•Radical Uncertainty
•Small-world network
•AnyLogic simulation software
•Practical Planning Models
•Small World: Crafting an Inclusive Classroom (No Matter What You Teach)
Because it isn’t practical to fully model a CAS, we need to ask ourselves how we will make practical use of our understandings of the target systems and policies when we create and implement an advocacy initiative.
One common approach is to try to make our existing model of the system or policy more complete by iteratively making it more complex. Our assumption is that a model that is “more like” the target will provide us with a more usable base for an advocacy plan. Unfortunately, this isn’t true, and we need to understand why if we are to realistically use models as tools for change.
As the saying goes, “All models are wrong, some models are useful”. Even a very simple model can give us useful insights into the system or policy we are targeting, but we always need to remember that the model is not the target and that no amount of sophisticated design can ever make it the target. The model is an abstraction and can point us toward insights that will make our advocacy more effective, but it is never a true substitute for the target. And unfortunately, as we make the model more complex, it becomes increasingly useless as a guide to action.
We are so used to processing abstractions as a part of our thinking, planning, and change work, that it is easy to confuse those abstractions with the CAS.
A better approach is to use the frameworks of the practical sciences, like engineering, medicine, carpentry, plumbing, vehicle repair, and so on, as a guide to using our knowledge and experience in an organized way to increase the impact of our work.
These practical sciences are a mesh of theory, experience, training, and intuition that can be used to understand and change an issue in a CAS, even if that CAS is, say, your water, heating, and plumbing system. If water is accumulating in your house, it can be difficult to figure out why when the puddle might be a long distance from the source, and the source might be a long distance from the “cause”. We must diagnose, hypothesize, try small safe changes to see what happens, and apply different tools for different ways of changing the system.
All of this practical engineering becomes part of our long term learning about our target.
If we wish to approach the problem with a better tool, we improve an existing one. You want a better wrench, not a tool that eliminates the need for wrenches, screwdrivers, hammers, and so on. Such an omni-tool tool will only make diagnosis, testing hypotheses, and experimentation in general to solve the problem at hand more difficult, not easier.
Put another way, think Waze when you are trying to formulate a response to a travel problem, not the creation of the entire theory of human travel for the foreseeable future. A good model of traffic flow wouldn’t be based on a representative vehicle. Instead, it would be based on a small world model of typical travel outcomes in the larger travel space. Such a model allows for an interactive dynamic between different purposes to be formulated that reflects why people travel and not simply how they travel.
Creative Commons Attribution 4.0 International License
