Sunday, June 12, 2011

Grading PolitiFact (Georgia): Herman Cain and hiring Muslims

Words matter -- We pay close attention to the specific wording of a claim. Is it a precise statement? Does it contain mitigating words or phrases?
--Principles of PolitiFact and the Truth-O-Meter

The issue:

(image clipped from PolitiFact.com)


The fact checkers:

Willoughby Mariano:  writer, researcher
Jim Denery:  editor
Jim Tharpe:  editor


Analysis:

PolitiFact's fact check in this case consists of a shell game.  Whether or not the PolitiFact team consciously planned that type of result I cannot say.

The pea:  Herman Cain will not appoint Muslims.

Shell No. 1 (bold emphasis added):
(A) blogger for liberal ThinkProgress.org questioned him at the Conservative Principles Conference in Des Moines, Iowa.
"Would you be comfortable appointing a Muslim either in your Cabinet or as a federal judge?" the blogger asked.

"No, I will not," Cain replied. "And here’s why. There is this creeping attempt, there’s this attempt to gradually ease Sharia law and the Muslim faith into our government. It does not belong in our government."
Shell No. 2 (bold emphasis added):
This brings us back to Beck’s radio show, where Cain said his statement was "misconstrued."

"[The reporter] said, would you be comfortable with a Muslim in your Cabinet?" Cain told Beck. "And I immediately said, without thinking, ‘No, I would not be comfortable.’ I did not say that I would not have them in my Cabinet. Because if you look at my career, I have hired good people regardless of race, religion, sex, gender or orientation and this sort of thing."
Shell No. 3 (bold emphasis added):
The Monday after the news broke, Cain recounted what he said on Fox News’ "Your World with Neil Cavuto."

"A reporter asked me ‘Would I appoint a Muslim to my administration?’ I did say ‘no,’ " Cain told Cavuto.

"And here’s why ... I would have to have people totally committed to the Declaration of Independence and the Constitution of the United States. And many of the Muslims, they’re not totally dedicated to this country," he said. 
 Shell No. 4 (bold emphasis added):
In April, Cain repeated he would not hire a Muslim to radio host Bryan Fischer, who is also a conservative. We found an excerpt from the show on a website critical of the political right:

"[T]he comment I made that became controversial, and my staff keeps hoping will die, is that I wouldn’t have Muslims in my administration. And it’s real simple. The Constitution does not have room for Sharia law ... and to introduce that element as part of an administration when we’ve got all of these other issues, I think I have the right to say that I won’t," Cain said.  
The magicians at PolitiFact first show us shells No. 1 and No. 2, telling us that there is apparently no pea under either shell:
Indeed, the question the ThinkProgress.org blogger asked Cain was whether he would be "comfortable" with a Muslim in his Cabinet, not whether he would appoint one. If you take the video on its face, the explanation Cain gave on Beck’s show seems reasonable.
But PolitiFact informs us there is supposedly a problem.  There is a pea under shell No. 3 and another under shell No. 4--and apparently there was one under No. 1 after all:
(C)ontrary to his claim on Beck’s program, Cain did say he would not have Muslims in his Cabinet. Not once or twice, but three times in as many weeks to ThinkProgress.org, Cavuto and Fischer.
Consider PolitiFact's approach to this fact check.  The fact check team finds the claim from ThinkProgress consistent with the claim from the Beck program.  But they also find the ThinkProgress claim consistent with the statements from the Cavuto and Fischer programs.  The supposed consistency of the ThinkProgress material with the statements from each of the other venues individually shows that the original comment was ambiguous.

PolitiFact fails to plainly admit the ambiguity of the first comment.  Instead, it gets lumped in as a claim contrary to Cain's statement to Glen Beck.  But Cain's ThinkProgress and Beck comments cannot be both contradictory and non-contradictory.  That's a contradiction.  Yet that's exactly what PolitiFact concludes in the course of the fact check.

PolitiFact's logic fails

With PolitiFact's contradiction set aside we can more easily follow the movements of the entity manipulating the shells.

PolitiFact interprets two of Cain's statements (shell No. 3 and shell No. 4) as inconsistent with his statement to Beck (shell No. 2).  PolitiFact concludes that the statement to Beck was ridiculously false.  But somehow PolitiFact accomplished that feat without evaluating the other two statements as to their truth value.  If the other two statements are not true then the conclusion doesn't follow.

PolitiFact can't even appeal to its fallacious "burden of proof" criterion since Cain ought to have the burden of proof for both statements.  In this case, the structure of the story suggests that PolitiFact arbitrarily ruled two of Cain's statements true in order to find the fact-checked statement false.  In principle, one may obtain clues as to the truth values for a pair of conflicting statements, but this story displays no specific evidence of that type of reasoning.

What was Cain saying?

This analysis sets aside the question of whether Cain could successfully distinguish between hiring a Muslim as itself a political act and hiring a Muslim as the result of hiring the best available person for a given job.  The latter understanding, in fact, makes the original question from the ThinkProgress blogger nearly incomprehensible.  It is foolish to hire a person one does not trust--regardless of his religious persuasion.

In the end, Cain's statements parse with difficulty because of ambiguity.  PolitiFact avoids the problem by assuming a lack of significant ambiguity.  That approach is inappropriate in fact checking but ordinarily readily acceptable in opinion journalism.  PolitiFact is supposed to represent the former.


The grades:

Willoughby Mariano:  F
Jim Denery:  F
Jim Tharpe:  F

Herman Cain was using language capable of significant nuance because of its ambiguity, somewhat similar to the distinctions often drawn between "listening" and "hearing."  PolitiFact ignores those distinctions.  Worse, PolitiFact takes statements by Cain that might have appropriately received treatment as a flip-flop and instead rules one of the statements as false with no other justification than the discrepancy.


Afters:

Some of the non-objective writing in this piece deserves special attention:
Metro Atlantans are accustomed to Cain as an agitator. He’s been goading liberals for years as a conservative talk show host on AM 750 and 95.5FM News/Talk WSB.
That, ladies and gentlemen, is the type of language that opinion journalists use.

A reporter would make sure the assessment above came from a third-party source, or would at least provide some sort of evidence in support of the assessment.  In this case we get none of that.  Instead, the PolitiFact team simply provides readers its own assessment of Cain's style on the radio.

Cain's radio colleague, Neal Boortz, does agitate and goad liberals.  But Cain's style, at least in my experience, is conversational and non-confrontational.  Based in part on PolitiFact Georgia's previous ratings of Cain, there is reason to doubt whether the PolitiFact team has ever listened to Cain's radio program.  It would not surprise me if Mariano and company conducted their research on Cain's radio career by asking around in the newsroom and taking the resulting poll as a suitable summary of Cain's radio career.  Don't buy the PolitiFact portrait of Cain on faith.

The following YouTube video shows Herman Cain engaging President Bill Clinton on Clinton's health care plan.  The video serves as a good example of Cain's style of communication.



If anything, the video shortchanges Cain with respect to the humor and good will he used to good effect on the radio.

No comments:

Post a Comment

Please remain on topic and keep coarse language to an absolute minimum. Comments in a language other than English will be assumed off topic.