by Rafe Sagarin
Babies and truly adaptable organizations use the same basic methods to deal with this information deficit and make sense of an ever-changing, ever-surprising world. This is also essentially how good natural historians like Darwin make sense out of an incredibly complex and variable biological world. Darwin, for his part, didn’t just dream up his theory of natural selection, nor did he have a revelation while staring at finches on the Galapagos. Rather, he spent painstaking years during and after his long voyage on the Beagle piecing together the variation of life. If you doubt his dedication, keep in mind just two of his lesser known works, the extensive and definitive guides to the world’s barnacles: Living Cirripedia, a Monograph on the Sub-class Cirripedia, with Figures of All the Species: The Lepadidæ; or, Pedunculated Cirripedes, Volume 1; and Living Cirripedia, the Balanidæ (or Sessile Cirripedes); the Verrucidæ, Volume 2. Not exactly world-changing books like On the Origin of Species or even delightful travelogues like The Voyage of the Beagle. But this painstaking observational study of barnacles was part of Darwin’s quest to figure out how each living thing played into the larger picture of the highly variable and ever-changing world he observed during his wandering years. As he reflected in his autobiography, “My mind seems to have become a kind of machine for grinding general laws out of large collections of facts.”26
Or, as the authors of the “Learning from Samples of One or Fewer” paper wrote, “Great organizational histories, like great novels, are written, not by first constructing interpretations of events and then filling in the details, but by first identifying the details and allowing the interpretations to emerge from them. As a result, openness to a variety of (possibly irrelevant) dimensions of experience and preference is often more valuable than a clear prior model and unambiguous objectives.”27
The commonality of these different perspectives on an inductive approach is that learning to see the world with an adaptable mind, as babies, naturalists, and adaptable organizations are able to do, requires both a dedication to observation and a rich imagination. True, some early naturalists in the Middle Ages let their imagination get away from them at times—mixing incredibly accurate portrayals of real species with descriptions of unicorns, mermaids, and dragons.28 Likewise, toddlers routinely talk to imaginary friends, weave elaborate tales that mix imaginary and real entities, and derive fanciful explanations for what they observe.29
But so do we.
The most memorable conclusion of the 9/11 Commission Report was that the security failures leading up to 9/11 represented a “failure of imagination,”30 and this was certainly an apt description of the security situation up until 9/11. Our problem now at times seems like the opposite. We imagine too much. We see monsters under every bed, and it makes us do things, sometimes at great expense, that are fairly ridiculous. We now have no problem imagining almost anything to be a security threat, from any package over 16 ounces (which can no longer be put in a mailbox) to my daughters’ yogurts, cruelly confiscated by TSA agents when we board a plane. For good reason, soldiers in Iraq and Afghanistan imagine almost any piece of roadside debris or unusual disturbance to be a threat, but many stricken with varying degrees of post-traumatic stress disorder continue to imagine these threats when they return.
How do we reconcile the active imagination needed to envision multiple states of an unknown future world with the need to spend resources on things that will truly make us safer? Babies do it by growing up. That is, babies’ imaginations are better, more finely tuned, and more adaptive than the “anything goes” imagination that most adults practiced after 9/11. Their imagination is contextualized and scientific. Like good naturalists, babies mix observation and imagination to create hypotheses—educated guesses—about how the world works and how it could work. They then test multiple alternative hypotheses: they see how a toy car moves when pushed and how it breaks when it falls down the steps, how other babies react when they’re touched gently or screamed at, and how adults react when they attempt to imitate different sounds—and they reject the hypotheses that don’t work for them and further test those that do.
This is exactly what I do when I begin to study a new tide pool or revisit a natural location that was studied long ago by one of my predecessors. I observe with all my senses, and I compare those observations to what I think I know from reading the past scientist’s notes or what I’ve experienced with similar species in another location. I use my imagination to tell some hypothetical stories about how this little corner of the world has changed. For example, I might find that compared to earlier studies, very few large snails still live in this tide pool; and I begin to compile some stories—a new predator came to town and ate all the large snails, warming ocean temperatures were disadvantageous to the large animals, humans began using the snails for bait, a mermaid loved the snails and brought them all to her magical undersea garden. Then I make more observations to figure out which of those hypothetical stories could be true, and which are most likely to be true. Like a baby, I am never completely correct. But I can’t let that stop me. I will never have enough data or observations to be 100 percent sure that I am telling the right story, but, like the baby, I need to find the plausible story so I can move on and grow. My needs to distill this imagination and observation into a plausible story are prosaic—I might need to get the work published so I can keep my university job—or potentially a little more important—I need to identify the most likely cause of change to help a community group that wants to enact a conservation plan. But for the baby and the security organization, this ability is a matter of life and death.
How can our security organizations and strategies grow up beyond the purely imaginative stage while maintaining an ability to understand and respond to unconventional threats? Hypothetical thinking is only one of three ways that “Learning from Samples of One or Fewer” suggests that organizations can cope with uncertainty. It also suggests that making multiple experiences out of a single experience by getting many people to look at it and deconstruct it is a way to deepen that experience and learn from it. This isn’t the typical way organizations work. Usually a single team is ordered from central management to analyze problems and develop solutions. This kind of centrally controlled organization was discovered to be at the root of most of the cases that Dominic Johnson and Elizabeth Madin examined in which organizations failed to learn until catastrophe struck.
By contrast, loosely related groups of individual humans, when sharing learning through networks, can adapt almost instantaneously. The best example of this is that until 9/11, the normal response to a plane hijacking was to put up no resistance: hijackers made demands that were eventually negotiable, and lethal threats were unlikely to be carried out. But within minutes of the mutation that saw terrorists starting to use passenger planes as weapons of mass destruction, humans used networked technology to share information about the change in hijackers’ tactics, and passengers on one hijacked plane immediately adapted a more active defense, risking their own security to protect a larger (and largely unrelated) group of humans. Subsequent airborne attack attempts by Richard Reid and Umar Abdulmutallab were similarly stopped by passengers.
To be adaptable, organizations need to learn like babies and naturalists and networked groups of well-informed adults, not like businesses and engineers, but they’re generally not set up to do so. Instead, we need to alter the genetic architecture of a security organization to be rearranged into a kind of chimera—a beast as adaptable and flexible as natural learning systems, but not so remotely different from our existing organizations that we have to destroy everything and rebuild it all. Fortunately, nature offers abundant clues on what this organization should look like. In fact, we can’t look at any part of nature without seeing templates for truly adaptable organizations. The next chapter shows how decentralized organizations are extremely common in nature—in fact, it is from this kind of organization that almost all of the remarkable capacity to adapt found throughout the natural world emerges.
/>
chapter four
ORGANIZED TO CHANGE
UNDERSTANDING HOW NATURAL security systems work requires keen observation of nature. But observing nature isn’t just about peering for hours into a forest grove to catch a glimpse of a rare bird to add to your life list, or collecting butterflies to pin into a shadow box. Useful nature observation looks at how organisms interact with one another and with the world they live in. It looks at how these interactions change through time—over a matter of minutes, through the seasons, and across eons of geologic time. And good observation engages all of the senses.
One of the best observers of nature I have ever met is paleontologist Geerat Vermeij. He can identify nearly any fossilized or living marine organism, and by carefully observing the shell or carapaces of any individual, he can tell you how it lived, how it died, and how it changed relative to its ancestors. Even the most seasoned naturalists are amazed by Geerat’s skills in the field. His abilities as a naturalist are even more impressive when you take into account that he has been blind since the age of three.
But just as his lack of eyesight hasn’t hampered his ability to observe the natural world, it hasn’t restricted his vision in connecting the pieces into a larger whole. In this he follows the tradition of Darwin, Ricketts, and Rachel Carson, who built grand holistic syntheses based on meticulous observation of the variation in nature. Vermeij uses his incredible tactile sense to build, in his mind, an immense catalog of biological variations from which emerges a holistic picture of patterns and trends in the evolution of life. Twenty years ago, Vermeij published one of the first treatments of arms races in the natural world in his book Evolution and Escalation, in which he analyzed millions of years of battles between snails episodically growing in bursts of stronger and more armored shells as their various crab predators developed ever more powerful claws.1
As Vermeij continued to observe the diversity of life recorded in fossils that span hundreds of millions of years, he gradually came to understand a prominent pattern of nature, one that has critical importance to the study of natural security systems. He found that for the most part, centrally controlled organizations do not thrive in nature. Rather, the job of sensing the environment is farmed out to multiple agents that have a great deal of power to respond on behalf of the larger organism. As with all of the natural security systems I discuss in this book, this decentralized pattern of organization occurs at every level of biological structure. Cells, individuals, and even ecosystems survive and adapt by letting localized agents detect environmental change and respond to it with little central control.
Starting small, the vertebrate immune system is a wonderful example of an adaptable organization. Although it serves a centralized purpose (to keep the body free of malicious invaders) it works by sending out multiple independent cells that identify invading organisms. These cells don’t have a predetermined “watch list” of intercellular terrorists handed down from the brain for them to check against. Rather, they map out the characters of anything entering the body and then call in helper agents to bind to the invaders and neutralizing agents to destroy them, if necessary. All of this happens continuously and without a lot of preplanning. Of course, past experience with certain invaders improves performance of this system, which is why vaccination using small doses of pathogens is so effective.
Like the immune system, our brains have to deal with a huge array of uncertain information. In the previous chapter I mentioned that one way the brain anticipates what will come of all this uncertainty is by organizing all past information into relational chunks. The brain does this with a highly decentralized storage system that places different aspects of memories (the smells, the colors, the emotional feeling associated with it) in different places. The ensuing network of related neurons is robust—the memory can’t be removed just by an injury to a particular part of the brain—but also flexible and adaptable.2 It’s been argued that at a higher level, human intelligence generally mirrors the intelligence system of a single brain—that is, our collective intelligence as a species is decentralized, but still a robust collection of knowledge, experience, and learning linked by social networks.3
Back at the level of whole individual organisms, one of the great advancement of birds over their dinosaur ancestors was decoupling bones connecting the legs and tail, which were essentially fused as a single locomotive unit in dinosaurs. The specialization allowed independent movement of the tail, which provided key stabilization to wings for a new life of living in the skies.4 Other organisms develop adaptive organizations by specializing individual clones. That is, species like corals, tunicates, and bryozoans start as genetically identical clones, dividing asexually from a single parent cell. As the colony smears out through division across a hard surface such as a submerged boulder or the skeletons of long dead corals, each unit can take on an important function—some becoming specialized for feeding, some for reproduction, and others along the edge of the colony armed with stinging cells to repel other colonies and gain precious space on the hard surface—and all contributing to the overall survival of the colony.
Decentralized organization doesn’t just benefit individuals. Populations of individuals also utilize this system of organization. For example, schools of fish maintain safety for the individual fish within them by independently responding to changing ocean conditions. No supreme leader fish tells all the others what to do—rather, each fish doing its own part to sense change and change itself leads to the appearance of a much larger whole organism.
Simon Levin of Princeton, who has built a career brilliantly uncovering linkages between the properties of mathematical functions, economic game theory, and ecosystem and societal organization, finds that this same decentralized organization is what gives ecosystems resiliency in the face of environmental change.5 Levin isn’t saying that ecosystems actually evolve as a unit, but rather that the individual organisms within the ecosystem—each doing their own thing to survive—impart survivability on the ecosystem as a whole. This turns our whole notion of a hierarchy on its head. We tend to think of hierarchies as controlled from the top, exerting power down through the system. Ed Ricketts saw this reversed hierarchy as a basic property of biological organization: “Each higher order, instead of ruling the ranks of individuals below, is actually ruled by them. Each rank is completely at the mercy of its subjects, dependent on their abundance or accessibility.”6
Levin sees the same patterns in social and political patterns that emerge when people in adjacent areas tend to have similar belief systems and vote in similar patterns. Levin calls the systems that emerge—whether ecosystems or voting blocs—“complex adaptive systems,” and their most remarkable characteristic is that they emerge from very simple and largely independent actions of individuals within the system.
Decentralized and distributed organizational systems are adaptable for three main reasons. First, multiple sensors all looking or experiencing the environment from their own perspective provide more opportunities to identify unusual changes and unexploited opportunities. When we let a single entity (say, TSA at airports) take complete charge of security, the number of observers goes down, along with the probability of identifying a threat to security. Second, multiple agents committed to the security mission in their own local area create opportunities to specialize tasks, so that energy isn’t wasted in having every part of the organism doing the same things; rather, those doing the most important things (e.g., providing defense when hostile enemies are around or reproducing when populations are low) get the resources to replicate their activities. We have often ignored this lesson in distributing resources for homeland security. Recently, governors throughout the United States were appalled to find that in order to receive federal funding from the Department of Homeland Security, they had to commit 25 percent of their budgets to defense against improvised explosive devices7—a huge threat in foreign conflicts but extremely low in importance relative to other threats facing the states. Third, d
istributed sensors respond to the most immediate environmental conditions in time and space—they see the environment for what it “is” rather than what it “should” be according to some preconceived notion. This way, the octopus that gets transferred to a lab tank isn’t paralyzed by the new environment. It simply uses its eight tentacles and thousands of suckers (which can smell, by the way) to feel out its new surroundings, search for food, and find an escape route.
To appreciate how vital these organizations are to adaptable security, consider what happens when we fail to use them.
THE FIRST POST-9/11 TEST OF HOMELAND SECURITY
August 29, 2005, started like any other day in post-9/11 America. On that Monday, roughly 1.8 million airline passengers were busy taking off 3.6 million shoes and putting them through X-ray machines, going through the security ritual necessary before boarding their flights. Nearly four years after 9/11 and then Richard Reid’s attempt to blow up an airplane with a shoe bomb, passengers were getting used to long lines and invasive searches at airports. The multi-billion-dollar Transportation Security Administration (TSA), part of our newly created Department of Homeland Security (DHS), could proudly say that there was virtually no danger from a shoe bomber on August 29, 2005. But by the end of the day, something else had happened: an entire U.S. city was underwater. Families were separated, and senior citizens were drowning as floodwaters rose and no one was there to save them. Chaos and disease were festering in the shattered shell of the New Orleans Superdome, once considered to be an engineering wonder of the modern world, now a makeshift shelter. From an ecosystem perspective, the destruction caused by Hurricane Katrina was inevitable. Just as the hardest hit areas of the Asian tsunami were those where protective mangrove forests had been removed for coastal development and aquaculture ponds, New Orleans, a city already built below sea level behind levee walls, had stripped most of its protective wetlands for its own economic development. Like ignoring animals that are running uphill before a tidal wave, actively destroying natural protective systems is an obvious failure of our security responses to respect nature’s experience.