Johnson

IT MUST have been excruciating for the National Museum of American History's archivists to have the earliest known recordings of the human voice but not to be able to listen to them. The records, made in the Volta Lab of Alexander Graham Bell in the early 1880s, were too fragile to play. But the Lawrence Berkeley National Laboratory figured out how to scan them optically and retrieve the sound, as described on the museum's website here. Six recordings have been released on YouTube. The most familiar text is this one:

Bell's lab was in competition with that of Thomas Edison, who invented the phonograph that became the standard way of recording sound. The first phonograph recordings available come from 1888, so this amazing work has pushed that boundary back a few years to the early 1880s. 

Of linguistic note is the accent. "Suffer" has the same lack of rhotic "r" after a vowel that one finds in standard British English today, but not in most American accents. The second is that r's in other positions are noticeably trilled, characteristic of old-timey, hyperarticulated stage voices. The vowels, too, have a British flavour.

The missing piece that we will never have, unfortunately, is truly spontaneous speech from the era. Linguists know that people talk differently when a recording device is prominent in front of them, and surely even more so when recorded sound was as rare and fantastic as it was in the 1880s. In fact, Candid Microphone, a comedy radio show from the 1940s, is one of the oldest large collections of people being recorded without knowing it. An example can be heard here.  

But the Volta recordings should nonetheless prove fascinating. The researchers think that they will isolate, over time, the voice of the great Bell himself.

(Via Language Hat.)

A sad, sad moment: the scrapping of the Ofsted judgment of “satisfactory”. For readers who live outside Britain, Ofsted is the schools inspectorate, and “satisfactory” has for years been one of the categories in which it puts schools, at least until now. As I noted a couple of years ago in an article for the print edition about education jargon, or “eduspeak”, in this context “satisfactory” means precisely “unsatisfactory”. It was hearing the then head of Ofsted give an interview in which she said that far too many of England’s schools were satisfactory—a statement that presumably provoked bewilderment in many visitors to the country—that inspired my article. 

I’m sad to see “satisfactory” go, but at least it went out with a bang: the prime minister, David Cameron, responded to the news with the priceless: “I don't want the word ‘satisfactory’ to exist in our education system.”

Any other examples spring to readers’ minds?

HERE we go again. The idea that language regulates thought is as old as the hills. (See The Economist's debate on the subject here.) But while experts debate the scientific particulars, laymen sometimes put it to such lazy and odious use that Johnson cannot help but respond. In this week’s issue of Mint Lounge—the weekend edition of an Indian business newspaper run in association with the Wall Street Journal—Aakar Patel, a columnist, argues in favour of restricting free speech in India. 

Mr Patel makes his case using the single most shameful event in 21st century Indian history: the 2002 riots in the state of Gujarat, during which hundreds of Muslims were slaughtered by Hindus as authorities stood aside and watched. The riots started when a train carriage filled with Hindu pilgrims was set ablaze at a station in Gujarat as they returned from a holy site in north India. 58 men, women and children were killed. Hindu mobs retaliated against the local Muslims, spurred on, says Mr Patel, by local newspapers. So free speech must be regulated because Hindi- and Gujarati-speaking Indians will believe anything they read in the papers. 

In a series of baseless assertions, Mr Patel claims that the reason all Muslims were blamed for the massacre is that Indians have a collective identity and behaviour. Why? How? He doesn’t say. But he adds:

It is difficult to explain to Indians the wrongness of collective punishment. This is because our identity is collective, and so is our behaviour. The understanding that this is wrong comes mainly to those who speak English. Individuals are more easily produced by English because it opens access to the world outside the tribe. It is able to place us outside the narrow definitions assigned to us by Gujarati and Hindi. [Emphasis mine]

Where to begin? The absence of a word from language A in language B is often used to prove that the latter lacks the concept. This is easily disproved: Schadenfreude means satisfaction at another’s misfortune. That no single word for this exists in English doesn’t mean English-speaking people are incapable of feeling the emotion. They just take a number of words to express it. That is a flimsy enough argument at the best of times. But Hindi does not lack the word for individual (व्यक्ति, vyakti; the Gujarati word is similar).

Maybe Mr Patel means individual rights cannot be expressed in Hindi. In that case perhaps he should read the Hindi version of the Universal Declaration of Human Rights. Article 2 would tell him that "Everyone is entitled to all the rights and freedoms set forth in this Declaration, without distinction of any kind, such as race, colour, sex, language, religion, ... or other status." And Article 3 would have told the murderous rioters in Gujarat, in their choice of Hindi or Gujarati, that “Everyone has the right to life, liberty and security of person.”

