Monday, May 23, 2011

"We'll Return To Our Regularly Scheduled Programming In Just a Minute..." (Let's Kill The Intelligence Cycle).

My intent today was to jump right into my series on the intelligence cycle and why we should get rid of it (put a wooden stake through its heart were the exact words I think I used...).

However, over the weekend, I received a torrent of emails and the post received a number of comments and it occurred to me that, before I got started in earnest, it might be useful to do a little wholly unscientific sentiment analysis on this issue.

Using the Swayable tool (which many of you have already tested here and here), I intend to first test the underlying assumption behind this study and second to ask two related but independent questions about your perceptions of the intelligence cycle and its place in intelligence theory.

The Assumption Check

The first question is:  "Is the traditional intelligence cycle a perfect representation of the current intelligence process?  By "perfect" I mean perfect -- does the intelligence cycle accurately model the intelligence process as it is currently done?  Trivial issues count here (we will deal with them later).


Something A Bit More Substantial


The second question addresses the degree to which the cycle is imperfect (assuming you thought it was imperfect in the first place): "Do the benefits derived from continuing to use the intelligence cycle as a depiction of the intelligence process outweigh the costs?" I would ask you to think carefully about both the costs and the benefits before answering.



Finally, I want to get at your beliefs: "Without reference to perfection (or imperfection), costs or benefits, do you believe that a better general description of the modern intelligence process is possible?" (Note: Extra credit for guessing why I chose pictures of Leibniz and Voltaire and double secret extra credit for knowing which is which...)



That's it. Please do not hesitate to pass this post and the series on to anyone who might be interested. In addition, please do not hesitate to join in the discussion by dropping me an email or posting a comment (comments are better as they can be seen by all but I understand if that is not possible).

Next: The Disconnect Between Theory And Practice

6 comments:

Danny said...

Personally, I found the cycle to be incredibly simplistic and unrealistic. We spend a great deal of time when we first get to Mercyhurst learning about the intel cycle, but it doesn't accurately depict what we do.

When I do the research for the assignments we get, I don't sit down, make a collection plan, and then go through each subsequent step. I do a quick search, gather basic information, develop a rough hypothesis or two, then go back for more information. Also, as far as open source information goes, I don’t think that I do much with the Processing and Exploitation stage (though I could accept the argument that by using open source data I am relying on information that someone else already processed).

I also think that our brains are wired to make a decision as soon as we see information. The intel cycle doesn’t really depict this. We try our best to be objective, but at the end of the day, I don't think that we can stop ourselves from judging data when we first see it. This is where structured methods become incredibly useful for me. I almost always use them to challenge the assumptions I have made.

In reality, gathering and analyzing intelligence is a much more fluid process than the intel cycle allows for. The intel cycle model is semi-useful as a very very basic introduction to intelligence, but once you move past that level it is very easy to see that it is a flawed system. I don’t know how to better depict it though, beyond throwing a ton of additional arrows in at every step of the cycle. I look forward to reading your suggestions.

Anonymous said...

This is a disingenuous question you ask. Anyone who answers other than "not perfect" either misunderstands the word "perfect" or cannot see through the fallacy of the logic you are using to ensure that you get the answer you want from your poll.

This is an embarrassment. It is like asking "when did you stop beating you wife?"

As for Danny's comment that we are wired to make a decision as soon as we see information is a reflection of human psychology and bias. Does Danny really want to create models of human biases? Is that what a model of the intelligence cycle ought to be? A series of human frailties? Until you get some sort of focus this sort of peripheral idea is bound to confuse.

Because this discussion is designed for reactions and "swarming" illogical ideas I can't see how you can achieve anything other than swirling muddy water. Rather than answering your loaded binary question with either A or B I am going to choose option C: not answer.

Why can't you ask questions that aren't loaded? Or maybe pose a thoughtful question?

I didn't post to begin with because at first it seemed clear that your BLOG posting was the product of inane thinking. You decided that you knew what the answer was and that you were hell-bent to make your opinion the right one after collecting information that supports your view.

I hope this isn't want students are being taught because it is demonstrating some pretty poor thinking.

I had to jump in now and challenge this whole exercise. If you want to fabricate in your mind an answer to a question and then prove it then by all means engage in your bias driven goal to make you satisfied with your own thinking.

Don't try to drag the rest of us down with illogical thinking.

