western areas and accusative dem in the rest. Written Nynorsk has generalized dei, but Bokmål upholds the distinction between de [di:] and dem. In my schooldays we were drilled in it. Naturally, what seems to be winning out is the hypercorrect de. Similar for hun “she”/henne “her” (though of course confused. My daughter knew I would be amused when she could tell me that her friend had just said henne slo hun “her hit she”).
I remember seeing a suggestion to write objective-case who as who’, as in “I know who’ he was looking for”, but it was never taken seriously.
@Lars
>Danish also doesn’t have anything like whoever
Actually, I think I was wrong before. Danish has
“Hvem Trump end nominerer, arver den undersøgelse.”
and although the “-ever” (“end”) is separated, “hvem” is indeed fused here. But “hvem” doesn’t conjugate for case in Danish, so confusion averted!
I think “hvem” used to be the oblique case, with “hvo” (as in the fossilized “Hvo intet vover, intet vinder”, “Nothing ventured, nothing gained”) being the nominative.
> I actually find it hard to grasp the semantic distinction between that and she/he who.
In Danish
“Den som Trump nominerer, arver den underøgelse.”
is possible too, but doesn’t make it clear that the inheritance is assured no matter who is nominated.
danichi. Ben’s first method is “to put that hole in the sentence aside for a moment and look at the rest of the clause. If takes an object at the end (“I was involved with … her”), use whomever; if it takes a subject at the beginning (“She … abandoned these chickens”), use whoever”. So that’s “matrix-clause wins”, isn’t it? Which is wrong, as the 2nd and 3rd of Ø’s 4 examples show.
In Swedish, it would be
Den som Trump nominerar kommer att ta över undersökningen.
If you want to stress the ever in whoever, you would need to split it up in the German way.
Just as mentioned above, the problem with who/whom reminds me too of the Swedish de/dem confusion. In some sentences, where de/dem is the subject in one phrase (?) and the object in the other, both would be considered correct.
Trump nominerar de/dem/dom som tar över undersökningen.
Although some people do have strong feelings on which one sound right and which one sounds wrong.
I find it very interesting to read about these kind of grammatical issues in English that confuses the native speakers. I remember my old high school English grammar, the book spent very little time on explaining who/whom, and recommended use of who unless a few very specific instances (that I instantly forgot), on the other hand, it spent several pages on the preposition “in”, giving countless examples on when to use it and when not to.
Some dialects do keep the distinction, mostly pronouncing de as “di” and dem as “dom”.
@Giacomo Ponzetto: A true pedant wouldn’t strand a preposition. No: it must be “But it is I for whom you are looking”.
Hvem Trump end nominerer, arver den undersøgelse.
True, that is acceptable in a formal register, maybe even preferred, but it’s certainly not what I would produce spontaneously (in speech). More like Lige meget hvem [som] Trump [så] vælger, så arver han/vedkommende sagen.
Which is to say, relative clauses as subject get dislocated so the question of fusedness becomes moot. And in less self-conscious speech, hvem can get an extra relativizer if the relative clause is ‘heavy’ — this case is borderline, adding som here may strike some listeners as ‘childish’. (But you could argue that it’s there anyway, just omitted like relativizers can be with short clauses in other constructions).
(And I actually don’t think the relativizer / indefinite pronoun hvem can be a subject (unlike hvem the interrogative pronoun). Even in the most formal register you would have hvem som intet vover, vinder intet for dainichi’s fossilized proverb above, leaving out som is not Danish any more (if you ask me)).
Not exactly the same type of construction, but: “Richard loves Richard; that is, I am I.” Which seems to have confused the typesetter, since this is an editorial correction of “I and I.”
The only time it is downright ungrammatical to use who instead of whom in Standard English is immediately after a preposition (the comparative particle than counts as a preposition here), and since it is always (I think) possible to strand the preposition instead, whom is never necessary. Examples:
“I know from whom it was stolen.” (traditional, formal)
“*I know from who it was stolen.”
“I know who it was stolen from.” (natural)
I concur.
