parole

A couple of days ago, the Toronto Star completely screwed up its explanation of the IELTS English proficiency test, by presenting as "an example of Part 1 of the writing test" some badly-designed material from a training booklet not even published by the test designers, asking questions of a kind that are apparently never found on the test.

Arnold Zwicky reminds me that the same newspaper did essentially the same thing a little more than a year ago, as Arnold documented in "Do you speak Canadian?", 6/4/2008.

It's shocking how badly some major newspapers sometimes misrepresent basic matters of fact — and how little attempt the editors apparently make to correct these errors and to prevent the same thing from happening all over again a few months later.

There's a weblog associated with Jerry Coyne's book Why Evolution is True. A couple of days ago, Jerry (or whoever writes on the blog under the name "whyevolutionistrue") posted a couple of familiar eggcorns, described as "two solecisms [that] have recently appeared on this site", and invited readers to "Feel free to contribute those mistakes that most irk you, making sure that—for our mutual edification—you give the correct usage as well."

The result, so far, is an outpouring of 251 comments. This is towards the upper end of the distribution for that weblog — the previous half-dozen posts posts are "Gnu atheism" (31 comments), "New York Times to readers: of course you have free will" (174 comments), "Frogmouths!" (14 comments), 'The free will experiment" (94 comments), "Vacation reading from Nature" (31 comments), "Interview with Hitchens" (12 comments), "Space pix" (13 comments) — confirming again that people love to share and discuss their linguistic crotchets and irks.

One of the puzzles of the whole "Plastic Bertrand" drama for Americans is that we don't like plastic. In a famous scene from The Graduate (1967), "plastics" is a one-word symbol for the emptiness of mainstream success:

In Mean Girls (2004), "the Plastics" are "an exclusive group of girls led by queen bee Regina George", who are depicted as shallow, arrogant, and thoughtless.

Allen Ginsberg's Friday the Thirteenth (1984) sums it up:

Slaves of Plastic! Leather-shoe chino-pants prisoners! Haircut junkies! Dacron-sniffers! Striped tie addicts! short hair monkeys on their backs! Whiskey freaks bombed out on 530 billion cigarettes a year— twenty Billion dollar advertising Dealers! lipstick skin-poppers & syndicate Garbage telex-Heads! Star-striped scoundrelesque flag-dopers! Car-smog hookers Fiendish on superhighways! Growth rate trippers hallucinating Everglade real estate!

But in Europe, the connotations of plastic seem to be more positive.

The Plastic People of the Universe were at the center of Prague's intellectual underground from 1968 onwards. When Roger Jouret left the Royal Conservatory of Music in Brussels to launch a pop-music career in 1977, he took the name Plastic Bertrand.

Maybe Europeans are just more ironic. Maybe plastic was associated with America and thereby acquired pop-culture coolness. Maybe European intellectuals are more likely than Americans to think of 20th-century technology as a positive and progressive force.

Anyhow, it's odd.

Dave of Balashon - Hebrew Language Detective (which I welcomed here and have since linked to less often than I should), has done a post—the last in a series on the five grains of the Land of Israel—on the Hebrew word כוסמת kusemet, which now means 'buckwheat' but once meant... well, that's not clear, but I urge you to read his thoughts on the subject. And his final paragraph describes an interesting morphological/semantic split:

As we mentioned, Ben Yehuda made no reference to this usage. And in halachic literature, kusemet continued to refer to spelt. But even heavyweights such as these didn't have control over the living language of Modern Hebrew. And the language seemed to come up with a solution of its own, and a strange on at that. Kusemet continued to be used for buckwheat, but the plural, kusmin כוסמין, was reserved for spelt - and you can actually find the two next to each other in the supermarket, even produced by the same company.

(In the course of his discussion, he links to this old LH post about emmer, spelt, and Italian farro; as usual, the thread wandered into a discussion of hats, snake goddesses, and what have you.)

Here's a case that  I'm hoping will turn out to be an epic example of journalistic misunderstanding.  Because the alternative is that the International English Language Testing System is a really, really bad way to measure English language proficiency.  And that would be a shame, because IELTS, a product of University of Cambridge ESOL Examinations and the British Council, is pretty much the standard English proficiency test outside the U.S.

