LARNet; The Cyber Journal of Applied Leisure and Recreation Research 

Role of Manager and Visitor Self-Interest in Wilderness Management:
Nordhouse Dunes and Limits of Acceptable Change
(September 2003)
Dennis B. Propst, Dept. of Park, Recreation & Tourism Resources, Michigan State University
Maureen H. McDonough, Dept. of Forestry Michigan State University
Ami Wiita, Dept. of Natural Resources, State of Alaska

Contact:  Dr. Dennis B. Propst,
131 Natural Resources Bldg.,
Michigan State University,
East Lansing MI  48824-1222
phone 517/353-5190, X119; fax 517/432-3597;

ABSTRACT: The traditional role of the resource manager as omniscient, autocratic expert is being challenged in conjunction with challenges to authority-based leadership models across society.  This trend poses a dilemma for the managers of eastern wilderness.  Some form of recreational use restriction may need to be applied to these small preserves in the populous East, but citizens and wilderness users increasingly demand a say in such decisions. Study objectives were to develop a conceptual model that integrates the motivator, self-interest, into the "Limits of Acceptable Change" planning system (LAC) and to illustrate how to use the model to resolve manager/user conflicts. A yearlong survey of visitors at Nordhouse Dunes Wilderness (Michigan) provided a case to which to apply the model. Survey results presented managers with conflicting information about managers’ predetermined management strategies.  A key finding was that managers and users differ in perceptions of crowding and appropriate behaviors at Nordhouse Dunes. Once researchers discussed the survey results with managers, assisted in the discovery of areas of mutual self-interest and facilitated a cooperative learning session, managers incorporated conflicting information into their decisions and changed preconceived management actions. The mutual self-interest model holds much promise for conflict resolution—before the conflicts begin.

KEYWORDS: outdoor recreation, wilderness, self-interest, common property resource management, limits of acceptable change, conflict resolution

ACKNOWLEDGMENTS:  The authors wish to thank the U.S. Forest Service, North Central Forest Experiment Station (Chicago, IL) for the funding that sponsored this research. We also express our appreciation to Nordhouse Dunes Wilderness managers, John Hojonowski and Ramona Venegas-DeGeorgio, for their hospitality and assistance in planning and implementing the study and communicating the results to others. The views and conclusions expressed are those of the authors and not necessarily those of the U.S. Forest Service or its employees.

Garrett Hardin, in his classic treatise on natural resource depletion, argues that the tragedy of the commons in wilderness, parks, and other reserves leads to increased visitor impacts and hence steady erosion in the values that visitors seek (Hardin, 1968).    Burgeoning populations coupled with trends such as urban expansion and greater mobility present threats to all resources.  Policies and new cultural norms that limit the absolute numbers of humans and their impacts are mandatory (Hardin & Baden, 1977; Keyfitz, 1989; Hardin, 1993; Miller, 1996) else “freedom in the commons brings ruin to all” (Hardin, 1968, p. 1244).  Carrying the commons argument further implies that unrestricted access to wilderness, parks and other reserves will inevitably lead to these protected areas being “loved to death” since, as Hardin implies, humans are incapable of limiting individual consumption of common-property resources for some greater collective good.

Though criticized as suffering from cultural myopia (Feeny, Burkes, McCay, & Acheson, 1990; Shepherd, 1988), Hardin’s work has led some natural resource scientists and managers to resist calls for management approaches involving collaboration with publics. Hardin concluded that one way to avoid the tragedy was through government control (Hardin, 1968; Hardin & Baden, 1977).  Namely, state institutions through the professionals they employ (e.g., resource managers) should regulate uses and users.  These arguments have been used to support a continuation of the conventional authority-based model of "leadership" (Heifitz & Sinder, 1988).  Yet, one often overlooked qualification that Hardin made is that, to be effective, any type of social arrangement such as restricting visitation should be “mutual coercion, mutually agreed upon by the majority of the people affected” (Hardin, 1968, p. 1247).  Mutual coercion is consistent with Leopold’s land ethic, which calls for interdependent individuals or groups to develop means of cooperation in order to impose limits on the human use of natural resources (Leopold, 1966, p. 238).

Hardin’s qualification poses an interesting dilemma for the management of America’s eastern wilderness areas.   Some form of recreational use restriction may need to be applied to these relatively small preserves in the populous East, but we live in an age and society in which members of the public increasingly demand a say in government decisions that affect their use of a resource (Feeny et al., 1990; Propst, Wellman, Campa, & McDonough, 2000; Tipple & Wellman 1989; Wellman & Tipple 1990; Wondolleck, 1988).  The trend toward decentralization, greater user participation in resource management, “co-management”, and “participatory management” reflect dissatisfaction with top-down, autocratic and bureaucratic models (McCay & Jentoft, 1996).   Furthermore, there is evidence that individuals who lack knowledge of technical scientific issues can quickly learn about their critical features and choose policy options similar to those chosen by scientists (Doble & Richardson, 1992; Ferguson, 1987) or are at least as likely to ask the right questions and find novel solutions (Kaplan & Kaplan, 1983). The traditional role of the resource manager as omniscient, autocratic expert is being challenged in conjunction with challenges to authority-based leadership models across society (Heifitz & Sinder, 1988; Sirmon, 1993; Brick, Snow & de Wetering, 2001).

Several bodies of literature converge on a common conclusion: experts possess a belief system that often conflicts with the belief systems of the people who they are trying to serve.  In the outdoor recreation literature, Manning (1999) summarizes numerous published studies indicating that professional outdoor recreation manager perceptions of problems and their solutions often differ markedly from those of visitors.   Similar findings have been observed in the health care profession (Ferguson, 1987), the park and landscape design field (Clare Cooper Marcus, 1990; Halpirn & Burns, 1974; Hester, 1984; Hultsman, Cottrell, & Zale-Hultsman, 1987; Whyte, 1980), business (Peters & Waterman, 1982), environmental planning (DeYoung & Kaplan, 1988) and urban forestry (Magill, 1988).

