Randomness isn't a valid explanation for modern life

     One often sees attempts to explain the modern world by undirected processes, such as randomness,  the wisdom of the crowds, etc.  Events and trends are not explained in terms of the aims of human beings, but rather in terms of many small incidents that add up or interact in some way.  Or inevitable structure that results from randomness.  Of course, just like polls and surveys all these explanations are meant to fool people and prevent them from noticing what is really going on.  

    That is not to say that these kinds of explanations have no validity whatsoever.  They do and they can be very useful within the proper domain.  But in practice what happens is that non-purposive explanations are used to explain everything, even what is actually caused by planning and purpose.  

    One example is with lotteries.  Some people win the lottery twice.  It's a rare event, but less rare than one might think at first.  The reason is that the chance of any specific person winning the lottery twice is astronomical, but if we consider all the lottery winners who still buy tickets, the chance of any one of them winning is greater, in particular when people buy multiple tickets.  

    One name for this kind of phenomenon is the Law of truly large numbers.  The encyclopedia article states in the intro:

    "The law of truly large numbers, attributed to Persi Diaconis and Frederick Mosteller, states that with a sample size large enough, any outrageous thing is likely to happen.  Because we never find it notable when likely events occur, we highlight unlikely events and notice them more.  The law seeks to debunk one element of supposed supernatural phenomenology.

    Notice that the introduction starts with a somewhat valid point, but then immediately jumps to an unjustified conclusion about supernatural events.  Forget the supernatural: lotteries aren't even a good analogy for unusual events in normal human life.   

    All possible outcomes of a lottery are proscribed in advance.  Even though the number of possible tickets is enormous, it is known ahead of time what outcomes can occur.  Furthermore, drawings occur regularly and there is no limit on the number of tickets that can be sold.  In addition, we can calculate precisely or at least get a good estimate of the possible outcomes.  

    This is far from ordinary life, where the unusual is often unexpected and where we have no good estimate, much less a precise calculation of the probability of any event.  Furthermore, we don't know how frequently an attempt to make an unusual event occur will happen (analogous to a lottery drawing).  

    Another example is the mathematical discipline of Ramsey theory, which studies under what circumstances certain structures inevitably occur.  The standard problem asks what is the smallest group of people necessary to ensure that at least one of the following two situations must occur: 

   1. At least three members of the group know each other 

    2. At least three members of the group do not know each other

    The answer turns out to be six.  In a group of six or more people, at least one of those situations will occur, regarless of how many fellow group members any particular person is acquainted with.  

    In every popular article on the subject, it seems to be obligatory to add a sentence like: "Ramsey theory shows that true disorder is impossible."  The implication being that if you see something strange, it's just an inevitable structure that must have occurred.  The problem is that, as with the lottery, we are taking an extremely restricted situation and applying it where it no longer applies.  

    In the example about the group of six people, the only thing we considered was whether two members knew each other or not.  There are only two possibilities.  By contrast, in a group in the real world, there are many more relationships and interactions that can take place.  So, while it is true that provided we do not have some undifferentiated mass, there will be inevitable structures, we have no way of knowing what they are or how to find them.  

    Taking these kinds of explanations too seriously is something of a clever silly behavior.  The best explanation for human behavior is the common sense explanation, that people do things because they want to.  

    But, explaining human life in terms of undirected processes has been very effective in preventing people from attempting to find the true explanations of what occurs.  The scientific or mathematical nature of the explanations dazzles some people and allows some to feel superior by believing an unintuitive explanation.  And for the scientists or mathematicians themselves, constantly refining a model that only works in special circumstances provides a powerful distraction.  In particular when the people who choose which explanations to promote and popularize do not care about the explanations themselves.  All they care about is using them to manipulate people.  

    In addition to the arguments above, there is another common sense reason not to put too much stock in undirected explanations: randomness and coordination don't look the same.  When the same thing happens in a sufficiently widespread area, or the same events happen time after time after time, then there is no reason to believe they are random.  Furthermore, when events are uncoordinated, even when they are largely similar, there are always small, though still significant differences.  

    Even when there is both coordination and independence, as frequently happens, there is a difference between an organic situation and when someone puts their thumb on the scales.  One way this happens is by setting up a situation where people can choose anything, but only within a predetermined range of choices.  Or when individuals are constantly steered in subtle ways to make certain choices or to avoid others.  And since this effect is often much more powerful than the individual choices, it is the most important factor to consider.

    And so, while non-purposive explanations are useful in certain situations, there is no reason to apply them to the world as a whole.

2 comments:

  1. Very good!

    It's a strange business that this non random world is (without thought) modelled by the invented abstraction of randomness - invented to calculate the odds in gambling card games, apparently.

    ReplyDelete
    Replies
    1. Thanks

      Yes, it's quite a leap and made worse by the fact that (as you have pointed out), many people no longer consider purposive explanations. They believe that only randomn or mechanistic explanations can explain the world.

      Delete

The real AI agenda

    On a post  by Wm Briggs, about artificial intelligence, a commenter with the monniker "ItsAllBullshit" writes:           "...