Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: CAUSATION IN PHYSICS



I agree with what John wrote (see below). And with what Bob
wrote (pasted at the end). I am happy that nobody objected to
my two "what is wrong with this?" declarations.

So what are we arguing about? About philosophy of science.
Actually, we are not arguing, we are expressing our thoughts.
And this is very interesting. A reminder that individual laws are
parts of large structures was very appropriate. We will never
know everything about the world.

The only reservation I have is Bob's statement about electrons.
"they are completely human constructions and are deeply mired
in taste, culture, and current modes of thought."

Yes, everything we can possibly say is "a human construction".
But some human construction are more real (detectable by our
laboratory instruments) than others. I am sure Bob also thinks
that electrons are quite real. And that they were real before we
had instruments to detect and mesa them. He does not think
that electrons appeared when we postulated their existence, or
that bacteria appeared when Pasteur invented them to explain
certain diseases. All this is obvious to research physicists.
Otherwise their strategies (in trying to make discoveries)
would be quite different from what they are.
Ludwik Kowalski

John Denker wrote:

a) Is nature governed by laws?

Yes and no. Mostly yes.

I say mostly yes, even though
-- There are parts of physics where God evidently does play dice.
We have no law to predict the outcome of any particular dice-roll.
-- There are parts of physics that we don't understand.
-- I suspect there will always be parts of physics we don't understand.

I say mostly yes, because there is a huge amount of stuff we do
understand. We can construct macroscopic technology that averages out
there dice-rolls and performs all sorts of wonderful predictable actions.

b) Are these laws causal?

That's an interesting question, but the wording is a bit ambiguous.
Let's break it into less-ambiguous sub-questions:

1) If we take all the laws of physics collectively, do they express
causality principles? Answer: Yes.

2) If we take one of the laws of physics individually, is that law
consistent with the overall causality principles? Answer: Yes. It
would be very bad to have one of the laws contradict the others.

3) If we take one of the laws of physics individually, does that
law by itself encode everything we know about causality? Answer:
In most cases, no.

In particular, the law F=ma does not encode _any_ of what we know
about causality.
-- F=ma does not encode any of what we know about the arrow of
time in thermodynamics.
-- F=ma does not encode any of what we know about light-cones in
special relativity.
-- F=ma does not encode any of what we know about neutral Kaon decay.

Summary: F=ma doesn't violate causality, but it doesn't enforce it,
either. It just goes along for the ride.

We have exhibited cases where it is more natural to calculate acceleration
in terms of F/m, and vice versa (F in terms of m*a); no law of mechanics
favors one over the other.

*********************************************
Bob Sciamanda wrote:

1.) The empirically useful content of Physics is in a form which asserts a
numerical equality between two functions of the readings of operationally
defined measuring devices (ie., an "equation", such as eg. PV=RT,
constructed by Boyle et al. to connect the readings of their thermometers,
manometers and meter sticks when "applied" to a rarified gas).

2.) These numerical relations (mathematical models) constitute the
*externally* useful, and experimentally testable content of Physics.

3.) To each of these numerical relations we are wont to attach an
*internally* useful content, ie., a conceptual model (eg., the point
particle ideal gas model). This brings into existence humanly invented
things (eg. atoms, electrons) and relations (eg. causality) among these
things which make their observed behavior "explainable" to us. These
conceptual models are internally satisfying and offer fodder for further,
fruitful speculation; but it must be remembered that they do not stand on
the firm, testable ground of the testable numerical relations; they are
completely human constructions and are deeply mired in taste, culture, and
current modes of thought. Our disputations regarding these conceptual
things and relations should therefore always be moderated with a
judicious respect for the dictum: "De gustibus non est disputandum".

4) In my personal view, "real" causality is in no non-trivial way
applicable to material reality (what could it mean?) . It is in no way
required for the success of our testable Physics; it is however a useful
(perhaps necessary) crutch to human, conceptual models.

5) A bottom line: The accelerations of a pair of binary stars are
numerically related to their relative, spatial positions. Indeed, the
accelerations of all astronomical objects are quantifiably related to
their relative positions. Our conceptual model of this serendipity
invents an attractive "force" as a *cause* of this behavior. That's
how we get our jollies :)