The central focus of language studies in India is, of course, Sanskrit. As a liturgical language of Hinduism, Buddhism, and Jainism, Sanskrit has played an outsized role in India’s linguistic development. Over its lifetime, Sanskrit traveled as far as Indonesia, Japan and Afghanistan on the backs of Hindu and Buddhist religious emissaries. The language’s name for itself, saṃskṛta vāk, means “perfected speech”—and its users genuinely believed that Sanskrit was indeed perfect. Sanskrit grammarians and authors looked down on commoners’ prākṛta, “natural”, languages as seriously deficient compared to Sanskrit. Rulers and other elites felt the same way. (These prākṛta languages, descendants of Sanskrit, eventually became most of the languages spoken in northern India, Sri Lanka, Pakistan and Bangladesh, including Hindi, Urdu, Bengali, Punjabi, Sinhalese and Marathi.) Because the native religious traditions of India highly value the precise oral recitation of scriptures, the liturgical language itself holds sacred importance. For thousands of years, Sanskrit persisted as a language of religion and elite education even as local vernaculars increasingly diverged from it. This relationship parallels the continued formal use of Latin in continental Europe through the Middle Ages despite the Romance languages developing apart from it, or the freezing of written and formal Arabic in its Koranic form as the spoken dialects became, in effect, new languages over the past 14 centuries.
Sanskrit’s position of prestige also allowed it to infiltrate the vocabularies of unrelated languages. This included the major languages of southern India, including Kannada, Telugu and Malayalam, as I wrote last week. Sanskrit also influenced (and was influenced by) Tamil, another major southern Indian language. More recently, Tamil-speakers have worked to shed the language of its Sanskrit borrowings, in part because of complex class and ethnic politics associated with the creation of modern India. Farther off, Sanskrit words penetrated deep into languages like Thai, Burmese, Khmer, Javanese, Balinese, Malay and Indonesian. For prominent examples, see Bangkok’s Suvarnabhumi Airport, from Sanskrit suvarṇabhūmi, “golden land”, or Singapore, from siṃhāpura, “lion city”.
Sanskrit’s star billing in these many languages doesn’t mean that they are related to, or descended from, Sanskrit. A language’s genealogy is much more fundamental. Figuring out whether two languages are related, however distantly, involves a thorough study of the structural features of a language. Linguists look at many things to determine structural relationships. How is a language’s grammar constructed? Are there vowel and consonant sound changes that have occurred in many words? Are there written records of intermediate forms of a language? Did ancient historians observe language change? Are there well-known social, class, ethnic and religious divisions that could have affected the way a language is shaped? Historical linguists spend decades piecing together the different ways languages could have changed over time. Persistent and systematic patterns usually provide the best clues.
It’s no secret that, say, Nepali is descended from Sanskrit, though. The job of figuring out more distant cross-continental relationships is altogether more difficult. Sanskrit, as it happens, played a central role in the development of historical linguistics. The existence of a language family stretching from Ireland in the west to Bangladesh in the east, now known as the Indo-European language family, was first proposed when an Anglo-English civil servant, William Jones, discovered persistent similarities between Sanskrit, Latin and Ancient Greek. Two centuries of thorough research has created a body of ironclad scholarship in Indo-European linguistics. The Indo-European relationship does not mean that Sanskrit came from European languages, or that European languages came from Sanskrit. It means that languages as different as Irish, Italian, Russian, Armenian, Farsi and Bengali all share a very distant ancestor, a language known as Proto-Indo-European (PIE). Over thousands of years, PIE and its successors spread across Eurasia. PIE’s linguistic descendants underwent natural sound change, absorbed other languages’ vocabulary and assumed unique characteristics. Over time, they became the hundreds of modern Indo-European languages.
In the Indian subcontinent, PIE’s descendant Sanskrit came into contact with Proto-Dravidian languages, the ancestors of today’s modern southern Indian languages. (Some Dravidian languages, like Brahui, are found in Pakistan, suggesting that the family was once more widespread across the subcontinent.) That long and fruitful exchange gave Sanskrit, among other features, a new set of common sounds—retroflex consonants—that aren’t found in many other Indo-European languages. In turn, Dravidian languages absorbed, and continue to absorb, Sanskrit sounds and vocabulary. But Dravidian languages are structurally unrelated to Indo-European languages. This fact gets obscured by the confusing relationship of Dravidian languages to Sanskrit. Lots of vocabulary has been adopted into Dravidian languages because of Sanskrit’s status as a prestige language, and the sound catalogue of some Dravidian languages has changed as a result of this contact. These exchanges don't change the genealogy of a language group. Dravidian languages are distinct from Indo-European languages, just as Japanese is distinct from Chinese despite borrowing some of its features, and just as Farsi is distinct from Arabic despite borrowing some of its features. Similarly, Burmese, Thai, Khmer, Malay, Indonesian, Javanese and Balinese have all absorbed a great deal of Sanskrit vocabulary. Just as in Dravidian languages, Sanskrit-derived terms are used in formal or ritual contexts in those languages. Linguists have studied these languages and deduced that (like Dravidian languages) the grammatical structure of each is fundamentally different from Sanskrit and other Indo-European languages.
When language communities interact, the product is hardly easy to categorise and parse. When these interactions happened ages before anyone bothered to record them, the task is much harder. Languages can absorb a great deal of another language without ever changing its structure. Distant linguistic relatives might even meet up again, unrecognised, as when Hindi absorbed a great deal of Farsi vocabulary during Mughal rule in India. Languages might meet up more than once, as in English's on-and-off relationship with Latin-derived vocabulary. And distinct language communities can have different layers of exchange. Far away and long ago, the medieval Indianisation of Southeast Asia was largely led by people who spoke Tamil, a major Dravidian language. They spread both Tamil and Sanskrit, along with religion, to places like Cambodia and the Indonesian archipelago. Nearly a thousand years later, Tamil-speaking people again reappeared in Southeast Asia, brought to places like Singapore and Malaysia as indentured servants for European colonists in the 1700s and 1800s. The Tamil-Southeast Asia cultural contact was reborn, adding a rich new layer of complexity to an already hybridised culture.
The serious study of Indo-European languages, just like the study of any language group, is not normally part of any political or social agenda. What we know about the Indo-European language family is the product of centuries of thorough research—not just in the Indian subcontinent, but in places like Iran and Europe, too. This has included the painstaking reconstruction of (an idealised form of) Proto-Indo-European, a language which was never written down, but which researchers know must have existed to account for the systematic similarities between Bengali, Russian, Portuguese and the rest.
In India, though, some people have been busily rewriting parts of Indian history to conform to jingoistic ideas about Indian exceptionalism and cultural superiority. They have attempted to cut out huge swaths of history involving the exchanges Indians have had with Greeks, Persians, East Asians, Arabs, Central Asians, Southeast Asians, and Western Europeans. They intend to write a story of Sanskrit and Hindu culture that is “pure” and devoid of foreign influence. Linguists know, based on reams of research, that a form of PIE, the language, did arrive in India from elsewhere, becoming Sanskrit over time. That fact doesn't have to diminish the "Indianness" of the language. Sanskrit's deep and longstanding cultural importance in the subcontinent is a strong enough connection. Its shared ancestry with farflung languages is just one of the many connections that have been made and remade over and over again in India's history.
This approach, of course, is nuanced and complex. Matters get complicated when religion and cultural identity is at stake, and Sanskrit isn’t alone in being used as an ideological tool. Hebrew, for another, has been touted as a “perfect” language and the source of all the world’s languages. Trained linguists describe the world’s languages as they are, not in the service of political, social or religious ends. It’s a shame that the conversation about India’s linguistic history gets twisted in ways that are at odds with what linguists and historians have deduced. Viewing India as a microcosm of the world’s diversity is far more fascinating. Seeing Indian languages as the product of many rich and varied cultural exchanges is far more exciting. These perspectives also have the virtue of being true to the facts.]]>
Now, the context. Three women had been missing in Cleveland for a decade. The man here, Charles Ramsey, rescued them after hearing a cry for help from a front door in his neighborhood. The area has been described as "rough" and working-class. From his speech you might guess that he is poor himself, and not highly educated.
Now, did you find him funny? Exotic? Stupid? If you found him funny, funny like an entertainer, or funny like entertainment?
I'll bet that how you see these things will correspond to some preconceived notions that you have. And one writer, Aisha Harris for Slate, found Mr Ramsey's rocket-ride to internet fame unsettling:
Ramsey has become the latest in a fairly recent trend of "hilarious" black neighbors, unwitting Internet celebrities whose appeal seems rooted in a "colorful" style that is always immediately recognizable as poor or working-class.
Well, yes. And why might that be?
It's difficult to watch these videos and not sense that their popularity has something to do with a persistent, if unconscious, desire to see black people perform. Even before the genuinely heroic Ramsey came along, some viewers had expressed concern that the laughter directed at people like Sweet Brown [another black internet celebrity] plays into the most basic stereotyping of blacks as simple-minded ramblers living in the "ghetto," socially out of step with the rest of educated America. Black or white, seeing Clark and Dodson merely as funny instances of random poor people talking nonsense is disrespectful at best. And shushing away the question of race seems like wishful thinking.
Now, did you see a "simple-minded rambler"? It's true that Mr Ramsey's accent, word-choice and grammar all suggest a black American without higher education:
I’m eatin’ my McDonald’s..
“it’s [ie, "there are"] some more girls up in that house”
You got some big testicles to pull this off, bro.
