If you're wondering what will be the next Big idea after reengineering, here's a prime candidate.

Chaos Inc. by Simon Caulkin

In 1987, recounts M. Mitchell Waldrop in his book Complexity (Simon & Schuster), a researcher at the nuclear physics lab at Los Alamos, N.M., caused a sensation by demonstrating a computer-generated flock of birdlike objects that he christened "boids". On the screen, starting from a random distribution, the boids formed into flocks: Just like the real thing, they swarmed, flowed smoothly around objects, and merged together again in a twisting, constantly changing formation.

Yet there was nothing in the software that told the boids to generate collective behavior. They were programmed individually with just three rules: Keep a minimum distance from other objects, try to fly at the same speed, and head where the concentration in densest in the vicinity. Flocking is one of the mysteries of avian !life, but here it was: a complex, ever varying pattern of behavior emerging from just three simple rules.

It's a crude analogy, but when Mike McMaster, managing director of Knowledge Based Development Ltd., a Berkshire, UK, management consultancy, recently undertook an assignment to increase productivity at the construction site of an offshore oil platform, he did something similar: With the help of a cross functional team, including pipe-fitters and welders, he distilled the project down to just four key principles. Counter to intuition, the aim was to produce not simplicity but a complex and adaptive intelligent entity that would adapt to changing conditions as swiftly and organically as a flock of birds.

Says McMaster: "The project is a complex entity already. The problem is, we typically kill the complexity with rules and rigid structures which block the natural flow of information. So what we do, we just look at the four principles and let the rest take care of itself, so that the creative complexity emerges."

The difficult trick, McMaster says, is to view the principles not as independent but as interactive, working together to produce a higher collective behavior that, as with the boids, is different than the sum of the parts. Why four principles? "Because you can't control four. Companies often say, can't we just have two? But that misses the point - with two you can't get the rich interplay that we're after.'

If this sounds like Alice in Wonderland management, buckle your seat belts - there's much more to come. For several years now, math Ph.D.s have been using the arcane calculations of chaos and complexity theory, pioneered at organizations such as the Santa Fe Institute (SFI) in New Mexico, to try to predict movements in trading environments, and there is at least one venture-capital-funded complexity start-up, the Santa Fe-based Prediction Co., in this field. Insurance companies, too, are deriving new approaches to risk analysis from chaos theory.

But beyond the first wave of mathematical theory, McMaster is one of an advance guard of explorers introducing the concepts to mainstream business. His forte is organizational design - and, he claims, although his clients aren't too hung up on the scientific theory, they're happy enough with the results. This stuff works. Productivity reportedly doubled at the construction yard, and a chemical company that had already sweated blood to get the cost of a plant down to economic levels took out a further 30 percent by working to the same principles, McMaster says.

Other hardy experimenters are reporting business gains all over the shop. Farm- and industrial-equipment manufacturer Deere & Co. is using insights from complexity theory to figure out how to handle rapid changes on its assembly lines. Understanding how components of a complex system interact could help it choose the right manufacturing technology, it believes. Coopers & Lybrand, an international consulting firm, is helping telecommunications clients handle fast-changing strategy issues with a sophisticated business-simulation game.

As for Citicorp, chairman and CEO John S. Reed was instrumental in urging SFI to undertake a pivotal economics program, as a result of which it is now gaining "hard operational performance improvements" from its better understanding of the behavior of customers and markets. Citicorp senior technology officer Colin Crook declines to be more specific: "Suffice it to say that results are so positive we sure as hell don't want to talk about them," he declares. Complexity, he predicts, will be "the next hot thing" in consultancy after reengineering - and when it appears, Citicorp aims to have ready "a solid cadre of expertise on the basis of our own experience." But complexity and chaos theory come with a built-in disclaimer: Don't expect it to be easy. McMaster prefaces each assignment with several days exploring fundamental operating and design principles of intelligent systems, and, he says, it takes "a lot of digging" to get down to the base-level principles that should drive a work site or multimillion-dollar project. As he and other workers in the cause admit, there's not yet any fully developed language or accepted set of metaphors to describe the complex intelligent organization. Cautions Crook: "This is nontrivial stuff. What we're talking about is a real restructuring of thinking about how the world works and the way to handle it. A lot of conventional organizations just won't get it."

