Scientific prescriptivism: Garner Pullumizes?
« previous post | next post »
The publisher's blurb for the fourth edition of Garner's Modern English Usage introduces a new feature:
With more than a thousand new entries and more than 2,300 word-frequency ratios, the magisterial fourth edition of this book — now renamed Garner's Modern English Usage (GMEU)-reflects usage lexicography at its finest. […]
The judgments here are backed up not just by a lifetime of study but also by an empirical grounding in the largest linguistic corpus ever available. In this fourth edition, Garner has made extensive use of corpus linguistics to include ratios of standard terms as compared against variants in modern print sources.
The largest linguistic corpus ever available, of course, is the Google Books ngram collection. And "word-frequency ratio" means, for example, the observations that in pluralizing corpus, corpora outnumbers corpuses by 69:1.
I applaud this move in the direction of empirically-based prescriptivism — for some earlier discussion of the idea, see "Prescriptivist science", 5/30/2008, or "A test kitchen for stylistic recipes", 6/1/2008. And as a result of this development, it seems that Bryan Garner and Geoff Pullum are now on the same team, at least as viewed from a sufficiently distant perspective. Consider this sentence from the blurb for Garner's book:
Garner liberates English from two extremes: both from the hidebound "purists" who mistakenly believe that split infinitives and sentence-ending prepositions are malfeasances and from the linguistic relativists who believe that whatever people say or write must necessarily be accepted.
This is parallel to Geoff Pullum's point in "'Everything is Correct' vs. 'Nothing is Relevant'", 1/26/2005, where he objects to a writer who
… cannot see any possibility of a position other than two extremes: on the left, that all honest efforts at uttering sentences are ipso facto correct; and on the right, that rules of grammar have an authority that derives from something independent of what any users of the language actually do.
But there had better be a third position, because these two extreme ones are both utterly insane.
And Garner seems even more Pullum-like in his interview with Daniel McMahon ("How the world's leading authority on the English language used Google to write the most comprehensive treatment of English usage ever published", Business Insider 4/26/2016):
Daniel McMahon: This new fourth edition of your main usage book, "Garner's Modern English Usage," replaces the third-edition "Garner's Modern American Usage." Besides the title, which we'll get to, what's different in the new book?
Bryan A. Garner: The biggest change is the level of empiricism underlying all the judgments. I made extensive use of corpus linguistics, and especially of Google Books and the ngrams, to assess the judgments that I've made in previous editions, and it was a most enlightening process. I've added almost 2,500 ratios of the most current available information about how many times one form — the standard form, let's say — would appear in relation to a variant form. That's enormously useful information for the connoisseur. But even for a less serious aficionado, those ratios can be extremely interesting.
If you want to know how often, for example, "between you and I" occurs in comparison with "between you and me" in print sources or current books, that information is now available to us, whereas previous lexicographers and usage writers simply had to guess. There's a lot of that empirical evidence spread throughout the book, and in some cases my judgments about terms changed.
There are clearly remaining differences in the attitude towards the prescriptive enterprise, despite a similar reliance on interpreting usage counts:
McMahon: What led you to add so much more empirical evidence? Did you feel challenged on some of your usage preferences?
Garner: Not really. Once the ngrams became available, it took me a little time to start playing with ngrams and realize this is absolutely revolutionary in the field of lexicography. The moment I played with a couple of ngrams, I realized this fundamentally changes the nature of usage lexicography. For a long time, some descriptive linguists have complained that usage books with a prescriptive bent are written by people who just sit back and say, "I like this better than I like that," and I don't think that's ever been so, because the best usage books, even prescriptive ones, have been based on lifetimes of study — when you consider people like H.W. Fowler and Wilson Follet and Theodore Bernstein and others.
And then there's this:
Garner: I find Bernie Sanders's dialect to be very unpleasant to listen to. I could also understand why so many people in New England considered George W. Bush to be unlistenable, because he overdid the Texas twang. And in fact even to a Texan — it made this Texan cringe. But Bernie Sanders is very difficult to listen to because one doesn't expect an educated American to have that kind of accent.
It's not surprising that Bryan Garner has this kind of reaction to regional varieties of English. Shaw wrote in the preface to Pygmalion, "It is impossible for an Englishman to open his mouth without making some other Englishman hate or despise him". And despite the Atlantic ocean and the intervening 100 years, most Americans still have similar visceral responses to the speech patterns of some Others.
What's surprising, at least to me, is that Garner sees this reaction not as a prejudice that he ought to try to overcome, but as a fault for Sanders and Bush to remedy. Of course, Shaw's position was similar: It never seems to have occurred to him to make Cockney or Birmingham accents acceptable, rather than train their speakers to imitate "the noble English of Forbes Robertson".
Compare Geoff Pullum's perspective on these matters, as expressed in his 2004 MLA address "Ideology, Power, and Linguistic Theory". His conclusions about the nature and role of standard languages are not very far, in practical terms, from Garner's views. But the underlying attitudes are clearly still different.