Commenting on this morning's post "Language tests for immigrants in Canada", IS pointed us to a .pdf of a "sample test" on the web site of The Toronto Star.  It includes four questions — and I bet you can't get more than half of them right without peeking at the answer sheet.  I certainly couldn't. Here's the test, minus the answer sheet:

Let's try to correct these sentences one at a time. The first one is

1. Tomoko spends an equal amount of money on rent, food, study materials and entertainment.

I thought of several ways to "correct" this sentence, which is ambiguous but false under all available interpretations. One alternative among many would be:

1a. Tomoko spends more money on rent and food than she does on study materials and entertainment.

But according to the answer key, the (only?) right answer is:

1b. Tomoko spends an unequal amount of money on rent, food, study materials and entertainment.

I'm not sure that this is even a grammatical English sentence. I could say "Tomoko spends unequal amounts of money on A, B, C and D", but  "… spends an unequal amount of money on A, B, C(,) and D" just seems wrong to me, with or without the extra comma.  And at best, the "corrected" sentence strikes me as awkward and uninformative.

I got the second sentence right. It simply gives the wrong percentage for "study materials", and so the obvious thing to do is to substitute 25 for 15:

2. Tomoko spends 45 per cent of her money on rent and food, but she only spends 15 25 per cent of her money on study materials.

However, some people might prefer to move only so that it's adjacent to its focus: "… but she spends only 25 per cent of her money on study materials". Apparently this would be marked as a wrong answer, even though some prescriptive authorities would mandate the change.

I also got the third sentence right, though it was a near thing:

3. Tomoko spends more less on clothes than she does on study materials.

I was tempted to leave more and invert the order of the categories compared.  And again, apparently that would have been wrong — though in making that correction, I'd demonstrate that I understand the text and the pie chart, and can compose an English sentence to express my understanding.

4. Tomoko spends as much on rent and food as she does on everything else put together.

I got this one wrong. I decided to add "almost" before "as much". But the right answer, apparently, is to change "as much" to "less". I like my edit better, although it would apparently have been marked wrong.

So my score was 50%. How'd *you* do?

Getting back to the reason that I'm hoping for misleading journalism, it would be nice if the cited answers were simply examples of possible correct answers, rather than the only answers accepted as correct. Then the fact that the Toronto Star presented the answers as shown below would be rather misleading, but the questions and answers could still be part of a plausible evaluation of English proficiency.

I'm hoping that someone who knows how the IELTS is graded — maybe even someone who works as an IELTS grader — will tell me that I would have scored 100%, not 50%. But I'd need some favorable odds to be willing to bet money on that outcome.

According to Nicholas Keung, "All immigrants face mandatory language test", The Star, 7/20/2010:

Born and raised in New York, Dodi Robbins graduated from Harvard University and has been practising law for 13 years.

Her first language is English. Yet like all other skilled immigrants applying to settle in Canada, the American corporate lawyer must now take a language test to prove her English is good enough to settle here.

“I was outraged, insulted and floored,” said Robbins, who obtained her law degree at Benjamin N. Cardozo Law School in New York. A mother of two, she has been working in Toronto on a work permit for four years as compliance and regulations counsel for an international financial services company.

“I almost fell off the chair. I’ve been practising law here for years and I have to prove my proficiency in English?”

Last month Ottawa made its language proficiency test mandatory for all skilled immigrant applicants, including native English and French speakers. The so-called “ministerial instructions” stipulate officials are not to process applications without language test results, starting June 26.

There seems to be some substantive controversy over the way the policy was introduced:

Critics say the government is now trying to use the ministerial instructions to circumvent public scrutiny and consultation, ramming through changes without parliamentary oversight.

But Ms. Robbins' case seems to be an odd one to lead with.  It's legitimate for her to be annoyed at having to spend a half a day and $285 taking the IELTS. But the article describes her as sweating the outcome of the test:

Robbins says she is juggling her full-time job and two kids to prepare for the IELTS test in August.

Does a native speaker with a college education really need to "prepare for the IELTS test"?  If so, it must not be a very well-designed instrument.

I recognize that "language exams" can be (and sometimes are) designed to test something other than language proficiency. When I was a graduate student, we needed to demonstrate proficiency in two languages. In principle, all that was required was the ability to translate a linguistics article,  with access to a dictionary. Having achieved roughly that level of competence in German, I was planned to take the German exam. Then one of my fellow grad students, a native speaker of German who had an undergraduate degree from a well-regarded institution in Austria, told me that she had failed the exam.