But I dwell on technicalities. Mr Patel must surely mean that it is an education in English—and therefore with western, enlightenment values—that produces smart, rational beings. “The damage is done by a Hindi-medium world view,” he writes, whereas English “opens access to the world outside the tribe”. Without getting into the wisdom of calling the 240m-odd speakers of Hindi (and mutually comprehensible Urdu) or 46m speakers of Gujarati “tribes”, perhaps Mr Patel is getting his causality the wrong way around.

The Indians who speak English already have access to the outside world. The 2001 census found that a mere 230,000 Indians spoke it as their native tongue. These, presumably, are the people that speak the language well enough to be influenced by the underlying culture that Mr Patel argues in favour of. But to speak English as a native language means going to an expensive private school in a city. The conditions for that ensure that these English-speakers come from wealthy or least upper middle class families. Having money means having security. That, in turn, means you don’t blame your poor neighbours for everything that is wrong with your life.

In her remarkable new book about life in a Mumbai slum, Katherine Boo, a New Yorker staffer, concludes that the world’s unequal societies don’t explode into violent insurrection because poor people pick on other poor people, not the rich. Just as the wealth flowing into India has yet to trickle down to its very poorest (though it has already lifted hundreds of millions out of poverty), the troubles of the poor leave the better off unaffected. Hundreds of Muslims died in riots in 2002 because poor Hindus expressed their frustrations at day-to-day life against equally poor Muslims. In 1992, when similar riots spread across India, it was the poorest towns and neighbourhoods that became war zones. The nice bits of town remained relatively peaceful.

The barbaric behaviour seen in Indian riots has everything to do with poverty and despair and nothing to do with language. Columns such as those by Mr Patel in business newspapers for English-speaking city people only blame the poor for not lifting themselves out of poverty, in this case by schooling their children in their own language.

GETTING the gist of a news story across in six or so words is difficult, which is one reason editing at a newspaper is a thankless task. (That, and cleaning up the prose, the facts and the argument of a writer who then gets 100% of the byline.) Headsup is one blog largely devoted to this difficult art of headline-writing. And the folks at Language Log have written often about "crash blossoms", headlines that are confusing to analyse because non-content words like articles, prepositions and so forth are left out, leaving a headline that can be read several ways. Many crash blossoms come from Britain, where editors often assume the reader is familiar with the characters and plot of an ongoing story, and so pile up long noun-noun compounds that tell the whole chapter, like "Sex quiz cricket ace in hotel suicide leap".

Having acknowledged that headline writing is hard, I was surprised to find not one or two but three significantly ambiguous headlines in the Wall Street Journal I read on my phone on my morning commute today. Here goes:

Unilever Sees Small Profit Rise: Was it a small profit that rose? Or was it a small rise in an otherwise big profit? The first words of the story confirm the second reading, which I figured was the more likely. I probably wouldn't have thought much more about this, but then I read the next one. 

PetroChina Buys Stake in Shell Canada Asset: To some, this might seem easy, to others not so much. If you follow energy, you will probably have guessed that PetroChina, an energy company, bought an energy asset (a shale-gas prospect) from Shell Canada. But if you follow financial markets, you know that Chinese companies have often bought shells of listed American companies and then merged with them, a controversial "reverse merger" that puts the Chinese company on American markets without pre-IPO vetting. Perhaps since American regulators were cracking down, Chinese companies were buying Canadian shell companies (assets) instead. That was wrong—again, thinking about it for a few extra seconds I guessed correctly. The web headline had the extra word "shale", which would have cleared this up, but my phone did not.  Again, I might have shrugged, but then I read the third.

Swiss Watch Exports Hit Record High: This was a legitimate 50-50 toss-up. A headline reading "Belgians watch exports hit record high" would have been unambiguous: the Belgians have seen their country's exports hit an all-time record. But it just so happens the Swiss are known for their watchmaking. In boom times for the very rich, have sales of Patek Philippes and Rolexes taken back off?  Take a guess, click through, and see if you were right.

Many crash blossoms are funny because they invite logically possible but otherwise hilarious readings ("Squad helps dog bite victim"). Others are funny because they are nearly impenetrable ("Bright sparks weather gala night power cut to party on"). A third class of crash blossoms is the possible but neither likely nor funny: just another from the Journal today is "New York Times Profit Falls". Did the city of New York time the plummeting of certain kinds of profits? No, since New York doesn't have the power, nor the incentive, to do this. A newspaper, the New York Times, has seen its profits fall. Most people wouldn't be hung up on this one for long on the wrong reading. 