But the totality of the interview suggests a fast-thinking and clever man. What I, like many others, will remember best is the end of the interview:
Newscaster: What was the reaction on the girl’s faces? I can't imagine, to see the sunlight...
A: Bro, I knew something was wrong when a little pretty white girl ran into a black man’s arms. "Something is wrong here." Dead giveaway.
Of course many people are forwarding the video eagerly, in part because Mr Ramsey doesn't speak like the co-workers in their office towers. But they're also forwarding it because it's proof that a poor person is not dumb by virtue of the fact that he doesn't speak the Queen's (or, as we say in America, Broadcast) English. On the contrary, he's clearly quick on his feet in addition to being the kind of person who runs to save strangers. (In this video, not as funny, his thinking is on even clearer display.) I'd like—and I imagine Slate's Aisha Harris would like—everyone to remember the lesson that a heavy African-American accent or dialect has nothing to do with anything but dialect. Judge someone negatively because (for example) he says police and you just might misjudge a clever man—or a hero.]]>
The Post buried the real news, though: what the new paper does is claim this as evidence that 7 modern language families, not yet conclusively shown to be related, are part of an Ur-family called proto-Eurasiatic. By their theory, the Indo-European, Uralic, Altaic, Inuit-Yupik, Dravidian, Chukchi-Kamchatkan and Kartvelian languages all share a common ancestor. The descendants of these proto-languages are spoken in a vast territory covering most of Eurasia including the Indian subcontinent today.
What the Post doesn't even brush on is how controversial this is likely to be. Historical linguists have not just established the existence of proto-families. They have elaborately reconstructed them. By contrast, the authors of the latest PNAS paper have, apparently, found just 23 words they think are shared among at least four of the seven families in the putative Eurasiatic. Clever statistical analysis can make a stab at answering how likely this is to be due to chance. But such analysis after 150 centuries of language change can hardly give certainty.
I don't have the paper yet, but hope to get it and offer further thoughts. In the meantime, listen to audio recordings of the proto-words (link now fixed) as reconstructed in the intermediate families (Indo-European, Dravidian and the rest). The similarities are indeed striking. Are they also persuasive?
Addendum: Just last week, Piotr Gasiorowski wrote this, roughly the established view:
If we ever manage to prove that the IE [Indo-European] languages are related to some other established family, the reconstructed features of the common ancestor will naturally be even harder to constrain, and the protolanguage itself more elusive and fragmentary. It is hard to predict how far back in time our best reconstructive methods can take us before the notion od "protolanguage" becomes too vague to be meaningful. We can only resolve this question empirically, by putting our methods to extreme tests. If we consistently fail, it may mean that we have already reached the limit. Fortunately, there is no shortage of enthusiasts undaunted by the difficulties of long-range comparative research. Their efforts are necessary and praiseworthy, but the results so far have been rather disappointing. Only time can tell if further progress can be achieved.
But my favourite must be the Norman invasion of 1066. When the Normans, who spoke a dialect of Old French, ruled over England, they changed the face of English. Over the ensuing two centuries, thousands of Old French words entered English. Because the ruling class spoke Old French, that set of vocabulary became synonymous with the elite. Everyone else used Old English. During this period, England's society was diglossic: one community, two language sets with distinct social spheres. Today, English-speakers pick and choose from the different word sets—Latinate (largely Old French borrowings) and Germanic (mostly Old English-derived words)—depending on the occasion. Although English is no longer in a diglossic relationship with another language, the Norman-era diglossia remains reflected in the way we choose and mix vocabulary. In informal chat, for example, we might go on to ask something, but in formal speech we’d proceed to inquire. There are hundreds of such pairs: match/correspond, mean/intend, see/perceive, speak/converse. Most of us choose one or the other without even thinking about the history behind the split. Germanic words are often described as earthier, simpler, and friendlier. Latinate vocabulary, on the other hand, is lofty and elite. It’s amazing that nine hundred years later, the social and political structure of 12th-century England still affects how we think about and use English.
English isn’t alone in having this sort of split personality. Halfway across the world, languages spoken in southern India underwent similar changes. Kannada, Malayalam, Tamil, and Telugu, the four major languages spoken there, are Dravidian languages. They are structurally unrelated to the languages of northern India, which are Indo-European. But Sanskrit, an Indo-European language of ancient India and the liturgical language of Hinduism, has held prestige all over the subcontinent for over two thousand years. Kannada, Telugu, and Malayalam—and to a lesser extent Tamil—have absorbed, and continue to absorb, thousands of Sanskrit words. (A relatively recent movement among Tamil-speakers aimed to expunge the Sanskrit borrowings.) Much of southern India, just like Norman England, was diglossic between Sanskrit (used ritually and formally by Hindu elites) and vernacular Dravidian languages. Today, that diglossia is gone, but Sanskrit-derived vocabulary still forms an upper crust, mostly pulled out for formal speech or writing.
Some writing, especially poetry, still slants toward native vocabulary. Two influential religious movements among Hindu Kannada-speakers, the 12th-century Lingāyat and the 16th-century Haridāsa movements, treasured simple Kannada poetry. These movements arose in part to spread religious teachings beyond Sanskrit-educated elites to the common people. Works written then are largely devoid of obvious Sanskrit borrowings. To many Kannada-speakers, those works are softer and folksier than stiffer Sanskrit-heavy works. But caste and class politics didn’t end then, of course. Sanskrit still holds sway in India today, officially one of the "scheduled" languages listed in the constitution. It sometimes seems like any Kannada newscaster or speechwriter worth his salt swears by a Sanskrit dictionary. Sanskrit borrowings are used all over the place in order to sound proper, even when it sounds strange. (My favourite example of strained usage is the upscaling of “toilet” to shauchālaya, “abode of cleanliness”.) In the most tortured formal writing, Sanskrit words might just be strung together with Kannada grammatical endings. This has the strange consequence of allowing speakers of unrelated languages like Hindi to take a stab at translating the text. (Hindi, as it happens, is also split between the Sanskrit-heavy shuddh, “pure”, Hindi, popular in government and academia, and colloquial Hindi, which makes greater use of Arabic and Persian borrowings.) There’s some sweet spot in the middle of both extremes. Good writers seem to get it best.
It has always fascinated me how the Sanskrit/Dravidian divide in Kannada is so strikingly similar to the Latinate/Germanic divide in English. In English, word choice is often used to judge someone's class or education. In Kannada, caste is also mixed in. Picking certain words over others can have social consequences, branding the speaker or writer according to his vocabulary. In both languages, older borrowings underwent sound and spelling changes, but newer borrowings keep the roots intact. (In English, these old pre-Norman borrowings are mainly religious terms, like "nun", "monk", or "priest".) “Native” terms are considered earthier and Sanskrit/Latin-derived borrowings are stuffier. But there are interesting differences, too. English didn’t descend from Latin, though they’re both Indo-European. Dravidian languages, in contrast, aren’t related to Sanskrit at all. In Kannada, Telugu and Malayalam, the alphabet had to expand dramatically to incorporate Sanskrit sounds like voicing and aspiration. The shift was so complete that each language's alphabet, while written completely distinctly, contains nearly all of the same sounds as the Sanskrit-descended Hindi.
Many languages have "high" and "low" layers of vocabulary. But in most other languages, the two sets are drawn from the same source. By contrast, contact between Old English and French, Dravidian languages and Sanskrit, Japanese and Chinese, Persian and Arabic, and other pairings around the world have created fascinatingly hybrid languages. These mixed lexicons are, for linguistic and social historians, akin to the layers of fossils that teach paleontologists and archaeologists so much about eras gone by.
Some people even think English is descended from Latin, or Kannada from Sanskrit. That’s frustrating not only because it’s wrong, but also because the reality is far more interesting.]]>
The teachers — most of them from other countries — teach regular subjects like mathematics and reading and social studies, only speaking exclusively in a foreign language. At first, they pantomime and use pictures and videos to get their point across, but they say the students can understand them within a few weeks.
Utahns are quickly catching on.
Parents, wary at first, have rushed to enter lotteries to place their children in the programs. Some school districts have waiting lists 100 students long. Some parents drive 30 miles to bring their children to class, or have even moved to be closer to an immersion school.
This seems to mark a friendlier attitude towards foreign languages more generally. The English-only law was approved by nearly 70% of voters in 2000, while a majority of state legislators voiced their opposition at the time. But in 2003, exceptions to the law were carved out for Utah's public colleges and universities. Utah is one of the whitest states in the country, but immigration in the last decade (in part driven by Mormon converts) is changing the face of the state. Now, public schools' immersion programs could change the way Utahns think of foreign languages in their midst. In contrast, Utah's southern neighbor, Arizona, is still struggling with how to implement Spanish-English bilingualism in schools, even though (or perhaps because) it is roughly one-third Hispanic. Utah's language programs are an important step forward. In the last two decades, Salt Lake City has become more diverse than most Americans realise. With such a public embrace of multilingualism, the state could shed its monocultural reputation entirely.]]>
At some point L. Ron Hubbard takes to the sea and he moves the main people in Scientology to the sea with him. ... So at some point he decides to come back to land. He needs a safe place to be and a place where Scientology can flourish and he chooses Clearwater, Florida.
Mr Yagoda concludes that describing the past this way is a crutch: "it's essentially a novelty item. It's tacky. Give it a rest." I don't quite agree, but his description of the historical present prompted this digression on another use of the present tense that he points out: jokes. (More specifically, jokes in the form of a funny story.)