Attempts to institutionalize the knowledge of expert/nonexpert differences into professional natural resource management training are as scarce as some of our natural resources (Wellman & Tipple, 1992; Propst et al., 2000).  The philosophical and ethical beginning points for such training exists in the natural resource and environmental policy literature (Shannon, 1991; Wellman & Tipple, 1990, 1993).  Shannon argues for the need for professional resource managers to build trust with citizens by engaging in ongoing “civic conversation”.   Wellman and Tipple (1990) argue for “direct democracy” where, instead of deference to the traditional role of the expert, recreation managers (in this case) should encourage active participation and hence contribute to the growth or perhaps rebirth of democracy.  Hefitz and Sinder (1988), who argue that in a democracy real leadership means turning the work of the community back to the community, support their position1 .

One way to implement the Heifitz and Sinder “communities of interest” approach to natural resource management is suggested in a report for the Forest Policy Center of the former American Forestry Association (Wellman & Tipple, 1992).   In this report, Wellman and Tipple challenge professionals to become managers of participation as well as managers of resources.  Managers of participation develop active citizens who, in turn, frame alternatives and assist in cooperatively managing natural resources.   Citing scholarly efforts at redefining the relationship between public administrators and the citizens they serve (Reich, 1988; Wamsley, 1990), Wellman & Tipple (1992) offer some practical guidelines for developing constructive partnerships between communities and organizations.   In addition, Wellman and Tipple (1990) and Propst et al. (2000) offer several ways in which education can change the worldview of the resource manager from traditional autocrat (passive citizenry mode) to facilitator of public interest and responsibility (active citizenry mode).  Knight and Landres (1998), Clark, Willard and Cromley (2000) and Brick et al. (2001) provide case studies of successes and failures in “collaborative conservation”, or the attempt to bring together a diverse coalition of interests to manage natural resources by voluntary consent and compliance rather than enforcement by legal and regulatory coercion.

Our purpose is to extend such recommendations for institutional and educational reform into a specific public policy arena.  We start modestly by arguing for the need to integrate the motivator, self-interest, into the “Limits of Acceptable Change” management framework as applied to eastern wilderness in the U.S.  Building upon the knowledge that resource managers’ worldviews often differ substantially from those of the people they serve and adding contributions from psychology and political science, we make recommendations regarding (a) a self-interest approach to the management of the Nordhouse Dunes Wilderness, (b) the role of “mutual self-interest” in forging effective partnerships in resource management, and (c) how the professional resource manager can implement and fulfill the challenges of this new role in a society where citizens demand changes in the way the expert goes about his/her business.

Literature Review
Researchers have known about the disparity in expert versus client views of the world for over 25 years (Manning, 1999).  Yet altering the way resource managers interact with publics has not been a foregone consequence of such knowledge.  Why not?  The cognitive psychology literature on human decisionmaking  provides a useful beginning for understanding this question.

One approach to understanding the mechanisms by which humans learn, make decisions and act upon their decisions is the limited capacity/cognitive map model of human behavior (Downs & Stea, 1973; Kaplan & Kaplan, 1983; Kaplan & Petersen, 1993).  Humans cannot think about all of their stored knowledge at once.  In other words, there is “a limit to how much activity the cognitive system can support at a time” (Kaplan & Petersen, 1993: 46). Familiarity with a class of stimuli and appropriate responses is helpful to organisms with limited capacity to process the amount and complexity of information with which they are constantly bombarded.   We can act quickly and appropriately with a minimum of storage capacity utilized, thereby freeing up additional capacity to process environmental information with which we have little familiarity and experience.

The cognitive map model of human brain functioning has widespread support (Lynch, 1960; Downs & Stea, 1973; Kaplan & Kaplan, 1983; Kotulak, 1996).  One implication relevant to this discussion is the potential bias towards making decisions quickly and with limited information rather than optimally or rationally.   Such a bias is supported by evidence concerning how managers and other professionals actually make decisions.  Namely, when faced with complicated decisions, professional often “muddle through” (Lindblom, 1979) or “grope their way along” (Behn, 1987; Peters & Waterman, 1982).

Furthermore, human cognition is said to be “conservative” in the sense that, once dense associations of familiar representations and their successful responses have been formed, there is great reluctance on the part of the organism to alter his or her cognitive map--hence the all too familiar resistant response to change, “but we’ve always done it this way before” (Kaplan, Kaplan, and Ryan, 1998).

Herein lies the cost of this limited capacity/cognitive map/quick response system that has evolved over the millennia.  Experts within any profession, including natural resource management, because of their training and experience, become familiar with various classes of technical, scientific information as well as a range of responses as to what to do when presented with such information.  This potentially limited range of responses is resistant to change not only because of the cognitive maps that have formed over time but also because of the motivational coding of values that goes along those maps (Kaplan & Kaplan, 1983; Gilovich, 1991).  Thus, experts possess a powerful cognitive and emotional self-interest in the actions and decisions that stem from their professional knowledge and experience.  This cognitive self-interest is bolstered by the shared values and purposes inherent in professional groups (e.g. foresters, wilderness managers, park rangers) (Orren, 1988).  Indeed, organizations such as the U.S. Forest Service work very hard to develop and maintain these shared values in their employees (Bullis & Tompkins, 1989).