I have a feeling I posted this before in a similar discussion, but starting around 0:50:
As he went inside, he was merrily greeted/by the girl with who he was in love”
“whom” would sound mighty strange in this context, I think
No it wouldn’t; in fact the Byrds sing it that way in this 1970 version (better than Little Feat’s, to my taste, though from a lousy Byrds record).
Well, I like the Little Feat version better. And who knew the Byrds were such grammar purists!
I don’t think it’s purism so much as natural spoken grammar; as John Cowan said way up there: “The only time it is downright ungrammatical to use who instead of whom in Standard English is immediately after a preposition.” To me, “with who” sounds weird and wrong. Idiolects do, of course, differ, and you and I are living proof of that.
I agree that “with who” in general sounds wrong. I guess I am so accustomed to the way Lowell George sings the song that “with who” is baked into my brain and “with whom” would just sound strange.
@rosie:
Even the first rule is very poorly formulated. The question is what clause he means in “the rest of the clause”. The full text suggests looking at “I was involved with … her” right after the example “When things would start to go south with whomever I was involved with”, which suggests he means the relative clause. But agreed, it’s not very clear.
@JC
>it is always (I think) possible to strand the preposition instead
– It was stolen from [mumble].
– From who?
IANA native speaker, but I suspect there are native speakers whose ideolects don’t have “whom” and who have no problem with this question.
@Lars:
I agree on everything you said, including that it has to be “hvem som” (or “hvem der”) in the nominative. Also, the interrogative pronoun used in indirect questions (which are not archaic in any way) definitely has to be “hvem der” in the nominative. (I think “hvem som” is impossible, or at least marked here.)
Jeg ved ikke hvem der gjorde det = I don’t know who did it
Interestingly, in this similar, but syntactically slightly different construction, I suspect a lot fewer native English speakers would consider using “whom” on the grounds that “it’s the object of ‘know'”.
IANA native speaker, but I suspect there are native speakers whose ideolects don’t have “whom” and who have no problem with this question.
I am a native speaker, and I find I have no problem with it in that context. Thanks for complicating my view of English!
@Hat
> I have no problem with it in that context.
I suspect this is what’s going on: Choosing not to strand the preposition when you can bumps you into a more formal register where “whom” is required after prepositions. But in this case, where stranding wasn’t an option to begin with, no such bumping takes place, and no “whom” is required.
Sure stranding is an option: I have no trouble with “Who from?” here. But yes, uninverted questions are much more acceptable in a meta-linguistic context: I might ask ask “It was stolen from who?” if I didn’t hear or have trouble believing the name, but “Who was it stolen from?” is the normal full-sentence option if the information wasn’t given in the first place.
That’s also how German solves this problem: aber das, was du suchst, bin ich, literally “but that which you seek am I”.
This holds in short form, too: das bin ich is the most common way to say “it’s me”, followed by ich bin es.
@David Marjanović, making the antecedent neuter would solve the problem at the root in Latin too, because then nominative and accusative become indistinguishable. However, I’m not sure if a neuter antecedent would be correct in Latin; perhaps, as in Italian, it would be grammatically correct but stilted.
Some literal-minded German has translated Plautus: doch der, den du suchst, bin ich. Stilted as this may sound (I cannot judge), it avoids Plautus’s irregularity, which I suppose German wouldn’t tolerate: doch den, den du suchst, bin ich.
I think the reason I have trouble with “the person with who I spoke” is because of the clash of registers. In everyday speech and writing (and pretty much any context where I knew my writing wouldn’t have to pass through a pedantic editor), I’d use “the person who I spoke with.” So “with who” violates both the “don’t pied-pipe prepositions” rule in my usual grammar and the “use whom in accusative environments” rule I’ve internalized from school.
@JC:
I see how you could call “Who from?” stranding, but is the distribution of “Who from?” vs. “From who?” really the same as the distribution of stranding vs. pied-piping? That would go again my intuition (which has been wrong before).
Also note that this doesn’t only happen in meta-linguistic contexts. My examples could just as well have been about ellipsis:
– It was stolen.
