When big companies threaten lawsuits to suppress the timely reporting of newsworthy information, media organizations often rush to the barricades waving the First Amendment as their banner.

Not so now, when it is a group of media organizations doing the suppressing. A consortium called the Voter News Service conducts Election Day exit polls for its six members: ABC News, CBS News, NBC News, Fox News, CNN, and the Associated Press. VNS also sells data to more than 100 paying subscribers. Threats by VNS to sue have stopped the online magazines Slate and National Review Online from continuing to break the taboo against publishing leaked VNS exit poll numbers before the polls close.

This little-noticed dispute could have large implications for First Amendment rights; for the future of exit polling; and for Election Day television reporting, which is steeped in well-intentioned but corrosive deception of viewers about what the networks know and how they know it.

By combining its exit polls with other data, VNS is able to project election results with remarkable accuracy long before all the votes have been cast. The exit polls also collect demographic and other information about voting patterns that many scholars and journalists consider invaluable. VNS distributes the exit poll numbers and projections as they are compiled to its members and paying subscribers, including many newspapers and magazines but not Slate and National Review Online.

In keeping with a 1985 promise by network executives to Congress-which was worried that prospective voters who hear winners and losers announced might decide to stay home-VNS members and subscribers commit by contract to delay broadcasting the exit poll numbers from any state until most polls in the state have closed.

Slate deputy editor Jack Shafer began thumbing his nose at the VNS embargo on the afternoon of the New Hampshire primary, when he put raw, early exit poll numbers gleaned from other journalists on Slate’s Web site. Shafer says he did so to dramatize his larger point, that much Election Day television reporting is "a fraud," perpetrated by news anchors and correspondents who know the results of many elections hours before the polls have closed but who pretend otherwise. They resort to ever-more-transparent hints and innuendoes to signal who is winning long before they disclose the exit-poll data underlying their analyses or the reasons for the delay.

Meanwhile, VNS exit polls circulate so widely on the journalistic and political grapevines that they are readily available to journalists such as Shafer, who is not bound by the embargo because Slate is not a VNS subscriber.

Although a VNS spokeswoman refused to comment, the threatened VNS lawsuit would have accused the magazines of "misappropriation" of intellectual property.

VNS and the networks are hardly alone in resenting Slate’s conduct. Richard Morin, The Washington Post’s polling director, warned in an op-ed column that Shafer and other "self-aggrandizing Net journalists who gleefully have reported the early results of exit polling even before the polls close" will provoke Congress "to kill off exit polling for good."

Shafer is even more outraged-but at what he calls "the VNS information cartel" and "the sort of behavior that I would expect from Food Lion, or a defense contractor, or a tobacco company, not a news organization."

Nonetheless, Slate Editor Michael Kinsley-on the advice of attorneys for Microsoft Corp., Slate’s owner-decided not to risk litigation, and stopped publishing exit poll data in late February. In the meantime, other Web publishers have jumped into the breach by posting early exit poll numbers on the Virginia and Super Tuesday primaries. VNS also delayed the release of its findings to subscribers, from 2 p.m. to 4 p.m., on Super Tuesday to minimize leaks.

In my view, VNS has good reasons to want to keep its exit polls secret until its members are prepared to make the numbers public. But Shafer was doing the kind of reporting that good journalists do every day. And news organizations should hesitate to haul other journalists into court for publishing news. Any such lawsuit by VNS would smack of hypocrisy and set a precedent that could haunt its members.

(Full disclosure: Kinsley is an old friend; he and Shafer have been helpful to me in various ways; and I have done free-lance writing for Slate.)

Indeed, one of the legal claims threatened by VNS-interference with contractual relations-echoes the claim that Brown & Williamson Tobacco Corp. used to scare CBS News (a VNS member) into delaying publication of Mike Wallace’s famous 60 Minutes interview with tobacco whistle-blower Jeffrey Wigand.

VNS’ most legally plausible claim is that Slate violated its intellectual property rights under the so-called "hot news" doctrine. This doctrine dates to a 1918 Supreme Court decision upholding a lawsuit by AP against another news wire that had made a business of immediately republishing facts from AP dispatches without paying for them.

But the courts have construed the hot news doctrine quite narrowly in recent years. And it should be. If VNS and its members-who sit on their "hot news" until it has cooled-succeed in court, the precedent would crimp the First Amendment rights of journalists to report on other news organizations, which would be claiming a unique and unseemly legal right to block journalistic scrutiny. Such scrutiny often involves discussion of not-yet-published stories.

So what should VNS and its members do? Kinsley says they should publish their own exit poll data as soon as they are available. He stresses that there is no solid evidence that this would discourage people from voting, let alone tip election outcomes. And he argues that it is inexcusably patronizing, as well as deceptive, for journalists to keep prospective voters in the dark for their own good.

Because we all know that a single vote almost never tips an election, Kinsley adds, voting is less an effort to affect outcomes than "an act of democratic faith." It would thus be illogical for voters to stay away from the polls just because the winners have already been announced.

These are powerful points. But "democratic faith" is too fragile to be reduced to hard, Kinsleyesque logic. It seems plausible (if unproven) that instantaneous publication of VNS exit polls and projections might have the unfortunate effect of depressing voter turnout.

And while Kinsley may be right in principle, the networks have good reason to worry that Congress, if provoked, might act to curb exit polling. It could, for example, ban anyone from approaching a voter within 200 feet of a polling place. Although courts have struck down state laws of this nature, there’s no guarantee that the Supreme Court would reach the same result.

The best solution to the exit poll dilemma would probably be for VNS and its members to stop circulating the data so widely, and treat them as real secrets until they are ready to make them public.

"They sort of kill their own embargo when they leak it all over the place," says Warren J. Mitofsky, who headed the CBS News election and survey unit for many years and who ran VNS’ predecessor consortium from 1990-93. He now consults for CBS and CNN.

The only sure way to prevent leaks to nonsubscribers such as Shafer, says Mitofsky, "is not to distribute the [VNS] information to the members until late afternoon." I’d amend that to shortly before the polls close. This would also have the happy side effect of ending the deception of viewers by keeping exit polls out of the hands of network (and other) journalists until they are ready to clue in the public.

This idea is unlikely to appeal to the networks, which say their anchors and correspondents need early access to the exit polls to plan their evening coverage. But Mitofsky, who knows whereof he speaks, dismisses this as "B.S." And the other early uses of exit poll data-enabling correspondents to offer confident "analyses" without disclosing the bases for them, and building a contrived suspense and sense of drama-add entertainment value at the expense of candor. Network correspondents should not be put in the position of playing games with their viewers about how much they know.

In the long run, it would bolster the networks’ credibility if the need to plug leaks forced them to keep the exit polls from their correspondents until they are ready to make them public. If Jack Shafer has pushed the networks in that direction, he may have done them a favor.