Perloff (1987), in his review of psychological, economic and evolutionary literature pertaining to human behavior, argues that the variable of self-interest is a powerful and effective force underlying our actions.   Perloff argues that despite its malignment as antithetical to community goals, self-interest is central to American character, contributes positively to personal well-being, and hence is supportive of the public good.   Furthermore, social support (in the form of welfare provided by the state) often works against self-interest.  Giving advice and assistance, no matter how well-intentioned, may be excessive, untimely and inappropriate thereby creating unnecessary dependencies that limit self-help and initiative (Perloff, 1987).   In the political science literature, the same argument is made by Heifitz and Sinder (1988) when they propose that when leaders act as authorities who can solve problems, citizens no longer feel any personal responsibility for the problem or its solution. In a similar vein, Perloff qualifies the definition of self-interest by linking it to personal responsibility, noting that in the case of cancer patients, for example, self-interest leading to personal responsibility for one’s health leads to more effective coping strategies than the passive acceptance of the inevitability of the disease.   Perloff concludes his review by stating that self-interest enhances performance (a strong motivator), conserves one’s resources, and helps cope with illness and traumatic life events.  Less self-interest leads to lowered feelings of personal responsibility.

The Limits of Acceptable Change (LAC) planning process is one attempt to infuse public self-interest into the management of wilderness areas (Stankey, Cole, Lucas, Petersen, and Frissell, 1985; Stankey, 1999; Slover, 1996).   LAC and similar frameworks are even being integrated into the guiding principles and operational policies of entire national park systems (Graefe, 1996; Manning, 1999).  However fortified wilderness managers may be with policy or visitor data regarding the acceptability of various management practices, they are still left with the nagging question of how far to go in applying the LAC results versus their own expertise.  How should social data be integrated with the trend toward participatory management?  What should be done when visitor and manager views of what constitutes acceptable change are not the same?

Manager self-interest can be separated into technical expertise (learned through training) and values and beliefs regarding wilderness (some of which attract people to the profession and some of which are learned as members of the profession).  Part of the challenge to answering the two questions posed above is that it is cognitively difficult for managers to separate their technical expertise from their own values and beliefs (Figure 1).

Thus, manager self-interest is used to make decisions that seem rational to them, but not to the visitors they are attempting to serve.  Since wilderness is a social construction,  the contention of management that its role is to maintain “purity” is clearly one-sided and even delusional (Cronon, 1995).   Managers often either assume no visitor self-interest or that visitor self-interest conflicts with management-defined purity standards.  In addition, managers may not see how the application of their technical expertise is colored by their values.  In short, managers often presume that visitors cannot sustainably manage the commons.  Indeed, the typical reaction is to dismiss user input that does not agree with manager views.  Building upon Perloff’s synthesis and Hardin’s “mutual coercion” qualification regarding the commons, the framework adopted in this paper (Figure 1) calls for a balance of visitor self-interest (i.e., what visitors value and help themselves to achieve) and manager self-interest (i.e., social support as defined and provided by management).  In natural resource management, social support would include the provision of services and facilities for visitors’ convenience based on managers’ perceptions of what is best for the visitor.  The implication is that by leaving visitors out of the decisionmaking context, managers may inadvertently create unnecessary dependencies and visitor expectations of not being involved in natural resource management in a way that Leopold called for in his land ethic.  Consistent with Figure 1, we interpret mutual coercion to mean, “behavioral restrictions mutually agreed upon by all groups affected,” in this case by management decisions. The advantage of this framework is that efforts to limit resource use may not seem coercive in a negative sense (cf. DeYoung & Kaplan, 1988) once points of mutual self-interest are identified.

The Case Of Nordhouse Dunes
A study at Nordhouse Dunes Wilderness Area (NDWA) in Michigan provides a case to which the mutual self-interest model may be applied.  Nordhouse Dunes is the only designated wilderness area in Michigan’s Lower Peninsula.  It is within a six-hour drive of roughly 14 million people, including the cities of Chicago (IL) and Detroit, Ann Arbor, Lansing, Flint, Saginaw and Grand Rapids (MI). This wilderness area, designated in 1987 primarily because of its unique dunal ecosystem and endangered flora and fauna, consists of only 3,450 acres. The western border of the wilderness consists of 7,300 feet of undeveloped Lake Michigan shoreline.  Located within the Huron-Manistee National Forest, the wilderness shares its eastern border with the Nordhouse semiprimitive motorized area and its western border with the Lake Michigan Recreation Area, which contains a highly developed campground.  The U.S.D.A. Forest Service manages all three areas. Ludington State Park borders the wilderness to the south and is managed by the Michigan Department of Natural Resources.

The location of Nordhouse Dunes creates multiple management questions for the wilderness. These include use pressure, undesignated main entry points into the wilderness from surrounding recreation areas, and impacts of surrounding developed recreation areas, which differ in purpose from the wilderness, on wilderness use.  Also, prior to its wilderness designation, there were traditional/historical uses of the area such as off-road vehicles and snowmobiling.  Forest Service managers expressed concern that these uses, which were incompatible with a wilderness designation, were continuing to occur in the area.

To cope with these management challenges, staff of the Huron-Manistee National Forest began to develop a Limits of Acceptable Change (LAC) management plan.  To help develop the plan, the U.S. Forest Service contracted with researchers at Michigan State University to supply the necessary visitor information. The purpose of the visitor study was to estimate visitation to the wilderness area and describe its users and their perceptions of and preferences for the NDWA.  A full analysis of all data collected in this study can be found in McDonough, Propst and Wiita (1995) and Wiita (1998).

A combination of stratified and random sampling procedures was employed in the visitor study.  Sampling was conducted from May 1993 to March 1994.   The desired sample size was based on Forest Service estimates of 15,000 to 20,000 visitors per year.  A sample size of 3% was used to set a goal of collecting 500 to 700 surveys over the life of the project.

