More than 25 years ago, a young organisational psychologist named Karl Weick published a highly original and nearly impenetrable book called The Social Psychology of Organising. Alone in the wilderness, Weick promoted the adaptive advantages of chaotic systems, distributed authority, and "sensemaking." "Stamp out utility" and "Complicate yourself" were among his battle cries. Weick's approach to his own work has been novel: while most social scientists massage colourless statistical data, Weick hangs out with the Utrecht Jazz Orchestra to study how organisations function. The Rensis Likert Collegiate Professor of Organisational Behaviour and Psychology at the University of Michigan, he has written on topics as diverse as labour strikes in outer space (the Skylab crew) and the Naskapi Indians of Labrador. He spoke with John Geirland about why, in a wired world of constant change, "chaotic action is preferable to orderly inaction."
Wired: What will organisations be like in the future?
Weick: I think you're going to lose middles. People talk about the disappearing middle class; there's no more middle management; and midsize organisations really don't exist any more. More importantly, there'll be a lot of chronic ambiguity. For instance, many organisations have stopped publishing organisational charts because they become obsolete the day they get circulated. Ed McCracken, chairman and CEO of Silicon Graphics, says flat out that it's a waste of time trying to project technological trends more than a few months out. If you take chaos theory seriously, it asserts that the world is both unknowable and unpredictable. All you can do is engage in transient moments of sensemaking.
The kind of sensemaking you've written about is a social activity, not something we do alone.
Problems are too complex these days for individual minds to comprehend. Look at air-traffic controllers, who sit around radar screens. You've always got pairs of people there, and there's an awful lot of chatter back and forth between controller and assistant. I don't think the mind is located inside a single head, anyway.
Problems are more complex, but we also have more "knowledge workers" for dealing with them. Management philosopher Peter Drucker says that knowledge workers can't be supervised. Do you agree?
Back in 1973, the third Skylab crew had a tight schedule of experiments to run. NASA kept leaning on them to take on more experiments. The crew got more behind, more overloaded, so it turned off the microphone for 24 hours and spent some time reading and looking out the window. This says something about how companies blend control and autonomy. People are better able to get complex assign-ments done when given more discretion within a frame- work of common values.
Most organisations try to simplify and streamline. But you say that organisations should become more complicated.
It's the law of requisite variety, which says that if you want to make sense of a complex world, you've got to have an internal system that is equally complex. A good example is the Naskapi Indians of Labrador. Their problem is where to hunt for caribou. The hunter holds the shoulder blade of a caribou over a fire until it develops cracks. Then somebody reads those cracks to see where the caribou are likely to be. The wisdom of this practice is that it randomises the hunter's behaviour, making it harder for the caribou to learn where the hunter is likely to be. It also ensures that some areas don't become overhunted. The translation should be clear to people running businesses. In fact, there are examples in Asian management practices of ancient rituals being given considerable stature.
Another one of your rallying cries is "Stamp out utility!" What does that mean?
Stamping out utility is holding on to practices that are inefficient in the short term. During the Gulf War, the US Army had PCs and it was all wired up for handling logistical operations. But Gus Pagonis, who was in charge of logistics, reports that most Gulf War requisitions were made using 3-by-5-inch cards. The cards didn't have utility in the sense of speed or central storage, but you could make them up at ease, carry them to different locations, discard them once the requisitions had been filled. Stamping out utility is the idea that you never know what is going to crop up next, so you ought to have some things in the system that, given your current problems, are useless.
You've written about "galumphing" - a kind of purposeful playfulness - as a source of innovation and increased adaptiveness for organisations. If I were the president of a multimedia start-up, how might I go about galumphing?
I'd give you a video camera and say, Go make a video that represents what is on your computer screen right now, or Make a video of a person who has never seen a PC before, or Shut down all the PCs in your company for the day and have your people work with pencil and paper. Galumphing is a way to come up with novel combinations of technology or ideas that may give you the capability to craft a new multimedia application in the future.
You also say "discrediting" - purposely turning your back on what has worked in the past to avoid future traps - is a strategy for organisational survival and innovation. How should Bill Gates use discrediting?
Firefighters are most likely to get killed or injured in their 10th year on the job, when they think they've seen pretty much everything there is to see on the fires. They become less open to new information that would allow them to update their models.
Discrediting tries to solve these problems of hubris or seeming infallibility.
An ultimate use of this strategy would be to ask Gates, "Do you think you're the person - with your blind spots and vision - to take Microsoft through the next phase?" If I were Gates, I'd think about the possibility that the good times were making me look wiser than I may in fact be.
John Geirland (jgeirland@aol.com) is a management consultant and novelist who works with organisations on collaboration, creativity, and managing change.