Norwood

« previous post | next post »

In discussing the relatively low rate of contraction in Charles Portis's novel True Grit, I suggested several different explanations. It might be false archaism, or it might be a way to bring out the personality of the narrator, Mattie Ross. Another option, of course, would be that it's a quirk of the author's style. We can eliminate this last possibility by checking another of his novels, Norwood, which (according to Wikipedia)

… follows its namesake protagonist on a misadventurous road trip from his hometown of Ralph, Texas, to New York City and back. During the trip, Norwood is exposed to a comic array of personalities and lifestyles. The novel is a noteworthy example of Portis' particular skill rendering Southern dialect and conversation.

The only easy way to estimate contraction rates is to use the page-counts in Google Books, which gives us the following counts and percentages:

Norwood True Grit
don't/do not 105/2 (98%) 14/11 (56%)
won't/will not 14/1 (93%) 6/6 (50%)
wouldn't/would not 24/1 (96%) 1/11 (8%)
it's/it is 56/23 (71%) 2/21 (9%)
that's/that is 56/3 (94%) 2/17 (11%)
I'm/I am 68/9 (88%) 0/26 (0%)

You may notice that the counts are somewhat different from those I gave in my earlier post. This reflects the bizarre and puzzling non-determinism of Google Books. Thus when I checked yesterday evening, GB told me that True Grit had 5 pages containing won't and 9 containing will not — this morning it tells me that the counts are 6 and 6. Yesterday evening, True Grit had 12 pages containing don't and 8 containing do not — this morning, it has 14 and 11 respectively. And so on. But despite these peculiarities, I'm going to trust that the large reported differences in contraction-percentage between Norwood and True Grit are real.

The contraction rates estimated for Norwood are simiar to those that we saw in Tom Sawyer and Swan Peak. And the distribution of variants also follows the sociolinguistic patterns that we expect: for example, the only two examples of "do not" in Norwood are these:

… and took down the "I Do Not Loan Tools" sign from the persimmon tree.
… although the signs said DO NOT TALK TO OPERATOR he started right in cracking jokes and carrying on with the passengers.

So I conclude that the low contraction rates in True Grit represent an artistic choice by Portis. Unfortunately, none of his other novels are set in the 19th century, so we can't determine whether he believed that people talked differently in this respect back them, or simply felt that lower contraction rates suited Mattie. Since they do, I'm going to give him the benefit of the doubt on this.



11 Comments

  1. Proudhon said,

    December 30, 2010 @ 1:35 pm

    It seems that no one here has actually read the book, only counted contractions. True Grit is narrated, in its entirety, by Mattie Ross, some forty-odd years after the events she depicts occurred. The dialogue is never presented by Portis as it may actually have been spoken, but as Mattie herself, now in her sixties, wrote it. Might Mattie have presented the dialogue in a stilted fashion simply because she didn't do dialogue very well? Sure. It was the only time she ever did this kind of thing. Might she have remembered a formality that wasn't really there, giving the highlight of an otherwise uneventful life something of a chivalric aura. Sure. Are there other possibilities that make this book (and all of Portis's work) worth reading for reasons beyond the number of contractions contained therein? I would not be surprised.

    [(myl) It seems that you have not read the posts that you're commenting on. From the cited discussion:

    Charles Portis originally published "True Grit" as a serial in the Saturday Evening Post in 1968, and in somewhat changed form as a book. The story is narrated in the first person by Mattie Ross as an old woman in 1928, describing her adventures many years earlier at the age of fourteen. […]

    Is this degree of variation [in contraction percentages] … an accurate representation of 19th-century American speech patterns? Or was Portis using an exaggerated (if variable) lack of contraction as a way to project the personality of Mattie Ross, an elderly woman whose strong notions of right and wrong may have included some prescriptive feelings about contractions?

    And Otter wrote in a comment:

    I noticed the absence of contractions in Maddie's speech, and it seemed as much specific to her character as related to the period. A precocious girl, schooled (by herself or someone else) with higher-class pretensions and large helpings of scripture, might tend to avoid contractions, and the fierce front she puts on when dealing with adults and trying to gain their respect and compliance might well increase this tendency.

    However, our double-your-money-back guarantee remains in force, and the LL public relations department has prepared a check for twice your subscription fee, which they will cheerfully hand you on your way out the door.]

