Complexity

When Simplification Goes Wrong

Those in the business of reducing complexity assume that simple rules and structures lead to simple organizations with a low level of complexity, but insights from the fields of mathematics, economics, physics, and biology show us how misleading this assumption is.  

An interdisciplinary group of researchers in Santa Fe, New Mexico, has found that simple rules create highly complex systems. It can be demonstrated, for instance, that simple calculation rules, when carried out again and again, form complex self-similar structures.  

The calculation rule z(n + l) = [z(n)²] + c, for instance, yields complex sequences of numbers that exhibit similar but never identical structures. In explaining this phenomenon, which can also be found in chemistry, astronomy, and economics, John H. Holland, a scientist at the Santa Fe Institute, points to games that have very few rules but that, when played, become highly complex. Although, or rather because, chess has very few rules, it is such a complicated game that even grandmasters and powerful computers only partially grasp its complexity. William Brian Arthur, an economist at Stanford University and one of Holland’s colleagues at the Santa Fe Institute, has an even more striking way of explaining the phenomenon. Spill a bit of water on a smooth surface and it will form a complex structure of droplets, not because highly complex rules are involved but because two relatively simple rules are complementing each other: on the one hand, gravity tends to push the water apart and to cover the surface with a thin film of water; on the other, the surface tension of the water molecules tends to unite them into large, compact spheres. The effect of these two simple rules working together produces a complex and unique pattern of droplets. Every time the experiment is repeated, an entirely new arrangement results. 

The desperate battle fought by managers against hyper-complexity and what drives it is being waged against the very same fact that makes chess such a highly complex game and that creates such unique patterns in the water on the flat surface: complexity results from the combined effect of a few simple rules; it is not the result of there being too many rules. The development of Taylorism and of lean management both show that attempts at reducing complexity are ultimately doomed to fail in the face of an environment that is becoming more and more complex. Every attempt to reduce complexity leads to renewed and increased complexity. 

This means that the traditional goals of a management that feels overwhelmed by proliferating organizational structures, constantly growing ranges of products, and drawn-out decision-making processes, whether these goals are called slimming down, reduction, or trimming, will only lead to a further increase in complexity. The systems theorist Dirk Baecker describes this paradox as follows: ‘Every simplification increases complexity, a complexity that doesn’t crop up just anywhere, but rather precisely at the point where the simplification was carried out … Simplicity is not the antonym of complexity; instead, it is a moment of overcoming complexity that contributes to the increase of complexity.’ Contrary to what common sense suggests, simplicity is thus not the opposite of complexity. Businesses are not simply faced with a choice between strategies that increase complexity and strategies that reduce it. An organization is not either highly complex or simple, and by introducing simplification strategies it may actually increase its complexity. 

If we resist the conventional view of simplicity and complexity as opposites, we can begin to understand why ideas about how to reduce complexity rest on questionable premises. A strategy of rationalization that reduces temporal leeway increases the risk of disruption. Kaizen, the continuous improvement approach, results in a total loss of flexibility. Perfecting the production process leads to increased complexity. 

The complexity dilemma for post-bureaucratic businesses consists in the fact that, given the obscurity and complexity inside and outside the organization, employees crave simple, lean structures that reduce complexity, but these are precisely the structures that lead to a further increase in obscurity. The fact that even highly bureaucratic businesses strive to reduce complexity is cold comfort for post-bureaucratic businesses. That even the employees of businesses which have spent years in organizational paralysis still seek more simplicity and security is, for the employees of post-bureaucratic businesses, a taste of things to come. 

The threat of the dilemma – that the sought-after measures for reducing complexity will only increase it – lies not so much in the dilemma itself but in an unwillingness or inability to acknowledge it. The current trend for simplification strategies in management circles carries risks – even ‘deadly risks’: because people expect these strategies to reduce complexity, the increased complexity that results is not registered. The complexity that results from simplification strategies can then no longer be mastered by the organization. The forced simplifications run counter to the system’s own rules. The system ‘defends’ itself and develops a complexity which the management is no longer able to control.