Introduction to Chaos Theory
People have been managing formal organizations for several thousand years, beginning with armies, religious institutions, and secular government. Yet despite the concerted efforts of hundreds of academics, practitioners, and forward-thinking scientists, futurists, and philosophers, there appear to be no constant, simple, applicable rules for managers to follow that are universal. The science of management has come a long way in just 20 years, but compared to what we know about the physical world, the science of managing people is still in its infancy.
And it may always be so. People are so individualistic, interactions between people are so unpredictable, and so many variables appear to affect what occurs in an organization that outcomes do not appear to be predictable, given the initial starting conditions and information about events that change those conditions. That is why many people consider managing organizations more art than science and why one of the prevailing models of decision-making in organizations is Charles Lindblom’s concept of “The ‘Science’ of Muddling Through,” the title of a classic paper he wrote in 1959 that suggested that there are limits to rational decision-making by managers faced with too many competing values, too many changing variables, and only so much gray matter in the cerebral cortex available to process it.
Science can tell us a lot about management, however. The work of management thinkers, such as Herbert Simon, Max Weber, Peter Drucker, and Robert Golembiewski, have provided managers with some general principles about management.
For centuries, ideas about managing people have paralleled ideas about the nature of the physical world. The science of chaos theory is providing managers with a new perspective on how to manage their organizations and to consider the effect of even the smallest changes they make.
Every experienced manager has at some time experienced a small glitch that somehow gets magnified and causes a major catastrophe. An employee gets sick, a major project goes uncompleted as a result, and dire consequences befall the organization. Or a freak storm requires the cancellation of a board meeting and, as a result, one particularly argumentative, hostile board member shows up at the rescheduled meeting and changes the course of the organization forever. There is an aphorism that appears in a somewhat edited form in Benjamin Franklin’s 1735 edition of Poor Richard’s Almanac that expresses this thought:
“For want of a nail, the shoe was lost;
For want of a shoe, the horse was lost;
For want of a horse, the rider was lost;
For want of a rider, the battle was lost;
For want of a battle, the kingdom was lost!”
Or, conversely, a chance meeting on the street results in a collaboration that transforms the organization in a positive way. Or, perhaps, you were planning to spend the day having a picnic, but it started to pour, and you rushed into the local mall, found this book in your local bookstore and decided to buy it and have your mind opened up to new ideas, implementing TQM and BPR simultaneously in your organization.
Chaos theory is concerned with the performance of dynamic, non-linear systems. The word “chaos” in this context is a misnomer. The implication of the theory is that seemingly random behavior in nature shows structure and pattern.
A linear system is one that reacts to an outside stimulus in a proportional way. Non-linear systems are those that react to even small outside stimuli in a large, disproportionate way, and may even react to a small change in a more pronounced way than to a large change. Systems that exhibit chaotic behavior show output that appears random and disorderly, but yet is random and disorderly within clearly denoted boundaries. When a manager comes home from the office and writes down what occurred during the day, typically the day was filled with putting out fires, unanticipated distractions, and wildly diverging demands for the manager’s time. Yet the manager is able to return to the office day after day and experience much of the same thing, even if the “thing” is constantly changing.
Roots of Chaos Theory
Chaos theory has its roots in modern physics, chemistry, and biology. Quantum mechanics and relativity theory have required much of classical physics to be revised, and to be applicable only under certain conditions. For many years, the flight of electrons around an atomic nucleus was thought to resemble the apparent clockwork stability of planets orbiting the solar system. We know now that this is not the case; randomness and uncertainty seem to permeate both the microcosm and the macrocosm. In chemistry, complex chemical compounds demonstrate nonlinearity when subjected to internal and external events, but they often transform as a result of breaking apart by forming new and even more complex compounds (as, for example, noted in the discoveries of Hya Prigogine, the 1977 Nobel Laureate in chemistry).
And in biology, the science of evolution, once thought to be the result of slow, steady, incremental changes in genetic material, has had to be modified to account for apparently quantum leaps in the complexity of new species within a relatively short time frame.
