• Welcome to droqen's forum-shaped notebook. Please log in.

Thinking in Systems (A Primer)

Started by droqen, February 11, 2023, 04:23:59 PM

Previous topic - Next topic


Regarding Donella H. Meadows'
"Thinking in Systems"


I neglected to keep good notes for this book, so I'll have to go back and find the things that leapt out at me. There have been several. Compared to my recent reading I am finding this book dry and obvious.. but great

EDIT: So many connections! I created a whole forum in order to better take the type of notes that were arising from its reading.

~ Discernible macrobehaviours of a system. (Avoiding teleological thinking.) ". . . a system must consist of three kinds of things: elements, interconnections, and a function or purpose"
~ Self sufficient subsystems, and evolution. "Why the Universe Is Organized into Hierarchies--a Fable"


Quote from: page 116
When various actors try to pull a system stock toward various goals, the result can be policy resistance. Any new policy, especially if it's effective, . . . produces additional resistance, with results no one likes, but that everyone expends considerable effort in maintaining.
Let go. Bring in all the actors and use the energy formerly expended on resistance to seek out mutually satisfactory ways for all goals to be realized--or redefinitions of larger and more important goals that everyone can pull toward together.


The compatibility:

"Do without doing." In Ursula K. Le Guin's translation of Tao Te Ching, this phrase or something like it repeats itself a few times. Actually many times. "Act without acting." There is something about "Shape without sawing" or "Shape without cutting"? Something like that.

In The Nature of Order, or actually more in The Timeless Way of Building, there is talk about resolving forces. The idea behind patterns is that a conflict is at play.

Thinking in Systems presents a language for more precisely describing the nature of certain conflicts. I think the language is very intuitive to me, it's no big surprise, it's like filling in the cracks with a slightly more precise shape of what was already there all along, given to me by Nicky Case's "LOOPY" and perhaps some of Dan Cook's design diagrams [todo: find/add link].

The One-Straw Revolution describes systems too -- the boundlessness of systems, the damage of assuming one is understood...


Quote from: page 12[One example of a non-system is] a conglomeration without any particular interconnections or function. Sand scattered on a road by happenstance is not, itself, a system. . . .

Quote from: page 12When a living creature dies, it loses its "system-ness." The multiple interactions that held it together no longer function, and it dissipates, although its material remains part of a larger food-web system.

So early in this book! Defining what a system is in a way that I was not a hundred percent on board with.

Meadows says that a system has a "function" or "purpose", which rubs up against my discomfort with the way these sorts of terms lead into a teleological viewpoint, as if a system exists to do something, rather than... merely doing it... I would prefer less intention- or design-coded words like "result" or "outcome" or something of that nature.

(It reminds me of Apocalypse World's impulse, i.e. "spawning pool (impulse: to generate badness)" So it is clearly a useful, intuitive, human way of looking at things.)


Wow, these are some excellent descriptions of 'traps' and 'ways out' of them... This is incredible. I also skipped ahead and saw - this is the book that coined all those 'intervention points in a system' that I've read about so long ago! Holy shit! This is that book! What a damn book


Quote from: p137
Rules to govern a system can lead to rule beating--perverse behavior that [obeys the rules, but] distorts the system
Design, or redesign, rules to release creativity not in the direction of beating the rules, but in the direction of achieving the purpose of the rules

I've written up a quote but I think I'll just summarize all of them at the end, and maybe re-summarize where I think they can be lumped together.


Quote from: p140If you define the goal of a society as GNP, that society will do its best to produce GNP. It will not produce welfare, equality, justice, or efficiency unless you define a goal and regularly measure and report the state of welfare, equity, justice, or efficiency.


All the "System Traps . . .
and Opportunities"
from Chapter 5
(pages 111-141)

A system is kept stable by existing forces that respond to discrepancy.
THE TRAP: Trying to change a self-stabilizing system by application of a new force will cause those existing forces to increase. (Actually, this trap is probably what causes such a system to arise in the first place.)
- Don't add new forces; bring in all existing forces and acknowledge their individual needs.
- Reframe all actors' goals in a unifying way; find a new common goal.

If an action is available which directly benefits the actor and indirectly harms everyone (e.g. via the titular 'commons'), everyone will suffer as a result of taking individually intuitive actions.
- Increase everyone's perspective, teach them to inherently value the health of the commons.
- Also "restore or strengthen the missing feedback link" perhaps via "mutual coercion, mutually agreed upon": Create a form of feedback which translates the indirect cost to everyone into a direct cost to the actor.

When the goal is to maintain the current state, and the current state slips negatively, standards and performance will perpetually drop.
- Set goals/standards that do not respond to recent performance (make them 'absolute')
- OR, set goals/standards based on the best recent performance rather than the average or the worst.