Apparently the gentleman who administered the German exam had a chip on his shoulder about all the grad students who didn't take the courses his department offered. In any event, he apparently set my friend to translate a particularly fiendish passage from von Humboldt, which she found so impenetrable that she occasionally got confused about who did what to whom.  Or perhaps she suffered the fate that Mark Twain described in The Awful German Language:

You observe how far that verb is from the reader's base of operations; well, in a German newspaper they put their verb away over on the next page; and I have heard that sometimes after stringing along the exciting preliminaries and parentheses for a column or two, they get in a hurry and have to go to press without getting to the verb at all. Of course, then, the reader is left in a very exhausted and ignorant state.

So I abruptly changed course and arranged to take my German exam in Latin.

And then there's the traditional Japanese method of determining English proficiency, which apparently is a version of the cloze test that in effect requires students to commit large numbers of classic works to memory.

But I find it hard to believe that the IELTS is designed in such a way that a highly educated native speaker really needs to study for it. Can someone who's taken it recently comment?

If I understand the situation correctly, this is roughly the Canadian equivalent of a U.S. H-1B visa, for which a language proficiency exam is not required, rather than the Canadian equivalent of the U.S. Naturalization process, which does have a language proficiency requirement, though a rather minimal one:

During your interview, a USCIS officer will test your ability to read, write, and speak English and your knowledge of civics.  You must read one sentence out of three sentences correctly in English, and you must write one sentence out of three sentences correctly in English.  Your ability to speak English is determined during your interview on your naturalization application.  Finally, you must answer 6 out of 10 civics questions correctly to achieve a passing score.

By Ndesanjo Macha

“Thanks to our Facebook page, African music fans from South African and Kenya were able to get the story behind this late Miriam Makeba song,” writes Chale of African music blog, Museke. The song is in Swahili and Xhosa.

According to the Guardian,

The Belgian singer Plastic Bertrand has admitted that the voice that gave the world the 1977 Euro-punk anthem Ça Plane Pour Moi was not his. Roger Jouret, the man behind the Plastic Bertrand persona, had previously denied that he was not the singer on the record. But in an interview with the Belgian newspaper Le Soir, he admitted it had been another singer – and laid the blame at the door of his former producer, Lou Deprijck. His admission came a day after a linguist commissioned by a judge concluded that the singer's accent did not match the voice on the record.

To set the stage, here's a YouTube version of the song:

(A slightly less plasticized performance is here.)

Another report has Plastic Bertrand advancing a hypothesis about who the actual singer was, and gives a bit more detail on the dialect issue:

Plastic Bertrand n'est pas l'interprète de Ca plane pour moi selon plusieurs experts mandatés par la justice belge. Ce jugement n'a pas l'air d'avoir ébranlé Plastic plus que ça, qui a réagi hier soir sur RTL. "Tout ce que le rapport dit, c'est que c'est quelqu'un qui a l'accent ch'ti ou picard qui l'a chanté. Alors c'est Dany Boon", a-t-il déclaré.

Plastic Bertrand is not the singer on Ca plane pour moi according to several experts commissioned by a Belgian court. This judgment doesn't seem to have shaken Plastic up very much, as he responded yesterday evening on RTL, "All that the report says is that it's someone who has a ch'ti or picard accent who sang. So it's Dany Boon", he said.

Some other necessary background: the court case in question seems to be based on a lawsuit started by the record producer Lou Deprijck, apparently aimed at establishing that the voice on the hit song belonged to Deprijck himself. (More background on the song is here, including the amazing fact that it was used as background music in National Lampoon's European Vacation, Ferris Bueller's Day off, and an episode of What's New, Scooby Doo?, as well as other far-flung cultural connections including Extreme Championship Wrestling and Australian Mars Bars commercials.)

According to rtbr.be ("Plastic Bertrand reconnaît que ce n'était pas sa voix", 7/28/2010), Plastic was rather out of the loop on the whole recording process, explaining why he might still be in the position of having to guess who the actual singer was:

"Mais c'est moi la victime. Je voulais chanter, mais il (Lou Deprijck, ndlr) m'interdisait l'accès au studio", affirme-t-il. Et d'ajouter : "Le jour où j'ai quitté RKM (la firme de disques qui avait produit les premiers albums à l'époque, ndlr) pour gagner ma liberté, il a gerbé sur moi".

"But I was the victim. I wanted to sing, but he [Lou Deprijck] barred me from the studio", he asserted. And he added: "The day I quit RKM [the record company] in order to free myself, he vomited (?) on me."