Frank Schwimmer

Anonymous said...

Lets be honest. Isn't the intelligence cycle simply the scientific method (slightly revised). It seems to me to be. This is the core problem (there are many problems with it) - it more aligns with science than it does with intelligence. This only lends itself to increased confusion over exactly what it is intelligence professionals do. A more accurate model would depict how what we do is different from what scientists, researchers, policy analysts (any other analyst, really), soothsayers, etc.... do. Too bad we can't put our finger on that...

Anonymous said...

Some opinions...
The Intelligence Cycle has never gone through an Operations Research-based Systems Design process. The Intelligence Cycle has evolved over time through the "Expert Systems" system design approach. Attempts at formalizing the process have failed because the qualified advanced systems design engineers have never been developed internally or brought in from the outside of the IC. The Systems Engineering process as it is practiced today in the IC bears no resemblance to the systems engineering processes that build the advanced airborne platforms and space programs.


A consequence of such a practice was the equivalent of "pilot's" system designing an advanced platform. Pilot's are able to "talk the walk but are unable to walk the walk". The popular OODA Loop in the aviation community is another example of an ill-conceived ill-logicical process causing a slow down in developing of cost-effective advanced systems.

For the problem at hand, the absence of advanced systems designers has resulted in the non-development of Information Age science and engineering. This has resulted in an IC community unable to recognize major scientific and engineering advances in Information Science and the corresponding major changes in Scientific Methodology and Philosophy.

There is a proven theory of closed-loop, self-correcting processes that has been applied successfully to non-intelligence systems.

Thank you.
Very respectfully,
Achteck

Anonymous said...

Frank,

While I tend to agree that some of the questions are, in fact, leading (and perhaps skewed and unnecessary), I find myself hard-pressed to see it as anything more than setting a baseline, or perhaps serving as an opener -- if the system isn't "perfect" given the considerable number of facets involved, where can we look to improve it? In fact, I believe the cost vs. benefit question is useful; as you've no doubt noticed, responses are more or less split evenly on that one. Of course I can't speak for Kris, the author, as to his own motivations, but I can at least throw in my own thoughts. Though I'd be willing to wager his opinion is that this particular blog post is a "little wholly unscientific sentiment analysis on this issue."

Even so, it seems patently clear that at some point in time he developed the hypothesis that the cycle may be flawed and set out to prove it, though from reading the first part of the series I see no evidence that he failed to take an... evidential route. Or, at least, I'd argue that we surely cannot judge that he didn't until a later point in the "series." Did he throw in some sensationalism? Yes, definitely, but that does not by necessity equate to fallacious logic or flawed research.

In reference to Danny's comment, you're right in that it needs to be fleshed out further, but confronting that task is a far throw from rejoicing about "a series of human frailties." If human bias and rational imperfection are things we confront on a daily basis, and general adherence to the set and structured conceptual model or method is in practice difficult, we may do well in taking an integrative approach to those "frailties." Implicitly, you seem to be suggesting the same, that it's an area worthy of a serious degree of "focus," though I do not know how consideration of biases plays into Kris' research.

To be clear, I think you're right in providing a sort of challenge to matters of analytic process and theory, but I think we're presently quite far from being able to assert that it's simply a "bias-driven goal to make [him] satisfied with [his] own thinking."

Regards,
Nick

Anonymous said...

Kris,

Voltaire's most famous quote for me is "The best is the enemy of the good." ("Le mieux est l'ennemi du bien.")

Leibniz is reported to have said: "“Although the world is not perfect, it is yet the best that is possible.”

And Sting was wrote: "The search for perfection is all very well, but to look for heaven is to live here in hell."

You are recasting the famous oppositional viewpoints that led Voltaire to write "Candide", with Leibniz as Pangloss, with a very closed, narrow view of an ideal world, and that we should accept the imperfections because a higher authority has put them there for us to contend with and thereby improve ourselves as a society.

Leibniz held that the world as created (by God) was perfect, and we shouldn't fiddle with it. Axiomatically, the intel process as handed down to us from on high and ages past is holy writ. We must not tamper with it.

Voltaire says, "Why not?" Given that there are imperfections, that the world (and art) has changed and evolved, why would we not try to make a realistic improvement to the existing process to reflect the realities and vagaries of intelligence today.

Yours,

Nick Psaki