iOS app Android app More

Jay Rosen

Jay Rosen

Posted April 16, 2009 | 11:35 AM (EST)

"He Said, She Said" Journalism: Are We Done With That Yet?


There I am, sitting at the breakfast table, with my coffee and a copy of the New York Times, in the classic newspaper reading position from before the Web. And I come to this article, headlined "Ex-Chairman of A.I.G. Says Bailout Has Failed." I immediately recognize in it the signs of a he said, she said account.

Quick definition: "He said, she said" journalism means...

* There's a public dispute.
* The dispute makes news.
* No real attempt is made to assess clashing truth claims in the story, even though they are in some sense the reason for the story. (Under the "conflict makes news" test.)
* The means for assessment do exist, so it's possible to exert a factual check on some of the claims, but for whatever reason the report declines to make use of them.
* The symmetry of two sides making opposite claims puts the reporter in the middle between polarized extremes.

When these five conditions are met, the genre is in gear. The he said part might sound like this:

Mr. Greenberg asserted that he would have reduced or at least hedged A.I.G.'s exposure to credit-default swaps in 2005, when A.I.G.'s credit rating was reduced.


"A.I.G.'s business model did not fail; its management did," he asserted.

Followed by the "she" said...

That provoked another scornful counterattack from his former company, saying that Mr. Greenberg's assertions were "implausible," "not grounded in reality" and at odds with his track record of not hedging A.I.G.'s bets on credit-default swaps.

I had read enough of the Times coverage of Mr. Greenberg to wonder why the editors would run something so lame. Their business columnists have been (excuse the expression) kicking ass on meltdown coverage, including A.I.G. But here there was no attempt to assess clashing truth claims, even though Times journalism was available to do just that. Instead Hank Greenberg got to star in a game of "you say black, I say white."

It seemed strange to me that in 2009 stories like that were still being waved on through. On Twitter I sometimes talk to Ryan Chittum, who writes The Audit column for Columbia Journalism Review. It's a running critique of the business press after the banking meltdown. So I asked Ryan, "is this the best the Times can do?" because he knows a lot more about the coverage than I do. A few hours later he answered me at CJR.

This one's easy: No. The Times's story offers no analysis and forces readers--95 percent of whom know little or nothing about Greenberg's tenure at AIG--to try to guess who's right.

Which is why these stories are so frustrating: we're left helpless by them. I want to quote the rest of his judgment because it helps nail down what is meant by he said, she said, not just at the New York Times, which has no special purchase on the form, but anywhere. The means are available to do better, but these are not employed. Chittum:

There's no attempt to try to separate out who's right here, even though everybody but Hank Greenberg knows he has major responsibility for driving AIG into the ground.


Here's some stuff that helps explain why. I just culled it from the excellent Washington Post three-parter on AIG in December (if you haven't read that yet, make sure you do):

He created the Financial Products division in 1987 with traders from soon-to-be disgraced Drexel Burnham Lambert, approved its entry into the credit-default swap market in 1998, empowered Joseph Cassano, oversaw FP when it set up "sham" companies that resulted in tens of millions in fines, was an unindicted co-conspirator in a huge fraud at AIG, oversaw the company's credit downgrade from AAA, was in charge when half of the company's $80 billion in CDS on subprime CDOs were written. Apparently, Cassano and FP stopped issuing CDS within months of Greenberg's exit in 2005.

How much more evidence do you need to tell your readers that this guy has significant responsibility for the disaster that came to his his company and the entire economy--to not let him spin away?

"How much more evidence do you need?" is the kind of exasperation a lot of us have felt with what he calls "false balance," which is another name for the pattern I'm describing.

So far so good. I told you what he said, she said is, and gave you an example. CJR chimed in, and told the New York Times it could do way better, showing how. Press criticism lives! (Twitter helps.) But this does not tell us why he said, she said reporting still exists, or ever existed. To understand that we have to cut deeper into news practice, American style.

Turn the question around for a moment: what are the advantages of the newswriting formula I have derisively labeled "he said, she said?" Rather than treat it as a problem, approach it as a kind of solution to quandaries common on the reporting trail. When, for example, a screaming fight breaks out at the city council meeting and you don't know who's right, but you have to report it, he said, she said makes the story instantly writable. Not a problem, but a solution to the reporter's (deadline!) problem.

When you kinda sorta recall that Hank Greenberg is a guy who shouldn't necessarily get the benefit of the doubt in a dispute like this, but you don't know the history well enough to import it into your account without a high risk of error, and yet you have to produce an error-free account for tomorrow's paper because your editor expects of you just that... he said, she said gets you there.