Command, Control, Complexity

Complexity goes against the grain of just about everything that most managers believe in. Up till now, the history of management can be seen as a constant striving for order, control, and predictability. Today's "enlightened" policies of individual empowerment and flat organizations are no different. The aim is still to proceed in an orderly fashion from the A of now to the B of the company's strategic vision: It's just that nowadays freeing up the troops to make their own decisions is assumed to be a better means of getting there than direct command and control ..

The disorienting implication of complexity, on the other hand, is that there's no sure way of getting to B at all. Complexity says that long term outcomes for a complex entity such as a company, a market, or an economy are essentially unknowable. This is because the relationships between actions and outcomes are nonlinear; through intricate feedbacks causes become effects and effects causes, so that in practice causal links can't be traced. And depending on initial conditions, apparently insignificant variations can be magnified by positive feedback into huge consequences.

It may be that if atmospheric conditions are right, the heat from a jogger in Hyde Park will set off a hurricane weeks later in Florida - but no computer will ever be able to track the chain of causes.

As if that weren't enough, managers have to cope with a second dimension of flux and uncertainty, explains Ralph Stacey, professor of management at Hertford Business School and author of two studies on management and chaos theory. "Purely physical or chemical systems are deterministic - the rules stay the same, even if actual outcomes are impossible to predict precisely. But complex organic systems such as species or ecologies or societies are adaptive rather than deterministic in that the rules change in the light of the consequences of the behavior they produce", he points out. They evolve to different levels, often in sharply unexpected directions, at which point a new set of rules applies.

One simulation project at SFI is an artificial stock exchange in which traders "learn," evaluating behavior and modifying their operating instructions according to results: The simulation produces all the hallmarks of a real stock market, including surges, periods of little movement, and crashes.

Some managers, like Citicorp's Crook, respond eagerly to this radically different description of the way the world operates indeed, it was dissatisfaction with conventional models of economic behavior that led CEO Reed to SFI in the first place. Others are deeply upset by its implications. "You have just set management back 15 or 20 years!" one apoplectic British manager reportedly exploded in one of Stacey's seminars. His reaction is not surprising. Liberating for some, complexity makes of management a fundamentally paradoxical activity, a confusing mixture of order and chaos.

For a company to be capable of creativity and novelty, it must operate in the area of "bounded instability" or "edge of chaos" where alone novelty is possible - yet without compromising the order needed to accomplish day-to-day tasks. "It's a knife-edge", admits Stacey. "If a company tips over into total randomness, it disintegrates completely; if it gives in to the very powerful forces to adapt to a given situation and become a single-equilibrium organization, it ossifies and dies like the dinosaurs."

Adapting to Dynamic Conditions

If the complexity theorists are right, some of management's most cherished shibboleths are in need of an overhaul. For example:

• Conventional doctrine says that management is a negative-feedback activity setting a strategic aim and moving the company toward it by correcting deviations from plan. Under complexity the picture is more complicated. The conventional theory is right for day-to-day activities - routine order processing or manufacture of standard parts. But for creative activities like long-term strategy setting, it is dangerously misguided.

  The distant outcome of actions can't be plotted, because the structure of the system makes the future unknowable. The corollary is that viable strategy is not something that is the result of prior intent by a foresightful leader. Rather, it emerges from the multiple possibilities thrown up by messy group dynamics in organizations in collision with the environment. Accordingly, says Stacey, managers should think of themselves as gardeners rather than executives - "instead of intending it they must let it happen."

• In the management literature, executives control the company through orderly structures and procedures. If that's all they do, in a complex world the company is destined to go the way of Tyrannosaurus rex. Attempts to make the system stable work only at the expense of making it incapable of interacting with the environment to create an alternative future. Result: stagnation and death. For a cautionary example, look no further than the economy of what used to be the Soviet Union. In a chaotic dynamic, power must be distributed to the maximum - "real empowerment, taken not given" - which is another reason why the concepts are less than popular at the top of large organizations.

• Consultants emphasize that strongly shared cultures and values are essential to steer the company into the future. Again, this is good for honing standard routines. But in dynamic conditions, where the future consists of multiple, shifting agendas, a monolithic top management will likely fail to generate the creativity to give the company adequate options ahead. For this, diversity of opinion and approach is needed to fight consensus. Straight-ahead thinking untested by different visions may be one of the most important contributory factors to the fall from grace of so many once "excellent" firms over the past turbulent decade.