The sampling methodology was structured to obtain a representative sample of wilderness users across all times of day, days of the week, entry points and times of the year. Seven main entry points into the wilderness were identified and sampled. Because of lower visitation and a reduction in the number of available interviewers, sampling frequency decreased during the fall and winter sampling periods. With the exception of winter, all possible time blocks during daylight hours were selected to assure representatives across days of the week and entry points.  The result was a sampling design, which was representative of all possible daylight time periods and access points from May to March.

Interviews were conducted on site to reduce costs, increase response rates, counteract recall bias and obtain valid estimates of use.  Each person in a party over the age of 15 exiting the wilderness for the day was interviewed to gather the necessary limits of acceptable change information.  Interviewing only those groups who were exiting for the day minimized the double-counting that would occur if both entering and exiting parties were interviewed.

Data collection instruments used in other Eastern wilderness areas were provided by Dr. Alan Watson (Wilderness Management Unit, Forest Sciences Lab, Missoula, Montana).  Most questions for the instrument in this study were taken directly from these instruments to enable the direct comparison of these results to those of other wilderness studies. The instrument contained questions focusing on visitor group characteristics, trip and wilderness visitation characteristics, wilderness activities, visitor perceptions and expectations, level of acceptable encounters, crowding norms and visitor demographics.

Interviewers contacted 772 visitors.  A total of 506 visitors constituting 285 groups were interviewed.  The response rate for the study was 75% as 166 people refused.

Managers’ perspectives were assessed from a variety of sources throughout the study. Specifically, manager views were based on qualitative data generated through content analysis of recorded notes from a lengthy planning meeting and verified with a variety of printed materials.  During the planning stages of the research design, researchers and Nordhouse managers met to agree upon research objectives and content of the survey instrument.  A note taker kept a detailed written record of the meeting.  As recommended by Henderson (1991) as a method of increasing reliability (“dependability”) in qualitative data, the notes from this meeting were independently evaluated by three separate researchers who then met to reach consensus regarding consistent themes.  These themes were recorded and shared with the managers to check for discrepancies.  This process created an “audit trail”, which could be checked for consistency by managers and other researchers (Henderson, 1991).  As an added check for consistency, researchers and managers reviewed newspaper accounts and internal planning and public involvement documents that had been prepared by Nordhouse staff.   Participation by multiple researchers and agency representatives in sampling, data collection, coding, analysis, and reporting was used to enhance the confirmability, or objectivity, of the qualitative findings (Henderson, 1991).

Manager Perceptions of Nordhouse Dunes

A consistent theme from the qualitative analysis was that managers perceived Nordhouse Dunes Wilderness Area to be overused and overcrowded, especially along the Lake Michigan shoreline.  Quotes from an article that appeared in the major newspaper of the state capital (Lansing) exemplify this theme:

"It's [Nordhouse] just overused.  People are loving it to death…They [visitors] don't find the solitude they're looking for, especially along Lake Michigan" (Ingells, 1991).
Use was estimated by managers to be approximately 15,000 - 20,000 visitors, figures exclusive of the significant number of users the Forest Service believed enter the wilderness by watercraft (Table 1).  Management was concerned with the perceived continued illegal use of motorized vehicles such as watercraft, and especially, snowmobiles.  There was also a concern about indiscreet nude sunbathers and nude windsurfers who were thought to create discomfort for others, especially families with children.  Encounters among trail hikers and campers were estimated by managers to be “high”.  Managers described a wide variety of user types including day users, boaters of all types, nude sunbathers, horse riders, motorized vehicle users, large groups, and others whose presence and/or activity might conflict with other users. To managers, ecological and scientific values were the primary characteristics of wilderness that needed to be protected (Gobster, 1993; U.S. Forest Service, n.d.). Managers suggested, based on their use estimates, that "there could be a new class called 'Urban' Wilderness for the Nordhouse Dunes from mid-June through Labor Day" (Ingells, 1991).
Table 1 - Comparison of manager versus user views of Nordhouse Dunes Wilderness, Michigan (1995)
Managers' Viewsa
Visitors' Views
Recreational Use Overused and overcrowded Not overcrowded (83%)b

Visitors do not find solitude Place to be alone (87%)b

15-20,000visitors/year 7-10,000 visitors/year

Many conflicts Conflicts minimal

Nudism a serious problem Nudism not a problem
Minor activity (13%)c
Major activity (6%)d
Expectations nonexistent or not exceeded

Significant problems with watercraft Watercraft on beach not a problem 

Few watercraft users (20%)b

No watercraft observed by interviewers
Management controls Need more controls on horses, nudism Current controls adequate

Reduce control (10%)b

Increase controls IF overuse occurs (40%)b

Experience good  (95%)b

Wilderness quality staying the same or improving (89%)b
Meanings Reclassify as “urban wilderness”  Definitions of wilderness consistent with Wilderness Act of 1964 (83%):e
• Forces of nature predominate
• Minimal human influence

 Meanings outside of Wilderness Act (17%):e
• Spiritual/tranquility
• Sense of being away

Overall, Nordhouse meets my personal definition of wilderness (86%)b

a Manager's views were not based on a random sample of managers but upon qualitative data from personal communications, transcribed meeting minutes (Gobster 1993), Nordhouse Dunes published literature (US Forest Service, n.d.) and newspaper articles (e.g., Ingells, 1991)
b ( ) = percentage of visitors agreeing with the statement
c 19% participated in nude sunbathing;13% listed it as a secondary (“minor”) activity
d 6% participating in nude sunbathing as the primary (“major”) reason for visiting the area
e 83% of visitor definitions matched Wilderness Act legal definitions closely; 17% were personal definitions not specifically stated in the Wilderness Act

