Home | Archives | About | Login | Submissions | Notify | Contact | Search

 ES Home > Vol. 7, No. 1 > Art. 12

Copyright © 2003 by the author(s). Published here under license by The Resilience Alliance.
Go to the pdf version of this article.

The following is the established format for referencing this article:
Toupal, R. S. 2003. Cultural landscapes as a methodology for understanding natural resource management impacts in the western United States. Conservation Ecology 7(1): 12. [online] URL: http://www.consecol.org/vol7/iss1/art12/


Cultural Landscapes as a Methodology for Understanding Natural Resource Management Impacts in the Western United States

Rebecca S. Toupal

University of Arizona


Multicultural demands on public lands in the United States continue to challenge federal land managers to address social and cultural concerns in their planning efforts. Specifically, they lack adequate knowledge of cultural concerns, as well as a consistent strategy for acquiring that knowledge for use in decision-making. Current federal approaches to understanding such issues as access, use, and control of resources include public participation, conservation partnerships, government-to-government consultations with American Indian tribes, cultural resource inventories, and landscape analysis. Given that cultural knowledge arises from human–nature relationships and shared perceptions of natural environments, and that landscapes are the ultimate expression of such knowledge, an exploratory methodology was developed to provide a different approach to understanding cultural concerns through landscape perceptions. Using cultural landscape theories and applications from the natural and social sciences, this study examines the landscape perceptions of four groups concerned with management planning of the Baboquivari Wilderness Area in southern Arizona: the Bureau of Land Management, the landowners of the Altar Valley, recreationists, and members of the Tohono O’odham Nation. The methodology is based on a human–nature relationship rather than cultural aspects or features. It takes a holistic approach that differs from other perception studies in that it includes: emic aspects of data collection and analysis; a spatial component (triangulation of data collection through narrative and graphic descriptions); ethnographic, on-site interviews; and cultural consensus analysis and small-sample theory. The results include: verification of four cultural groups; two levels of consensus (in the population of concern, and in each group) that overlap in some aspects of landscape perception; descriptions of four cultural landscapes that illustrate similarities and differences among the groups, and include patterns and representations of spatial relationships; and an effective methodology for revealing cultural concerns that are not identified through public forums, and which has potential for application by agencies at the field-office level.

KEY WORDS: American Indians, cultural landscapes, ethnographic data, landscape perceptions, natural resource management, public participation.

Published: June 3, 2003


The western United States presents a mosaic of public, private, and tribal land ownership fraught with many cultural perspectives to which federal land managers must respond. Legislation and policies, which result from these perspectives as groups pursue particular agendas, may be incompatible with local cultural systems, giving rise to a self-perpetuating cycle of public input and legislative output (Fig. 1).

The federal manager’s challenge to address diverse cultural concerns effectively is compounded by the traditional dichotomous approach to management and policies, which considers natural and cultural resources independent of each other. Methods of public participation, partnerships, and government-to-government consultations with American Indian tribes are the traditional federal responses to cultural concerns. Each approach has specific weaknesses, however, that limit identification of cultural issues and often result in 1) hegemonious decisions that favor more powerful interests over others, 2) public dissatisfaction with government dominance over local communities, and 3) lawsuits.

Public participation meetings tend to become exercises in one-way communication, and the partnership strategy tends to be homogeneous, predominantly involving Euro-American groups and individuals. Frequently missing from both are the American Indian tribes who occupy approximately 53.5 million acres of land (about 7%) in the western states. Decades of failed relations, concerns about protecting sensitive cultural information, and cultural-based norms of decision-making commonly impede participation by American Indian tribes, who have legal standing as dependent sovereign nations. Federal agencies, consequently, must consult with them on a government-to-government basis that excludes public interaction (Appendix 1).

Technical approaches to identification of cultural issues include the Cultural Landscape Inventory (CLI) (Page 1998) and the Ethnographic Resource Inventory (ERI) (Roberts and Evans 2001) developed by the National Park Service (NPS), and Landscape Analysis and Design (LAD) developed by the Forest Service (USFS) (Diaz and Apostol 1993). The CLI is an in-house tool that focuses on material culture and provides a basis for selecting one of potentially many cultures for management and interpretation of resources. The ERI, another in-house tool, relies on historic and contemporary ethnographic data about landscapes, natural resources, objects, and places. The LAD provides the greatest opportunity to identify cultural concerns in that local citizens can participate throughout the inventory, planning, and decision-making processes. However, in addition to having cultural limitations similar to those of partnerships, this approach remains confined to a few USFS planning activities.

Through discussions with the Bureau of Land Management (BLM) and NPS, I identified two specific needs: the need for more cultural knowledge and the need for a consistent strategy to obtain such knowledge for decision-making (Juen personal communication, Walker personal communication). These needs arise in part from a lack of understanding of human–nature relationships and a segregated approach to interacting with the public and managing the landscape.

Inadequate cultural knowledge stems from the current approaches, the formats of which do not bring out deeper understanding of other cultural groups’ relationships with the land. In a public forum, for example, an Indian person might state that the mountain is sacred, which non-Indian people hear as religion. Although religion is part of what is meant by sacred, it is not the entire or even dominant meaning. The lack of understanding results from using one culture’s term for another culture’s concept, and from the reluctance to express intimate feelings or share sensitive knowledge in a public forum (Appendix 2).

The lack of a consistent strategy to obtain cultural knowledge stems from the traditional management approach that relies on biophysical sciences to understand natural resources, and on social sciences, such as history, archaeology, and sociology, to understand cultural resources. Each science may provide an understanding of a resource, but those understandings often are not contextualized or are restricted to material or man-made cultural items.

Seeking alternatives to these problems, I examined landscape and perception theories and applications from natural and social science disciplines, with an emphasis on human–nature relationships, cognitive maps, and cultural constructs of nature. I merged research techniques from these subjects to develop an exploratory methodology with which to describe landscape perceptions held by different cultural groups of a given area. Although the term culture has many definitions, for the purposes of this research, I chose the definition of shared beliefs, values, and norms about a particular knowledge domain, in this case the landscape. I adopted Kempton et al.’s (1997) description of cultural perceptions: that when the mental models of individuals are widely shared, they become cultural models. Through shared experiences, groups of individuals develop, transmit, and, consequently, share beliefs, norms, and knowledge about a specific realm, forming a “consensus” that is interpreted as “cultural” (Holland and Quinn 1987).

