Knowing the past?

[Kenelm Philip]

	Sally asked:

> How can anyone possibly know what happened thousands or even millions of 
> years ago?

	(I would have phrased that 'millions or even thousands'. :-)  )

	This is all based on what might be called circumstantial evidence--
but, to quote Thoreau: "Some circumstantial evidence is very strong, as
when you find a trout in the milk."

	The basic idea behind dating the past is to assume that processes
that can be observed to be going on now, and whose rates can be measured,
were also going on in the past. Some such processes are: radioactive decay,
erosion and deposition of sediments, continental drift (yes--this drift has
recently been directly measured, and its speed is known), changes in the
earth's rotational speed caused by tidal forces generated by the moon
and sun, growth of trees (one ring per year) and coral (daily and annual
variations), etc. There is also the field of astrophysics, where stellar
evolution (based on nuclear physics and the Hertzspring-Russell diagram)
has given us estimates for the ages of stars.

	Obviously, most of these do not allow _precise_ dating--but even
with large error bars it's clear that 500 million years (plus or minus,
say, 50 million?) is not 6000 years. Furthermore, the underlying pattern
is also clear: these diverse means of dating the past are in basic
_agreement_ with each other. An interconnected web of circumstantial
evidence is far more convincing than any single strand in that web might
be on its own.

	Creationists (young earth variety) will dispute the validity of
_all_ of these methods for ascertaining age. Since they do in fact
support one another, it would be exceedingly strange if every single one
of these, coming from totally different fields of science, were completely
invalid--and scientists who work with them and know how to use them were
all deluded.

							Ken Philip
fnkwp at uaf.edu