Memo on the use of confidential sources in the Washington Post, February 2004.
The Washington Post is committed to disclosing to its readers the sources of the information in its stories to the maximum possible extent. We want to make our reporting as transparent to the readers as possible so they may know how and where we got our information. Transparency is honest and fair, two values we cherish. Whenever questions arise about how to convey the transparency of our reporting to the reader, consult with editors.
Sources often insist that we agree not to name them in the newspaper before they agree to talk with us. We must be reluctant to grant their wish. When we use an unnamed source, we are asking our readers to take an extra step to trust the credibility of the information we are providing. We must be certain in our own minds that the benefit to readers is worth the cost in credibility.
In some circumstances, we will have no choice but to grant confidentiality to sources. We recognize that there are situations in which we can give our readers better, fuller information by allowing sources to remain unnamed than if we insist on naming them. We realize that in many circumstances, sources will be unwilling to reveal to us information about corruption in their own organizations, or high-level policy disagreements, for example, if disclosing their identities could cost them their jobs or expose them to harm. Nevertheless, granting anonymity to a source should not be done casually or automatically.
Named sources are vastly to be preferred to unnamed sources. Reporters should press to have sources go on the record. We have learned over the years that persistently pushing sources to identify themselves actually works – not always, of course, but more often than many reporters initially expect. If a particular source refuses to allow us to identify him or her, the reporter should consider seeking the information elsewhere.
Editors have an obligation to know the identity of unnamed sources used in a story, so that editors and reporters can jointly assess the appropriateness of using them. Some sources may insist that a reporter not reveal their identity to her editors; we should resist this. When it happens, the reporter should make clear that information so obtained cannot be published. The source of anything that appears in the paper will be known to at least one editor.
We prefer at least two sources for factual information in Post stories that depends on confidential informants, and those sources should be independent of each other. We prefer sources with first-hand or direct knowledge of the information. A relevant document can sometimes serve as a second source. There are situations in which we will publish information from a single source, but we should only do so after deliberations involving the executive editor, the managing editor or the appropriate AME. The judgment to use a single source depends on the source’s reliability and the basis for the source’s information.
We must strive to tell our readers as much as we can about why our unnamed sources deserve our confidence. Our obligation is to serve readers, not sources. This means avoiding attributions to “sources” or “informed sources.” Instead we should try to give the reader something more, such as “sources familiar with the thinking of defense lawyers in the case,” or “sources whose work brings them into contact with the county executive,” or “sources on the governor’s staff who disagree with his policy.”
In the age of news management, some government agencies and private companies order their employees never to speak to a reporter on the record. In politics, locally and nationally, people who work as aides to prominent figures are often reluctant to speak on the record on behalf of their bosses. But we can still help readers with specific, if anonymous, attribution. So “a senior aide to a Democratic senator on the Commerce Committee” is more helpful to the reader than “a Senate Democratic source”.
When sources refuse to be identified, it is often helpful to show readers that we tried to identify them, and explain why we could not. We should write, for example, that a source “spoke only on the condition that he or she not be named,” rather than saying that a source “asked not to be identified.” Merely asking should not be sufficient to become anonymous in our stories.
It is nearly always possible to provide some useful information about a confidential source. In rare cases when you think a story simply cannot do this, you should seek the approval of the relevant AME, the managing editor or the executive editor before printing information attributed to a totally unidentified source.
Spokespersons, by virtue of their role and title, should be on the record when they are giving briefings or calling us with information. When they decline to be quoted by name in such situations, we should protest, ask for a publishable explanation as to why, and tell readers what happened, if appropriate.
When we call spokespersons in search of guidance, confirmation or information that we need, of course, we can accept that information on background or not for attribution.
In cases where a source is actively trying to persuade us to put something into the paper, but refuses to be identified, we should request a publishable reason for concealing the source’s identity. In an era when more and more of the people we deal with want to remain hidden from view, we can help readers by telling them the reasons why they are not seeing more on-the-record quotes. Indeed, this is one of our best weapons against excessive secrecy. A sentence that reminds readers, for example, that a particular government agency has internal rules forbidding most of its officials from speaking on the record, and has refused to explain this policy publicly, is a good addition to any story where it is appropriate.
Obviously, it is important to share with readers any direct interest our sources may have in the story we are writing. When sources have axes to grind, we should let our readers know what their interest is. Smart sources, particularly in government and increasingly in business, know how to tempt reporters with juicy stories. Smart reporters and editors know how to avoid letting them spin us for their own purposes. Attribution must be truthful. If a story refers to sources (plural), we must have multiple sources. If the story says someone declined to comment, that must be the truth. We cannot offer to protect sources by writing inaccurately that they refused to talk to us. This is not always easy; clever sources sometimes try to pressure us to write things that are untrue or misleading in order to protect their identity. We must resist their efforts. Sometimes we may have to avoid any reference to asource who has helped us but refuses to be identified. Sometimes we may be able to find a formulation that protects the source, but to fulfill our obligations to readers, the attribution must be true. Situations of this kind should be discussed between reporters and editors before anything is put in the paper. As a general rule, any reporter who feels she or he is at risk of misleading the reader about the sources of information in a story has an obligation to discuss this with his/her editor.
