Presentation at American Anthropological Association, November 28 - December 2,
This talk describes on-going work for “SimShoBan,” an educational simulation created in collaboration with teachers, students and tribal culture representatives at the Shoshone-Bannock reservation in southern Idaho. The project goal is to use computer simulation to translate some of the sophisticated traditional knowledge systems of the Shoshone and Bannock communities -- which included botany, zoology, astronomy, and number and geometric patterns – into the framework of contemporary math and science education. Rather than assume the usual “one-way bridge” across the digital divide – the characterization of one side as technologically rich and the other as technologically poor – we use a “computational ethnography” framework to illuminate both the culture of our software programmers and the technology of Shoshone tradition. In addition to immediate practical application at the tribal school, and possible replication by other indigenous groups, our dissemination of this research can be used to improve software design practices by illustrating new possibilities for a collaborative approach to simulation.
The characterization of inadequate information technology resources in disadvantaged communities as a "digital divide" was a useful wake-up call. At the same time however, this metaphor is often taken to imply a problematic solution: the "one-way bridge." We have all the technology, they have nothing. This one-way bridge perpetuates stereotypes, and overlooks the valuable resources that disadvantaged groups can offer. This proposal describes a “two-way” bridge approach in which we illuminate both technology and culture on each side of the divide. By deploying an anthropological framework, our collaborative design for simulation of Shoshone-Bannock traditional knowledge systems will enable a more symmetric epistemological basis, empowering local education while offering new approaches for participatory software design in high-tech frontiers.
We refer to this new anthropological framework as “computational ethnography”. Anthropology has become increasingly involved in computational modeling, often using ethnographic data as the basis of its claims. The postmodern revolt in anthropology has often pointed out the pitfalls of modeling, in particular:
a. Reducing cultural complexity to
b. Using the authority of scientific objectivity to enforce neocolonialism (eg via IMF, World Bank. etc).
Postmodern ethnography has attempted to fight against reductionism by questioning the objectivity of data, and insisting on more acknowledgement of the subjective and political side of ethnographic inquiry. But this two negative consequences. First, it has distanced ethnography from technological access; it relinquishes the power of computing to those least likely to raise critiques against reductionism and authoritarian objectivism. Second, it implies romantic organicism, setting up an opposition between “bad” artificial representations and “good” human representations. Computational ethnography attempts to bring together the cultural critique of postmodern anthropology with power of technological access typically associated with reductionist modeling. It neither idolizes the analytic power of machines, nor romanticizes the intuition and discursive complexity of human beings, but instead asks for a collaborative relations between the two.
By creating a synthesis between the participatory approach of traditional ethnography and the technological power of computational resources, computational ethnography offers new possibilities for the application of anthropology to contemporary problems. We envision a wide range of methodologies within computational ethnography, including the bottom-up "artificial societies" approach of complexity theory, modeling scenarios based on traditional ethnographic approaches, and the development of simulations in collaboration with the subjects of investigation. It is the latter possibility that we explore in this essay.
Our case study for computational ethnography is “SimShoBan,” an educational simulation created in collaboration with teachers and students at the Shoshone-Bannock reservation. The Northern Shoshone originally inhabited a wide area extending from the Cache valley in northern Utah into Idaho. Restricted by both US colonization and environmental degradation due to cattle ranching, they are now located on a reservation in nearby Ft. Hall, Idaho, and have purchased some of the surrounding ranch land for return to its natural state. Shoshone science teacher Ed Galindo has pioneered innovative techniques for restoration of the local ecosystems, particularly re-introduction of salmon. Working with various community members--Mr. Galindo and other local teachers, students, and tribal culture representatives--we are creating a computer simulation that can model natural and social features of traditional Shoshone life. Most importantly, it provides the opportunity to translate the indigenous knowledge systems -- which included botany, zoology, astronomy, and number and geometric patterns – into the framework of contemporary math and science education.
We began in the spring of 2000 when a local university professor invited us to participate in research on Shoshone ethnomathematics. After an enthusiastic meeting with science and mathematics teachers at the tribal secondary school, we suggested that we expand the work to include other indigenous knowledge systems besides mathematics, and place the research in a computer simulation environment. Science teacher Ed Galindo was particularly enthusiastic about the project and offered to provide local leadership.
Back at RPI, we assembled a team of four undergraduates—two computer science majors and two electronic arts students—as well as a faculty member in ecology. We created a storyboard for how the simulation might be run, and a software prototype for the ethnomathematics material, the “Virtual Bead Loom” (VBL). In November 2000 we took the storyboard and software to the tribal school.
The VBL (see Virtual Bead Loom at this link) was a clear success. The VBL screen begins by showing the prevalence of four-fold symmetry in Native American design, where the “four directions” concept, an indigenous analogue to the Cartesian coordinate system, structures astronomical observations, calendars, numeric systems, and other knowledge domains. We then move to the Shoshone beadloom, showing the underlying Cartesian structure of its grid, and finally to the virtual loom. Here students can enter numeric coordinates for bead position; along with color choice this enables pattern capabilities similar to the indigenous loom. Students and teachers immediately understood the implications for showing the algebraic and geometric content in traditional bead patterns, all based on the deep cultural theme of the “four directions.”
The simulation storyboard (figure 1), in contrast to the VBL, was a near disaster.
