Easing in with one of my interests: language. Below an article on accents, which I copy in full, as the South China Morning Post requires you to log in:
Say what? Online archive holds clues to world accents |
An online archive of the way different nationalities speak English is giving a better understanding of accents |
| "Please call Stella. Ask her to bring these things with her from the store: six spoons of fresh snow peas, five thick slabs of blue cheese, and maybe a snack for her brother Bob. We also need a small plastic snake and a big toy frog for the kids. She can scoop these things into three red bags, and we will go meet her Wednesday at the train station." You can hear these words recited 1,300 times at the online speech accent archive at George Mason University in Fairfax, Virginia - and each time it is different. The archive was set up to exhibit "a large set of speech accents from a variety of language backgrounds". Native and non-native English speakers are recorded - or record themselves - reading the passage, chosen because it contains most of the consonants, vowels and clusters of Standard American English. Steven Weinberger, associate professor and director of linguistics at the university, is the administrator of the archive, which has been on the Web since 1999. Anyone can submit a sample. "We get them all of the time, from all over the globe," he says. "We ask for CD-quality recordings, and we get some very good ones. We also get noisy, bad quality ones, which we discard. If the recording is good, and the data confirmed, we add it to the archive." Contributors answer questions about their demographic and linguistic backgrounds to enable archive users "to determine which variables are key predictors of each accent". If you are, say, a 56-year-old male, native English-speaker from the north of England and the archive already has a sample matching that, you will not be rejected. "All speakers are slightly different from each other," Weinberger says. "If we have 75 Spanish speakers, there are a number of different countries represented. The same goes for Arabic. "There are loads of gaps, mainly in the less common languages such as those spoken on small islands [Tahitian, Balangingi, etc] or Native American languages. "I find the older speakers most instructive. Many of our speakers who are older than 70 seem to have the most 'archetypal' accents." You can search the archive by language or geography, or just enjoy a browse; alongside each recording is a phonetic transcription. So, for example, you can compare the accent of a female native Afrikaans speaker aged 27, who learned to speak English at nine, with a 43-year-old man, from a different region of South Africa, who learned English at four; or you can hear accents of native Arabic speakers from Egypt, Israel, Iraq or Syria. Crucial to understanding accents is the fact that they are "systematic rather than mistaken speech", Weinberger says. This can counter "biased social judgments" based on accents. "When we understand that accents are not due to `errors' or faulty learning, we may be more sympathetic to the speakers. But biases are hard to unlearn." So how and when do we acquire our accent? "We have an automatic ability to listen to a snippet of speech and quickly determine whether that speaker is from our community: five-year-olds can do this. "When it comes to non-native speakers, there seems to be something that all French speakers share, all Mandarin speakers share, etc. French speakers of English substitute, alter, delete and add sounds to their English, making it different from that of a native English speaker. When we distil what they do, we see patterns: French speakers sound French because they use French sounds and structures," Weinberger says. "It is systematic. Most French speakers of English can be shown to do these specific [French] things, Swahili speakers to do Swahili things, and so on. So what we hear in an accent is the grammar system from the talker's native language," he says. But "there are still other things about accents that may be more idiosyncratic. "Everyone has an accent, we are wired to have an accent. Most linguists believe there is a critical period when humans can acquire a language perfectly. After this age [around six years old] humans learn a language incompletely. This shows up most often in the pronunciation. "The archive generally confirms this notion of a critical period. It is only the very young learners who pass for native speakers. So a Korean who starts learning English at age 11, and lives in the US for 20 years speaking English, will still have a Korean accent. But a Korean who starts English at four, and moves to the US and lives there for five years, will not have a Korean accent." As well as linguists and phoneticians, users of the archive range from teachers of English as a foreign language, and makers of speech recognition machines, to speech pathologists and actors who need to learn an accent. Does Weinberger get sick of hearing about snow peas and blue cheese? "Sure, and so do my family and students. But some like it - just Google `please call Stella ...' People have used it for all sorts of things." There are ringtones, it's on YouTube, and composer Cathal Roche has written saxophone pieces based on it. "There have been academic papers and theses based on the data too, but I think most people just like to listen to accented speech. I am sure there are plenty of drinking games based on the archive!" Had the archive been around at the time, we might have been spared Dick Van Dyke and Sean Connery, who topped Empire magazine's poll of the worst accents in cinema history for Mary Poppins (1964) and The Untouchables (1987). "We get thanks from actors who are working with scripts that require obscure accents," Weinberger says. "We are getting ready for better maps via Google, more searchable sounds, and phonetic inventories. We are also putting together a database of the syllable structures available in the world's languages. "We have also made a device that will automatically compare two accents and reveal the specific speech patterns that make one accent different from another. It will be available, free, to anyone." In the meantime, he carries a voice recorder with him all the time. "I never know when I will encounter an interesting accent." Guardian News & Media |