Wikipedia explains that

Picard is a language closely related to French, … spoken in two regions in the far north of France – Nord-Pas-de-Calais and Picardy – and in parts of the Belgian region Wallonia, district of Tournai (Wallonie Picarde) and a piece of district of Mons (toward Tournai and France frontier).

Picard is known by several different names. Residents of Picardie call it picard; but in Nord-Pas-de-Calais its dialects are more commonly known as chti or chtimi, in and around the towns of Valenciennes and Lille as rouchi; or simply as patois by Northerners in general. Linguists group all of these under the name Picard.

Some questions for our Francophone readers:

What is the etymology of ch'ti(mi)?

What aspects of the singer's performance in Ça plane pour moi identify it as Picard?

What is Plastic Bertrand's own variety of French? What features does he have (or lack) that the singer lacks (or has)?

Who are the "plusieurs experts" who analyzed the accent, and is a copy of their report available?

What does "il a gerbé sur moi" actually mean? Do Belgian record producers really vomit on their musicians to express annoyance, or is this just another of the many pieces of French slang that I don't know?

And finally, why would anyone expect that someone who adopted the stage name "Plastic Bertrand" would sing his own songs?

Yesterday here in Prince George I overheard a young woman on her cell phone complaining about the heat: "It's plus 29 here!". [That's 84.2 in Antique American temperature units.] I suspect that this would not be felicitous in, say, Phoenix or Riyadh.

Another grammar of an Australian language is to be launched: Amanda Lissarague's grammar and dictionary of Gathang (see 2008 blog-post on launch of other books published by the ever-productive Muurrbay and Many Rivers Language Centre).

Copy Editing at The New Yorker with Mary Norris. As I said here, "That was interesting, although I rapidly tired of the interviewer’s snarky-twelve-year-old style (apparently mandatory these days). But from her description of the painstaking process of editing and fact-checking, you’d never guess how error-ridden the magazine is these days."

What It's Really Like To Be A Copy Editor, by Lori Fradkin. As I said here:

That was amusing, and I certainly identified with some of her stories, though starting off with the “douche bag” business can only reinforce the standard image of copy editors as humorless pedants who wield dictionaries as bludgeons. I agree with the commenter who said “I enforce Chicago and Webster’s 11th with shock and awe, though I am flexible and respectful of variance and alternatives, as long as they are consistent.” To my mind, a slang term like douchebag is a prime candidate for flexibility, especially at a popular magazine like New York. Me, I would have issued a memo the first time the subject came up, saying “Look, guys, Webster’s says it’s two words; if it’s important to you to spell it as one, I understand and will abide by it, but I want it on record that I provided the dictionary spelling.” And then I would have let it go.

And a response to the previous one, What it's really like to be copy-edited, by R.L.G. As I said here:

Continue reading "SOME LINKS ON COPYEDITING."

By Raphael Tsavkko Garcia

The first monolingual Galician online newspaper Vieiros [gz] has closed down after 15 years because of financial problems. In a melancholy post, Galician reporter McShuíbhne says the loss of language online amounts to a loss of nation.

Janet Randall wrote:

I am faced with a query from someone at a pretty high level at Public TV who is objecting to an employee's use of the preposition "acrosst".  I looked for some dialect information about this variant of "across" but haven't been able to find it on language log, or anywhere else, without spending more time than I have, so I thought you might know of a post on the log, or know who might know whether this is regional, etc.

I replied:

The OED has an entry for "acrost", listing as U.S. dial. and colloq., with citations going back to 1759:

1759 in Essex Inst. Hist. Coll. (1882) XIX. 145 Ye enemy fird at our men a Crost ye River. 1779 W. MCKENDRY Jrnl. 6 Sept. in Proc. Mass. Hist. Soc. (1886) II. 467 The Lake..is..about 8 miles acrost.

But I felt this was not really an adequate answer, so I forwarded Janet's query to Bill Kretzschmar, who responded:

I'm in Europe and so don't have access to my paper files in Athens [GA], where I could address the particular findings on this point.

