Speech Science: Tongue Twisters and Valley Girls
IRA FLATOW, HOST:
If you thought saying she sells seashells by the seashore was hard, my next guest has created even more difficult tongue twisters. She also studies what is happening when your tongues do get tangled in knots, and boy, do my tongue get - you know. She just presented her findings this week at the annual Acoustical Society of Americans Conference in San Francisco. Stefanie Shattuck-Hufnagel is a psycholinguist in the speech communication group at the Massachusetts Institute of Technology in Cambridge.
Welcome to SCIENCE FRIDAY.
STEFANIE SHATTUCK-HUFNAGEL: Thank you, Ira.
FLATOW: Why is it so hard for us to say some of those tongue twisters?
SHATTUCK-HUFNAGEL: Well, we have some idea of the answer to that question, but we certainly don't have a complete idea yet. There are two factors that we think about: One is, what are the sounds themselves? So there's something about th- and sh- that are particularly difficult to say in sequence and so she sells seashells or the sixth sick sheik of the six sixth sheep's sick.
Those kinds of twisters are particularly hard partly because of the sound, the particular sounds that are involved. But there's another reason why things are hard to say, and that is the pattern with which the sounds occur. So if you think of she sells seashells, the s/sh are at the beginnings of those words, are alternating in one pattern.
And the e/l of the rest of the word is alternating in the opposite pattern, and it's kind of like rubbing your stomach and patting your head at the same time. Your brain just doesn't seem to be able to handle two alternating patterns in the same utterance very well.
FLATOW: Are there two different distinct types of tongue twisters?
SHATTUCK-HUFNAGEL: No, it's more, I think - well, tongue twisters themselves have been an object of study across languages for a long time and people have tried to figure out why they're hard to say. The way we look at it is that there are quite a few different factors that contribute to something being hard to say and these are two of the factors. And then there are additional factors that you can manipulate independently.
You can create tongue twisters that have one or two or four or five of these different factors and I hasten to say that I make no claims to have found the hardest, the mother of all difficult tongue twisters.
FLATOW: Really? There isn't one.
SHATTUCK-HUFNAGEL: No, I'm not making that claim at all.
FLATOW: Well, let's try out a few. Start with the word list. Give us a tongue twister word list.
SHATTUCK-HUFNAGEL: All right. Here's a word list. Pod cod, pod cod, pod cod, pod cod, pod cod.
FLATOW: I don't even know what that second word was.
SHATTUCK-HUFNAGEL: It's the fish, the famous fish.
FLATOW: Oh, the cod. Oh. Pod cod, pod cod, pod cod, pod cod.
SHATTUCK-HUFNAGEL: Faster, faster.
FLATOW: Faster, faster. Oh, you want me to read it faster, faster. Pod cod, pod cod, pod cod, pod cod, pod cod.
SHATTUCK-HUFNAGEL: You're very good. Try this one. Top cop very fast.
FLATOW: Oh, this is a tough one. Top cop, top cop, top cop, top cop, top cop, top cop. You got me. You got me on that one.
SHATTUCK-HUFNAGEL: Right. And one of the things that you probably just did, it was a little hard to hear, is you started to say t-cop. Can you hear me saying two sounds at the beginning of that word instead of just one?
FLATOW: Right. Yeah, I did.
SHATTUCK-HUFNAGEL: That's what often happens when you're trying to say those two words over and over again. You're alternating the two words, you're not saying it as you would say a sentence; there are many, many of them in every second of speech. There are a lot of characteristics of that string of words that are different from typical speech.
Nonetheless, you make a certain kind of error and one of the kinds of errors you make is you try to say two sounds at the beginning of the word, so you come out with t-cop or p-cod. But now if I give you a sentence to say, which has the same words in it, at least in our experiments, we find that there's a different kind of error that occurs. So if you say the top cop saw a cop top, the cop top saw a cop top.
Try that and let's see what happens.
FLATOW: Give me the sentence again.
SHATTUCK-HUFNAGEL: The top cop saw a cop top.
FLATOW: The top cop saw a cop top? The top cop saw a cop top; is that right?
FLATOW: The top cop saw a cop top, the top cop saw a cop top.
SHATTUCK-HUFNAGEL: That's very good. What our subjects, our participants did was they tended to substitute one whole sound for a different sound. So if they were trying to say the top cop, they would say the cop top. They would produce a T for a K and a K for a T.