• Most strategists believe that success is the result of maintaining adaptive equilibrium with the environment. If this were true, managerial free would be reduced to the choice of getting the fit right or getting it wrong. In the world of complexity, the stakes are much higher. First, equilibrium is death. Second, in unstable, evolving conditions, the environment adapts to the company as much as the other way about. The implications are that the company can't blame the environment for failure - and the successful business is vertiginously free to create its own future. The trade-off, of course, is that it then must be able to ride the bucking bronco it has created.

• Most firms accept the traditional view of markets and economies as self-correcting mechanisms: Major changes are damped down by the reactions they themselves engender. The resulting equilibrium is the best possible outcome in the circumstances. Over the last few years, economists like W. Brian Arthur, Citibank professor at SFI, have begun to develop an alternative view of economic processes that emphasize the operation of positive feedbacks, or increasing returns (or vicious or virtuous circles), in many markets. To the discomfiture of conventional economics, this makes multiple and apparently arbitrary equilibria feasible. Take VHS videocassette recorders or the DOS operating system for personal computers. Both won out over the opposition not through an optimum technical solution but when initial early gains in market share - whether through corporate skill, accident, or external circumstance- suddenly snowballed. Increasing returns ensured that the "winner takes all," locking the future into the past - as the fortunes of Microsoft Corp., for one, confirm.

Chaos and Order: Visa and the Internet

Citicorp, like almost all other corporations dipping their toes in these waters, is still recognizably a company built on traditional, precomplexity lines. So what would a full-fledged, distributed, self-organizing intelligent organization look like? One role model might be the Internet, a complex and creative organizational form that is evolving its own rules as it goes along, without benefit of headquarters, central ownership, or long-term strategy. But perhaps you don't see chairmen of the Fortune 500 buying something that well - way out. In that case, try an example a little closer to home:

Imagine an organization with a globally respected brand name that is used by 23,000 institutions in 200 countries and territories; whose market-leading products are used by 355 million people to make more than 7 billion transactions worth $650 billion annually - the single largest block of consumer purchasing power in the world economy; that would have a value of around $150 billion in stock-market terms but that can't be bought, sold, or raided, since the 23,000 institutions that create its products are also its owners, members, customers, subjects, and superiors, holding ownership in the form of perpetual membership rights. Note, however, that that portion of the business created by each member is its sole property, its value reflected in that member's stock price. Oh yes, and that for a quarter of a century has been growing between 20 percent and 50 percent compound per annum.

The name of the riddle? "A little ubiquitous thing called Visa," chuckles Dee W. Hock, founder and former president of Visa International. Contrarian Hock has argued for decades that Newtonian, hierarchical organization in whatever sphere is an "aberration of the Industrial Age, antithetical to the human spirit, destructive of the biosphere, and structurally contrary to the whole history and methods of physical and biological evolution" - not just irrelevant but a menace likely to lead to an epidemic of institutional failure. Convinced there was an alternative, Hock has spent much of his life patiently building an organization on a biological rather than mechanical metaphor, capable of evolution and self-determination.

Now the world seems to be falling in step. Organizationally, Visa parallels to an uncanny degree the scientific precepts of complexity and chaos - indeed, Hock dubs Visa and the Internet "chaords," seamless blendings of the principles of chaos and order, competition, and cooperation. And where in the past people scoffed or yawned, he now finds an eager audience for his views. "The seeds of chaordic thinking are sprouting everywhere," he told the conference of the Bionomics Institute in San Francisco last October. "It is my personal belief, although I would be hard-pressed to prove it, that we are at that very point in time when a 400-year-old age is dying and another struggling to be born; a shifting of culture, science, society, and institutions enormously greater than the world has ever experienced."

With institutional backing, Hock is currently researching how the pump might be primed by creating four of five more influential "chaords," stretching over the fields of education, government, social services, and commerce. Others take a more evolutionary line. But those who've worked at it agree that there is only one direction to move in. "Complexity is going to be hard for people to take and turn into the hack stuff that consultants are selling in the process reengineering ares," emphasizes Crook. "Those who do understand it will be differentially advantaged. In my experience, this really is the way the world works."