Or when the Congressional Budget Office issues a report on ethanol and what it's costing us in higher food prices, the AP reporter to whom the story is given could just summarize the report, but that's a little too much like stenography, isn't it? So the AP adds reactions from organized groups that are primed to react.

This is a low cost way of going beyond the report itself. A familiar battle of interpretations follows, with critics of ethanol underlining the costs and supporters stressing the benefits. Of course, the AP could try to sort out those competing claims, but that would take more time and background knowledge than it probably has available for a simple "CBO report issued" story. "Supporters of ethanol disagreed, saying the report was good news..." gets the job done.

These are some of the strengths of the he said, she said genre, a newsroom workhorse for forty years. (Think it's easy? You try making any dispute story in the world writable on deadline...)

The best description I've read of the problem to which devices like he said, she said are a solution comes from former Washington Post reporter Paul Taylor, who covered national politics. Here's a comment about it that I left at the New York Times Opinionator blog. It was an attempt to explain a phrase I use to describe the kind of distortion that he said, she said can produce: "regression toward a phony mean."

Journalists associate the middle with truth, when there may be no reason to.


In his 1990 book, See How They Run, former Washington Post reporter Paul Taylor (once seen as heir to David Broder) explained why regression toward a phony mean is so common in journalism. It answers to a need for what he calls "refuge." Here is what he said:

"Sometimes I worry that my squeamishness about making sharp judgments, pro or con, makes me unfit for the slam-bang world of daily journalism. Other times I conclude that it makes me ideally suited for newspapering- certainly for the rigors and conventions of modern 'objective' journalism. For I can dispose of my dilemmas by writing stories straight down the middle. I can search for the halfway point between the best and the worst that might be said about someone (or some policy or idea) and write my story in that fair-minded place. By aiming for the golden mean, I probably land near the best approximation of truth more often than if I were guided by any other set of compasses- partisan, ideological, pyschological, whatever... Yes, I am seeking truth. But I'm also seeking refuge. I'm taking a pass on the toughest calls I face."

Clearly, there can be something extreme about this squeamishness, too. Clearly, the desire for refuge can get out hand. Writing the news so that it lands somewhere near the "halfway point between the best and the worst that might be said about someone" is not a truthtelling impulse at all, but a refuge-seeking one, and it's possible that this ritual will distort a given story.

Like the "straight down the middle" impulse that Taylor writes about, he said, she said is not so much a truth-telling strategy as refuge-seeking behavior that fits well into newsroom production demands. "Taking a pass" on the tougher calls (like who's blowing more smoke) is economical. It's seen as risk-reduction, as well, because the account declines to explicitly endorse or actively mistrust any claim that is made in the account. Isn't it safer to report, "Rumsfeld said...," letting Democrats in Congress howl at him (and report that) than it would be to report, "Rumsfeld said, erroneously..." and try to debunk the claim yourself? The first strategy doesn't put your own authority at risk, the second does, but for a reason.

We need journalists who understand that reason. And I think many do. But a lot don't.

He said, she said reporting appears to be risk-reducing, but this is exactly what's changing on the press. For a given report about, say, former counter-terrorism official Richard Clarke, "the halfway point between the best and the worst that might be said about someone" is no more likely to be accurate than the one-fifth mark, especially when you factor in the reality of the Overton Window and the general pattern we know as "working the refs." The halfway point is a miserable guideline but it can sound pretty good when you are trying to advertise to all that you have no skin in the game. This is how I think of he said, she said reporting. Besides being easy to operate, and requiring the fewest imports of knowledge, it's a way of reporting the news that advertises the producer's even handedness. The ad counts as much as the info. We report, you decide.

"Ex-Chairman of A.I.G. Says Bailout Has Failed" was a text most likely intended for the print edition of the New York Times business pages. The newswriting formula that produced it dates from before the Web made all news and reference pages equidistant from the user. He said, she said might have been seen as good enough when it was difficult for others to check what had previously been reported about the ex-chairman of A.I.G., but that is simply not the case for a New York Times reporter in April, 2009.

There has been a loss of refuge. And this is why he said, she said journalism is in decline, even though you still see plenty of it around. Today, any well informed blogger, competing journalist or alert press critic can easily find the materials to point out an instance of false balance or the lame acceptance of fact-free spin. Professional opinion has therefore shifted and among the better journalists, some of whom I know, it is no longer acceptable to defend he said, she said treatments when the materials are available to call out distortions and untruths. (That doesn't mean the practice has halted; I'm talking about a shifts in the terms of legitimacy among journalists, and about efforts like this.)

