|
|
The Systematic Collection of TEKJeffrey C. Johnson East Carolina University Traditional ecological knowledge (TEK) is an example of only one of the many kinds of knowledge that researchers are interested in collecting. There are other types of knowledge including, more generally, cultural knowledge, expert knowledge, beliefs, perceptions, feelings, emotions, and so on. These types of data are related in that they all involve the use of similar kinds of elicitation interviewing techniques (see the Johnson and Weller attachment). There is an important issue in the collection of knowledge type data that needs to be considered. This primarily concerns the ability to make valid comparisons of responses across informants. In part, our ability to make valid comparisons aids in delineating just what is core knowledge versus that knowledge that may be unique to a given informant or subset of informants. To what extent is knowledge shared among a set of informants and how does that knowledge vary within a given group (e.g., gender, expertise)? Comparison is facilitated by the use of systematic elicitation methods that allow informants to respond in a subjective manner to a similar set of stimuli or statements. Although open-ended and pure narrative approaches are important in eliciting valid stimuli, constructs, elements and so on, they inhibit our ability to make valid comparisons among informant responses. Just because an informant, for example, did not list something or mention something in the course of an open-ended interview does not mean that she/he does not have knowledge of that particular thing. Humans vary in their ability to recall or remember bits of knowledge and personal style often influences the amount a given informant talks (e.g., introvert versus extrovert). Thus, if an informant did not mention a particular thing was it because they lacked such knowledge or was it due to other factors? With respect to TEK we may be particularly interested in the development of traditional models of ecological dynamics (Atran, in press). What forms of reasoning do people use in providing for an understanding of specific ecological processes? What elements of cause and effect come into play when accounting for ecological change? Are these models shared and, if so, how does the understanding of a shared cultural knowledge vary according to, for example, age and the sexual division of labor? The methods described in this section provide examples directed at answering these questions and for testing hypotheses and constructing models related to TEK. As a matter of course in the design of the research for this type, one would tend to move from a more exploratory approach to one that is more explanatory (Johnson, 1998). The ethnographic research should move from a phase of data collection that is more qualitative and less directed to a phase employing more directed and systematic forms of data collection. This will ensure a higher degree of reliability and validity with respect to the operationalization and measurement of theoretical constructs and aid in curbing investigator biases. In addition, it will allow for more valid comparisons between informants that contributes to a better understanding intracultural variation. The following paragraphs discuss a sequence of methods that can be used in the collection of TEK. The important feature of this sequence of methods is that there is a movement from more exploratory or emergent methods to more structured or systematic ones. This is, of course, just one of many hybrid approaches one can use in studying TEK. The following is a brief description of a sequence that can be viewed as just one example of many possible sequences that can be employed. Exploratory Phase: In an exploratory phase, one would generally collect ecological narratives from an extensive sample of informants, selected according to theoretically important categories (Johnson 1990). These individuals would be interviewed in-depth about ecological knowledge of interest (e.g., local food webs, relationship between ice cover and the availability and behavior of animals, perceived trends in ecological processes--possibly as it relates to ice and changing ice cover over time), and, possibly, the place of modern technology in economic or subsistence activities. Visual-stimuli such as maps and photographs can be used to aid in the facilitation of discussions about these processes (Johnson and Griffith 1998). Care should be taken in considering the types of informants interviewed in this phase of the research. Depending on the objectives of the study, for example, an n cell sampling strategy in an age (young adults, adults, elders) by gender format controlling for expertise (expert vs. nonexpert) can be employed with approximately z (e.g., 5) informants per cell being interviewed (with possibly more than z experts being interviewed) in this exploratory phase. Expertise can be determined on the basis of a snowball sampling technique in which experts are use to help in identifying other experts (Johnson 1990). Sampling