The "st" ending is known, especially in Ireland, to be applied to "while" yielding "whilst", and some usage mavens don't like it. A similar case that I know about is the choice between "toward" and "towards", as when you "turn toward(s) an event." I wrote about our Atlas evidence on that one for a Bill Safire "On Language" column years ago (it turns out that the Mid-Atlantic speakers who don't like "towards" are the only ones not to prefer the final s!).  In Michigan, people who work for the Ford Motor Company often say that they work for "Fords," which might be interpreted as a strange genitive construction but probably isn't. So, the presence of an extra -s or -st is a commonly occurring variant pronunciation in English. No need to get excited about it. It is not substandard, just a bit different, like many other variations that naturally occur in the language of every educated person.  If somebody with enough authority at MPTV doesn't like it, then she can ask the employee not to do it. Now, if somebody then asked me to find such variations in the speech of the MPTV honcho, I suspect it would be easy to find them….  We've all got them. Might be better not to start down the road of correcting every little thing we hear.

After Bill got back to Georgia, he added this:

Back in Athens last night. I've had a look at the paper records, and can tell you that we recorded the -st form of "across" from about 20% of the people we interviewed in Maryland (11/62). The form is found throughout the state, from Eastern Shore to Baltimore to up north near the panhandle. It's used by both men and women, both middle age and older (we didn't interview too many young people), different job types.  I did not see it among the most educated, but did find it regularly among high-school educated people. It is also pretty common in Pennsylvania, less so in Virginia. Kurath and McDavid produce a map (#179) in their *Pronunciation of English in the Atlantic States* of the places that add a -t to the words "once" and "twice" which shows that the pronunciation is widespread in the Midland and South (except the Virginia piedmont), and "widely used by middle-class speakers" in these areas besides by less-educated people. So, what they say about "oncet" and "twicet" matches what we found for "acrosst".

So, the evidence says that this is a historical pronunciation in Maryland and adjoining states, not particularly marked as uneducated. Our interviews are 20th. c., so not contemporary, but these things usually don't disappear.

Alexa O writes:

My daughter, who is two, loves corn. She loves it so much that she talks about it all the time. Since she is two and also loves eggs, she calls it "eggcorn."

On a whim, my mother wiki-ed "eggcorn" and, lo and behold, we discovered your term.

I can't tell you how excited I was, because I love saying "eggcorn" (was there ever a more satisfying set of syllables?) and now I have an excuse to do so even after my daugther stops asking to eat it.

I was even more excited when I found the various websites, including the Language Log and The Eggcorn Database, that gave so many delicious examples of said phenomenon.

So Alexa wrote an eggcorn-enriched blog post, "The Word of the Week, or: Eggcorn, Mommy, Eggcorn!", 7/28/2010. Which I recommend, even though she calls us

the famous (in dork circles) Language Log, which is an online blog for linguists. (Not for the feint-hearted.  This is technical writing at its most dense.)

Dense? Please. Don't tempt me.

By Rezwan

Aminul Islam Sajib explains why he had to write a letter in English memorized from the book instead of writing creatively during his school examination.

I'm not at all surprised that Mark's posts on regional variation in American English (here) and (here) have stirred up such reader interest, because speech variability seems to be one of the first things people notice, even if they can't pinpoint exactly what it is. It's not as well understood that there is a long tradition of studying variation in the languages of the world, even in the United States. But there was a time when the study of linguistic geography was an important part of most linguistics departments. In the 1950s and 1960s you could study with nationally prominent linguists at universities in Ann Arbor, Chicago, New York, Washington, Providence, Berkeley, Cleveland, Madison, Seattle, Austin, and other places. The BIG names in linguistics back then included dialectologists such as Hans Kurath, Raven McDavid, Fred Cassidy, Albert Marckwardt, Harold Allen, Carroll Reed, E. Bagby Atwood, W. Nelson Francis, Uriel Weinreich, David Reed, James Sledd, and others. Their papers about regional dialects were prominent features at annual meetings of the Linguistic Society of America.

The impact of these giants (and those in Europe as well) was strongly felt by the entire field of linguistics, which in those days was primarily descriptive. In fact, language variability was what attracted me to linguistics in the first place. I had begun my PhD studies in English literature until I took a grad seminar on Chaucer, during which I got excited (not "interested," but "excited") about the dialects of that period. I am deeply indebted to that professor, Morton Bloomfield, for encouraging me to change the focus of my grad studies to study with McDavid, who also taught me the joys of doing fieldwork, first in Illinois and later in Michigan and Washington DC. In the late 1950s he started me off by doing Linguistic Atlas interviews in all the rural counties of Illinois, an experience that changed my life in many ways.

