Thinking in Systems, Ch. 3: Resilience, Self-Organization and Hierarchy

so why do systems work so well and this this video is me basically going through chapter three from this book and she has three reasons why systems work so well one is resilience the second is self self-organization and the third is hierarchy and she makes so many really important points in this chapter so i'm just going to sort of go through each of these one by one so first what is resilience and let me just read her favorite definition she acknowledges that different fields have different definitions for this but hers is pretty good she says

resilience is the ability to bounce back or spring back into shape position etc after being pressed or stretched elasticity the ability to recover strength spirits good humor or any other aspect quickly the opposite of resilience is rigidity or brittleness and if you ask how do systems actually become resilient she has an answer to that too which is that systems have a rich structure of feedback loops within them so the feedback loops are really key she also points out that these feedback loops have redundancy built in so that if one part of the system fails or

one of the feedback loops doesn't quite work doesn't quite get the right signal then there are other feedback loops that sort of kick into gear to make the system work so redundancy of feedback loops is important and a lot of times that redundancy is feedback loops that operate under different time scales where some of them are fast repairing feedback loops and some of them are feedback loops that maybe operate on a on a longer time scale like a year instead of a week or a week instead of a day she also talks about meta resilience

which is what is sometimes known as anti-fragility in nassim taleb's book by the same name and meta resilience is basically a system's ability to learn from its mistakes or its ability to learn from negative shocks and negative events so that next time that negative thing happens again the system is better able to handle it and you might think of allergy shots as a version of meta resilience where allergy shots is where they give you a little bit of serum that has uh dog dander or whatever you're allergic to in it in a small amount in

the first week and a bigger amount the second week and an even bigger amount the week after that so that your body develops an ability to recognize that thing you're allergic to and it develops systems for handling that allergy so meta resilience is anti-fragility means the system will get stronger the more it encounters negative shocks and of course the the truth there is that actually um in an ideal situation the system gets stronger at the right pace so that you give people or give systems the the shocks that the system can actually learn from there's

always going to be a shock that is way too big for the system to handle that will kill the system so in some ways a lot of systems will work in terms of handling smaller shocks developing resistance to those shocks learning from the smaller shocks and encountering bigger and bigger shocks over time she points out that systems are dynamic they're ever changing they're ever moving ever responding to things if a system has static stability in the sense that it's just stayed the same for a long time that is actually a very fragile system and it's

unlikely to be able to handle shocks to the system well because it's sort of been stuck and it's lost its dynamic element and we can even think of this one in terms of war for example if you have a country that's constantly in little skirmishes little wars where the army has to respond to different things and they have their skills sort of built up for responding to that that country may be better able to handle a bigger war than a country that's just had a long period of peace where nobody has actually experienced real war

she points out that a common breakdown in systems is some kind of breakdown in the information flow or in the signals that are sent that guide the different feedback loops so distortion of information can cause a system to break down and i really loved her example of this she said chronic illness is often times the body's inability to get signals to keep signals up to date and therefore um it's basically a weakening of the immune system in the body where the immune system is the series of feedback loops that help our body respond to negative

things that happen when the information signals throughout our body start to be less effective at communicating with other systems in our body then our immune system breaks down and that can cause chronic illness and then the last part of this section of the chapter on resilience that was important i thought was her analogy with the plateau where she said she thought of resilience as the plateau of situations that a particular system can handle um can it handle only a small area of uh of sort of other things that happen or can it move around on

the plateau without falling off and she said a lot of times people focus on the strategy of a system which is basically the game play on any one given part of that plateau whereas resilience is more it's not about the gameplay in that one section but it's about can the system move around to different situations on that plateau different different scenarios and operate perhaps using different strategies so i think if we're trying to imagine the plateau she's uh she's talking about here it's a plateau where you might have different rules of the game that are

going to to work differently at different parts of that plateau and the more adaptable a system is the more easily it can change between one set of rules to another set of rules depending on the scenarios it needs to respond to so the second thing about systems that makes them work well is the property of self-organization which she defines as a system's capacity to make its own structure more complex and of course this relates to anti-fragility or what she's calling meta resilience it's the ability for the system to sort of reorganize to handle new situations

