Accessibility links

Breaking News

January 12, 2005 - Interview with William Labov: Sound Change, Part 1 - 2005-01-11

First broadcast: January 12, 2005

AA: I'm Avi Arditti with Rosanne Skirble, and this week on Wordmaster: the sounds of change.

RS: If you want a good example of how language changes, just picture a "mouse." Are you thinking of a rodent -- or a device for moving the cursor on a computer screen?

AA: Shifts in language also involve pronunciation. In fact, our friend Ali the English teacher in Iran suggested this topic, since regional dialects can be confusing for English learners.

RS: Native speakers, too. If someone in Chicago offered us some "sacks," we might think it was for carrying our groceries, not for keeping our feet warm. "Sacks" is how Chicagoans pronounce what we call "socks."

AA: The linguist William Labov at the University of Pennsylvania cites this example of a sound change in the United States called the "Northern Cities Shift."

WILLIAM LABOV: "Now what happens here is the short-a becomes 'ai' [like in "yeah"] in every single word, so that people have, say, 'theaht' and 'feahct'. In the meantime, the short words spelled with short-o like 'socks' or 'block' or 'cot' move into the position that was formerly occupied by 'ah.' So the man's name [John becomes] 'Jahn' -- that's a man's name, 'Jahn.' And the girl's name [Jan] becomes 'Jain.' So this is like a game of musical chairs. We call it chain shifting, in which five or six vowels all change places."

RS: William Labov and other researchers have been tracking sound changes for a big project, the Atlas of North American English, to be published in a few months.

WILLIAM LABOV: "Well, American dialects have been studied for a hundred years or so. But unlike European countries, America has never finished a map of the United States, only the eastern United States is covered and a few spots here and there. So we included both Canada and the United States in a study we started in 1992. And in about six years we covered the entire continent using a telephone survey of all the urbanized areas, the big cities. So it covers about two-thirds of the population of North America, and it represents them with about 800 speakers.

"And what's most remarkable is that the rapidly changing dialects of the United States form a very solid, clear picture. And instead of getting a pepper-and-salt effect, we find very clear and sharp divisions between the dialects of the United States, which are getting more different from each other as time goes on. The most important differences have developed in this huge area around the Great Lakes region which we call the Inland North, going from Buffalo, Syracuse, Cleveland, Detroit, Chicago, Milwaukee. Those great cities occupying about 35 million people are all moving in a very different direction from the rest of the United States."

RS: "How do you go about describing how people talk? How do you put together an atlas -- you have to actually hear the people talk."

WILLIAM LABOV: "Well, yes, our Web site is in construction now, is being created by people at the University of Marburg [in Germany], is going to be quite a remarkable innovation in dialectology because on this Web site you will see the maps, you will see all the cities. And when you click on any one point you will be able to hear maybe a minute or two of speech from each person in that area. And, furthermore, you can hear the same word pronounced in many, many different cities across North America.

"The answer to your question about sounds is that we can measure sounds acoustically and the difference between 'mad,' 'maad' and 'maaad,' the difference between 'go' and 'gow' and 'gaow,' will show up very clearly in these measurements, which is one of our main businesses. So about 440 speakers of our 800 have been analyzed in that way."

RS: "So these are all from telephone conversations?"


AA: "And you give them a list of words to repeat?"

WILLIAM LABOV: "We actually have mailed people word lists. We focus upon pairs of words very often which are the same in some areas and different in other areas. For about half the geographic area of North America, the words 'cot' and 'caught' are pronounced the same way, [and] 'Don' and 'Dawn.' So they will say 'Don Hock married Don Hock' whereas the people in New York might saw 'Dawn Hock married Don Hawk.'"

RS: We continue our conversation with University of Pennsylvania linguist William Labov next week on Wordmaster.

AA: We will also have a link to where you can learn more about the forthcoming Atlas of North American English, at our Web site,

RS: And our e-mail address is With Avi Arditti, I'm Rosanne Skirble.