New York Times Anonymous Sources Project

September 2008
By: ALICIA TEJADA and CASSANDRA LIZAIRE

Last fall, Clark Hoyt, Public Editor of The New York Times, spoke to Professor Richard Wald's Critical Issues in Journalism class at Columbia University's Graduate School of Journalism. He talked about the use of anonymous sources and presented the idea of a research project to evaluate how closely reporters adhere to the newspaper's anonymous sources policy.

Drawing on his experience as the former Washington D.C. Bureau chief of Knight Ridder, Hoyt reflected the need for anonymity to protect the identity of sources who fearing reprisal might not otherwise come forward. While the device buffers against a chilling effect on the press, Hoyt also criticized hurried reporting that abuses anonymous sourcing for the sake of the big scoop.

Anonymous sources have long served as the anchors for many an investigative story. In 2005 Josh White of the Washington Post broke news of the possible detention of “ghost” prisoners at the Abu Ghraib prison in Iraq. The Post's coverage supplemented information from military documents, with the disclosures of unnamed prison guards and “Defense Department officials” to report on the illicit confinement of unlisted prisoners in Iraq by the U.S. Army and the CIA.

But as the public demands more transparency from the press, not least because of a series of recent fabricated scandals, there is a freshly growing need to provide readers with more information.

For instance, the Times article about John McCain published last February cited conversations with "former campaign associates" and insinuated the presidential candidate once had an affair with a female lobbyist. The story should have stood on the merits of questionable quid pro quo—instead, the reporters overreached with their claims, and undermined the credibility of the story.

The argument presented by critics of the article was that if reporters are going to break the rules by introducing an anonymous source, they should better communicate why information obtained and cited that way is reliable.

As a point of departure for the study, students used the policy outlined by Times Executive Editor Bill Keller in a memo dated February 25, 2004.

Under the new policy, if sources could not be identified by name and title, explanation must be given as to their authority, possible views and motivation and reason for confidentiality.

Ideally, the anonymity is not granted to sources making personal attacks, or to those without "firsthand knowledge" of asserted facts. And in the case where anonymous sources broached "less verifiable" facts potentially damaging to another to one side, corroborating sources should be used.

Anonymous sourcing can only be used as a "last resort" on stories that are particularly newsworthy, yet sensitive, in order to protect the source and maintain the public's trust.

Students read two sets of New York Times papers: The first set of 6 papers was published just before Keller issued the 2004 memo and the second set of 6 was published in the fall of 2007.

After much discussion, we agreed to document an anonymous source as one where information was attributed to someone whose identity is indiscernible to the lay reader. This includes attributions such as "sources said" or "experts said," without further description of the speaker or the reasons for granting anonymity.

A comparison of citations found in the sample sets includes the following observations:

• Ultimately 79% of anonymous sources in the Times do not meet the requirements of the 2004 memo, which was improvement from the pre-policy memo where 87% of citations were inadequate.

• Post-policy, levels of uncorroborated anonymous sources went up by 7%.

• 42% of anonymous source citations were used in conjunction with an opinion versus a statement of fact – up from 38% in the pre-policy sample.

• The daily average use of anonymous sources was cut by half after the 2004 memo was circulated.

• Also, anonymous sources were less likely to appear on the front page of the NYT in the 2007 sample, than in pre-policy 2004 set.

The students' findings were presented to top editors at the Times, including Clark Hoyt, Bill Keller and Craig Whitney, the paper's standards editor. Keller noted that he would be "more likely to use examples than the metrics," underscoring the difficulty of assessing progress on a subjective issue like anonymous sourcing and the importance of remaining flexible in judging the merits of its use.