What Science Needs: Interpretation Issues

I made the claim that science needs “at least three things” to be fixed before it can function safely as a religion (or as a replacement for religion, for those who are seriously committed to the myth of mythlessness). The first of these three things is the moral compass. The second involves a problem that essentially requires – is there any way to say this that doesn’t sound confusing? – to recognize the data loss that occurs when we translate non-discrete information into discrete data sets. To try to put that into ordinary language: Science can only deal with things that are discrete. Facts are either true or false. But the world doesn’t come that way. Facts have to be “rounded up” or “rounded down”.

  • A fact that we might view as having a 40% probability of being true is treated as the same as a fact that has 2% probability of being true.
  • A fact that we might view as having a 65% probability of being true is treated as having the same truth value as a fact that has 99.999% probability of being true.

The problem with this is that we aren’t recognizing the implications: we aren’t demanding that the results include relevant information on just how probable all the facts and assumptions really are. In short: science is leading us to false conclusion through what you might call rounding errors.