"and himself jail"

« previous post | next post »

In "More Cohen Businesses Coming to Light," on Talking Points Memo, Josh Marshall writes:

The biggest taxi operator in New York, Evgeny “Gene” Friedman, now manages Cohen’s 30+ NYC medallions or at least did the last time we spoke to him. Friedman has been struggling for the last year to keep his taxi businesses out of bankruptcy and himself jail.

The final three words of the boldfaced clause present a weird, and dare I say unusual, case of double ellipsis. The semantic content communicated by those three words (in the context of the sentence) is richer than you'd think could be expressed by only three words, especially given that one of them is merely the conjunction and. That content can be represented as follows, with the struck-through text standing for the content that the reader must infer:

Friedman has been struggling for the last year to keep his taxi businesses out of bankruptcy and to keep himself out of jail.

There's nothing unusual about the first omission; I don't see anything wrong with the clause to keep his taxi businesses out of bankruptcy and himself out of jail. But the omission of out of strikes me as very strange, and what's even stranger is that to my ear, the clause is worse if to keep is put back:

* Friedman has been struggling for the last year to keep his taxi businesses out of bankruptcy and to keep himself jail.

I'm reasonably comfortable marking that sentence with a full-blow asterisk rather than an equivocal "?*". As far as I can tell, the Cambridge Grammar of the English Language doesn't even address the possibility of prepositions being ellipted, which can be read to suggest that such ellipting never occurs, or at least is vanishingly rare. And CGEL says that even when ellipsis is otherwise grammatical (I went by car and Bill went by bus) it is ungrammatical if the ellipted chunk ends with a preposition (*I went by car and Bill went by bus).

So what are we to make of my feeling that the actual sentence, with its two gaps, is less awkward than the version with only one? It's of course possible that my intuition is idiosyncratic, and if that's the case, I'm sure that someone will say so in the comments. However, let's assume that my intuition is shared by others.

The issue here is one of psycholinguistics. But if in fact our poor monkey brains encounter more trouble with the more-explicit variant than the less-explicit one, that seems counterintuitive. Why wouldn't the greater level of explicitness (= less semantic content that must be inferred) result in a smaller cognitive-processing load, and therefore less of a sense of anomaly?

Maybe I'm making a category error, by assuming that there's a direct relationship between degree of explicitness and perceived acceptability. Hell, I'm not a psycholinguist, I just play one on the internet. But since you've asked, I'll tell you what I think.

In the original sentence, the first gap—the acceptable one—kicks the comprehension process into Inferring Mode, or maybe I should call it Reactivating Mode. The chunk keep his taxi businesses out of bankruptcy having been mentally processed only moments earlier, the brain can easily reactivate the representation for keep. And having done that, it's probably not all that difficult to reactivate the representation for out of when it's necessary to do so in order to make sense of himself jail. So in the terminology of psycholinguistics, the Reactivating Mode is primed by the first gap. It's warmed up and ready to go.

But in the variant with only one gap, the Reactivating Mode doesn't get triggered at all until the comprehension process hits himself jail, and all of a sudden, everything is all, "wtf?" until the Reactivating Mode gets turned on.



31 Comments

  1. Chad Nilep said,

    April 19, 2018 @ 12:48 am

    My intuitions agree with yours completely. In fact, my poor monkey brain initially read the quote as "…and himself out of jail", which struck me as a perfectly cromulent ellipses. I had to read it twice to notice how truly weird the quote actually is.

  2. DTI said,

    April 19, 2018 @ 12:50 am

    Heh. When you read Josh’s original post you saw it as an ellipsis, I typo.

    I do leave at least one of those things pretty much any time I go back to revise something I’ve written. So I read it as “fossil” evidence of a hastily rewritten phrase.

    I knew I knew what he meant. I really appreciate your explanation of HOW I knew. Thanks!

  3. rosie said,

    April 19, 2018 @ 1:05 am

    Your paraphrase of CGEL: "CGEL says that even when ellipsis is otherwise grammatical it is ungrammatical if the ellipted chunk ends with a preposition". A chunk consisting solely of a preposition ends with a preposition. So it seems to me that you've provided evidence that CGEL endorses your asterisk.

