Complex Systems And The Hubris Is Easy. Also Profitable.

Another reason the elites are so fundamentally wrong so often, and of the blindness that leads to their Hubris to the Power of Dunning-Kruger, are the properties of complex systems* with their intrinsic network effects and the ease of ‘explaining’ them after the fact, of gaming systems in them.

Systems are everything that isn’t obviously a machine.  Complex systems have subsystems, subsystems may have sub-subsystems, …

Network effects are a chain of this-affects-that that can ultimately change everything in a network.  Chaos theory‘s ‘butterfly effect’ is one manifestation, the idea that complex systems are internally connected via non-linear feedback such that tiny changes can produce large effects, e.g. a butterfly flapping its wings in Asia ‘causing’ a hurricane in the Caribbean some weeks later.

‘Causing’ because tracing causal chains through a network isn’t normally possible with any precision, nearly every effect has many possible causes.  Tracing forward from cause to effect is possible, but predictability drops off quickly as you step through the chain.  That is the source of side-effects.  It is easy to predict a business boom from low interest rates, but not particular bubbles nor social effects such growth of oligarchy, centralization of political power and loss of the middle class.

Tracing backward from effect to cause is NOT possible, because you can’t know all of the possible causes of an effect.  A wonderful recent example is the cause of the west’s crime wave.  That is a plausible causal relation, but very likely not the last such link that ‘explains’ crime rates.

Reasoning using forward chains depends on valid Cause and Effect relationships, which are very difficult to establish in complex systems.  Most of the understanding we have about those is in the form of correlation, X has been associated with Y to some degree in the past.  That is not useless, we run our lives on that kind and quality of information.

But C&E data is much more powerful, and also difficult to establish.  C&E relationships can only be established by experiments where one variable is changed at a time and that is compared to the condition of no changes. That process is social and can easily be biased by social and economic incentives.   All sciences have outright fraud.   Even without fraud, the best-studied error-rate and -type is medical research, in which 25% to 80% of research papers in peer-reviewed journals cannot be replicated and one replication does not ensure another.

Medicine has 30 or so allied medical sciences that provide cause-and-effect for understanding the open, evolving, complex system that is human physiology, cross-checks of hypotheses from many points of view.  That is a huge advantage relative to softer science.  Nevertheless 15% of people who die in major teaching hospitals do so from undiagnosed causes. Psychology, sociology, cultural anthropology, political science, economics, business, … all of the softer sciences must be worse, as they have few or no such supports.

This has very direct consequences for any area that deals with complex systems, e.g. society, the economy, government, politics, law, … In complex systems, it is extremely easy to go wrong, and extremely easy to support the claim “I was right, my plan and implementation works” with chosen questions answered with selected evidence favorably interpreted.  Human minds do that selection, filtering and interpreting in their own interests, very well, often unconsciously.  Bureaucracies are brilliant at that, search for “<your favorite social program> is a great success” for confirming instances.

The constant in systems involving people is people.  The range of human mental limits, motives, impulsiveness, and all other characteristics do not appear to have changed much from Roman times.  Our political class is very good at exploiting its’ population’s mental and emotional limitations.  Systems are hard to understand for anyone, so a politically-persuasive narrative and claims of understanding are easy to agree with, however preposterous they are when considered in the light of history and evidence such as the above.

Everywhere you look is evidence that there are deep reasons we humans do things in the way we do, an example I just happened across is the way cities are organized.  Self-organized, actually.  City councils are good at protecting the interests of people they know and those who come to city council meetings or contribute to their campaigns.  But Houston has no zoning, seems to work well, so the case for city councils protecting constituents is poor.

Or the way well-run businesses** manage themselves.  That turns out to be what works in any group or small community, a clear view of the opportunities and dangers (vision), best ways to handle them, and build the team.  Management as a service for the group with a lot of bottom-up information flow, rather than the standard elitist “do what I say, you aren’t paid to think”.  Who could have guessed?

The situation is the same in all complex systems involving people : they self-organize and self-manage, when allowed.  The best solution for our human system’s management and evolution has been free and open discussion from as many points of view as possible and allowing people to go their own way in mutual respect so long as they do the same. That does not produce a perfect system, even if you can define a perfect system.  It does not provide a system that is optimal in any area, that is not possible because the real world is constraints, trade-offs are built into it.   But it is an implementation-by-humans of a well-tested algorithm for finding solutions in a multi-dimensional system such as a village with cooperating and competing agents doing agriculture, hunting, local forest, number of people, average weather through the year, efficiency of plows, …  Such algorithms implemented on computers are used very widely in running factories and processing facilities.

If we are to do well for ourselves, our families and communities, we need to think very clearly and critically, beginning with the quality of the evidence and the probability we are misleading ourselves.  If we are to remove the Status Quo’s command of our lives, we need to challenge their claims with better evidence and understandings.

“Pass a law, make the world a better place”, the claim of rulers everywhere, has brought us to this dismal state of system failure.  We must not repeat that mistake if we are to recover and build a better, more stable, civilization.

To reverse their control, we must begin by repealing laws, many, many laws.

Few laws, but many understandings based on shared values in a community are the natural Conservative position.

*Systems are one description of elements of the complex reality we live within.  High-Dimensioned Games are another, and whichever it was convenient to adopt, the reasoning would be very similar and would reach the same result.

**There are bad things to say about Google, but they are serious about thinking clearly and using the best evidence to do so.  They study what works and adopt it.  Also, this is not the first time those points had been re-discovered, they were old when In Search of Excellence was published, and Drucker put them into the best companies, e.g. IBM and HP before that.

Advertisements

2 thoughts on “Complex Systems And The Hubris Is Easy. Also Profitable.

  1. Pingback: thinkpatriot

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s