Anonymous sources, slippery slopes

Charles G. Bakaly 3d, the spokesman for Mr. Starr, declined to discuss the matter. "We will not discuss the plans of this office or the plans of the grand jury in any way, shape, or form," he said.

These two declarative sentences appeared in a page-one exclusive by Don Van Natta Jr. in The New York Times on January 31. The story arrived at the height of the Senate impeachment trial and created a stir. It said that Kenneth Starr had concluded that he "has the constitutional authority to seek a grand jury indictment of President Clinton before he leaves the White House." This was attributed to "several" sources, none of them, apparently, Charles G. Bakaly III.

On March 12 came another story, this one widely covered: Bakaly abruptly resigned and hired a lawyer as Starr turned over to the Justice Department results of an internal investigation into the leak of that January 31 Times story. As the Times itself put it, "there was no doubt that Mr. Bakaly's departure and the unauthorized disclosures were linked." What gives?

On one level, this situation is murky. The independent counsel has been under investigation by federal judge Norma Holloway Johnson about possible leaks related to grand jury investigations, and why he might toss Bakaly off the sled on this non-grand-jury story is hard to say.

From a narrower, journalistic angle, however, two possibilities present themselves. Either Bakaly did not leak, just as he asserted via Van Natta. Or he did, and persuaded the newspaper of record to falsely quote him saying he didn't.

Exactly what happened can't be known without more information, which may never arrive. (Van Natta's bureau chief, Michael Oreskes, says he can't discuss "who the source was or who it wasn't" He also says that the use of a false no-comment would violate Times standards. He has yet to hear from Justice.) But the incident serves as a departure point for thinking about how we bargain with our anonymous sources.

Not long ago at The Philadelphia Inquirer, reporters and editors engaged in a yearlong series of frank, small-group conversations about fairness and accuracy. Editors were surprised to hear one of their columnists describe how she had put into a story that X had no comment when, in fact, X was a source. "She didn't see anything wrong with it," says Gene Foreman, former managing editor, now teaching at Penn State. Some Inquirer sportswriters admitted using deceptive I.D.s for their sources. Such practices were strongly discouraged. "I get quite exercised about it," Foreman says.

Journalists face ever-stiffer competition against outlets with wildly varying sets of standards. Sources, meanwhile, are increasingly savvy. So they have stronger bargaining position than they used to on such matters as how they'll be I.D.'d, says Bill Kovach, curator of the Nieman Foundation. "The leverage has shifted from the journalist to the source." Still, we need to bargain hard, and to remember who we bargain for.

Nobody who is realistic wants to outlaw anonymous sourcing. And journalists know that to limit it is to limit certain stories. Robert Kaiser, associate editor of The Washington Post, contends that the image of the anonymous source as a story manipulator is largely false: "More likely, that source is the young staff assistant who knows what's going on but is scared to death to talk about it" for the record. Some bargaining about identification, Kaiser says, "is normal."

Yet to use any anonymous source is to walk gingerly along a slippery slope. That slope increases its pitch this way: Anonymous but well identified. Anonymous and weakly identified. Anonymous and no I.D. Anonymous but deceptively identified. Anonymous and protected with a false no-comment. I can imagine a rationale for each of these steps down the slope. But at the same time, each step exponentially gambles our store of credibility. You can argue persuasively that it is a long distance between deception and no deception, but not that reporters never cross that line. All the way to We deceived the readers in order to inform them.

The nature of deceptive sourcing makes its frequency impossible to gauge. It is seen as enough of a potential threat by edi tors at The Washington Post to put this in written ethics rules: "We cannot offer to protect sources by writing inaccurately that they refused to talk to us." The Wall Street Journal, too, has recently taken steps to guard against the same problem. he big test of how well we bargain came, of course, with the Monica mess. The Committee of Concerned Journalists (www.journalism.org/concern) studied both the frequency of anonymity during a slice of the scandal and the quality of source identification.

On the former point, it found that The New York Times seems to work harder at getting sources on the record than The Washington Post, with the Los Angeles Times in between.

On quality of identification, the committee analyzed more than 2,000 anonymously attributed "statements and allegations" in major media stories over four days in January and March '98, and found that six in ten were attributed to sources characterized "in the vaguest terms." Less than two in ten offered "even the slightest hint of the source's allegiances."