Chaos Unraveled

The graceful images on these pages were computer-generated by Clifford Pickover, a researcher at IBM and a prolific author on the subject of chaos and computers. In Chaos in Wonderland (St. Martin's) and three other books, Pickover invites readers of mathematical persuasion to generate their own chaos - with beautiful results. We asked him to demystify this complex subject for us:

To ancient man Chaos represents the unknown - menacing visions that reflect man's fear of the irrational and the need to give shape and form to his apprehensions. Chaos theory today involves the study of a range of phenomena where there is a sensitive dependence on initial conditions - i.e. where changing a parameter slightly in an equation or system can result in very different behavior. From toys with randomly blinking lights to eddies of cigarette smoke, chaotic behavior is generally irregular and disorderly. Other examples include weather patterns, some neurological and cardiac activity, the stock market, and certain electrical networks of computers.

Although chaos often seem totally "random" and unpredictable, it actually obeys strict mathematical rules derived from equations that can be formulated and studied. And, though it may never be possible to precisely predict phenomena like the weather or the stock market, one might foresee the global patterns of their behavior - the "order within the chaos."

Understanding Chaos

In the basement of the London School of Economics (LSE), notes Professor Paul Ormerod in his 1994 book The Death of Economics (Faber and Faber), there is an extraordinary machine built by a distinguished economist in the 1950's to teach the workings of the British economy to his students. "Levers are pulled, buttons pressed. Sluice gates open, and liquids of different colors rush around the tubes of the system in a controlled way."

For three centuries since Isaac Newton, scientists have pictured the world as fundamentally resembling the LSE's wondrous model. Governing this world were principles of regularity and order. All things were the sum of their parts; causes and effects were linked in direct, linear fashion; and systems moved in deterministic, predictable ways. Of course, scientists had long been aware of phenomena that appeared to contradict this linear logic: The swirling shapes of flames in a fire, eddies in a stream, and cloud formations, for example, can't be rendered by simple linear equations. And economies, of course obstinately refuse to obey the forecasting models. But still, scientists reckoned, these were exceptions that did not ultimately disturb the ruling Newtonian paradigm.

It took the development of chaos theory in the 1970s and 1980s to suggest a very different model of the way the world works. Says Ormerod: "The single most important scientific advance of the latter decades of the 20th century has been the perception that the world is fundamentally nonlinear."

Chaos in this sense is a misnomer. Part of what has become the broader science of nonlinear dynamics, or complexity theory, linking disciplines as diverse as physics, biology, chemistry, economics, and sociology, chaos denotes a unique area of "bounded instability" as entities move between equilibrium on one hand and complete randomness on the other. It is in this area alone that creative behavior occurs. Flames, eddies, and cloud formations are classic chaotic systems: Operating far from equilibrium, they are unpredictable and richly creative in detail, yet constrained within certain physical limits.

In nonlinear chaotic systems, the links between cause and effect vanish in amplifying feedback loops that can turn tiny initial variations into huge consequences - hence the famous example of the flap of a butterfly's wing in Beijing causing a hurricane in Hawaii weeks later. In effect, the future of such systems is unknowable.

It was quickly clear to adventurous thinkers in the social sciences that chaos and complexity theory potentially threw much light on human and biological organizations such as companies, markets, economies, and ecologies. They too were complex creative systems made up of many interacting agents with a seeming drive to self-organization - people into markets, birds into flocks, embryos into cells, simplicity into complexity. And unlike flames or clouds they are adaptive, so that the rules of their behavior change as they evolve and learn.

Points out M. Mitchell Waldrop in his 1992 book complexity (Simon & Schuster): "They actively try to turn whatever happens to their advantage....Species evolve for better survival in a changing environment - and so do corporations and industries. And the marketplace responds to changing tastes and lifestyles, immigration, technological developments, shifts in the price of raw materials, and a host of other factors."

This world id not the one represented to the economics students of LSE. Things are more than the sum of their parts; equilibrium is death, increasing, not diminishing, returns are the rule, caused are effects and effects causes; disorder and paradox are everywhere. In this new universe new rules are necessary. Nothing, sums up Dee W. Hock, founder and former president of Visa International, "can be made simpler without becoming more complex."-

July/August 1995 - Across the Board