[An important caveat: Bryan Garner's remarks on Sanders and Bush come to us through a journalistic transcription, which (as often noted here) may be inaccurate, incomplete, or otherwise misleading.]
For more from Bryan Garner on empirical methods, see "How to incorporate big data into your day-to-day toolkit", ABA Journal 5/1/2016.
And one last note: If you try to interpret the ratios offered by the Google ngram index — e.g. (corpora)/(corpuses) — note that the y-axis is denominated in percentages, which means that you need to divide by 100 to get interpretable ratios.
Jerry Friedman said,
May 8, 2016 @ 10:00 am
This is not all that new for Garner. In the Preface to the First Edition (as I found it in the 2009 edition), he describes his use of NEXIS and WESTLAW.
I certainly applaud the approach that falls between what Prof. Pullum called the insane extremes, and I hope it has led Garner away from his more laughable examples of hidebound conservatism, none of which I can think of at the moment.
mike said,
May 8, 2016 @ 10:16 am
"If you want to know how often, for example, 'between you and I' occurs in comparison with "between you and me" in print sources or current books, that information is now available to us, whereas previous lexicographers and usage writers simply had to guess."
This makes it sound like he's the first "lexicographer and usage writer" to have thought of using corpus linguistics. I suspect that some folks at, say, Merriam-Webster might find this statement odd.
I guess it depends on what he means by "previous." Previous to him?
Barrie England said,
May 8, 2016 @ 11:18 am
‘Between you and I’ and ‘between you and me’ probably occur more in intimate speech than in formal prose, so any account of their relative frequencies in Ngrams, which draw only on printed material, is likely to be misleading.
Mark Young said,
May 8, 2016 @ 12:11 pm
Google Ngrams reports:
No valid ngrams to plot!
Ngrams not found: " between you and I ", " between you and me ", (" between you and me ") / (" between you and I ")
Barrie England said,
May 8, 2016 @ 12:38 pm
https://books.google.com/ngrams/graph?content=between+you+and+I%2C+between+you+and+me&year_start=1800&year_end=2000&corpus=15&smoothing=3&share=&direct_url=t1%3B%2Cbetween%20you%20and%20I%3B%2Cc0%3B.t1%3B%2Cbetween%20you%20and%20me%3B%2Cc0
Ralph J Hickok said,
May 8, 2016 @ 2:25 pm
"But Bernie Sanders is very difficult to listen to because one doesn't expect an educated American to have that kind of accent."
I went to Harvard with several people who had the same accent as Bernie Sanders. At least two of them graduated summa cum laude.
tangent said,
May 8, 2016 @ 11:55 pm
Surely the empirical question is what fraction of listeners don't expect an educated American to have that kind of accent?
Rubrick said,
May 9, 2016 @ 12:38 am
I personally feel that while Google Books and the Ngrams clearly had talent, and their early hits were quite listenable, their career was ultimately held back by poor management and an unfortunate choice of name.
Yuval said,
May 9, 2016 @ 8:40 am
I wonder how useful this is, a reference book using a very accessible online tool as its own reference. Do you expect there to be many queries that a reference-seeker wouldn't come up with on their own in the common scenarios?
Dick Enzyan said,
May 9, 2016 @ 9:30 am
I am no unquestioning Shaw supporter, but I think you are being, in passing, a bit hard on him.
A quote from a letter from Shaw to The Times March 19, 1949:
The fact that no two people have the same vowels any more than the same fingerprints does not matter provided they understand one another’s speech.
(The Letters of Bernard Shaw to The Times 1898-1950, Ronald Ford, Irish Academic Press, Dublin, 2007)
[(myl) The cited preface by Shaw is pretty explicit:
Finally, and for the encouragement of people troubled with accents that cut them off from all high employment, I may add that the change wrought by Professor Higgins in the flower girl is neither impossible nor uncommon. The modern concierge's daughter who fulfils her ambition by playing the Queen of Spain in Ruy Blas at the Théâtre Français is only one of many thousands of men and women who have sloughed off their native dialects and acquired a new tongue. But the thing has to be done scientifically, or the last state of the aspirant may be worse than the first. An honest and natural slum dialect is more tolerable than the attempt of a phonetically untaught person to imitate the vulgar dialect of the golf club; and I am sorry to say that in spite of the efforts of our Academy of Dramatic Art, there is still too much sham golfing English on our stage, and too little of the noble English of Forbes Robertson.
]
Robert Coren said,
May 9, 2016 @ 10:19 am
@Dick Enzyan: Yes, I was wondering if what Mark was describing as Shaw's attitude is really the attitude of Prof. Higgins, gently mocked by Shaw.
Barrie England said,
May 9, 2016 @ 11:01 am
Higgins was, I believe, modelled on the British phonetician, Daniel Jones.
