A group of MIT Media Lab researchers have published Radiotalk, a massive corpus of talk radio audio with machine-generated transcriptions, with a total of 240,000 hours' worth of speech, marked up with machine-readable metadata. Read the rest

Conversational language is not the same as formal language: chatter over the dinner table does not follow the same rules as a speech from a podium. Informal language follows its own fluid, fast-moving rules, and most of what we know about historic informal language has been gleaned from written fragments, like old letters and diaries -- but now, the internet has produced a wealth of linguistic data on informal language, which is explored in Canadian linguist Gretchen McCulloch's new book, Because Internet: Understanding the New Rules of Language. Read the rest

These people are pretty bad at recognizing different languages. And so am I.

Bonus facts I happened to find on Ethnologue, a fascinating directory of languages: Apparently there are more than 7,000 living languages in the world. Half the world's population speaks one of the top 23 most-used languages, with the top five being Chinese, Spanish, English, Arabic, and Hindi. One-third of the 7,000 languages are endangered, "that is, loss of all individuals who continue to identify the language as being related to their identity." Read the rest

We often unconsciously mirror the behavior of people we interact with. This can include mirroring posture, gestures, and voice patterns. A recent paper in Current Biology reports that we can mirror a smile based on speech alone, and even do so without actually detecting the smile.

The researchers applied a signal processing technique for altering recorded speech under a neutral mouth position to what it would have sounded like had the speaker been smiling. They played 60 such recordings (some manipulated, some not) to 35 subjects, and asked them to judge whether the speaker was smiling. The researchers also measured the responses of two subject muscle groups while listening, the zygomatic (smiling) muscles and the corrugator (frowning) muscles.

When the subjects correctly reported neutral expression or smiling in the speech, both of their muscle groups accurately mirrored the speech while listening (e.g., for smiling speakers, zygomatic tensing and corrugator relaxing). Interestingly, even when the subjects were wrong, their zygomatic muscles still mirrored correctly. This was not true for the corrugators, which instead reflected the subjects' report.

Our mirroring capabilities go well beyond what we see, or even perceive. Read the rest

In the forests of the Amazon, West Africa, and Asia, villagers often beat on large drums to send messages miles away. While you may think that the patterns are similar to Morse Code, they're actually simplified versions of the villagers' spoken languages, "without consonants or vowels but with enough connection to the original language that speakers can reliably interpret what they mean." In newly published research, University of Cologne linguist Frank Seifart and his colleagues reveal how it's done. From Science:

All but one of the 20 or so drummed speech systems come from tonal languages, including Yoruba in Nigeria, Banda-Linda in the Central African Republic, and Chin in Myanmar. Spoken Bora has two tones, which are recreated using two different drums made from hollowed logs, called manguaré. The thinner “male” has a higher tone, and the thicker “female” has a lower one.

But tone alone isn’t enough to distinguish all the words a drummer might want to say. So Seifart and his colleagues looked at what he calls a “neglected” quality in linguistics—-rhythm...

The intervals between beats changed in length depending on the sounds that followed each vowel. If a sound segment consisted of just one vowel, the time after the beat was quite short. But if that vowel was followed by a consonant, the time after the beat went up an average of 80 milliseconds. Two vowels followed by a consonant added another 40 milliseconds. And a vowel followed by two consonants added a final 30 milliseconds.

These short durations are enough to distinguish the drummed messages for “go fishing” and “bring firewood,” which are identical in tone, but not in their ordering of consonants and vowels.

Emma Byrne, a science writer and artificial intelligence researcher, has just published a new book called Swearing is Good for You: The Amazing Science of Bad Language and it sounds fucking great. "If you ask people what they think about swearing, they tend to insist that it diminishes the speaker’s credibility and persuasiveness—-especially if the speaker is a woman," Byrne writes. But actually, a presenter's swears can sometimes make them damn more convincing. From Smithsonian:

In the book, Byrne cites one study that examined the rhetorical effects of swearing on an audience that was already sympathetic to the speaker’s message. For the study, psychologists Cory Scherer of Penn State University and Brad Sagarin from Northern Illinois University showed videotaped speeches to 88 undergraduate students. Participants listened to one of three different versions of a speech about lowering tuition rates at a university—one with no swearing, one that had a “damn” thrown in the middle, and one that opened with a “damn.” The rest of the speech was unchanged.

“The students who saw the video with the swearing at the beginning or in the middle rated the speaker as more intense, but no less credible, than the ones who saw the speech with no swearing,” Byrne summarizes in her book. “What’s more, the students who saw the videos with the swearing were significantly more in favor of lowering tuition fees after seeing the video than the students who didn’t hear the swear word.”

Byrne delineates between what she calls propositional swearing, which is deliberate and planned, and non-propositional swearing, which can happen when we’re surprised, or among friends or confidants.

Dern tootin', Reese! (Vanity Fair)

Read the rest

There's a fascinating linguistic fight brewing in Kazakhstan, due to the president's decision to adopt a new alphabet for writing their language, Kazakh.

The problem? It's got too many apostrophes!

For decades, Kazakhs have used the Cyrillic alphabet, which was imposed on them by the USSR back in the 30s. Now that Kazakhstan has started moving away from Russia -- including making Kazakh more central in education and public life -- the president decided he wanted to adopt a new alphabet, too. He wanted it based on the Latin one.

But! Kazakh has many unique sounds that can't be easily denoted using a Latin-style alphabet.

Kazakhstan's neighbors solved that problem by following the example of Turkey, where they use umlauts and phonetic symbols. But Kazkhstan's president didn't want that -- and instead has pushed for the use of tons of apostrophes instead.