Here's a well known example, which inspired the name of a book on punctuation that sold in the millions:
A panda walks into a café. He orders a sandwich, eats it, then draws a gun and proceeds to fire it at the other patrons.
"Why?" asks the confused, surviving waiter amidst the carnage, as the panda makes towards the exit. The panda produces a badly punctuated wildlife manual and tosses it over his shoulder.
"Well, I'm a panda," he says, at the door. "Look it up."
The waiter turns to the relevant entry in the manual and, sure enough, finds an explanation. "Panda. Large black-and-white bear-like mammal, native to China. Eats, shoots and leaves."
The panda walks in, orders, eats, and draws a gun. It would be weird to begin a joke—in English—with "a panda walked into a bar and ordered a sandwich."
But that's not how all languages work. In looking around at joke websites, I found that conventions vary a bit. This joke, in Portuguese, sets the scene in the imperfect (a tense used for continuous or repeated actions in the past), before moving to what we might call the jocular present:
Um homem estava com a família visitando o zoológico, quando chega um funcionário todo afobado e diz:
— Senhor, senhor!
O homem responde:
— O que foi? Qual é o problema?
— Uma desgraça! Sua sogra caiu no poço dos jacarés.
O homem, na maior calma, diz para o funcionário:
— Não quero nem saber! Vocês é que tratem de salvar os jacarés.
A man was visiting the zoo with his family, when a flustered employee comes up and says
- Sir, sir!
The man responds
- What happened? What's the matter?
- A disaster! Your mother-in-law fell into the alligator pool!
The man, supremely calm, says to the worker
- Not my problem! *You* try to save those alligators.
A look at a couple of Spanish joke websites show some jokes beginning in the imperfect, while others start in the present tense. German websites (like this one) show the English-style preference for the "jocular present":
Zwei Männer im Supermarkt stoßen zusammen. Meint der eine völlig aufgelöst:
"Entschuldige, aber ich bin total durcheinander, ich suche meine Frau!"
Darauf der andere: "Mir geht es auch so, seit 30 Minuten suche ich schon.
Wie sieht Deine denn aus?"
"Meine hat schwarze lange Haare, ist 1.80 m groß, braungebrannt, vollbusig, schlanke Figur, hat einen superkurzen Mini an, ein weißes enges Top ohne BH und Schuhe mit sehr hohen Absätzen. Und wie sieht Deine aus?"
"Scheiß drauf, wir suchen Deine."
Two men bump into each other in the supermarket. The first one says, distraught
"I'm sorry, but I'm completely flustered; I'm looking for my wife!"
The second replies "Me too; I've been looking for 30 minutes."
"What does yours look like?"
"Mine has long black hair, 5'11", tan, busty, thin, has a super-short miniskirt on, a tight white top and no bra, and high heels. What does yours look like?"
"Screw it; let's look for yours."
(No comment on the quality of the jokes here. This is strictly hard-core linguistic analysis.)
I would have assumed all languages use the jocular present if not for the memory of one session with my Arabic tutor. My exercise was to read a corny joke (which I can no longer remember) in a newspaper, and then tell it to my tutor without looking at the paper. An twist was that it was written in modern standard Arabic in the paper, but I had to deliver it in (Palestinian) spoken dialect, which involves a non-trivial translation process on the fly. This wasn't easy in the first place, but what really threw me was when my teacher repeatedly stopped me and had me render everything in the past. Every verb, every bit: "A man walked into a restaurant and said... And the woman said back to him..." It was surprisingly frustrating. I kept naturally switching to the present, and my teacher, equally annoyed, kept making me switch back. "Why are you saying everything in the present tense?"
Not sure that my teacher wasn't just unusual, I did a little searching and found this Arabic joke:
سأل المعلم تلميذاً
ما هي الرياح؟
فأجاب التلميذ : هواء مستعجل
The teacher asked a pupil: What is wind?
The pupil replied: air in a hurry.
This is interesting, because Arabic sometimes uses the present tense to describe past events.
The world's best-known linguist, Noam Chomsky, has staked a career on the theory that all natural languages share a universal grammar. Whether he is right or not, the jocular present doesn't seem to be a part of it.
If you speak another language well enough to know jokes, what tense do you use in telling them?]]>
Done? The answer, according to GlobalEnglish, is probably not what you think. (GlobalEnglish is owned by Pearson, which part-owns The Economist).
Remember that the survey tested non-native speakers, so don't be too distracted by the placement of the Anglophone countries. Still, Global English's results are very strange. If you thought that the Philippines has the best business English in the world, and that Germany would miss the top 25, you haven't had the same work and travel experiences I have.
Global English says this about its Business English Index:
The GOE [Globalisation of English] and BEI [Business English Index] together give us a complete picture of the trends, achievements and challenges in business communication and the importance of Business English in the workplace.
If this is a complete picture, we have a genuine stop-the-presses moment here: Madagascar, Bulgaria and Romania well ahead of Denmark, Switzerland and Germany? Spain just behind Angola? If this doesn't violate your common sense about the relative quality of English around the world, look at a few neighboring and demographically similar countries. Slovenia miles ahead of Slovakia and the Czech Republic? Argentina and Uruguay leagues ahead of Chile?
In mild amazement, I wrote to Global English's press-relations person, asking how the study had been compiled. His reply:
The Business English Index is based on the placement test that GlobalEnglish subscribers take within GlobalEnglish Edge, an on-demand Business English development solution, to assess the Business English proficiency level of each respondent. GlobalEnglish later analyzes the data based on respondents’ demographics, location and employer information to produce the annual report.
So what we're looking at is not "speakers of business English", but paying GlobalEnglish subscribers.
Statistics 101 should have taught the compilers of this study to frame these results very carefully. The idea that GlobalEnglish subscribers are representative of business-English speakers generally is perhaps completely backward: companies that subscribe to GlobalEnglish presumably feel the need to invest a lot of money to improve their business English. GlobalEnglish may well do that job superbly. But the obvious corollary is that many companies will not sign up for GlobalEnglish's services because their employees already speak good English. And those companies tend to be in the rich countries of northern Europe, which is why another study of English proficiency by EF, another language-teaching company, found a more predictable top 5: Sweden, Denmark, Netherlands, Finland and Norway, with another clutch of wealthier European and Asian countries making up the next tier. EF's index isn't perfect either, since, as EF notes, "the test-taking population represented in this index is self-selected and not guaranteed to be representative of the country as a whole." But at least EF included this straightforward caution. The GlobalEnglish paper includes no methodological note.
I missed the release of the GlobalEnglish rankings last year, but a few journalists didn't, uncritically swallowing the Philippines' number-one ranking then. Here's Yahoo News:
Well, people will now have to think twice before mocking Pinoys' use of the English language.
The Philippines was named the world's best country in business English proficiency, even beating the United States, according to a recent study by GlobalEnglish Corporation.
GlobalEnglish has released early this month the results of its annual Business English Index (BEI), the only index that measures business English proficiency in the workplace.
You won't catch me mocking Pinoys' use of the English language, but that's because that's a stupid and spiteful thing to do. I doubt very much that they have the best business English on earth. Getting and analysing good data on language proficiency is time-consuming and expensive, but if you're going to do it, it's worth doing right.]]>
And the [masculine] worker, who for twelve hours weaves, spins, drills, turns, builds, shovels, breaks stones, carries loads etc. - does he hold this twelve hours' weaving, spinning, drilling, turning, building, shoveling, stone breaking to be an expresion of his life, as life? [Und der Arbeiter, der zwölf Stunden webt, spinnt, bohrt, dreht, baut, schaufelt, Steine klopft, trägt usw. - gilt ihm dies zwölfstündige Weben, Spinnen, Bohren, Drehen, Bauen, Schaufeln, Steinklopfen als Äußerung seines Lebens, als Leben?]
Note "weaving" and "spinning". Marx was not referring to the work that only men did in his time. But nonetheless, the German of the mid-19th century called for a "he" when referring to "the worker". Language, it seems, was in sexist chains for centuries.
That was then. Today comes news that the state of Washington has finished a long project cleaning out all male-exclusive terms in its statutes, when those terms should apply to both sexes. No more "signalman": the law now refers to a "signal operator". "Freshmen" in college and high school will now be "first-year students". In such cases, the change was straightforward. As we noted before, some gender-neutral terms like "police officer" and "flight attendant" are now well known.
But the Washington overhaul has pressed into service some awkward coinages. "Fishermen" will now be "fishers", a word I can recall only ever having seen in the Bible ("Come and I will make you fishers of men"), and even then only to avoid the awkward "fishermen of men". An "ombudsman" will now be an "ombuds". "Ombudsmand", a Scandinavian word, has the etymological meaning a "man who is asked for something", ie, help or redress. Washington has shorn the title down to a meaningless "ask-for".
A "journeyman plumber" will become a "journey-level plumber". But what will a simple "journeyman" be? The bill does not specify a "journeyperson". "Penmanship" will now be "handwriting", though they aren't quite the same thing: "penmanship" connotes an acquired skill, while "handwriting" can mean even untutored or ugly script.
And in some cases, Washington could not change its statutes at all: there was simply no good replacement for "manhole", for example. And "airmen" and "seamen" are officially so called in America's air force and navy, so Washington (state) cannot rename them without Washington (DC's) approval.