  2. Marcus Bales said,

    December 30, 2010 @ 6:28 pm

    Damon Runyon did the same kind of thing in his Broadway stories. His thugs and hoods spoke without contractions (and Frank Loesser picked this peculiarity up in "Guys and Dolls") as a deliberate choice to make us think about the slurring slang of the underworld without actually doing a Francois Villon (here translated by WE Henley):

    Villon’s Straight Tip to All Cros Coves
    “Tout aux tavernes et aux fiells”
    WE Henley

    Suppose you screeve? Or go cheap-jack?
    Or fake the broads? Or fig a nag?
    Or thimble-rig? Or knap a yack?
    Or pitch a snide? Or smash a rag?
    Suppose you duff? Or nose and lag?
    Or get the straight, and land your pot?
    How do you melt the multy swag?
    Booze and the blowens cop the lot.

    Fiddle, or fence, or mace, or mack:
    Or moskeneer, or flash the drag;
    Dead-lurk a crib, or do a crack;
    Pad with a slang, or chuck a fag;
    Bonnet, or tout, or mump and gag;
    Rattle the tats, or mark the spot;
    You cannot bag a single stag;
    Booze and the blowens cop the lot.

    Suppose you try a different tack,
    And on the square you flash your flag?
    At penny-a-lining make your whack,
    Or with the mummers mug and gag?
    For nix, for nix the dibbs you bag!
    At any graft, no matter what,
    Your merry goblins soon stravag:
    Booze and the blowens cop the lot.

    Envoi:
    It’s up the spout and Charley Wag
    With wipes and tickers and what not
    Until the squeezer nips your scrag,
    Booze and the blowens cop the lot.

  3. Bryan said,

    December 30, 2010 @ 9:11 pm

    Ápropos of talking about Google and contractions I present a strange little mystery I discovered:

    http://ngrams.googlelabs.com/graph?content=i'm,i+am&year_start=1700&year_end=1800&corpus=0&smoothing=3

    http://ngrams.googlelabs.com/graph?content=i'm,i+am&year_start=1700&year_end=2010&corpus=0&smoothing=3

    A graph of "I'm" vs. "I am" has a very strong spike in "I'm" for 1708.

  4. Chris said,

    December 30, 2010 @ 9:43 pm

    But despite these peculiarities, I'm going to trust that the large reported differences in contraction-percentage between Norwood and True Grit are real.

    Why? It's abundantly clear that Google's "word-counts" aren't actually the result of counting words, so why would you then go on to assume that they are? If you're interested in providing evidence of word frequency, don't use Google, or your readers will assume you're just making up evidence.

    Sorry, I've just had it with linguistics arguments made using Google's "data." It's not repeatable, or verifiable with any data source besides Google. It's not science, it's an appeal to an authority that's demonstrably wrong.

  5. Clayton Burns said,

    December 30, 2010 @ 9:47 pm

    From a review at thestar.com: "Joel and Ethan Cohen’s casting staff spent 18 months seeing 'thousands' of young actresses in their search for a girl who could handle the difficult King James Bible-inspired formal and contraction-free dialogue of True Grit, what co-star Barry Pepper dubbed 'American Shakespeare.'"

    On pages 122 and 123 of The Overlook Press paper "True Grit," with an afterword by Donna Tartt, there are over a dozen contractions (Rooster and Quincy arguing). Therefore, at The Star it is safe to discuss books you have not read.

    Formalism in Mattie's modal use is an interesting subject. A factor that might make us reluctant to attribute the formal style purely to Mattie is that we have a 14-page "transcript" of court proceedings and a page- long letter from lawyer Daggett.

    Somehow, I think, the issues of prose style in Shakespeare and in the King James Bible as influences here remain unfocused (to a degree). Perhaps Robert Alter could (re)read the book and watch the film twice so as to offer an informed opinion.

    Even if the novel lacks depth (to some extent), both the novel and the new film seem to be getting a lot of traction. It is a bit like the case of the ivory-billed woodpecker. As long as you mention it, you know you will get a lot of links.

    What we really have to have is Mark's comprehensive analysis of the film. I look forward to that with real anticipation.

