I have followed the thread of the last few days with great
interest and finally felt compelled to add my experience.  I
decided to cross-post it to ARCH-l and HISTARCH since the
debate has spilled over to both.  I believe Dwight Read
responded cogently to the attack on the use of GIS in
archaeology but I would like to address the question posed
by Larry Mckee as to the use of geophysical remote sensing
by archaeologists.
 
For background , I am primarily a "dirt" archaeologist
interested in the Middle Mississippian period in SW Indiana.
I got involved in geophysical remote sensing because my lab
had bought a gradiometer, resistivity meter, conductivity
meter, and a SUN workstation the same year I showed up for
grad school.  The opportunity was there so I took advantage
of it.  I have spent the last six years doing geophysical
surveys, trying to improve my field methodology and ground
truthing the results.  So, as an archaeologist I would like
to point out some of my pet peeves as to how archaeologist
deal with remote sensing surveys:
 
(1) They fail to involve themselves at the beginning in the
discussion of the survey methodology.  Decisions as to the
proper sample interval, or sensitivity level is left to the
geophysicist, or technician, who may have no experience with
the type of feature being sought or the specific geological
context.  Often this reticence is due to a sense of
ignorance on the part of the archaeologist, but there are
enough introductory articles available on geophysics to
remedy that. There has to be archaeological inputfor an
effective survey design.  A one meter sample interval may be
effective for a Roman fort but it is next to useless for
most of the sites I have surveyed.
 
(2) The Rorschach effect- I can't count the number of times
I have handed a dot-density map to collegues and they
immediately "see" houses, hearths stockade lines.  I know
archaeologists live on hope but there is a tendency to get
carried away.  I have an interesting resistivity survey map
from a Mississippian site which I always have to remember to
turn sideways during slide presentations since the dot-
density map is a dead ringer for the Shroud of Turin!
 
(3) The use of a proposed remote-sensing survey as a hook to
get funding
 
(4) The mistaken idea that remote sensing will detect all
sub-surfacefeatures.
 
As far as the use of geophysical remote sensing by my
institution, at this point it has become a standard part of
our pre-excavation strategy. This is due to the capital
investment in the equipment and a familiarity with what a
remote sensing survey will or will not tell you.  I have had
no difficulty in identifying sub-surface burned wattle and
daub cabins or stockade trenches less than a meter wide.
Under particularly favorable conditions, a light sandy soil
matrix, a gradiometer survey identified every pit (17 in
all) within the excavation area that contained over 100
grams of low fired pottery.  On the other hand it completely
missed a stockade trench which did not contain any
magnetically enhanced material. At this site the geophysical
surveys were effective at finding trash pits but were of no
help in the detection of house floors.  The prehistoric
occupants were not courteous enough to use wall trenches or
daub their structures.  I have found that cross-referencing
signals (in my case magnetic and resistivity) dramatically
improved any attempt at feature interpretation.  In regard
to the detection of graves; why would you use GPR (expensive
and complicated) when a resistivity survey (inexpensive and
simple) would detect the same properties.  I have only
surveyed one cemetery, an early 19th century pioneer
cemetery in Indiana. The location of the graves were quite
obvious from a resistivity survey performed at a .25 meter
interval.
 
The main selling point to remote-sensing is not its economy
or ease of use.  Remote sensing is non-invasive and non-
destructive.  It often takes more time and money to do
geophysical surveys than it would to swiss-cheese the site
with a power auger.  The auger survey would provide more
accurate information as to sub-surface deposits and would
also provide material remains for analysis.  Geophysical
surveys are tools that are among a suite of tools that
archaeologists bring to bear on the archaeological record.
GIS is a tool as well.  I have been using a GIS system for
the past six years.  I have never published or given a paper
on it because at this point I have not performed any
analysis which was deserving of public perusal.  The reason
there is so much crap coming out in GIS applications is
because there are too many archaeologists ignorant of its
limitations and therefore incapable of constructive
criticism.  A poor GIS analysis can slide by in a way that a
faulty statistical analysis or a flawed pottery seriation
cannot.
 
Sorry I appear to have rambled on a bit!
 
  Stephen Ball
  Glenn A. Black Laboratory of Archaeology
  Indaiana University