I wouldn't be surprised if few have yet to realize this

« previous post | next post »

Lauren Collins, "Haiku Herman", New Yorker 3/31/2014:

When asked later about the role that poetry had played in Kiev's Independence Square — protesters waved portraits of the nineteenth-century poet Taras Shevchenko —  Van Rompuy said, "I wouldn't be surprised if this struggle and this tragedy had not inspired people there."

Sometimes, as in that example, the construction "I wouldn't be surprised if X had not Y'ed" means something like "I believe that X probably Y'ed", with the extra not presumably due to some combination of negative concord and the difficulty of keeping track of multiple negations. But about equally often, it means roughly the opposite: "I believe that X probably didn't Y". For instance, "Cup final booklet sold for £3,000", BBC News 9/26/2008:

Auctioneer Andrew Bullock said: "The amazing thing about this programme is its condition.

"It was tucked inside a book published in 1906 and I wouldn't be surprised if it had not seen the light of day since the game was played almost 100 years ago.

The versions with negative concord are quite idiomatic, and are used by some excellent writers — thus Graham Greene, The Third Man:

"She claims to be Austrian, but I suspect she's Hungarian. She works at the Josefstadt. wouldn't be surprised if Lime had not helped her with her papers. She calls herself Schmidt. Anna Schmidt. You can't imagine a young English actress calling herself Smith, can you? And a pretty one, too. It always struck me as a bit too anonymous to be true."

This sort of thing generally arises in structures involving a negated propositional attitude verb, with a complement sentence that may or may not be negated. Thus Joe Nocera, "Amazon’s ‘Bullying’ Tactics", NYT 5/30/2014:

As for Amazon's "pro-consumer culture", I am wary of that expression as applied to any profit-seeking entity. It's a business, not a charity. If Amazon could jack up prices by 50%, don't think they wouldn't hesitate.

Again, versions both with and without the extra negation are can also be found:

Don’t think they wouldn’t hesitate to re-instate the draft to keep our Arab friends happy.

Don’t think Google wouldn’t hesitate to shut down rather than pay up.

I have copies of all of our PMs and your bribe offers so don't think I would hesitate to post them all at the drop of a hat.

I know what to do in the event of an emergency and don't think I would hesitate calling 911 should the need arise.

But cases like this one seem to require a semantically or pragmatically negative verb like "hesitate", which allows the impulse towards negative concord to be reinforced by the principle that multiplex negatio confundit.

Negative concord isn't involved at all in this next example — Andy Greenberg, "Darkcoin, the Shadowy Cousin of Bitcoin, Is Booming", Wired 5/21/2014:

In only a month, the little-known bitcoin alternative known as Darkcoin has rocketed nearly tenfold in value–from around 75 cents a coin to almost seven dollars. Its selling point: Darkcoin offers far greater anonymity than bitcoin, mixing up users’ transactions so that it’s incredibly difficult to trace a payment to a person. And though few have yet to accept that more-anonymous coin for actual goods and services, the promise of Darkcoin’s privacy features seems to have sparked a miniature boom. It’s one of the fastest growing among the wave of cryptocurrencies that’s followed bitcoin’s success, with the total value of its combined coins topping out at nearly $30 million.

This one is also Out There:

Few have yet to realize that in order to truly make our species interplanetary, we have to have babies on other planets.

A few new players are entering or trying to; some came in at the peak and have been clobbered; few have yet to establish a solid track record.

For Solazyme, the ability to harness a very productive strain of algae has allowed the company to excel where few have yet to explore.

Few have yet to understand Vasconcelos in terms of the absolute logic with which he perceived himself to respond to Spencer's elevation of Anglo-Saxon man as the culmination of History …

While marketers are improving their customer knowledge and value propositions, few have yet to see higher conversion rates

Although these proposed criteria are clearly an improvement over the current DSM-IV-TR criteria for children, few have yet to use them in clinical practice.

After all, we’ve become more than used to hearing corporations cry the blues about what healthcare reform is doing to their businesses, despite the fact that few have yet to be impacted by the law.

As evidence that there's something going on here, note that "most" can be substituted for "few" in those examples without changing the intended meaning.

And I wouldn't be surprised if a few commenters don't find something to say about these other misnegation examples:

[link] While many players are quick to pull the ripcord before a major, Wozniacki hung on against Roberta Vinci today despite hobbling with a bad left knee for much of the match. According to a TV translation, she complained to her father about the pain on a changeover, while he told her to go after Vinci’s backhand. In the end, Wozniacki served and volleyed her way to a 6-3, 6-3 loss. Hopefully not before she did any more damage to her knee.

