Death before syntax?

« previous post | next post »

Ursula K. LeGuin, "Introducing Myself":

What it comes down to, I guess, is that I am just not manly. Like Ernest Hemingway was manly. The beard and the guns and the wives and the little short sentences. I do try. I have this sort of beardoid thing that keeps trying to grow, nine or ten hairs on my chin, sometimes even more; but what do I do with the hairs? I tweak them out. Would a man do that? Men don’t tweak. Men shave. Anyhow white men shave, being hairy, and I have even less choice about being white or not than I do about being a man or not. I am white whether I like being white or not. The doctors can do nothing for me. But I do my best not to be white, I guess, under the circumstances, since I don’t shave. I tweak. But it doesn’t mean anything because I don’t really have a real beard that amounts to anything. And I don’t have a gun and I don’t have even one wife and my sentences tend to go on and on and on, with all this syntax in them. Ernest Hemingway would have died rather than have syntax. Or semicolons. I use a whole lot of half-assed semicolons; there was one of them just now; that was a semicolon after “semicolons,” and another one after “now.”

It's a great essay — you should go buy the book immediately. But the thing is, Hemingway having "little short sentences" is one of those curious backwards stereotypes, like the one about Obama using first-person singular pronouns to an inordinate degree, or the one about women talking more than men.

The first 554 words of Ursula K. LeGuin's wonderful novel The Dispossessed contain 39 sentences, for an average length of 14.2 words per sentence. The first 572 words of Ernest Hemingway's justly famous novel For Whom the Bell Tolls (excluding dialogue) contain 20 sentences, for an average length of 28.6 words per sentence.

The first 759 words of Ursula K. LeGuin's essay "Introducing Myself" contain 51 sentences, for an average length of 14.9 words per sentence. The first 786 words of Ernest Hemingway's essay "Miss Stein Instructs"  contain 31 sentences, for an average length of 25.4 words per sentence.

The 598 words of Ursula K. LeGuin's essay "Being Taken for Granite" contain 46 sentences, for an average of 13 words per sentence. The first 764 words of The Old Man and the Sea (excluding dialogue) contain 28 sentences, for an average of 27.3 words per sentence.

As for whether Hemingway "would rather have died than have syntax", again it's a long-standing stereotype that he avoided clausal embedding — see the 1927 Thurber parody quoted in "Homo Hemingwayensis" (1/9/2005), which is also an early expression of the "little short sentences" meme. But it's really not true, though Hemingway's generation rejected the elaborate syntactic gingerbread of late-19th-century prose in favor of a plainer and more paratactic style. To say that Hemingway "would rather have died than have syntax" is a neat hyperbolic sketch of that stylistic difference — but I suspect that by any plausible measure, the Hemingway passages cited above will have more "syntax" than the LeGuin passages.

Nor does Hemingway definitively prefer death to semicolons — the 2,122 words of the three Hemingway passages quoted above include 5 semicolons, for a rate of one per 424 words. The 1,911 words of the three LeGuin passages quoted above also include 5 semicolons, for a rate of one per 382 words, so that in this sample, LeGuin is a bit more prone to semicolonize than Hemingway is. But death before semicolons? Not hardly.

 



17 Comments

  1. Robot Therapist said,

    October 20, 2014 @ 4:59 am

    I spent an afternoon once with Ursula LeGuin in a punt on the river in Oxford. She was smoking a pipe. Happy days.

  2. edithcuth said,

    October 20, 2014 @ 5:19 am

    I can't believe I don't know the answer to this after so many years of school, but: what do you count as a sentence? Is it simply the words between two periods (or question marks, or exclamation marks), or is it something else?

    [(myl) It's not quite as simple as that, since (for example) "We went 3.2 miles." is not two sentences, and "Wow!!!" is not three. Empirically, I wrote a script some years ago that tries to divide English text into sentences in a way that conforms to most people's intuitions, and the results for the works cited are here, here, here, here, here, here.

    The program isn't perfect, but it's good enough for stuff like this.]

  3. Deirdre said,

    October 20, 2014 @ 6:19 am

    @edithcuth
    I thought it was whether the phrasing included a finite verb (plus certain sentence substitutes such as 'Yes.'). I'm interested to know if that definition is not longer considered correct or complete.

  4. Bill Benzon said,

    October 20, 2014 @ 7:15 am

    Thanks for this, Mark. I don't know where this false truism started or why it persists (because English lit scholars don't know syntax?). A couple of years ago I took a close look at two sentences from Death in the Afternoon, a set of essays on bullfighting. One was 170 words long and the other 131. I didn't choose them for their length, but the length is probably not incidental to what he was doing with them. I argued, in one case, that his prose rhythm "imitated" the cape pass he was describing. Whatever he was doing, he was a master of prose rhythm:

    While any written document will have some kind of rhythm by virtue of the fact that language unfolds in time. But most people do not deliberately manipulate prose rhythm. Hemingway does. Is there a statistical difference between writing where the author deliberately manipulates rhythm and writing where the rhythm is not taken into account? Is the relative lack of subordinating conjunctions – assuming it to be statistically real – in Hemingway’s prose a device that gives him greater rhythmic freedom? That is, subordination places emphasis on logical relations within a sentence, relations that must be satisfied. If you drop such relations you have more freedom with word placement.