But the fourth category, that in question today, might be the rarest of all: a headline, overseen by professional editors, that has two totally plausible and very different real-world meanings, both of which might make stories in the Wall Street Journal. If it hadn't happened three times in a day, I'd have remembered just how hard editing is and given the Journal's folks a pass, but this might warrant a bit of their extra attention. 

ONE of the top stories on the New York Times website last week was that of Alejandrina Cabrera, a Latina citizen of San Luis, Arizona. Ms Cabrera was running for city council, but was stopped by a judge from appearing on the ballot because she doesn't speak fluent enough English. Arizona law requires officeholders to do so. A linguist from Brigham Young University, hired as an expert the court, found that she had "basic survival level" English, and not more. 

The Times seems to back up the linguist's assessment. She was asked, on the witness stand at her hearing, where she went to high school. She was unable to answer. Her later explanation to the Times, did not help matters: "My brain, my mind was white. That was my first time in court." By "white", though, she meant "blank". Both words are blanco in Spanish. And a city council is a formal public setting; if she was terrified by a courtroom into speechlessness, it is hard to imagine success on the city council.

Another of her explanations also didn't help her case. She said that the linguist who examined her, an Australian, pronounced "summer" as "summa", and so she thought he meant Somerton, a nearby town. If her grip on the context was so loose that a common foreign accent made her confuse a season with a town, this is on her own shoulders.

Finally, Ms Cabrera later told CNN en Español (video in Spanish) that she had subsequently had her hearing tested; her doctor found (on a 1-5 scale) that one ear rated only a 2, the other a 3. But she seemed to have no trouble understanding the CNN interviewers over a fuzzy Skype connection. 

Commenters from the left see skullduggery in the decision to keep Ms Cabrera off the ballot. John McIntyre of the Baltimore Sun asks

I wonder how a judge would have ruled had he been called upon to decide the qualifications of the elder Richard Daley to be mayor of Chicago on the basis of his mastery of English. 

Well, you can see Daley here; my guess is that Arizona's Judge Nelson would have let him run for city council. What's more, I don't think Arizona's law is unreasonable. It sounds like Ms Cabrera speaks only basic English, not to a level where voters could trust she could handle her duties in it. I haven't been able to find a video of her speaking English, but I have no reason to distrust the court's expert. 

Does this mean, as many conservatives fret, that Spanish is taking over regions of the country, even official domains like city councils?  Hardly.  This is one person. She does speak English, merely not well enough, probably because she spent much of her childhood in Mexico. And she was not allowed to run for office in the end. Most importantly, this story from a small border town was rare enough that it made the Times, the Wall Street Journal and CNN. In other words, much more than this will be needed to prove that Spanish poses any real threat to English in America.

ANOTHER question from readers: Why does the "theta" sound in Spanish Spanish not prevail in Latin America?  For those unfamiliar, the letters c (when followed by e or i) and z (always) are pronounced like the English "th" as in "thin", in Spain. (It's often called theta after the Greek letter θしーた, which is pronounced the same way.) My favorite example is civilización, which comes out thivilithathion in Spanish Spanish but not in Latin America. Nate in Phoenix, Arizona asked if this was because, as he had heard, Spanish settlers in Latin America came from the south of Spain (Andalucia), where the theta is not used? 

I didn't know the answer, so wrote to John McWhorter, an expert in language contact and change.  His reply:

While theta happened in Castille, some of Andalusia developed into just /s/ while other places there, and elsewhere, had an opposition between /s/ and an apicoalveolar /s'/ between /s/ and theta. Both of the latter are still around in Latin America, as opposed to just /s/, and thus the idea that Latin America has /s/ from Andalusia fails from both sides of the pond. But it's true that there's no theta in Latin America because Castillians didn't come over much.

So Nate's conjecture was almost but not quite right.

SEVERAL commenters have detected what they decry as a drift towards prescriptivism here at Johnson.  Well, guilty as charged: we all think some ways of writing and speaking are better than others, and a language blog that never criticised anything would be unusual. We've also been relentlessly descriptivist, bringing usage facts to bear on peeves we feel are unjustifiable. Johnson, like The Economist, has time for both facts and opinions.