Chaos theory is a new science that accounts for the fact that very small changes in initial conditions of a system can lead to very divergent outcomes. This theory is completely revolutionizing the physical sciences, and recently has gained credence as having important ramifications to management science as well. Organizations are viewed now as non-linear systems. Some days are stable and calm, and some days are filled with random, unpredictable “turbulence” that results in events that completely change the organization in a way that, unlike systems in an orderly, mechanistic, deterministic model, are not reversible.
The theory’s modern origins date back to 1960 when a meteorologist, Edward Lorentz, was using a computer to make weather predictions. He set up a computer model based on 12 equations. Wanting to save some time and paper while running the program a second time, he rounded off a single data point from .506127 to .506. He expected to get results approximately equal to the first outcome, but instead, found a completely unexpected, counterintuitive, and massive divergence from the first outcome.
Adherents of chaos theory have labeled this phenomenon the “butterfly effect.” Half in jest and half seriously, it refers to a hypothetical situation in which a butterfly in Tokyo happens to flap its wings creating barely measurable air turbulence, and as a direct result from a cascade of events, a storm occurs in New York that otherwise would not have. A more precise mathematical term for the butterfly effect is for a system to demonstrate a “sensitive dependence to initial conditions.”
There are many common phenomena in nature that seem to verify the tenets of chaos theory. Every snowflake is unique, a result of the fact that minute changes in humidity, air pressure, and wind have large effects on the formation of the crystal that creates each flake. Each person’s fingerprint is unique. The structure of clouds, tree branches, and coastlines show the phenomenon of scaling, in which the micro structure has the same general appearance as the macro structure. The stock market is considered to display chaotic behavior, in that it is possible to predict trends but not data points day to day.
Chaos theory looks at what appear to be random data and randomly patterned physical structures, and surprisingly finds patterns that otherwise might not be readily discernible. For example, making a graph of Lorenz’s three equation model for a simple system that also displays sensitivity to initial conditions, creates the equations that precisely describe a water wheel. The equations, when graphed, never repeat, are not periodic, and appear to be random, except for the fact that they form a double spiral curve that never repeats, and looks like a butterfly on the page. The graph is called a Lorenz attractor, and management theorists who apply chaos theory, such as L. Douglas Kiel, encourage managers to find attractors for their work processes and draw conclusions from looking at these graphs.
Looking at the output of an organization, some days might show excellent productivity and other days minimal productivity. Each day might appear to be unique, and random. Yet chaos theory provides tools to find the structure of even apparently random output of an organization, and use that structure to extract useful information. Just as Lorenz graphed attractors to see the pattern of random data, non-profit organization managers can do the same.
An attractor is a graph of output for a variable over time. It is called an attractor because while any data point seems random, looking at the graph suggests that these points are “attracted” to a region on the graph. The attractor shows how fast or slow the data points are changing compared to the previous data points; that is, the incremental difference in their value. Rather than being plotted as a line graph, the data points are graphed on a Cartesian graph. This Cartesian graph consists of four quadrants. If the value of each data point is continuing to increase, for example, most of the points on the graph will be in the upper right quadrant. A chaotic attractor in which the data oscillates randomly throughout its attractor boundary is graphed as the butterfly-shaped figure first shown by Lorenz. You can see pictures of these attractors by visiting some of the Web sites listed in the Internet Resources section of this book.
Chaos theory upholds that even in inorganic systems, it is impossible to predict outcomes with any certainty because of the inability to make small enough measurements to counter the wildly changing outcomes that result from these small differences in initial conditions. For example, even if computer models improved enough and valid data could be collected from points in the atmosphere at one-foot intervals in three dimensions over the entire globe, long-term weather forecasting would still be impossible according to chaos theory.
