The nanosecond of uncertainty

A couple of years ago, Neal Whitman and Mark Liberman scrutinized a claim by James J. Kilpatrick. From Mark’s summary, here:

James Kilpatrick complained in print about the “horrid” headline “Mass Transit Not An Option for All Drivers”, on the grounds that “if mass transit is not an option for ‘all’ drivers, it cannot be an option for even one driver”. He added, “Even a little ambiguity is a dangerous thing. The problem with this Horrid Example is that it creates a nanosecond of uncertainty.”

Neal Whitman and I ignored the “nanosecond of uncertainty” business, since a literal application of this idea would put pretty much all of the English language off limits.

Mark and Neal focused instead on Kilpatrick’s treatment of negation and quantification (and Jan Freeman joined in with a discussion of another example from this point of view). Here I’m going to go a bit further with the “nanosecond of uncertainty” matter and the dangers of “even a little ambiguity”.

Mark is right, of course, that avoiding the potential for ambiguity is a hopeless goal; potential ambiguities are an omnipresent feature of languages (as I’ve noted in several Language Log postings, for instance here, here, and here). As a result, language processing is just one damn nanosecond (well, millisecond) of uncertainty after another. Almost all of these moments are below the level of consciousness, though some people are a bit more sensitive to the possibility of multiple interpretations than others are.

Now for some cases (recently collected, all in print) where I was aware of a moment of uncertainty.

Coordination. First, one (from Dan Mahaffey) that required a serious moment of rethinking:

Several groups trying to re-ignite New England’s faith are theologically conservative, such as the Southern Baptists, Presbyterian Church in America and the Conservative Baptists’ Mission Northeast. They say a reason for the region’s hollowed-out faith is a pervasive theology that departs from traditional Biblical interpretation on issues such as the divinity of Jesus, the exclusivity of Christianity as a path to salvation and homosexuality. (link)

As Mahaffey pointed out in e-mail, a serial comma (after salvation) would make it clear that homosexuality is the third conjunct in a series and that salvation and homosexuality is not a constituent. But then the Associated Press doesn’t use serial commas.

Now one from “Colors From a World of Black and White” (by Holland Cotter, New York Times Week in Review, October 11):

Thomas was born in Columbus, Ga., in 1891,
and moved to Washington [D.C.] in her teens… Her par-
ents had relocated for two reasons: racial vio-
lence was on the rise in Georgia and Washington
had excellent public schools.

I’ve reproduced the line divisions here, because they contribute to the moment of uncertainty, in which Georgia and Washington is interpreted, nonsensically, as a constituent. A comma separating the two conjoined clauses would have helped things, but the Times doesn’t require this comma when the conjoined clauses are short.

Modifier attachment. From the NYT again (“N.F.L. Players With Head Injuries Find a Voice”, by Alan Schwarz, October 28):

Sitting at a restaurant here Friday, she reconnected with a few Buccaneers retirees. There was Richard Wood, the fearsome linebacker known as Batman whose searing migraines and tendency to get lost in his own neighborhood leave him scared for his future. Across the table was Scot Brantley, an even harder hitter through the 1980s whose short-term memory is gone. Then there was Brandy Winans, former wife of Buccaneers lineman Jeff Winans, who slipped into such inexplicable depression, fogginess and fury several years ago that their marriage splintered.

At first I took that final relative clause to be modifying Brandy Winans, even though it comes right after Buccaneers lineman Jeff Winans — probably because there are three sentences in sequence, the first about Richard Wood, the second about Scot Brantley, the third about Brandy Winans.

Then one involving a modifying adverbial (NYT, “Transcripts of Defeat” by Victor Sebestyen, October 29):

The Soviet leaders realized that they had blundered soon after the invasion.

That is, soon after the invasion, the Soviet leaders realized that they had blundered. This is clear in the context of the op-ed piece, but attachment to the verb blundered (rather than realized) is tempting.

Lexical ambiguity. And, finally, in a headline in the NYT on October 29:

More Britons for Afghan War

At first I read this as saying that the number of Britons in favor of the Afghan War has increased. But then the body of the teaser went on:

Prime Minister Gordon Brown set conditions for sending 500 more British troops to Afghanistan.