[link] Noah Webster published his first dictionary, A Compendious Dictionary of the English Language, back in 1806. From then, dissatisfied with the breadth of what he had conceived, he embarked on decades of intensive work to expand his groundbreaking creation into a more comprehensive reference, An American Dictionary of the English Language. No mean slouch, according to his own account, he learned 26 languages (including my favorite, Aramaic) to unearth etymologies and tease out root sources of many of the words we use now without a second thought.



  1. Pflaumbaum said,

    June 1, 2014 @ 8:08 am

    I'm not quite getting why it's necessary to partially invoke the difficulty of keeping track of negations, along with negative concord, for the first example –

    "I wouldn't be surprised if this struggle and this tragedy had not inspired people there."

    Isn't it a textbook example of negative concord? CGEL's example of a 'pleonastic subordinate negative' is –

    I shouldn't be surprised if it didn't rain (p845)

  2. Pflaumbaum said,

    June 1, 2014 @ 8:09 am

    Sorry wouldn't, in fact, not shouldn't.

  3. Ben Hemmens said,

    June 1, 2014 @ 8:42 am

    There's a remnant of negative concord in German in the form of a pattern like: Keine Nachspeise bevor du nicht deinen Fisch aufgegessen hasst (literally: no dessert before you haven't eaten up your fish). I think this one with bevor is prevalent to the point of being basically correct (possibly not in really formal writing); analogous pattens certainly occur in speech with other things, such as the sentence quoted from Van Rompuy. Bear in mind that he seems to be a Flemish speaker, so I wonder if there's something like that in Flemish too.

  4. Jerry Friedman said,

    June 1, 2014 @ 9:16 am

    It's funny that "few have yet to accept" means the opposite of "few have yet accepted".

    Pflaumbaum: Is that "pleonastic subordinate negative" considered a type of negative concord? I thought negative concord was what we see in "I ain't never seen none of 'em nowhere."

  5. Pflaumbaum said,

    June 1, 2014 @ 9:24 am

    @ Jerry

    That's an example of negative concord in non-standard English (or I suppose standard Russian, Italian etc.).

    Mine was an example of a type of negative concord that's considered standard. Others include neither… nor… constructions and phrases like not in my car you don't!

  6. Orin Hargraves said,

    June 1, 2014 @ 10:42 am

    If you parse "I wouldn't be surprised if this struggle and this tragedy had not inspired people there." with the Stanford parser and draw the resulting tree at http://ironcreek.net/phpsyntaxtree/, you can read across the bottom of the tree to get the heart of the semantics with the opining aspect stripped away: "this struggle/this tragedy inspired people there." It would be interesting to collect some of these convoluted negations and see if there's an algorithmic shortcut that dependably pulls off the clutter.

  7. J.W. Brewer said,

    June 1, 2014 @ 10:56 am

    While Mr. Van Rompuy no doubt has excellent command of English, he apparently still composes his haiku in his L1 Dutch although they are available translated into English, French, German, and . . . (wait for it) . . . Latin. This sort of construction seems common enough for L1 Anglophones that I wouldn't place too much stock in the influence of another language, but if e.g. Dutch had obligatory negative concord in a similar context in a way that French didn't I suppose you could try to do some datacrunching to see if Flemings used a parallel construction in English more frequently than Walloons did.

  8. Michael Watts said,

    June 1, 2014 @ 11:06 am

    Few have yet to realize that in order to truly make our species interplanetary, we have to have babies on other planets.

    I tend to agree that this says the opposite of what the writer meant, but I don't exactly disagree with what it says. I think you could poll a lot of agreement with the idea "if there are no babies anywhere else but earth, we're not really an interplanetary species".

  9. Eric P Smith said,

    June 1, 2014 @ 11:44 am

    CGEL gives four constructions in Standard English where "a single semantic negative is expressed more than once":

    • Their action was neither illegal nor immoral
    • They aren't here, I don't think
    • Not in my car, you're not
    • I wouldn't be surprised if it didn't rain
    It seems to me that the fourth of these differs from the first three, because only in the fourth is the one negative within the syntactic scope of the other. Therefore the first three strike me as logical, whereas the fourth, 'pleonastic subordinate negative', strikes me as illogical. I don't know how pleonastic subordinate negative first arose, but I wouldn't be surprised if it did (!) through a few monkey brains and a few hundred million sheep, or parrots depending on your preferred metaphor.
    I fully accept that pleonastic subordinate negative is regarded as standard, and that natural language need not always be "logical" in the above sense. But I think that pleonastic subordinate negative is remarkable, because as far as I can see it is the only example of that particular illogicality in what is regarded as Standard English.

