Ask Language Log: why is "whether or not" more frequent?

« previous post | next post »

Ton van der Wouden asks:

The Google Ngram viewer shows a tenfold increase in the frequency of the string "whether or not". Can the readers of language log think of any explanation for this growth? Can it perhaps be traced back to some prescriptive source? Is it perhaps accompanied by a comparable decrease of the frequency of the variant with postposed "or not", as in "whether you like it or not" — a string that is too long a search term for the Ngram viewer.



17 Comments

  1. Stan Carey said,

    May 27, 2016 @ 12:01 pm

    Given the many complex factors involved, it can be hard to nail down the cause of a change in local whether patterns.

  2. JS said,

    May 27, 2016 @ 12:33 pm

    It seems to be in part a function of what comes after… the earlier examples seem all to be "whether or not S", with the newer much more varied. Thus the trend towards "whether or not INF" is still more dramatic, as reflected by results for "whether or not to".
    whether or not to

  3. Ralph said,

    May 27, 2016 @ 12:53 pm

    And it's not just the "or not" construction. Compare "whether to" and "whether he" – they have gone in opposite directions, increasing and decreasing by a factor of about 3.

  4. Sarah said,

    May 27, 2016 @ 1:07 pm

    What I can't stand is the increasingly prevalent use of 'whether or not X or not'. I tend to have a knee-jerk reaction to speech patterns which I find particularly annoying and this one just does it every time. I don't know why; it's probably something to do with the inherent redundancy.

    (@ Stan: ba-dum-CHHH.)

  5. AndrewD said,

    May 27, 2016 @ 1:10 pm

    Is this a U.S. thing, as in my U.K. english usage the phrase " whether or not" is a reasonably well used formulation

  6. marc ettlinger said,

    May 27, 2016 @ 1:31 pm

    You get extraordinarily similar change with "regardless of":
    https://books.google.com/ngrams/graph?content=whether+or+not%2C+regardless+of&year_start=1800&year_end=2000&corpus=15&smoothing=3&share=&direct_url=t1%3B%2Cwhether%20or%20not%3B%2Cc0%3B.t1%3B%2Cregardless%20of%3B%2Cc0

    So, I'm inclined to think this has less to do with usage patterns of "whether or not" and more to do with either well-known issues in the n-gram viewer (e.g., bi-gram counting, changes in the genre-mix over time) or with some larger semantic/syntactic trend.

  7. marc ettlinger said,

    May 27, 2016 @ 1:37 pm

    …and to add to my previous comment, also consider "as a result of", "even if":
    https://books.google.com/ngrams/graph?content=whether+or+not%2Cregardless+of%2Cas+a+result+of%2C+even+if&year_start=1800&year_end=2000&corpus=15&smoothing=3&share=&direct_url=t1%3B%2Cwhether%20or%20not%3B%2Cc0%3B.t1%3B%2Cregardless%20of%3B%2Cc0%3B.t1%3B%2Cas%20a%20result%20of%3B%2Cc0%3B.t1%3B%2Ceven%20if%3B%2Cc0

  8. Stephen F. Moore said,

    May 27, 2016 @ 2:23 pm

    Looks to me like the increased usage almost inversely tracks the decline of "whether or no".

  9. Greg Malivuk said,

    May 27, 2016 @ 2:51 pm

    @AndrewD: It's not just a US thing. The Google Books ngram viewer shows the same increase in British English as in American English (though perhaps delayed slightly). It's an increase over the course of the past century or more, though, so it isn't something you'd necessarily have noticed in your lifetime on either continent.

  10. Milan said,

    May 27, 2016 @ 7:18 pm

    Maybe there is some kind of semantic bleaching going on? The "decision" element in the meaning of 'whether' is becoming weaker, so that it needs to be re-enforced with "or not.

  11. Jason Eisner said,

    May 27, 2016 @ 11:13 pm

    Oh, during the period graphed, I personally shifted away from the postposed usage in writing. To improve readability, I started rephrasing "whether X or not" into "whether or not X," particularly for lengthy X.
    Dunno whether or not you can blame it all on me.

  12. David Weman said,

    May 28, 2016 @ 3:43 am

    I play around with ngram viewer a lot and I feel like you never have just a straight line for anything. There's always trends.

  13. Ellen K. said,

    May 28, 2016 @ 9:02 am

    I notice the Ngram shows a decrease since 1980. Which drops it back to the mid-60s level, still many times higher than in the 1800s.

  14. David Morris said,

    May 28, 2016 @ 9:49 am

    Whether the whether be fine
    Or whether the whether be not
    Whether the whether be cold
    Or whether the whether be hot
    We'll weather the whether
    Whatever the whether
    Whether or not we like it.

  15. Bloix said,

    May 30, 2016 @ 1:32 pm

    I went to the ngram, and many of the examples seem vaguely ungrammatical or at least clumsy:
    "The results of this study led the researcher to reassess whether or not the academy could be a true smaller learning community."
    Ick.

  16. Clay Beckner said,

    May 30, 2016 @ 10:54 pm

    I just took a look at COHA, and found the same overall rise in "whether or not." I've been looking at individual examples, too… in the 1810s, often there was an "or [not]" later in the same clause, but not immediately after the "whether". (For example:"whether you have any regard for me or not"; and "Whether this objection be substantial or frivolous.") It appears that the "or not" gradually got glommed onto the "whether," and the sequence became a discourse unit of some kind. There also may have been other grammatical changes involving "whether" — note the subjunctive in the first example above.

  17. Tom V said,

    June 4, 2016 @ 8:33 pm

    @David Morris
    That last line was just plain diabolical. Fie upon thee!!!

RSS feed for comments on this post