– From who?
or
– It was stolen, but I don’t know from who.
In those cases, as I was trying to say above, I would use “Who from?” and “but I don’t know who from” respectively.
E. J. Favilla, A World Without “Whom”, now available at a bookstore near you.
since it is always (I think) possible to strand the preposition instead, whom is never necessary.
To whom it may concern: I beg to differ.
It doesn’t. In fact, this is the idiomatic way to do it. The neuter antecedent sounds awkward; I kept it to be as literal as possible.
To whom it may concern
Great example! Of course the to can’t be stranded there, because the matrix sentence is missing: this is short for I write this to whom(ever) it may concern. But more subtly, “*I write this whoever it may concern to” doesn’t work either, and I don’t know why not. I looked around trying to find out when preposition stranding is impossible, finding examples but no clarity. Just why is it that You are using the wrong word, to start with cannot be pied-piped as *You are using the wrong word with which to start? Or rather it can, but that means something else.
But what I did find was cool data on non-standard German and equally non-standard North American French, with such shocking sentences as Ich kann mir da nichts von leisten, or even worse Da kann ich mir nichts von kaufen, with sentential stress on da. Note that while the inseparability of davon is shattered, the V2 rule remains intact. Similarly on the interrogative side we have Wo hat Marie das Kleid her bekommen? I’m not even sure whether to write that with a hyphen after her or not. Similar French examples are Tu connais pas la fille que je te parle de, Robert a été parlé beaucoup de au meeting, and Qui est-ce que tu as fait le gâteau pour?. Weird.
In Dutch, though, preposition stranding is quite normal: Welk bos [forest] liep hij in?
OT: I remember the German translation of Often, often, often, comes the Christ in the stranger’s guise (from the anonymous Celtic Rune of Hospitality) as Oft, oft, oft, kommt Christus im Kleide des Fremdens, which is nicely metrical, but googling finds only the lame im Kleid eines Fremden (it’s the stranger not a stranger, people) or the downright unmetrical im Gewand eines Fremden. I know the original is a tad archaic, but why shouldn’t it be?
Of course the ambiguity of guise is lost (Deckmantel vs. Auftreten), but I suppose that is unavoidable, as they are too long as well as too specific. How about Mantel, or is that always literally ‘coat, cloak’ even in poetry?
Oh yes: with whom‘s doom, can we expect solvet saeclum in Favilla?
It would be “to whoever it may concern”. The “to” in “to whom it may concern” isn’t part of the relative clause: It seems to me to be a case of a fused relative, not of pied-piping.
To whom it may concern is another case of using one relative pronoun instead of two. But at least they’d be identical; in German, one would be dative and the other accusative…
Davon is routinely separated like this north of the White Sausage Equator (or wherever). Southeast of it, the da part is instead duplicated and only the duplicate is moved. This allows topic-verb-comment sentences like Da davon kann ich mir nichts kaufen “I can’t by anything from that [small amount of money]”, aided by the fact that in the dialects da doesn’t have the same vowel as either of the ones in davon. (…And von has a different one again, but I digress.)
Hin & her, and even their composites like hinauf, have colloquially become separable verb prefixes; wohin gehst du has become wo gehst du hin, and ich bin dort[ˌ]hin ge[ˈ]gangen has long been reanalyzed as ich bin dort [ˈ]hingegangen – in other words, there are now such verbs as hingehen, hinaufgehen, herbekommen/herkriegen…
The neuter antecedent sounds awkward; I kept it to be as literal as possible.
But as Giacomo Ponzetti has pointed out, there is no neuter antecedent in the Latin – istum is a male accusative, not a neuter (that would be istud ). So a literal translation into German would be den, den du suchst, bin ich .
While both a male nominative and a neuter antecedent would be idiomatic in German, this kind of attraction wouldn’t.
Oops. That’s a trap I hadn’t fallen into for a long time.
He who knows not; and knows not that he knows not is a fool, shun him.
It occurs to me that this is a condensation into a single line of “DFW Demolished”.