By revealing how and to what degree landscapes are important to and valued by different cultural groups, I hoped to provide a deeper understanding of cultural concerns that would allow integration of cultural issues with natural resource management. I intended the methodology to be operable at the field-office level, given some practical training.


The concept of cultural landscapes, which has its roots in cultural geography (Sauer 1931, 1963), provides an organizing construct within which cultural perceptions of and relationships with a natural environment may be characterized by components, uses, and meanings. To make the concept operational, I drew from landscape architecture and cultural anthropology for theories and applications capable of considering multiple temporal scales (Stoffle et al. 1997), intangible processes and phenomena (Zube et al. 1975, 1982, Saarinen 1976, Zube 1980, Kaplan and Kaplan 1982, Austin 1994, Greider and Garkovich 1994, Romney 1994), material culture and the built environment (Page 1998), and qualitative and quantitative methods (Reichardt and Cook 1979, Bernard 1995). I applied this mixed methods approach to sampling, data collection, and statistical analysis.

As the federal land manager is more concerned with multiple interests on a project-by-project basis than with comparing the cultural interests of two or more projects, I chose a single case study to test my methodology. I began with three criteria: (1) a contested management unit with (2) federal jurisdiction that had (3) at least three identifiable cultural groups, including an American Indian tribe. I decided that similarities and differences between groups could be better understood by including the managing agency as a fourth group, given that personal cultural filters hinder an understanding of other cultural groups.

A timely planning mandate for the Baboquivari Wilderness Area in southern Arizona provided an ideal case (Fig. 2). Baboquivari Peak, the most prominent feature of the wilderness area, is highly contested. The BLM has management responsibility for the area, and local landowners, recreationists, and the Tohono O’odham Nation (Papago) are the predominant interest groups. I developed my study design during the course of three public meetings, which confirmed the interest groups as those most affected by the BLM’s decision-making and provided access to individuals and organizations representing those interests. These meetings also produced a ranking of issues and concerns voiced by the participants (Fig. 3).


I examined previous landscape, perception, and cognition studies (McHarg 1969, Kaplan and Kaplan 1982, Turner 1988, Stoffle et al. 1990, 1997, Berlin 1992, Diaz and Apostol 1993, Austin 1994, Greider and Garkovich 1994, Lansing et al. 1998, Page 1998), ethnographic techniques, and small sample theory to develop my sampling strategy. Comparatively small ethnographic samples (fewer than 20) of key informants (someone recognized as knowledgeable about the area of investigation) are sufficient to obtain confidence that members of a culture agree on the array of items or symbols that constitute a cultural domain or category (Arnold 1970, Zube 1974, Daniel and Boster 1976, Schroeder and Daniel 1980, Romney et al. 1986, Boster 1987, D’Andrade 1995), or on the core attributes used to identify similarities among those items and to contrast them with other cultural domains (Kronenfeld 1996). Several studies provided guidance to determine when sample sizes are appropriate for obtaining data about cultural knowledge (Arnold 1970, Romney et al. 1986, Cohen 1988, Handwerker 1998, Kramer and Rosenthal 1999). Kramer and Rosenthal (1999) and Cohen (1988), in particular, have addressed the potential for inferential problems from small samples with effect size, a statistic that, when considered with significance, allows one to determine the adequacy of a sample size (Appendix 3). I applied this technique to all t-test results for the samples in this study.

I referenced Arnold’s (1970) dimensional sampling framework of universe, dimensions of variability, and typology of values. Based on BLM records and participation at public meetings, I defined the universe, or population of concern, as landowners, recreationists, Tohono O’odham people, and the BLM. My dimensions of variability were Altar Valley landowners who occupy the valley east of the Baboquivari Mountains, recreationists from Phoenix, Tucson, and Green Valley, Arizona, the Tohono O’odham Nation to the west of the mountain range, and BLM employees from the Tucson Field Office and Phoenix State Office. For a typology of values, I selected variables of gender and knowledge base, which are two key life experiences that contribute to cultural knowledge (Gupta and Ferguson 1997, Handwerker 1998).

I determined that sample sizes of eight to 16 informants for each group were adequate (Arnold 1970, Romney et al. 1986, Handwerker 1998); the landowner, recreationist, and Tohono O’odham groups each had 16 informants, and the BLM sample was reduced to eight because not enough employees met the typology (Fig. 4). Snowball sampling allowed me to identify key informants among the landowners, recreationists, and the Tohono O’odham, whom I selected for gender and knowledge-base variables.

Guided by the Tohono O’odham distinction of elder and non-elder knowledge, I determined knowledge base according to criteria regarding contact with the landscape, cognition of the landscape, and age, and placed individuals in the historic/traditional (H/T) group if they met any two of the criteria (Fig. 5).

Data collection

My strategy for data collection involved a triangulated, ethnographic approach of semi-structured, personal interviews held on site with key informants. Ethnographic data provide the cultural contexts within which resources are used and “thick descriptions” (Appendix 4) of human–nature relationships can complement archaeological, historical, and folklorist studies of material culture.

Triangulation can occur within a method, such as with open- and closed-response questions, or between types of methods (Denzin 1978, Beebe 1995). My “within” methods were open- and closed-response questions; the “between” methods were an interview instrument of these questions and a GIS-based map of the Baboquivari Mountain range and adjacent valleys (Fig. 6). Using these qualitative and quantitative methods of data collection, I was able to explore the parameters of the cultural landscape phenomenon in a way that provided valid numerical data (Ingersoll 1983, Fielding and Fielding 1986, Henderson 1991, Schensul et al. 1999).

I based the interview instrument on a cultural landscape paradigm of resources identified by Indian and non-Indian people during natural and social science landscape studies (Fig. 7). In addition to these resources (plants, animals, water, archaeological features, spiritual aspects and features, topography, historic aspects and features, views, sounds, sky, and places), I included a 12th category, other features, to give informants an opportunity to identify missed resources. I also based the instrument on the premise that contested landscapes present the federal land manager with a need to make decisions on what resources are used, who uses the resources, when they use them, where they use them, and how they use them.

I then developed multiple-choice questions to obtain characterization, uses, and meanings of resources for quantitative analysis and qualitative descriptions (Appendix 5). This structure ensured systematic data collection from each informant, allowing me to reduce problems of variation within and between groups (Bernard 1995, Wilson 1996). The structure of the interview instrument allowed informants to talk about the resources in a landscape context and to the extent that they desired.

