Three Universal Processes

The Three Ps of Adaptive (Evo-Devo) Foresight Practice

Recall the Three Ps foresight model (above), which we introduced in Chapter 1. We will now see why this model is grounded in apparently universal processes of adaptation, and can be used not only for how foresight is produced in biological and social systems but for any complex adaptive system.

The Evo-Devo Foresight model proposes we can identify three basic processes of adaptive change, and thus three basic processes of foresight, that operate in any replicating biological, societal, technological or universal systems:

3. Evolutionary processes that generate novelty and variety in a system over time. These processes are largely unpredictable, except that they add more variety to the environment over time. Graphically, evolutionary processes look like trees, with increasing branches and interactions over time.
   
   

   Evolutionary Tree
   (The Tree of Life

   For examples, think of Darwin’s phylogenetic tree (“tree of life”), the way living systems create new varieties and species of offspring. Physically, evolutionary processes generate branching novelty and variety in ways that are contingent and unpredictable. For the universe, think of quantum mechanics (branching changes based on observation), nonlinear dynamics that may branch into to “chaos”, and high-energy physics (symmetry-breaking in the early universe). For human societies, think of all the varieties of human subcultures, and all the unique new ideas, products, services, and behaviors we create every year. Mathematically, there are many early models for novelty-generating branching processes. Stochastics, combinatorics, and diffusion-limited aggregation are among our current tools and models. Exponential growth, scale-free networks, and fractals are common patterns seen in such processes. See Ball’s Branches (2011) for one account. Modeling the emergence of new ideas, technologies and businesses in the marketplace is very much in its infancy. Rogers’ Diffusion of Innovations (2005), Christensen’s The Innovator’s DNA (2011), Sawyer’s Explaining Creativity: The Science of Human Innovation (2012), and North’s Novelty: A History of the New (2013), are a few good books for foresight professionals who want to get better at understanding, generating, and managing novelty and variety in their practice.

4. Developmental processes that conserve and converge a system over time. These processes create predictable future states. Graphically, developmental processes look like funnels, guiding the system to one particular set of future states.
   
   

   Developmental Funnel
   (Protein Folding)

   For examples, think of the energy landscape of protein folding (picture right) which funnels an astronomical number of possible 3D protein sequences into a few reliable shape-charge assemblies, over and over again in the cell. Think next of how biological development creates genetically-identical twins, which are funneled so similarly to their future states by developmental processes that you cannot tell them apart from across the room. Think also of predictable stages of psychological development. Think of the many predictable ways that economies, societies, and technologies develop. Physically, developmental processes conserve and converge (“funnel”) complex systems toward probabilistically predictable future states. For the universe, think of the laws of classical mechanics (which determine the far future motions of planets), relativity (determining the emergence of black holes), and thermodynamics (determining an irreversible increase in entropy for the entire system). For human societies, think of any general, predictable patterns we see in social, economic, and technical development. Mathematically, processes of development are even less understood than novelty creation. Reaction-diffusion systems exponential decay, power laws, and learning curves are among our useful models. Normal and log-normal distributions (as in Gibrat’s law for the log-normal development rate of organizations, and cities) are among the regular patterns seen in development. Think of the normal distribution of IQ or height in a developing organism. See Developmental Bio: A Very Short Intro (2011), for what we know today about biological development. Social and economic development are large fields with simple models, and technological development is a small corner of science and technology studies. See Wright’s Nonzero (2001), Morris’s Why The West Rules, For Now (2011), and Pinker’s The Better Angels of our Nature (2012) for a few very good examples of predictable global patterns of social, economic, and technological development.

5. Evo-Devo or Adaptive processes that combine evolution and development, and involve competition and cooperation between systems involved in a life cycle (birth, growth, replication, and death), and try to adapt to their environment. Such systems are subject to natural selection, a process regulated by both evolution and development. Graphically, their changes can be drawn as selection peaks and valleys on an adaptive landscape, as in Wright’s evolutionary landscapes. Multiple adaptive peaks representing competing or cooperating systems may merge, split into more peaks, or rise or fall in adaptation, depending on their changing abilities and the selective environment. Physically, the “complex adaptive systems” that engage in evo-devo processes can show branching, funneling, cycling, accelerating, decelerating, self-organized criticality, and other complex behaviors, depending on where they are in their life cycle, and what else is happening in their environment. All such systems are “dissipative structures,” which means they maintain a resilient adaptive state using energy flows, in a far-from-equilibrium condition.
   