"This was a much larger effort than I had envisioned," Jeanne Kohl-Welles, the senator who sponsored the bill, told Reuters. The effort is laudable, and of course it's impossible to please everyone. One Washington radio host complained sarcastically that "I won't like our society until every human being in the Puget Sound area is completely, totally and utterly androgynous. Forget about the language, I don't think we should have men or women anymore." Ranting is good fun, but Johnson doesn't agree that gender-neutrality is a risible waste of time. Language matters. Still, in all things, balance is best. With some of these bizarre coinages, the 475-page bill seems to have gone a bit farther than it absolutely had to. Readers who disagree are invited to contact our ombuds.]]>
AT THE Lingua Franca blog, Ben Yagoda describes a conversation Ruth Fraklin of the New Republic over anti-Semitic code language in America before and during the second world war. "Restricted" is perhaps the baldest of all the terms (as used by a thuggish detective in the Coen brothers' 1991 masterpiece, Barton Fink, above). Apparently, "no Jews" code was particularly common in hotel advertisements. Mr Yagoda and Ms Franklin discuss "exclusive" and "selected clientele" among other euphemisms. Shockingly, ads like these persisted into the years of America's participation in the war against Hitler.
At least a silver lining is that, on some level, people know naked racism is wrong, wrong enough to disguise in euphemism anyway. Anti-black racism needed no code in the pre-1960s era: "Whites Only", etc. Now, people know that it's not acceptable to reminisce about the good old days of Jim Crow. But several years ago I responded to a reader who, I thought, protested far too much in proclaiming that "ghetto" as an adjective ("that's so ghetto") had no racial overtones. Of course it does.
When I moved to New York in 1999, I was lightly outraged to find that some personal ads included a bit of code. Living in Brooklyn, I would have been excluded from consideration by anyone who wrote a personal ad looking for "212 only". That's the area code for Manhattan; the outer boroughs mostly shared 718. People seeking "212 only" wanted no "bridge-and-tunnel" types, presumably lower-class. 212 snobbery featured in an episode of "Seinfeld", where a man loses interest in Elaine after discovering she has a 646 number (then new, thanks to a shortage of 212 numbers).
ELAINE: It's a new area code.
MAN: What area? New Jersey?
ELAINE: No, no. It's right here in the city. It's the same as 212. They just multiplied it by 3, and then they added one to the middle number. It's the same.
MAN: Do I have to dial a one first?
(Elaine nods and the man crumples up her number.)
MAN: I'm really kinda seein' somebody.
ELAINE: Yeah? Well, so am I!
A decade and a bit later, t-shirts can be brought in oh-so-hip Brooklyn emblazoned with a proud "718", and anyone who still actually has a 718 number can usually brag of having moved to the borough before it took off. Throw "212 only" in the bin with "selective clientele".
I'm sure readers have many more tales to tell of coded language like this.]]>
In part because of this longstanding division, some Flemish and Wallonian laws are fiercely protective of Dutch and French. But the European Court of Justice (ECJ) thinks at least one of those laws has unacceptable consequences. Flemish law had previously considered only Dutch-language contracts authentic. Contracts in other languages would be nullified. Anton Las, from the Netherlands, had been contracted to be chief financial officer of PSA Antwerp, a Belgian subsidiary of a Singaporean port operator. His contract was in English. Unfortunately for Mr Las, his contract was nullified under the Dutch-only law, and he was booted from the company. Denied a job, he sued. A Belgian court, unsure of how to answer, asked the ECJ to make a preliminary ruling on that section of Flemish law. On Tuesday, the ECJ decided that the policy violated EU law.
The ECJ agreed that preserving and promoting a country’s language is important. But the laws must be proportionate to the need. Because EU law protects workers' freedom of movement, restrictions on that freedom must be carefully crafted. The court worried that a Dutch-only law would dissuade people from moving to Flanders and taking up work.
[T]he objective of promoting and encouraging the use of Dutch, which is one of the official languages of the Kingdom of Belgium, constitutes a legitimate interest which, in principle, justifies a restriction on the obligations imposed by Article 45 TFEU [the EU law covering freedom of movement for workers].
... [But] in order to satisfy the requirements laid down by European Union law, legislation ... must be proportionate to those objectives.
The court recommended allowing foreign transactors to use a mutually intelligible language:
[P]arties to a cross-border employment contract do not necessarily have knowledge of [Dutch]. In such a situation, the establishment of free and informed consent between the parties requires those parties to be able to draft their contract in a language other than [Dutch].
The ECJ’s preliminary rulings are binding, so the Flemish government will have to rejigger its policies. The law in question here was written with no flexibility at all, suggesting that it served linguistic nationalism more than genuine need. Belgium has outsized language worries for its size, so it will be curious to see how Flemish businesses react to new rules. I suspect that more flexibility will revitalise, not doom, business there.]]>
But what about isolation by langauge? Few have acknowledged the effects of language barriers in prisons, which can cut off prisoners from meaningful human contact for long stretches. The closest America has come to addressing this issue is through its Crime Control Act of 1990, which requires that inmates in federal prisons take English language classes until they reach the proficiency of an eighth grader. But this does not apply to individual states, nor to irregular prisons like the one at Guantánamo Bay, where a prisoner may not be considered "isolated" if others are physically present. Anyone who has gone even a day without speaking to another person, though, knows that this is real solitude, even when others are around. How exactly does linguistic isolation affect the psyche?
In a recent paper, Peter Jan Honigsberg of the University of San Francisco law school has taken a step toward answering this question. In "Alone in a Sea of Voices: Recognizing a New Form of Isolation by Language Barriers", Mr Honigsberg compares physical and linguistic isolation, and found the similarities uncanny. In the most notable example, Honigsberg described a 16-year-old boy who spent eight years at Guantánamo after 9/11 because no country would have him, despite his quickly-discovered innocence. During those years, he was surrounded by prisoners who only spoke English or Arabic, while he knew only Uzbek. He was given no materials to help him learn English or Arabic, and no translator after his initial imprisonment. This resulted in a loneliness so acute that he cried every time he woke up. Mr Honigsberg wrote that this situation is akin to the anguish experienced by a stroke victim who is surrounded by conversations, but cannot participate. In immigration facilities, the same problems pertain, but are less severe, as prisoners spend less time there. Nonetheless, those suffering from language isolation experience disorientation and a decline in their decision-making ability.
This has not gone completely unnoticed. The European Council's European Committee in Crime Problems, for example, has said that the "inability to communicate in the language most commonly spoken in a prison is a severe barrier to foreign prisoners' ability to participate in prison life. It is the root cause of many problems such as isolation." But the council can only recommend, not require, its member states to take action.
Of course, language isolation will affect different people in different ways. If someone is naturally quick at picking up languages, his or her isolation will be short-lived and therefore not as hard to bear. But many prisoners will not be so gifted. And the problem is exacerbated in prisons that allow only one language to be used. Mr Honigsberg gives the example of a European Court of Human Rights case in which a Tajik inmate in a Russian prison was only allowed to speak Russian, even when his family came to visit. The court found that this "violated Article 3 of the European Convention of Human Rights, prohibiting torture or inhuman or degrading treatment or punishment".
The UN's 2008 report on solitary confinement states that "the practice has a clearly documented negative impact on mental health," mentioning insomnia, confusion and hallucinations, "and therefore should be used only in exceptional circumstances." While the isolation felt due to language barriers may be dulled slightly because there are other people nearby, policymakers should nonetheless take it seriously. Humans have an inherent need to communicate, and to be denied this need is to be denied a fundamental right.]]>
Pity the New York Times. Activists will always try to get journalists to use their preferred language. But few outlets actually make decisions that matter. Britain's cutthroat newspaper market means that the papers seem to compete to use the most inflammatory words they can. America is different; its papers pride themselves on objectivity first and foremost. And the Times matters like no other American broadsheet. Whatever words the paper chooses in describing controversial topics, someone is going to be unhappy. So it seems that each Times public editor (the person in charge of responding to reader complaints, independent of the rest of the editorship) must dedicate a column to the language of the "newspaper of record".
In this case, the three terms above have annoyed Times readers, who think that Times reporters have bought government spin uncritically. Do they have a point?
On "harsh interrogation techniques", Margaret Sullivan, the public editor, got this response from Philip Corbett, the "standards editor" (a title that makes me wonder what the other editors are responsible for):
The word torture, aside from its common sense meaning, has specific legal meaning and ramifications. Part of the debate is on that very point.
This is, itself, a legalistic response worthy of a cunning lawyer. Mr Corbett is saying that torture is illegal, and calling American government action "torture" might falsely convey the appearance of a legal conviction.
But the law defines all kinds of terms, including some very common ones, for the purposes of applying the law. Journalists, however, aren't executing laws, and so needn't ignore a common-sense usage to privilege the legal one. The "harsh interrogation techniques" in question, used especially in the early years after September 11th, 2001, are well known. Waterboarding is the best known. Were they torture? What do most people consider torture in the real world? Here's a simple definition: inflicting treatment on prisoners that they cannot stand, such that they will do almost anything to get the treatment to stop. This could be ripping out toenails or electrocution, but "torture" can also leave no physical mark. It could be sleep deprivation; it could be putting a gun to a prisoner's child's head and pulling the hammer back. As for waterboarding, everyone who has undergone it (even as a simulation, with friendly trainers) concedes that a few seconds are enough to induce a terror that will cause the sufferer to do anything to stop it. The Economist's position is clear: torture it is, and can never be justified.
What about "detainee"? One reader complains that people are "detained" for a few minutes for a minor reason. Some of the men at Guantánamo Bay have been there for over a decade. Here, Mr Corbett admits that "detainee" is awkward, but says the word "prisoner" is inapt because those held “are in such an unusual situation – they are not serving a prison term, they are in an unusual status of limbo.”
