Grading PolitiFact: Ann Coulter and terrorist attacks

The issue:

The fact checkers:

Louis Jacobson:  writer, researcher
Martha Hamilton:  editor

Analysis:

Comparing this fact check of Coulter with a recent fact check of President Obama helps illustrate PolitiFact's layered problem with selection bias.

In the case of Coulter, PolitiFact identifies a specific point to fact check and an underlying argument.  The underlying argument:

Is there a website somewhere listing everything that encourages terrorist recruiting?" She went on to knock Andrew Sullivan, an iconoclastic conservative pundit, for suggesting that if Obama was elected, his racial background and life story could help counter jihadist recruiting efforts.

"It didn't work out that way," Coulter wrote.

The fact check:

"There have been more terrorist attacks on U.S. soil by these allegedly calmed Muslims in Obama's first 18 months in office than in the six years under Bush after he invaded Iraq. Also, as I recall, there was no Guantanamo, no Afghanistan war and no Iraq war on Sept. 10, 2001. And yet, somehow, Osama bin Ladin (sic) had no trouble recruiting back then. Can we retire the 'it will help them recruit' argument yet?"

How I know what the fact check will supposedly be:

But the only part of her comment that's checkable is her claim that "there have been more terrorist attacks on U.S. soil by these allegedly calmed Muslims in Obama's first 18 months in office than in the six years under Bush after he invaded Iraq." So we looked into it.

Humor me while I sketch a comparison with the story about Obama.  The underlying argument:

President Barack Obama has been making a case against Republicans in Congress, hoping to help Democrats in the midterm elections.

One of his points is that the Democrats' policies on the economy are better than the Republicans.

"Along with tax cuts for the wealthy, the other party's main economic proposal is that they'll stop government spending," Obama said. "Of course, they are right to be concerned about the long-term deficit -– if we don't get a handle on it soon, it can endanger our future. And at a time when folks are tightening their belts at home, I understand why a lot of Americans feel it's time for government to show some discipline too."

 The fact check:

"But let's look at the facts. When these same Republicans –- including Mr. Boehner –- were in charge, the number of earmarks and pet projects went up, not down."

Again, how I know what will supposedly make up the fact check:

We knew earmarks had gone up during the Bush administration, but we weren't sure how they tracked Republican majorities in Congress. Republicans held control of Congress for most of the period from 1995 to 2006. (In 2001, Democrats briefly re-took control of the Senate but lost it again in 2003.) So we decided to check it out.

In Obama's case, PolitiFact stuck with its narrow mission, confirming that earmark expenditures were up while Republicans controlled Congress and only marking the president down to "Mostly True" because he named Boehner as part of the problem despite the fact that Boehner eschews earmarks.

Watch out for mission creep in Coulter's case.

PolitiFact confirmed Coulter's claim:

So, Coulter's numbers may be meticulously circumscribed, but they're right: There were zero attacks during the six years cited under Bush, and two attacks in the first 16 months under Obama.

Jacobson mentions no caveat parallel to Obama's name-dropping of Rep. Boehner. The underlying argument does seem to reappear, however:

But terrorism experts told us that the more important question is what those statistics signify. The answer may be: not much.

But that's not relevant when the only checkable part of her claim had to do with the number of attacks.  Is it?

Observe the conclusion:

Reviewing the evidence, Coulter's numbers are correct -- there have been more terrorist attacks on U.S. soil under Obama than under Bush. But this statistic is of questionable value. It's based on carefully constructed parameters that could have produced a different comparison if those parameters had been tweaked. It suggests that terrorist plots can be blamed on a specific presidency, even though plots can take years to come together. And it suggests that the only way to measure success in combating terrorism is by comparing deadly attacks on U.S. soil, even though "it's immensely relevant whether someone has become motivated to attack U.S. persons or interests anywhere in the world, including Afghanistan and Iraq," said Robert Chesney, a law professor and terrorism specialist at the University of Texas. Ultimately, we think the statement is accurate but doesn't provide needed context. We rate it Half True.

I thought the context consisted of claims by folks such as Andrew Sullivan that electing Obama would hamper recruiting efforts by radical Islamists.  And I thought PolitiFact determined that only Coulter's statement regarding the number of attacks was "checkable."  Instead, we get an expanded set of underlying arguments that supposedly undermine Coulter's truthfulness.

1)  "It's based on carefully constructed parameters that could have produced a different comparison if those parameters had been tweaked."

Yes, if the comparision had been different then it would have been a different comparison.  Sound the tautology alert.

2)   "It suggests that terrorist plots can be blamed on a specific presidency, even though plots can take years to come together."

This is amazing.  The above conclusion seems to appear out of thin air, as Coulter's point runs counter to it.  She argues, fairly or not, that people like Sullivan are nuts if they expect the identity of the president to have any serious effect on terrorist recruitment.  Coulter even points out in her column that terrorists launched attacks well before they had Guantanamo and the Iraq invasion to complain about.  Her column has nothing to do with supporting the notion that specific presidencies are to blame for terrorist plots.  It is the reverse, if anything.

3)  "And it suggests that the only way to measure success in combating terrorism is by comparing deadly attacks on U.S. soil ..."

Again, Coulter's column was not about success in combating terrorism.  It was about the argument that various activities enable terrorist recruitment.  That's it.  One might very well question whether Coulter's metric serves to significantly measure terrorist recruitment.  But once it is asserted that such things are not "checkable" does it make sense to use similar things in the final rating as PolitiFact ultimately does?

Regardless of the correct answer to that question, journalistic fairness arises as an issue when one fact check concerns itself with an underlying argument while another does not.  The Obama fact check mentioned above, written by Angie Drobnic Holan but sharing the same editor as the Coulter piece, specifically noted the broad underlying argument that Democrats were better on the economy and the narrower underlying argument that earmark spending served to indicate that Republicans were less apt to exercise fiscal restraint than Democrats.  PolitiFact opted not to explore whether Obama's facts reasonably supported those underlying arguments, and arguably lent support to Obama's underlying arguments by failing to question them.

The grades:

Louis Jacobson: F
Martha Hamilton: F-

The wild flights of fancy undertaken by this duo for the sake of faulting Coulter's underlying argument bury them in "F" territory. Hamilton warrants the dreaded minus with her "F" because she presided over the disparate treatment between Coulter and Obama. That's something the editor ought to take into account.