Contingency

In a lot of ways, Taylorism was the last gasp of a worldview that saw the world as an infinitely complex clock that could be taken apart, understood, and optimized. Modernism was coming, even for business, and researchers started to not focus just on breaking businesses and work processes down into their smallest component parts, but to look at them in context. Zoom out and ask, what is this process part of? What is that part of? And it argues that awareness of context is what will make the right management decisions. 

This makes sense when we compare the rote movements in Gilbreith’s motion studies to the invention process that Conway talked about in his paper on invention. A major criticism of Scientific Management is that it’s inaccurate and inhumane to ask workers to only do, not to think. And I don’t mean to imply that it’s gone, it’s exactly how “unskilled” labor like warehouse workers are treated today. You can see, though, how there’s a fundamental tension here. You can’t optimize the precise actions you want an employee to take when their task is to solve a problem. You can’t do a motion study on thinking.  

So a new way of thinking about management and business organization began to take form in the 1950s, known as contingency theory. Contingency theory supposes that: 

1. There is no one best way to organize
2. But that doesn’t mean that all ways of organizing are equally effective in every context. (Galbraith, 1973, p. 2)

This is a fancy way of saying it depends. But because we are good architects, we have to be able to say what it depends on.

Contingency theory, and the explanation I’m using here is adapted from this 1973 book on organization design that I am not making you read because I am a kind and benevolent professor, argues that complex organizations, like large companies, can be understood as information processing networks. He suggests the main thing it depends on is how uncertain your context is: Structures can arise organically or be determined mechanistically. We would put the Scientific Management approaches we were talking about earlier in the “mechanistic” camp.

Research suggests that centrally-determined, designed forms are most effective in stable contexts, while organic, emergent forms are most effective in rapidly changing contexts. This goes for technology too: Bespoke, fluid technologies suit flat organizations without much supervisory staff; mass-production, standardized tech lends itself to tall organizations. (Galbraith, 1973)

I think it’s important to note here that organic, fluid, emergent structures are not Freeman’s structurelessness. These are formal, if minimal, structures that are flexible and determined by local conditions, not imposed by some general idea of how things can be optimized. Think of Xerox letting PARC do it’s thing.

As information people, we tend to be biased toward order, mechanisms, mass production, standardization. It can take some effort to see the benefits of chaos.

References