But once again, "prisoner" has a strictly legal meaning and a common-sense meaning. The Times referred in 2008 to an Austrian father who "imprisoned" his daughter in his basement for 24 years, and there was certainly no due process there. The men at Guantánamo are being held in a prison. The Economist has called them "prisoners" many times. That said, we have referred to "detainees" at Guantánamo even more. There's a good case for reversing the ratio, and preferring "prisoners".
Finally, is "targeted killing" really "assassination"? Here's where the Times's readers have the weakest case. The Times says that assassination should be of a political figure. The Economist reserves "assassination" for "the killing of a prominent person, usually for a political reason". That definition is obviously open to interpretation. Was Anwar al-Awlaki, the American-born al-Qaeda propagandist killed by an American missile in Yemen, "a prominent person" killed for a "political reason"? He had not himself committed terrorist acts, but he certainly incited them. (The Obama administration also says he was planning attacks to come.)
Here again, though, "assassination" has a common-sense meaning. If you surveyed people by asking them quickly to name famous assassinations, they would name people like Kennedy, Lincoln or Martin Luther King, not Anwar al-Awlaki or even the much more famous Osama bin Laden. Someone "assassinated" is not, typically, a fighting man. And there's second reason "assassination" is inapt, having to do with the other end of the gun. In the common understanding, uniformed military forces don't "assassinate". That's the job of a lone gunman, a secret society or, perhaps, irregular or covert forces working for a government. (Many people certainly think that the CIA "assassinated" John Kennedy, for example.) When critics call military personnel "assassins", the point is not so much accuracy as the attempt to compare them to the likes of John Wikles Booth or James Earl Ray.
Say what you like about the policies themselves—they can certainly be fairly criticised. But even in criticism, plain is best. "Killing" is not only relatively neutral, but vivid enough to remind people what is going on. "Targeted" can be useful to distinguish attacks meant for a single person from the hurly-burly of general combat, a very different thing both operationally and morally. Those opposed to drone strikes can leave off "targeted" if they like, or refer to drone strikes specifically. And they can and should raise hell about the unintended consequences, including many dead civilians. But for clarity, "assassination" is best kept to its traditional meaning: irregulars killing politicians, and not the other way round.
Update: With perfect timing, today's Times carries this story: "U.S. Practiced Torture After 9/11, Nonpartisan Review Concludes":
A nonpartisan, independent review of interrogation and detention programs in the years after the Sept. 11, 2001, terrorist attacks concludes that "it is indisputable that the United States engaged in the practice of torture" and that the nation's highest officials bore ultimate responsibility for it.
The constitution is clear about which language to use in India's senior courts. In the Supreme Court and High Courts, English is used in all official documents. This makes sense. The language of the law in India is English. At the appellate level, only lawyers, who are mostly competent in English, present to the court. (The trial courts, which hear directly from witnesses and litigants, can use a state’s other official languages.) Lawmakers can allow another language in a High Court, but so far only four High Courts—in the states of Rajasthan, Bihar, Uttar Pradesh, and Madhya Pradesh—have done so, in each case with Hindi.
The unique allowance made for Hindi in some High Courts smarts for the country’s Tamil-speakers, who have long resented Hindi encroachment from the north. On Sunday, the chief minister of Tamil Nadu called upon national lawmakers to allow Tamil to be introduced as an official language in the Madras High Court. She argued that “If we are to take the administration of justice genuinely closer to people, then it is absolutely imperative that the local language is used in the High Court, as is already being done by the state government and legislature."
Tamil, India’s fifth most-spoken language, is a Dravidian language with few links to Hindi, an Indo-European language. Unlike other Dravidian languages, Tamil has largely resisted borrowings from Sanskrit, the ancestor of modern north Indian languages. This makes Tamil particularly different from India’s other major languages. These differences are a source of pride. In some ways, the country’s north-south divide is sharpest in Tamil Nadu. Between 1937 and 1986, Tamil-speakers repeatedly protested against the broad adoption of Hindi in India’s central government. Indian states were largely reorganised in 1956 to take account of language. National leaders had planned to keep Hindi and phase out English soon after independence. But pro-Tamil protests catalysed the adoption of the Official Languages (Amendment) Act of 1967, which ensured the survival of the central government’s official bilingualism, a practical recognition that English remained the only workable lingua franca for central government.
This hardly put an end to the squabbling, especially at the state level. Internal immigration since then has led to a sharp rise in Hindi use in major cities, such as traditionally Kannada-speaking Bangalore, and to resistance by speakers of the regional languages. Of course, some of this tension is manufactured. Though there is genuine popular resentment against Hindi in some quarters, regional parties (many of which rose to prominence in the wake of these protests) have sometimes waved the flag of language nationalism to distract from more pressing matters.
Given this history, it’s apparent why Tamil Nadu lawmakers would want Tamil in the Madras High Court. It’s less clear that it is “absolutely imperative”. (Why is Hindi necessary in those four High Courts, either?) High Courts do sometimes hear trial-level cases, but these are rare. Language exceptions for such cases might make sense. But why extend the services to appeals, which make up most of High Courts' dockets? Only lawyers participate in appellate proceedings. Indian lawyers—even Tamil-speaking ones—learn the law in English. In trial courts, interpretation and translation are indeed vital. In appeals-courts which mostly host exchanges only between English-educated judges and lawyers, Tamil doesn't seem so necessary. Are cultural preservation arguments persuasive enough to justify the expenditure? If Tamil and Hindi, why not all other languages? Is there a genuine and unique need for Tamil (or, in fairness, Hindi) in a higher court? Indian readers and appellate lawyers are particularly welcome to jump in here with their thoughts.]]>
we're looking at code-switching a little more broadly: many of us subtly, reflexively change the way we express ourselves all the time. We're hop-scotching between different cultural and linguistic spaces and different parts of our own identities — sometimes within a single interaction.
Many readers will be familiar with this phenomenon:
Your mom or your friend or your partner calls on the phone and you answer. And without thinking, you start talking to them in an entirely different voice — still distinctly your voice, but a certain kind of your voice less suited for the office. You drop the g's at the end of your verbs. Your previously undetectable accent — your easy Southern drawl or your sing-songy Caribbean lilt or your Spanish-inflected vowels or your New Yawker — is suddenly turned way, way up.
Public figures do it too. Mr Demby offers this 2009 video of Barack Obama in a chili-dog restaurant in a black neighborhood in Chicago. His voice is distinctly that of a black man comfortable in black Chicago, not that of the man who would give a soaring, formal inaugural address soon after. Asked if he needed any change for his chili dog, he tells the cashier, "Naw, we straight!"
A certain kind of politician can get away with this. I always enjoyed watching Bill Clinton dial up his southern-fried accent as needed for a barbecue or down for a speech to the United Nations. But I cringed just as much watching Hillary Clinton try to do the same. Here she is quoting some gospel lyrics to a black church in Selma, Alabama in 2007. Her corny southern accent sounds more like George Wallace, Alabama's segregationist governor and presidential candidate, than any of the black churchgoers in the audience. Americans like their politicians to have a common touch, but it should't fit like a crooked fake Groucho Marx moustache.
A good comedian can, of course, call in accents as needed, and the black comedy duo of Key and Peele are very good comedians. Watch the two accents gradually deepen as they negotiate their black identities through an order at a soul-food restaurant.
Click through to Mr Demby's post to see another Key and Peele skit, in which code-switching acts more like a shield than a sign of solidarity.
Language is a proxy for identity, and so code-switching is an apt metaphor for handling more than one identity. That makes it a great name for a blog on race, and I'll look forward to following it.]]>
Mr Bernstein was a long-time editor for the New York Times, and a respected usage writer in his day. His first popular book, 1958's "Watch Your Language", was his best-known (and helps explain the title of the follow-up that I had found on the ground). Much of the book could be used as-is today. But "More Language that Needs Watching" is also a fun introduction to what has changed in a half-century of writing advice.
Bernstein's short introduction to newspaper journalism in 1962 is a charming time-capsule in itself. In his typical day, describing the reporting of a crime story, an old clipping on the suspect is retrieved from the newspaper's "morgue", a term I'd never heard in this context. The writer hands his story to the copy desk bit by bit (in "takes"), so that the story can be edited as he continues pounding away. And it was a "he": "newspaperman" was Bernstein's term.
Then there are the bits of Bernstein's advice themselves. Here he is on a headline including the word anent:
"Retort to Democrats Anent U-2 Cites Statements by President". Except in legal usage, "anent" is archaic and semiprecious. Why not "about"?
I struggle to imagine an age in which journalists had to be chided not to use "anent". Today, it is fully obsolete in general usage, and counts as archaic and precious even among lawyers.
A few pages along, Bernstein quibbles about "co-ed":
"Co-ed commutes to Night Club." The caption referred to a Smith College girl and therefore was wrong. A co-ed, so the standard and slang dictionaries all tell us, is a female student in a coeducational institution, which Smith is not.
It's very hard to imagine the New York Times referring to a "co-ed" these days. The condescending term fit an age in which women at formerly male institutions were still unusual enough to merit their own word. Today, "co-ed" is mostly limited to the likes of Playboy.
Just a little further on, we find a battle that had just begun:
CONVINCE: "Three unidentified person who had taken the girl to the airport tried futilely to convince her to take her seat on the plane." "Persuade" would be the proper word in this construction. "Convince" may be followed by an "of" phrase or a "that" clause, but not by a "to" infinitive.
