Trust, and Magical Management Methods
Avery has been writing about group dynamics, which is a subject that I've put some thought into. I've always been interested in how to get people to do things that are good for them: whether it's getting kids to eat broccoli, or getting friends to invest in RRSPs.
After a while, I sort of gave up on direct coercion; that almost never works well in the end. What you basically have to do is get people to think of these things for themselves, and then they will be happy to do them.
How does this factor into Avery's dilemma of organisational messiness? Let's examine his argument, as challenged:
He's correct in asserting that one person can only do a certain amount of work. That's why groups of people can accomplish tasks no single person could ever manage in a lifetime. There are plenty of examples where co-operation has led to Great Works.
I have to agree with tying responsibility to power. Responsibilty for something you have no control over is just a way of shifting blame. And I'll even give Avery the premise that only one person should be responsible for any one thing, because otherwise the communications overhead will result in some responsibilities being unfulfilled.
But the conclusions he begins to draw are a little weird. These are not if-and-only-if statements.
If we work together to solve large problems (Limited Energy), and only one person can be responsible for a particular outcome (Exclusivity), and the person responsible for an outcome has to have power to control that outcome (Balance), then a logical conclusion is this: for any large problem, a single person must have final responsibility to tell other people on the project what to do, and the power to enforce his decision.
That's quite a leap to get to. A far more reasonable conclusion is: a single person must have the power to determine what contributed work is used, and what work is thrown away. And Avery realises this, because nobody really has true power to tell someone what to do.
This messes up our nice, clean system. Quantization says you can't give someone power over someone else's actions. Balance says you can't give someone responsibility for someone else's actions if you can't give them the power too. Exclusivity says, in that case, that your manager has no power at all. The reason we wanted him to have power was because Limited Energy says he can't do it all himself, and Exclusivity says only a single person must be held responsible for the overall result.
So how do successful managers convince people to assemble large projects? Because it is possible, and often desirable, to acquiesce to someone with a bigger picture. But a hierarchy that subjugates people causes them to malfunction, because they will rebel. And co-operatives often lead to in-fighting because people believe nobody should be permanently in charge.
Which is why I'm enamoured in so-called communities of interest. Where people with similar goals get together to co-operate towards a common outcome. It seems on the surface to be a dysfunctional co-operative, where people argue about things endlessly. People join and people quit, and there's rarely any overall stability. If you look a little deeper, you'll notice that there is typically a stable core of people making decisions. Not just technical decisions, but also social decisions. Oft-times they are difficult and unpleasant ones. And these decisions are respected! So underneath, it looks like a hierarchy. Why do they work this way? Because this core group has proven itself to make good decisions before. If they didn't, people would just stop listening to them.
If you look at business books that tell you how to be a successful manager, you'll notice plenty of "case studies" or anecdotes about how the best managers stay out of the way of their charges. Moreover, they talk about how they try very hard to remove obsticles to work, so that the team can get around to doing the things it should be doing, instead of being distracted all the time. Often, the team delegates unpleasant decisions or tasks to this manager that the manager is good at doing, or doesn't hate doing. Which basically sounds a lot like what communities of interest do spontaneously.
So the pattern I see is that real power is always handed from bottom-up, although most organisations try to wield arbitrary power from top-down. Leading to the inevitable clash between "lowly employees" and "upper management." Because "upper management" has lost, or never had, 天命. Typically, these things turn nasty. Because the true power is always in the employees' hands, and the only thing upper management can do is throw away this power through firings. This is not completely wise, because it hurts both the employee and the organization; if it didn't, unions would be mostly powerless.
Now, I have to be careful here, because I don't want to be arguing for the tyranny of the majority. It is very easy for people who don't have global information to make very good local optimisations that just suck for a large organisation. You see this in large companies where "the left hand doesn't know what the right hand is doing." So what is the correct thing to do?
I'm not really sure, but at minimum, it is obvious that all levels of management must be built on top of a stable foundation of trust. The people being managed need to be able to trust that more global decisions are being made to further the collective goal. If this trust is created and is maintained, it is possible to magically raise the amount of energy a person can direct, because it is seen as a useful end. This is why charismatic leaders are so effective at organising huge movements: they convince people that their goal is laudable and people work towards it because they trust their leader's direction.