of this kind ensures that the researcher captures the range of variability in knowledge of a given domain within a particular group. Of course, much of this depends on the particular objectives of the study. The various interviews would produce data that can be used to design and construct more structured interviewing tasks for the collection of systematic traditional ecological data that facilitates the production of traditional models of ecological processes and dynamics (Boster and Johnson 1989). In addition, the ecological narratives can be used in constructing cultural models of ecological processes and change (Quinn 1996; Strauss and Quinn n.d.). Cultural Models: Following the work of D'Andrade (1989,1995), Quinn (1996), Strauss and Quinn (n.d.), and Hutchins (1995) on cultural models, one can examine native reasoning about ecological change and ecological dynamics. From informants' ecological narratives, as well as native biographies (e.g. Bodfish 1991; Blackman 1989) and ethnographic sources, reasoning about dynamics can be determined through an examination of the logical relations among sets of native ecological propositions (based on an examination of metaphors) concerning, for example, causes of change. These logical relations constitute a cultural model that will inform us of native reasoning about causes, effects, and consequences of ecological change. Explanatory Phase: The narrative data allows for the systematic study of the relation between indigenous perceptions of the causes of ecological change and impacts of such change as, for example, on 1) local flora and fauna, 2) local ecological conditions and human subsistence activities, including traditional rituals, and 3) food web relationships, among other concerns. This can be achieved by developing a number of tasks and instruments for the systematic collection of traditional ecological data. These can take the form of cultural consensus analysis (Romney, Weller, and Batchelder, 1986; Johnson and Griffith 1996), sentence completion tasks (Johnson and Griffith 1985; Johnson et al. 1987; Boster and Johnson 1989; Weller and Romney 1988; Griffith and Dyer, 1996), pile sort tasks (Johnson and Griffith 1985; Boster and Johnson 1989; Weller and Romney 1988) and tasks that elicit predator/prey interactions or other kinds of ecological relationships. Each of these methods facilitate the testing of hypotheses regarding cultural and intracultural variation in knowledge as a function of, for example, age, expertise, and the sexual division of labor (Boster 1985; Johnson and Griffith 1996). In addition, collection of this data will allow for a comparison of traditional and other systems of knowledge such as scientific beliefs (e.g., comparisons of scientific versus traditional ways of knowing). Consensus Analysis: Often we are interested in the extent to which there is a shared cultural understanding of ecological processes. Based on the ecological narratives a series of consensus statements can be developed after the methodology described by Romney, Weller and Batchelder (1986), Johnson and Griffith (1996) and Johnson and Weller (attachment). Such statements might consist of native propositions concerning ecological dynamics and perceived causes of ecological change. The cultural consensus model allows for evaluation of the extent to which TEK is shared. If there is a shared understanding, the ratio of the first to second eigenvalue in a minimum residual factor analysis will be greater than three and the scores on the first factor should all be positive and relatively high (greater than .5). In addition, the consensus analysis allows for the means for determining how variation in knowledge is a function of age, expertise, gender (i.e., social roles), etc. through a comparison of scores using ANOVA and MANOVA. Also, separate analyses can be run on different strata to explore the expectation of higher levels of agreement among experts (See Johnson and Griffith 1996). Further, the statements can be administered to a sample of other groups of interest (e.g., fishers in other regions, scientists) for the sake of comparison. Sentence Completion Tasks: Traditional ecological knowledge concerning the relationships between items of interest (e.g., local species) and the local ecology can also be explored through the use of sentence completion tasks (Weller and Romney 1988; Johnson and Griffith 1985) in which informants are asked to associate items (e.g., species) with various ecological statements relating to such things as ice cover, temperature, weather, etc. An example could be "When ice covers a large area in the spring, _______ are usually more abundant." These data are amenable to a number of multivariate modeling techniques that allow for the construction of models that capture items/conditions relationships and allow for the determination of logical relations between both items and conditions (see Johnson and Griffith 1985 and Boster and Johnson 1989 for discussion of the use of multidimensioal Guttman scaling in producing such models). Successive