As the semi-structured interview approach is most effective when conducted under conditions most relevant to and revealing about the system being investigated (Beebe 1995), I held the interviews on site at locales chosen by each informant. I began each interview by asking each informant to complete an informed consent form that reaffirmed the purpose of my study and the informant’s willingness to participate. I then asked each informant to describe his or her relationship with the Baboquivari landscape, and to describe that landscape. We positioned the GIS-based map on the ground in front of us so that the informant could provide a graphic description of the landscape as well. I oriented each informant to the peak and our interview location, so that those who were unfamiliar with reading maps, particularly the Tohono O’odham elders, had little trouble responding. I continued the interview with questions delving into characterization, use, and meaning of each resource, asking only about those resources that the informant identified as part of his/her Baboquivari landscape.

Statistical analysis

My analysis strategy required a determination of cultural consensus before I could run t-tests to examine possible significant differences in the ways in which informants characterize, use, and assign meaning to landscape resources. I chose consensus analysis for the closed-response data because it is an established method of examining cultural data for similarities and differences, both of which are important to understanding the cultural domains being examined (Romney 1994, Kempton et al. 1997, Romney et al. 2000) (Appendix 6).

The consensus model assesses internal validity by testing for the existence of cultural consensus and the presence and basis of cultural differences (Handwerker 1998). It works with true–false, multiple-choice, and fill-in-the-blank questions, to provide estimates of informants’ cultural competence, or knowledge, and of the correct answer to each question asked of informants (Romney et al. 1986). As Baer et al. (1999) describe it, this procedure “determines whether the responses of a group of individuals indicate a “consensus” in beliefs. [It] also estimates individual respondents’ level of cultural knowledge [and] estimates the culturally “correct” answers to the questionnaire (the normative cultural beliefs) [providing] a probabilistic confidence level for the classification of each item.”

Cultural competence scores are a function of the level of agreement among individuals (Romney et al. 1986). Romney (1999) defines cultural competence as “the proportion of the cultural questions for which the correct answer, or normative cultural belief, is known by the respondent.” He does not assume competence scores to be some “truth” but rather as representative of agreement or culturally shared concepts among the informants in the sample. The estimated answer key represents a simple majority, the answers most often given (Borgatti personal communication).


Coding the responses nominally, I ran cultural consensus analyses, one-sample t-tests, and independent t-tests, and calculated scores for characterization, use, and meaning of resources by group. I began with a consensus test of the responses of all 56 informants, which confirmed the sample fit the cultural consensus model. I then ran consensus tests on the responses of each subgroup, also confirming each as fitting the model. These results allowed me to proceed with one-sample t-tests between each subgroup and the total sample, and independent t-tests between variables of the landowners, recreationists, and Tohono O’odham subgroups.

Consensus results

The overall group had a 7:1 ratio between the first and second eigenvalues, and a cultural competence (shared knowledge of the landscape) of 0.687 or approximately 69% (Fig. 8). The subgroup consensus tests resulted in higher ratios than that of the overall group, and knowledge scores that were equal to or higher than their respective averages within the overall results. Respectively, the ratios for the BLM (Fig. 9), the recreationists (Fig. 10), the landowners (Fig. 11), and the Tohono O’odham (Fig. 12) were 10.6:1, 7.4:1, 8.6:1, and 9:1; the knowledge scores were 74, 69.6, 69.2, and 74%, respectively. The changes in eigenvalue ratios and shared knowledge reveal different cultural knowledge domains that are nested within or overlap each other, consequently contributing to similarities and differences in landscape perceptions.

One-sample t-test results

I identified specific differences pertaining to each resource by running one-sample t-tests between each subgroup and the overall sample for each of the closed-response questions. The Tohono O’odham had the most differences with 67 (14.3%) throughout 12 resources, followed by the landowners and recreationists each with 18 (3.8%) involving eight resources, and the BLM with 16 (3.4%) over five resources (Fig. 13).

I then calculated characterization, use, and meaning scores from five to seven questions each (Appendix 5), for each resource by subgroup. The scores reflect relative levels of complexity for each group’s relationship with the resource; higher scores reflect more complex relationships and stronger group connections to the landscape. Complexities and connections not being equal, I examined the relationship between t-test results and composite scores by charting the scores and highlighting those within which a significant difference occurred in at least one of the five or seven questions constituting its score.

An interesting pattern of increasing differences emerged as my analysis proceeded from presence through characterization, use, and meaning, despite more questions per category. I found one difference in the presence of elements (Fig. 14), with few Tohono O’odham informants including “other features” as a distinct resource, ten differences in characterization (Fig. 15), 43 differences in use (Fig. 16), and 66 differences in meaning (Fig. 17).

Independent t-test results

The within-group results I obtained for the recreationists, landowners, and Tohono O’odham revealed significant differences in characterizations, uses, and meanings of resources that are attributable to gender or knowledge base. Following this analysis, the question arose as to the influence of type of recreation. I found that seven of the eight men are climbers, as are two of the women. I also found that six of the H/T group are climbers and only three are in the contemporary/recent (C/R) group. Distinguishing between climbers and non-climbers, I ran additional t-tests for this variable looking for possible overlap with gender or knowledge base. As with the previous results, I have indicated significant differences by highlighting characterization, use, or meaning scores; higher scores reflect greater complexity and stronger connections to the landscape.

Among the recreationists, I found 44 (21.6%) differences (Fig. 18, 19, 20, 21, with the most in use, followed by meaning and characterization (Fig. 22). Knowledge base was the greatest contributor to differences in characterization, use, and meaning, followed closely by recreation type, and then gender. The only overlap in variables that I detected was for the condition of the plants and water, and whether all or part of the plants are used. Both the climbers’ and men’s groups considered plant and water condition to be satisfactory more than the non-climbers and the women, and they only use parts of the plants. As the evaluations and use involve these resources primarily during climbing treks, I suspect recreation type is the responsible variable.

Of the 29 (14.2%) differences that I found among landowners (Fig. 23, 24, 25, 26), most occurred in meaning, followed by use and characterization (Fig. 27). The characterization and use differences were more attributable to gender than to knowledge base, but more of the meaning differences were due to knowledge base.

I detected only 14 (6.8%) differences (Fig. 28, 29, 30, 31) among the Tohono O’odham, with gender having more impact on resource use and meaning (Fig. 32). Both variables contributed equally to differences of characterization.

For the most part, consideration of effect size indicates that the sample sizes of eight to 16 for the between-group analyses, and of four to eight for the within-group tests are adequate. At a confidence level of .95, only six informants with average knowledge of 0.7 are needed to achieve correct answers 95% of the time, and only four informants are needed to achieve correct responses 85% of the time. A few exceptions occurred in the BLM responses, suggesting a possible inadequacy of the BLM sample size of eight for the questions involved. This finding illustrates the advantage of the effect size–significance evaluation by pointing out specific questions that may require further investigation. The possible inadequacy of sample size is limited to these few cases, and does not apply to other BLM responses, nor to the other groups. Additionally, the questionable results involved acceptance of the null hypothesis rather than rejection of it. Greater differences, however, may exist between the agency and the other three groups.

Based on established estimates (Romney et al. 1986, Handwerker 1998), the reliability and validity of the results are quite high. Sixteen informants with an average level of agreement of 0.7 have an estimated reliability of 0.97 and a validity of 0.99; eight informants with the same level of agreement have a reliability of 0.95 and a validity of 0.97, and four informants have a reliability of 0.90 and a validity of 0.95.

Map data

When I asked the informants to describe the Baboquivari landscape on the GIS-based maps, I began by orienting them to our interview site and Baboquivari Peak. From these points, they began marking features, trails, use areas, and other details that define their landscape. Although the data are concentrated on the mountain range, these extend to the boundaries of their landscapes. Consolidating informants’ maps into group maps, I determined that the landowners and Tohono O’odham have more detailed and complex landscapes than the BLM and recreationists. I noted three details (Fig. 33) for the BLM, 11 for the recreationists (Fig. 34), 14 for the landowners Fig. 35), and 18 for the Tohono O’odham (Fig. 36). I also found group-specific emphases, which include the wilderness area for the BLM, the Altar Valley watershed for the landowners, access and the peak for the climbers, and Baboquivari Peak and its associated viewshed, which includes most of southern Arizona and northwest Mexico, for the Tohono O’odham.

I detected a positive correlation between spatial and temporal aspects of the subgroups’ maps. Spatially, the BLM has the fewest details in the smallest landscape, followed by the recreationists, the landowners, and the Tohono O’odham. The temporal aspect of the subgroups’ maps has to do with their respective histories or relationships with the area. The BLM has been in the landscape for the shortest period, with its management responsibility beginning 40 to 60 years ago. The recreationists’ claim a 105-year history that began in 1895 when Robert Forbes made the first documented ascent of Baboquivari Peak. The landowners date their group by the multi-generational ranching families in the Altar Valley and consider their history to range from 150 to 300 years. The Tohono O’odham informants identify a relationship with the landscape since “time immemorial.” Although anthropologists debate what this means, at least one noted archaeologist has documented detailed evidence of cultural remains in the nearby Pinacate area that date back more than 40 000 years (Hayden and Dykinga 1988). A local, informal consensus of 10 000 years is less controversial, however, we can say irrefutably that it is at least 500 years since the first European presence in the area was documented, approximately in the mid-1500s.

Qualitative data

Each informant provided qualitative data throughout his or her interview, which give life to the preceding results. Their narratives serve as portals to their relationships with the landscape, and provide insight to the roles of specific resources.

From the BLM informants, I heard first a managerial response about their relationships with the landscape that often was followed by a personal response (Appendix 7). As an agency, the BLM’s relationship with the Baboquivari focuses on resolving multiple, conflicting demands within the context of mandates and policies. As individuals, informants struggle with reconciling their professional responsibilities with their personal ethics.

I heard from the recreationists concerns that center on access to the wilderness area for hiking, climbing, and respite, as well as concerns about the possibility of increased use should access improvements become part of the wilderness plan (Appendix 8). It is the rock climbers, however, who experience the area in ways difficult to comprehend without having gone one-on-one with the peak. They respect the spiritual nature of the peak and experience it more intensely with each climb; the power of the mountain encompasses them without relief until they return to the base. This is the attraction, the spirituality and the power; not a desire to conquer nature. I also found that the power is not limited to the peak, although it is concentrated there. As the climbers retreat from the intensity of their experience, back through ecological zones to the built environment, they bring some aspect of that power with them. All the recreationists expressed similar feelings of renewal that nourishes them until their next opportunity to visit Baboquivari.

“It’s home.” Repeatedly, the landowners told me how the Baboquivari landscape, the mountain range, and its sentinel of a peak dominate the lives of those who live in Altar Valley (Appendix 9). It is a source of reassurance, information, pleasure, and spiritual feeling, and can be seen from practically every corner of the watershed, providing a bearing for anyone who looks beyond the roadways or trails in the valley. Wherever residents may be returning from, be it Puerto Peñasco, Mexico, the Santa Cruz Valley, Tucson, or Phoenix, whether by car or airplane, the peak is what they look for and when they first see it, no matter how far away, they know they are home.

The Tohono O’odham informants told me that, at one time, all the O’odham villages throughout the traditional lands had a view of Baboquivari Peak; everyone knew what it was and what it meant. They see this as explaining how the Baboquivari landscape is O’odham and the O’odham are the Baboquivari landscape. They describe their relationship as involving prayers, guidance, spirituality, identity, moral lessons, teaching, learning, and personal and cultural survival. It is infused and sustained through resource uses such as plants and animals for medicines, food, clothing, and tools, through kinship, ceremonies, stories, songs, prayers, and legends (Appendix 10).

When they have lost touch with the O’odham that is self, they return to Baboquivari to seek advice and to regain their identity through interaction with the mountain and I’itoi. It is where tribal leaders and elders seek strength and guidance to help their people. Individuals also turn to Baboquivari when some experience, such as military service or alcoholism, leaves them feeling lost and out of touch with themselves, family, and friends. They remember the traditional relationship and return to Baboquivari, sometimes spending several nights there alone and in contemplation. Always they return with a restored sense of identity and purpose, sometimes giving their lives to their people to help them overcome today’s challenges.

Many O’odham believe that their culture is dying and that the younger generations are not interested in the traditional ways because they have lost much of that relationship through the past 100 years of impacts from government agencies and religious institutions. They know, however, that as long as Baboquivari and the O’odham people exist, it is possible to rebuild that relationship.


All the informants have some awareness and a general understanding of the importance of the landscape to the other subgroups and to stakeholders outside the sample of this study. It is a management area, for example, to the BLM, a vista and recreational retreat for recreationists and some landowners, a livelihood for local ranchers, and a sacred site to the Tohono O’odham. For the Arizona Game and Fish, and the U. S. Fish and Wildlife Service, it is a wildlife management area; for the U. S. Border Patrol, a corridor used by illegal immigrants and drug smugglers; and for the military, Border Patrol, and Pima County Sheriff’s Department, a flight path. Beyond this general understanding, however, lies the substance of each group’s relationship with the Baboquivari Wilderness Area.

At their most generalized, the results indicate that management decisions for an isolated 2065-acre wilderness will have direct and meaningful impacts throughout a much larger area that is bounded by Tucson, Phoenix, and Green Valley, Arizona, Pozo Verde, Mexico, and the Sea of Cortez. In practical terms, the results indicate that each group brings significant filters to the public forum.

The BLM is guided by natural sciences, special interests, and legislation. It makes minimal use of and has minimal physical contact with the land and resources. Recreationists seek healthy ecology, biodiversity, and pristine conditions. They emphasize individual well-being and are hobby oriented. The landowners prefer to apply their knowledge of the land and resources. They are economic and production oriented, and they emphasize family and community in their landscape. The Tohono O’odham view the land and people as one. They engage in reciprocal relationships with the land and resources, and know they have responsibility to care for the land. Considering relationships within the results, I found implications for management, use of the public forum, and the original research problems.


The consensus results illustrate the relationship between different bodies of knowledge (Fig. 37). The agreement of the overall group reflects the shared perception of the public forum while subgroups maintain distinct landscape knowledge and perceptions that have some overlap with the public forum. The shared knowledge of the overall group is composed predominantly of presence and characterization of resources. Within the subgroups, shared knowledge includes, in addition to these parameters, uses and meanings.

The different layers of consensus illustrate the advantage of examining subgroups separately from the overall group. Additional interaction between the agency and each subgroup outside the traditional public forum has the potential to significantly enhance the agency’s understanding of cultural issues and the group’s understanding of agency issues.

Examination of between-group differences and within-group differences reveals little overlap. The landowners, for example, differ from the overall group regarding connections between the spiritual resources and events, stories, and songs. Within-group, however, the women indicated much use and many details of use. The men in turn were reluctant to acknowledge or discuss spirituality as part of their landscape. Discussion with the women about their families’ relationship with the landscape suggests otherwise, that the landscape in fact holds a great spiritual component for the men as well. The lack of discussion or acknowledgment of this resource by the male landowners may lead others to believe that these men view the landscape as nothing more than an economic base. Impassioned disagreement that may arise between the landowners and others in the public forum may be attributable to this aspect of the landowner relationship with the landscape, but because it is unvoiced, other reasons may be assigned that do little to improve the situation (Appendix 11).

Between-group differences for the landowners and recreationists, which are concentrated in uses and meanings, are fewer than their within-group differences. The latter may have less significance or pertinence to management but remain important in that they help explain some of the differences that occur in the public forum. By recognizing and considering the perspectives of men, women, those more experienced with the landscape, and those less so, the agency gains a deeper understanding of the cultural issues and potential impacts of and responses to its decisions.

The Tohono O’odham, however, have many more between-group differences than within-group differences; most of the latter are attributable to gender and most occur in meanings. This finding confirms a level of cultural continuity that reflects gender-specific traditional knowledge. From a management standpoint, these within-group differences may be quite pertinent as O’odham men and women continue to honor and practice their respective relationships with and responsibilities for the landscape.

I had anticipated more knowledge-base differences among the Tohono O’odham informants than I obtained, however, as I reviewed the dates of the interviews, I found that most of the interviews with elders were not held during the winter months. The Tohono O’odham believe the winter months are the safe and appropriate time to tell stories because the snakes are hibernating and will not be a danger to people; consequently, as they were interviewed outside of the winter months, their responses were limited in detail.

The pattern of increasing differences through categories, from presence in the landscape to general characterizations, aspects of use, and meanings that are associated with the resources, takes on more meaning when considered with the spatial and temporal correlation of the map data, that between the size and details of each group’s landscape, and the “age” of each group’s landscape. Assigning an age to a cultural landscape is possible because such landscapes represent perceptual and physical relationships between humans and nature that take time to develop; the pattern of increasing differences noted above suggests a process by which this development may occur.

The BLM group then, has the youngest, smallest, and least detailed landscape, within which they emphasize management. The recreationists, with the next youngest landscape, have more details and a broader landscape; they emphasize access, recreation, ecology, and connections. The landowners, with the next oldest landscape, have even more details and broader landscapes. They emphasize multiple boundaries, communities, habitats, topography, and multiple uses. The Tohono O’odham, with the oldest and broadest landscape, and the most details, emphasize multiple boundaries, communities, cultural use areas, topography, connections, and song trails. The latter finding is of particular interest because people who are inexperienced in reading maps are not expected to provide much information. I believe this finding reflects the capacity of the Tohono O’odhams’cognitive maps, their cultural models, of the landscape.

The relationship between the interview data and the map data is one in which the graphic informs the narrative and vice versa. From individual responses, group perceptions are identified; from individual mental models, cultural models are revealed. Landscape resources are described by aspects of characterization, use, and meaning, then given a geographic basis of interactions, which grounds the relationships of the people with the land. Use areas and multiple landscape boundaries identified on the map also indicate the potential for much broader impacts from management decisions, and the interview data elucidate details about those impacts.

Overlaying the graphic descriptions for the four groups reveals a mountain range of layered cultural landscapes. As a place where different cultural relationships are concentrated, the Baboquivari Mountains represent both conflict and accord. The data enhance our understanding of what and where these are and, from this understanding, topical and physical areas of conflict can be clarified and opportunities for cooperation can be identified.

How can land managers use this information in planning and decision-making? Given the agency’s limited use of the resources, it provides them with a way to better understand local landscape relationships and the reasoning behind cultural priorities assigned to landscape use. Rather than trying to respond to individual demands, the agency can substantiate a decision to respond to and use a cultural group’s local knowledge over that of a single voice, when considering both is not possible.

The results also provide support for incorporation of local knowledge in the planning and decision-making process. This is more than the ethically desirable goal voiced by some BLM informants. When local, indigenous, and scientific knowledge systems are combined, they inform and stimulate each other, consequently providing the best chance of sustainable natural resource use and management (DeWalt 1994).

