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



On 11/18/2003 10:01 AM, Rick Tarara wrote:
> I think the view expressed by many here has not been to look at F =
> ma as a cause and effect equation, but rather to contend that
> accelerations themselves are caused by forces (but forces are not
> caused by accelerations).

The second half of that sentence seems to contradict
the first half. But what follows is clearer:

> At least in the Newtonian scheme, all accelerations require a net
> force

Sure.

But "requiring" is not the same as "causing" or "being
caused by". You must never try to infer causation from
statistics. The rooster does not cause the dawn. For
thousands of years "post hoc ergo propter hoc" has been
recognized as a fallacy.

http://www.av8n.com/physics/causation.htm

> and one or more agents that provide the force(s).

I don't know about that. I've never heard about "agents"
until this morning. A force is a force. An acceleration is
an acceleration. I need the laws of physics. I don't need
agents.

> I push the table, it accelerates. Seems like a cause and effect
> relationship?

Yes, there is a cause-and-effect relationship there.
But the relationship has nothing to do with F=ma. It
has to do with the human *choice* to control the force
on the table. It is fairly easy to use muscles to provide
a gentle more-or-less constant force. So it makes sense
to choose to apply a constant force to the table.

Meanwhile, it is easy to come up with other scenarios
where force has not been chosen as the "controlled"
variable ... perhaps due to a change of mechanism or
perhaps only a change of viewpoint.

> In what scenarios (likely to be encountered in intro physics
> instruction) is this not the case?

Fair question.

1) Once I was in charge of programming a Hollywood
motion-control system (basically a robot) where all the
actuators were stepper motors. I had absolute control
over position, and hence velocity and acceleration. I
had no way to predict or even measure how much force
was being developed. It never even occurred to me that
these unknown forces were the "cause" of the known
accelerations. That would have seemed (and still seems)
quite perverse.

2) Ludwik's loop-de-loop is a fine example (likely to
be encountered in intro physics) where it is recommended
to use the known acceleration to calculate the hitherto-
unknown force.