Pile Sort Tasks: Pile sort tasks have been a primary tool for the collection of data on ethnobiological classification (Boster 1988; Johnson and Griffith 1985; Boster and Johnson 1989). Free lists of local fauna can be elicited, photographed, and encased in plastic as described by Johnson and Griffith (1998). Informants can be asked to perform two tasks. In the first, informants engage in an unconstrained pile sort task in which species, or other environmental items of interest (e.g., photographs of different ecosystems), are placed into piles according to "how similar they are to one another" (Weller and Romney 1988). In the second, for example, informants may be asked to engage in a constrained pile sort task in which they place species into piles according to whether species sharing pile membership "eat the same or similar species". The first technique focuses on ethnobiological taxonomic relations while the second focuses on ethnobiological food web relations. In addition, data of this type can be modeled graphically using graph theoretic approaches and compared visually across different groups (Johnson et al. 1999). The attachment by Johnson and Weller provides a more in-depth discussion of these methods and other approaches. References Atran, S. In press. Managing the Mayan commons: The value of local knowledge. In: Ethnoecology: Situated Knowledge/Located Lives (Nazarea, V.D., ed.). Berkes, Fikret, Carl Folke, and Johan Colding (eds). 1998. Linking Social and Ecological Systems. Cambridge: Cambridge Univeristy Press. Berkes, F., C. Folke and M. Gadgil 1995. Traditional ecological knowledge, biodiversity, resilience, and sustainability. In C. Perrings et al. (eds.), Biodiversity Conservation. Kluwer: Dordrecht. Berkes, F. 1988. Traditional ecological knowledge in perspective. In Traditional Ecological Knowledge: Concepts and Cases (Inglis, J.T., ed.). Ottawa: International Program on Traditional Ecological Knowledge and International Development Research Centre. Blackman, Margaret. 1989. Sadie Brower Neakik: an Inupiaq woman. Seattle & London: University of Washington Press. Bodfish, Waldo, Sr. 1991. Kusiq: an Eskimo Life History from the Arctic Coast of Alaska. Fairbanks: University of Alaska Press. Boster, J.S. 1985. Requiem for the omniscient informan: There is life in the old girl yet. In Directions in Cognitive Anthropology. J. Dougherty (ed.) Univ. Of Illinois Press. Pp 177-197. Boster J.S. 1988. Natural sources of internal category structure: Typicality, familiarity, and similarity of birds. Memory and Cognition 16(3)258-270. Boster, J.S., B. Berlin, and J.P. O'Neil 1986. The correpondence of Jivaroan to Scientific Ornithology. American Anthropologist 88(3):914-919 Boster, J.S. and J.C. Johnson 1989. Form or function: A comparison of expert and novice judgment of similarity among fish. American Anthropologist 91:866-889. Boster, J.S. and J.C. Johnson 1992. Network structure and roles in isolated groups. Paper presented at the 12th D'Andrade, R.G. 1989. Culturally based reasoning. In: Cognition and Social Worlds (Gellatly, A, D. Rogers, and J. A. Sloboda, eds). Oxford: Clarendon Press. D'Andrade, R.G. 1995. The Development of Cognitive Anthropology. Cambridge, England: Cambridge University Press. Huntington, H.P. and N. Mymrin n.d. Beluga whale TEK. Arctic Social Sciences. Hutchins, E. 1995. Cognition in the Wild. Cambridge, Mass. MIT Press. Johnson, J.C. 1990. Selecting Ethnographic Informants. Sage: Newbury Park. Johnson, J.C. and D.C. Griffith 1985. Perceptions and Preferences for Marine Fish: A Study of Recrea-tional Fishermen in the Southeast Region. UNC-SG-85-01, 104 pages, University of North Carolina Sea Grant: Raleigh. Johnson, J.C. and D.C. Griffith 1996. Pollution, food safety, and the distribution of knowledge. Human Ecology 24(1):87-108. Johnson, J.C., D.C. Griffith, and J.D. Murray 1987. Encouraging the use of underutilized marine fishes by southeastern U.S. anglers. Part I: The research. Marine Fisheries Review 49(2):122-137. Johnson, J.C. (1998) Research design and research strategies in cultural anthropology. In: The Handbook of method in Cultural Anthropology . (R. Bernard, ed.) Altimira Press. Johnson, J.C. and D.C. Griffith. 1998. Visual data: Collection, analysis and representation. In: Introduction Lessons in Methods and Analysis: A Reader (V. de Mjmck & E. Sabo, eds). Altamira Press: Walnut Creek, CA. Johnson, J.C., S. Borgatti, J. Luczkovich, and M. Everett 1999. Network Models of Food Webs Part I. Paper presented at the International Social Network Conference, Charleston, SC, Feb. Quinn, N. 1996. Culture and contradiction: The case of Americans' reasoning about marriage. Ethos 24(3):391-425. Romney, A.K., S. Weller, and W. Batchelder 1986. Culture as consensus: A theory of cultural and informant accuracy. American Anthropologist 88(2):313-338. Strauss, C. and N. Quinn n.d. A Cognitive Theory of Cultural Meaning. Unpublished manuscript. Weller S.C. and A.K. Romney 1988. Systematic Data Collection. Qualitative research methods series, Volume 10. Newbury Park: Sage Pubs.
|