Google's N-gram viewer confirms that the phrase "convinced him to" had just begun its ascent when Bernstein was writing. (The red tick shows 1962).
Today, Bryan Garner's "Modern American Usage" and The Economist's style book still prefer "persuade ____ to", but it looks as though the rise of "convince ____ to" may be unstoppable.
Beside the entries that have become whiskery with time are many that are still useful for the reader today: clarifying the distinction between "careen" and "career"; reminding that "decimate" means "to destroy a large portion of", not "to annihilate utterly"; or pointing out that the diplomatic title is "counselor"/"counsellor", not "councilor". Overall, Bernstein is jovial fun, and generally shows the good sense of a man who would not have expected the language to be exactly the same 51 years after publishing his little book. My thanks go to whoever left it on the pavement for me.]]>
Domestic courts in America should take a leaf from this book. Several courts, especially in diverse cities and in the south, do offer multilingual services. (Johnson was delighted to find a sign advertising interpretation in 19 languages in a New York courthouse in January.) Behind the friendly sign, though, the scene in the courtroom isn’t so rosy. Speed and quality lag behind their international counterparts, even in Spanish. Both of us have witnessed awkward, stilted courtroom exchanges—in Los Angeles, southern Arizona, New York and Connecticut—in which Spanish-English interpretation was so slow and confused that nobody knew how to proceed fairly. Whatever one’s stance on language politics in America, it’s apparent that half-baked language services can seriously harm the project of justice.
Take, for example, testimony by a witness who speaks only Spanish. The court-provided interpreter must translate the lawyers’ questions from English to Spanish, then the witness’s answers back from Spanish to English. Questions and answers are often long and complex. When the interpreter is unable to manage simultaneous interpretation—whether due to the interpreter’s limitations or the courtroom’s technology—the result is often a paraphrasing. This is, in short, bad. Given the fine-toothed comb that lawyers will later take to the record, the low fidelity of much in-court interpretation causes problems. A word added, omitted, or even just slightly different from the witness’s actual testimony can wreak havoc on a case.
This is not to mention the role that these issues play in the administration of justice elsewhere in our legal system. Interpretation and translation are often central to proceedings for deportation, asylum, or welfare removal. Non-English speakers in these kinds of hearings “have a right to competent translation services” (as the Ninth Circuit, a federal appeals court, put it). But the same problems that plague interpretation of witness testimony often arise here as well. And even when a court knows the interpretation was faulty, relief is only granted when “better translation would have made the difference in the outcome of the hearing”. Proving this sort of prejudice is an extremely high bar to clear, meaning that only the worst instances will be corrected.
These issues aren’t just hypothetical. He v Ashcroft was heard by the Ninth Circuit. A Chinese alien, He Wang, appealed an adverse deportation and asylum decision in part by claiming defective interpretation. The appeals court noted (quite soberly) that “some of the evidentiary problems in this case appear[ed] to stem from interpreting difficulties.” The interpreter that had been provided to Mr He spoke Mandarin, while Mr He spoke the distinct Chinese language of his hometown of Fuqing, which is mutually unintelligible with Mandarin. (The court had no Fuqing interpreters.) Though Mr He apparently indicated that he did speak some Mandarin, his actual ability was limited. The two had difficulty even communicating with one another. The resulting interpretation of Mr He’s testimony was bad enough for the court to remark that “some portions of the transcript read like ‘Who’s on First.’” To its credit, the court recognized that “faulty or unreliable translations can undermine the evidence” on which a decision is based.
It’s great that the Ninth Circuit identified the problem. Now American courts should work toward solutions. What can they learn from international courts? In courts that regularly work with non-English speakers, simultaneous interpretation should become widely available for some common languages. (Even if it's present in some courts, it isn't everywhere.) Simultaneous interpretation, when done well, moves proceedings along quicker. More important, it reduces paraphrasing and increases fidelity. But it’s not foolproof. A second court stenographer should transcribe non-English testimony along with English. If there’s an important discrepancy between the two, the non-English transcript will provide a definitive original source. Right now, only English is transcribed, and only English goes into the official record—so if the interpreter gets it wrong, the record is wrong. In providing these resources, courts probably can’t be overly ambitious. But in most communities there are obvious places to start, like Spanish or Mandarin.
Of course, both of these recommendations require serious improvements in courts’ technology and capacity. It’ll be impossible to reach every language and dialect, so someone like Mr He might still be out of luck. Some upgrades will be expensive. Most American courts weren’t built with today’s technology in mind. But baseline improvements are possible. The international courts’ language apparatuses show that it’s possible to integrate newer interpretation services into a modern courtroom effectively. The costs are worth it. Taxpayers are already paying for interpretation services that are often inadequate (and hardly cheap). And as interpretation and transcription services mature, proceedings will in turn become quicker, smoother, and (most important) fairer. Due process is at the heart of any judicial system. Courts should be committed to ensuring justice regardless of language.]]>
So sensitive were liberal Angelenos to the possibility of appearing xenophobic that they almost invariably used the term "undocumented worker" rather than "illegal alien," which made contravention of the immigration law sound like some trivial problem of paperwork rather than, for better or worse, a breach of the laws of the United States.
A couple of decades later the linguistic tastes of LA's Westside have conquered swathes of America's media. A big victory came this week when the Associated Press decided to ditch the term "illegal immigrant" from its stylebook. AP journalists should instead refer to people who are "living in a country illegally" or who "entered the country without permission". ("Undocumented" is also rejected, on the grounds of imprecision.) Thanks partly to the shrinking of American newsrooms, AP stories appear widely in newspapers, and many adopt AP's style guide for their own stories. "The dominoes will start falling" at other publications, said one campaigner. Expect "people who are living in the country illegally" to be coming to a newsstand near you soon.
The Economist's style guide contains no such directives. Over the years we have tended to stick with "illegal immigrant", although on occasion we have simply referred to "illegals", a word many find offensively reductive (my colleague explored the distinction in more detail a couple of years ago). The phrase "illegal immigrant" has the virtues of concision, clarity and familiarity, although not necessarily precision: it does not distinguish between someone who has crossed the border illegally, someone who has overstayed a visa, a worker violating the terms of his visa arrangements, etc.
But the real objection to "illegal immigrant" is not far off the objection to "illegals": that it reduces well-rounded human beings to avatars of lawlessness. The word "illegal," according to AP's executive editor, should be used to describe only actions, not people. This is more or less what the campaigners pushing for this change have argued all along, as this short Slate piece explains. Describing someone as "living in a country illegally" may use up more of a journalist's word count, but it leaves space for that person's humanity and character.
The discussion over what campaigners call the "I-word" comes, of course, as Congress begins a much larger debate: over how to construct path-to-citizenship (or at least path-to-legalisation) rules that mean for most illegal immigrants in America "illegal" will be rendered not only impolite but incorrect. The timing is not coincidental: it is an expression of the growing political clout of America's Latinos, at the ballot box, inside the academy and elsewhere. After AP's decision this week, a contributor to an e-mail discussion group for Latino political scientists referred to "a multi sector national, online and offline campaign to finish the job".
The biggest prize, however, awaits. As recently as last October Margaret Sullivan, the New York Times's public editor, said that banning the term "illegal immigrants" would not serve readers well. Now the newspaper appears to be softening its stance. "Undocumented", which the paper has previously considered euphemistic, may be allowed, and other terms will be "encouraged", says Ms Sullivan (who has no decision-making role on style matters). The style gurus are expected to issue a verdict shortly.
This line of thinking is not to everyone's taste. After AP's declaration William Gheen, the president of an anti-immigration pressure group, Americans for Legal Immigration, said the group would adopt the term "illegal invaders" in response to what he called "political correctness on steroids". But as with the debate on immigration reform, the debate seems to be moving away from the likes of Mr Gheen. This morning the Los Angeles Times, which in darker times once referred to local Mexican workers as "ignorant peons", said it too was reconsidering its policy on "illegal immigrant". The long march of Santa Monica liberals is nearly complete.
Two authors have pulled off the balancing act in recent columns. Henry Hitchings, on the New York Times website, has written "Those irritating verbs-as-nouns". But the word "irritating" (probably written by an editor) doesn't quite get Mr Hitchings' attitude. He doesn't like "fail" and "reveal" and "take-away" as nouns, as he makes clear. But he takes a long view ("reveal" has been a noun since the 16th century). And he concedes that sometimes the nominalised verb is better than, or distinct from, the noun. (A "reveal" isn't quite a "revelation".) He sums up with this nugget:
Aesthetics will always play a part in the decisions we make about how to express ourselves — and in our assessment of other people’s expression — but sometimes we need to do things that are aesthetically unpleasant in order to achieve other effects, be they polemical or diplomatic.
A descriptive prescriptivist in this instance, then. (Mr Hitchings has, incidentally, written up the descriptive-vs-prescriptive battle in his book "The Language Wars".)
A bit of prescriptive descriptivism comes from Tom Chatfield. Having just published new book, "Netymology", he wrote "Why do tech neologisms make people angry?" for the BBC. He notes that a surge of new Latin-based words into English, like "portentious", "homicide" and "destructive", annoyed 16th-century commentators much as the glut of new technology-derived vocabulary does today. Mostly, Mr Chatfield is chronicling, not condemning. But he says "When I first heard tweet as a term, I sneered at it." He calls gamification "particularly cumbersome". He disparages friend/unfriend and follow/unfollow as "corporate coinages at their most reductive". In other words, accurately describing what is going on in language doesn't have to mean mean blindly accepting every change.
