Language Log

Email from Ann Syrdal on behalf of the S3-WG91 Standards Working Group:

The "Text-to-Speech Synthesis Technology" ASA Standards working group (S3-WG91) is conducting a web-based test that applies the method it will be proposing as an ANSI standard for evaluating TTS intelligibility.  It is an open-response test ("type what you hear"). The test uses syntactically correct but semantically meaningless sentences, Semantically Unpredictable Sentences (SUS).

To take the test, click here.

Back in September of 2008, a Seattle-based start-up named SpinSpotter offered a tool that promised to detect "spin" or "bias" in news stories. The press release about the "Spinoculars" browser toolbar was persuasive enough to generate credulous and positive stories at the New York Times and at Business Week. But ironically, these very stories immediately set off BS detectors at Headsup: The Blog ("The King's Camelopard, or …", 9/8/2008) and at Language Log  ("Dumb mag buys grammar goof spin spot fraud", 9/10/2008), and subsequent investigation verified that there was essentially nothing behind the curtain ("SpinSpotter unspun", 9/10/2008). SpinSpotter was either a joke, a fraud, or a runaway piece of "demoware" meant to create enough buzz to attract some venture funding. Within six months, SpinSpotter was an ex-venture.

An article in yesterday's Nieman Journalism Lab (Andrew Phelps, "Bull beware: Truth goggles sniff out suspicious sentences in news", 11/22/2011) illustrates the same kind of breathless journalistic credulity ("A graduate student at the MIT Media Lab is writing software that can highlight false claims in articles, just like spell check.")  But the factual background in this case involves weaker claims (a thesis proposal, rather than a product release) that are more likely to be workable (matching news-story fragments against fact-checking database entries, rather than recognizing phrases that involve things like "disregarded context" and "selective disclosure").

My colleague Geoff Pullum has objected to the selection of squeezed middle as Oxford Dictionaries' 2011 Word of the Year on the grounds that "the 'Word of the Year' should be a word." Allow me to provide a counterpoint to this view.

The Oxford Dictionaries organization (responsible for marketing the Oxford English Dictionary and its many spinoffs and abridgments) picks a word at the end of each year that they think epitomizes the main currents of what happened in the world (or the anglophone parts of it). Or to be more accurate, they pick either a word or a phrase. And two years running they have picked phrases. I want to argue that this is a mistake, not just because they have chosen an utterly undistinguished item, but because what they have chosen is a straightforwardly compositional phrase, one that couldn't be argued to be a lexical item at all.

The lower-grade newspapers in Britain this morning have been making much of what happened to a group of birdwatchers, gathered excitedly in a coastal area for a rare chance to photograph a Hume's leaf warbler. It seems they happened upon a calendar photo shoot and had a rare chance to also snap a blonde model, draped over a motorcycle, wearing nothing but a thong.

While we're talking about the politics of language peevers, I can't resist sharing with you the opening of Time Magazine's 1946 review of E.B. White's The Wild Flag:

E. B. White plugs federal world government with the dazed urgency of an Esperanto salesman. He has the same high purpose, the same rosy vision, the same conviction that all it needs is a try.

I applaud Mark for taking on the question of left- and right-wing linguistic moralism. It encourages me to add some snippets from the disorganized drawer of Thoughts I have on this topic, some of them from stuff I wrote but never published. I leave the insertion of transitions as an exercise for the reader.

In the first place, doesn't make sense to think of this question other than historically. The distinction between "prescriptivism" and "descriptivism" is a twentieth-century invention, and an unfortunate one, I think, since it implies that this is a coherent philosophical controversy with antique roots. In fact both terms are so vague and internally inconsistent that we'd be better off discarding them, and to impose those categories on the eighteenth-century grammarians, say, is gross presentism. So let me talk about "language criticism," both because it's closer to the mark, and because what linguists describe as "prescriptivism" in most of the Western languages is by-and-large just a stream of the critical tradition. (Language criticism, it has struck me, is the dream-work of culture.) And the politics of both have always been in flux.

In a comment on yesterday's post on "Momentarily", Alan asked

Is there any difference between the language peeves of left-wing authoritarian moralists and right-wing authoritarian moralists? Do they tend to peeve about different kinds of usage?

I don't have a large enough sample to make confident generalizations, but my impression is that peevers across the political spectrum have similar "crotchets and irks" (to use right-wing peever James Kilpatrick's felicitous phrase). The only things I've seen that seem likely to be systematic differences lie in the area of blame.

Wednesday, Thursday and Friday of last week I was at the University of Maryland, giving the first edition of the newly-endowed Baggett Lectures. The first of the three lectures was on The Linguistic Culture Wars, and most of its content will be familiar to regular LL readers. But in the course of preparing it, I found a few new things that may be of interest. For example, I decided to use the now-available web resources to look for the origins of momentarily in the sense "at any moment; in a moment; soon". This came up because I quoted Dick Cavett's NYT Opinionator column "It's Only Language", 2/4/2007, as an instance of left-wing authoritarian moralist peeving.

Encountering the headline "Whip rules furore claims first victim" on the Guardian's front page, Ian Preston (who has plenty of experience with British Headlinese) confesses to interpretive problems:

At first I thought a government parliamentary official (a "whip") had issued a ruling either regarding a victim of claims about a furore or decreeing that a furore had claimed a victim.  Neither turns out to be the case.  It is a story about horse racing and a controversy regarding rules about (non-metaphorical) whipping has led to a resignation.  I think the problem is that "whip", "rules" and "claims" are all words which could be either nouns or verbs – in fact, it is not until "claims" that you reach a verb here but that's not immediately obvious.

"Wikipedia woos India with local languages", Hindustan Times, 11/19/2011:

Adding an article on a local Maharashtrian delicacy or making changes to an existing page on ‘Misal Pav’ on the Marathi Wikipedia could now earn you the title of the ‘Global Wikipedian of the Year’. The Wikimedia Foundation has instituted an annual award for regular contributors and editors of the online community resource. “Regular contributions need to be acknowledged. From this year, the foundation will honour one Wikipedian for his effort,” said Jimmy Wales, the site’s founder, who addressed 700 Wikipedians on the first day of the WikiConference 2011 in [Mumbai] on Friday.

This year’s award will go to a Wikipedian from Kazakhstan, most prolific amongst the contributors of his community. In 2011 the number of editors making five edits on the Kazakh Wikipage per day rose from 15 to 231, said Wales. The winner will get a direct entry to the next WikiConference with a fully paid ‘delegate pass’.

Forget Watson. Forget Siri. Forget even Twitterology in the New York Times (though make sure to read Ben Zimmer's article first). You know natural language processing has really hit the big time when it's featured in a story in Entertainment Weekly.

The government of Canada, along with no doubt many others, frowns upon companies making health claims for which they have no evidence. This is supposed to nip in the bud deceptive practices like those exhibited in this pre-regulation 1652 handbill proclaiming the "vertues of coffee drink", in which the advertisement's author touted coffee as a prevention and cure for everything ranging from miscarriage to gout to "hypochondriack winds", whatever those may be. In that document, the claims were overt and brazen, with statements such as:

"It is excellent to prevent and cure the Dropsy, Gout and Scurvy."
"It is very good to prevent Mis-Carryings in Child-Bearing Women."

Yup, those are claims.

But in a recent case that's made headlines here in Canada, the Canadian Food Inspection Agency has determined that the names of two brands of infant formula made by Enfamil, A+ and Gentlease A+, also amount to claims, the former constituting a claim about nutritional superiority to other brands, and the latter an additional claim about ease of digestibility.

Which begs the question: What counts as a claim?