Nidal Hasan and the problem of connecting the dots
Was Nidal Hasan’s atrocity at Fort Hood an act of domestic terrorism?
According to some news reports, he did yell the violent Islamic radical terrorist’s apparently obligatory “Allahu Akbar” before murdering and wounding almost 50 Americans.
Or did Nidal Hasan snap under what has been termed “emotional, ideological and religious pressures?”
Are there some other reason we don’t know yet?
Should the FBI be blamed for not investigating months ago the “… red flags, and … signs that should have raised alarms?” asked one of the many officials “speaking [to the media] on condition of anonymity.”
Or was it the military who “missed the danger signs?”
Or maybe the entire intelligence community?
Where have we heard this before?
The 9/11 Commission famously wrote:
The importance of integrated, all source analysis cannot be overstated. Without it, it is not possible to “connect the dots.”
How come no one connected the dots before Hasan’s brutal act? How many more dots are out there in the military and the civilian worlds that we should be connecting?
People are already asking those questions. Google “Nidal Hasan” and “connect the dots” and you will find hundreds of hits. Maybe thousands by the time you read this.
—————————–
I value the 9/11 Commission Report. I think it is the best government report I’ve ever read. It is not perfect. But it does set a standard for clarity and — in many sections — comprehensiveness.
I do fault the Report for starting – or at least fertilizing — homeland security’s “connect the dots” meme.
I think asking for dot connection confuses real life with what children do with pencils on the already-outlined images of windmills and teddy bears in their Toys R Us booklets. I think it creates false expectations about what the nation can expect from its intelligence community. And in the process lets the rest of us off the hook when it comes to paying attention.
The newly rising echo about dots reminded me of the September 2004 essay that first caused me to doubt the appropriateness of the metaphor for homeland security. The article, by Max Boisot, is titled “Connecting the Dots: from data processing to pattern processing.” It is from an electronic journal called Omnipedia that I don’t think is being published any more. But its owners, The International Futures Forum, have kindly made the 14 issues they did publish available (here). The “Connecting the Dots” article can be download in pdf format here. But since it is brief, I am taking the liberty of republishing a substantial part of it below.
—————————–
“…we believe that pleading for more dots is to mistake the nature of the problem posed by international terrorism, and that even recognizing the significance of the information is a task that exceeds the capacity of a single organization such as the CIA.
Consider Figure 1 [below] where the relationship between dots, links and patterns is highlighted [dots on the left, links in the middle, patterns on the right]. An arithmetic increase in the number of dots to play with - high quality or otherwise – leads to a geometric increase in the possible connections or links that one can establish between them and to an exponential increase in the number of patterns that can be generated from connected dots.
From this figure we can derive Table 1 below :
If 4 dots lead to 6 possible links – as indicated in Figure 1 – it also generates 64 possible patterns. Add 6 further dots and you get 10 dots generating 45 possible links and approximately 3.5 trillion possible patterns. Now add a further 2 dots and you are dealing with 66 possible links but with approximately 4,700 quadrillion possible patterns.
The implication of this table is that whatever we feel about the need for more high quality dots to connect – and we are not denying that the need is a real one – if we are not to drown in a sea of unprocessable data, we also need to find a way of identifying meaningful patterns among the much larger number of those that are meaningless.
Even if you don’t get the pattern, you get the picture.
This need for variety reduction brings us to an important if much misunderstood distinction between data, information, and knowledge. Simplifying somewhat, dots are environmental signals that register with an agent as data — things that attract our attention because they are anomalous. Links between dots constitute information that an agent extracts from the data, and the patterns that can be derived from such information when properly filtered can become actionable knowledge for the agent. The intelligence challenge is thus one of pattern processing rather than of simple data processing — i.e., processing dots. The two are really quite different. The latter might just be carried out by one organization, no matter how intelligent. The former needs to be distributed to those best placed to sift out significant from insignificant patterns: those closest to the action.
If pattern processing is so much more challenging than data processing, it cannot be done by the intelligence services alone. The only way to deal with the challenge is to apply the concept of neighbourhood watch to problems of terrorism and security and to do so on a global basis. It will pose in a novel way the question: who is my neighbour?
Hopefully it will be someone located in Peshawar who is as disgusted with the terrorist phenomenon as I am. But how, then, do I relocate my potential neighbour in Peshawar into a “love” loop and out of a “fear” loop? Clearly, we are dealing here with a hearts-and-minds issue, and unfortunately, not one that current institutional players have proved very attentive to or competent in dealing with.
