How I got into linguistics, and what I got out of it (2)

by William Labov,

University of Pennsylvania

In 1970, I moved from Columbia to Penn, mostly because the Philadelphia dialect offers an ideal laboratory for the study of changes in sounds: two thirds of the Philadelphia vowels are involved in a complex game of musical chairs. Here at Penn, I was joined by my colleague Gillian Sankoff, who had developed these methods even further in her study of the French of Montreal, and broken new ground in the study of Tok Pisin, the newly formed national language of New Guinea. We developed the Linguistics Laboratory, a place people come to from here, there and everywhere to learn how to work with language in a scientific and realistic way. We work with one foot in the university, and one in the community. In the course on "The Study of the Speech Community," students learn how to cross the line that separates the university from the world around it. They make friends in the local neighborhoods, gather data on social life, and analyze it by quantitative techniques.

If this empirical approach were the dominant way of doing linguistics and linguistic theory, I would certainly have lost sight of the academic adventure that once inspired me. Fortunately, this is not the case. Linguists are still basking in their own ideas, still finding the answers by asking themselves questions, and most of them are frightened by any number larger than six. But the general impression in the field is that if you want to study how people actually use the language, and if you want to measure what you are studying, you should come to Penn and work with Sankoff, Kroch, Prince and Labov. We have a growing number of students who are afraid neither of people nor of mathematical symbols. The technology becomes more exciting all the time. We've derived equations that give some insight into why language keeps changing, and who changes it. And it's turning out that our knowledge of dialect diversity has important applications for automatic speech recognition. If a computer is going to understand how human beings speak, it has to understand Chicago speech as well as New York speech. And we've now succeeded in mapping these sound changes through Telsur, the telephone survey that has produced the Phonological Atlas of North America.

All of this technology could easily carry us away from the human issues involved in the use of language. From my point of view, that might win the game but lose the match. I spend a great deal of my time in the laboratory, at the office, or in class. But the work that I really want to do, the excitement and adventure of the field, comes in meeting the speakers of the language face to face, entering their homes, hanging out on corners, porches, taverns, pubs and bars. I remember one time a fourteen-year-old in Albuquerque said to me, "Let me get this straight. Your job is going anywhere in the world, talking to anybody about anything you want?" I said, "Yeah." He said, "I want that job!"

Once you find out what you want to do, you have to convince somebody else to pay you to do it. You can defend any piece or research by saying that it is "theoretical" and "basic" research, and you may be able to get the grants you need. I myself have always felt that theory can only be justified if it fits the facts, and that some facts--the ones that affect people's life chances--are more important than others. Four years ago, I organized another research group to return to the problem of black/white differences in Philadelphia, and we discovered that the differences we found in Harlem are not growing less. On the contrary, the home languages of blacks and whites are growing more and more different from each other. This became a national news story, and we were able to use the facts to underline the dilemma that Ted Hershberg of the Penn Philadelphia Social History Project has shown us: that increasing segregation in the northern cities is depriving the black community of its basic resources, and is in danger of creating a permanent underclass. Sociolinguists have now discovered that this is true in every city in the country: while the white dialects are continuing to develop and diverge from each other, the black community of the inner city holds aloof from all this, and has developed a nationally uniform grammar that is more and more distinct from that of the surrounding white dialects.

This year the renewed controversy about African American English surfaced in the "Ebonics" Controversy. When all the furor died down, it become clear that the African American community of Oakland has finally decided, as a whole, that it is time to stop blaming children for the failure of the schools, and time to improve our methods of teaching reading by using our knowledge of the language that children actually speak. After thirty years of effort, there is now a distinct possibility that the knowledge we have gained can be put to work, and here at Penn we are once again putting our shoulders to the wheel.

In 1987, I had another opportunity to test the usefulness of linguistics on a matter that was vital to a single person. A number of bomb threats were made in repeated telephone calls to the Pan American counter at the Los Angeles airport. Paul Prinzivalli, a cargo handler who was thought by Pan American to be a "disgruntled employee," was accused of the crime, and he was jailed. The evidence was that his voice sounded like the tape recordings of the bomb threat caller. The defense sent me the tapes because Prinzivalli was a New Yorker, and they thought I might be able to distinguish two different kinds of New York City accents. The moment I heard the recordings I was sure that he was innocent; the man who made the bomb threats plainly did not come from New York at all, but from the Boston area of Eastern New England. The problem was to prove this in court to a West Coast judge who could hear no difference between Boston and New York City speech!

All of the work and all of the theory that I had developed since Martha's Vineyard flowed into the testimony that I gave in court to establish the fact that Paul Prinzivalli did not and could not have made those telephone calls. It was almost as if my entire career had been shaped to make the most effective testimony on this one case. The next day, the judge asked the prosecuting attorney if he really wanted to continue. He refused to hear further statements from the defense. He found the defendant not guilty on the basis of the linguistic evidence, which he found "objective" and "powerful."

Afterwards, Prinzivalli sent me a card saying that he had spent fifteen months in jail waiting for someone to separate fact from fiction. I have had many scientific results where the convergence of evidence was so strong that I felt that I had laid my hands on the reality behind the surface, but nothing could be more satisfactory for any scientific career than to separate fact from fiction in this case. By means of linguistic evidence, one man could be freed from the corporate enemies who had assailed him, and another could sleep soundly on the conviction that he had made a just decision.