Oh, those prepositions!

Note. I wouldn’t go so far as to classify any of these examples as outright mistakes, or to insist that they should have been re-written for greater clarity. I’m probably more sensitive to the possibility of multiple interpretations than most people, and in any case once you start looking for things like potential attachment ambiguities, you’ll find them everywhere.

6 Responses to “The nanosecond of uncertainty”

  1. Z. D. Smith Says:

    I wish you hadn’t started off with that quotation. Even though the original absurd position isn’t the matter at hand, I get so irritated at people who fling around such judgments as ‘horrid’ and ‘awful’ and ‘pathetic’ and so on, with regard to constructions that are so clearly part of the mainstream of English utterance that any reader who would use that ‘nanosecond’ to apply their standard complement of pragmatic heuristics (that is, all of them, save the damaged and lonely pundit who goes looking for ambiguity where there is none), would inevitably, without fail, understand the headline as intended. Truth be told, it irks me so because I see it as an act of willful ignorance, in fact. It is a willful violation, on the part of the reader, of the conversational maxims, and those make me terribly upset. The act of taking a good faith utterance, discarding all pragmatic shading, and attacking the speaker for what is ‘rationally’ implied or denoted by the utterance, when the speaker has no recourse except the always weak-sounding ‘but that’s obviously not what I _meant_’, is the most cowardly and contemptible thing. But now I’m all farklemt.

  2. Z. D. Smith Says:

    Incidentally, to explicitly avoid tarring you with that same brush, I should say that it is quite easy for me to see the ambiguity in each one of your own sample sentences. Many of those I myself misread at first run-through. The difference is that you are demonstrating incompatibilities, or at least tensions, between the natural, or potential, first reading of a sentence and the necessarily artificial house style of that sentence’s publisher—language and composition of any sort will never be quite so systematic or rational as the deviser of any blanket rules of style would prefer—whereas Mr. Kilpatrick is attempting to exploit the incompatibility between the natural first reading of a sentence and the notional, theoretical reading of what that sentence SHOULD mean, if indeed the English language operated according to one man’s opinion of the systematic and regular operation of the modifier ‘all’; but it does not.

  3. mollymooly Says:

    The potential for ambiguity is greater in writing than speech, because visual cues and prosody are stripped away and no possible regime of punctuation or other orthographic buttressing can compensate.

    The consequences of ambiguity are more severe in writing than conversation, because there is no audience feedback to help identify when a misconstrual has arisen.

    Be careful out there.

  4. Z. D. Smith Says:

    Too true. And that is exactly why nearly all of the above-provided examples hinge on deficiencies or matters of punctuation or prosody. But as for Kilpatrick’s example, the one that gets me so worked up, I cannot imagine any naturally-uttered version that would lead me to the interpretation that he claims to favor. If one wanted to use the construction ‘Mass transit not an option’ in the same sense as ‘Failure is not an option’, they simply would not coordinate with ‘all drivers’. ‘Anybody’, maybe.

  5. arnoldzwicky Says:

    To Z. D. Smith: the postings by Liberman, Whitman, and Freeman savage Kilpatrick along the lines of your critiques above. On the claim about the scoping of negation and all, my summary is here:

    Mark Liberman mused on KR [“Kilpatrick’s Rule”] on Language Log …, disputing Kilpatrick’s claim, as did Neal Whitman and Jan Freeman. All three writers maintained that wide-scope negation was by far the most natural reading for examples like [the mass transit headline], and Mark provided a pile of examples, from a variety of respected writers over the centuries, in which negation scopes over the quantifier all, and he hadn’t found any examples with the other scoping.

    Liberman also tried to plumb the mind-set that would lead to absurd pronouncements like Kilpatrick’s. Read the postings (links in my neg-quant posting).

  6. Edward I as Oliver Cromwell « Arnold Zwicky's Blog Says:

    […] 10/31/09: The nanosecond of uncertainty (link): temporary ambiguity in parsing induced by avoiding the serial comma (in an example from Dan […]

Leave a Reply

%d bloggers like this: