Re: Kevin later wrote:

Kevin O'Brien (
Sun, 7 Mar 1999 21:31:49 -0700

>At 05:28 AM 2/24/99 -0700, Kevin wrote:
>>>>>What you wrote, "But there has never, to my knowledge, been a case
>>>>>when a
>>>>>physical law was found to be false by new evidence," is still an
>>>>And yet you still cannot or will not give even one example to prove me
>>>>wrong, or explain what physical law Einstein or any other scientist
>>>>wrong. Making bald assertions you either cannot or will not defend is
>>>>the height -- or should I say depth -- of absurdity.
>>>OK, how about Descartes' law of refraction?
>>I suppose that means that when Descartes' law was shown to be incorrect,
>>refraction as a physical phenomenon was also shown to be false.
>>Yeah, right, and I have seafront property for sale in Colorado.
>>Let's not confuse the physical phenomenon itself with our abstract
>>mathematical description of it, even though we tend to call both a
>>law". When Burgy and I talk about falsifying a physical law with new
>>evidence, we mean demonstrating that the physical phenomenon itself, not
>>just our abstract mathematical description of it, is false. Refraction is
>>still recognized as a legitimate physical phenomenon; what's more, it is
>>also recognized that it is caused by the decrease in the speed of light as
>>that light passes through dense material, such that as a general rule of
>>thumb the greater the density, the greater the angle of refraction.
>>Descartes' attempt to describe this phenomenon may have been shown to be
>>incorrect, but the phenomenon itself is still very much real.
>But what you said was physical law, not physical phenomena. Even
>if Newton's law had been overturned by Einstein (I agree with you
>that it wasn't) it would have no bearing on the physical phenomena
>of gravity. The absence of any mathematical description (law)
>for how gravity operates would have no bearing on the physical
>phenomena of gravity. As Stan so eloquently pointed out, the
>previous discussion and arguments make no sense if it is indeed
>physical phenomena (as opposed to our mathematical descriptions
>of them) that is the issue.

I'm sorry fellows, but I'm rather surprised that two scientists don't
understand this issue better. Perhaps I had better start with some basics.

First, a caveat: there simply are no terms available that will make this
discussion easier, so I ask that you bear with me and not nit-pick over
terms. It is the concepts behind the terms that are important, not the
terms themselves.

Alright, before any science can be done, first there has to be either an
observable phenomenon or a deducible phenomenon. Observable phenomena are
readily apparent; examples would include Boyle's and Charles' gas laws, as
well as refraction. Deducible phenomena are not readily apparent, but can
be deduced from the observation of apparent phenomena; examples would
include all the phenomena described by Newton's laws of motion. Most
phenomena are fairly limited, but some seem to be more general, in that they
seem to occur under a wide variety of circumstances, but always associated
with specific events. For example, the phenomena behind Boyle's and
Charles' gas laws only appear when observing ideal gases, but they occur
under a wide variety of experimental conditions. Refraction is a property
of light (i.e., electromagnetic radiation) only, but it occurs whenever and
wherever light passes from one medium to another. These general phenomena
can be called universal, since (within certain limitations) you would expect
them to be observable or deducible anywhere in the universe under any set of
local conditions. (Indeed, though limited to the non-quantum macroscopic
world, the phenomena behind Newton's laws of motion can be readily deduced
under any set conditions, even relativistic conditions. Though limited to
ideal gases, the phenomena behind the gas laws can be readily observed for
any ideal gas regardless of chemical composition or molecular weight at any
temperature, pressure or volume.)

These universal phenomena are often called physical laws, because they
dictate the behavior of other, more complex phenomena, such as mechanics,
ideal gases or light. Once the veracity of these universal phenomena has
been established, these phenomena never change and they can never be
refuted. Hence the physical laws are, for all intents and purposes,
immutable. It is theoretically possible for the deduced laws to be refuted,
if it can be shown that the deductions were wrong, but such laws tend to be
so basic that they in fact affect much of what we call science. The
refutation of any one of these deduced laws would have such a catastrophic
effect on science as a whole that it is virtually inconceivable that these
laws can be wrong.

So first we must recognize the existence of a physical law. Next we try to
describe it in some general, basic, abstract manner that usually involves
mathematics. For example, Boyle described the physical law he observed by
saying that the pressure exerted by any ideal gas at constant temperature is
inversely proportional to the volume of that gas. Mathematically this is
expressed as P=k/V, where "k" is a proportionality constant. This
description and formula are collectively known as Boyle's law, but they are
actually only a model of the physical law. We can call this model a
mathematical law, but the model is only as good as the observations done.
Boyle was unable to observe the behavior of gases at extremely low
temperatures, extremely small volumes or under extremely high pressures; if
he could he would have produced a better model that had a more detailed
description with a more accurate formula. Boyle at least knew that these
extreme conditions were possible, which was why he said he was modeling the
behavior of an "ideal" gas, but since at standard temperature and pressure
ideal conditions reign, Boyle's law is sufficient to describe one part of
gas behavior under most conditions.

The point is that Boyle's law is only a basic model of the physical law he
observed, thus it uses a formula that can only approximate correct gas
behavior. A better model based on a more detailed description had to wait
until the technology became available to observe gases under extreme
conditions; then a more precise formula was developed. Even so, the basic
description from Boyle's model still applies, even at those extreme
conditions, which confirms the universal nature of the physical law. This
is also why, if you use the more precise formula derived from this more
detailed description to calculate pressure values under ideal gas
conditions, you get the same result as you do if you use Boyle's law.