and of course our third trait is going to be hierarchy and there is going to be a little bit of tension between the self-organization property of systems and the hierarchy property of systems because self-organization is actually more of a disorganized process where the system has the flexibility to sort of figure out things on its own and to to build its own structure and building its own structure will start with chaos essentially in this section she points out the fact that systems will oftentimes fail because they aim at short-term goals instead of long-term goals and they

get really caught up on bolstering short-term goals in a way that actually makes the system more fragile and less less able to organize itself in the future and when i'm thinking about this i think about entrenchment where a lot of times in these large organizations whether it's a corporation or a government or whatnot um they'll have some situation within the company or within the government organization and they'll put out these rigid rules to handle that particular situation and sometimes the situation is arising because of a certain personality type that has a certain role and the

rules are meant to sort of hold that personality type in check but it's responding to the specific person in that role rather than thinking about do these rules actually work to serve the organization even when a new person comes into that role and of course if you have those rules on the books and if you have people whose job it is to enforce those little rules that were meant to deal with that one person then when that person retires or moves on you have this rigid structure that doesn't do a good job of self-organizing in

the next phase of the company when you have a different personality type there so that would be an example of a short-term versus long-term focus where the short-term focus and the rigid structure actually made the system less able to self-organize in the future because of the rules and really entrenchment is the opposite of this it's making the system so rigid that it cannot self-organize in the future she also talks about the problem where the system tries to treat people like robots or like algorithms or machines where it sets up a set of incentives and doesn't

allow the people within the system to have their own agency or to work on their own to achieve their own goals and when it does that then the system is is less able to use the creativity and ingenuity that real human beings can have when they're incentivized a little bit but given some freedom to do their own thing the final piece here is the hierarchy and she argues that systems naturally arrange themselves in hierarchy with systems and subsystems and subsystems to that within a body system there's subsystems that perform different functions like your nervous system

and your liver and your bladder in your stomach each of those has its own role in corporations the corporation divides down into departments and that would happen naturally if you try to figure out how do we actually function she points out that when systems evolve hierarchy naturally a lot of times it's from the bottom up it's a group of families that live in the same region coming together to form a mini city or a mini town and the group of towns coming together to form a state and oftentimes the building of that hierarchy is aimed

at serving the individual parts so that the individual families they want more efficiency in the way they do farming and that's why they form into coalitions that become towns why is hierarchy so efficient well one of the reasons is it reduces the amount of information that each node needs to keep track of so that when you divide it into many little parts you can keep really close track of the people in your little family or your little town but it would be impossible to keep track of everyone in an entire huge nation and all of

their characteristics and traits and economic advantages all of that she also points out that there are denser connections between people or between nodes at the smaller level there's a whole bunch of really intense connections at that smaller level and weaker connections going across different systems so if you look at your body the cells and the chemicals within your stomach are constantly communicating with other cells within the stomach but they may not have a as close connection with the cells in your foot or the cells in your brain even though there are some connections so there's

just denser channels of communication there's more more frequent channels of communication within a smaller subunit same thing with the army you really get to know the other people in your 10 person tent but you don't get to know everyone in your squadron quite as well there's just more connections the closer the closer knit within the system now the last thing she points out is that each little subsystem is going to have its own goals and those goals may actually conflict with the goals of other subsystems or what the goals of the system overall and if

you think about an organization that has an accounting department which is trying to make sure everybody everybody keeps rigid track of things and a marketing department that's more creative and the individuals within those systems they're motivated differently they're different types and you could imagine there could be conflicting goals like the creativity department is going to want a lot of freedom to not have to track every little thing um the the accounting department wants everything tracked and that creates a conflict and those two the conflict between those two subparts might might really not help out the

overall system the overall corporation so there are conflicts between different levels and conflicts between the lower levels and the higher levels and for a system to work well it needs to align the incentives across the different levels allowing for different motivations at the lower more minute parts of the system but making sure that the incentives align this is obviously one of the biggest jobs of a ceo of a company is looking at the different parts within the company and making sure they're each properly incentivized to serve the goals to maximize profit for the company