Malou Innocent, a foreign policy analyst at the Cato Institute, writes on The National Interest’s The Skeptics blog that commentators “buzzing” over the WikiLeaks revelations that some Arab leaders appear to be in favor of U.S. military action Iran are misinterpreting both the messages contained in the WikiLeaks cables and overlooking recent polling data from the region.

Innocent writes (my emphasis):

According to the Brookings Institution’s 2010 Arab Public Opinion Poll, 77 percent regard Israel as the biggest threat, 80 percent regard the United States as the biggest threat, and only 10 percent regard Iran as the biggest threat. Fifty-seven percent think the region will be better off if Iran had nuclear weapons. Despite those remarkably high numbers, it is the opinion of brutal Arab dictators who want America to bomb Iran that has dominated news stories. Meanwhile, the opinions of other Arab leaders in Oman, Qatar, Syria, Kuwait, and Jordan,who have expressed concern about the repercussions of an attack on Iran, have all but been ignored. Of course, it’s not a coincidence that the same news organizations overselling U.S.-Arab solidarity against Iran were the same ones that overhyped Iranian involvement during the Iraq War.

Innocent makes a valuable point that WikiLeaks cables offer far from conclusive evidence that the Arab world would support a U.S. or Israeli military strike on Iranian nuclear facilities. Inter Press Service‘s Gareth Porter addressed this very issue and took a critical look at the New York Times‘s reporting on the WikiLeaks cables.

He found:

In fact, the cables show that most Gulf Arab regimes – including Saudi Arabia itself – have been seriously concerned about the consequences of a strike against Iran for their own security, in sharp contrast to Israel’s open advocacy of such a strike. They also show the United Arab Emirates (UAE) and Kuwait expressing that concern with greater urgency in the past two years than previously.

Those facts were completely ignored, however, in the Times’ account.