FLATOW: Well, you know, I can do the tongue twisters, but I can't remember anybody's names. It's a whole different problem.
SHATTUCK-HUFNAGEL: We need to get you into the laboratory and study you very carefully.
FLATOW: Well, are there other people who are better at tongue twisters than others?
SHATTUCK-HUFNAGEL: Well, you know, we've never studied that in our lab, but as a result of this paper, I have been getting emails from people all over the English-speaking world with tape recordings of - or not tape recordings, but digitized sound files of themselves trying to say these things, and I have to say that I think there might be quite a bit if individual variation in how accurately people can produce these things.
One thing, of course, that we're not controlling in those recordings that people are sending me is the rate, and one way you can certainly help is to speak more slowly.
FLATOW: You know what I discovered that helps is also I close my eyes and I think more about the word...
SHATTUCK-HUFNAGEL: You're kidding.
FLATOW: The word coming toward me, the word is coming at me, you know.
FLATOW: Maybe I'm not distracted and I can focus more on the words.
SHATTUCK-HUFNAGEL: I wonder if you slow down. Can I come and record you?
FLATOW: Sure. Why not? Come in your lab, you know, put those electrodes on my head.
SHATTUCK-HUFNAGEL: Let's do it.
FLATOW: Well, where do you go next with your study? What would you like to know?
SHATTUCK-HUFNAGEL: Yes. What we would like to know, if you - if you think about the difference between these two kinds of utterances, on the one hand top cop, top cop, top cop, and on the other hand the top cop saw a cop top, they're different in many different ways. And because of the fact that we get different proportions of different error types when we do word lists versus sentences, we would like to understand better which differences between the word lists and the sentences are causing the error differences between the word lists and the sentences.
And the reason for that, very briefly, the reason that's interesting for us is that if you have different types of errors that are elicited by different types of speech, what that means is that there's more than one place in the planning process where the errors can actually occur.
When we start to talk, we have the impression that we don't have to do much.
SHATTUCK-HUFNAGEL: We have the idea, the mouth starts to move, the sound comes out, and we don't have a sense of a lot of steps in the process.
SHATTUCK-HUFNAGEL: But it's clearly a very complicated process.
FLATOW: Yeah. And that's what I noticed. You have to really focus on it. You know, you know the words. You think they'll just come out, but no. You have to focus on each word as you're saying it.
SHATTUCK-HUFNAGEL: That's right.
SHATTUCK-HUFNAGEL: And our hypothesis is - our motivation for these studies is that if we can isolate the characteristics of the speech you're trying to produce, that actually caused one type of error versus another, will be on the way to looking at brain processes.
FLATOW: Well, Doctor, thank you very much for enlightening us.
SHATTUCK-HUFNAGEL: You're welcome. And good luck with pod cod.
FLATOW: Right. Stefanie Shattuck-Hufnagel is a psycholinguist - there's a tongue twister - in the speech communication group at MIT in Cambridge.
(SOUNDBITE OF MOVIE, "CLUELESS")
ALICIA SILVERSTONE: (as Cher Horowitz) So, OK, like, right now, for example, the Haitians need to come to America. But some people are all, what about the strain on our resources? But it's, like, when I had this garden party for my father's birthday, right, people came that, like, did not R.S.V.P. So I was, like, totally bugging.
FLATOW: That's a clip from the movie "Clueless," with Alicia Silverstone. You recognize that Valley Girl way of speaking? You know, where every statement sounds like a question? And when you hear that way of speaking, you know exactly where the person is from, Southern California? But the Valley Girl way of speaking is not just for girls. It turns out that boys use it, too. And my next guest researched this SoCal dialect.
Amalia Arvaniti is a professor of linguistics, University of Kent in Canterbury, Kent in the United Kingdom. Welcome to SCIENCE FRIDAY.
AMALIA ARVANITI: Hi. Good to hear from you.
FLATOW: So boys do it also.
ARVANITI: Yes, they do.
FLATOW: I'm almost starting to talk like it now.
ARVANITI: They do it. They just don't do it as often as the girls do it.
FLATOW: Why is that?
ARVANITI: We do not know. The only thing we can hypothesize is that because it is a feature that it is considered to be a girl thing to do. Boys are kind of reluctant to adopt it. If you ask them, if in class, I would ask my students when I was at the University of California in San Diego whether they use that talk. They would say, no, no. Only girls do that. But, in fact, we haven't found anybody in our sample who did not use that talk at some point.
