Evolving Systems Starting Points: Beckmann

Peter Beckmann
Starting Points re Evolving Systems and Emergence
27 June 2009

(1) The basic idea behind evolution (in the sense that we mean it here) may very well be the greatest single intellectual achievement of humankind. 

(2)  I take the operational view that evolution emerges from fundamental rules appropriate for our universe.  I assume, until a better idea comes along, that these (reductionist) rules govern what happens at the most basic level and they, along with a continuously emerging set of initial conditions at every infinitesimal point in space-time, result in that which we call evolution.  There is no "direction" to evolution as an emergent property.

(3) The study of emergent systems perhaps can be defined as those for which reductionist approaches have not (yet) been possible.  However, communications among thinking people is somewhat hampered by the lack of an generally agreed upon operational understanding of the word "emergence." 

(4) "We are mammalian bipeds wandering the planet trying to make sense of our surroundings."  (I attribute this quote to Paul Grobstein.)  We seek food, shelter, sex, clothing, love, happiness, Guinness, Shostakovich symphonies, etc. 

(5)  Absolutism is not helpful.  With our everyday understanding of words like "apples," "fall," and "up," we can say that the statement "apples fall up" is false, or factually wrong, etc.  But it is not helpful to apply the notions of right and wrong, true and false, etc. to complex (emergent?) models of complex (very many parameters) systems.  More significantly, is it even unhelpful to apply these words to simpler systems that can be modeled from "first principles."  "Better" and "okay, but not as good" are better words to use.  Example: In what sense is a heliocentric model of the planets better than an earth-centered model?  Both satisfy all we know about fundamental laws of physics.  The answer can be either aesthetic (I choose the model that makes me feel better) or mathematical (I choose the model with the smallest number of adjustable parameters).  Neither is wrong.  [As an aside, our great grandchildren are likely to come home from their elementary school class and make fun of us because we use to think that the earth went around the sun when everybody knows that the earth travels in a geodesic (a straight line) in a curved space time.  We must continuously remind ourselves: we have not arrived.]

(6) Reductionism is an immensely powerful "direction" for human beings to take.  Its approach has resulted in virtually all the good stuff we have (e.g., iPhones and insulin).  You have not arrived at a reductionist model until the model can be expressed with mathematics with no recourse to a culturally based human language.  The best example is Newton's mathematical unification of "earthly goings-on" and "celestial goings-on" to produce a new cultural thing (understanding?, idea? phenomenon?): gravity.  Another example is quantum superposition for which nobody has founds words that convey what the mathematics conveys.

(7) As far as I know one cannot (yet) "derive" Duchamp's A Nude Descending a Staircase or Shostakovich's Eleventh Symphony (not to speak of love, cancer, etc.) using reductionist procedures.  I take the view that this is because we are stupid, not because it is in principle not possible. We need emergent approaches for these things.  I take the attitude that an emergent system is one that is not solvable by reductionist approaches because there are too many time-dependent adjustable parameters and we cannot determine initial conditions. 

(8)  We need to do our science from both ends (the reductionist end and the emergence end) and try to connect them in the middle.  The best example I know of the success of this approach is the First and Second Laws of Thermodynamics.  From the late 1600's to the late 1800's (think steam engines and good scotch), thermodynamics was surely one of the very best examples of an emergent set of rules.  (The First Law says that there is a mysterious entity that remains the same and the Second Law says that there is another mysterious entity that tends to increase.)  Boltzmann, Gibbs (the first internationally recognized American theoretical physicist), Maxwell, and others then derived these laws from the fundamental (reductionist) mathematical modeling of classical systems (now called classical statistical mechanics).  For more than 100 years, we measured and controlled temperature and pressure (emergent properties at that time) without a reductionist understanding of what they were.  (They are the average squared speed of the particles and the force exerted by the particles per unit wall area, respectively.)  The final chapter in this saga is Emmy Noether' contributions in the 1920's, but that's another story.  The point is the emergent properties of temperature and pressure (things you measured on thermometers and pressure gauges) were very helpful.  But, eventually, emergent models gave way to reductionist models.

(9)  We need to learn to say "we don't know."  This is a subtle point and not one that evolution encourages.  So, human students have to learn this.  It is not natural.  The idea that we seek the best-working model rather than the "right" model is surely one that is extremely difficult for humans to accept.  Perhaps each of us need three score years on the planet to really understand this.  Creating a supreme being as a stand in for that which is not known seems unnecessary to me though its historical aesthetic appeal is clear.  However, it doesn't help us in creating better models.  Nor does it result in iPhones, insulin, Guinness, etc.

(10) The traditional discipline of philosophy does not help me, personally, at any level with any of the issues discussed here.  And I include the famous folks like the Whiteheads and the Russells (both of whom also did extremely fine reductionist science and mathematics in addition to "philosophy!")  I suspect this is going to get me into trouble.  So be it.

(11) Organizational human structures that work out the best models (or strategies for determining the best models) with a "bottom up" strategy work best.  That is, the resulting models usually (but not always) work well.  Indeed, this is the procedure that evolution "uses."  Organizational structures involving a self-select few selfish people that determine models and strategies for the rest of us using a "top down" approach don't (usually) work.  That is, the resulting models don't (usually) work.  Sadly, it can take centuries and millions of lives before this is realized.  Biological evolution doesn't help here; it makes sense that "me" (and my gene pool) are all that matters.  So, we need social evolution as well.   How about eliminating the adjectives and just having "evolution?"

(12) It is both arrogant and stupid to assume that we have a special place in the universe.  It is extraordinarily limiting in developing better models. 

(13)  All educated caring people need to help in the never-ending quest to eliminate the unhelpful (and, indeed, damaging) Cartesian approach that mind and body are separate.  A lot of serious neuroscientists know this but our children don't.  We have to teach it.  It is an extraordinarily destructive and regressive model.

(14) In America, blacks, women, and many (but not all) minorities are faring better than they did only a couple of generations ago.  (Alas, we need to do so much more on all fronts.)  There is (at least) one more major battle.  We all need to join the battle in making our students understand that there is a continuum of biological sexual orientation between the traditional endpoints of "male" and "female."  Alas,  the traditional pedagogical approach concerning the "biology of sex" is not helpful here.  We need an interdisciplinary emergent approach.  Because human sexual properties are emergent properties, they are difficult for us to properly model.  But culture is an emergent property.  One goal of "culture" is to help us when our more basic modeling does not work. 

(15) To end by going (coming?) full circle, our students need to understand that they have not arrived.  The human individual is not evolution's final product.  We are a "work in progress."  Every course we teach can address this in some manner or other.  Physicists have become a humble lot, of late.  Our two most successful models, quantum mechanics and general relativity, are fundamentally incompatible.  We can't find a reasonable model for what mass is.  We don't know whether we're in the universe or whether ours is one universe in a multiverse.  We have no meaningful model for why the expansion of the universe is accelerating.  What a mess.  Talk about "a work in progress!"