Every student in statistics learns about the bell curve, and about the law of large numbers, which essentially states that everything eventually evens out given enough time. (This law actually states: “If the number of times an experiment is repeated is increased, the ratio of the number of successful outcomes to the number of trials will tend to approach the theoretical probability of the outcome for an individual event.”) For example, flip a fair coin four times and it would not be unusual for heads to occur more or less than half. But flip a coin a billion times and a graph showing the amount of outcomes will always approach 50% heads and 50% tails.
Chaos theory directly contradicts this frame of reference, suggesting that outcomes do not necessarily even out over time, and that outliers on the bell curve (that is, events or data points that vary greatly from the average) often disrupt the equilibrium of systems enough to completely change them into new systems.
Those who watched the popular film Jurassic Park heard actor Jeff Goldblum explain chaos theory for the first time in the popular media. Our very existence as a species may well be the result of chaos theory at work. Sixty-five million years ago, an asteroid may have hit the earth and completely wiped out the dominant species on the planet, the dinosaur. Had this highly improbable, chance event not occurred, humans would not have evolved into the apparently dominant species.
Management and Chaos Theory
What does this say about management? One oft-quoted piece of advice for those who run organizations is “Don’t sweat the small stuff.” Chaos theory suggests that you should sweat the small stuff.
Management science and its physical science counterparts were on a parallel course for many years. Taylorism in management was a mirror image of Newtonian physics—a mechanistic, reductionist viewpoint in which the component parts of a system could be broken down and analyzed, and adjustments could be made. And the outcome would reflect a linear relationship to the input. For example, increasing the 10-worker production line by one worker would, approximately, increase production by 10%. But chaos theory predicts that even in complex systems that are inorganic (such as weather), it is impossible to predict outcomes based on a small change in input. And in organic systems, such as organizations, there is even greater likelihood that small interactions result in disproportionate effects.
“ Efforts to make the workplace more productive, such as diversity training or organizational development techniques to promote communication, can be viewed as efforts to change that employees’ use in the workplace,” writes Kiel in the 1994 book Managing Chaos and Complexity in Government. “These changes, however, can generate surprises for management. The nature of non-linear interactions suggests that even such positive workplace interventions may bring on a new set of unintended outcomes.”
The fact is that the reductionist view of reality may no longer apply with validity. Everything is connected to everything else. Making a small change here affects the entire system, and changing the system affects connected systems.
Organizations are, of course, organic. The variables involved in even a small group of people are infinite. A good illustration of this point is a baseball game. Virtually everything about this game is controlled. The field is always the same size. The pitcher’s mound is exactly the same distance away. The ball is the same size and weighs the same. The lineups of the players may be the same. The rules don’t change. So why is it that every game you see is totally different (that is what makes baseball so interesting, of course) even if the conditions are purposely duplicated by having the same lineups, same pitcher, same field, same rules, and so on?
The reason is that chaos theory is at work. Everything may look the same. But the pitcher may have had toast with his coffee in the morning instead of eggs. A fan may have yelled something distracting to the batter. The umpire may blink at an inopportune time. A runner may have a toenail brushing against his shoe in a way that makes him a split second slower getting to first, causing an out rather than a hit. To the discerning fan, every game has a different outcome. And it is not unusual to see something strange and memorable at a game even after watching thousands of them, and perhaps even see something that has never happened before.
And this description is for a system in which the conditions are highly controlled compared to that of a typical organization office environment where people get sick, telephone calls disrupt, weather changes moods, and people interact in a myriad of unpredictable ways.
Change in Organizations
Erich Jantsch, in a 1980 book, The Self-Organizing Universe: Scientific and Human Implications of the Emerging Paradigm of Evolution, identified three stages of looking at change—deterministic change, equilibrium-based change, and dissipative or transformational change. Management theorists have looked at his model and tried to apply them to organizational theory.
Deterministic change is related to the Newton model of physics that dominated science (and management as well) until the 20th century. Apply a force to an object after knowing the initial mass and velocity, and you can predict where that object will be at any future point. A manager knowing that a worker produced 10 widgets on the average could produce an additional 10 widgets by simply hiring one worker. Under this model, a grand theory of management would be possible if one simply had enough information, and could then apply this information to any organization, and essentially could expect the same outcome after doing so. Under this model, management’s job was to control workers, and tinker here and there with policies, so output would increase.
A second stage of thinking in the Jantsch model is equilibrium-based change. This model had its roots in biological systems rather than physical systems. It is an “organic” model in that unlike the deterministic system, the equilibrium system adapts to, or resists, the external stimulus in order to maintain a balance or equilibrium. It does so to avoid instability, large variation, and disorder. Organizations under this model resist change, and change can be accomplished only by small, incremental steps. For example, the father of scientific management, Frederick Taylor, identified a phenomenon called “soldiering,” in which workers informally agree to maintain a minimal, acceptable level of output among themselves so as to avoid group discord.
The third stage in the model is transformational change. It is this stage that is consistent with chaos theory’s prediction that random disorder and instability act to increase the possibility for transformation that makes radical, dramatic changes in an organization. Reengineering is a strategy that seeks to take advantage of this model. Managers don’t have enough information to predict what will happen from their management interventions, and thus cannot seek to control their workers. What managers can do is to liberate their workers to control themselves, and provide them with the tools and technology they need to manage themselves.
How should the manager deal with what chaos theory teaches?
First, as pointed out by management thinkers well before chaos theory became in vogue, communication is a key to organizational survival rather than control and rational efficiency. Communication permits the organization’s members to adapt quickly to changing conditions. Even the most astute manager simply doesn’t have enough information to control all staff efficiently (and even if all of the information were available, it would be of such quantity to militate against being able to rationally process it) compared to giving staff authority to organize and control themselves to adjust to the turbulence of changing conditions.
Second, managers need to overcome their bias in favor of mechanistic, linear relationships among their human resources and other inputs and see that non-linear relationships are more likely to describe how people and other resources interact with each other. It means that it is almost impossible to predict the future of an organization because that future is affected by unexpected events. For example, the loss of a single employee in a large organization may have devastating effects on productivity well beyond the value-added contribution of that worker. Increased communication may act to limit the disproportionate deleterious effects of small changes.
Third, recognizing that small changes in processes can have disproportionate overall effects, managers should be looking for these opportunities that some academics have called “lever points” or “leverage.” TQM is one such possible source of leverage. Making a small, incremental change in some process might well provide the straw that breaks the camel’s back and provides a quantum improvement in a process. Or, as one writer pointed out, it is possible that someone adopting statistical process control may be graphing data points and finds a point outside of the upper control limit that just happened at random. Looking at what caused that data point, a change in the process may be suggested that reengineers the process.
Of course, you could make the argument that as a result of finding that random point, a process that was actually in control and working quite well may have been reengineered, causing harm just as easily as benefit. But that is what chaos theory predicts—a small change results in disproportionate results, without a bias as to whether the change is positive or negative to the system. Managers need to be aware of this, plan for it, and perhaps even embrace it.
While it is easy to look at chaos theory in a negative light from the perspective of managers who want stability in their organizations, chaos theory can provide positive motivation for change as well. It is often the “not average” events that occur that permit creative managers to reshape and reengineer their organizations so that they can adapt to changing conditions. Good managers don’t just want to see their organizations survive. They should strive to enhance their organization’s ability to meet the needs of society. Chaos theory provides the unexpected events that permit creative managers to evolve their organizations, and find opportunities that would not otherwise present themselves.
Chaos theory is still developing as a science, and applications of this new frame of reference are being found almost universally in every field of human endeavor. As managers take steps to change their organizations to improve quality and performance, they should be aware of what chaos theory says about even small, incremental interventions.
Up Close: Dr. Douglas Kiel
Dr. Douglas Kiel, the author of Managing Chaos and Complexity in Government, is the director of the Master of Public Affairs program for the University of Texas at Dallas. An associate professor of government/politics and political economy, Dr. Kiel also serves as the campus’s internal total quality management consultant. He has written extensively on chaos theory and its applications to organizations, and believes that non-profit organizations should pay more attention to the lessons that chaos theory teaches, particularly if they are considering introducing a change management strategy.