   

   Adaptive Landscape
   (Population Genetics)

   For examples, think of any replicating, varying, and interacting systems, including replicating stars and prelife chemistry in the universe, replicating organisms on Earth, ideas that replicate in and between brains (“memes”), and technology applications and algorithms that replicate in economies (“temes”). Mathematically, S-curves, predator-prey, and game theory interactions are some of many patterns we see in such systems. Carroll’s Endless Forms Most Beautiful (2006) and Laubichler’s, From Embryology to Evo-Devo (2009) are good intros to evo-devo biology. To understand how both life and our universe seem to balance the use of both evolution and development to create adaptation, we must today go to systems theorists. Some leading meta-Darwinian models for living and universal systems include Wesson’s Beyond Natural Selection (1991), Salthe’s Development and Evolution (1993), Kauffman’s At Home in the Universe (1996), Denton’s Nature’s Destiny (1998), Smolin’s The Life of the Cosmos (1999), Conway Morris’s Life’s Solution (2005), Corning’s Holistic Darwinism (2005), Ried’s Biological Emergences (2011), McGhee’s Convergent Evolution (2011), and Pross’s What is Life? (2012). For more on how memes and temes compete and cooperate in society, see Wright’s Nonzero (2000), Aunger’s The Electric Meme (2002), Brandenberger’s Co-opetition (1997), and Kelly’s excellent What Technology Wants (2011).

Campbell (2015)

An excellent book that explains why possibility, probability, and preference processes are universal is the EDU-affiliated scholar John Campbell’s Darwin Does Physics (2015). Campbell describes how all complex adaptive systems, from quantum wave functions to brains to science processes to the cosmos itself, can be understood as physical systems that encode models of themselves and their environment, and which undergo constant selection to improve the accuracy and usefulness of those models. Campbell reminds us that Shannon’s information theory is a measure of the ignorance (surprise-potential) of our models. Knowledge, by contrast, involves accurate (surprise-free) prediction of events, and its mathematics is Bayesian inference, models based on conditional probabilities, which are continually updated with new information.

Campbell calls his perspective Universal Darwinism. This is a good description for how the universe works if we focus on the selection (adaptation) aspect of knowledge accumulation, but in my view Universal Evolutionary Development is an even more useful description, as it forces us to think about all three aspects of physical change: evolution, development, and adaptation.

The fundamental processes that influence both change and knowledge accumulation in our universe are the possible physical states of a system (evolutionary options), the probable future states, whether anyone knows them or not (developmental option-reductions), and the (evo-devo) adaptiveness of the model states (preferences) of all complex systems, as they interact with each other in the world.

Understanding the universe doesn’t get any simpler than this, in my view, and I highly recommend Campbell’s book if you are interested in how our physical, biological, and social sciences must continue to converge around issues of modeling, knowledge accumulation, and adaptation in coming years. EDU scholar Michael Price has published a nice short interview with John Campbell, Why Physics Needs Darwin (2017) which you might also enjoy.

For foresight professionals, the key takeaways from the Evo-Devo Foresight model is that our world is made up of a mix of unpredictable, tree-like, and predictable, funnel-like processes that interact to create the adaptive landscapes (preference maps) we observe all around us. The model also offers us some useful graphical aids (trees, funnels, and selection peaks) which we see in the diagram at right.

Chance, Necessity, and Utility (Evo, Devo, and Adaptive Change)

We can understand the world as a privileged set of competing and cooperating peaks (high fitness configurations) and a minefield of threatening valleys (low fitness configurations). As foresight professionals, it is our job to continually track and rebuild useful preference maps, and to help our clients find and steer toward the peaks while avoiding the valleys. Adapting well is a worthy and ceaseless challenge, and is much easier discussed in abstract than done in reality. We wish you well in that work.