    What you identify as the first gap is so natural, is it even a gap? Does "to keep driving taxis and earning money" have a gap?

  4. loonquawl said,

    April 19, 2018 @ 1:14 am

    … spare his business bankruptcy and himself jail. -> Then rewritten (spare -> keep out of) ?

  5. CNH said,

    April 19, 2018 @ 2:23 am

    Is it an example of a zugma?

  6. martin schwartz said,

    April 19, 2018 @ 2:27 am

    I prefer a tmesis: She blew my nose and mind.

  7. martin schwartz said,

    April 19, 2018 @ 2:29 am

    I prefer a tmesis: She blew my nose and mind.
    Or, my nose and mind she blew and never returned.

  8. Neal Goldfarb said,

    April 19, 2018 @ 3:05 am

    @DTI:

    Heh. When you read Josh’s original post you saw it as an ellipsis, I typo.

    It took me a minute, but Heh, indeed.

    I do leave at least one of those things pretty much any time I go back to revise something I’ve written. So I read it as “fossil” evidence of a hastily rewritten phrase.

    It's quite possible that you're right about this, and if so the sentence wouldn't be counterevidence to the judgment that the construction is ungrammatical. But I don't think that would change the way that the sentence is understood by readers.

  9. Keith Clarke said,

    April 19, 2018 @ 5:17 am

    (*I went by car and Bill went by bus).

    [I can't do strike-through in the comments box.]

    I'd (British English) be OK with eliding the second "went by" if the ellipsis were marked with a pause, e.g.

    I went by car and Bill, bus.

    I'm almost sure I've read a construction like this, but have no idea who the writer(s) would be. It draws attention to "bus", so would work even better with "but" rather than "and", or a colon rather than comma.

  10. Ursa Major said,

    April 19, 2018 @ 6:08 am

    @Keith Clarke

    "I'd (British English) be OK with eliding the second "went by" if the ellipsis were marked with a pause, e.g.

    I went by car and Bill, bus."

    This and the original sentence sound to me like old-fashioned British, or perhaps I should say they sound like something my grandmother would have said, and my dad would sometimes say. Not being British myself I think I'd always say "and Bill by bus", but I recognise the briefer version.

  11. richardelguru said,

    April 19, 2018 @ 6:46 am

    martin schwartz
    'Tmesis'? Surely bath-bloody-os…

  12. John Shutt said,

    April 19, 2018 @ 6:59 am

    I would typeset it with a comma between "himself" and "jail" (and, just to keep that comma from feeling too heavy, I'd probably also put a comma before "and himself"). I also, tbh, don' t see it as double ellipsis because I don't find it natural to think of "to keep his taxi businesses out of bankruptcy and himself out of jail" as having /any/ ellipsis in it. I think what's happening isn't ellipsis at all, it's just that two different endings are being provided (even that sounds more syntactic than what I think is going on there), which is why the first commenter didn't by instinct even see the "first ellipsis". The idea that there's an ellipsis before "himself", I see as a fiction of transformational grammar.

  13. Neal Goldfarb said,

    April 19, 2018 @ 10:08 am

    @John Shutt:

    I also, tbh, don' t see it as double ellipsis because I don't find it natural to think of "to keep his taxi businesses out of bankruptcy and himself out of jail" as having /any/ ellipsis in it. I think what's happening isn't ellipsis at all, it's just that two different endings are being provided (even that sounds more syntactic than what I think is going on there),

    Although you think your description "sounds more syntactic than what I think is going on," the reality is that your description isn't syntactic enough. In fact, it's not really a syntactic description at all, except in the sense that in a word-order language, anything having to do with the linear arrangement of words has syntactic implications. As far as I'm aware, thee is no syntactic category called "an ending."

    The idea that there's an ellipsis before "himself", I see as a fiction of transformational grammar.

    The concept of "ellipsis" as a linguistic phenomenon is an analytical construct that is used in order to describe regularities in linguistic expression. So are linguistic phenomena such as "nouns," "verbs," "tense," "transitivity"—in fact, pretty much every linguistic category that has a name. So I suppose you could call the phenomenon of ellipsis a fiction, but that would mean throwing out a lot of babies with a very small amount of bathwater.