“ Chaos theory can tell us a lot about the challenges of implementing new management strategies such as TQM and reengineering,” he asserts. “The obvious danger is that any change effort creates new uncertainties and the potential for chaos.”
However, even chaos bounces around within defined parameters, he observes.
“ These parameters represent the ‘order in chaos.’ What managers need to know is what the parameters are that they can expect, or perhaps tolerate, during change efforts,” he relates. “If we had a better understanding of the parameters that our work systems function in we might better appreciate what will be lost and gained during change efforts,” he adds. “We might begin to realize that our systems are pretty chaotic already and thus see that these new approaches might actually help us settle things down and make managing easier.”
Despite a popular notion that many of the effects of chaos theory at work are negative and tend to wreak havoc in organizations, Kiel views chaos theory as a positive weapon in the manager’s arsenal that he or she can proactively harness to improve his or her organization’s quality and performance.
“ Chaos theory teaches us that instability, disorder, and uncertainty can be of real value to managers and organizations,” Kiel says. “The dominant management model has a strong control orientation that tries to avoid instability, disorder, and uncertainty. Yet, how many practicing managers think they can really avoid the chaos of organizational life? When we begin to recognize that instability, disorder, and even a little chaos can lead to new opportunities to create new forms of organization and management, we may find that we can develop new methods for dealing with the turbulence of the current non-profit environment.”
Kiel feels that inducing dynamic instability is a positive force to keep an organization “on its toes” to sustain its overall viability. He also points out that chaos theory also gives us new insight about how we view strategic planning.
“ It seems that we thought we could implement and achieve strategic goals with some kind of engineering accuracy,” he observes. “But the reality of managing is that there may be multiple pathways to achieving our strategic goals.” Non-profit organizations are increasingly turning to these formal planning efforts, but often find unpredicted events thwarting their efforts. He offers two examples of “butterflies” that changed non-profit organizations, one apparently negative and one positive, and which even the most professional strategic planning effort would not have taken into account.
The first example is the infuriating scandal of the United Way of America in 1992, where ethical lapses by its CEO had serious repercussions not only for the national association and its innocent local affiliates, but for the entire voluntary sector. But Kiel also sees the silver lining of this episode. “Didn’t this incident force the new leadership to really rethink its strategies and tactics? Can’t we see this example of disorder as an opportunity for an organization to really reconsider what it is trying to accomplish?”
His second example is the story of the American Paralysis Association.
“ Consider how Christopher Reeve became the lead spokesperson for this important organization,” he points out. “It took just one very quick and small chance event when riding his horse that totally changed his life. This single, small butterfly of an event provided the American Paralysis Association with a respected, articulate, and highly visible spokesperson. In this case, Mr. Reeve’s tragedy can be seen as an organizational butterfly. Such butterflies need not result from tragedy, but most importantly, such stories remind us that small events can have amplifying effects.”
What can non-profit managers learn from all of this?
“ Chaos theory has much to offer non-profit managers if we are willing to re-examine our views of change. Usually we think of change efforts as proportionate with the amount of change we hope to achieve,” he says. “Chaos theory tells us, however, that we may be able to identify leverage points, or butterflies if you will, in organizations and with our employees that allow relatively small efforts to amplify change. The challenge is to look deeply into our organizational and work structures for those points,” he implores.
“ We need to examine the deepest held beliefs and commitments to processes that people hold and consider what changing these values might do,” he summarizes. “We also need to be more thoughtful about how our organizations move in time. For example, we need to be thinking if there is a point in time where our people may be more responsive to such leverage relative to other points in time.”
The challenges of introducing change management into an organization are perplexing enough, but managers willing to take the risks of doing so should consider the confounding effects predicted by chaos theory. Kiel’s research and writing is on the cutting edge of this promising new science, which has lessons for even the most experienced non-profit organization manager.