In fact, it's taken a long time to get to this point. Back in 2004 setting a higher standard than he said, she said was still a novel idea. Chris Mooney wrote about it in the context of science coverage under Bush. ("How 'Balanced' Coverage Lets the Scientific Fringe Hijack Reality.") As CJR's Campaign Desk noted...

The candidate makes a statement. You write it down, then you call the other side for a response. It's one of journalism's fundamentals. Tell us what he said, tell us what she said, and you're covered, right?


Well, no. Given the amount of spin this election year, the old rules don't apply any more. Campaign Desk herewith proposes a new ground rule: "He said/she said/we said."

... With a variety of Internet research tools readily at hand, it has never been easier for reporters to draw an independent assessment on any given day of who is right, who is wrong, and in what way.

The tools are there to make an independent assessment of who is right: for journalists, that is the critical point. (See also my post from 2004, He Said, She Said, We Said and Rethinking Objectivity by Brent Cunningham from 2003.) Because of that--and because of working the refs, the Overton Window, the failures of the political press under Bush--he said, she said no longer has the acceptance rates it once did. Which is why it was so easy to get Ryan Chittum to answer my question, "is this the best the Times can do?"

It wasn't. And it's easier than ever to show that. More people are involved in showing it, too. This raises the question of whether a he said, she said treatment loses you more in user disgust with your lameness than any informational gain in having fresh news to report about Hank Greenberg trading barbs with A.I.G. Do people want to feel helpless in sorting out who's bullshitting them more? Is that the news media's role, to increase that feeling? Is such a practice even sustainable in the Web era?

That it may not be (and the industry knows it) is shown by what The Politico called a "high-stakes experiment" at the AP's Washington bureau. The plan was to move "from its signature neutral and detached tone" to a more aggressive style of newswriting that bureau chief Ron Fournier calls "cutting through the clutter."

In the stories the new boss is encouraging, first-person writing and emotive language are okay.


So is scrapping the stonefaced approach to journalism that accepts politicians' statements at face value and offers equal treatment to all sides of an argument. Instead, reporters are encouraged to throw away the weasel words and call it like they see it when they think public officials have revealed themselves as phonies or flip-floppers.

In other words, we can't skate by on he said, she said any more. Call it like they see it is, in fact, a successor principle but this means that AP reporters are now involved in acts of political judgment that can easily go awry, and their own politics can be at issue.

Time to wrap this up.

Part of the problem is that American journalism as an occupational scene has never gone for the candor Paul Taylor showed in his comments on searching for the halfway point between the best and the worst that might be said. The pro system talks about the reporting of news as a truth-telling enterprise, but not a difference-splitting or dilemma-disposing one. It says: we're the source of "the most authoritative news coverage," as the AP recently put it. But it rarely mentions the refuge-seeking part, which subtly undermines that authority.

As I tried to explain in Why Campaign Coverage Sucks (published at TomDispatch.com and Salon, January 2008) there is an "innocence agenda" at work in the mainstream press. It favors certain practices:

Who's-gonna-win is portable, reusable from cycle to cycle, and easily learned by newcomers to the press pack. Journalists believe it brings readers to the page and eyeballs to the screen. It [plays] well on television, because it generates an endless series of puzzles toward which journalists can gesture as they display their savviness, which is the unofficial religion of the mainstream press.


But the biggest advantage of horse-race journalism is that it permits reporters and pundits to play up their detachment. Focusing on the race advertises the political innocence of the press because "who's gonna win?" is not an ideological question. By asking it you reaffirm that yours is not an ideological profession.

In its heyday he said, she said was like a stamping plant in the factory of news. It recognized that production demands trumped truthtelling requirements. But these were the production demands of a beast that is now changing. Refusing to serve as a check on Hank Greenberg's power to distort the news when the means for a such a check are available-- this too can have a cost, just as importing the knowledge to do the check has a cost. At a certain point in this dynamic, he said, she said journalism loses its utility and becomes one of the things dragging the news business down. But as the industry sheds people and newsrooms thin out, there could be greater reliance on a more and more bankrupt and trust-rotting practice. That's a downward spiral.

Criticism of he said, she said practices and the flippancy that comes with it should therefore continue. The other day, Paul Kane of the Washington Post said it was too much to expect him to import into his account the background knowledge that a Republican Senator warning about the dangers to Senate comity of proceeding with only 50 votes had voted to do the same thing when her party held the majority but not 60 votes. (Matthew Yglesias picked up on it.)

Kane said he was astonished by this demand; he couldn't figure out where it was coming from. "We reported what Olympia Snowe said. That's what she said. That's what Republicans are saying. I really don't know what you want of us."

If he's not just blowing smoke, and he really doesn't know-- that is a problem for the Washington Post.