Blog of Life-Affirming Anecdotes: Ben Munson

Intonational Meaning, 'Uptalk', and The Lynda Barry Experience

In the inaugural entry of this blog, I bragged that I would post material that would supplement my classroom teaching. That's an easy claim to make at the beginning of the summer, when days and days and days stretch ahead in which, ideally, I can spend hours writing and developing course materials.

Well, summer 2005 just didn't work that way. In fact, I have never spent a busier summer, and had less to show at the end, than summer 2005. What did I do? It seemed like the days were as relentlessly busy as the busiest academic semester, but I only wrote one new manuscript. (Two if you count the manuscript that Sarah Jefferson, Elizabeth McDonald, and I submitted in May, and three if you count the chapter that Jan Edwards and I began writing in August.) In summer 2003 I think I wrote three new manuscripts (real ones, not book chapters), and that summer felt very relaxing and carefree. I wrote three in summer 2004, too. (Actually, more like 4 in 2004, though one of them was a review article of which Mary Beckman was the first author.) Of course, there were grant applications in summer 2005, but HOLY MOE, that shouldn't take all summer, should it?

Whether or not it should, it did. I'm pleased to report that the grants were submitted, and I was pretty pleased with them. Who knows their ultimate fate, but I'm not ashamed of them.

The summer is long over, and I'm already dealing with the disappointment of not achieving one-quarter of what I hoped to achieve during the academic semester. I am over halfway through teaching phonetics (yup, at 8 am, and on Thursday I have the pleasure once again of being in 170 Tate Lab of Physics, home of rude physics students busting in mid-class because they can't read Noelle Padgitt's door signs telling them not to do otherwise), and this first blog of the fall semester is intended to supplement the phonetics class. It is also written for a general audience, since my boundless ego imagines that there are people who read my blog faithfully, even though I have only had handful of postings, and only one of them had any real original content in it, and the theme threaded throughout seems to be "I have a cat."

The topic of the blog is intonational meaning. Pedantic rant starts here.

Speech communicates multiple types of information in parallel. Take a single utterance, say "I could be a mud doctor, checking out the Earth, underneath." (Highbrow cineastes will recognize this as a quote from the 1978 movie "Days of Heaven.") One type of meaning that this utterance conveys is the kind of meaning that formal semanticists would show using quantifiers and brackets and Boolean operators and things that I never learned, even with an entire quarter-long class on mathematical linguistics from Ed Keenan. It would look something like "there is a possibility, such that for this possibility there exists a set of potential outcomes of being a doctor of the variety of …" In short, this would be the meaning that you could garner from the statement by reading it without any context: the person acknowledges that there is the possibility that she could be a mud doctor (whatever that is), and that this mud doctor would do some sort of quasi-medical exam on dirt.

That's one type of meaning. But, of course, spoken language typically occurs in the context of social interaction. One of the things that a spoken utterance conveys is information on how the talker intends the listener to interpret the utterance in the context of the ongoing communication. That is, when we speak in social interactions, we guide our conversational partner to interpret the message we are saying. One way that we do this is through intonation. Intonation is far more complicated than its textbook and dictionary definitions would lead us to believe. More than being merely a sing-song quality of the voice showing the difference between questions and statements, intonation is a complex, subtle system of changes in things like pitch and duration that talkers use to guide listeners to specific interpretations of spoken utterances as they relate to the broader discourse.

One of the ways we use intonation is to clue listeners on whether or not what we are saying should be interpreted with respect to information that is coming later in the discourse. This makes perfect sense. You want your listener to know how they should parse the discourse: should they tie things up, or should they wait for more? Historically, people have done this by raising the pitch of their voice at the end of an utterance (or, in an utterance that consists of many smaller chunks, at the end of the phrase-internal chunks, which we can call intermediate phrases). We can call these continuation rises.