But back to complaining.  One commenter asked, in our "What should we write about?" post, about the word "issue", used nowadays often for "problem". We don't like the drift. 

"Issue" seems to have migrated in an unusual direction, from official euphemism right down to the kids.  Beginning with "Our servers are currently experiencing outage issues," it's now unremarkable to hear a teenaged girl saying "I have an issue with you."

And "issue"-as-"problem" seems here to stay. The first results for "I have an issue with" in Google include things like "a barking dog", "a website that is potentially posting copyrighted material", "activating MS office 2010" and the like. Many are technical. A problem is a type of issue, of course: in politics we talk about the Iranian nuclear issue, the unemployment issue, the Israel-Palestinian issue. But an issue did not always need to be negative. The economy is an issue in boom times, too, even if a less salient one. 

But our style book, in a ruling this Johnson endorses, wishes that people with "issues" might just come out with it:

"The Economist has issues–51 a year–but if you think you have issues with The Economist, you probably mean you have complaints, irritations or delivery problems. If you disagree with The Economist, you may take issue with it. Be precise."

If your issue is a problem, and you want to bring it up with someone, it might be a good idea to go ahead and use the P-word. This doesn't have to be a cause for unpleasantness, but it does communicate your feelings and expectations more pointedly than "issue". Get it out there.

(But if you do have problems with your delivery, comments here are not the best way to get them addressed.)

BACK in Britain for a few days, and often the capital city doesn't feel very British, from the Russians at the hotel to the Italians at Pret to the American nonsense on television. It's good when you still find a nice old British tradition, then: a warm and flavourful ale, a black cab with a brilliant driver, or a newspaper article shot through with unconscious language prejudice. 

Today's article in the Independent is about the rise in people taking elocution lessons. Mind, this probably wouldn't make the business section, as the "rise" is documented by one tutor whose business is booming. But if a true trend, it would be interesting to hear that more people are taking classes to learn to speak differently. (The boom is said to be fuelled by anxious job-seekers in a weak economy.) My complaint is the constant refrain that people are aiming to "lose their accents": "Since Annette Burgess began her elocution lessons last autumn, she feels she has made huge strides and has ambitions to discard her accent completely," and so on, several times.

Not that the journalist or his sources are Henry Higgins-style snobs:

In what we like to think of as an increasingly classless society, and at a time when the distinctive regional accents are gradually being melded and lost, it seems a shame that there are so many people anxious to lose their accents. "I get a lot of requests from people looking to reduce their regional accents, Midwinter says. "I think as long as people speak clearly, if they have an accent, that's OK, as long as they can be understood."

But there it is, "lose their accents" again. The mental frame is that the speech of Devon is an "accent", an encumbrance to be lost, while someone who uses the Received Pronunciation native to southeastern England has "no accent". Only the mute have no accent. When I fly home, the first thing I'll notice at the airport is that General American is an accent, too, unless I get an earful of New York English.

The saddest thing is to see people turn the prejudice on themselves: 

"I also wanted lessons to help me do away with my Devonian accent. I originally come from Plymouth, so there is a Devon twang to some of my speech that I would like to lose. I felt it was holding me back in terms of forward progression within my career."

Don't let the snobs get you down!  By all means, anyone who wishes should to take lessons to enunciate clearly, project, slow down, choose words effectively and conquer the common fear of public speaking. But a clever argument charmingly delivered—as I can attest from talking to colleagues who don't speak in RP—will always carry the day.

I understand journalistic shorthand. But this article reads very differently—correctly, not just compassionately—if the first mention of "losing an accent" were replaced with "learning the Received Pronunciation of southeastern England", and every subsequent mention with "learning RP". Then the picture would become clear: some people feel they must learn a pronunciation not native to them in order to meet others' prejudices. That's quite a different story, one that would be better at home in the left-leaning Independent.

READERS have, in the past, been responsible for some of our most interesting posts by simply asking a question. k.a.gardner's recent question about the comma-splice led to a post. So given that today is a travel day and there's no time to write here (not to mention yesterday's American holiday), help a Johnson out. What would you like to see us answer? 

ONE could just put this picture in the what-is-the-world-coming-to file and leave it at that. But the more I think about it, the more I wonder whether it was a mere infelicity on the part of an ill-educated copywriter for New York's Metropolitan Transity Authority, or a calculated act.