Thanks primarily to the tremendous and enduring influence of Bill Labov in the 1960s, the older focus on regional dialectology expanded to language variability that went far beyond the older, rural, relatively uneducated white male subjects to the exciting language variation of people of all ages, races, localities, education level, social status, and gender. Labov opened the door to the linguistic reality that individual pronunciations and grammatical features are used variably even by the same speaker, explaining for example, why I, a native North Midland dialect speaker, sometimes use the /ih/ vowel before nasal consonants and sometimes the /eh/ vowel. Today, the field is now called sociolinguistics. The same rigorous descriptive work continues today, but much more broadly than before and with much more useful applications to social issues in the real world.

The multitude of comments to Mark's posts on "pin/pen" variation are illustrative of the deep interest people have on language variability of all types, as well as language change in progress.  We seem to want to talk like each other in order to be understood, but different enough from each other to maintain our individual and social identities.

Although this sign over a children's clothing shop in Hangzhou is fairly simple, it offers much food for thought.

On the left, which we will naturally read first because of the directionality of the writing, we see "les enphants," a clever blend of "les éléphants" ("elephants") and "les enfants" ("children"). The notion of "elephant," of course, makes for a very cute logo, which is situated in the middle of the sign. To the right, we see Lì yīng fáng 麗嬰房 ("Beautiful Baby Shop").

It is difficult to say for certain whether "les enfants / enphants" inspired "Lì yīng fáng" or vice-versa, but I suspect that the proprietors started with "les enfants" and came up with "Lì yīng fáng" to match it, then playfully embellished "enfants" by substituting -ph- for -f-. My reasoning for making this surmise is that "les enfants" is a fixed expression in French, whereas "Lì yīng fáng" is not a fixed expression in Chinese. If you run "Lì yīng" through Babel Fish, it offers only "Li infant," not knowing what to do with the first syllable. And if you run it through Google Translate, it comes up with "Korea baby," not "beautiful baby." This is not so dumb as it may seem, since GT is thinking of the Lì as short for Gāolí 高麗 ("Korea"). Furthermore, although fáng (usually meaning "house") can be used to convey the idea of a shop or store, there are at least a dozen other terms that are more likely to be used before it. For all of these reasons, it seems to me that the shopkeepers started out with "les enfants" and thought up "Lì yīng fáng," both to match the sound of the French expression and to convey an appropirate, felicitous meaning in Chinese.

When I first began this post, I thought that I would propose christening the French equivalent of Chinglish as "Chinçais," unless there were already an established term for spoken or written French that is similarly influenced by Chinese. Strictly speaking, however, the relationship between "les enfants" and "Lì yīng fáng" would appear to be one of French influencing Chinese, rather than the other way around. Might we call it "Zhongçais" or "Franwen"?

[Thanks to Ian Mair for another great photo from Hangzhou.]

This year’s AAA meetings have the highest number of registrants on record. As one of the student assistants on the Executive Program Committee, the level of interest was both reassuring and daunting. Last month, I traveled to Washington DC to the AAA offices to help with the enormous task of scheduling the academic program. It was great to see the “backstage” spaces of the organization and to understand the labour process of putting together the meetings.

Staff members, Jason Watkins and Carla Fernandez, had already been hard at work preparing for Dr. Monica Heller, Dr. Rob Albro and myself to arrive. In a complex web of flip charts, coloured labels and index cards we sorted and slotted an enormous amount of academic knowledge. I can’t say much more, but I can say SLA members will be intrigued to see the ways in which the conference theme Circulation has been taken up. There are also some other exciting program elements which I will talk about in upcoming posts. The full program will be online very soon!

Among the comments on yesterday's pin-pen post, Eric (one of several) asked:

Hey academic linguists, I have a nerdy question. I assume that in phonetics "field research" or whatever, lots of scenarios have several investigators listen to a speaker, make independent IPA transcriptions, and then check their transcriptions against each other. And then when the various transcriptions show some level of convergence, that's taken to be the correct phonetic description of the speech. But are there ever scenarios where the results of the investigator's transcription is checked, not against the transcriptions of other listeners/investigators, but against the speaker's own belief about her pronunciation? As someone who merges like 90% of the pairs mentioned in this thread, I'm interested in pushing a radically skeptical line: that speakers are often subjectively convinced they make a phonetic distinction (like Mary v. marry) which objective investigation would dis-confirm…

Actually, Eric, your skepticism about the relation between how people speak and how they think they speak is not nearly radical enough. And there are actually three things to consider: not only how people speak and how they think they speak — which may be bizarrely different –  but also how they hear.