The idea that continuation rises are used by talkers to indicate that the listener should interpret information with respect to upcoming information in discourse can be seen in a really great example that I found on a CD called "The Lynda Barry Experience." I have already blogged about Lynda Barry. She is an author and cartoonist. She lives in rural Wisconsin, having grown up in Beacon Hill, a neighborhood in South Seattle currently home to Richard Wright (or, as Kevin Burk calls him "Richard King"). The Lynda Barry Experience is a spoken-word CD (actually, I first bought it on cassette), of stories about her childhood, all true—except for the ones that are big, honking lies (so she says). One of the stories (actually, the first on the CD) talks about her experience in middle school learning about female hygiene. The sentence that is relevant is repeated below.

Our ugly teacher, Miss Burnside, who wore a fall and a miniskirt and played Where Have All the Flowers Gone? on her guitar while sitting on a stool so we could see up to where she stopped shaving her legs which gave me nightmares, passed out the little special day booklets to put in our underwear drawers at home.

You can listen to the sound file here. Download file

You can also see an acoustic representation of the pitch of the voice in the following screen shots from the Praat signal-processing program. Praat is a wonderful—nay, super-wonderful—acoustic analysis and synthesis tool that Paul Boersma and David Weenink have developed for the community of laboratory phonologists, experimental phoneticians, speech scientists, and speech industry types. These Praat print-outs show the waveform, and the spectrogram (the black and white blobs) overlaid with a blue line showing the fundamental frequency of the voice. Fundamental frequency is related to our perception of pitch, though the actual sensation of pitch is the result of a complex interaction between the signal itself and the various warpings that the ear and brain do to it. I have done a VERY ROUGH segmentation of individual words, where possible, and tried to divide the utterance into intermediate phrases. When the intermediate phrases end in a H%, that means a rise at the end of the chunk of speech, and L% means a fall at the end of a chunk. Take a look at the pitch tracks.

As you can see (and hear in the audio file), the pitch rises at the end of each intermediate phrases in the noun phrase until Lynda reaches the point where she's ready to say the verb phrase. This jibes with the traditional description that pitch rises are ways that talkers tacitly convince listeners to wait for information coming up in the discourse.

This intonational pattern has gotten a lot of press lately, and has been given the label 'uptalk'. The Lynda Barry example is a classic example of the form of uptalk that doesn't appear to really upset people terribly much. She is using phrase-internal rises to hold your attention and to show you that she was on a topic that wasn't over. She wasn't done describing the subject of her noun phrase (the ugly teacher Miss Burnside) until she finished telling you about the hairy legs and the nightmares they caused. You can buy the Lynda Barry Experience at Marlys Magaine, a great Lynda Barry fan website. I highly recommend it. (In addition to giving some great examples of uptalk, you get to learn about the Aswang, and hear some great examples of Lynda imitating her grandmother's Tagalog-accented English.)

The 'problem' is that people appear to have generalized this pattern so that it now occurs in the places where people low boundary tone--a fall in pitch--used to occur. Pitch rises are called high-rise terminals when they occur at the end of phrases. The following E-mail from Noelle (Poelle) illustrates this:

Oh yeah, I meant to tell you something about the Dar Williams concert. She's all into local community action and everything, and the concert was partially a fund raiser for Second Harvest. So anyway, at intermission this guy came out to give a pitch for second harvest, and he was speaking with a continuation rise at the end of his [intermediate] phrases. So he said something like, "we've raised over $5000, which provides meals for 100,000 people" --and there was this awkward silence. He was expecting everyone to clap, but since he used a [high-rise terminal], the audience was waiting for more. It was kind of interesting.

It is this generalization of phrase-medial pitch rises to phase-final position that seems to be upsetting people so much, because of exactly the reasons that Noelle was getting at in her E-mail. People used to use pitch rises to cue listeners to hold off on making processing decisions about a string of phrases. Now they're using rises at the end of utterances, when listeners are meant to be making processing decisions. This, not surprisingly, causes a little confusion.