As Mark Liberman and Geoff Pullum explain, it's far from true that "fewer" must always be used with plural nouns; "less" is correct when the noun refers to something divisible, as in "seven dollars less", and may be allowed even with something indivisible when it's being considered as part of a "mass-like quantity" (for instance, "a margin of 5,000 votes or less"). So could hitches be considered a mass-like quantity? Well, maybe, if you think of them as contributing to some aggregate quotient of unhappiness in the life of the New York commuter. But I somehow doubt the copywriter looked up Messrs Liberman and Pullum before brushing the cake crumbs off his or her keyboard and getting down to work.

So that leaves two more explanations, besides sheer ignorance. The first is that the writer felt that "New switches, fewer hitches" just didn't scan as well. Personally I think it scans just fine. "Newer switches, fewer hitches" would be perfect, of course, but I suspect the MTA wouldn't want its switches to be called merely "newer", as if to imply that they bought a second-hand job lot off the London Underground and had them shipped across the Atlantic, to replace the decidedly long-in-the-tooth ones bequeathed as a gift from the Paris Metro back in the 1970s.

And the last option is just that "less" is the way people talk, and the MTA doesn't want to appear snooty and out-of-touch. Which would be a shame, because a message that stares you in the face on your way to and from work every day is bound to make an impression, and maybe it's paternalistic of me, but I think that public authorities (unlike, say, entertainers) ought to take an educational stance, not a populist one.

But I'm curious what you think. Was this a deliberate error, or an accidental one? And if deliberate, was it justified?

SEVERAL months ago I was surprised to see Arnold Zwicky, a linguist, use a comma splice. A few commenters took me to task for being over-picky. The question came up again in the comments several days ago, when k.a.gardner, a frequent commenter, asked for a post on the comma splice. One of my colleagues quickly replied that "The comma-splice rule is totally arbitrary," and a back-and-forth ensued.

What is a comma splice?  Prof Zwicky wrote back in July

"this is not even a tempest in a teapot, it’s a fuss in a thimbleful of spit."

That's two independent clauses joined only by a comma, or a comma splice, sometimes called a "comma fault".  Ashbird, another commenter, was taught as I was: the comma-splice is an error.  My senior English teacher marked down any paper with even a single comma splice by two letter grades, so that an otherwise perfect A paper would receive a C. (She applied the same rule for fragments and run-on sentences.)  My colleague, however, says it's a matter of style.  The Economist doesn't have a ruling on comma splices in the style book, but I don't recall ever having seen one in the newspaper. And I would notice; my English teacher's injunction gave me a terror-loathing of comma splices that has never left me.

But, as ever, there are facts to be had, and in cases like this, the go-to reference is the Merriam-Webster Dictionary of English Usage. (It is hard to praise this book enough.) Sure enough, the comma splice was once part of the best English usage:

As to the old one, I knew not what to do with him, he was so fierce I durst not go into the pit to him — (Daniel Defoe, Robinson Crusoe, 1719)

The New Jersey job was obtained, I contrived a copperplate press for it — (Benjamin Franklin, Autobiography, 1771)

By the nineteenth century, MWDEU gives examples by Lord Byron, Jane Austin and Lewis Carroll, but only in personal letters. This trend continued into the 20th century, where Ronald Reagan and E.B. White both used comma splices in letters. (Yes, that is the E.B. White who put his name, in "The Elements of Style", to this crisp injunction: "Do not join independent clauses with a comma," though "Elements" allows for rare exceptions if the sentences are very short and closely related.)

By the 20th century, rulebooks commonly warned against the comma splice, with the effect that it now seems limited to informal writing such as letters, or in reported speech.

The Ambassador...responded with a blast of enthusiasm. "Those weren't tough questions, those were kid-glove questions..." — John Updike, Bech is Back, 1982

MWDEU, which often debunks sticklerish rules with massed evidence from indisputably great writers, says "uncorrected examples are so hard to find in print" that "You should not try the device [of the comma splice] unless you are very sure of what you want it to accomplish."

I agree. Editors have made the comma splice so rare that they leap off the page (unpleasantly so, for me) when I spot one. The comma splice is unnecessary; a brief pause between two related thoughts can be accomplished by a semicolon like the one in this sentence. A full stop separates two thoughts more cleanly. Unless you're being aphoristic ("Man proposes, God disposes") or intentionally seeking a loose-knit style, beware that a comma splice is probably not worth the readers it will irritate. 

Update: Stan Carey, in a thoughtful post with lots of evidence, was softer on the comma splice than I am. He also reminds me that "Elements of Style" allows for limited comma splicing, so I added a bit to this effect above.