User Perceptions Of Nordhouse Dunes
In contrast to manager perceptions, 83% of the visitors to Nordhouse Dunes said that the area was not overcrowded and 87% felt Nordhouse was a place to be alone (Table 1).  Actual use was estimated to be 7-10,000 visitors annually or approximately half of manager estimates (McDonough, 1995).  Visitors felt that with the exception of hunting, visitor conflicts were minimal. Visitors did not consider nudism--seen by the managers as a serious problem for other visitors--to be a problem at all.  Nineteen percent (19%) of visitors said they participated in nude sunbathing at Nordhouse Dunes. Thirteen percent (13%) listed nude sunbathing as a minor activity, while 6% listed it as a major activity.  In response to the question, “how did the number of nude sunbathers you saw compare to the number you expected to see”, 1% said they saw more nude sunbathers than expected, 17% said they saw what they expected, a third said they saw fewer than expected, and half had no expectation regarding nude sunbathing.  Furthermore, visitors did not note problems with watercraft on the beach.  Less than 20% reported using a boat of any kind adjacent to the wilderness and none were observed by interviewers.

With respect to management controls, most visitors felt management controls were adequate. In fact, 10% wanted management controls reduced.  Only 6% of visitors wanted management to reduce visitation.  However, 40% wanted more management controls if overuse began to occur. Ninety five percent (95%) of visitors rated their experience at Nordhouse Dunes as very good or good and 89% of repeat visitors said wilderness quality was either stable or improving.

In an open-ended question, respondents were asked to describe what wilderness meant to them. Of the responses, 83% were well within the bounds of the Wilderness Act of 1964 (McDonough et al., 1995).  Many of their definitions were consistent with those of managers, particularly with regards to wilderness as a place where the forces of nature predominate and where the influence of humans was minimal.  Visitors also mentioned spiritual values and a sense of being away.  Unlike managers, visitors infrequently stated ecological and scientific values as part of their definitions.  Another discrepancy was that minimal human influence had a broader meaning for visitors ranging from completely untouched by humans to rustic accommodations.  Overall, 86% of the visitors agreed that Nordhouse Dunes met their personal definition of a wilderness.  The visitor data did not support management’s contention that Nordhouse should be reclassified as an “urban wilderness”.

Applying the Mutual Self-interest Framework

Managers see Nordhouse through the screen of their training and experience and so interpret what they see as overcrowded and problematic.   They then tend to assume that visitors feel the same way.  Visitors share with managers some of the same meanings of wilderness but did not interpret the current conditions as reasons for alarm.  These differences in interpretation of current conditions pose critical questions relative to the LAC process as well as other modes of public input.  How can managers understand the causes of the discrepancy between their perspective and that of the visitors?  For example, which of their proposals are a reflection of their technical expertise and which are a reflection of their personal values?  How can managers recognize the mutual self-interest they have with visitors who share wilderness definitions with them, think Nordhouse meets these definitions and want to see it remain in its current condition?  What are the appropriate management responses once awareness of these differences and similarities occurs?

One way to answer these questions is to reassess management issues at Nordhouse Dunes in a mutual self-interest framework.  As described earlier, the proposed framework builds upon the work of Perloff (1987) and Heifitz and Sinder (1988). The mutual self-interest framework has five parts (Figure 1). Part A is visitor self-interest, B is manager self-interest based on technical expertise, C is manager self-interest based on personal values and D is mutual self-interest.  As defined by Perloff (1987), manager self-interest manifests itself as “social support” provided by the state.  Following Perloff’s reasoning, when social support becomes excessive or inappropriate, dependencies develop and visitors no longer feel personal responsibility for resource management.   Visitor self-interest may manifest itself as complaints to agency officials and legislators, rule-breaking, resource damage and, in the extreme, lawsuits which temporarily or permanently restrain management activity.  Mutual self-interest is analogous to Hardin’s (1968) “mutual coercion” in the sense that  “democracy is governed by restrictive laws that can be legitimately described as exerting mutual coercion, mutually agreed upon” (Hardin, 1993: 297).  The authors’ conceptualization of mutual self-interest differs from Hardin in that rules and regulations may be developed jointly by manager and visitor dialogue rather than by laws and litigation.  One example is visitor/manager agreement on rules and restrictions pertaining to recreational activities at Cape Cod National Seashore (Canzanelli & Reynolds, 1996).   In the case of Cape Cod, visitors promulgated rules that were more restrictive than those that the managers were considering.   A similar event—user groups agreeing upon more restrictive regulations than those initially offered by managers--occurred at the Charles C. Deam Wilderness on the Hoosier (Indiana) National Forest (Slover, 1996).

In the following discussion, we demonstrate how the mutual self-interest framework works by applying it to four potential conflicts at from Nordhouse Dunes.  Three (crowding, horses, and nude sunbathing) were identified by managers as significant and the fourth, hunting, was identified by visitors.

