『要找到新的研究課題，一些語言學家只要開闊研究領域，從記錄瀕臨消失的語言開始�！�

　　Say what?

　　人類語言

　　Sep 10th 2011| from The Economist

　　WHERE in the world is the largest number of different languages spoken? Most linguists would probably plump for New Guinea, an island that has 830 recognised tongues scattered around its isolated, jungle-covered valleys. But a place on the other side of the world runs it close. The five boroughs of New York City are reckoned to be home to speakers of around 800 languages, many of them close to extinction.

　　New York is also home, of course, to a lot of academic linguists, and three of them have got together to create an organisation called the Endangered Language Alliance (ELA), which is ferreting out speakers of unusual tongues from the city’s huddled immigrant masses. The ELA, which was set up last year by Daniel Kaufman, Juliette Blevins and Bob Holman, has worked in detail on 12 languages since its inception. It has codified their grammars, their pronunciations and their word-formation patterns, as well as their songs and legends.

　　Each volunteer speaker of a language of interest is first tested with what is known as a Swadesh list. This is a set of 207 high-frequency, slow-to-change words such as parts of the body, colours and basic verbs like eat, drink, sleep and kill. The Swadesh list is intended to ascertain an individual’s fluency before he is taken on. Once he has been accepted, Dr Kaufman and his colleagues start chipping away at the language’s phonology (the sounds of which it is composed) and its syntax (how its meaning is changed by the order of words and phrases). This sort of analysis is the bread and butter of linguistics.

　　Every so often, though, the researchers come across a bit of jam. The Mahongwe word manono, for example, means “I like” when spoken soft and flat, and “I don’t like” when the first syllable is a tad sharper in tone. Similarly, mbaza could be either “chest” or “council house”. In both cases, the two words are nearly indistinguishable to an English speaker, but yield starkly different patterns when run through a spectrograph. Manono is a particular linguistic oddity, since it uses only tone to differentiate an affirmative from a negative—a phenomenon the ELA has since discovered applies to all verbs in Mahongwe.

　　Such niceties are interesting to experts. But the ELA is attempting to understand more than just the nuts and bolts of languages. It is collecting stories and other verbal material specific to the cultures of the participants. One volunteer, for example, wants to write a storybook for children in her language (Shughni), and also a recipe book. That means creating a written form of the language, which the researchers do using what is known as the International Phonetic Alphabet.

　　Many of Dr Kaufman’s better finds, he says, have come from “hanging out at street corners with a clipboard on Roosevelt Avenue”—a street in the borough of Queens that he describes as the “epicentre of the epicentre” of linguistic New York. How long it will remain so is moot. The world’s languages, which number about 6,900, are reckoned to be dying out at the rate of one a fortnight. The reason is precisely the sort of cultural mixing that New York epitomises. The value of learning any particular language is increased by the number of people who already speak it. Conversely, the value of a minority language is diminished as people abandon it. To those languages that hath, in other words, shall be given. From those that hath not, shall the last speakers soon be taken away. (587 words)