Public forum and the ethnographic approach

During public meetings, landowners, recreationists, and Tohono O’odham people met with the BLM to identify and rank issues and concerns about management of the Baboquivari Wilderness Area. The participants realize, however, that a body of obscure knowledge remains unaddressed; much of this information is inaccessible in the public forum because of the sensitive or deeply personal nature of the cultural issues.

Although the meetings provided limited information about social concerns, cultural issues, and potential impacts from management decisions, the results of this research enhance understanding of some issues, and reveal significant concerns not brought out in the public forum. The results overlap with and contextualize three concerns from the public forum: access, management, and sacredness (Fig. 38).

In the public forum, for example, access is identified as a concern of the recreationists, and is assumed to mean physical access for hiking, climbing, bird watching, etc. In the interviews, however, I found access to be a need of both recreationists and the Tohono O’odham; it supplies the recreationists with spiritual renewal and escape from urban pressures, and it provides the Tohono O’odham with physical and spiritual renewal, connection to the “truth,” hope for continuance of their culture, and a means of teaching traditions to the youth. The agency now has more cultural knowledge to aid decision-making within the context of such diverse mandates as the Wilderness Act (1964), the American Indian Religious Freedom Act (1973), the Native American Graves Protection and Repatriation Act (1990), Executive Order 11593 (Protection and Enhancement of the Cultural Environment) (1971), Executive Order 13007 (Sacred Sites) (1996), and Executive Order 13175 (Consultation and Coordination with Indian Tribal Governments) (2000).

As another example, the Tohono O’odham people speak of the sacredness of the mountain and the peak at public meetings. Some have related abbreviated versions of how they felt lost when they came home from war and how they turned to the mountain for help. Away from the public forum, these individuals and others launch into discussions of plants, animals, medicines, ceremonies, and family that seem tangential at first but soon reveal the web within which sacredness and the meaning of sacredness is held. Within the same account, however, the youth are brought into the picture. They do not have the traditional foundations that the elders and other adults retain, nor do they seek such knowledge and philosophy. The traditional people are at a loss as to how to get through to the youth before it is too late and more of the culture is lost to the next generation. Were they to feel less constrained by outside forces that have some control over Baboquivari, they could pursue the traditional lessons, the passing on of knowledge in the appropriate setting and with the appropriate state of mind.

The original research problems

Federal land managers are hindered in their efforts to respond effectively to the diverse public they serve by a lack of cultural knowledge and of a systematic way to obtain it. At the planning level, where the brunt of resource management conflicts are felt, the federal resource manager faces specific challenges of natural and cultural resource management issues (Juen personal communication, Walker personal communication):

  • Field staff at all levels lack adequate knowledge and understanding of cultural issues that would allow them to do a better job;
  • Field staff at all levels lack skills, patience, and tolerance to work with local people or groups of diverse people;
  • Managers lack of a consistently effective way, including quantifiable methods, to combine multicultural values and concerns with current management tools and concepts for decision-making;
  • Personnel may consult but find it difficult, with their current level of understanding, to incorporate Native American concerns, issues, and needs into decision-making; for example, statements like “it is important, sacred to us” do not provide enough information for the manager to make an appropriate decision about where to site a mining operation;
  • Field staff at all levels lack knowledge and understanding of the uniquely federal relationship of tribes and federal agencies, of sovereignty, government-to-government relationships, and trust responsibilities.

The results of this research address these issues by providing new knowledge pertinent to wilderness management decisions, and by enhancing knowledge obtained through the public forum. The methodology, from sampling to data collection and analysis, provides a consistent strategy for obtaining cultural knowledge. The results also show how and to what degree the importance and value of landscapes to different cultural groups can be described. The potential benefits of this research to the BLM in its planning for the Baboquivari Wilderness Area include (Juen personal communication):

  • the ability to write a plan that incorporates cultural concerns with BLM decision-making strategies to effect an integral conservation of social and natural systems;
  • a learning opportunity for BLM employees;
  • an opportunity to inform the general public of cultural concerns without unnecessarily revealing sensitive cultural information;
  • placing the BLM in a better position to mitigate existing and anticipated conflicts;
  • a way to identify and incorporate diversity for more equitable management responses to the needs of diverse cultural groups; and
  • an opportunity for stakeholders to learn about the concept of cultural landscapes and its potential as a natural resource management tool.

The methodology is applicable at the field-office level. Compared with current methods of gathering cultural information, some training in ethnographic techniques and analysis may be required, however, agencies already employ social scientists, and have GIS technology, databases, and spreadsheet software to conduct the mapping and analysis. The time required to implement this methodology is comparable to the time frames of current methods, and just as dependent upon the public response to a given planning activity.

As the methodology is based on human–nature relationships from which cultural landscapes develop, it has broad applicability in situations matching the case study criteria of this research. The results, however, are not transferable to similar situations because Baboquivari is a wilderness area, which raises different values than non-wilderness areas. Each new case will be as individual as its landscape and population of concern, and will produce its own unique findings. The Montana rancher with a Scandinavian heritage, for example, will have a different landscape than the Colorado rancher with a German heritage. So too will the Lakota Sioux have a different relationship with their landscape than that of Western Shoshone tribes.

It should be noted as well that this study was limited to the predominant interest groups who have maintained communication with the BLM regarding the Baboquivari Wilderness Area. Other federal agencies, state agencies, and environmental organizations may have interest in the area, yet, having not interacted with the BLM directly or through the public forum, they remain wild cards in the BLM’s planning process. This research, therefore, represents a significant advance in the BLM’s understanding of the more vocal interests in the Baboquivari Wilderness Area, rather than a panacea.

An additional benefit of this methodology is its potential to improve federal agencies’ relations with the public. On-site interviewing, as conducted with the participants in this study, provides an informal and neutral ground conducive to an enjoyable learning experience for both parties. This benefit will become increasingly important as our population increases, and as environmental and political interests continue to complicate land uses and management.

Although the public forum has a legislated basis and role, the ranking of issues is more an artifact of that social environment than a reliable reflection of social and cultural issues. Consensus of the overall group, and through the public forum in the form of prioritized issues, differs from consensus of each cultural group as identified through systematic social science. The two levels of agreement have some overlap, however, significant differences exist between the two, and cultural consensus within each subgroup, as it represents agreement and shared knowledge, is greater than overall group consensus. Subgroup consensus and knowledge are useful to the federal manager in that these can inform and enhance public forum rankings, and aid the BLM’s negotiation of differences.