    As for whether it's a fiction of transformational grammar, one wonders why, if that's the case, the concept is used in nontransformational approaches to grammar, such as both CGELs (Cambridge and Comprehensive [note the ellipses]) and HPSG, not to mention approaches that predate generative grammar.

  14. Kyle said,

    April 19, 2018 @ 10:10 am

    I feel like it's a question of precedence.

    without elision:
    (
    to keep ((his taxi businesses) out of (bankruptcy))
    ) and (
    to keep ((himself) out of (jail))
    )

    with elision of "to keep":
    to keep (
    ((his taxi businesses) out of (bankruptcy))
    and
    ((himself) out of (jail))
    )

    with elision of "out of":
    (
    to keep ((his taxi businesses) out of (bankruptcy))
    ) and (
    to keep ((himself) (jail))
    )

    Treating elision as an implied backreference to a previously-used word, the "to keep"-elision involves jumping out of a "to keep()" clause to find an appropriate preposition to use, while the "out of"-elision stays within the same verb clause.

    (Apologies for probably messing up the linguistics jargon; I'm a programmer by trade, as evidenced by those tangles of parentheses)

  15. Nick Fleisher said,

    April 19, 2018 @ 11:41 am

    We appear to have gapping ellipsis in this example, and gapping always involves ellipsis of the verb. So, against the idea that there is a psycholinguistic puzzle of increased explicitness being tied to reduced acceptability, one could just say that these examples reflect a known syntactic property of gapping. Whatever ultimately proves to be the right explanation for gapping will take care of these cases as well.

    For a great and extensive overview of gapping, see Kyle Johnson's recent manuscript here: http://people.umass.edu/kbj/homepage/Content/gapping.pdf

  16. Nick Fleisher said,

    April 19, 2018 @ 12:29 pm

    In other words, it’s plausible/likely that this is about our (specifically) human brains, not our poor monkey brains.

  17. Daniel Barkalow said,

    April 19, 2018 @ 12:46 pm

    It's okay to omit a verb with a particle: "I went by the store and Bill the candy shop" (in the reading where you're explaining which temptations you resisted rather than your routes). So maybe "and himself jail" being better is a matter of accepting "keep out of" as a verb with particles in the speaker's dialect. I'd accept "I passed the class out their tests", so two objects with a particle between them isn't impossible. While I doubt the author actually has "keep out of" as a unit like "pass out", that makes this a weird open-class lexical item rather than structurally impossible, just like "I passed the class out their tests and the TAs (*out) pens."

  18. martin schwartz said,

    April 19, 2018 @ 3:23 pm

    @richardelguru:
    Egads, my remark may be bathos, but what I meant was not tmesis but zeugma. More pathetic than bathetic, so to speak. Thanks for sending my memory to the bath.
    MS

  19. Marty Gentillon said,

    April 19, 2018 @ 4:08 pm

    I am going to agree with the no double ellipses theory. Grammatically, I parse the latter half of the sentence as being just a compound object and not that much different than "[he [gave John [[a book] and [a car]]]]". We don't need ellipses to describe the verb having a more complicated indirect object. Objects can be combined with and. Nothing more complicated is happening there.

    Now, he ellipses is a little odd, but doesn't strike me (as an Idahoan) as being ungrammatical so much as uncommon. Something that one might choose to use when the place, jail, has such startling implications as an actual jail has. Though, I do think that it is improperly typeset — missing a comma or em-dash.

  20. Marty Gentillon said,

    April 19, 2018 @ 4:16 pm

    *sigh* s/he ellipses/the ellipses/g

    Apparently I can't type today.

  21. Julian Hook said,

    April 19, 2018 @ 8:06 pm

    If you search for examples of zeugma online, one that pops up on several websites is "As Virgil guided Dante through Inferno, the Sibyl Aeneas Avernus." (Attributed to Roger D. Scott.) That appears to involve a double ellipsis similar to the one under discussion here—that is, "the Sibyl [guided] Aeneas [through] Avernus."

