PDA

View Full Version : DNI to intel analysts - try talking to people outside the IC for a change



Entropy
07-31-2008, 02:43 AM
Someone finally got hit with the Captain Obvious club (http://www.fas.org/blog/secrecy/2008/07/dni_outside_ties.html):


In a new directive that challenges the insular culture of U.S. intelligence agencies, Director of National Intelligence J. Michael McConnell has ordered analysts to cultivate relationships with outside experts “whenever possible” in order to improve the quality of intelligence analysis.

The DNI’s July 16 directive on “Analytic Outreach” (pdf) establishes procedures for implementing such outreach, including incentives and rewards for successful performance.

“Analytic outreach is the open, overt, and deliberate act of an IC [intelligence community] analyst engaging with an individual outside the IC to explore ideas and alternate perspectives, gain new insights, generate new knowledge, or obtain new information,” the directive states.

“Elements of the IC should use outside experts whenever possible to contribute to, critique, and challenge internal products and analysis….”

“Sound intelligence analysis requires that analysts… develop trusted relationships” with “experts in academia; think tanks; industry; non-governmental organizations; the scientific world; …and elsewhere.”

Tom Odom
07-31-2008, 10:40 AM
Someone finally got hit with the Captain Obvious club (http://www.fas.org/blog/secrecy/2008/07/dni_outside_ties.html):

But what if they don't agree? :wry:


Good for the DNI.....

Maybe he should make them all watch, "Remember the Titans"

William F. Owen
07-31-2008, 12:00 PM
Considering the majority of intelligence agencies bureaucratic product is about supporting the policy maker, rather than informing him, I'm not sure how this will work.

It's like Newton observed about studying dung heaps. Interesting, but seldom useful. Likewise, intelligence should never seek to gather opinions, just facts.

Sorry to be a cynic!

Tom Odom
07-31-2008, 12:21 PM
Considering the majority of intelligence agencies bureaucratic product is about supporting the policy maker, rather than informing him, I'm not sure how this will work.

It's like Newton observed about studying dung heaps. Interesting, but seldom useful. Likewise, intelligence should never seek to gather opinions, just facts.

Sorry to be a cynic!

That may be your opinion but it is not fact and is not even close to the mark. As for gathering opinions, much of HUMINT is based on opinion. In a larger sense, an intelligence estimate from a battalion S2 to a national intelligence estimate is an opinion.

As for facts and their gathering, just deciding what is important enough to gather is very much an opinion. We call them collection plans.

Cynic is ok. Accurate cynic is much better in my opinion.

Tom

Ron Humphrey
07-31-2008, 12:37 PM
But if there isn't some kind of effort to make it much simpler to define whats ok for conversation vs not (and this needs to be well thought out and not hippocket )decided, then the first time someone gets slammed for sharing the wrong things or more likely someone decides after the fact that something was actually of value; game over:(

Entropy
07-31-2008, 02:03 PM
Considering the majority of intelligence agencies bureaucratic product is about supporting the policy maker, rather than informing him, I'm not sure how this will work.

It's like Newton observed about studying dung heaps. Interesting, but seldom useful. Likewise, intelligence should never seek to gather opinions, just facts.

Sorry to be a cynic!

I think it depends on what you consider a "fact." Are enemy doctrine, TTP, goals and motivations, etc. facts? A more useful term I'd use is "evidence."

There are other things to consider. Often facts are ambiguous or one set of facts will be contradicted by another. Without extensive subject matter knowledge, a fact is likely to be misinterpreted. Too often analysts don't recognize the importance of certain facts or evidence because they are ignorant of the nature of the problem.

Back when the intelligence target set was limited, the IC could hire and/or develop subject matter experts on almost every facet of a problem in-house. But today, the IC's problem set has expanded to such an extent that outside expertise is often a necessity. Some of this comes from open source material, but there's no substitute for meeting and talking with actual people.

Finally, most intel people don't like the term "opinion" and believe, justifiably I think, that what they do is more rigorous that mere opinion. Opinion needs no justification - an intelligence assessment or estimate has to be rigorously defended with supporting evidence and arguments clearly laid out.

William F. Owen
07-31-2008, 03:45 PM
That may be your opinion but it is not fact and is not even close to the mark. As for gathering opinions, much of HUMINT is based on opinion. In a larger sense, an intelligence estimate from a battalion S2 to a national intelligence estimate is an opinion.

As for facts and their gathering, just deciding what is important enough to gather is very much an opinion. We call them collection plans.

Cynic is ok. Accurate cynic is much better in my opinion.

Tom

I agree that product should be "opinions" or "guesses" - based on a sound analysis. An S2 is paid to express an opinion. His opinions. Not others.

... but I reject the notion that anyone should seek to gather opinion, when you are dealing in issues, that are ambiguous, time sensitive and prone to deception. I might get "opinion" fed into my collection, but if a source gives me an opinion (what he thinks) then my main effort will go to confirming the facts that created that opinion.

