Send a draft of your story to a source? Some journalists do it — but be mindful of the risks.

A Washington Post reporter’s decision to share a draft of a story with his sources is provoking a healthy discussion in newsrooms about when, if ever, it’s wise to circulate unpublished material for comments.

As originally reported in the Texas Observer, a Post education writer gave the media-relations department at the University of Texas at Austin an opportunity to review and suggest changes to a March 14 story about standardized testing. The story focused on an exam designed at UT-Austin to measure how much college students learn and retain, a practice with some prominent detractors. (The reporter’s exchanges with UT showed up in emails gathered by the Observer under Texas’ Public Information Act.)

According to The Poynter Institute, the Observer‘s report is causing the Post to reevaluate and tighten its standards for when letting a source preview a story is acceptable.

While it’s difficult to say “never, ever,” a story should never — ever — be sent to anyone outside the newsroom without careful consideration of the legal and ethical perils.

The main one is this: Once a reporter establishes a certain fact-checking protocol, then any deviation from that fact-checking protocol risks appearing, to a judge and jury, like a lapse in diligence. Lapses in diligence are, under the law of defamation, a bad thing.

If a jury is convinced that a reasonably thorough reporter would have shown the unpublished story to the source in advance — and that showing the story to the source would have prevented a serious error — then failing to have the story pre-reviewed could appear negligent.

An aggressive libel lawyer could tie a reporter in knots on the witness stand with this line of questioning:

“Isn’t it true, Mr. Reporter, that you have shown stories to sources on numerous occasions when you were concerned about accuracy?”

“And isn’t it true, Mr. Reporter, that you failed to follow your routine practice with this story?”

“And isn’t is true, Mr. Reporter, that you would have shown this story to my client if you cared about getting the facts right?”

You see how uncomfortable this could get. It is safest to be able to testify truthfully that circulating unpublished drafts is against policy and is never done.

Emailing a draft back and forth also invites the risk of creating a paper trail — perhaps an incomplete and misleading one — that could be manipulated to suggest carelessness with the facts.

Suppose the source writes back, “You can’t say XYZ, that’s false.” The journalist carefully retraces his research, calls additional sources, convinces himself that XYZ is true and publishes it anyway.

And it turns out to be wrong.

The email now becomes Exhibit A that the journalist disregarded red flags that the fact was inaccurate. Under the law of defamation, disregarding red flags is a bad thing.

If a highly controversial story is to be shared in advance, then it is safest to share it with all sides in the controversy. If the story criticizes Acme Corp., and it was shown only to the company’s critics, then — if Acme is able to prove the story false — the one-sided reviewing process might be used as evidence of a predisposition to shade the facts.

And if a story is to be shared in advance, then the comments accompanying it should be made with great care. A stray remark in the margin of a draft, like “I’m not sure this is right,” could indicate that a fact was published with doubts about its veracity.

Of equal or greater concern than libel is the risk that distributing unpublished material outside of the newsroom might result in waiving a claim of reporter’s privilege.

The reporter’s privilege enables journalists to lawfully refuse demands from attorneys or law enforcement for confidential newsgathering material. The privilege always covers material leading to the discovery of confidential sources, but in many states the privilege goes even further and protects all unpublished material — notes, photos, interview recordings and, yes, drafts of stories.

Privilege can be waived if the person seeking access to the journalist’s material demonstrates that the material was shown to outsiders beyond the newsroom. That is exactly what happened to student journalists at Northwestern University, when an Illinois judge ruled that their internal memos prepared in connection with investigating a potentially wrongful murder conviction had to be turned over to prosecutors, because privilege was broken when the memos were sent to the inmate’s defense lawyers.

Of course, journalists can make careless errors in verbal interactions with sources just as as they can make mistakes in emailing around their drafts. Verbally spilling an unpublished confidence can waive privilege. Verbally expressing disregard for the truth can be used as evidence in a defamation suit.

Still, written communications are more durable than witness’ memories, and are likely to carry more persuasive value with jurors. Every interaction with sources should always be a cautious and professional one. But a written interaction is more likely to backfire hurtfully.

Half-measures short of sharing entire drafts are more widely accepted in the industry. Reading back quotes over the phone if there is uncertainty, or reviewing limited passages of stories with experts if the material is unfamiliar and technical, are generally safer alternatives.

A final takeaway from the Texas Observer story: Journalists who correspond via email with government officials should be mindful that those emails can show up in a freedom-of-information request and get published. Comments that might sound perfectly fine during an informal hallway chat with a source can, fairly or unfairly, take on an entirely new life when memorialized in writing. Journalists have been zinging public officials for years over squirm-inducing emails sent in thoughtless moments, and they should be prepared to be held to the same standard themselves.