/LMAR=0/XLINE=3/FONT#0=BAXL30/FONT#1=BAXM30/FONT#2=BAXB30/FONT#3=SUB/FONT#4=SUP/FONT#5=BASL35/FONT#6=NGR25/FONT#7=MATH30/FONT#8=FIX25/FONT#9=GRK30



␈↓ ↓H␈↓α␈↓ ∧⊃ARCHAEOLOGICAL CHROMATOGRAPHY

␈↓ ↓H␈↓␈↓ α_This␈α∪idea␈α∪was␈α∪triggered␈α∪by␈α∪an␈α∪article␈α∪in␈α∪which␈α∪an␈α∪archaeologist,␈α∪or␈α∪perhaps␈α∪a␈α∪mere
␈↓ ↓H␈↓journalist,␈αwent␈α
on␈αabout␈α
how␈αthe␈α
archaeologists␈αof␈α
the␈αpast␈α
were␈αblunderers␈α
compared␈αto␈αthose␈α
of
␈↓ ↓H␈↓today␈α↔and␈α↔carelessly␈α↔ruined␈α↔important␈α⊗sites␈α↔losing␈α↔important␈α↔information.␈α↔ Maybe␈α⊗present
␈↓ ↓H␈↓archaeologists␈α⊂also␈α⊃lose␈α⊂so␈α⊂much␈α⊃information␈α⊂that␈α⊂they␈α⊃will␈α⊂be␈α⊂cursed␈α⊃by␈α⊂archaeologist␈α⊃of␈α⊂the
␈↓ ↓H␈↓future.␈α∪ For␈α∪example,␈α∪today␈α∪archaeologists␈α∪destroy␈α∪information␈α∪that␈α∪might␈α∪be␈α∀retrievable␈α∪by
␈↓ ↓H␈↓␈↓↓archaeological chromatography␈↓.

␈↓ ↓H␈↓␈↓ α_Present␈αpaper␈αchromatography␈αis␈αbased␈αon␈αthe␈αfact␈αthat␈αwhen␈αa␈αliquid␈αcontaining␈αdissolved
␈↓ ↓H␈↓substances␈α⊃is␈α⊃poured␈α⊃on␈α⊃a␈α⊃column␈α⊃of␈α⊂filter␈α⊃paper,␈α⊃the␈α⊃different␈α⊃substances␈α⊃migrate␈α⊃down␈α⊂the
␈↓ ↓H␈↓column␈αat␈αdifferent␈α
rates.␈α This␈αpermits␈α
separating␈αsubstances␈αby␈α
sectioning␈αthe␈αcolumn␈αafter␈α
some
␈↓ ↓H␈↓percolation has taken place.

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

␈↓ ↓H␈↓␈↓ α_This␈α∪information␈α∪can␈α∩in␈α∪principle␈α∪be␈α∩extracted␈α∪by␈α∪collecting␈α∩each␈α∪cubic␈α∪centimeter␈α∩of
␈↓ ↓H␈↓material␈αunder␈α
the␈αruin␈α
to␈αa␈α
depth␈αof␈α
say␈αten␈α
meters␈αand␈α
analyzing␈αit␈α
for␈αtrace␈α
substances.␈α The
␈↓ ↓H␈↓amount␈αof␈αinformation␈αobtained␈αwill␈αdepend␈αon␈αthe␈αstate␈αof␈αthe␈αart␈αin␈αchemical␈αanalysis.␈α A␈αmole
␈↓ ↓H␈↓ccontains␈α∩6 ␈↓πx␈↓ 10␈↓∧23␈↓␈α∩molecules,␈α∩and␈α∩even␈α⊃single␈α∩molecules,␈α∩if␈α∩sufficiently␈α∩complex,␈α∩may␈α⊃contain
␈↓ ↓H␈↓significant␈α⊃information.␈α⊃ (There␈α⊃is␈α⊃no␈α∩fundamental␈α⊃law␈α⊃that␈α⊃precludes␈α⊃the␈α⊃detection␈α∩of␈α⊃single
␈↓ ↓H␈↓molecules, but that is another argument).

␈↓ ↓H␈↓␈↓ α_We␈αcan't␈αnow␈αanalyze␈αeach␈αcubic␈αcentimeter␈αof␈αdirt,␈αnor␈αcan␈αarchaeologists␈αafford␈αto␈αcollect
␈↓ ↓H␈↓and␈α∂tag␈α∂each␈α∂cubic␈α∂centimeter.␈α∂ Therefore,␈α∂if␈α∂we␈α∂want␈α∂to␈α∂maximize␈α∂the␈α⊂information␈α∂eventually
␈↓ ↓H␈↓obtained,␈α∂there␈α∂should␈α∂be␈α∂a␈α∂moratorium␈α∂on␈α∂archaeology␈α∂except␈α∂for␈α∂sites␈α∂that␈α∂will␈α∂be␈α∂destroyed
␈↓ ↓H␈↓anyway.␈α
 A␈αless␈α
drastic␈α
solution␈αwould␈α
be␈α
to␈αreserve␈α
most␈αof␈α
each␈α
site␈αfor␈α
future␈α
methods␈αtaking
␈↓ ↓H␈↓account␈αthe␈αfact␈αthat␈αarchaeologists␈αof␈αthe␈αfuture␈αwill␈αalso␈αhave␈αto␈αreserve␈αmost␈αof␈αwhat␈αis␈αleft␈αfor
␈↓ ↓H␈↓their successors.

␈↓ ↓H␈↓John McCarthy
␈↓ ↓H␈↓Artificial Intelligence Laboratory
␈↓ ↓H␈↓Stanford University
␈↓ ↓H␈↓Stanford, California 94305
␈↓ ↓H␈↓USA