Crowding. The issue of crowding has two components: biophysical (e.g., potential ecological damage to the wilderness from overuse) and social (e.g., the feeling of being crowded).  Managers believe that Nordhouse Dunes is crowded and that crowding is affecting the quality of visitor experiences. These perceptions are based partially on inaccurate estimates of actual use (B in Fig. 1) and partially on manager’s own definitions of crowding (C in Fig. 1).  Visitors do not think Nordhouse Dunes is crowded and they rate their experiences very highly (A in Fig. 1).  However, they do not want to see use increase.  Mutual self-interest lies not in controlling the current “crowding” which visitors do not perceive but rather in addressing two questions.  First, can use levels be kept at their current level?  Second, what should be done in the future if use levels do rise?  To hold future use to current levels, managers must first apply their technical expertise to monitor use closely and carefully (B in Fig. 1).  If use rises beyond current levels, managers and visitors will need to develop mutual use restriction strategies (D in Fig. 1). In addition, managers and visitors agree that some heavily used areas need to be managed (lower use, rehabilitate).  Due to their technical expertise (B in Fig. 1), managers are best at identifying ecological damage along trails and in campsites.  If ecological damage is identified by managers and exceeds certain standards, a mutual self-interest approach would suggest that managers and users develop cooperative regulations.

Horses.  At the time of this study, managers were proposing regulations because they projected horse use as a future problem (B and C in Fig. 1).  Visitors did not see this as a problem yet and, in fact, little horse use occurs in the wilderness area (A in Fig. 1).  Manager training is in identifying ecological damage (B in Fi.g. 1). Visitor self-interest is in maintaining the current condition as it meets their needs and their definition of wilderness.  As with crowding, mutual self-interest would be realized in two steps: (1) ecological damage begins to occur and is identified and documented by managers (B in Fig. 1), and (2) visitors and managers participate in cooperative development of regulations (D in Fig. 1).

Nude sunbathing.  Managers define this as a problem for visitors.  The data as well as comments written on the survey indicate nude sunbathing is not a concern of visitors.  The definition of a “problem” is likely a reflection of manager values (C in Fig. 1) rather than technical expertise (B in Fig. 1) as nude sunbathing causes little damage to the resource and in this case, to visitor experiences.  Increasing regulation by managers would have a negative effect on visitor self-interest, as visitors do not want more regulation (A in Fig. 1).  Mutual self-interest in regulating nude sunbathing (D in Fig. 1) might exist if visitor protection or safety issues existed but they do not.  There appears to be little mutual self-interest.  Thus, no regulation is necessary, as the problem does not appear to exist for visitors.

Hunting.  In this case, there are three sets of self-interest instead of two as visitor self-interest is divided.  While only 19% of the visitors to Nordhouse Dunes are hunters, the highest concentrations of visitors occur during firearm deer season (McDonough et al., 1995).  The views of hunters and nonhunters diverge substantially.  For example, 54% of nonhunting visitors agreed that there were too many hunters and that hunting should be banned.  On the other hand, only 15% of hunters agreed that there should be fewer hunters and 7% felt that hunting should be banned.  Legally, hunting is allowed and so managers must provide for this opportunity (B in Fig. 1) regardless of personal self-interest (C in Fig. 1).  Most hunters are present only during firearm deer season so there is no face-to-face conflict as user groups who object are separated in time.  In fact, most of the people who objected to hunting in the wilderness area were visiting in the summer months while firearm deer season is in November. It appears that both sets of visitor self-interest (A in Fig. 1) are being met through the separation in time.  If nonhunting use pressure increases in the fall, mutual self-interest might be achieved by having all three groups decide how to reduce conflict (D in Fig. 1).   That is, the principle of mutual coercion would suggest that all parties participate in cooperative development of further hunting regulations.


The purpose of this paper was to propose and test a model for improving conflict resolution in wilderness management.  The model integrates the powerful motivator, self-interest, into the Limits of Acceptable Change planning process and calls upon managers to facilitate public discourse with the goal of finding areas of mutual self-interest when determining appropriate management strategies. As recommended by the literature on participatory democracy, such a model asks public managers to manage citizen participation as actively and purposefully as they manage natural resources.  To the extent that this framework is implemented, resource managers follow another recommendation of the literature, which is the evolution from autocrats to facilitators of public responsibility in resource management.  The upshot pertains to a third recommendation of the literature, experimentation with collaborative management.

Collaboration is not a panacea for solving messy resource management issues, and it is not a way of relieving public managers of their obligations to make difficult decisions.  Nonetheless, collaborative efforts can at least solve some of the problems, avoid some of the costs (social as well as financial) of litigation, propose innovations in public land management, and contribute to direct democracy (Brick et al., 2001).  Collaboration is increasing becoming functionally integrated into the natural resource policy making process in the U.S. (Wellman and Propst, in press).

Collaboration is fundamentally different from traditional public involvement processes in which the flow of information is usually one-way, most often in the form of public hearings or requests for written comments on draft plans. The basis of the mutual self-interest model is two-way communications.  As with other collaborative processes, there is no guarantee of a short or predictable outcome. We recommend that resource managers implement and fulfill the challenges of the new role implied by this and other participatory democracy models by conducting small experiments in expanding public discourse and developing processes with which they are most comfortable.  The Nordhouse Dunes case exemplifies one way in which managers are realizing their new role.

At Nordhouse Dunes, the process toward identifying and implementing points of mutual self-interest has been ongoing since the researchers presented the results of this study.   Because of the LAC study, managers obtained a great deal of empirical data about the values, interests, expectations and characteristics of Nordhouse visitors. Having been exposed to these data, managers identified their own self-interest by engaging researchers in a dialogue, which attempted to clarify the distinction between personal values and technical expertise2.  This dialogue resulted in a formal presentation at a national conference during which time both researchers and managers demonstrated how the social science data from the LAC study had affected management decisions by refocusing resources on areas of mutual interest (Propst, 1998).  In addition, managers have engaged visitors in a dialogue about the results of the study and the need for use restrictions pertaining to specific behaviors in specific areas of the wilderness (John Hojonowski, personal communication, February 19, 2002).

