Complexity and systems thinking

When explaining how to get to a destination, it doesn’t necessarily mean that more information is better, right?

For example, “Pass the store with the blue sign 300 meters ahead, turn right at the building with the red roof just past it, and take the road next to the white vending machine” is a very detailed explanation. On the other hand, “Turn left at the second traffic light, then go straight and turn right at the next intersection” contains less information.

However, it relies on fewer variables, and the information you need to focus on is simpler.

So, doesn’t providing information that strays from the essence risk distorting our understanding of that essence?

The Red Queen hypothesis is a hypothesis in evolutionary biology proposed in 1973, that species must constantly adapt, evolve, and proliferate in order to survive while pitted against ever-evolving opposing species

Basically saying that every species is constantly evolving towards a local optimum (it's niche), but that since that niche is made up of other species who are also evolving, the local optimum is always moving, so it becomes kind of like the three-body problem.

cross posting here because there was some interesting discussion in some of the comment threads.

"Inevitably under conventional positivist frameworks, the majority of the most pressing issues are ignored: non-linear dynamics, multi-stable states, multi-scale behavior, slow/fast variable dynamics on the systems side and adaptive change, surprise and inherent unpredictability on the policy side. As the Aussies say, the “science” in traditional ecology is typically quadratescience small scale, short term. For it is the only way to pretend that one can be certain. In essence the primary root to success under conventional science is to define a trivial question, use a replicated experimental protocol, and avoid type one error. I saw this vividly among the 60 editors of Conservation Ecology (a journal begun in 1997 that has come to be called Ecology and Society). Those who function at scales below a few meters or a few decades (from population genetics to quadrate ecology, population ecology and community ecology) tended to react to papers by wanting precise answers rather than interesting questions. They searched for what is wrong in a paper, rather than what might be relevant and novel. Their ignorance of the broad scientific literature, or even other areas of ecology and environmental science, was profound. Above that scale, the editors were just about the opposite. Their backgrounds were multidisciplinary with strong roots in one discipline, interests in both theory and practice that extended across large scale and cross scale systems.

CS Holling, quoted in:

Curtin, C. G., & Allen, T. F. H. (Eds.). (2018). Aggregation in Complex Systems. In Complex Ecology: Foundational Perspectives on Dynamic Approaches to Ecology and Conservation (pp. 10–147). Cambridge University Press. https://doi.org/10.1017/9781108235754.003

Holling was an ecologist and founder of ecological economics.

I started a list with all the books I know focused on complexity. Others can add to it.

Opening session of Waves Forum for Changemakers 2024 in Helsinki, Finland.

In this fireside chat with Nora Bateson, International Bateson Institute, and Dave Snowden, Cynefin Company, hosted by Sara Lindeman, Leapfrog, we explore what changemakers can learn from complexity science to better understand change in complex social systems.

I found this chat packed full of useful wisdom, and way too short.

The second talk is also pretty interesting, I fell off on the third one.

Conway's game of life (like many complex automatons) is Turing complete, which means that you can program in it, which means that you can program itself in itself.

This is the basis for Greg Egan's 1995 scifi Permutation City, which is a fun read.

If you want to understand how this works in depth, there's an explainer on youtube too: https://www.youtube.com/watch?v=Kk2MH9O4pXY

I thought this was a really good high level introduction to how to approach modelling complex systems. The "systems reductionism" idea seems particularly useful to me.

A new proposal by an interdisciplinary team of researchers .. [proposes] nothing less than a new law of nature, according to which the complexity of entities in the universe increases over time with an inexorability comparable to the second law of thermodynamics — the law that dictates an inevitable rise in entropy, a measure of disorder. If they’re right, complex and intelligent life should be widespread.

Complexity economics is a heterodox alternative to neoclassical and mainstream economics, that starts with an assumption of complex system dynamics, and attempts to model economic system behaviour from that.

From the page:

Complexity economics is the application of complexity science to the problems of economics. It relaxes several common assumptions in economics, including general equilibrium theory. While it does not reject the existence of an equilibrium, it features a non-equilibrium approach and sees such equilibria as a special case and as an emergent property resulting from complex interactions between economic agents. The complexity science approach has also been applied to computational economics.

Good simple introduction to an early computer simulation showing complex emergent behaviour from simple rules.

This paper is an excellent overview of the history of systems thinking from mid 20th century onward. It covers 4 waves of systems thinking, and DSRP.