... otherwise when I right my report, I can't say I think FNU SNU is "lying" or "telling the truth". If I merely pass on opinion, then I would suggest I've failed.

If I am tasked to find out "What someone thinks about X" then I'd pretty much limit myself to the targets words and actions. If I go and ask his sister what "he thinks" then I've probably screwed up.

Yes, this is a generalisation, so there are exceptions.

Rex Brynen
07-31-2008, 03:52 PM
Considering the majority of intelligence agencies bureaucratic product is about supporting the policy maker, rather than informing him, I'm not sure how this will work.

It's like Newton observed about studying dung heaps. Interesting, but seldom useful. Likewise, intelligence should never seek to gather opinions, just facts.

Sorry to be a cynic!

As Entropy has suggested, I think very few facts are facts--or, more accurately, they mean very little in the absence of contextual interpretation.

Moreover, having worked for many years with intel shops producing bureaucratic product (of the political and strategic assessment type) for senior decision-makers, I can say that relatively little of it is about "supporting the policy maker, rather than informing him." On the contrary, while particular collections and analysis may be driven by current policies and interests, a good analysis manager is careful to support analysts in telling it how it is, rather than how leaderships would like it to be. That often means offering up conclusions that are the ones that policy-makers won't want to hear, but need to anyway.

Of course, it doesn't always work that way, as we saw with the Iraq/WMD case. There can be problems of political pressure, ideological group-think, poor management, and--of course--decision-makers ignoring the analysis they are provided with because they believe they know better (sometimes they do, often they don't).

The issue of providing raw facts and underlying data to senior decision-makers is often a sensitive one in the IC. Of course part of this is a combination of nerves and ego, and analysts not wanting to be second-guessed. Some of it concerns the need to protect sources and methods, and the fear that those outside the IC (and not simply politicians) might unknowingly compromise these. However, it is also the case that analysts spend long, long hours deeply immersed in topics, with access to a broad range of contextual information and confirmatory (or contradictory) intel material that senior decision-makers don't have. In short, they should be--IF the system is working--better positioned to interpret what a "fact" might mean.

To return to the original issue, namely DIA broadening its contacts and consultations outside its own shop: frankly, I think this is an excellent idea. We've seen CIA do much more of it too, and the ODNI has encouraged this and supported it through the outreach efforts of the NIC. You also see it in other initiatives too, ranging from the wikis in the OSC to the (international, cross-agency, unclassified) Global Futures Forum (https://www.globalfuturesforum.org).

There was an enormous amount of stovepiping in the US intelligence community, often justified on OPSEC grounds but also simply a function of having large organizations with multiple analysts and a corresponding sense that you didn't need to go outside the building to have the discussions that might be useful.

Yet those other discussions can be very useful indeed. Other government departments may have very different and informative perspectives. Returning and former diplomats do too. Academics may have been working on an issue area since before some analysts were born, known the primary actors personally, and have types of access that neither foreign ministries nor intelligence collectors have. Aid and NGO workers who have spent years on the ground, negotiating militia checkpoints and dealing with local powerbrokers, have insights that neither embassy reporting nor COMINT is likely to supply. And so on.

By contrast, most other NATO countries (with the partial exception of the British, perhaps) the analytical community simply isn't large enough to sustain this sort of stovepiping--if you want an intelligent discussion, you often have to go out of the building. In Canada, for example, it wouldn't be unusual to find colleagues from CIDA, DFAIT, DND, other government departments, and even suitably cleared academics :eek:, meeting to contribute feedback on draft analytical product before it is formally submitted to senior managers and policy-makers. It is a problem when you get to the higher levels of classification, since the potential pool of colleagues who can contribute to the product shrinks dramatically. However that can usually be addressed by discussing a lower-classification version of the assessment in more mixed company.

This is not a "small is better" argument, by the way--us non-superpowers would love to have rooms of people working on "country-of-interest X". But since we don't, it does require a sort of compensatory networking that has real advantages, which can be evident in working meetings with US IC colleagues.

J Wolfsberger
07-31-2008, 04:00 PM
On my side of the house, I have to distinguish between "data" and "information." Data is measurable and quantifiable - it comes from test and observation. Information comes from modeling and simulation. While it is quantifiable, it also has to be understood for what it is - an evaluation based on code developer and user assumptions about how the world operates. Both of them feed into analysis - which is a process that occurs in a human brain.

It seems to me, that the DNI is asking for outside opinions (analysis) as to what the data means.

Rex Brynen
07-31-2008, 04:03 PM
Wilf must type faster than me, since he partially addressed my concerns before I finished posting my reply!

However, on this, I continue to disagree.. I think:


... but I reject the notion that anyone should seek to gather opinion, when you are dealing in issues, that are ambiguous, time sensitive and prone to deception. I might get "opinion" fed into my collection, but if a source gives me an opinion (what he thinks) then my main effort will go to confirming the facts that created that opinion.

