Language Log

Today's SMBC:

Mouseover text: "Fun fact: In the middle of every Derrida book, there are nuclear launch codes, the recipe for Coca Cola, and the location of Blackbeard's gold."

Mark Liberman's "Real tone" (2/7/18), replying to "Tones for real" (2/5/18), is a nice demonstration of what's happening in real speech.  The question for John McWhorter and all serious language teachers / learners is how much of it can be systematized and regularized?  In other words, how much of it can be taught / learned?

I will be blunt:  I don't think that the discernment and production of speech can be taught / learned at this ad hoc, microphonemic level.  That is why the very best teachers and best students I know do not focus on lexemes or morphemes, but rather on phrases, clauses, or even whole sentences.

In 'Tones for real", 2/5/2018, John McWhorter expresses his frustration as an American learner of Chinese: "How much must I attend to the damned tones in a sentence, as opposed to in citation, to really speak this language?"

As John very well knows (when he's not frustrated by the difficulties of learning a new language), his question has the same answer as the analogous question "How much must I attend to the damned consonants and vowels in a sentence, as opposed to in citation, to really speak this language?" Fluent native speakers almost never use standard citation forms in fluent speech — sometimes the fluent versions are reduced or assimilated or dissimilated versions of the citation forms, and something they're just variably different. This is partly because informal speech is variably non-standard, but mostly because of the complex effects of linguistic and communicative contexts on the phonetic realization of phonological categories.

Unfortunately for language learners, these complex effects (though in some sense "natural") are different in different languages and dialects/varieties, so you can't just use your normal phonetic habits and expect the results to sound right.  And we can use John's own pronunciation of English to illustrate some of these contextual effects.

Alex Baumans sends a link to a new album Double Negative by the band The Men That Will Not Be Blamed For Nothing.

Negative Concord would be a better title, in my opinion.

During today's episode of "Angelo Cataldi and the Morning Gang" on WIP sports talk radio, there was an interview with Doug Pederson, the coach of the Philadelphia Eagles.

One exchange caught my linguistic (as opposed to sports fan) attention:

 

I've recently noticed an uptick in spam with good graphical quality but terrible proofreading. A few random examples are below.

In "Text-as-data journalism? Highlights from a decade of SOTU speech coverage" (Online Journalism Blog 2/5/2018), Barbara Maseda surveys some of the ways that "media has used text-as-data to cover State of the Union addresses over the last decade".

When Erica Hendry asked me for thoughts about features of Donald Trump's style in last week's SOTU, the only contribution I could think to make to her article ("Trump’s language shifts from ‘I’ to ‘we’ in State of the Union address",  PBS News Hour 1/31/2018) was the thought that in a speech like that one, which the president delivered but probably didn't write, the main indications of his personal rhetorical style would be the place where what he said deviated from the RAPFD ("Remarks As Prepared for Delivery").

For several years, John McWhorter has been studying Mandarin very seriously.  He and I have, from time to time, corresponded about the best, most effective, most efficient way to do that.  After years of assiduous learning, it seems that he has recently experienced a kind of satori about one of the most challenging aspects of acquiring fluency in spoken Mandarin:  the tones.

There was a big city-wide party last night here in Philadelphia, but the Philadelphia Orchestra, got on board back in early December:

"Alexa Loses Her Voice" won USA Today's Super Bowl Ad Meter:

I believe that this was also the first Super Bowl ad to raise a technical question about speech technology.

Douglas Hofstadter has a critical article in the latest issue of The Atlantic (1/30/18):

"The Shallowness of Google Translate:  The program uses state-of-the-art AI techniques, but simple tests show that it's a long way from real understanding." (1/30/18).

Hofstadter criticizes GT for not being as good as himself at translating from French, German, and Chinese into English.  I will let others respond to his critique of the French and German translations, but I will comment on his critique of the Chinese to English translation.

Yesterday's edition of the comedy radio news quiz "Wait, wait, don't tell me" featured some discussion of the Talking Orcas story that Geoff Pullum discussed a few days ago in "Orca emits speech-like sound; reporters go insane", 1/31/2018. The whole discussion is worth a listen:

On the weekend of January 19-20, 2018, there was a Tangut Workshop at Yale University.  Organized by Valerie Hansen and sponsored by the Yale Council of East Asian Studies, this was an intense, exciting learning experience for the 35 or so people who were in the room most of the time.

Many readers may be scratching their heads and asking, "Tangut?  What's that?  And why should we at Language Log be concerned with it?"