Well done to both writers, for sensible contributions to an all-to-often senseless debate.]]>
If only the Ethiopians and Italians had modern translators at their side. Treaty translation is big business today. The European Union, for example, spends an estimated €300m annually on translating between its 23 official languages. (While this is a big chunk of money, it’s less than 1% of the EU’s annual budget.) Three of those—English, French, and German—are working languages in most meetings. In reality, English (to the chagrin of the French) is most commonly used. But because each document must be faithfully recreated in each of the EU’s 23 languages, creating authentic versions can be expensive and time-consuming. Thankfully, most problems are dealt with in procès-verbal, a way to introduce technical corrections to treaties without revisiting negotiations. It might still delay matters. Last year, for example, Ireland’s ratification of an EU treaty was delayed by grammatical errors in the Irish version. There are obvious trade-offs to language equality, but the EU has calculated that the delays and costs are worth it.
The United Nations should revisit its own calculations. It has just six official and two working languages. The task of translation here in Geneva, home to most UN organs, is thus decidedly simpler. The UN’s official languages are geographically diverse—combined, native speakers of Arabic, English, French, Mandarin, Russian and Spanish number over 2.2 billion. But the two working languages are bound to tradition. The persistence of French is attributed to its history as the “language of diplomacy”. In the hallways of the New York headquarters, English is (naturally) favored, and French is preferred in Geneva. Treaties registered with the United Nations Treaty Series are always translated into French and English. Documents are always provided in French and English. This city’s Geneva Conventions, written in equally authentic French and English versions, laid part of the groundwork for the international system.
But for all its history, today's preference for French is anachronistic. At 74m native speakers, French is much smaller than languages like Hindi, Portuguese, and Japanese. To be fair, French has geographic diversity to its credit. Languages with many more speakers, like Indonesian and Bengali, are spoken mainly on their home turf. But Spanish and Arabic are geographically spread, too, and they’re also numerically great. Spanish-speaking countries, unlike many of the Arabic-speaking ones, are enthusiastic participants in international bodies. Nearly all of Latin America, for example, claims membership at the UN-affiliated International Criminal Court. The Inter-American Commission on Human Rights and the Inter-American Court of Human Rights, with their centres of gravity in Latin America, are strong and active. A contact at the UN office of the High Commissioner for Human Rights recently told me that Spanish is more commonly heard than French there, despite the office's location in French-speaking Geneva.
So what is it costing the UN to hang on to French? What would it cost the UN to add another working language, or to replace French? Because French has remained a working language of diplomacy for at least a century, French-language diplomatic education is excellent. The French translation apparatus at international organisations is well oiled. Many traditionalists in Europe hew to French, not English. Many African countries are officially Francophone (although French usage is mostly limited to elites). Important organisations, like Médecins Sans Frontières and the International Committee of the Red Cross, prefer French. There are decent reasons to keep French around, more than the obvious fear of hurting French-speakers’ feelings.
But the balance of power is different from a century ago. French is no longer representative of the international community. If the UN's budget is tightening, its choice of working languages should be more efficient. Of the official languages, Arabic, Mandarin, Russian don't make sense as successors to French. Arabic-speaking countries are comparatively listless participants in global affairs, often preferring instead to work through the introverted Arab League. Mandarin is enormous, but mostly plays a home game. Since the end of the cold war, Russian's domain has shrunk. All three use non-Latin scripts, so introducing one as a working language would require a dramatic and expensive overhaul of the UN's language apparatus. Spanish is the only logical replacement. It makes sense not only in comparison to Arabic, Mandarin and Russian, but also by itself. The UN should reward Spanish-speakers’ increased economic and social clout—and their outsized commitments to the international system—with a bigger seat at the language table.
In truth, these conversations are probably moot: as long as Geneva plays host to some of the UN’s most important organs, French isn’t going anywhere. But conversations about fairer language distributions are happening at the UN, even if mostly in private. Demographic data on the growth of Spanish show that matters will only become more urgent with time. As many international institutions mature despite limited budgets and resources, the UN system is overdue for a language reshuffle.]]>
Sign languages are fully fledged means of communication, complete with an intricate syntax. What they lack in vocal features, they make up in other ways. Take "referential shift", in which signers assign portions of space around them to refer to different objects. So, in the imagined conversation, every time you referred to one friend, you would shift your torso to the left, say, whereas turning to the right would indicate a reference to the other person in the story.
But what if you were recounting it to several signers at a time? That would probably mean that they would need to perform a fair amount of mental gymnastics in order to interpret at least some of the signs from an awkward angle.
This involves marshalling a lot of spatial information, says Ann Senghas, of Barnard College, in New York. As a result, Dr Senghas suspected, signers may possess above-average spatial-reasoning skills. To test the idea she and her colleagues Amber Martin and Jennie Pyers looked at users of Nicaraguan Sign Language (NSL), which uses referential shift explicitly.
NSL was created spontaneously in the 1980s by pupils at a school attended by many of Nicaragua's deaf children. It has since been passed on from older students to younger ones. Because it is not taught outside the one school, Dr Senghas could determine precisely how long the participants of her study were exposed to it before she tested them. She could also test whether the age at which children enrolled, and began picking it up, mattered.
To find out, the researchers presented 33 signers and 16 non-signers of different ages with a set of tasks. These included determining what an L-shaped Lego block or a doll with one arm raised would look like rotated by 120 degrees, either vertically or horizontally.
They found that NSL users were indeed better than non-singers at tasks involving spatial reasoning. And the younger their subjects were when first exposed to NSL, the better they did. In other words, it was not how many years a person has spent studying sign language that mattered, but rather how young he was when he learned it—a truth all too familiar to any adult struggling to master a foreign tongue.
Correction: In an earlier version of this article we misrepresented why NSL makes for a good controlled experiment, wrongly suggesting that it had to do with controlling for the quality of teaching and selectivity of schools. This has now been corrected. Apologies.]]>
Until now, that is. First, Tesco, a British supermarket chain, took out advertisements in national newspapers apologising for the discovery of horsemeat in their processed food. These looked like they had been written in lines of free verse, with statements wound down over separate lines:
This is it.
We are changing.
Then, a couple of weeks later, Twitter made it possible to include line-breaks in tweets. Suddenly enjambment was trending on social media.
But many poetry geeks (this Johnson included) will be quick to point out that Tesco, and many Twitter users, failed to exploit enjambment's full potential (or, for that matter, that they are very good poetry). In “The Force of Poetry”, Christopher Ricks, formerly the Oxford Professor of Poetry who is now at Boston University, writes elegantly of the way enjambment can make language seem elastic:
Lineation in verse creates units which may or may not turn out to be units of sense; the "flicker of hesitation"…as to what the unit of sense actually is—a flicker resolved when we round the corner into the next line—can create nuances which are central to the poet’s enterprise.
Mr Ricks cites some fine examples, such as John Milton's use of the device to turn an intransitive verb to a transitive one in two lines from “Paradise Lost”:
Then feed on thoughts, that involuntary move
Add rhyme and enjambment can alter the sense of a poem even more deeply. Even outside of strict metered verse, such as in the poetry of e.e. cummings, it can make poetry surprising and playful.
Tesco's stab was certainly surprising. But likening it, as some British newspapers have, to something that might be compared to a Shakespearean sonnet or analysed by a Cambridge professor is a stretch. The line endings seemed more pompous rather than playful and to be dictated more by the startling white space of the background than a clear poetic goal. To be fair, Twitter’s offerings were more fun, albeit still somewhat unsophisticated:
And the rest was...
only a line break
away from decrepitude
and loss and all that's better off
unmentioned and relegated
Mainstream media are unlikely to wax eloquent about enjambment again anytime soon. But who knows what experimentation in the social sort will bring.]]>
Translating between languages is a hurdle for all of the international courts: the pace of the courtroom can creep, even with simultaneous interpreting. The ICTY is faster than, say, its younger and bigger successor, the International Criminal Court (whose language changes with each case). This is partly because it focuses on one linguistic region. Its purview is mostly limited to three languages—the two working languages of the United Nations, English and French, and what the court terms "BCS": Bosnian/Croatian/Serbian. (For crimes committed in Kosovo and Macedonia, the court temporarily introduced Albanian and Macedonian, but English, French, and BCS form its permanent core.)
The politics of BCS is complex. Some people separate the group into Serbo-Croatian and Bosnian; others distinguish all three. Serbs use the Cyrllic alphabet; Croats and Bosnians the Roman alphabet. There are several major dialects, but these do not line up neatly along ethno-nationalist lines. The Croatian dialects are quite distinct from one another. In any case, the various dialects are all mutually intelligible, the main reason why linguists traditionally considered them a single language. The break-up of Yugoslavia has naturally led to the increased tendency for nationalists to insist that they are distinct. The court brings them all together in official documentation under the label BCS. However they are counted, however, BCS cover essentially all lawyers, judges, defendants and witnesses.
As a result, the ICTY is able to deliver high-quality, quick translations, so fast that translated courtroom exchanges proceed almost as fast as monolingual dialogue. It is a bit like watching a dubbed movie in real life, but at least it is smooth. Some hearings at the International Criminal Court (ICC) are choppier; a dubbed movie whose audio is half a minute out of sync, if you will. BCS typically follows a subject-verb-object structure, making it easier to translate into English (though unlike English it is highly inflected). At the Special Tribunal for Lebanon (STL), also in The Hague, Arabic interpreters work more slowly. Hearings are harder to follow.