However, I'm not going to discuss all three of these topics. A nerdy question deserves a nerdy answer, and so I need to start by pointing out that your picture of phonetic investigation is incomplete. And it's going to take me long enough to sketch an answer to the first question — how to characterize how someone speaks — that I'll leave the other two parts for later posts.

It's not enough to "listen to a speaker", because the way you talk depends on the role you're playing, your audience, and your psycho-physiological state. Are you giving a speech? Reading out loud to a child? Reading a list of words in an acoustically-isolated laboratory chamber? Telling a joke to a friend? Arguing with a family member?  Giving directions to a foreigner? Are you happy and exuberant, tired and depressed, or somewhere in between?

Linguists use various techniques to get recordings of different sorts of speech — and it's common to compare speech from the same speaker in different settings.  Here's a graph (from Labov, The study of nonstandard speech, 1969) showing the percentage of "g-dropping" in three speech styles from members of four socio-economic classes in New York City:

(For more discussion, see here.)

"G-dropping" is a categorical choice — whether to use a coronal nasal [n] or a velar nasal [ŋ] in the gerund-participle ending -ing.  As common sense tells you, and as the plot above suggests, speakers are variable. They don't always use [n] or always use [ŋ]. Instead, they mix them up in proportions that depend on lots of things, including degree of formality.

This variability takes on a new aspect when we look at a gradient linguistic choice, such as where to place a particular vowel, on a particular occasion, in a continuous articulatory and acoustic space.

For the study of phonetic variation in vowel pronunciation, IPA transcription is not very helpful, partly because it's a subjective description with imperfect inter-transcriber agreement, but mostly because it forces us to assign tokens to one of a small number of distinct categories. Instead, the standard approach is to measure the "formants", or resonance frequencies of the vocal tract, which are a useful quantitative proxy for vowel quality. (Sociolinguists generally characterize an individual vowel in terms of a F1 and F2 at a single point deemed characteristic of its quality. This abstracts away from the vowel's duration and time-varying properties, and from F3 and other spectral characteristics, in a way that is sometimes problematic — but introducing these additional complexities wouldn't change the discussion below in significant ways.)

For an excellent example of such an analysis, see chapter 6 (pp. 132-194) in Keelan Evanini, "The permeability of dialect boundaries: a case study of the region surrounding Erie, Pennsylvania", Publicly accessible Penn Dissertations, Paper 86 (2009).  This chapter is all about the merger of the vowel categories in cot and caught. Here's Keelan's explanation of the background:

The short-o vowel is represented here by the symbol /o/, following the notation in the ANAE, and it corresponds to the LOT vowel class in Wells (1982). It is descended primarily from short o in Middle English, and occurs in nearly all segmental environments. Some examples of words with /o/ include lock, pot, god, and stop.

In most dialects of North American English, /o/ has been unrounded and lowered to [ɑ]. In many of these dialects, /o/ has moved towards the front, and is unrounded. In these dialects, the best phonetic representation would be [a]. This is especially the case in the North where the fronting of /o/ as the second stage of the Northern Cities Shift has caused /o/ to move close to the position formerly occupied by /æ/. In other dialects, /o/ has maintained its roundedness, merging with /oh/ in the low back position. This is the case for the Western Pennsylvania dialect centered around Pittsburgh.

The symbol /oh/ is used to represent the long open-o class, and corresponds to Wells’ THOUGHT lexical set. It is derived primarily from the monophthongization of the Middle English diphthong au, which itself was derived from a variety of sources (such as Old English /aw/, OE /a/ + /x/, as in fought, vocalization of OE coda /g/, as in draw, and Middle French loan words, as in applaud). Another large source for /oh/ words was the lengthening of /o/ to /oh/ before voiceless fricatives, as in lost, and the velar nasal, as in strong. The distribution of /oh/ is severely restricted, and it occurs before only a small number of consonants, mainly before /t/, /d/, /k/, /z/, /n/, /l/, and word-finally. Some examples of words with /oh/ include thought, hawk, caught, and law.