One likely possibility is that this phenomenon represents a sound change in progress. Sounds change over time--we don't speak Elizabethan English, after all--and the sentence-final pitch rises that people call uptalk might just be an intonational sound change in progress. The origins of the sound change leading to uptalk being more common are unclear. Like most sound changes, it was likely to have been initiated by young women. One can easily concoct a folk-linguistic explanation for it. Perhaps young women noticed that the continuation rises were effective in maintaining their conversational partners' attention, so they generalized using them to where they were not used by others. This is hardly a testable explanation. A more likely scenario is that it may have arisen more spontaneously as a way of conveying social-group membership among a group of 'stylistic icons' in a speech community. Though I am only a sociolinguist by avocation and not by vocation, this is a likely scenario for many sound changes. Of course, the world of sociolinguistics is not without its different camps, with those proposing that sound changes are driven more by structural factors, and those seeing it as process that cannot be understood separate from the sociocultural causes and consequences of linguistic variation. I will remain agnostic, and point you to Penny Eckert's website and Bill Labov for those who want original-source materials.

A further complication comes because utterance-final pitch rises are also used in yes-no questions. Consequently, some people—primarily people trying to sell their voice-coach services on the web—have interpreted the 'uptalk' pitch rises as a string of yes-no questions, and have claimed that it is reaching "epidemic proportion" (Yipes! Sounds like bird flu!) and will almost certainly lead to people failing in business because it belies a lack of inner fortitude. Or something like that.

Well, this strikes me as a real straw man, and I don't think that claims like those in the previous paragraph really warrant a serious response. Listen again to Lynda Barry's example above. The pitch rises that you hear in the early part of that utterance don't sound like questions. Try saying the sentence yourself as if each intermediate phrase ("our ugly teacher," "Miss Burnside," etc.) were a question. Clearly, this would sound considerably different from the way Lynda actually says the sentence. The most-stressed syllable in each of the intermediate phrases (The "tea" in "teacher," the "Burn" in "Burnside") would have a much lower pitch than they do Lynda's actual articulation of them.

Moreover, I want to point out that there are a few pieces of evidence that uptalk is not the same things as a yes-no question. There is acoustic evidence that the kind of final rises that you see in "uptalk" are different from those used in yes-no questions. You can find some of this in a conference paper by Janet Fletcher, available on-line if you click here. Fletcher and colleagues' work is on Australian English, where 'uptalk' is purportedly more frequent than it is in American English. In some analyses it is the default intonation contour when individuals are presenting new information. Put simply, the pitch properties of the rises that are characteristic of the 'uptalk' that people find so offensive are actually measurably different from the yes-no questions that people seem to associate with uptalk. I did a fairly cursory search for references on uptalk in American English, and couldn't find any. Amalia Arvaniti reports an in-preparation manuscript on this topic in an in-press paper to be published in Papers in Laboratory Phonology 9, clickable link here. (Amalia's paper, co-written by native Minnesotan Gina Garding, gives a very kind acknowledgment to me, though I can't in truth remember offering anything to this paper other than a few questions at the Labphon 9 conference. I will also have a sole-authored paper in that volume, and a coauthored paper with Beckman and Edwards, both of which were written during bucolic summer 2004.) Moreover, not all questions are produced with utterance-final pitch rises. Wh-questions are produced with a pitch fall similar to those seen in neutral declarative utterances, yet they are clearly, unambiguously questions.

A few recent stories have cast uptalk in a fairer light. See one of them at the Guardian (UK): Click here There was a good discussion of uptalk in the recent "Do you Speak American?" series on PBS.
If you are interested in learning more about intonation meaning, then I strongly encourage you to read Pierrehumbert and Hirschberg's 1990 chapter "The meaning of intonation contours in the interpretation of discourse." It's an information-rich chapter, and rather heavy going for the novice. Still, who doesn't love a challenge? As far as intonational meaning as it relates specifically to "uptalk," I suppose Arvaniti's in-preparation paper will have something, though I suspect that it will focus on the structure of the intonational contours, rather than on their meaning.

In the meantime, for misinformation about uptalk, simply surf the web.

