Date: 01 February 2002
Speaker: Trevor Harris
**Common Critiques of GIS:
1. GIS is a tool for epistemological assimilation. It imposes a way of
knowing--spatial primitives, topologies, defined spaces, etc.
2. Seen as and used as (if it were) "objective" technology. Promotes a
naive empiricism. Most evident in cases when data is seen as discovered
rather than as created.
-Actually, GIS is socially and politically constructed. People
create the data, and they create it for a certain
audience/client/customer.
3. Privileges modern, positivist science.
4. Ignores multiple realities.
5. Emphasizes cartographic space. Focuses on the physical.
6. GIS might be used (abused) as a tool for surveillance. Data
convergence through GIS.
**Trevor's Work:
Key concepts -- structural knowledge distortion; local
knowledge/geographies; sensuous GIS
Trevor's recent research in South Africa exposes aspects of the political
economy of GIS. The South African apartheid government didn't properly
document the forced removals of blacks from their tribal lands. The
government didn't record much data regarding these relocations, and now
there exists conflicting reports as to which lands were seized from blacks
and given to whites. In addition, lands currently held by blacks are
under-mapped. The government doesn't have a clear sense of the relative
fertility/potential production of much of its land.
This situation points to a primary danger of GIS: "structural knowledge
distortion." SKD describes the situation in which a region is under
mapped because insufficient data exists to properly handle it in a GIS.
No data = no geography. A lack of data leads to large holes in the map.
If a tribe wasn't "mapped" before it was relocated, and if the relocation
wasn't recorded, then the tribe has no apparent claim to its land.
Trevor and his team conducted field research in South Africa. Working
with the tribes and with the white farmers, they attempted to fill-in the
gaps in the maps. Trying to capture local knowledge, Harris and team
attempted to create maps more accurate to the inhabitants' senses of the
land. By relying on local knowledge, Harris hopes to combat some of the
dangers of GIS (noted above).
Trevor is also interested in sensuous GIS, in GIS that draws on more
intuitive visualizations (particularly 3-D) and on a wider range of senses
(touch and smell), multi-sensual GIS. He demonstrated various projects,
including a 3-D fly-over of central WV and VA (area where new highway is
planned) and a 3-D fly-over reconstruction of an ancient Indian burial
ground. He also pointed to some technologies currently used in video
games as potential tools for GIS.
Trevor also pointed towards OOGIS (object-oriented GIS) as a promising
area of development.
**Assignments/Projects:
1. Have students re-imagine a GIS software package. Re-think its
premises and the practices it encourages. GIS was developed at Harvard
and is very much an American, academic technology. To get students
started, ask them to imagine what GIS software would look like if it had
first been developed in China.
2. Teach students Exploratory Data Analysis (EDA) instead of traditional
statistics. The danger of much statistical analysis is that it encourages
students to find the magic number--the median, the mean, etc. EDA teaches
a process much more aligned with humanities research.
3. Electoral/Political Geography. Take GIS census data and ask students
to generate congressional districts based on that data. Requires students
to devise an objective and then to make the map that would support that
goal.
**Common Mistakes Made by Non-Geographers:
1. Scale. People often collect data at different resolutions and then
merge the data sets into one map at one scale. This created misleading
representations. Trevor used the example of a winding road and a nearby
river. If you capture data about the road and about the river but do so
at different resolutions, your combined map might seem to indicate that
road crosses the river at several places. This would lead one to assume
that there are several bridges that aren't actually there.
**Notes on M.A./Ph.D. program at WVU:
--15 Ph.D.'s and 30 M.A.'s
--1st semester everyone takes "Theories of Geography"; 2nd semester
everyone takes "Methodologies of Geography"
--in the methodologies course, students develop proposals for projects,
applying what they learned in the theories course
--most students don't teach but instead are research assistants
--the program does offer an elective course on GIS statistics and
packages. Here students deconstruct the models and algorithms used in
popular GIS packages
**Bibliographies
Trace History of GIS:
[1988] Chrisman, Nicholas. Paper presented at Auto-Carto 8 (?).
Annotation: Argues that GIS must be socially, economically and politically
responsible.
[1991] Openshaw, Stan. "A view on the GIS crisis in geography, or, using
GIS to put Humpty-Dumpty back together again." Environment and Planning A
v. 23 (May '91) p. 621-8.
Annotation: Violent criticism GIS as positivist and naive empiricism.
[1993] NCGIA Friday Harbor Conference
Meeting of cartographers and GIS.
[1995] Papers from above conference. _GIS and Environmental Modeling:
Progress and Research Issues_.
[1995] _Ground Truth: The Social Implications of Geographic Information
Systems_. Ed. John Pickles. New York: Guilford Press.
Annotation: Probably the most important book on GIS.
[2002] _Public Participation and GIS_
Miscellaneous:
Tukey, John. _Exploratory Data Analysis_. 1977.
Cleveland, William. _Visualizing Data_. 1993.
This archive was generated by hypermail 2b30 : Sun Feb 17 2002 - 18:03:11 EST