If the world we lived in, or more specifically, if the organizations we work in and with, were mostly symmetrical and homogenous, there would be a number of advantages. They would be stable and their behavior would be predictable. It would also be possible to model them accurately, and thus to understand them fundamentally. "Knowing" them would lead to the possibility of controlling them. The problem is, such a world or such organizations could only be very uninteresting. Living things and complex social systems are by their nature heterogeneous and asymmetrical. Complex systems are made up of a multitude of non-linear interactions that cannot be simplified.1 They are unpredictable and full of surprises. There are serious difficulties involved in understanding, let alone modeling, them.
But perhaps the complex behavior of such systems is only epiphenomenal. Perhaps, underneath the multifaceted surface, there are general principles to which the seemingly contingent behavior could be reduced. This would allow us to model the essential behavior of these systems, and not be distracted by the contingencies. Finding these internal regularities was the hope of what could generically be called Modernism.2 This strategy was governed by the ideal to find universal, ahistorical, and non-contingent principles that would describe complex systems accurately and thus allow for prediction and control.
If such an ideal was the guiding principle, diversity would be a problem. It would complicate our understanding and interfere with our planning. It would confront us with the surface of things, not with their essence. It would force us to deal with a countless number of factors, too many to handle. I shall argue, however, that such an understanding of diversity is not only misguided, but dangerous. Diversity is not a problem to be solved; it is the precondition for the existence of any interesting behavior.
The notion of "diversity" is used here in the context of post-structural theories of meaning and of the characteristics of complex systems. These contexts will be unpacked in more detail later, but my general argument is that in a poststructural understanding of language, meaning results from the differences between all the signs in the system. Sameness does not generate meaning. The richness of the system is a function of the differences it contains. Similarly, complex systems are made up of the nonlinear interactions amongst large amounts of elements which are not necessarily complex in themselves. These interactions produce the "emergent" properties of the system, the higher order properties that make the system what it is. A good example is the way in which consciousness emerges from the interaction between neurons in the brain. For this to take place, there must be a large number of neurons which are non-linearly and asymmetrically connected. A small amount of homogenous neurons will just not do it.3
This "necessity" of diversity can also be explained by looking at an organization. To be able to fulfill its role and to cope with a challenging and changing environment, an organization needs diverse resources. The functions of the different components of the organization are not simply interchangeable. The crane operator cannot do the job of the financial manager and vice versa. The more complex the role of the organization is, the more diversity is required to perform it.
The problem to be addressed should now begin to emerge. For an organization to have vital and dynamic properties, it needs a lot of diversity. If, however, we want to describe, understand, control, or manage such an organization, the diversity becomes a problem. We cannot reduce rich, nonlinear difference to simple descriptions, but we need descriptions nonetheless. It was the hope of Modernism that such simplified descriptions - descriptions which are accurate and contain the essence of the matter - could be found. The poststructural argument and the argument from a critical understanding of complexity is that such reductive strategies are seriously flawed. …