[(myl) No, the cited preface by Shaw is explicit in citing Henry Sweet as the model:
The reformer England needs today is an energetic phonetic enthusiast: that is why I have made such a one the hero of a popular play. There have been heroes of that kind crying in the wilderness for many years past. When I became interested in the subject towards the end of the eighteen-seventies, Melville Bell was dead; but Alexander J. Ellis was still a living patriarch, with an impressive head always covered by a velvet skull cap, for which he would apologize to public meetings in a very courtly manner. He and Tito Pagliardini, another phonetic veteran, were men whom it was impossible to dislike. Henry Sweet, then a young man, lacked their sweetness of character: he was about as conciliatory to conventional mortals as Ibsen or Samuel Butler. His great ability as a phonetician (he was, I think, the best of them all at his job) would have entitled him to high official recognition, and perhaps enabled him to popularize his subject, but for his Satanic contempt for all academic dignitaries and persons in general who thought more of Greek than of phonetics. […]
Those who knew him will recognize in my third act the allusion to the patent shorthand in which he used to write postcards, and which may be acquired from a four and six-penny manual published by the Clarendon Press. […]
Pygmalion Higgins is not a portrait of Sweet, to whom the adventure of Eliza Doolittle would have been impossible; still, as will be seen, there are touches of Sweet in the play. With Higgins's physique and temperament Sweet might have set the Thames on fire. As it was, he impressed himself professionally on Europe to an extent that made his comparative personal obscurity, and the failure of Oxford to do justice to his eminence, a puzzle to foreign specialists in his subject.
]
BZ said,
May 9, 2016 @ 12:18 pm
Isn't it true that all college educated Americans (in relatively prestigious colleges) would be exposed to very similar accents during the course of their education? Doesn't it also mean that they would at least be capable of speaking in such an accent? Personally, I don't find Sanders's accent that unusual. Bush, well, there is enough baggage there that I can't pretend to be objective, but let me just point out that, in addition to his education, neither his father, mother, wife, nor brother speak with the same accent he does, which does lead to certain conclusions.
Guy said,
May 10, 2016 @ 1:22 pm
@BZ
But people mostly get their accents from their peer groups, mainly before adulthood, not from their families. For example, I'm either partially or completely cot-caught merged even though my mother definitely isn't.
Bloix said,
May 10, 2016 @ 2:38 pm
"But Bernie Sanders is very difficult to listen to because one doesn't expect an educated American to have that kind of accent."
There must be a hundred Nobel Laureates with that accent.
I guess Texans still don't much like New York Jews.
I personally have a strong bias about certain accents. I think people from the Virginia Piedmont sound utterly charming, that people from Minnesota sound honest and well-meaning, and that people from Texas sound like sons of bitches.
Jonathon Owen said,
May 10, 2016 @ 3:10 pm
Agreed. Pam Peters's Cambridge Guide to English Usage relied heavily on corpus data, and it was published twelve years ago, shortly after's Garner's 2nd. And MWDEU might not have been built using a modern corpus, but it relied on Merriam-Webster's huge citation files. Garner's hardly the first to use real data in writing about usage, though he may be the first to use Google Books Ngrams in particular.
And anyway, Erin Brenner, editor of Copyediting newsletter, found that Garner's 4th is not much better than the 3rd. He may be a little more transparent about his methodology now, but if the 4th is anything like the 3rd, then it's safe to say that he freely mixes fact and opinion and ignores facts when they're inconvenient.
chris said,
May 10, 2016 @ 6:17 pm
For a long time, some descriptive linguists have complained that usage books with a prescriptive bent are written by people who just sit back and say, "I like this better than I like that," and I don't think that's ever been so, because the best usage books, even prescriptive ones, have been based on lifetimes of study
So it would be fairer to describe it as "Based on a lifetime of study, I like this better than I like that"?
"Complained" seems like an oddly dismissive verb for the documentation of usage changes over time and contemporary with/antedating the usage authority deprecating them that you can find on nearly every page of, say, MWDEU. Which Garner could have read before he wrote his own usage book, if he had wanted to.
Barrie England said,
May 11, 2016 @ 1:39 am
Thank you for that correction, Mark.
The ‘Longman Grammar of Spoken and Written English’ (1999) is also corpus-based.
Dick Enzyan said,
May 11, 2016 @ 6:30 am
Re. George Bernard Shaw:
The charge is that Shaw regarded regional accents as a fault to be remedied (by the speakers of those accents) rather than as a prejudice to be overcome by speakers of other accents.
In this part of the Preface, Shaw is giving practical advice: what to do if your accent is cutting you off from all high employment. His advice is: change it, but do it properly – “the thing is to be done scientifically”. Certainly, we would offer different advice nowadays, when speakers of regional varieties demand equal treatment: anything else would be seen as craven or defeatist, and unnecessarily so. However, this is practical advice to the troubled (in his terminology) in 1920’s Britain; he is not commenting on the rightness or wrongness of the situation. He does not defend, in this passage, the prejudice involved in language elitism or language snobbery.
(The effective language of the stage, and silly golfing club accents, are another matter.)
I’d say the charge was, if not “Not guilty”, then at least “Not proven”.