  10. D.O. said,

    June 1, 2014 @ 11:51 am

    I will be one of those few (how many commenters make a few?) who do not find anything to say about the last 2 examples. But what about the introductory sentence? There are always expected to be some commenters who write about a specific topic and some who don't. Either category can be thought of as "a few". Hence whether you interpret the introductory sentence according to the standard compositional rules or as negative concord/misnegation example you'll probably have a true statement. Is it the intended effect? Am I ruining the joke by trying to explain it? I wouldn't be surprised.

  11. Brian said,

    June 1, 2014 @ 12:26 pm

    Of the last two: I have no problem with the former, but I'm completely unable to see how the latter is at all notable or even fits the subject matter. Am I completely misinformed about the meaning of the idiom "no mean slouch"? I would have said that it translates roughly to "Not a dull underachiever", and is always used just before reporting someone's unusual accomplishment. The second part of the emphasized text just indicates that either we don't have secondhand confirmation of his linguistic mastery, or maybe that he was quick to blow his own horn. Someone please take pity on this mean slouch and explain what I'm missing.

  12. Andrew (not the same one) said,

    June 1, 2014 @ 1:21 pm

    I think 'no mean X' normally means 'a good X', 'not a minor or insignificant X', as in 'no mean achievement'. Hence, 'no mean slouch' should mean someone who is good at slouching. 'Mean' by itself can mean either 'stingy' or just 'nasty', but I don't think it can mean 'dull'; 'a mean X' is this context is an inferior example of Xness.

    However, the use of 'no mean…' with a term which is negative in force is a bit paradoxical anyway, so it's not surprising that the meaning here becomes unclear.

  13. Brian said,

    June 1, 2014 @ 2:00 pm

    > Hence, 'no mean slouch' should mean someone who is good at slouching.

    I can see how that might be a valid interpretation if it wasn't already a well-established idiom. But I'm pretty sure the prevailing meaning is "not a minor/insignificant underachiever". (I used "dull" because it was the best single synonym for the meaning in question in the reference book I had at hand, but I'll take "minor/insignificant" to avoid confusion.)

    A web search for "no mean slouch" in quotes seems to support my idea that it typically precedes a statement of admiration.

  14. M.N. said,

    June 1, 2014 @ 2:23 pm

    I think the fourth example in Eric's list ("I wouldn't be surprised if it didn't rain") is actually ambiguous between the negative-concord reading and the "logical" reading. In fact, my intuition is that the literal reading (a lack of rain would be unsurprising) is more prominent, and the "negative concord" reading comes about some other way – maybe by the same mechanism that generates both "Let's see if we can solve this problem" and "Let's see if we can't solve this problem" with the same meaning.*

    The obvious difference, though, is that that pair has the negation in an embedded question, and the original example has it in the antecedent of a conditional. It's possibly also relevant that "If it didn't rain, I wouldn't be surprised" only has the logical reading as far as I can tell.

    *I think the German phenomenon that Ben refers to has the same optionality: i.e., you can say either Das weiß man nicht, bis man nicht hingeschaut hat or Das weiß man nicht, bis man hingeschaut hat and mean exactly the same thing. I'm not a native speaker, though, so I could be wrong. There's some paper by Krifka about this, or at least mentioning it.

  15. Andrew (not the same one) said,

    June 1, 2014 @ 2:35 pm

    I can see how that might be a valid interpretation if it wasn't already a well-established idiom. But I'm pretty sure the prevailing meaning is "not a minor/insignificant underachiever".

    Yes, of course. But the point is that this idiom reverses the ordinary meaning of 'no mean…' Normally 'no mean X' is an outstanding example of X. In this case it's someone who isn't an X at all.

  16. Pflaumbaum said,

    June 1, 2014 @ 3:27 pm

    @MN –

    Yes that one is certainly ambiguous, and I believe the CGEL section mentions this. For me personally the negative concord interpretation is the more salient, perhaps because the other one involves more processing of 'real' negations. But I wouldn't be at all surprised if I wasn't in the minority there ;)

  17. Brian said,

    June 1, 2014 @ 3:28 pm

    All right, I can see that. But one could easily argue the same thing in a phrase like "No slapdash water-color, the painting is prized the world round." I doubt many would read the introductory phrase as suggesting that the artwork is a meticulous water-color.

    But in any case, idioms are idioms; they don't have to make sense.