Humming busily in the background is an electronic management system, E-court, which was introduced in 2006. Some of E-court's most significant changes involved language in the court. A live transcript (translated into English and French) of the proceedings can be flashed on one of the many computer screens in front of everyone in the courtroom. Parties can challenge the translation immediately, which could be crucial for a case. The translations of testimony and proceedings not in French or English must be kept in those two languages, so accuracy is paramount. E-court also allows witnesses to remain abroad and send their testimony via a video link, with simultaneous translation happening remotely. (This technology is used in the ICC, too.)
Despite BCS's copious use of diacritical marks, the ICTY does not have to deal with a totally different script, as in the ICC or the STL, both of which must work partly in Arabic now. (The ICTY must use Cyrillic to process Serbian documents, but it delivers BCS in Roman. There are brief exceptions: during the Macedonian cases, Cyrillic was used.) At the STL, Arabic proves daunting for Western lawyers tasked with leafing through thousands of pages of handwritten Arabic-language evidence. Doing language tech right isn't easy.
The ICTY's legal legacy will be debated, of course. But as the first court of its kind in 50 years, it seems at least to have dealt well with language, one of international lawyers' biggest sources of difficulty.]]>
Outrage at local authority plans to abolish apostrophe [Guardian]
Council 'murders' apostrophe in street-sign row [Scotsman. "Move condemned as 'appalling, disgusting and pointless'"]
End of the road for the misunderstood apostrophe [Times]
Residents' fury as council bans APOSTROPHES from street signs 'to avoid confusion' [Mail]
Outrage! Murder! Disgusting! Fury! End of the road! What fresh hell is this?
Few of the headline-writers seem to have read the source of the story itself, this 18-page proposed set of changes to regularise street names in Mid Devon. The section on punctuation reads, in full:
7. New street names shall not end in “s” where it can be construed as either a possessive or plural; neither shall they commence with the word “The”.
8. All punctuation, including apostrophes, shall be avoided
That's it. There are currently three streets in Mid Devon which have apostrophes in their names. It was not clear (I've sent several questions to the council) whether these three were to be renamed, or whether the changes applied only to future street namings. But the Guardian wrote that "Mid Devon District Council's plan is presumably to avoid errors such as this," showing a picture of a sign in a park reading "Childrens' Garden". No. The policy had nothing to do with such signs; it was only to avoid newly naming streets in ways that have confusing apostrophes. But never mind: the council has now reversed its decision in the wake of the national outcry.
Wasn't this just still dumbing down in a society that can no longer be bothered to teach punctuation? Wrong again. The possessive apostrophe is legitimately confusing to educated adults in proper names. How many people know that Queen's College, Oxford is named after one queen and Queens' College, Cambridge after two, leading to the different placement of the apostrophe? Even if you're a native Londoner, can you pass John Wells's test by correctly stating whether or not apostrophes appear in the following Underground station names?
Barons Court has no apostrophe, but Earl's Court does. Americans celebrate Veterans Day but Mother's Day. Presidents' Day is usually written thus, but the White House website has it as President's Day.
In other words, this is tricky. Mid-Devon never proposed scrapping punctuation. It merely proposed avoiding new street names that were likely to confuse (citing, for example, the possibility of confusing the emergency services). This had almost nothing to do with grammar, and everything to do with a few chest-beating commentators' desire to bewail the death of civilisation yet again. With Mid Devon council's meek retreat, the chest-beaters won.
Mr Wells, in calling the whole affair a "moral panic", called for scrapping the possessive apostrophe. He's right about the moral panic part, and he's even right that the possessive apostrophe could, in theory, be done without. We don't hear apostrophes in speech, and yet we can easily rely on context or a bit of extra disambiguating language to resolve any problems. If we reinvented the writing of English today, we might rethink apostrophes.
But we won't be reinventing anything, neither punctuation nor English's nightmare spelling. The reason is that there is no one to do so. There is no official English Academy, and so even good suggestions have no endowed body to put them forward or to enforce them. And yet, somehow, the language self-regulates, and continues its spread round the world. Small changes will happen over time (like the lost hyphen in "to-day" or the gradual decline of the comma splice). But they don't happen by fiat; nor should they. The fact that nearly everyone has learned the old rules is a good reason for not abandoning them, even if the old rules can be irrational. Officially imposed changes can be chaotic, as with the mess that accompanied modest changes in German spelling.
So, punctuation panickers, panic not. The apostrophe is alive and well, even in Mid Devon. And regulators, find something else to regulate. Languages don't take to it, and they get along just fine anyway. That's something we can celebrate. As Kory Stamper wrote (and the Guardian had the good sense to re-publish), "English may be a shifty whore, but she's our shifty whore."]]>
To those who don't follow sport, this sort of attention might seem silly—crazy, even. Fans might proudly agree. Indeed, the sport commentariat regularly turns to the vocabulary of mental illness to describe fans. Words that were once clinically used to describe patients before the terms became impolite—crazy, insane, mental, mad—have found a second life in the bleacher section. Even the term "fan" in its longer form, "fanatic", sounds positively rabid. When Jeremy Lin became an NBA sensation, commentators called the phenomenon "Linsanity". The more passionate the fans, the more respect and attention (however grudging) they seem to earn. For example, the Duke University Blue Devils, four times national champions, are followed by the deranged Cameron Crazies, who are reviled but legendary for frothing so copiously at their blue-painted mouths for the team. Even the tournament's nickname "March Madness" suggests its viewers are unhinged.
This trend isn't new, but it's still curious that craziness has become a virtue in the stands while real mental illness is as stigmatised as ever. The definitions of words like "crazy" and "insane" have evolved enough beyond specialised usage that it's impolite, even cruel—but still not uncommon—to use them to refer to real patients. Somehow these words didn't meet the same fate as, say, "idiot" to refer to mentally disabled people, a usage that now seems strange and archaic (as we've previously written). Nor did they become pejorative. Words once to describe mental illness have acquired positive shades--I'm mad about you, That dunk was insane, She's crazy smart, and the like—while words to describe mental disability mostly haven't. ("Retarded" is a rare exception, having both positive and very negative meanings in slang.) Perhaps that's a symptom of our relationship to each: as a general matter, people tend to take mental illness less seriously than cognitive disability.
It seems pedantic to take issue with our use of "crazy", since English is so saturated with it and its synonyms. But is it troubling? In the context of a national debate over the scant resources and attention we afford to mental illness, perhaps so. I'm not about to pick a fight over "crazy", but maybe it's something to think about. Still, for better or worse, there doesn't seem to be a righter way to describe the sort of pulsating, rabid mobs we'll see in the bleachers over the next three weeks.]]>
It was particularly striking that he chose a new name. My colleagues and I were going over the list of old names, taking bets on common names like Gregory and Leo, and wondering if we'd get a symbolic or resonant name like Innocent or Clement. A few were hoping that Pope Lando II would emerge from behind the curtains. (Yes, there really was a Pope Lando.) Instead, Jorge Bergoglio became the first pope since Pope Lando himself to choose an entirely new name. A look at the list of papal names shows that most of the one-off names are quite rare: Telesphorus and Hormisdas and such. Among the few names now in common use, but which have been used by only one pope, are Mark, Peter, Zachary—and now Francis. We need refer to him as only Pope Francis for now; with no other Francis, there's no need to call him Francis I.
Much of the attention of the press focused on the pope's home country. He is the first non-European pope, coming from Argentina. But his parents came from Italy; his was hardly the most radical geographical/ethnic choice the cardinals could have made. (That would have been an African.) But the choice of a Latin American pope of European parentage is another bit of nice bridge-building symbolism. Pope Francis addressed the adoring crowd, which chanted "Francesco! Francesco!", in comfortable, fluent Italian. It will be of even bigger import that he can address the Catholic Church's most important region, the Americas, in flawless Spanish. (Although not "accentless" Spanish; nobody is accentless, and the porteño accent of Buenos Aires is very distinctive.)
It should be no surprise if the new pope is decent at languages. He is the first Jesuit pope. The Jesuits have an egghead reputation for their passion for learning. As part of their evangelising mission, they have contributed much to the learning and transmission of languages. Johnson has already reported on Alexandre de Rhodes, the Jesuit who gave Vietnamese the modified Roman alphabet used to write that language today. Jesuits were also among the first to take native American languages seriously (the better to convert the locals). And they transmitted knowledge of Japanese, Chinese and Sanskrit to Europe. Like the Mormons, the Jesuits remain keen religious language-learners today.
This, of course, represents a gradual shift from the old, Latin-only policies of the church. The world's religions may now be usefully divided into those that expect the pious to read the sacred texts in their original languages (the Arabic of the Koran and the Hebrew of the Torah, especially); those that lean not on the scriptural language but on another liturgical language (Church Slavonic in some Orthodox Christian countries, Latin among traditional Catholics); and those who embrace any linguistic means available (most of Christianity and Buddhism, for example).
Finally, we have the second pope to Tweet. Francis will inherit his predecessor's nine Twitter accounts, in English, Spanish, Portuguese, French, Italian, Latin, Polish, German and Arabic. All nine accounts had Benedict XVI's old tweets erased and archived elsewhere. The various @pontifex accounts lay empty (reading "Sede Vacante" on the profile page) for the past few weeks. Now, they all now have just one tweet. Perhaps tellingly, it is in Latin: HABEMUS PAPAM FRANCISCUM.]]>