Now you see it, now you don't
Matthew Yglesias quotes Barack Obama as saying:
By reporting that Iran halted its nuclear weapon development program four years ago because of international pressure, the new National Intelligence Estimate makes a compelling case for less saber-rattling and more direct diplomacy. The juxtaposition of this NIE with the president's suggestion of World War III serves as an important reminder of what we learned with the 2002 National Intelligence Estimate on Iraq: members of Congress must carefully read the intelligence before giving the President any justification to use military force.
There's a very subtle dig here at Hillary Clinton and John Edwards, noting that both of them, unlike you or I or Barack Obama, had access to the classified version of the 2002 NIE on Iraq, a document that debunked substantial elements of the administration's case for war, but which neither Clinton nor Edwards (nor a great many other members of congress) bothered to read before voting to authorize the use of force.
But what would they have read in the earlier report? Norman Podhoretz describes what is publicly known.
This latest NIE “judges with high confidence that in fall 2003 Tehran halted its nuclear weapons program”; it “judges with high confidence that the halt was directed primarily in response to increasing international scrutiny and pressure resulting from exposure of Iran’s previously undeclared nuclear work” ... These findings are startling, not least because in key respects they represent a 180-degree turn from the conclusions of the last NIE on Iran’s nuclear program. For that one, issued in May 2005, assessed “with high confidence that Iran currently is determined to develop nuclear weapons” and to press on “despite its international obligations and international pressure.”
What changed the picture was new information, not the fact that people had developed better habits at reading NIEs. The content of the current NIE was different from the content of the old NIE. The AFP reports:
President George W. Bush said Tuesday that a "great discovery" as recently as August prompted the US intelligence community's stunning reversal of its long-held view that Iran had an active nuclear weapons program.
Bush provided no details on the nature of the new intelligence, which set off an in-depth intelligence review of the evidence and assumptions that underpinned a 2005 assessment, which had held with "high confidence" that Iran was determined to acquire nuclear weapons.
Mike McConnell, the director of national intelligence, came to Bush in August and said: "We have some new information."
"He didn't tell me what the information was. He did tell me it was going to take a while to analyze," Bush said at a White House news conference in describing the encounter.
Granting that the "new information" actually exists and the intelligence process hasn't been corrupted to the point where information is custom-manufactured to suit a political agenda, revising an estimate is exactly what analysts should do when in possession of new facts. In other words, if we grant Mike McConnell the benefit of the doubt with respect to his integrity, revising the estimate is what his duties compelled him to do.
The deeper question, one raised by Norman Podhoretz and Michael Ledeen at Pajamas, is whether the "new information" isn't politically driven or otherwise invalid. I obviously can't say. What is known from the press, however, suggests that the information is "new information" come to light only in August, "great discovery" of such detail that it "was going to take a while to analyze".
What does this suggest?
In intelligence analysis information is not only examined for comprehension but for reliability and credibility. In other words the intelligence community going to look at how credible this "new information" is and which parts, if any, are believable. Apparently the initial take gives them "great confidence". But it's always good to remember that only a few years earlier they had "great confidence" that the opposite was true.
And then there is the question of interpretation. Israeli intelligence believes that the Iranian nuclear weapons program stopped for a while, then it got going again.
In Israel, Defense Minister Ehud Barak said "it's apparently true" that Iran stopped pursuing its military nuclear program in 2003. "But in our opinion, since then it has apparently continued that program," Barak told Army Radio. "There are differences in the assessments of different organizations in the world about this, and only time will tell who is right."
"Only time will tell who is right," is a charming way to put it. That puts me in mind of song about finding out if something is there or not by waiting to see what happens.
My Bonny peered into the gas tank
The heart of its contents to see.
I lighted a match to assist her,
Oh bring back my Bonny to me.
Leaving aside questions of intelligence corruption, failure etc, partisan politics by looking for consistency in estimates has damaged the way the public should look at intelligence. The intelligence picture changes all the time. Any student of military history knows that grease pencils are used on map overlays because information is constantly updated. Initial reports are often inaccurate. Corrections are messaged in. If battle maps were treated like NIEs they would be inked into the map itself instead of marked in grease pencil on plastic overlays. A politician's idea of a battlefield situation map would be one with enemy formations already typeset into the printing plates before being issued to the troops.
There are many reasons to be concerned about recent changes in the NIE estimate, though sadly, few of them have been raised by the politcians. The first is: how could the intelligence community have been so wrong? A revision of this magnitude indicates a blind spot or inadequacy which needs to be addressed. Any manager faced with the prospect of having to revise his balance sheet 180 degrees around would probably ask his accounting department how such a thing could happen. The second question is if the same standards of reliability failed so miserably in detecting what is now regarded as wrong information in the past, then how can we be so sure of it's ability to judge new information now? Think of it: politicians are quite ready to believe as gospel truth statements the diametrical opposite of statements made by the same intelligence agencies. Would they be ready to believe a new estimate that flipped 180 degrees again if newer information became available in a few weeks time? I would, if I had confidence in the process. If not, I'd be looking to fix the process. These are the questions I'm interested in, though like the rest of the public, I'll have to live another 10 years before I read about the explanation in one of Bob Woodward's books. That's where the truth eventually winds up doesn't it? Doesn't it?