WASHINGTON -- For many years now, professional pollsters have defended their business not as a predictor of things to come but as, in their favorite phrase, "a single snapshot of a moment in time." In other words, polls are a measurement of what people think today, not tomorrow or next month, on a range of issues or personalities. The definition is an excellent buffer against post-poll criticism, but it happens to be valid.
The limitation has not stopped readers of polls in politics, in the news media and among the public from using polling results as crystal balls, particularly in the so-called horse race aspect of political campaigns -- not only who's ahead and who's behind, but who is going to win on Election Day.
Campaign fund-raisers have also made it common practice to use political polls -- and increasingly to commission polls themselves -- to produce numbers that will persuade contributors to open their wallets, the contents of which are then used to buy more polls to generate more campaign funds, and so on and so on.
Beyond that practice, there are polls and then there are polls. In the era of talk radio and television, the phenomenon of call-in polls -- dial the host and give him or her your opinion or candidate preference -- has too often reduced polling to a totally unscientific and unreliable parlor game. The loudmouth school of talk show hosting, personified by conservative polemicist Rush Limbaugh, mostly draws angry mimicry of the hosts' opinions; Limbaugh himself calls his faithful "dittoheads."
Now comes a University of Texas political scientist named James Fishkin, in collaboration with the Public Broadcasting Service, the "MacNeil/Lehrer Newshour," 11 presidential libraries and some of the most highly regarded poll-taking organizations in the country, to provide an alternative to what Fishkin calls tabloid television polling.
He is heading a very ambitious effort to gauge informed American opinion -- as opposed to "top-of-the-head attitudes" -- on the eve of the 1996 presidential primary season.
By so doing, he hopes to raise the level of public discourse and provide a much better guide to it for the presidential candidates as they seek to address the major issues confronting the nation.
With a $4 million budget raised privately, half of it already in hand, the project intends to identify, listen to and inform 600 eligible voters around the country, then bring them to the Lyndon B. Johnson Presidential Library in Austin for a long weekend of briefings, discussion and deliberation in January 1996.
As part of the exercise, the major presidential candidates at the time will be invited to respond to questions hammered out as a result of the discussions among the 600, held first in small groups and then at the "national issues convention" in Austin.
The 600, who themselves will have answered a questionnaire at the outset without the benefit of the briefings, papers and discussions designed to better inform them, will fill out another after the process, for what Fishkin calls a "deliberative poll."
The objective, he says, is to gauge "what people would think if they had a chance to think" armed with balanced information on which to make their judgments, as opposed to the "tabloid sound bites" fed to them through other avenues.
The convention and the deliberations are to be televised over PBS, not simply the weekend sessions but edited segments of smaller discussions among the participants that have led to the specific questions asked of the candidates, providing, Fishkin says, "a poll with a human face."
Unlike talk radio call-in polls, he says, "instead of just an angry voice, we mean to get deliberative voices."
When asked whether the project would go forward if only some or none of the major candidates agreed to take part, Fishkin seemed hesitant. Then he said that it could be done without HTC them, but that they should have something to say about issues raised by such a process, "and if they don't come, God help the country."
Candidates, however, do what suits their own interests. This exercise should do that, but if done right with or without them, it should serve the public interest of an informed electorate. So it's worth trying whether the candidates join in or walk away.