  6. Twitter Trackbacks for Yesterday evening, True Grit had 12 pages containing don't and 8 containing do not — this morning, it has 14 and 11." [upenn.edu] on Topsy.com said,

    December 30, 2010 @ 10:48 pm

    […] Yesterday evening, True Grit had 12 pages containing don't and 8 containing do not — this morning,… languagelog.ldc.upenn.edu/nll/?p=2875 – view page – cached December 30, 2010 @ 12:07 pm · Filed by Mark Liberman under Language and […]

  7. Bloix said,

    December 30, 2010 @ 11:21 pm

    IIRC, Laura Ingalls Wilder"s style is characterized by an infrequent use of contractions.

  8. Bloix said,

    December 31, 2010 @ 10:00 pm

    Well, IRI. The narrator of the Little House books (who is, presumably, the adult Laura) uses formal diction without contractions, but the characters use them freely.

    [(myl) Interesting. Though perhaps not in such an extreme form, this is typical of fiction over many decades, and to some extent up to the present day, since formal written style has traditionally discouraged contractions, and in principle fiction represents spoken style in its dialogue and written style in its narration.

    First-person narration is more ambiguous, since it might be taken as a form of speech.

    And (this aspect of) the contrast is gradually diminishing, as contractions come to be used at greater rates in written styles.]

  9. Mark Mandel said,

    January 1, 2011 @ 1:29 am

    Bryan@Dec. 30, 9:11: Look closely at that "I'm" spike. Get the Google books range for that year and examine the text, if you can: http://www.google.com/search?q=%22i%27m%22&tbs=bks:1,cdr:1,cd_min:1708,cd_max:1708&lr=lang_en

    The first work listed, with its snippet, is:
    The true characters of, viz: A deceitful pretty-fogger, vulgarly … – Page 14
    1708 – 16 pages – Full view
    … Mii' l| 'i|l ni| "| M' 'yi i' X i V r i'i 7'iV i ii"i i"| l'lpi" "> '/"l' J 'v i' m ' '/* " i i' \i'i tiii … 'H |'|«lM|l M| |M|\\ "l""i' M' |'iH i'l 'l | \|'i|W iii':iiii i'U "' 7'" i lV ' \ '\H'| V lW\l M'ff iM l'Ml'OV o …

    That atrocious gibberish comes from OCR scanning of an image which is so washed-out that barely a letter, let alone a word, can be distinguished on it. O click and see!

    There are indeed some real cases of "I'm" in the snippets, such as "He did not make so long a Speech to me, I'm sure oft, tho I brought him Apples* Bel."

    I have no intention of searching through the 90 results, many of which allow "No preview". But I am quite confident that that huge spike in "I'm" is entirely valueless.

    Chris@Dec.30 9:43 pm: Not only is it true that 'Google's "word-counts" aren't actually the result of counting words': even the words they start their hand-waving from aren't words.

  10. Clayton Burns said,

    January 1, 2011 @ 2:43 am

    What emerges from repeated viewing of "True Grit" (I have seen it four times) is that dramatic texture is created by variations in patterns of contracted and full forms.

    This a a feature of the book that is sharpened in the film, not systematically, but certainly for the film's equivalent of the Texas Road dugout scenes in the book where the Quincy-Moon action happens.

    In The Overlook Press version of the novel in the stores here, the Quincy-Moon sequence begins on page 118. If in watching the film you concentrate on contractions, you will note that the first major cluster occurs in the dugout, vehemently, violently (in the book, the harrowing encounter with Quincy and Moon attracts about 35 "don't" or "didn't" contractions, even drawing Mattie in so that she responds with three of her own pointed comments containing "don't" (126), to Quincy.

    In the film, these contractions in the Quincy-Moon scene have surreal power, partially because so far there were no such clusters (I have read about the Pullum update, but I do not think that the terminology, although interesting, is decisive). Obviously, the power of the dugout scene emerges out of great acting, but the language is weirdly effective.

  11. Contractions and Verisimilitude in “True Grit” « English Grammar Blog said,

    January 24, 2011 @ 11:18 am

    […] question of the contractions comes up in two blog entries by the linguist Mark Liberman here and here. Liberman does an interesting analysis of the near complete lack of contractions in not only the […]

RSS feed for comments on this post