One of the key concepts that we use in our discussions is this idea of premature load bearing. And as those words suggest it means that we’re essentially asking our systems to do a lot of very difficult work but before they’re really ready for that. And that turns out to be something that’s really important because the systems that we have for managing our modern world themselves are very sophisticated, but they took a long time to get that way.
And perhaps one analogy or one way of thinking about this is to see the capability of organizations as kind of similar to how we learn languages or how we learn to play an instrument. All of these tasks take a long time to acquire and you’re not very good at them until you practice and until you start getting good at them.
And if you’re asked to give a lecture in Swahili and you’ve never spoken it before you’re asking someone to do something that is way beyond what they’re actually capable of. And, just being able to take a few quick classes at it doesn’t actually get you where you need to go. You have to do the hard work day in, day out, for a long period of time to acquire the capability to be able to do those kinds of tasks.
Now, what we often find ourselves doing too often in development, unfortunately, is making our systems go too quickly to an end state where they’re being asked to perform tasks that are way beyond what their actual capability actually is. So, we have lots of otherwise very noble aspirations to build the rule of law, to engage in climate change, to be able to protect workers in the domestic work place, or to engage in trying to stop domestic violence. All of these things are very difficult things for organizations to do, and the fact that they happen in some countries is often a product of decades if not centuries of learning how to do those kinds of tasks, not just by individuals but with thousands even hundreds of thousands of people simultaneously knowing how to form and perform those particular tasks.
So, a simple example for this could just be an example of a bridge, if you wanted to cross from one side of a river to another. From a distance you could build something that looks like a bridge and that could be photographed, it can have lots of admirers looking at it from afar as a wonderful example of what a bridge could be. But if anybody actually tried to walk on that bridge, perhaps they would fall through. And if it turned out that the bridge was made of paper rather than that of bricks and concrete, it just wouldn’t be able to perform the task that it needed to do.
So, I think this is one of the big ideas that we’re struggling with in this class and hopefully that you’re working through as well, is how do we build systems that can eventually run but that we help them initially by helping them to walk, by being able to do these initial tasks in a way that requires this very particular kind of skill to be able to perform.
And if we do this too fast, then a lot of problems occur. One of them is that you will, probably in all likelihood fail and that could delegitimize the very task that you’re trying to engage in. If it’s very complicated and there’s a lot of money and a lot of political resourcing behind this particular task, and then it doesn’t work because you’ve asked it to do too much too soon, then we could be back, not to just square one and starting all over again, you might be back to square minus one. You might actually have compromised the very legitimacy and idea that you could engage in such a task.
So getting that sequencing right is very important. Getting the expectations right is very important. But from a programmatic and practical point of view ultimately what we’re trying to do is to help the systems acquire this capability to do what is ultimately very difficult kind of work.
And at the end of the day that makes it very similar to the way you learn to play the violin, the way you learn to speak a foreign language. You have to learn over a long period of time and you have to persist with it. And you have to recognize that often times there’s going to be a lot of low level performance that’s required ultimately before you build that capability.
But for organizations to do this, that task just becomes even more complicated.