COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	.require "memo.pub[let,jmc]" source
C00007 ENDMK
C⊗;
.require "memo.pub[let,jmc]" source;
.cb ARCHAEOLOGICAL CHROMATOGRAPHY


	This idea was triggered by an article in which an archaeologist,
or perhaps a mere journalist, went on about how the archaeologists of
the past were blunderers compared to those of today and carelessly ruined
important sites losing important information.  Maybe present
archaeologists also lose so much information that they will be
cursed by archaeologist of the future.
For example, today archaeologists destroy information that might be
retrievable by
%2archaeological chromatography%1.

	Present paper chromatography is based on the fact that when a
liquid containing dissolved substances is poured on a column of filter
paper, the different substances migrate down the column at different
rates.  This permits separating substances by sectioning the column
after some percolation has taken place.

	It seems plausible that Nature performs chromatography experiments
for us.  Namely, a building falls down and the site is rained on.  The
water dissolves substances from the ruin and deposits them in the ground
underneath.  Subsequent rains move the substances further.
Thus the ground underneath a ruin contains information about the substances
and even the objects that the ruin contained.

	This information can in principle be extracted by collecting each
cubic centimeter of material under the ruin to a depth of say ten meters
and analyzing it for trace substances.  The amount of information obtained
will depend on the state of the art in chemical analysis.  A mole ccontains
6_%8x%1_10%523%1 molecules, and even single molecules, if sufficiently complex,
may contain significant information.  (There is no fundamental law that
precludes the detection of single molecules, but that is another argument).

	We can't now analyze each cubic centimeter of dirt, nor can
archaeologists afford to collect and tag each cubic centimeter.
Therefore, if we want to maximize the information eventually obtained,
there should be a moratorium on archaeology except for sites that
will be destroyed anyway.  A less drastic solution
would be to reserve most of each site for future methods taking account
the fact that archaeologists of the future will also have to reserve most
of what is left for their successors.
.verbatim

John McCarthy
Artificial Intelligence Laboratory
Stanford University
Stanford, California 94305
USA