Thanks to Google, the Creepy Cat Persona Lives on

The World Wide Web is a network of such vast interconnectedness that nobody (except maybe Al Gore) could have imagined. It is helped spur academic research on the nature of networks. I would pass along a reference that Dave Pisoni gave to me, Linked by a guy named Barbarasi. It's good reading, and has implications for the kinds of network that we pose the mental lexicon to be. Mike Vitevitch has been doing work on that--look for it in print soon, I hope.

All I know is that today the web is more like the literal web spun by a spider, perhaps a black widow. Google the name "Ben Munson" (as of today, November 6, 2005) and what shows up?

Ben Munson's Blog
Ben Munson's Blog ...Phonological development and disorders, ... -Ben Munson.
PS Look at little woogums helping her daddy write his blog! She's sho shweet. ...
blog.lib.umn.edu/munso005/httpbloglibumneduMunsonblog/ - 38k - Cached - Similar pages

That's right! I'm number one, here, folks, and the quote that accompanies my name is "Look at little woogums helping her daddy write his blog! She's sho shweet."

Beautiful.

I need to work on that uptalk entry.

The Mystery of the Appearing Water Bottles

I always wanted one of those fancy Nalgene bottles, but they charge an arm and leg for them. So it was incredibly good fortune when I was teaching in 133 Tate Lab in Fall '04 and a student in the class the night before left one there. The lesson we learned in grade school still applies as adults: finder—keepers; losers—weepers. I don't know if the poor sap who left her/his bottle was weeping, but I was keeping. I took it home, swirled bleach around in it, and boiled it for two hours to get rid of any residual germs and bad karma. Then, I put it in the dishwasher for good measure. Unfortunately, I put it on the bottom shelf, and during the heated dry cycle, the cap melted slightly—so much for taking it on my trip to Mount Everest. Still, who am I to complain? I didn't pay a red cent for the thing. For over a year, it served me well. Then, earlier this week, I noticed it had gone missing. Meh. Another grade school lesson: easy come, easy go. Of course, I wasn't about to go out and spend my money on one of those things. If there's one thing I learned from my dad (other than how to conduct DNA electrophoresis for various high school science projects examining the influence of household chemicals on DNA) it's that a real man is cheap, cheap, cheap. I'm nothing if not a real man. Whether I'm stocking up on Splenda brand sweetener at my local Starbucks brand coffee café or stocking up on dollars pilfered from purses in the coat room at a dinner party, I know that a real man saves money like it was 1933 all over again. Of course, I mentioned to my classes that the bottle had gone missing, in the hopes that it would show up. Another swirl of bleach, another boiling, and the thing would be as good as nearly new once again. Who am I to turn up my nose?

Then, on November 2, 2004, a University of Minnesota Nalgene bottle, puce colored, shows up in my mailbox in the department, with only a mysterious note attached:

Ben Munson's New (never-been-used-by-another-person) Water Bottle (Use it in Good Health).

A free Nalgene bottle? What is this, Christmas and my birthday combined? I was suspicious. I took it home, swirled it out with bleach, boiled it for 2 hours (actually, more like 1 hour 15 minutes), then filled it the cap with water and let my cat drink out of it. She didn't die, so I figured it was OK. Then I cleaned it again—cat germs, you know—and brought it to school today, 11/3/05. Well, I made a big deal out of it in the morning phonetics course. Then, about 2 hours later, BAM, -> ANOTHER <- water bottle shows up in my box with the same message. I'm running out of bleach here, guys!

I haven't been this flummoxed since the whole canolli fracas back in '03.

Who are the suspects? I think they know who they are. Let's call one of them "Ms. H from Bloomington." Another could be referred to as "the former Miss Munson, currently of Afton." "Poëlle Nadgitt" is the code name of another likely suspect. Also there are two guys in phonetics—I think their names are both James—who seemed veeery interested in the whole business. They're on the list FOR SURE. The font in the first note was Comic Sans, a favorite font of Dr. Peggy Nelson.

There is a reward for any information that leads to the definitive identification of the perpetrators. The reward is a couple of singles and a bunch of packets of Splenda. I can spare them.