Examining 22 years worth of evidence pertaining to common property resource management, Feeny et al. (1990) conclude that Hardin’s insights were critical but insufficient.  According to the authors, Hardin’s model overlooks the enormous role of nongovernmental institutional restrictions and cultural influences on behavior.  For example, shared governance or state regulation--jointly with user self-management--are viable options (Feeny et al. 1990).

The latter option, state regulation jointly with user self-management, can be implemented by taking advantage of a common human trait: self-interest.  The self-interest framework posits that mutual self-interest, the agreement zone in Figure 1, is achieved when managers and visitors mutually agree upon the actions necessary to achieve certain visitor experiences or restrict certain visitor behaviors.  This approach calls for managers to examine their own values and when they conflict with user values, seek a civic conversation that leads to agreement about appropriate strategies.  Management’s role is then to apply its strength--technical expertise--to the implementation of those strategies.  Reinforcing visitor self-interest through collaborative decisionmaking will transfer more personal responsibility to the users and diminish the view of wilderness as a “commons” than must be protected from humans who use it3.   Public participation will be ineffective unless all three types of self-interest are given balanced attention and institutionalized into the policies, educational programs and actions that affect natural resource decisionmaking. The mutual self-interest model holds much promise for conflict resolution—before the conflicts begin.

Literature Cited

Behn, R. (1988).  Management by groping along. Journal of Policy Analysis and Management, 7(4), 643-663.

Brick, P., Snow, D., & Van De Wetering, S. (Eds.). (2001). Across the Great Divide: Explorations in collaborative conservation and the American West.
           Washington, DC: Island Press.

Bullis, C.A. & Tompkins, P.K. (1989). The forest ranger revisited: A study of control practices and identification. Communication Monographs, 56, 287-306.

<>Bultena, G.L. (1974). Self-interest groups and human emotion as adaptive mechanisms. In D. R. Field, J.C. Barron & B.F. Long (Eds.), Water and community
           development: Social and economic perspectives. (pp. 151-169).  Ann Arbor, MI: Ann Arbor Science. 

Canzanelli, L. & Reynolds, M. (1996). Negotiated rule making as a resource and visitor management tool. Park Science, 16(2), 1-3.

Clare Cooper Marcus, C. A. F. (1990). People places: design guidelines for urban open space.  New York: Van Nostrand Reinhold.

Clark, T.W., Willard, A.R., & Cromley, C.M. (Eds.). (2000). Foundations of natural resources policy and management. New Haven: Yale University Press.

Cronon, W. (1995). The trouble with wilderness. The New York Times Magazine, Aug. 16, 1995, section 6.

De Young, R. & Kaplan, S. (1988). On averting the tragedy of the commons. Environmental Management, 12(3), 273-283.

Downs, R.M. & Stea, D. (Eds). (1973). Image and environment: Cognitive mapping and spatial behavior.  Chicago: Aldine.

Feeny, D., Berkes, F., McCay, B. J., & Acheson, J. M. (1990). The tragedy of the commons: Twenty-two years later.  Human Ecology, 18(1), 1-19.

Gilovich, T. (1991).  How we know what isn’t so: The fallibility of human reason in everyday life.  New York: The Free Press.

Gobster, P. (1993). Notes on January 27 meeting for proposed Nordhouse Dunes study. Unpublished meeting minutes. Chicago, IL: USDA Forest Service, North
           Central Forest Experiment Station.

Halprin, L. & Burns, J. (1974). Taking part: a workshop approach to collective creativity.  Cambridge, MA: MIT Press.

Hardin, G. (1968).  The tragedy of the commons. Science, 162, 1243-1248.

Hardin, G. & Baden, J. (Eds.). (1977). Managing the commons. San Francisco: Freemen.

Hardin, G. (1993). Living within limits: Ecology, economic and population taboos. Oxford University Press, New York.

Heifitz, R.A. & Sinder, R. M. (1988). Political leadership: Managing the public’s problem solving.  In Reich, R.B. (Ed.), The power of public ideas (pp. 179-205).
           Cambridge, MA: Harvard University Press.

Henderson, K. A. (1991). Dimensions of choice: A qualitative approach to recreation, parks, and leisure research. State College, PA: Venture.

Hester, R. T. (1984). Planning neighborhood space with people. New York: Van Nostrand Reinhold.

Hultsman, J., Cottrell, R.L., & Zales-Hultsman, W. (1987). Planning parks for people. State College, PA: Venture.

Ingells, N. (1991). Parks in peril. Lansing State Journal, July 7, 1D, 5D.

Kaplan, S. & Kaplan, R. (1983). Cognition and Environment: Functioning in an Uncertain World.  Ann Arbor, MI: Ulrich’s.

Kaplan, S. & Peterson, C. (1993). Health and environment: A psychological analysis. Landscape and Urban Planning, 26, 17-23.

Kaplan, R., Kaplan, S., & Ryan, R.L. (1998). With people in mind: Design and management of everyday nature. Washington, DC: Island Press.

Knight, R.L. & Landres, P.B. (Eds.). (1998). Stewardship across boundaries. Washington, DC: Island Press.

Kotulak, R. (1996). Inside the brain: Revolutionary discoveries of how the mind works.  Kansas City: Andrews and McNeel.

Leopold, A. (1966).  Sand County almanac. New York: Ballantine.

Lindblom, C.E. (1979). Still muddling, not yet through. Public Administration Review, 39, 517-526.

Lynch, K. (1960). The image of the city. Cambridge: MIT Press.

Magill, A.W. (1988). Natural resource professionals: The reluctant public servants.  The Environmental Professional, 10, 295-303.

McDonough, M.H., Propst, D.B. & Wiita, A.L. (1995). Nordhouse Dunes Wilderness user study.  Final Report. Chicago, IL: U.S.D.A. Forest Service, North
           Central Experiment Station and East Lansing, MI: Michigan State University, Department of Forestry.