This one seems like the sibling of the above... When everyone's goal is to surpass everyone else, a "reinforcing feedback loop" causes efforts to perpetually increase.
THE WAY OUT, ONCE CAUGHT (but try to avoid this!):
- Refuse to compete
- Negotiate a new system with balancing loops to control the escalation [see also THE WAY OUT for POLICY RESISTANCE, above]

"If the winners of a competition are systematically rewarded with the means to win again, a reinforcing feedback loop is created"
- If losers can create a new field for competition, this diversification can help them escape
- Limit the rewards any one actor can win [or, a negative feedback loop to compete against the reinforcing one]
- Offer rewards that do not help their actors win more

An outside force concerned with reducing the symptoms or appearance of a systemic ill may make things worse, especially if that outside force takes over the existing internal maintenance force. The internal system deteriorates, and the outside force has to work harder and take more control. Reinforcing feedback loop.
- Avoid symptom-relieving or signal-denying policies or practices.
- Again, just notice the system

Rules can be exploited, which can lead to a distortion of the goals which called for the rules in the first place. This cannot be solved by enforcing the rules better, rules can always be evaded [see this as another form of POLICY RESISTANCE]
- Rule beating takes effort; rule beaters are doing so because of systemic forces at play; again, see THE WAY OUT of POLICY RESISTANCE
- Because it's so relevant to game design, I will quote this explicitly. "Design, or redesign, rules to release creativity not in the direction of beating the rules, but in the direction of achieving the purpose of the rules." (p.137)

Also on rule beating, "Rule beating becomes a problem only when it leads a system into large distortions, unnatural behaviors that would make no sense at all in the absence of the rules. If it gets out of hand, rule beating can cause systems to produce very damaging behaviour . . . Rule beating that distorts nature, the economy, organizations, and the human spirit can be destructive." (p.136)

". . . design the [rules] with the whole system, including its self-organizing evasive possibilities, in mind." (p.137)

If a system's goals, if "the indicators of satisfaction of the rules" "are defined inaccurately or incompletely," the system will produce the wrong results.
- Goals, and indicators, must relate to the actual, "real welfare of the system."

[See: SCORE in games. A part of this trap: Don't choose indicators based on ease of implementation. Choose indicators based on accuracy and completeness.]


Bad Feedback
- In Tragedy of the Commons, the actor is not receiving feedback representative of their impact on others.
- In Drift, too-weak feedback leads to worse and worse performance

Surface Goals
- In Shifting the Burden, the outside actor has the wrong goal
- Rule Beating alludes to a bad response to rule beating, which is to enforce the rules. Enforcing rules is a bad surface goal.
- This is the core of Seeking the Wrong Goal, of course

Loss of Agency
- In Success to the Successful, a lack of agency leads to further loss of agency; this is bad
- In Shifting the Burden, the one most affected - and most qualified to provide good goals, has lost (or given up) agency

How does Escalation fit in? actors are concerned with competition rather than the bigger picture? Is that Bad Feedback or Surface Goals? It must be its own thing.


Leverage Points ---
Places to Intervene in a System

(pages 145-165)

Yes, I've read this list before, but having read this whole chapter: damn, the full text is good. The list itself is nice as a reminder for this chapter, the standalone list is not nearly as useful. Get that full-fat version, read the damn book

Quote from: p. 162[One] of Jay Forrester's famous systems sayings goes: It doesn't matter how the tax law of a country is written. There is a shared idea in the minds of the society about what a "fair" distribution of the tax load is. Whatever the laws say, by fair means or foul, by complications, cheating, exemptions or deductions, by constant sniping at the rules, actual tax payments will push right up against the accepted idea of "fairness."

Quote from: p. 163Paradigms [shared social agreements about the nature of reality] are the sources of [systems goals and information flows, feedbacks, stocks, flows, and everything else about] systems.


My notes on the places to intervene.

-Changing numbers, buffers (stock sizes relative to flow rates), the plumbing (how stocks and flows connect), and delays in the system are lower-order intervention points. They are often physical and expensive to change one by one, and besides, they don't really get at the heart of the problem. Pay attention to these mechanisms, of course. They matter.

-Balancing and reinforcing feedback loops are powerful and more can be done with them. They are atomic in nature, each one the result of a simple "If X is higher/lower, do Y more/less". If you can change one of these relationships (the right one, in the right direction) it can have a big impact.

-Information flows, rules, and self-organization-- the ability to change the rules and structure. Education and enforcement and agency over the system itself. The previous item suggests that changing numbers directly is expensive and low-impact; instead, change the flows, change the loops. The numbers will work themselves out. The idea with this item, is that you don't have to change the flows or loops if you can change the nature of change itself. How is change effected in this system? If it's in the right hands, if everyone knows what's going on and why it's going on and how to change it (which correspond respectively to information, rules, and self-organization), the system will change for the better of everyone.

-Goals, paradigms, and transcending paradigms. Again, this flows down the list. In order to change a system, we understand its goals, goals come from paradigms, and transcending paradigms allow us to play with paradigms...

Go up another level, another level, another level. Appeal to the biggest arena you have real access to. Make changes there, while considering every subsystem. I would not be as comfortable thinking like this without the backing of everything I learned about local adaptation and living systems from The Nature of Order.



This chapter is great, really lovely. I'm going to be drawing more connections. I need a better way to draw connections than these links and backlinks... The thing is it's connected but not

I guess I just have a lot of books tied together by a huge web. Maybe these are "Systems Thinking" books, but they're not all about systems thinking in the way that this book is, they're just... thinking about things, and everything is a system... Anyway, I'm going to go one by one through all the "systems wisdoms" that this final chapter tries to describe. They are all good.


Quote from: p166People who are raised in the industrial world and who get enthused about systems thinking are likely to make a terrible mistake. They are likely to assume that here, in systems, analysis, in interconnection and complication, in the power of the computer, here at last, is the key to prediction and control. This mistake is likely because the mind-set of the industrial world assumes that there is a key to prediction and control.

~ The Nature of Order, "20th-century mechanistic viewpoint" (A bad specific link, because I can't quiiite find anything specific - it's crystal clear in my mind, but not here. I need some connective tissue.)

EDIT: So I created a new forum for this! I'm excited for it.

~ The harmful industrial mechanistic viewpoint.