Connecting the Dots: President Obama
Connecting the dots is Mission Impossible
The idea of âconnecting the dotsâ in intelligence is a clichĂ© that should have been retired years ago. However, given that the strategic intelligence analysis community still has no working theory of its capabilities and limitations, we will continue to see it used.
With the attempted bombing of North West Flight 253 on Christmas Day by Umar Farouk Abdulmutallab, the intelligence community is again being told that it has failed to âconnect the dots.â The issue of connecting the dots has been repeatedly raised in the past by a variety of intelligence commissions designed to study this perceived failure. None of the commissions has ever explained how the intelligence community is to successfully âconnect the dotsâ in the future â and neither will any investigation into this latest bombing attempt.
The first thing that should be noted is that connecting the dots, looking backwards, is an simple task for the social sciences, the hard science and for politicians and pundits. When looking at past events, it is an easy task to see the four or five dots that should have been connected to form the picture that was required. Looking forward, however, the problem is quite different. Here is an extract from a previous publication that takes a look at the true nature of the âconnect the dotsâ problem.
Consider an analytical problem where there are only four variables at play. To put the problem in a different light, consider that there are four variables and it is your job to âconnect the four dotsâ in such a way that your intelligence consumers can understand the outcome of this connection. With only four dots to connect, there are six different ways of linking up the dots. With these four dots and six different links, it is then possible to see that there are sixty four different possible patterns that could be formed. Your intelligence consumer can now see that there are sixty four different ways of connecting the dots or sixty four different possible outcomes to a problem that only has four variables.
An intelligence problem with only four variables is an unlikely event, however. Perhaps a more realistic intelligence analysis problem would have ten variables. How many different ways are there to connect the dots? The answer is quite startling. With only ten different dots to connect, there would be a total of forty five different dot patterns that could be created. What is stunning, however, is that the ten dots and forty five patterns can produce 3.47 trillion different outcomes.
Needless to say, even with the most advanced of computers, doing the calculations for 3.47 trillion different possibilities is a daunting task. Even this assumes that each variable can be reduced to a mathematical form so that the binary decision making process of a computer can handle the problem. At the end, all you would have is a report to your consumers that says there are a few trillion different possible outcomes.
With all of this understood, it should be clear that each real world intelligence problem in national security is likely to have hundreds or maybe even thousands of different variables at play. The number of possible outcomes in a real world situation would be astronomical.
It is not clear who the advisors were that suggested President Obama use this reason (excuse?) for the security breakdown, but it is probably fair to say that they have never studied the process of intelligence or any form of futures forecasting.
The more likely reasons for the security and intelligence failure are to be found in areas which are already known for years and not addressed. Among the ongoing and unsolved problems are:
a. Too many intelligence agencies,
b. Too many layers of bureaucracy compounded by overlaps and confusion,
c. Sharing at the classified and unclassified level is still a difficult process,
d. International sharing (in a meaningful way) is limited by classification systems and a lack of trust outside of the âthree eyesâ or âfive eyesâ community,
e. Analysts are frequently restricted from travel to foreign countries for âsecurityâ reasons, and
f. Analysts are frequently refused training for cost reasons, yet there is always enough money to buy some more computers or collection systems.
However, as noted above, the most significant problem for the intelligence community may be its own lack of a working theory, especially for strategic level intelligence analysis. With no working theory available, it is difficult to frame discussion on how reliable intelligence assessment (i.e. intelligence as a form of knowledge) can be. Although a limited theoretical basis does exist, the field of intelligence theory remains a work in progress.
With no overall theory to guide it, the role of strategic intelligence analysis remains imprecise and unable to determine its own strengths and weaknesses. Foreknowledge is constantly demanded by policy makers and they fiercely criticize the intelligence community when a perceived âfailureâ occurs. But could even a perfectly functioning intelligence community predict future discrete events and âconnect the dotsâ? Without the ability to answer these types of questions, the intelligence community will not be able to define its capabilities nor will it be able to establish its reliability under a given set of circumstances.
The opinions expressed in this blog are personal and do not reflect the views of Global Brief or the Glendon School of Public and International Affairs.
(The extract is from Seeing the Invisible, National Security Intelligence in an Uncertain Age, Thomas Quiggin, Singapore, RSIS, World Scientific, 2007)