Responses to this article are invited. If accepted for publication, your response will be hyperlinked to the article. To submit a comment, follow this link. To read comments already accepted, follow this link.


My research would not have been possible without the guidance and support of my committee: Dr. H. R. Gimblett, Dr. W. W. Shaw, Mr. M. D. Johnson, Dr. R. W. Stoffle, and Dr. J. S. Lansing. The openness and enthusiasm of the BLM Tucson Field Office for this project was most gratifying. Documents from this office and the BLM Phoenix State Office were instrumental to understanding the management history of the Baboquivari Wilderness. I am indebted as well to the Altar Valley landowners, the recreationists, and the Tohono O’odham Nation and members who participated in this study, and thank them for sharing their landscapes.


American Indian Religious Freedom Act (AIRFA). 1973. 42 U.S.C. § 1996.

Analytic Technologies. 2002. ANTHROPAC, Version 4.98. Last revised: 29 September 2001. Last accessed: 10 December 2002. (Available online.) URL: http://www.analytictech.com/apacdesc.htm

Archaeological Resources Protection Act (ARPA). 1979. 16 U.S.C. 470aa – 470mm.

Arnold, D. O. 1970. Dimensional sampling: an approach for studying a small number of cases. American Sociologist 5:147–150.

Austin, D. E. 1994. Exploring perceptions of hazardous waste facility proposals in Indian country: an application of the active symbol cognitive map model. Dissertation, University of Michigan, East Lansing, Michigan, USA.

Austin, D. E. 1998. Cultural knowledge and the cognitive map. Praticing Anthropology 203:21–24.

Baer, R. D., S. C. Weller, L. Pachter, R. Trotter, J. G. D. Garcia, M. Glazer, R. Klein, L. Deitrick, D. F. Baker, L. Brown, K. Khan-Gordon, S. R. Martin, J. Nichols, and J. Ruggiero. 1999. Cross-cultural perspectives on the common cold: data from five populations. Human Organization 58:251–260.

Beebe, J. 1995. Basic concepts and techniques of rapid appraisal. Human Organization 54:42–51.

Berlin, B. 1992. Ethnobiological classification: principles of categorization of plants and animals in traditional societies. Princeton University Press, Princeton, New Jersey, USA.

Bernard, H. R. 1995. Research methods in anthropology: qualitative and quantitative approaches. Alta Mira Press, Walnut Creek, California, USA.

Bierwert, C. 1999. Brushed by cedar, living by the river: Coast Salish figures of power. University of Arizona Press, Tucson, Arizona, USA.

Boster, J. S. 1987. Agreement between biological classification systems is not dependent on cultural transmission. American Anthropologist 89:914–920.

Cohen, J. 1988. Statistical power analysis for the behavioral sciences. Lawrence Earlbaum Associates, Hillsdale, New Jersey, USA.

D'Andrade, R. G. 1995. The development of cognitive anthropology. Cambridge University Press, Cambridge, UK and New York, New York, USA.

Daniel, T. C., and R. S. Boster. 1976. Measuring landscape esthetics: the scenic beauty estimation. USDA Forest Service, Rocky Mountain Forest and Range Experiment Station, Fort Collins, Colorado, USA.

Denzin, N. K. 1978. The research act. McGraw Hill Book Company, New York, New York, USA.

DeWalt, B. R. 1994. Using indigenous knowledge to improve agriculture and natural resource management. Human Organization 53:123–131.

Diaz, N., and D. Apostol. 1993. Forest landscape analysis and design: a process for developing and implementing land management objectives for landscape patterns. Government Printing Office, Washington, D. C., USA.

Executive Order 11593. 1971. Protection and enhancement of the cultural environment (May 13, 36 FR 8921).

Executive Order 13007. 1996. Protection and accommodation of access to Indian sacred sites (May 24, 61 FR 26771 [42 USC 1996]).

Executive Order 13175. 2000. Consultation and coordination with Indian tribal governments (November 9, 65 FR 67249 [25 USC 450]).

Fielding, N. G., and J. L. Fielding. 1986. Linking data. Sage Publications, Beverly Hills, California, USA.

Geertz, C. 1973. The interpretation of cultures. Basic Books, New York, New York, USA.

Greider, T., and L. Garkovich. 1994. Landscapes: the social construction of nature and the environment. Rural Sociology 59:1–24.

Grenier, L. 1998. Working with indigenous knowledge: a guide for researchers. International Development Research Centre, Ottawa, Ontario, Canada.

Gupta, A., and J. Ferguson, editors. 1997. Culture, power, place: explorations in critical anthropology. Duke University Press, Durham, North Carolina, USA.

Handwerker, W. P. 1998. Consensus analysis: sampling frames for valid, generalizable research findings. Pages 165–178 in V. C. de Munck and E. J. Sobo, editors. Using methods in the field: a practical introduction and casebook. Altamira Press, Walnut Creek, California, USA.

Hayden, J. D., and J. W. Dykinga. 1988. The Sierra Pinacate. Southwest Center Series, University of Arizona Press, Tucson, Arizona, USA.

Henderson, K. A. 1991. Dimensions of choice: a qualitative approach to recreation, parks, and leisure research. Venture Publishing, Inc., State College, Pennsylvania, USA.

Holland, D., and N. Quinn. 1987. Cultural models in language and thought. Cambridge University Press, London, UK.

Ingersoll, B. 1983. Approaches to combining quantitative and qualitative social support research. Paper presented at the Annual Convention of the American Psychological Association, Anaheim, California, USA.

Johnson, M. 1992. Research on traditional environmental knowledge: its development and its role. Pages 2–22 in M. Johnson. Lore: capturing traditional environmental knowledge. Dene Cultural Institute and the International Development Research Centre, Hay River, Northwest Territories, Canada.

Kaplan, S., and R. Kaplan. 1982. Cognition and environment: functioning in an uncertain world. Prager Publishing, New York, New York, USA.

Kempton, W., J. S. Boster, and J. A. Hartley. 1997. Environmental values in American culture. The Massachusetts Institute of Technology Press, Cambridge, Massachussetts, USA.

Kramer, S. H., and R. Rosenthal. 1999. Effect sizes and significance levels in small-sample research. Pages 59–79 in R. H. Hoyle, editor. Statistical strategies for small sample research. Sage Publications Inc., Thousand Oaks, California, USA.

