A different kind of 'sound change' research at Wisconsin

Our local Wisconsin Week paper — the university weekly just ran a piece about some work on campus by Christian Stilp and Keith Kluender in Psychology. It appeared in PNAS recently(abstract here) and though I'd seen the piece, I suddenly realized we hadn't called attention to it here.

For the readers into speech sounds, this work aims to provide some underpinning for the notion of sonority and it also, I think, more generally focuses on a really dynamic approach to understanding sounds, how we perceive in terms of change. The image below (from the paper) will fill in some detail on what's in the abstract. Pretty cool.


I'm heading out the door but this is something that we should have a meaty post about ...

Voicing, with tongue in cheek

There's too much to blog about these days,* so this probably counts as randomness: Over at wonkette, as part of a snarko-screed called "Lo! The Wretched Ancient Saga Of Liz Cheney",we find this description of the voiced sibilant /z/ in English:

According to several popular children’s ghost stories books, centuries ago Dick Cheney created another in the image of himself. He called it “Liz,” for he liked to draw out the zzzz and allow the vibrations produced by the humans’ language tickle his tongue-organ.

Wow, pretty poetic description of vocal fold vibration, there, Wonkette.

*For instance. this piece on standardized testing. Or this remembrance of Safire by John McWhorter.

Speech synthesis and the Simpsons

OK, here's the rare case where I'm unwilling to assume that the writers for the Simpsons had full knowledge of just how funny something was. I was talking about speech sounds yesterday in class and paraphrased this opening bit from Nittrouer and Lowenstein's 2008 paper in the Journal of the Acoustical Society of America:

Perhaps the trouble all started in 1944 when Frank Cooper and Al Liberman decided to build a reading machine for the blind. At that time they adopted what Liberman would later call the “horizontal view” in his book, Speech: A Special Code 1996 . According to this view separate segments are aligned in the speech signal in a linear fashion, strictly auditory perceptual processes recover the acoustic character of each segment, and cognitive processes then translate those acoustic descriptors into phonemic units, void of physical attributes. Assuming this much about the acoustic speech signal, Cooper and Liberman turned their attention to what they saw as the truly difficult problem: optically isolating the letters on the page that would need to be converted into acoustic segments. But their own experiments soon revealed the intractable problem that listeners are unable to recognize separate acoustic elements presented at a rate replicating typical speech production.

A student started grinning immediately. When we reached an appropriate pause, I asked what was up and he said that there was a Simpsons' episode ("Smart and Smarter", it turns out) where Maggie gets a toy. Here's a key passage (from here):

Homer: Look what they sent over. A talking dealy. His name is Phonics Frog.
(Homer presses A, B, and C)
Phonics Frog: Ah-Buh-Cuh…
(Homer types his name)
Phonics Frog: Huh-Oh-Muh-Eh-Ur
Homer: That's me! Huh-Oh-Muh-Eh-Ur.

Of course, HeiDeas covered this (here), quite reasonably under the rubric of 'hooked on phonics, but the thought of 1940s speech science matches the Simpsons is nice. Just don't let me get started on the South Park "Hooked on Monkey Fonics" episode.

The edges of alliteration

In yesterday's column, Safire continues his recent language play — trope-a-dope might be the best example — but actually gave a datum that I'd been half wondering about. In discussing media names for the fiscal catastrophes of late, he notes:

The Guardian, turned to tried-and-true alliteration in dating days of dreary drops, with “Meltdown Monday,” which came along “nearly every other week nowadays, along with ‘Frantic Friday’ and ‘Tsunami Tuesday’ and ‘black’ any old day.”

Then he goes into Manic Monday, sadly without reference to the great 80's song. (Raise your hand if you remember the Bangles.)

When I'd first heard and read Tsunami Tuesday, I definitely did not think of it as alliterative. In fact, it struck me as not quite working. For most Americans, I imagine it's [su:] + [tu:] or [tju:]. For those who get closest to pure alliteration, people who pronounce the affricate of the original form, it's presumably [tsu:] + [tu:]. (That is, I'm assuming nobody says [tu:] + [tu:] here, certainly not Safire.)

Does that feel like alliteration to English speakers? There aren't many parallels for those who use the affricate, maybe tsetse flies teeming or something.

Image from this wonderful website, How Stuff Works.