... seeking simple answers to complex problems, and in the process, disrupting the status quo in technology, art and neuroscience.

Thursday, March 27, 2014

Standards of Truth




If absolutes are an illusion of the human mind, how do we gauge relative truth?  When does a generalization become useful?

The American justice system takes this illusion into account and presents a good example of something useful with its two standards of proof. "More likely than not" is required to indict (bring to the court).  "Beyond a reasonable doubt" is required to convict (and proceed with sentence and execution).  Both evaluate evidence but ultimately require subjective conviction.  If all proof was truly objective, judges and juries wouldn't be needed.

"More likely than not" simply means anything with a probability of greater than 50 percent, even by just a little.  This reflects nature's effort to exceed the average. The test covers half the spectrum of probability and also reflects how the intuitive mind reaches conclusions.  Evidence is presented, but "feelings" one way or the other are used to describe our ultimate conclusion.

"Beyond a reasonable doubt" is a far stronger test. It means almost certain and requires substantial evidence, logic and nothing contrary that raises "reasonable doubt". Still, mistakes are made and innocent people are put to death.  Most convicted ARE guilty, but there are exceptions.

These are the same tests we unconsciously use to filter our experience.  If something makes sense, we find it more likely than not to be true.  We put it in the probable category and wait for more evidence.  Once we get multiple independent verification, we move beyond a reasonable doubt to conviction.  It's conviction that drives our action. But even after conviction we sometimes discover error. Does it make us more careful? Usually not. Generally we rationalize our action as based on bad data.  We blame someone else and move on.

Still, these two standards of truth are powerful tools and the best we have in critical thinking.  Keep them handy.  Weight the evidence.  Be ready to be wrong.