  18. Jacob said,

    June 1, 2014 @ 3:59 pm

    The Noah Webster example puts me in mind of "Caddyshack":

    JUDGE SMAILS: You know, you should play with Dr. Beeper and myself. I mean, he's been club champion for three years running, and I'm no slouch myself.

    TY WEBB: Don't sell yourself short, Judge. You're a tremendous slouch.

  19. Jerry Friedman said,

    June 1, 2014 @ 5:06 pm

    Pflaumbaum: What "negative concord" suggests to me is that everything in the clause that can become negative has to, just as in some languages dependents on a noun have to agree with it in gender, etc. But if linguists use it for any situation where a negative is "illogical" (as Eric P Smith puts it), I can't object.

  20. Pflaumbaum said,

    June 1, 2014 @ 6:07 pm

    @ Jerry –

    I don't think it's that they use it where a negative is illogical. It's that they use it where some agreement takes place, without stipulating that everything has to agree.

    Even within the non-standard realm, a sentence like None of us seen nothin', which is perfectly possible, would be classified as negative concord, even though the verb is not negated.

    I also think it's a mistake to regard this as a matter of logic (I know that wasn't your word). It's taking too narrow a view of the meaning of 'negative' words. Take a standard construction like

    I didn't try any wine.

    Here any has a different meaning from its meaning in Try any wine you like. It's a 'negative polarity item', in concord with the negated verb because for some reason words like 'some' aren't normally permitted in negative clauses in standard English.

    But if someone claimed this was a corruption of the true meaning of any, and that the logical reading of the sentence was I didn't try just ANY wine… (i.e., I tried a particular one), that would sound daft. Yet that's essentially the same argument as the 'double negatives are illogical' one. In non-standard English (or Russian or Italian etc.), words like nobody, nothing etc. are not 'negatives' in this context – they're negative polarity items.

    That's as far as I understand it anyway, apologies if I've got any of that wrong.

  21. Eric P Smith said,

    June 1, 2014 @ 6:44 pm

    The word 'illogical' is coming in for some criticism here, and as it was I who introduced the word to the thread, I just want to clarify what I said.

    I am not saying outright that negative concord is illogical. There is nothing illogical about the French "Je n'ai rien fait." There is nothing illogical about the AAVE "I ain't done nothing." But "I wouldn't be surprised if it didn't rain" (meaning, "I wouldn't be surprised if it did rain") strikes me as illogical because in Standard English wouldn't and didn't are unambiguously negatives (not having any of the ambipolar nature of Standard English any or French ne), and the occurrence of didn't is within the syntactic scope of the occurrence of wouldn't in the quoted sentence, and it is a general rule of Standard English (unlike French or AAVE) that there is no concord between two negatives when one of them is within the scope of the other.

    And I'm certainly not equating 'illogical' with 'bad'. More like 'sturdy indefensible' which is the term Fowler used.

  22. Pflaumbaum said,

    June 1, 2014 @ 7:45 pm

    Point taken, Eric.

    Re your scope point, is it definitely the case that one negative isn't within the scope of the other in the second CGEL example, They aren't here, I don't think?

  23. Eric P Smith said,

    June 1, 2014 @ 11:52 pm

    @Pflaumbaum: Thankyou.

    Is it definitely the case that one negative isn't within the scope of the other in the second CGEL example, They aren't here, I don't think?

    I think so. CGEL calls They aren't here the anchor clause and I don't think as the parenthetical clause (p845), that is, it is added parenthetically to the anchor clause (p895). In such cases CGEL denies that the anchor clause is subordinate to the parenthetical clause (p896).

  24. Michael said,

    June 2, 2014 @ 3:00 am

    "I shouldn't be surprised if it didn't rain (p845)"

    The real problem is that such a sentence can legitimately mean two entirely different things, depending on the context! The reader is rightly confounded…

  25. Rod Johnson said,

    June 2, 2014 @ 8:37 am

    Re "I [w]ouldn't be surprised if it didn't rain": my intuition is that in the negative concord sense (i.e., the one that means it is likely to rain), the "didn't" is very lightly stressed, almost elided. If it gets enough stress to be prominent in the sentence ("I wouldn't be surprised if it DIDN'T rain") it has to mean the speaker thinks it won't rain. Intuitions like this are highly unreliable, of course.

  26. KathrynM said,

    June 2, 2014 @ 1:21 pm

    Brian: But, just how well-established an idiom is "no mean slouch"? It doesn't appear in Google ngrams at all, and Google search results show less than 125 actual hits (disregarding this post and the blog comment cited in it).

RSS feed for comments on this post