Re: Surviving Einstein.

From: John E. Hadstate (nospam_at_null.nil)
Date: 06/29/03 

Date: Sun, 29 Jun 2003 10:13:02 -0400

"Douglas A. Gwyn" <DAGwyn@null.net> wrote in message
news:3EFE4279.6080700@null.net...
> John E. Hadstate wrote:
> > Hidden variables have been ruled out by whom? By people who can't
figure
> > out what the hidden variables might be? That is circular reasoning.
>
> Not at all. If there are hidden variables they would have
> certain consequences. Experiments have been performed to
> test which way it goes.
>

After thinking about this for a while, I realized that 'hidden variables'
are not necessarily implied by the inability of a model to fully predict the
behavior of a process.

Consider a process that is characterized by a decaying exponential y =
exp(-at). (I'm talking about the real exp() function, not some digital
approximation of it.) Let's go a little futher and claim that we know why
the process should be characterized this way.

We can't fully evaluate the exp() function by any means at our disposal.
That is to say, we have no mathematics that precisely computes the value of
the exp() function for every possible input. Consequently, all we can do it
model it to some arbitrary precision with arithmetic that we can do.

At any point in 't', our model will exhibit random deviations from our
process, but to us it will look like the process is random compared to our
model.

After enought 't' passes, our process measurement will be indistinguishable
from the noise generated in our measurement device (which we also can't
model) and we will no longer know or care whether our model conforms to
reality.

> Note that the kind of probability involved in fundamental
> quantum phenomena is not the same kind of probability that
> is involved in incomplete description of classical
> deterministic systems. The easiest way to see that is to
> consider the famous two-slit experiment, wherein providing
> additional paths to the goal actually reduces the chance
> of certain formerly likely outcomes to zero. That simply
> contradicts classical probability theory. Whatever is
> behind the phenomena, it is not simply lack of knowledge
> of additional parameters.
>

Your last sentence beginning with "Whatever is..." really distills the
essence of what I claimed, "Show me a truly random process and I'll show you
a process we don't fully understand." If we did fully understand it, it's
behavior wouldn't appear to be random. I'm not claiming to know why we
don't fully understand it.

If 'hidden variables' or 'inadequate parameterization' don't explain the
deviations from predicted behavior, well, that's just a fact of life. It's
also an indication that the whole system on which the model is built is
deficient.

We don't know what is behind a random phenomenon and pretending that
randomness is explained by randomness is folly. Randomness is merely an
observed deviation from prediction; by itself, it doesn't explain anything.