FLATOW: Let's listen to some - let's listen to a female and a male clip from your study. Here's an example. I'm going to play the female and male up-talking.
UNIDENTIFIED WOMAN: OK. So you start at Revelle. Keep going till you pass Robeks Juice, not in the Roosevelt direction, but towards La Jolla Mesa.
UNIDENTIFIED MAN: And then from there, you have to go all the way to right in the middle where it says Canyon Hills?
FLATOW: Wow. Now that - quite similar.
ARVANITI: Yes. They - there are some subtle differences that we find overall in that the women were a little bit - started their rises relatively later compared to the young man. And, of course, the women use a much higher, larger pitch range. But apart from that, they are really doing the same thing. Yeah.
FLATOW: Let's play another clip, something called a vocal fry. What is that?
ARVANITI: This is what happens when speakers move their vocal folds really - that they make them really tight. So they're almost completely closed, and only one small part of the vocal folds opens and closes. We often enclose are vocal folds when we talk to produce voicing. It's something that you cannot really feel if you put your hand on your Adam's apple.
FLATOW: OK. Let's hear a female and a male vocal fry.
UNIDENTIFIED MAN: OK. So, basically, the scene was about this lady who want to get prepared for her med school interviews.
UNIDENTIFIED WOMAN: And he wasn't impressed with the information she brought.
FLATOW: They're - they have a very distinct youthful sound to them.
ARVANITI: Yes. It seems to be spreading quite a lot. When I first encountered it, I asked one of my students if he had ever been to the doctor to make sure there is nothing wrong with his vocal folds, because I could not imagine that anybody would do this habitually, but they do. It is very widespread.
FLATOW: This is SCIENCE FRIDAY, from NPR. I'm Ira Flatow, talking with Amalia Arvaniti about - well, we've talked about vocal fry clips. Why does it spread? How does it spread? Is it just virally, like, on the Internet or - what happens?
ARVANITI: Recent wisdom is that things like that cannot spread over the Internet or through the media, that you have to - that they spread only through personal contact. However, there is recent research to be published in the journal of the Linguistic Society of America language by Jane Stuart-Smith at the University of Glasgow, which suggest that, actually, speakers can pick up features from things they see on TV or, you know, YouTube and so on and so forth. So the jury is still out on that one. It could be that simply some features are used by, you know, as you saw in "Clueless." It's the cool kids that do these things.
FLATOW: Yeah. And when did we first - when did scientists first - can you put a data on when all these started to happen?
ARVANITI: No. But that's a very good question. I think the only way to do this would be to go back and listen to materials that were recorded, you know, in the '50s and the '60s and so on and so forth, to see if we can actually document the beginning of the change. But right now, as far as I know, this kind of research has not been done. Another thing that we can do is record those older speakers to see what they are doing. Although it is possible for the older speakers also to pick it up from the younger speakers.
FLATOW: And when did they - when did the speech, like, a like, you know, come into - into play?
ARVANITI: The like has been around for quite some time, as well. So I think the first studies are now 15 years old on this one or so.
FLATOW: Yeah. It's just so...
ARVANITI: Well, it became so common that people started studying it.
FLATOW: Is this speech all - is it only a Southern California thing? Is it only a Valley Girl sort of dialect?
ARVANITI: No. I don't think so. In fact, I was listening to a clip from the Boston radio, NPR. They had my co-author Amanda Ritchart there. And although everybody was saying how they hate up-talk and it sounds awful, everybody was actually using it at some point. So I think it's much more widespread than we think.
Perhaps what makes it so salient in California is that the speakers in California use it much more than speakers elsewhere. So, you know, you hear more of it, so you begin to notice it, and it may become irritating to people who are not from the area and are used to it. So...
FLATOW: Well, Dr. Arvaniti, thank you very much for coming...
ARVANITI: You're welcome.
FLATOW: ...coming right on and talking to us.
ARVANITI: You're welcome.
FLATOW: Amalia Arvaniti is a professor of linguistics at the University of Kent in Canterbury, Kent, and that's in the UK.
NPR transcripts are created on a rush deadline by a contractor for NPR, and accuracy and availability may vary. This text may not be in its final form and may be updated or revised in the future. Please be aware that the authoritative record of NPR's programming is the audio.