Orren, G.R. (1988). Beyond self-interest. In Reich, R.B. (Ed.), The power of public ideas (pp.13-30).  Cambridge, MA: Harvard University Press.

Perloff, R. (1987). Self-interest and personal responsibility redux. American Psychologist, 42(1), 3-11.

Peters, T.J. & Waterman, R.H. (1982). In search of excellence: Lessons from America’s best-run companies.  New York: Harper and Row.

Propst, D.B. (1998). “Nordhouse Dunes Wilderness user study: Management and LAC implications.” Society of American Foresters National Convention, Sept.
           19-23, Traverse City, MI.

Propst, D.B., Wellman, J.D., Campa, H.R. III, & McDonough, M.H. (2000). Citizen participation trends and their educational implications for natural resource
           professionals. In D.W. Lime, W.C. Gartner, & J. Thompson (Eds.), Trends in outdoor recreation and tourism (pp. 383-393).  Wallingford, Oxon, UK: CAB

Propst, D.B. & Bentley, S.C. (2000). Trends in citizen participation in outdoor recreation and resource management: Manager vs. citizen perspectives. In
           Proceedings of the Fifth Outdoor Recreation and Tourism Trends Symposium (pp. 202-212). East Lansing, MI: Michigan State University, Department of Park,
           Recreation and Tourism Resources and Michigan Agricultural Experiment Station.

Reich, R. B. (Ed.). (1988). The power of public ideas. Cambridge, MA:Harvard University Press.

Shannon, M.A. (1991). Resource managers as policy entrerpreneurs: Governance challenges of the urban-forest interface. Journal of Forestry, 89(6), 27-30.

Shepherd, G. (1988). The reality of the commons: Answering Hardin from Somalia. Network Paper 6d.  London: Overseas Development Institute.

Sirmon, J.M. (1993). National leadership. In J.K. Berry & J.C. Gordon (Eds.), Environmental leadership: Developing effective skills and styles (pp. 165-185).
           Washington, DC: Island Press.

Sirmon, J. M., Shands, W.E., & Liggitt, C. (1993). Communities of interest and open decision-making. Journal of Forestry, (91)7, 17-21.

Stankey, G.H., Cole, D.N., Lucas, R.C., Petersen, M.E., & Frissell, S.S. (1985). The limits of acceptable change (LAC) system for wilderness planning. General
           Technical Report INT-176. Ogden, UT: USDA Forest Service, Intermountain Research Station.

Slover, B. L. (1996). A music of opinions: Collaborative planning for the Charles C. Deam Wilderness. Journal of Forestry, 94(3), 18-23

Tipple, T.J. & Wellman, J.D. (1989). Life in the fishbowl: Public participation rewrites public foresters’ job descriptions. Journal of Forestry, 87(3), 24-30.

Wamsley, G.L. (1990). Refounding Public Administration. Sage, Newbury Park, CA.

Weiner, J. (1995). Common property resource management and northern protected areas.  In D.L. Peterson & D.R. Johnson (Eds.), Human ecology and climate
           change: People and resources in the Far North (pp. 207-217). New York: Taylor & Francis.

Wellman, J.D. & Tipple, T.J. (1990). Public forestry and direct democracy. The Environmental Professional, 12, 77-86.

Wellman, J.D. & Tipple, T.J. (1992).  Working with the community: A conceptual framework for urban forest managers.  Forest Policy Center Report 92-3.
           Washington, DC: American Forests.

Wellman, J.D. & Tipple, T.J. (1993). Governance in the wildland-urban interface: A normative guide for natural resource managers.  In A.W. Ewert, D.J. Chavez,
           & A.W. Magill (Eds.), Culture, Conflict and Communication in the Wildland-Urban Interface (pp. 337-347).  Boulder, CO: Westview Press.

Wellman, J. D. and Propst, D.B. (in press). Wildland recreation policy,  2nd Edition. Malabar, FL: Krieger.

Wiita, A. L. (1998). Evaluation of managers’ and visitors’ perceptions of wilderness conditions at the Nordhouse Dunes Wilderness Area. Unpublished MS Thesis.
           East Lansing, MI: Michigan State University, Department of Forestry.

Wondolleck, J. M. (1988). Public lands conflict and resolution: Managing national forest disputes. Plenum, New York.

U.S. Forest Service. (no date). Welcome to Nordhouse Dunes Wilderness.  Unpublished transcript of slide presentation. Manistee, MI: USDA Forest Service,
           Manistee Ranger District. 

1 Heifitz and Sinder’s “communities of interest” model has been applied to natural resource planning in a preliminary manner with mixed results (Sirmon, 1993; Sirmon et al., 1993; Brick et al., 2001).

2 After managers had time to digest the data upon which this study is based, they participated in an interactive session with the principal investigators.  During this session, managers worked in small groups to develop management implications based on the new data. Each group then presented its implications to the entire group for further discussion and a revised list of implications for crowding, horse use, nude sunbathing, and hunting were developed (McDonough, Propst, & Wiita, 1995).

3 Feeny et al. (1990) emphasize the importance of clearly distinguishing between the type of common property resources--open access, common property, private property, state property--when evaluating Hardin.  Feeny et. al and others (e.g.,  Weiner, 1995) make the case that Hardin’s original chastisement of human inability to manage the commons referred primarily to open access resources (e.g., the Internet or fisheries where it is almost impossible to control access).  The Nordhouse Dunes Wilderness would fall under the category of “state property”.  We are aware of the differences among types of common property resources, but argue that many in the natural resource management profession feel that humans are incapable of adequately managing state as well as open access property.