Dealing With Sources
We strive to treat sources fairly. This means putting statements we quote into context, and summarizing the arguments of people we quote in ways that are recognizably fair and accurate. Potentially controversial statements by public figures and others should be quoted in a complete sentence or paragraph when possible, and in context. In some cases this will mean making clear what question was being answered when the statement was made.
When seeking comment from persons who are the subject of a story, we should give them a reasonable opportunity to respond to us. This means not calling at the last minute before deadline if we have any choice about timing.
We do not promise sources that we will refrain from additional reporting or efforts to verify the information they may give us.
We should not publish ad hominem quotations from unnamed sources. Sources who want to take a shot at someone in our columns should do so in their own names.
We should avoid blind quotations whose only purpose is to add color to a story.
We do not use pseudonyms, and we do not mislead our readers about the identities of people who appear in our stories. In the rare situations when we decide to identify someone by other than their full name, we do so in a straightforward manner – by using a first name only, for example. Editors must participate in decisions to provide less than a full name, and we must explain to readers why we are not using full names.
We do not fool or mislead sources. When identifying ourselves, we say we are reporters for The Post. Our reporting should be honorable; we should be prepared to explain publicly anything we do to get a story.
We must be careful about our use of the word “said.” Standing alone, it should be used only when the reporter heard the source say the words quoted, either in person, on television or radio. When we quote a written statement of any kind, we should explain what we are doing accurately in our attribution [e.g. “…the White House said in a written statement.”]
When we put a source’s words inside quotation marks, those exact words should have been uttered in precisely that form. Sometimes we will want to avoid humiliating a speaker by paraphrasing in grammatical form an ungrammatical statement, or by presenting in a form acceptable for publication a statement that includes profanities. When we do so, however, we should not use quotation marks. A paraphrase should not be treated as a quotation. At the same time, we should not deprive our readers of the statements of legitimate news sources who characteristically speak so ungrammatically, or use such profane language, that we cannot quote them verbatim. This may mean stretching our rules occasionally to publish profane language we would normally avoid. When in doubt about how to quote a source, consult an editor.
When quoting people for whom English is not their first language, special care should be taken. If such quotations make the speaker look stupid or foolish, we should consider paraphrasing them (outside of quotation marks of course). When appropriate, a story should note that a source was struggling with English.
We must be truthful about the source of our information. Facts and quotations in a story that were not produced by our own reporting must be attributed. Readers should be able to distinguish between what the reporter saw and what the reporter obtained from other sources such as wire services, pool reporters, e-mail, websites, etc.
We place a premium value on original reporting. We expect Washington Post reporters to see as much as they can of the story they are reporting, and to talk to as many participants as possible. Reporters should consider the advantages of reporting from the scene of events they are covering whenever that is possible.
If a reporter was not present at a scene described in a story, the story should make that clear. Assertions that something actually happened although it was unseen by the reporter should be attributed, so the narrative device of describing an event as it was recounted to us by witnesses must include attribution. If we reconstruct statements or exchanges between people based on the recollections of those people or witnesses who heard them speak, we must attribute those recollections transparently. If you are unsure about the application of these guidelines in a particular situation, discuss it with your editors.
In some circumstances where a source has allowed us to see something that reporters would not otherwise be able to observe, special problems of attribution may arise. They should always be discussed with editors. It is not always necessary to interrupt a narrative constantly to attribute small details to specific sources. It is sometimes possible to attribute the details of a narrative in a single sentence or paragraph. Here are two examples of acceptable ways to attribute information efficiently. Both appeared in the paper:
1) “The information in this story came from interviews with several administration officials, direct observation and discussions with outside advisers privy to debates inside the government.”
2) “This account is drawn from interviews in November and December in Mogadishu with 19 Somalis – nearly half of them militia fighters – and 30 U. S. Army soldiers. Additionally, more than two dozen U. S. military personnel from the special task force sent to Somalia to capture Aideed consented to interviews this month in the United States, and several wounded soldiers were interviewed in October at the U. S. Army hospital in Landstuhl, Germany. In some cases, individuals agreed to talk on condition they not be identified. Although American and Somali versions coincide to a large extent, some accounts are impossible to verify independently.” This kind of collective attribution is not appropriate in routine news stories, but in takeouts, investigative reports and other special stories it can serve readers well. Before using such a device, discuss it with your editor. We have to be sure such attribution is honest, full, and appropriate whenever we use it.
Any significant reporting by a stringer, staff member, or other Post employee should be credited in a byline or a tagline at the end of a story. When such people take notes from broadcasts of news events on radio or television, conduct basic research or check routine facts, they need not be credited.