    [(myl) The "little short sentences" thing doesn't depend on any knowledge of syntax, but simply on the ability to count. Since even literary scholars in our culture can count, I suspect that the issue is the difference between generalizations based on overall impressions (often influenced by cultural expectations), and generalizations held responsible in some way to empirical observation.

    Of course, LeGuin is just making a joke about her self-perceived failure to match stereotypes of laconic masculinity, with a false assumption about Hemingway's sentence lengths as collateral damage. But I don't think she'd be very happy if the the tables were turned in some way, for example in a joke based on falsely attributing to her writing some questionable stereotype about the science fiction genre.]

  5. Jerry Friedman said,

    October 20, 2014 @ 9:22 am

    Some of Hemingway's short stories have conspicuously long sentences, as I recall, although they tend to be paratactic. "A Clean Well-Lighted Place"?

    One thing I notice about the page in The Old Man and the Sea that GB takes me to is that a lot of the sentences begin with the subject noun phrases, not adverbs or prepositional phrases or subordinate clauses, and those noun phrases tend to be very simple. Three consecutive sentences begin "They", all referring to the shark's teeth.

    I suppose I should count. Eighteen sentences begin with the subject, which in no case is more than three words, and six begin in some other way. There are also two very short sentences of the old man's thoughts that have no main verb; each begins with a noun phrase.

    A random page from Le Guin's Always Coming Home has fifteen sentences starting with the subject noun phrase, the longest of which is eleven words, and seven starting in some other way. A random low-dialogue page from City of Illusions has ten starting with the subject and six starting some other way, including two of dialogue. So the difference isn't that big, but maybe there is one.

    (And what's this thing where the software doesn't remember my name and e-mail address any more? Just something I did?)

  6. cameron said,

    October 20, 2014 @ 10:08 am

    My first question on reading the text quoted above was "how widespread is this usage of 'tweak' in lieu of 'tweeze'?"

  7. J. W. Brewer said,

    October 20, 2014 @ 11:07 am

    http://talltalestogo.wordpress.com/2012/01/14/the-writing-life-the-point-of-the-long-and-winding-sentence/ lists a bunch of (male) writers who are stereotypically associated with long sentences and a general sort of maximalist/convoluted/non-laconic style. Who knows if any of those stereotypes are based on any sort of actual empirical counting versus vague impressions, but I think it's telling that it includes both writers where that sort of style might be pointed to as supporting evidence by someone trying to characterize the writer as effete or unmanly (Proust, H. James), and others where the complexity of the syntax is an inherent part of a somewhat aggressive/show-offy/in-your-face style (Melville, Pynchon) that could easily, and indeed pejoratively, be dismissed by a hypothetical critic as characteristically masculine. It's not unlike the phenomenon where employing a lot of first-person-singular pronouns might be self-effacing as used by one writer/speaker in one context but self-aggrandizing as used by a different writer/speaker in another, making the count itself (even if done accurately) essentially meaningless as a diagnostic tool.

  8. Gregory Kusnick said,

    October 20, 2014 @ 2:01 pm

    Hemingway himself may be at least partly to blame for the "simpler is better" stereotype of his prose. Consider his reply to Faulkner's charge that Hemingway "has never been known to use a word that might send a reader to the dictionary":

    Poor Faulkner. Does he really think big emotions come from big words? He thinks I don’t know the ten-dollar words. I know them all right. But there are older and simpler and better words, and those are the ones I use.

  9. Lance said,

    October 20, 2014 @ 2:40 pm

    If you take Hemingway's novels, you get the kind of results Mark did, but what if you take his short stories? "The Killers" is the one that always comes to mind, and the text of it is (surely entirely legally) here. The opening:

    The door of Henry's lunchroom opened and two men came in. They sat down at the counter.

    "What's yours?" George asked them.

    "I don't know," one of the men said. "What do you want to eat, Al?"

    "I don't know," said Al. "I don't know what I want to eat."

    Outside it was getting dark. The street-light came on outside the window. The two men at the counter read the menu. From the other end of the counter Nick Adams watched them. He had been talking to George when they came in.

    Even setting aside the dialogue (though, granted, most of the story is dialogue), you're definitely talking about short sentences. The last of them does have an embedded clause, so it contains syntax, but it's a very short one.

    Similarly, the opening paragraph of "Hills Like White Elephants" has sentences that are not wholly short, but their lengths are to some extent a factor of two sentences, or at least two predicates, being joined by "and" (with no semicolon in sight). Things like It was very hot and the express from Barcelona would come in forty minutes. It stopped at this junction for two minutes and went on to Madrid., or a few paragraphs later She put the felt pads and the beer glasses on the table on looked at the man and the girl.