In dialects of North American English where /o/ and /oh/ have not merged, /oh/ has changed in three different directions: 1) In the Mid-Atlantic region and New York City it has raised substantially and developed a central offglide, 2) In many areas of the South, it has developed a back upglide, and 3) In the North, it has lowered and fronted as Stage 3 of the Northern Cities Shift. In dialects where /o/ and /oh/ have merged, /oh/ can become unrounded and rather front, especially in the West.

Here's an F1/F2 scatterplot of 56 /o/ vowels and 24 /oh/ vowels from the speech of someone who maintains a robust distinction:

Figure 6.1: /o/ and /oh/ from Walter K., born 1927 in Buffalo, Mean(/o/) = (841, 1451), N=56; Mean(/oh/) = (684, 1044), N=24; Dist(/o/, /oh/) = 436

In such plots, the origin is by convention in the upper right-hand corner, which makes the dimensions of vowel quality run the same way that they do in the IPA vowel chart:  front-to-back on the horizontal axis, and low-to-high on the vertical axis. (These scatterplots only show the small area of the vowel space occupied by the collection of vowels being displayed.)

As the plot indicates, Walter K.'s /oh/-vowels are substantially backer and a bit higher than his /o/-vowels, though there's an indication of occasional overlap, perhaps especially for certain words.

Here, in contrast, is a similar plot for someone who pretty thoroughly merges the categories:

Figure 6.4: /o/ and /oh/ from Dan R., born 1912 in Erie, Mean(/o/) = (704, 1338), N=55; Mean(/oh/) = (707,1283), N=31; Dist(/o/, /oh/) = 55

Keelan's comment:

His means for /o/ and /oh/ are only separated by 10 Hz in the F1 dimension and 78 Hz in F2. The two vowel clouds show considerable overlap throughout their entire ranges. To complement this acoustic evidence, the minimal pair data from Dan R. also point to a complete merger. He produced the pairs cot / caught and Don / dawn identically and judged them both to be the same.

There's a hint in the scatterplot that Dan R. might still have some residual tendency towards a fuzzy distinction — and there are plenty of "transitional" speakers whose distributions are more clearly distinct, though still heavily overlapped. Here's an example:

Figure 6.7: /o/ and /oh/ from H. O. Hirt, born 1887 in Erie, Mean(/o/) = (745, 1311), N=36; Mean(/oh/) = (664, 1074), N=21; Dist(/o/, /oh/) = 250

So to sum up what we've got so far:

1) Speakers are variable. In the case of categorical choices, individual speakers rarely behave in a consistent way, taking a given alternative 0% of the time or 100% of the time. More often, their behavior is somewhere in the middle, and is modulated by circumstances in a complex way. And in the case of gradient choices like vowel quality, an individual speaker must be characterized as a "cloud" of possible outputs, a multi-dimensional probability distribution that again is modulated by circumstances in a complex way.

2) The "behavior cloud" corresponding to a particular speaker's propensity to pronounce a particular vowel category often overlaps with the same speaker's cloud for a nearby vowel category. This remains true even if we keep the circumstances as constant as we can.

As a result, the question of a whether an individual speaker is "merged" or "unmerged", with respect to a particular pair of vowel categories, may not have a clear, categorical answer. Their productions may overlap to a degree while remaining to some extent distinct; and the degree of cloud overlap will typically vary with style, speaking rate, vocal effort, formality or precision of articulation, and so on.

Next time: how well do speakers know themselves?

If you haven't quite yet gotten your fill after last week's refudiate-fest, I return to the Palinism in my latest Word Routes column on the Visual Thesaurus. An excerpt of interest to all you antedaters:

Some have observed that Palin isn't the first to invent the word refudiate. Patrick Galvin of Politico noted a couple of recent uses, such as Sen. Mike DeWine's statement on "Fox & Friends" in 2006: "I think anyone that is associated with him campaigning needs to refudiate these comments." And on Language Log, Mark Liberman points to a playful usage in John Sladek's 1984 collection of science-fiction short stories, The Lunatics of Terra.

Even earlier is this glaring example that I found in the Atlanta Constitution of June 21, 1925: a headline reading "Scandal Taint Refudiated In Teapot Case by Court, Fall Says in Statement."

The headline refers to a court ruling in the Teapot Dome scandal that rejected accusations of fraud against former Secretary of the Interior Albert B. Fall and his cronies in the oil business. A Fall press release interpreted the verdict as "refuting all taint of scandal," and the hurried headline writer must have mashed up refute with repudiate, just as Palin would 85 years later.

Read the rest here.