News: Kevin Quinn on How Words Shape Democracy
by Elizabeth Gehrman, Special to the Harvard News Office August 1, 2006
There has been much talk in the United States in recent years about what the New York Times Magazine recently called "The Framing Wars". That is, the subtle - and not-so-subtle - ways politicians and parties use language to promote their views on a given topic, and how word choice can change the nature of a debate. For instance, Republicans might say "unborn" where Democrats prefer "fetus." Democrats might use the phrase "healthcare for all," and the GOP, looking at the same plan, "socialized medicine." A recent example of this phenomenon came when George Bush's call for "personal" accounts to supplement Social Security was quickly turned around by Democrats who recast them, instead, as "private" accounts - to voters, a significant difference.
Such changes in rhetoric often lead to actual changes in the meaning of words, says Assistant Professor of Government and IQSS faculty resident Kevin Quinn. "In the 1950s and '60s," he points out, using a frequently cited example, "being a liberal was not necessarily a bad thing. Whereas now nobody in their right mind, unless they're running for maybe Cambridge City Council, would call themselves a liberal, or they'll get killed in the election."
Republicans, Quinn adds, have for decades been "much better" at this than Democrats. "Though Clinton was fairly good as well," he says. "Welfare reform, for example, was a Republican issue for a long time and he was able to make it a Democratic issue and win over a lot of voters with it. This is a perfect example of how issues are framed, or how parties might try to steal issues from other parties." From a linguistic standpoint, then, in politics, if our word wins, we win.
Yet despite the dust-ups these lexicon-bending strategies have created, few studies have actually examined the issue. "To be honest," says Kevin Quinn, an assistant professor in the Department of Government, "political scientists really have very little comprehensive data on how politicians have used language over even the past 10 years, much less over the past 50 to 100 years." Quinn is one of the principle investigators in a three-year National Science Foundation-funded research project that will analyze the legislative records of many countries over several years to determine not only how language is used by parties, but also by individuals when addressing others in the political elite.
Phase one of the $749,000 study will develop base data and analyses of the floor debates in prominent English-language legislatures: the U.S. House and Senate, the British House of Commons, the Irish Dail and Senate, the Indian Lok Sabha, the New Zealand Parliament, the Australian House and Senate, and the South African National Assembly.
"This group was chosen for the initial phase of the study because their data are all available electronically," Quinn says. "It's a tremendous amount of information. The Congressional Record alone can run hundreds of pages a day. So when people have tried to analyze it in the past, they've had to look at distinct pieces of data or bits of time that are very well-defined." Most studies, therefore, have examined only the use of individual "unique" words not taking into account the order of the words, sentence structure, or "who's speaking after whom."
Until recently, sheer volume was an insurmountable problem for political scientists, but data storage and computational analyses have finally advanced enough that Quinn and his colleagues can compare legislative linguistics not only over the years, but across the world. "One reason we're looking at many countries," Quinn explains, "is to compare the types of rhetorical choices made by politicians in various institutional settings. So, for example, if you're a politician in the Netherlands, which has a single national constituency and a very pure form of proportional representation, will you have the same incentives to engage in the same types of speeches as, say, a politician in a majoritarian system?"
They also hope to be able to track the way political information filters across the globe. "How are the concepts we're talking about in the U.S. perceived in Japan?" asks Burt Monroe, an associate professor of political science at Michigan State University, an adjunct faculty associate at the University of Michigan's Center for Political Studies, and a principle investigator on the study. "How has the topic of terrorism changed over the past four years? Is it different among countries that are our allies? Is it affected by events? Does a bad event in Iraq or a bombing in London change the way people talk about these topics just in those countries, or in other countries as well?"
Quinn and his colleagues maintain that this is more than an academic exercise. "We hope to not just look back," Monroe says, "but monitor these things in real time. We'd like to create a Web site where people can look up the new words Democrats and Republicans or leaders in other nations are using this week."
The news media, Monroe points out, noticed the partisan phraseology of the Social Security discussion probably weeks after the debate began - but seeing such strategies early on can help voters sort out the issues and the candidates' responses to them and lead to more informed choices. "At its heart what we're after is a better understanding of representation," he says.
"And that's what makes for a better democracy," Quinn adds.