If I'm working on an intel assessment of issue X, I certainly want to know:

1) What smart and well informed people think about issue X--that is, the (informed) opinion that they hold, and why they think it. I especially want to hear from people with different vantage points than my own.

2) What local leaders think about issue X, and how that opinion might shape their decision-making and subsequent actions.

3) What the local population thinks about issue X, and how that might affect their voting behaviour, political mobilization, support for insurgent group Y, or whatever.

Of course, those opinions are only part of the very broad array of information that one is utilizing, and yes it is nice to be able to confirm their grounding in "facts." However, they may be just as useful even if they aren't.

Case in point: Wilf and I know that Israel's 2006 war in Lebanon was a response to Hizbullah's cross-border abduction. However, in an assessment of Lebanese politics or Hizballah's possible future moves, the (incorrect) opinion of many Lebanese Shi'ites that the war was a preplanned act of Zionist aggression is hardly irrelevant to my analysis.

William F. Owen
07-31-2008, 04:07 PM
As Entropy has suggested, I think very few facts are facts--or, more accurately, they mean very little in the absence of contextual interpretation.

Moreover, having worked for many years with intel shops producing bureaucratic product (of the political and strategic assessment type) for senior decision-makers, I can say that relatively little of it is about "supporting the policy maker, rather than informing him."

Being all smart folk, I don't think we need to get too bothered about what constitutes a "fact." The word "evidence" fits equally well. Something supported by a body of evidence should reasonably qualify as a fact.

As to your other point, I'll defer to your experience, though I suggest the historical record has more than ample evidence for debate.

William F. Owen
07-31-2008, 04:16 PM
If I'm working on an intel assessment of issue X, I certainly want to know:

1) What smart and well informed people think about issue X--that is, the (informed) opinion that they hold, and why they think it. I especially want to hear from people with different vantage points than my own.

2) What local leaders think about issue X, and how that opinion might shape their decision-making and subsequent actions.

3) What the local population thinks about issue X, and how that might affect their voting behaviour, political mobilization, support for insurgent group Y, or whatever.


1. I suggest the area of interest is what information "informed" the opinion.

2. Their opinions could create any number of subsequent actions. I would limit myself to what they say and do. "If you come we will fight you," is a threat, not an opinion.

3. What a population thinks is indicated by words and actions. Assessing those words and actions are the critical part of the job. People who think stuff but do nothing generally don't make an impact.

...and it occurs to me that I may be defending a purely semantic position, but may be not. :)

Rex Brynen
07-31-2008, 04:32 PM
...and it occurs to me that I may be defending a purely semantic position, but may be not. :)

Well, I wouldn't want to be accused of being anti-semantic!

Tom Odom
07-31-2008, 05:33 PM
1. I suggest the area of interest is what information "informed" the opinion.

2. Their opinions could create any number of subsequent actions. I would limit myself to what they say and do. "If you come we will fight you," is a threat, not an opinion.

3. What a population thinks is indicated by words and actions. Assessing those words and actions are the critical part of the job. People who think stuff but do nothing generally don't make an impact.

...and it occurs to me that I may be defending a purely semantic position, but may be not. :)

Semantics stand aside :eek:

Words, facts, and actions are part of it. But at the critical moment the intelligence analyst has to take it all in, digest it, and in the end offer what it means. That means making an assessment of whether words are indeed threats or bluster. Sometimes it means assessing whether what is unstated is a threat or a sincere promise. that applies at the tactical level and at the strategic. It is an art and some have it. Others do not. When the commander or the policy maker trusts you he/she will ask "what do you think?" or words to that effect. If you do the classic, "the enemy has the capability" to do whatever, you have lost the commander's trust. He wants your best and that is an informed opinion.

Tom

John Gentry
07-31-2008, 05:34 PM
Reminders to intelligence analysts to widely seek good sources of information and insight are old, regular, and well known to good analysts. For example, SecDef Gates repeatedly chastised CIA analysts for insularity when he was CIA's Deputy Director for lntelligence in the 1980s and even published a critique of CIA analysis in Foreign Affairs. For a time he required his analysts to take a formal course of some sort outside the agency each year precisely to help do what the DNI wants.

There is always room for improvement and people sometimes need to be reminded of the obvious, but there is little remarkable in this missive.

Entropy
07-31-2008, 06:02 PM
You also see it in other initiatives too, ranging from the wikis in the OSC to the (international, cross-agency, unclassified) Global Futures Forum (https://www.globalfuturesforum.org).


Who does one need to blackmail/bribe for an invite? :)

Rex Brynen
07-31-2008, 06:26 PM
Who does one need to blackmail/bribe for an invite? :)

PM sent.

selil
07-31-2008, 08:03 PM
Who does one need to blackmail/bribe for an invite? :)

Yeah, what he said.