Recall that Aristotle (350 BCE) championed the universality of these three perspectives, in his model of human intellect as a mix of the theoretical, or truth-associated, the productive, or beauty-associated, and the practical, or goodness-associated categories of mind. The 95/5 rule lets us reorder Aristotle’s three values as “beauty, truth, and goodness” (evo, devo, evo-devo). Beauty is aways far more plentiful, and much easier to see, than truth. Goodness, in turn, is always a judicious mix of both beauty and truth.

Evolution’s mandate is to unpredictably fan out into perennial new diversity and variety. It produces a natural world of astonishing beauty. Development’s mandate is to predictably funnel all this chaos to a small set of predictable futures, invariant truths, expressed as critical informational or physical structures and functions will emerge, at developmentally appropriate times, in all corners of the universe. Evo-devo’s mandate is to combine these two processes to create local adaptiveness, some of which will turn out to be universally adaptive as well.

This is a great time to point out that evolutionary development is nothing like Aristotle’s scala naturae (Ladder of Nature, Great Chain of Being), model of life, where all the important processes are predestined by a Creator into a strict hierarchy of emergence. In the evo-devo model, a 5% developmental framework for universal complexification is statistically entrained in a special subset of initial parameters and conditions, carefully tuned and selected in previous developmental cycles. Just as in life, that framework limits the evolutionary exploration that is possible, without breaking the developmental cycle. Nevertheless, just as in living organisms, the great majority of initial parameters and conditions, and of emergent change, is still evolutionary. Every so often, some of that evolutionary learning is proven locally adaptive, and some subset is added to the developmental parameters, allowing them to support both more complexity and more adaptiveness. The multi-billion year journey on Earth from replicating bacteria to replicating human beings is the most stunning example of such an irreversible accretion of developmental complexity. We are special, not an accident. Human beings are particularly generally adaptive, among all life on Earth, due to our complex network intelligence, ethics, and empathy. We are also not only far more free (evolutionarily), we are vastly more constrained (developmentally), both on many new levels, than our Archean ancestors.

Nor is an evo-devo universe a Newtonian or Laplacian “clockwork universe” model, which proposes total physical predetermination, though it is a model with some statistically clockwork-like features, including the timing of various hierarchical emergences over the universe’s lifespan and death, just as we see in biological development. Neither the Aristotelian nor Laplacian models of the universe are developmental (positing statistically predetermined emergence and lifecycles) but rather caricatures of it, one-sided models that allow no room or role for evolution.predetermined to emerge. But that framework says nothing about the creative 95% evolutionary painting itself, which is the bulk of the work of art. Recall the all-important differences in tissue microarchitecture and mental processes and life choices between two genetically (developmentally) identical twins. Evolution is as or more important to adaptiveness as development, and evolution is far more of what we see.

In evolutionary development, the Universe is not just a Ladder of Nature (above) or a Random Experiment (standard evolutionary theory) but some useful combination of the two simpler models.

It appears that our universe is significantly more complex, intelligent, resilient, and interesting than any of these models suppose – it is predictable, constrained, and conservative in certain critical parts that are necessary for its function and replication, and it is intrinsically unpredictable and creative in all the rest of its parts. Furthermore, unpredictable evolution and predictable development may be constrained to work together in ways that maximize intelligence and adaptation, both for leading-edge systems, and for the universe as a system.Nor is an evo-devo universe the random, deaf-and-dumb Blind Tinkerer that universal evolutionists like Richard Dawkins (The Blind Watchmaker, 1996) portray. Blind Tinkerer models misunderstand convergent evolution, and are as incomplete in describing universal change as neo-Darwinian theory is in describing biological change today.

If we live in an evo-devo universe, it must be statistically near-impossible for Earth-like civilizations to not invent critical adaptive technologies like language, electricity, internal combustion engines, factories, computers, mobile phones, and soon, human-surpassing AI. But what is entirely within our choice is the path we take to and beyond each of these developmental destinations.

Evo-devo models can be applied to the universe as a system, and to any of its replicating internal complex systems, including stars, molecules, organisms, behaviors, ideas, algorithms, and technologies. In an evo-devo universe, our greatest moral responsibility lies both in foreseeing these predictable destinations (anticipation) and in creating (innovation) and choosing (management) good social, organizational, and personal paths among all the unpredictable futures in front of us.