US must invest in science of dot-connecting

The terrorists attacks on Sept. 11, 2001, demonstrated the urgent need for the US government to use insights from "complexity science" to better understand our interconnected world.

"Complexity thinking" is an essential 21st-century skill. It involves recognizing changes in the broadest context; taking a big-picture approach to intelligence-gathering and analysis; and developing deeper understanding of complex human systems and the dynamics influencing regional politics and conflicts.

By all rational assessments, 9/11 was a systems failure. While individuals and agencies within our intelligence and law-enforcement organizations may have performed exceptionally, there was a significant failure as a whole to recognize and respond decisively to the growing threats in the years and months leading up to 9/11.

This is the conclusion that the 9/11 commission seems to have reached, even before its final report is issued. It is also the answer given by Condoleezza Rice, President Bush's national security adviser, and other key witnesses before the commission, which is holding its last hearings this week.

The remedy for a major systems failure is not just moving the boxes around, as was done in creating the Department of Homeland Security. More important is changing the thinking of those inside the system. Many in the intelligence community - including Marvin Cetron, author of "Terror 2000," a 1993 report that predicted the use of planes to strike the World Trade Center and the Pentagon - say we just don't have the right kind of analysts.

The CIA showed it is finally getting the message by sponsoring a recent international conference, New Frontiers in Intelligence Analysis. The gathering in Rome was based on the premise that analysts must do a better job connecting the dots, seeing emerging patterns in disparate types of information, and helping policymakers understand the dangers earlier.

The US needs analysts capable of making sense of a rapidly changing, complex global environment. We need more "big picture" thinkers who can quickly synthesize information from a variety of sources; spot subtle connections, emerging patterns, and discontinuities; and ask questions even before the "dots" are fully formed. We saw after 9/11 that by the time dots start appearing, it may be too late to connect them in any immediately useful way.

An example of the need for new approaches to intelligence gathering, analysis, and reporting hit me in 2002, when a US Customs training coordinator contacted me about the possibility of developing an educational program for senior analysts based on complex systems thinking. When asked what type of training analysts were receiving - eight months after the biggest intelligence failure in recent history - he replied, "We're teaching them how to write reports." While report-writing is essential, it's more critical that analysts develop nonlinear thinking and pattern-recognition skills in order to recognize and quickly put together information about the evolving ecosystem of global terrorism.

Testimony before the 9/11 commission has emphasized that intelligence reports prior to the attacks lacked specific information about "an imminent attack, a specific threat, or actionable intelligence." This emphasis on fully formed plots reflects a paradox to overcome - the desire for certainty and predictability in a world characterized by uncertainty and risk. Policymakers and analysts need a new way of thinking, and new models for analysis and reporting that reflect the complex, nonlinear, and dynamic realities of the world in which we live.

The US spends billions annually on technology, simulations, multiagency exercises, and yes, reports designed to help us think, and thus anticipate, the unthinkable. Most are based on insights from complex adaptive systems research. Such systems - traffic, for example, or the weather, the Web, society, or the globalizing economy - have many interacting factors, and each evolves in relation to the larger environment. In recent years, scientists have learned a great deal about how they work and how to influence them.

The science of complex systems has provided new concepts, tools, and a set of questions that can help make sense of messy situations, emerging events, and rapidly changing circumstances. Questions such as: What are the attractors in the system - the connections, relationships, and patterns of interaction creating the structure beneath the visible activity? What's happening in the larger context that might influence the system? Where are the sensitivities, and how are they evolving? What's perking on the horizon that could dramatically influence the future?

Complexity science is moving us from a linear, mechanistic view of the world to one based on nonlinear dynamics, evolutionary development, and systems thinking. Yet this kind of thinking rarely bubbles up to those who need it most: policymakers and the analysts on whom they depend.

"We need to be thinking this way, but we don't understand it," a senior National Security Council staff member told me in December. He asked to meet with me because one of his biggest concerns was that many in the intelligence community still use a linear framework to analyze evolving threats against our country.

The challenge for the 9/11 commission is to lay out a blueprint that will propel the US out of its old 20th-century ways of analyzing problems.

Until we're willing to develop new ways of thinking, we will continue to shuffle the boxes, place a premium on analysts' ability to prepare wellwritten reports, and leave ourselves open to attack from those who see that the world's most powerful country is still mired in old-world ways.

• T. Irene Sanders, executive director of the Washington Center for Complexity and Public Policy, is author of 'Strategic Thinking and the New Science: Planning in the Midst of Chaos, Complexity and Change.'