  22. Neal Goldfarb said,

    April 20, 2018 @ 12:01 am

    @Nick Fleisher:

    We appear to have gapping ellipsis in this example, and gapping always involves ellipsis of the verb. So, against the idea that there is a psycholinguistic puzzle of increased explicitness being tied to reduced acceptability, one could just say that these examples reflect a known syntactic property of gapping. Whatever ultimately proves to be the right explanation for gapping will take care of these cases as well.

    I don't understand why you think that what I describe as a psycholinguistic puzzle is really just the result of "a known syntactic property of gapping." Or to put it another way, I don't understand what the syntactic properties of gapping have to do with the differences in acceptability that the two variants display.

    What you seem to be saying is that one of the syntactic properties of ellipting keep is that it results in an increase in acceptability if out of is also ellipted, despite the latter ellipsis being ungrammatical. But I don't see why that result should be attributed to something about the syntax of gapping rather than to processing factors—i.e., psycholinguistics.

  23. Neal Goldfarb said,

    April 20, 2018 @ 12:15 am

    @Marty Gentillon:

    I am going to agree with the no double ellipses theory. Grammatically, I parse the latter half of the sentence as being just a compound object and not that much different than "[he [gave John [[a book] and [a car]]]]". We don't need ellipses to describe the verb having a more complicated indirect object. Objects can be combined with and. Nothing more complicated is happening there.

    Indirect object? Are you suggesting that his businesses and are indirect objects? Meaning that out of bankruptcy and out of jail are direct objects?

    How can a prepositional phrase be a direct object?

  24. Nick Fleisher said,

    April 20, 2018 @ 12:33 am

    What I meant is that "out of" can only be ellipted through gapping, and if you have gapping then you have to ellipt the verb, too. Those are known syntactic facts about gapping, and I think they account for the pattern you've observed, in which "and to keep himself jail" (ellipsis of "out of" without ellipsis of the verb) is far worse than "and himself jail" (ellipsis of both; and fwiw I also find this example quite degraded, plausibly the result of an editing error, but in any case still interestingly better than the completely impossible "and to keep himself jail").

    On predicting an increase in acceptability: I think your sentence is ambiguous, so I will clarify what I meant. If "out of" is ellipted, then ellipting "keep" will improve acceptability, for the reasons just stated. On the other hand, if "keep" is ellipted, there is no prediction that ellipting "out of" will improve acceptability. (And in fact it may be possible to analyze such examples as involving coordination below "keep" without any ellipsis at all.)

  25. Chas Belov said,

    April 20, 2018 @ 1:18 am

    I'm okay with the construction and got the intended meaning on first read. No *

  26. Marty Gentillon said,

    April 20, 2018 @ 3:28 pm

    @Neal Goldfarb, You got me, my I used the wrong words there. I should have said direct objects. (they are what he is keeping: "his business out of bankruptcy" and "himself out of jail") I suppose that I feel the prepositional phrases modify the nouns more strongly than the verb and are therefore part of the noun phrase, and that noun phrases can normally be connected with 'and' without anything being elided.

  27. Jerry Friedman said,

    April 20, 2018 @ 5:26 pm

    Big-ass terisk for me.

  28. Viseguy said,

    April 21, 2018 @ 9:48 pm

    I also see a compound direct object, not an elision of "keep". It's the elision of "out of" that earns the asterisk in my book.

  29. Rodger C said,

    April 23, 2018 @ 7:48 am

    you have to ellipt the verb

    Ellipe?

  30. ajay said,

    April 23, 2018 @ 10:57 pm

    "She blew my nose and mind."

    She lowered her standards by raising her glass/ Her courage, her eyes – and his hopes…
    When he asked "What in heaven?" she made no reply, /Up her mind and a dash for the door!

  31. martin schwartz said,

    April 24, 2018 @ 1:46 am

    @ajay: Ho! An belated response to my zeugmatic "She blew my nose and my mind", a modification of the Rolling Stones line. I thought I was being silly and irrelevant. I see you've taken up an example from
    Flanders and Swann. (A trivium: Laura Flanders, radio/TV journalist,
    is Michael Flanders' daughter). Here is a zeugma about the city of Zeugma:
    "At the onslaught of Shapur I, the escaping Christians crossed the bridge and themselves. "
    As an Iranist, I can assert that it's true, but I won't. Likely, though.
    Martin Schwartz

RSS feed for comments on this post