Kronenfeld, D. B. 1996. Plastic glasses and church fathers. Oxford University Press, New York, New York, USA.

Lansing, J. S., P. S. Lansing, and J. S. Erazo. 1998. The value of a river. Journal of Political Ecology 5:1–22.

McHarg, I. L. 1969. Design with nature. John Wiley, New York, New York, USA.

National Environmental Policy Act (NEPA). 1969. 42 U.S.C. § 4321–4347.

National Historic Preservation Act (NHPA). 1966. Public Law 89-665, as amended by Public Law 102-575.

Native American Graves Protection and Repatriation Act (NAGPRA). 1990. Public Law 101-601.

Page, R. R. 1998. Cultural landscape inventory professional procedures guide. U. S. Department of the Interior, National Park Service, Park Historic Structures and Cultural Landscape Program, Washington, D. C., USA.

Reichardt, C. S., and T. D. Cook. 1979. Beyond qualitative versus quantitative methods. Pages 7–32 in T. D. Cook and C. S. Reichardt, editors. Qualitative and quantitative methods in evaluation research. Sage Publications, Beverly Hills, California, USA.

Roberts, A., and M. J. Evans. 2001. Ethnographic landscapes, resource management, and resource protection in the National Park Service. Paper presented at the Society for Applied Anthropology Annual Meeting, Merída, Yucatán, Mexico.

Romney, A. K. 1994. Cultural knowledge and cognitive structure. Pages 254–283 in G, Spindler, M. M. Suárez-Orozco, and L. Spindler, editors. The making of psychological anthropology II. Harcourt Brace and Company, New York, New York, USA.

Romney, A. K. 1999. Culture consensus as a statistical model. Current Anthropology 40(Supplement):S103–S115.

Romney, A. K., S. C. Weller, and W. H. Batchelder. 1986. Culture as consensus: a theory of culture and informant accuracy. American Anthropologist 88:313–338.

Romney, A. K., C. C. Moore, W. H. Batchelder, and T. Hsia. 2000. Statistical methods for characterizing similarities and differences between semantic structures. Proceedings of the National Academy of Sciences of the United States of America 97:518-523.

Rose, D. B. 1996. Nourishing terrains : Australian aboriginal views of landscape and wilderness. Australian Heritage Commission, Canberra, Australia.

Russo, K. W. 2000. Finding the middle ground: insights and applications of the value orientations method. Intercultural Press, Yarmouth, Maine, USA.

Ryle, G. 1949. The concept of mind. Barnes and Noble, New York, New York, USA.

Saarinen, T. F. 1976. Environmental planning: perception and behavior. Houghton Mifflin Company, Boston, Massachussetts, USA.

Sauer, C. O. 1931. Cultural geography. Encyclopedia of the Social Sciences 6:621–624.

Sauer, C. O. 1963. The morphology of landscape. Pages 315–350 in J. Leighly, editor. Land and life: a selection of the writings of Carl Sauer. (Originally published in University of California Publications in Geography (1925) 2:19–54; reprinted 1938.) University of California Press, Berkeley, California, USA.

Sangre de Cristo Resource Conservation and Development Council. 2002. Quarterly Newsletter 10(4) Fall.

Schensul, S. L., J. J. Schensul, and M. D. LeCompte. 1999. Essential ethnographic methods: observations, interviews, and questionnaires. Altamira Press, Walnut Creek, California, USA.

Schroeder, H. W., and T. C. Daniel. 1980. Predicting the scenic quality of forest road corridors. Environment and Behavior 12:349–366.

Stapp, D. C., and M. S. Burney. 2002. Tribal cultural resource management: the full circle of stewardship. Altamira Press, Walnut Creek, California, USA.

Stoffle, R. W., D. B. Halmo, and D. E. Austin. 1997. Cultural landscapes and traditional cultural properties: a southern Paiute view of the Grand Canyon and Colorado River. American Indian Quarterly 21:229–250.

Stoffle, R. W., D. B. Halmo, M. J. Evans, and J. E. Olmsted. 1990. Calculating the cultural significance of American Indian plants: Paiute and Shoshone ethnobotany at Yucca Mountain, Nevada. American Anthropologist 92:416–417.

Toupal, R. S. 2000. Cowboys and conservation: the influence of private landowners on the success of conservation partnerships in the western United States. High Plains Applied Anthropologist 20:53–66.

Tuan, Y.-F. 1974. Topophilia: a study of environmental perception, attitudes, and values. Columbia University Press, New York, New York, USA.

Turner, N. J. 1988. The importance of a rose: evaluating the cultural significance of plants in Thompson and Lillooet Interior Salish. American Anthropologist 90:272–290.

Wax, R., and R. K. Thomas. 1961. American Indians and white people. Phylon: The Atlanta University Review of Race and Culture 22:305–317.

Weller, S. C. 1987. Shared knowledge, intracultural variation, and knowledge aggregation. American Behavioral Scientist 31:178–193.

Wilderness Act. 1964. P. L. 88-577, 16 U.S.C. 1131 et seq.

Wilson, M. 1996. Asking questions. Pages 94–120 in R. Sapsford and V. Jupp, editors. Data collection and analysis. Sage Publications Ltd., Thousand Oaks, California, USA.

Zube, E. H. 1974. Cross-disciplinary and intermode agreement on the description and evaluation of landscape resources. Environment and Behavior 6:69–89.

Zube, E. H. 1980. Environmental evaluation: perception and public policy. Brooks/Cole Publishing Company, Monterey, California, USA.

Zube, E. H., D. G. Pitt, and T. W. Anderson. 1975. Perception and prediction of scenic resource values of the Northeast. Pages 151–167 in E. H. Zube, R. O Brush, and J. G. Fabos, editors. Landscape assessment: values, perceptions, and resources. Dowden, Hutchinson and Ross, Inc., Stroudsburg, Pennsylvania, USA.

Zube, E. H., J. L. Sell, and J. G. Taylor. 1982. Landscape perception: research, application, and theory. Landscape Planning 9:1–33.

Address of Correspondent:
Rebecca S. Toupal
Bureau of Applied Research in Anthropology,
The University of Arizona,
P.O. Box 210030,
Tucson, Arizona, USA 85721
Phone: (520) 621-9607
Fax: (520) 21-9608

Home | Archives | About | Login | Submissions | Notify | Contact | Search