The RPI students had based their simulation concepts on familiar games, in particular “Dark Ages,” in which players became medieval characters attempting to develop and defend their local village. The Shoshone-Bannock school students and teachers pointed out that a simulation in which everyone stayed rooted to one spot was replicating the reservation system—certainly not a representation of traditional life—and that the graphic characters might be too close to the offensive cartoon Indians portrayed in sports mascots. They suggested that a more accurate simulation would show how populations shifted to different areas with the seasons, and that a simulation player should learn about the technologies and activities associated with each. Many creative suggestions were then opened up for new kinds of graphical interface, some of which our programming team is still striving to understand. In addition to gaining new insight into how to provide more appropriate software for the Shoshone-Bannock students, this experience has also helped us illuminate the ways in which software subculture affects the supposedly “cultureless” world of computer programming (eg influence such as the “Dark Ages” game). It is important to note how computational ethnography is able to reverse the usual epistemological relations here: rather than see the computer programmers as the ones with all the technology, and the Shoshone as the ones weighed down with cultural baggage, we are seeking a portrait of Shoshone technology, and a cultural account of computer programmers.
We scrapped the original storyboard and started over again, beginning with their suggestion to focus on changes in the annual cycle. Drusilla Gould, a Shoshone-Bannock tribal member and professor of anthropology at Idaho State University, had worked with some local researchers on a publication that included a portrait of the Shoshone calendar (figure 2). The school teachers and students agree that this would make a good starting place. They are currently working on a color version of this calendar, which will be used as the front interface for the software. One interesting outcome of that project has been a compromise between traditional and contemporary views: the students get to choose the color coding, so while it is based on Shoshone culture it includes their own creative innovations as well.
Meanwhile, we decided to focus four food-gathering technologies, each associated with one season. Students will be able to explore the underlying geometry of these four structures by manipulating numeric parameters (see SimShoBan at this link). Three-dimensional simulations of woven structures has turned out to be a particularly challenging task in computational mathematics—and they have in turn illuminated new research directions, investigating how analogs to the polar coordinate system were used in conjunction with the Cartesian grid.
We have also found a new research area in the comparison between features of simulated bead patterns on the VBL and the details of traditional beading. The initial version of the VBL had only a single point tool. I asked my programmer, Lane DeNicola, to create a rectangle tool, which would take four corners as its coordinates, and a triangle tool that takes the coordinates for its three corners. Lane said that the rectangle tool would be easy, but asked me what the algorithm for filling in a triangle given any three corners would be. I told him that was trivial, and he replied “if its so trivial then tell me.” I thought about it, and said “Hey, its non-trivial!” So we looked up what is called a “scanning algorithm” – a standard algorithm for converting geometric shapes into pixel patterns on a screen—and just substituted beads for pixels. When we compared the resulting triangles on the virtual beadloom to the triangles in Shoshone beadwork, we got quite a surprise: the jagged, irregular edges in the virtual beadloom never appeared in the Shoshone version (figure 3).
After interviewing of the beadwork artisans, we have learned some of the indigenous algorithms for generating these shapes. The indigenous algorithms not only make more precise structures, they are also a better fit to standard school curricula, and they are more mathematically sophisticated.
Our first fieldtest of SimShoBan took place in June 2001 at Ed Galindo’s summer science camp. Located at 9,000 feet in Idaho’s Saw Tooth range, the camp makes use of the tribal efforts to restore Salmon as a teaching opportunity for students from the Shoshone-Bannock reservation. We set up laptops in the back of an SUV, and guided students through the use of both the virtual beadloom and an early prototype of the food-gathering technology simulations (follow this link to see the students’ creative manipulations of these virtual structures). Although we introduced the simulations as a “game” whose goal is to “get as close as possible to the original,” students quickly learned how to abuse the system, creatively modifying the simulations into forms unrecognizable as traditional artifacts. A particularly nice example was titled “the black bullet.”
The use of such anthropological materials in the classroom has several advantages. First, researchers have found that many minority students can improve their engagement with mathematics when it is better connected to their social background (Hanks 1998, Rasch 1994). Second, since many minority students report that they avoid math and science achievement because it is seen as “acting white” (Powell 1990), ethnomathematics offers a chance for youth to view math as a bridge to their heritage, rather than an barrier to this identity. Third, by using information technology to illuminate the sophisticated basis for indigenous knowledge—to translate indigenous knowledge into the “western” framework -- we can help combat the damaging myth of genetic determinism and offensive primitivist cultural portraits that saturate contemporary media.
In conclusion: Computer simulations always involve people--ether directly as subjects or indirectly as stakeholders. Participant Simulation is a methodology for providing greater involvement of the people affected by the simulation in its fundamental design.
Hankes, Judith E. Native American Pedagogy and Cognitive-Based Mathematics Instruction. New York: Garland Pub 1998.
Powell, L. "Factors associated with the underrepresentation of African Americans in mathematics and science." Journal of Negro Education, Vol 59, no 3 1990.
Keitel, C. and Ruthven, K. (eds) Learning from Computers: mathematics education and technology. Berlin: Springer-Verlag 1993.
Stiff LV, Johnson JL, and Johnson MR. "Cognitive issues in mathematics education." in PS Wilson (ed) Research ideas for the classroom: high school mathematics. NY: MacMillian 1993.
Means, Barbara. “Critical Issue: Using Technology to Enhance Engaged Learning for At-Risk Students.” Oak Brook, IL: North Central Regional Education Laboratory, 1997.
Witherspoon, Gary and Peterson, Glen. Dynamic Symmetry and Holistic Asymmetry in Navajo and Western Art and Cosmology. Bern and New York: Peter Lang Publishing, 1995.
Yerushalmy, M. “Using empirical information in geometry: students’ and designers’ expectations.” Journal of Computers in Mathematics and Science Teaching, vol 9(3) Spring 1990.