Kazakhstan's linguists intellectuals think this is nuts, as the New York Times reports:

The Republic of Kazakhstan, for example, will be written in Kazakh as Qazaqstan Respy’bli’kasy.

Others complained the use of apostrophes will make it impossible to do Google searches for many Kazakh words or to create hashtags on Twitter.

“Nobody knows where he got this terrible idea from,” said Timur Kocaoglu, a professor of international relations and Turkish studies at Michigan State, who visited Kazakhstan last year. “Kazakh intellectuals are all laughing and asking: How can you read anything written like this?”

The proposed script, he said, “makes your eyes hurt.” [snip]

Under this new system, the Kazakh word for cherry will be written as s’i’i’e, and pronounced she-ee-ye.

Vulgar constructs languages for fantasy fiction or whatever other purpose you can imagine, applying consistent rules to the custom phonemes you feed it. [via]

Vulgar's output models the regularities, irregularities and quirks of real world languages; phonology, grammar, and a 2000 unique word vocabulary. Trial the demo version online. Purchase the premium version to get access to the complete 2000 word output (with derivational words) and extra grammatical rules. ...

Vulgar generates ... based on a list of some of English's most common words. However, the program is more than just a one-to-one mapping of unique outputs to English words. In an effort to mimic real world languages, Vulgar also creates various homphones and overlapping senses inspired by examples from real world languages. For example:

Here's my language:

The Language of Puput /ˈpʰupʰutʰ/

...and he stood holding his hat and turned his wet face to the wind. ...u lu bunela une luch yafa u neba luch miku peb tul ye Pronunciation: /u lu bɯˈnela ˈune løtʃ ˈjafa u ˈneba løtʃ ˈmikʰø pʰeb t̪øl je/ Narrow pronunciation: [u lu bɯˈnela ˈune løtʃ ˈjafa u ˈneba løtʃ ˈmikʰø pʰe t̪øl je] Puput structure: and he stood holding his hat and turned his wet face the wind to

Seed for this language: 0.36384689368800394

The Puput word for "stuff" is "nut." I'll spare you details of the nominative and accusative case forms, but they're there. The full edition of the app is $20. Read the rest

2016's lawsuit between Paramount and the Trekkers who crowdfunded Axanar, a big-budged fan film set in the Trekverse, continues its slog through the courts, and continues to be enlivened by the interventions of the Language Creation Society, an organization of synthetic language enthusiasts, whose amicus briefs ask the court to reject Paramount's claim of a copyright in the synthetic language of Klingon, which has many speakers, including some who learned it as their first language. Read the rest

Last month, I wrote about Paramount's lawsuit against Axanar, a crowdfunded Star Trek fan-film. Read the rest

The BBC's in-house linguists have an interesting piece about pronouncing words in Xhosa — a major language spoken in the region of South Africa where Nelson Mandela grew up. (Helpfully included in the story: How to pronounce "Xhosa".) Read the rest

To linguists, the central feature of Valleygirl Dialect is the tendency to make a statement sound like a question. For decades, this has been considered not just part of Valleyspeak, but part of female speech. That's changing. Like, dudes are totally doing it, too. Read the rest

If you've got 10 minutes, you can learn the history of English — including some interesting background on where specific words and phrases came from. (If you don't have 10 minutes, you can also watch the whole thing one chapter at a time in less-than-two-minute segments.) Interesting note: The equal importance of both The King James Bible and early scientific publications/societies to the formation of English as we speak it today.

Video Link Read the rest

Schwa Fire is would-be magazine that hopes to publish long-form journalism about the science and sociology behind the way we talk to each other. It sounds like it has the potential to be totally awesome, melding great storytelling with a field — linguistics — that doesn't get nearly the attention it deserves. You can help fund the magazine through a Kickstarter. Check it out! Read the rest

English — along with a whole host of languages spoken in Europe, India, and the Middle East — can be traced back to an ancient language that scholars call Proto Indo-European. Now, for all intents and purposes, Proto Indo-European is an imaginary language. Sort of. It's not like Klingon or anything. It is reasonable to believe it once existed. But nobody every wrote it down so we don't know exactly what "it" really was. Instead, what we know is that there are hundreds of languages that share similarities in syntax and vocabulary, suggesting that they all evolved from a common ancestor.

Of course, that very quickly leads to attempts to reconstruct what said ancestral language might have sounded like. In the track above, you can listen to University of Kentucky linguist Andrew Byrd recite a fable in reconstructed Proto Indo-European. Archaeology magazine helpfully provides a translation: Read the rest

Max Barry's new technothriller Lexicon is a gripping conspiracy novel about a cabal of "poets" who have mastered the deep language of the human brain and can use it to boss the rest of us around. It's a pitch-perfect thriller, a jetpack of a plot that rocketed me from page one to page 400 in a single afternoon, and it kept me guessing right up to the end. Imagine Dan Brown written by someone a lot smarter and better at characterization and at hand-waving the places where the science shades into science fiction, and you've got something like Lexicon.

In particular, Lexicon captures a lot of the stuff that makes the myth of Neurolinguistic Programming so compelling -- the idea that smart people can figure out how to make others march in lockstep just by tricking their subconsciouses into thinking that that's what they wanted to do all along. And Barry carries through the power-fantasy to its inescapable end: a secretive, paranoid, power-maddened cabal that is its own worst enemy.

Full of surprises and grace notes, this is the kind of delightful thriller that's anything but a guilty pleasure, and just what you'd expect from the author of such great books as Jennifer Government and Machine Man.

Lexicon Read the rest