The four waves, roughly summarised are:

• First wave: systems as real-world things - system dynamics, physical science focus
• Second wave: systems as mental models - this wave was more social-science driven
• Third wave: critical systems thinking - bringing critical theory to systems thinking, focused on critique of boundary definitions, and on power in relationships
• Fourth wave: universalisation - trying to pull all the threads together from diverse approaches (this wave is starting, may not happen)

The authors suggest that DSRP is a potential core framework for the fourth wave. DSRP summarised:

• 4 structures, each made of 2 elements, mutually defining
  • A Distinction (D) is defined as identity (i) co-implying an other (o)
  • A System (S) is defined as part (p) co-implying a whole (w)
  • A Relationship (R) is defined as action (a) co-implying a reaction (r)
  • A Perspective (P) is defined as point (ṗ) co-implying a view (v)
  • My understanding is that the first two structures both define the boundaries of the system as a whole, and of the individual components of a system (e.g. actors).
  • DSRP kind of assumes that all system components are also entire systems within themselves (e.g. humans are actors in a social system, but a human is also a system of cells, etc.)
• M = I ⊗ T
  • a mental model (M) is the complex product (⊗) of information (I) and DSRP simple structural rules of thinking (T) (listed above)
• Four levels of depth of abstraction:
  • Atomic Structures: The universal components that can't be broken down further.
    • These are the 4x2 component structures of DSRP, listed above.
  • Molecular structures, or “jigs” or “moves”: Abstractions of mental models.
    • “Jig” as in a work-tool that helps you build something. “Move” as in martial arts or dancing - patterns that repeatedly appear.
    • These are thinking-patterns that are basically information-less models. Often these can be discovered by abstracting from the next level (conceptual models).
    • Examples: 2x2 table of considerations - this appears in lots of types of thinking.
    • There’s a big list at https://help.cabreraresearch.org/moves-glossary
  • Compound structures: Conceptual models that include molecular structures AND information.
    • e.g. any applied model. Traffic optimisation models. SWOT analysis (this is a specific example of a 2x2 table)
  • Any person, any system:
    • They claim this represents all knowledge, but I think it only represents all thinking (maybe).

This paper is an excellent overview of the history of systems thinking from mid 20th century onward. It covers 4 waves of systems thinking, and DSRP.

The four waves, roughly summarised are:

• First wave: systems as real-world things - system dynamics, physical science focus
• Second wave: systems as mental models - this wave was more social-science driven
• Third wave: critical systems thinking - bringing critical theory to systems thinking, focused on critique of boundary definitions, and on power in relationships
• Fourth wave: universalisation - trying to pull all the threads together from diverse approaches (this wave is starting, may not happen)

The authors suggest that DSRP is a potential core framework for the fourth wave. DSRP summarised:

• 4 structures, each made of 2 elements, mutually defining
  • A Distinction (D) is defined as identity (i) co-implying an other (o)
  • A System (S) is defined as part (p) co-implying a whole (w)
  • A Relationship (R) is defined as action (a) co-implying a reaction (r)
  • A Perspective (P) is defined as point (ṗ) co-implying a view (v)
  • My understanding is that the first two structures both define the boundaries of the system as a whole, and of the individual components of a system (e.g. actors).
  • DSRP kind of assumes that all system components are also entire systems within themselves (e.g. humans are actors in a social system, but a human is also a system of cells, etc.)
• M = I ⊗ T
  • a mental model (M) is the complex product (⊗) of information (I) and DSRP simple structural rules of thinking (T) (listed above)
• Four levels of depth of abstraction:
  • Atomic Structures: The universal components that can't be broken down further.
    • These are the 4x2 component structures of DSRP, listed above.
  • Molecular structures, or “jigs” or “moves”: Abstractions of mental models.
    • “Jig” as in a work-tool that helps you build something. “Move” as in martial arts or dancing - patterns that repeatedly appear.
    • These are thinking-patterns that are basically information-less models. Often these can be discovered by abstracting from the next level (conceptual models).
    • Examples: 2x2 table of considerations - this appears in lots of types of thinking.
    • There’s a big list at https://help.cabreraresearch.org/moves-glossary
  • Compound structures: Conceptual models that include molecular structures AND information.
    • e.g. any applied model. Traffic optimisation models. SWOT analysis (this is a specific example of a 2x2 table)
  • Any person, any system:
    • They claim this represents all knowledge, but I think it only represents all thinking (maybe).

This is an excellent introduction to system dynamics style system thinking.