    Categorizing writing as masculine or feminine is a fool's game, perhaps particularly in Le Guin's field where Robert Silverberg famously described James Tiptree, Jr.'s writing as "ineluctably masculine". (Tiptree later turned out to be a pseudonym for Alice Sheldon.) I wouldn't try to call short, simple sentences "masculine"; but I still wouldn't hesitate to call them "Hemingwayesque".

  10. J. W. Brewer said,

    October 20, 2014 @ 3:03 pm

    Because Hemingway had such an over-the-top manly-man public persona (see http://www.kentstateuniversitypress.com/2010/all-man/ for a more-entertaining-than-usual academic study of literary celebrity/reputation among a mass audience), it is perhaps a plausible mistake to associate other things (such as a distinctive prose style) that are viewed as Hemingwayesque with machismo. But it is still a mistake. For example, Hemingway had a reputation as a heavy drinker (like a lot of other writers, both tough-guy and louche/effete), but was an early popularizer of the blender-created frozen daiquiri, which is frankly, at the level of crude social stereotypes, not a very manly drink (except for the machismo associated in downing some ridiculously high number in a single sitting).

  11. Piyush said,

    October 20, 2014 @ 3:14 pm

    @Gregory Kusnick:

    It is always surprising that even well known writers are not above criticizing other writers for not employing "big words". The most bizarre case I have seen of this is article by Shashi Tharorr, a somewhat well known novelist in his own right, who criticizes R. K. Narayan (perhaps the most admired English fiction writer from India) for a "shallowness of … vocabulary" and for "inadequacy of the language used" to describe emotionally charged scenes, and even more so, does so in a piece which is supposed to be an obituary of sorts.

    I thought it was clear that these writers were not writing to get the highest grade in a primary school essay contest.

  12. aka said,

    October 20, 2014 @ 4:22 pm

    Decades ago a teacher assigned reading The Bear (Faulkner?) and I recall page after page with no periods. Was that a printing error? If not, do academics have a word for it?

  13. Douglas Bagnall said,

    October 20, 2014 @ 6:00 pm

    @edithcuth, myl

    You can do much the same thing as Mark's script using Python and NLTK, with something like this:

    from nltk import sent_tokenize
    sentences = sent_tokenize(text)
    print len(sentences)

  14. Eric P Smith said,

    October 20, 2014 @ 7:06 pm

    @Deirdre

    I'm not an expert, but I have a copy of Huddleston & Pullum’s The Cambridge Grammar of the English Language (CGEL), and I could swear by it.

    In matters of syntax, CGEL regards the clause as a more basic unit than the sentence. The meaning that you are aiming at in your comment is the meaning of what CGEL calls a “main clause”, and is basically still valid in CGEL’s terms subject to one change in terminology: CGEL refers to finite and non-finite clauses, not finite and non-finite verbs. CGEL does use the word “sentence” with a syntactic meaning – saying, for example, that a sentence may consist of a main clause or a coordination of main clauses – but does not define it, and indeed hints that it cannot be satisfactorily defined.

    By contrast, in matters of orthography, CGEL defines an orthographic sentence in terms of punctuation. Leaving aside complications such as the ones that Mark has indicated in his reply to @edithcuth's comment, an orthographic sentence is a unit of writing that begins with a capital letter and ends with a full stop, question mark, or exclamation mark.

    The difference between the syntactic meaning and the orthographic meaning is illustrated by an example given in CGEL:

    There's another reason why we should hesitate – it is likely that interest rates will rise again in a few months.

    Words 1 to 7 constitute a syntactic sentence, words 8 to 20 constitute a syntactic sentence, and the whole thing constitutes a syntactic sentence. But the only orthographic sentence is the whole thing.

    Hope this helps.

  15. J. F. said,

    October 20, 2014 @ 9:07 pm

    @cameron

    I'd never heard "tweak" used this way. There are only a little more than a dozen ghits for "tweak hairs", and well over ten thousand for "tweeze hairs".

  16. pj said,

    October 21, 2014 @ 8:13 am

    @cameron, J.F.
    The first definition for 'tweak' (the verb) in OED is

    To seize and pull sharply with a twisting movement; to pull at with a jerk; to twitch, wring, pluck

    You can do that to hairs, no? I don't see why tweezing is really coming into it. It doesn't mean the same thing (though you could, I suppose, use tweezers to tweak). She's not using one 'in lieu of' the other, any more than Hemingway is using 'lunchroom' in lieu of 'restaurant', say, in Lance's extract. He's using lunchroom because I take it he means lunchroom.

    She may, if you like, be doing one in lieu of the other, but that's a different charge entirely.

  17. Xmun said,

    October 21, 2014 @ 8:56 pm

    @Deirdre
    I haven't consulted any grammar book about your question, but (in brief) I don't think a main verb is necessarily present in a sentence. I can think of lots of perfectly intelligible, complete utterances that lack verbs, e.g.
    Everybody out!
    Fair enough!
    Just a moment.
    Nonsense, my dear.
    On with my coat and out into the night. (A line from a poem by Belloc.)
    No doubt the grammarians have a term for these utterances but I can't remember what it is.

RSS feed for comments on this post