All of this is true for Newton's 2nd law of motion as well. The basic
description is that a change in momentum is proportional to the magnitude
and in the same direction of the force being applied to the mass. Newton
could not do any experiments where a mass was travelling at relativistic
velocities, and his view that space, time and mass were absolute did not
permit him to deduce relativity. As such, the (approximate) mathematical
formula he derived from his basic description was F=ma. Einstein developed
a more detailed description of Newton's physical law, and thus created a
more accurate mathematical formula, but the basic description is still the
same because the physical law still applies even under relativistic
conditions. Even if we had to wait until Einstein before we had a model for
this physical law, the basic description would still be the same, because
even at relativistic velocities a change in momentum is proportional to the
magnitude and in the same direction of the force being applied to the mass.

So in a nutshell this is my position. We are simultaneously discussing two
versions of the term "law". One is the physical law itself, the universal
physical phenomenon that is readily observable or deducible and which is
immutable and irrefutable. The other is the mathematical law which is an
abstract model and is composed of a basic description and a mathematical
formula. To my knowledge no physical law has ever been overturned or
refuted. Mathematical models sometimes are, especially if they are
partially based on theoretical mechanisms meant to explain how the physical
law works. But to my knowledge no basic description has ever been refuted
as long as it was based solely on the physical law itself. Of course, the
formulas derived from those basic descriptions have often been shown to be
inaccurate, so more detailed descriptions are created that can produce more
accurate formulas. But the basic description itself remains unchanged,
because the physical law itself remains unchanged.

>Nevertheless :), I tried to pick an example that would resist all
>attempts at rebuttal. You say above that " is also recognized
>that it is caused by the decrease in the speed of light as that
>light passes through dense material...". This is the whole
>controversy. And it was truly a controversy on the grandest
>scale, one of the most interesting in the history of science.
>Imagine the furor among the Cartesians when a mathematician
>(Fermat) derived the correct form of the law by making the
>outrageous assumption that light travels along that path which
>minimizes the time of travel. Clerselier, a well known
>Caretesian of the day, wrote of Fermat's principle:
>"That path, which you reckon the shortest because it is the
>quickest, is only a path of error and bewilderment, which
>Nature in no way follows and cannot intend to follow."
>Now, returning to your description of the physical phenomena:
>" is also recognized that it is caused by the decrease in
>the speed of light as that light passes through dense material..."
>This may be recognized today, but it wasn't then. Descartes
>(and everyone else, including Leibniz) thought light travelled
>more quickly through a dense media (a property predicted by the
>law. Snell's law is obtained from Descartes' by inverting either
>the right or left hand side). Not only is the law incorrect,
>the physical phenomena associated with the law (that light
>travels more rapidly through a dense media) is incorrect.

OK, back to basics again. Refraction is a physical phenomenon in which
light, passing from a media of one density into a media of a different
density at any angle other than perpendicular to horizontal (normal),
changes its angle closer to normal if the new media is denser than the old
or further away from normal if the new media is less dense than the old.
This basic observation has never changed and has never been disproven.
Because this is a universal phenomenon, it is also a physical law that light
must obey.

Descartes, Leibniz, et al. tried to explain this observation by saying that
light speeded up as it passed through denser media. This, however, was a
theory that proposed a specific mechanism to explain this observation, not a
physical phenomenon as you claim above. They then created an abstract model
of the observed phenomenon that accurately described the observed action of
light during refraction (specifically the change in angle from one media to
another), but could not accurately predict the value of this change in
angle, because the theory the mathematical law was based on that tried to
explain the observed phenomenon had it wrong.

So, the observed phenomenon of refraction is a physical law, because light
obeys it more or less univerally, and the abstract model of this phenomenon
is called a law because it describes a basic reproducible observation of the
phenomenon itself. The theory that explains how the observed phenomenon
works is not a law, because it is not itself a phenomenon, nor is it a basic
description of a phenomenon; instead it is the proposed mechanism by which
the phenomenon operates. However, a theory that explains how a phenomenon
works can influence the way in which that phenomenon is described. Hence
any mathematical law based on that theory, while essentially correct since
it describes the basic observed phenomenon, may turn out to be incorrect in
certain critical details if the theory turns out to be incorrect. The basic
observed phenomenon, however, remains unchanged. As such, as we study these
observed phenomena more carefully over time, our theories as to the
mechanisms that operate them will change, and thus our abstract models of
these phenomena will also change. If a theory is ever proven totally wrong,
then our abstract mathematical descriptions will change radically as well.
But as long as the basic observed phenomena do not change or are never
refuted, then the basic abstract description will not change either.

In the case under discussion here, the observed phenomenon of refraction was
not refuted, but the mechanism proposed by Descartes, Leibniz et al. was.
As such, Descartes' abstract model of refraction was refuted, but the basic
description that formed the core of Descartes' model was not. And it served
as the core of the new model proposed by Snell that was based on a new

So, using the confusing inadequate language available to us, the physical
law of refraction has never been refuted, but the mathematical law of
refraction proposed by Descartes was, based as it was on a theoretical
mechanism that was itself refuted.

Kevin L. O'Brien