Journalistic ground rules can be confusing, but our goal is clarity in our dealings with sources and readers. This means explaining our ground rules to sources, and giving readers as much information as possible about how we learned the information in our stories. If a source is not on the record, it is important to establish ground rules at the beginning of a conversation. In a taped interview, it is preferable for the discussion of ground rules to be on the tape. We strongly prefer on-the-record interviews to all other types, but we recognize that getting sources on the record is not always possible. When it is not, we owe readers explanations as to why not, as discussed above.
We should start virtually all interviews with the presumption that they are on the record. Inexperienced sources – usually ordinary people who unexpectedly find themselves the news – should clearly understand that you are a reporter and should not be surprised to find themselves quoted in the newspaper.
In establishing ground rules, the following are The Washington Post’s definitions of various forms of attribution. People use these terms to mean different things, so if your dealings with a source are going to be anything other than “on the record”, you should have a discussion to clarify the terms before you begin an interview.
On the record: For quotation, attributable to the source by name.
On background, or not for attribution: These both mean the same thing: information that can be attributed to “a police department official” or “a player on the team” who is not named. We must be careful, when dealing with sources who say they want to provide information “on background,” to explain that to us that means we can quote the statement while maintaining the confidentiality of the source. Some sources will try to negotiate the terms of art in “background” attribution – for example, a State Department official may ask to be identified as “an administration official.” We should try to put the reader’s interest first. In a story about a fight between the Pentagon and the State Department, for example, quoting “an administration official” is useless to readers. Use good judgment, and press for maximum revelation in attribution.
Deep background: This is a tricky category, to be avoided if possible. Information accepted on “deep background” can be included in the story, but not attributed. That means there is no way to help readers understand where it is coming from, which is why we discourage the use of deep background. You can also use information received on deep background as the basis for further reporting.
Off the record: This is the trickiest of all, because so many people misuse the term. By our definition, off-the-record information cannot be used, either in the paper or in further reporting. But many sources, including some sophisticated officials, use the term when they really mean “not for attribution to me.” We must be very careful when dealing with sources who say they want to be “off the record.” If they mean “not for attribution to me,” we need to explain the difference, and discuss what the attribution in the paper will actually be. If they really mean off the record as we define the term, then in most circumstances, we should avoid listening to such information at all. We do not want to be hamstrung by a source who tells us something that becomes unusable because it is provided on an off-the-record basis.
One alternative to off-the-record is “for guidance.” A source may be willing to give us information for our guidance or to prompt further reporting, on the understanding that we will not use his comments as the basis for putting something in the paper. This, for example, was the relationship between Deep Throat and Bob Woodward and Carl Bernstein. Such guidance can be useful even if we can’t print it at once.
We do not allow sources to change the ground rules governing specific quotations after the fact. Once a quote is on-the-record, it remains there.
Sometimes sources will agree to be interviewed only if we promise to read quotations we plan to use back to the source before they are published. This can create difficult situations. We do not want to allow sources to change what was said in the original interview, but sometimes that cannot be avoided, or can be avoided only at the cost of losing an on-the-record quote from an important source. If you find yourself in this gray area, consult with your editor.
Some reporters have read stories back to sources before publication to insure accuracy on technical points or to try to catch any errors before they appear in the paper. For a science writer to read a story, or passage, about a complex subject to a source to make sure it is accurate is a routine occurrence. But it is not our policy to routinely read stories or parts of stories to sources or to share copy with outsiders before it has been fully edited by us. A reporter who isn’t sure whether to read something to a source before publication should consult first with his or her editor.
Datelines are used to indicate where the substantive information in a story was actually gathered, not where the events described in the story took place. We do not use a Hagerstown dateline because we have talked to Hagerstown residents on the telephone; we have to be there to use the Hagerstown dateline. Datelines that actually include a date may be used only if at least one of the authors of the story was present in the datelined location on the day mentioned. A reporter in one remote location writing a story about events in another remote location cannot use the dateline of the place he is writing about. So, for example, a bureau reporter in Chicago writing a news story from there about events in Michigan must use a Chicago dateline, or no dateline at all. Editors should decide which is more appropriate.
If a story has two bylines and it carries a dateline, with or without a date, only one reporter needs to have been in the place named in the dateline, provided a substantial amount of reporting and writing was done in that place. In that case, the byline of the reporter on the scene should come first at the top of the story. If the reporter on the scene did not do enough of the work on the story to justify having the first, principal byline, then the story should not carry a dateline. Such stories should include an italic note at the end explaining who reported from where. An italic note should also be used to give readers the names and locations of any other significant contributors, including stringers, staffers and other Post employees. If a reporter must leave a city before filing a story, as often happens during wars and political campaigns, the dateline may still be used if the reporter was in that place on the date mentioned in the dateline. If the story is held for a later date, it can still carry the name of the town or city from which it was reported and written, but not a date.