At Home in the Universe by Stuart Kauffman

Summary
  1. Natural selection is important, but it has not labored alone to craft the fine architectures of the biosphere. Self-organization is the root source of order and is not merely tinkered, but arises naturally and spontaneously because of the principles of self-organization. Self-organization works together with natural selection to help shape and drive evolution in species
Key Takeaways
  1. Science has taken away our paradise – purpose and values are ours alone to make – job today is to reinvent the sacred and Kauffman believes that complexity may contain the answer
  2. Complexity suggests that not all order is accidental and is responsible for much of the spontaneous order seen throughout the world
    1. May lie at the heart of the origin of life and leads to order found in organisms today
    2. Life, therefore, is to be expected and is not an accident if it arises out of fundamental self-organizing principles
    3. Spontaneous order and natural selection have always worked together
  3. Second law of thermodynamics – order tends to disappear in equilibrium systems
  4. Best models explain and predict but failure to predict does not equal failure to understand or explain, especially with chaotic systems. Can find deep theories without knowing every detail (don’t have to know every detail of ontogeny (development of an adult organism) but we can understand it – spontaneous order which then natural selection goes on to mold)
  5. For most systems, equilibrium = death
  6. Order for free – order arises spontaneously and naturally and leads to self-organized systems and emergent properties
    1. Life would then be able to emerge full-grown from a primordial soup and would not need to be built one component at a time – life emerges whole and not piece meal
    2. Life is a natural property of complex chemical systems and that when the number of different kinds of molecules in a chemical soup pass a certain threshold, a self-sustaining network of reactions – an autocatalytic metabolism – will suddenly appear
      1. Life did come from non-life – reduces biology to physics and chemistry
      2. Must pass the subcritical / supracritical threshold
    3. Life exists in between order and chaos – in a kind of phase transition where it is best able to coordinate complex activities and evolve
    4. The very nature of coevolution is to attain this edge of chaos, a self-organized criticality, a web of compromises where each species prospers as well as possible but where none can be sure if its best next step will set off a trickle or a landslide
      1. This world does not lend itself to long-term prediction, we cannot know the true consequences of our own best actions. All we players can do is be locally wise, not globally wise
  7. All living things seem to have a minimal complexity, below which it is impossible to go
    1. Matter must reach a threshold of complexity in order to spring to life – this is inherent to the very nature of life
  8. Living organisms are autocatalytic systems – living organisms began as a system of chemicals that had the capacity to catalyze its own self-maintaining and self-reproducing metabolism once a sufficiently diverse mix of molecules accumulates. Once this threshold is reached, a vast web of catalyzed reactions will crystallize. Such a web, it turns out, is almost certainly autocatalytic – almost certainly self-sustaining, alive. Life emerges as a phase transition once the subcritical threshold of reactions to chemicals is breached
    1. The spontaneous emergence of self-sustaining webs is so natural and robust that it is even deeper than the specific chemistry that happens to exist on earth; it is rooted in mathematics itself
    2. There is an inevitable relationship among spontaneous order, robustness, redundancy, gradualism, and correlated landscapes. Systems with redundancy have the property that many mutations cause no or only slight modifications in behavior. Redundancy yields gradualism. But another name for redundancy is robustness. Robust properties are ones that are insensitive to many detailed alterations. Robustness is precisely what allows such systems to be molded by gradual accumulation of variations – the stable structures and behaviors are ones that can be molded
  9. Homeostasis, the ability to survive small perturbations, required for life to survive
  10. Small-world, sparsely connected networks are extremely efficient at connecting agents and trend toward internal order
  11. Complexity – orderly enough to ensure stability but flexible enough to adapt and exhibit surprises – evolution takes life to the edge of chaos
    1. Organisms evolve to the subcritical-supracritical boundary which exhibit a power law distribution of events
  12. Be smart by being dumb – have a huge sample set and choose what serves your purpose (don’t be ideological, go with promising evidence over beautiful theory)
  13. Immune system is a universal tool box – ability to produce 100m + antibodies allows you to recognize and respond to any threat
  14. Cambrian pattern of evolution – It is a general principle that innovations are followed by rapid, dramatic improvements in a variety of very different directions followed by successive improvements that are less and less dramatic.
    1. Learning curve – After each improvement, the number of directions for further improvement falls by a constant fraction – an exponential slowing of improvement (applies to technology, evolution, business, mastering skills, any improvement!)
      1. The more complex the system, the more difficult it is to make and accumulate useful drastic changes through natural selection
      2. Correlation length – taking massive jumps can lead to fitter mutations if land at a fitter peak – explore and try vastly different areas to possibly get outsized rewards (deep fluency in many fields and iterate constantly with small bets and pursue promising areas – parallel traced scan)
        • When fitness is average, the fittest variants will be found far away but as fitness improves, the fittest variants will be found closer and closer to the current position. Expect to find dramatically different variants emerging during early stages of an adaptive process but later the fitter variants that emerge should be ever less different
        • When fitness is low, there are may directions uphill. As fitness improves, the number of directions uphill dwindles. Thus we expect the branching process to be bushy initially, branching widely at its base, and then branching less and less profusely as fitness increases
  1. Optimal solutions to one part of the overall design problem conflict with optimal solutions to other parts of the overall design. Then we must find compromise solutions to the joint problem that meet the conflicting restraints of the different subproblems
  2. Coevolution itself evolves over time as fitness landscape changes – maybe towards Red Queen or Evolutionary Stable Strategy
    1. Evolution pushes towards edge of chaos, towards phase transitions
    2. Highest fitness occurs right between chaos and order
  3. Mill-mistake – mistaking the familiar for the optimal
  4. A central directing agent is not necessary to life, life results as an emergent property
  5. The tools we make help us make tools that in turn afford us new ways to make tools we began with
  6. Technological revolution is coevolution – niche creation and combinatorial organization
    1. Diversity begets diversity and growth but must first cross the supracritical threshold to hit the autocatalytic phase transition
      1. Diversity (resources, goods, trade, skills, etc.) great predictor of economic growth
  7. Patch Procedure
    1. Take a hard, conflict-laden task in which many parts interact and divide it into a quilt of nonoverlapping patches. Try to optimize within each patch. As this occurs, the couplings between part in two patches across patch boundaries will mean that finding a “good” solution in on patch will change the problem to be solved by the parts in adjacent patches… – models coevolving ecosystems
      1. If a problem is complex and full of conflicting constraints, break it into patches and let each patch try to optimize such that all patches coevolve with one another
    2. May not give us the solution to the real problem but may teach us how to learn about the real problem, how to break it into quilt patches that coevolve to find excellent solutions
    3. Ignoring certain subsets of restraints may be helpful at times – should not please all of the people all of the time but you should pay attention to everyone some of the time
What I got out of it
  1. Spontaneous self-organization is a deep, fundamental principles of math, physics, life. Order for free, patch procedure, learning curves and the Cambrian diversity principle, subcritical and supracritical threshold breach is the same thing as phase transition, all we can do is be locally wise and not globally wise since our system is too complex to predict