Innumerable as the Starrs of Night,
Or Starrs of Morning,
Dew-drops, which the Sun
on every leaf and every flouer
NGC3132 ©
Beauty is truth, truth beauty,
— that is all
Ye know on earth, and all
ye need to know.

E = M

Energy is eternal delight.
William Blake

Impearls: 2004-04-18 Archive

Earthdate 2004-04-19

Scientific Laws and Theories

Reader Mike Zorn writes in reply to Impearls' recent piece “Battle-tested General Relativity”:

“Query:  Why then bother to examine alternative theories of gravity?
Reply:  To have foils against which to test Einstein's theory.”

Exactly.  That's what makes a theory: it has to be capable of being disproved.

I think that one of the problems the public has with science is that the public's definition of “theory” is completely different from the scientist's.  […]  [T]he public [has a] perception of “theory” as “a really good guess,” vs. science's definition as “the best explanation we have so far that fits all the facts we have so far.”  (As in, “evolution?  It's just a theory — why should we pay so much attention to it?”)

Mike makes good points here.  Many people think of science as little more than a gathering or encyclopedia of facts and, as Zorn notes, there's a perception that a scientific “theory” is merely a vague hypothesis or guess, as good (or bad) as any other.

These common perceptions are actually quite far from science.  Looking back on the requirements for a “viable” theory of gravity as described in the previous posting (see the link above), it should now be clear that a good theory is vastly more robust than than a mere guess.  It must be internally self-consistent, incorporate all of physics (and chemistry, etc.) beneath its theoretical umbrella, agree with every experiment ever performed, and predict a vast spectrum of observable phenomena for the future.

As Mike says, it must be disprovable.  The public has this funny idea that science proves things.  Rather, science is only capable of disproving laws or theories.  The “best man” left standing after each candidate has been tested by “trial by combat” against all the criteria — including, last but not least, the searing fires of experiment, past and present — is (provisionally) considered to be the victor.

Nor does evolution in particular belong on a qualitatively different and lower plane than, say, physics.  The fact that evolution to an extent draws its information from out of the distant past is not important in this regard.  Geology and astronomy similarly derive much of their data from the far past, yet these are quite decidedly “true,” reliable sciences.  Every fossil dug up out of the ground is a newly-detected signal from the past, readily able to disprove evolution if new results show that the painfully built up pattern of relationships within the organisms of the past is but an illusion.  (I'll not hold me breath waiting for that to happen!)  Every science, in fact, deals with (and only with) signals from the past.  A physics or chemical experiment on a lab bench — any physical measurement — observes the past, as it takes time for light or whatever the medium to propagate from the site of the reaction or event into the measuring apparatus.

Physicist and philosopher of science Jacob Bronowski put it this way, in his book The Common Sense of Science: 1

We are not merely observing and predicting facts; and that is why any philosophy which builds up science only from facts is mistaken.  We know, that is we find laws, and every human action uses these laws, and at the same time tests them and feels towards new laws.  It is not the form of these laws which matters.  The laws of science, like those which we use in our private behaviour, remain helpful and truthful whether they contain words like “always,” or only “more often than not.”  What matters is the recognition of the law in the facts.  It is the law which we verify: the pattern, the order, the structure of events.  This is why science is so full of the symbolism of numbers and geometry, which are the most familiar expressions of structural relations.

There is no sense at all in which science can be called a mere description of facts.  It is in no sense, as humanists sometimes pretend, a neutral record of what happens in an endless mechanical encyclopaedia.  This mistaken view goes back to the eighteenth century.  It pictures scientists as utilitarians still crying Let be! and still believing that the world runs best with no other regulating principles than natural gravitation and human self-interest.

But this picture of the world of Mandeville and Bentham and Dickens's Hard Times was never science.  For science is not the blank record of facts, but the search for order within the facts.  And the truth of science is not truth to fact, which can never be more than approximate, but the truth of the laws which we see within the facts.  And this kind of truth is as difficult and as human as the sense of truth in a painting which is not a photograph, or the feeling of emotional truth in a movement in music.  When we speak of truth, we make a judgment between what matters and what does not, and we feel the unity of its different parts.  We do this as much in science as in the arts or in daily life.

We make a judgment when we prefer one theory to another even in science, since there is always an endless number of theories which can account for all the known facts.  And the principles of this judgment have some deep appeal which is more than merely factual.  William of Ockham first suggested to scientists that they should prefer that theory which uses in its explanation the smallest number of unknown agents.  Science has held to this principle now for six hundred years.  But is there indeed any ground for it other than a kind of aesthetic satisfaction, much like that of sacrificing your queen at chess in order to mate with a knight?

We cannot define truth in science until we move from fact to law.  And within the body of laws in turn, what impresses us as truth is the orderly coherence of the pieces.  They fit together like the characters in a great novel, or like the words in a poem.  Indeed, we should keep that last analogy by us always.  For science is a language, and like a language, it defines its parts by the way they make up a meaning.  Every word in the sentence has some uncertainty of definition, and yet the sentence defines its own meaning and that of its words conclusively.

It is the internal unity and coherence of science which gives it truth, and which makes it a better system of prediction than any less orderly language.


1 J. Bronowski, The Common Sense of Science, 1951, Harvard University Press, Cambridge, Mass.; pp. 130-131.

UPDATE:  2004-04-29 23:50 UT:  A follow-up “Copernicus Dethroned” has been posted.

Labels: , , , , , ,


(Blank last screen)


Are you an

If you fancy
any piece
you may donate
to reprise!


Pajamas Media BlogRoll Member

[Powered by Blogger]

Impearls Archives

2002-11-03 2002-11-10 2002-11-17 2002-11-24 2002-12-01 2002-12-08 2002-12-15 2002-12-22 2002-12-29 2003-01-05 2003-01-12 2003-01-19 2003-01-26 2003-02-02 2003-02-16 2003-04-20 2003-04-27 2003-05-04 2003-05-11 2003-06-01 2003-06-15 2003-06-22 2003-06-29 2003-07-13 2003-07-20 2003-08-03 2003-08-10 2003-08-24 2003-08-31 2003-09-07 2003-09-28 2003-10-05 2003-10-26 2003-11-02 2003-11-16 2003-11-23 2003-11-30 2003-12-07 2003-12-14 2003-12-21 2003-12-28 2004-01-04 2004-01-11 2004-01-25 2004-02-01 2004-02-08 2004-02-29 2004-03-07 2004-03-14 2004-03-21 2004-03-28 2004-04-04 2004-04-11 2004-04-18 2004-04-25 2004-05-02 2004-05-16 2004-05-23 2004-05-30 2004-06-06 2004-06-13 2004-06-20 2004-07-11 2004-07-18 2004-07-25 2004-08-22 2004-09-05 2004-10-10 2005-06-12 2005-06-19 2005-06-26 2005-07-03 2005-07-10 2005-07-24 2005-08-07 2005-08-21 2005-08-28 2005-09-04 2005-09-11 2005-09-18 2005-10-02 2005-10-09 2005-10-16 2005-10-30 2005-11-06 2005-11-27 2006-04-02 2006-04-09 2006-07-02 2006-07-23 2006-07-30 2007-01-21 2007-02-04 2007-04-22 2007-05-13 2007-06-17 2007-09-09 2007-09-16 2007-09-23 2007-10-07 2007-10-21 2007-11-04 2009-06-28 2009-07-19 2009-08-23 2009-09-06 2009-09-20 2009-12-13 2011-03-27 2012-01-01 2012-02-05 2012-02-12