"Your scientific studies are impressive and your projections of future haze problems are sobering. However, I want you all to know that my decision to support or reject proposals will be based, not on these lengthy and complex scientific documents, but on the principles I walked in the door with. This is not about science, it's about values!" As the silver-tongued environmentalist sat down there was an audible sigh from several of the representatives of industry in the room and a few of the highly trained and dedicated scientists turned a bit pale.

The Public Advisory Committee was discussing how to develop consensus decisions about strategies for reducing air pollution, specifically visibility-impairing particles that cause haze. Many hoped that this would be a "data-driven process," that the more than $1 million spent on new studies and developing a computer-based model for the interstate transport of air pollutants would provide a scientifically-defensible foundation for proposals on how to reduce pollution. The passionate comment from the representative of the environmental group was discouraging to those who placed their hope in the science.

The facilitator for the group glanced over at me and another member of the facilitation team sitting at the side of the room and smiled. We had been wondering when someone in the group would finally expose the myth about a "data-driven process." We understood that the group needed good facts and reliable projections in order to make informed decisions. But, based on our experience with these kinds of processes, we knew that other factors would be driving the individual members of the group. This was one of the more diverse groups we had worked with, and it was clear that the science would not work for all members of the group. Values, culture, politics, and power would all have an impact-in addition to the clearly divergent interests of those in the room.

As the conversation continued, other participants rose to speak about how they would personally decide. Representatives of Indian tribes noted that they had been "invited to the table" late and that for many of them, the technical reports were tedious-and might even be designed to obscure the truth. It would not be the first time that the dominant culture would swindle the tribes. Tribal representatives would be considering the impacts of any proposals on Indian country. Lawyers and executives from utility industries insisted that decisions of the group would have to be supported by cost-benefit analyses that show that any proposed pollution control measures would be worth the sacrifices required to implement them.

The facilitation team remained confident that the group would eventually reach consensus on recommendations to the U.S. Environmental Protection Agency. But, at that moment, the conversation sounded like an orchestra in which each section was playing a different tune with a different rhythm and in different keys. Not a melodious enterprise. The rest of this chapter will describe the process of bringing people to playing the same tune and in relative-though not perfect-harmony. We will be using the case example of the Grand Canyon Visibility Transport Commission to illustrate how consensus-building works in public policy controversies.

The Grand Canyon Visibility Transport Commission (GCVTC) was established by the U.S. Congress in the Clean Air Act, as amended in 1990. The purpose of the Commission was to develop proposals for consideration by the Director of USEPA regarding measures needed to remedy existing visibility impairment (haze) and to prevent future impairment at sixteen designated national parks and wilderness areas in the Southwest. Because air pollution does not respect state boundaries, a regional approach was needed.

The Commission was composed of the Governors of eight Western States (Arizona, California, Colorado, Nevada, New Mexico, Oregon, Utah, Wyoming, ), four tribal leaders representing the Navajo, Hopi, Hualapai, and Acoma Pueblo, and ex officio representatives of five federal agencies (U.S. EPA, National Park Service, U.S. Forest Service, Bureau of Land Management, and the U.S. Fish & Wildlife Service), plus the Columbia River Inter-Tribal Fish Commission.

In order to develop widespread support for recommendations for remedying and preventing regional haze, the Commission established a variety of committees, some of the them devoted to conducting or overseeing scientific studies, and a Public Advisory Committee composed of approximately eighty representatives of stakeholder groups appointed by the Commissioners which began to meet in 1992. The PAC had responsibility for actually negotiating recommendations for consideration and approval by the Commission. The competing interests of private industry, citizens groups, environmental advocacy organizations, government (tribal, federal, state and local), and the scientific community were all represented on the PAC. See Figure 1 for a chart of the Commission structure.

The technical committees under the Commission structure amassed an impressive body of existing information about the patterns of pollution emission and transport, the chemistry of visibility impairing pollutants, and the composition of pollutants detected at sensors in the national parks and wilderness areas in question. They also commissioned additional studies and let contracts for the development of a computer-based program (the Integrated Assessment System) that was designed to project emissions into the future and estimate the potential gains and costs from various emission control mechanisms. All of this technical information was made available to the Commission, the PAC, all other Commission bodies, and the general public-for consideration in developing or responding to proposals.

Several factors distinguish the GCVTC process from other environmental policy development processes. First, the Commission was one of the first groups in the United States to attempt to develop policies on a regional basis. Since air pollution is transported by wind currents across state and tribal boundaries, no one state alone can implement successful strategies for controlling pollution. A regional approach is necessary. Particles causing haze at the Grand Canyon in Arizona might be generated in Los Angeles or Las Vegas-both in other states. While locally-generated pollutants are also critical, major portions of the problem are regional. When the GCVTC started, it was the first regional group to address air pollution policies; others have been established since.

A second distinguishing characteristic of the GCVTC was the active participation of tribal groups in policy development. While tribes had been involved, in recent years, in development of policies that affect Indian lands, this was the first time they had been invited to help generate policies with broader applications.

Participation of Indian tribes was not all smooth sailing, however. The tribal representatives were brought into the process late, especially some of the representatives on the PAC. Although their level of technical expertise was neither worse nor better than many of the other representatives of citizens' groups, many of them felt that technical information was difficult to comprehend, perhaps even deliberately so. In addition, aggregate information about the Western United States did not contain data detailed enough to enable precise analysis of the potential economic and social impacts of new policies on tribal areas.

There are approximately 211 different Indian tribes living on lands within the eleven-state "transport region" (the area from which pollutants are transported). Despite repeated efforts to invite participation of more tribes in the process, only about fifteen groups participated regularly. Most tribal governments have only a few people dedicated to working on environmental issues-who must juggle many competing priorities. Regional haze issues, while vitally important to some, was not a high priority for all groups.

Within the negotiation process itself, the tribal representatives began, over time, to develop mechanisms for unified responses and negotiating strategies. While some of the other groups (notably the utility industry and environmental groups) had considerable experience working as allies in environmental negotiations, this was the first time so many tribal groups had come together to engage in development of long-term government policies.

The process of developing policy recommendations through a negotiation process among diverse stakeholder groups represents a maturing of democracy. While authority for establishing policies still rests with government agencies, and ultimately with Congress, in recent years, agencies have been using consensus processes among stakeholder groups to develop policy recommendations. The recommendations of the PAC, after being confirmed by the Commission, were sent to the Administrator of EPA and, along with input from many other sources, became the basis for development of national rules regarding regulation of certain particles. The recommendations will also serve as policy guidance for an ongoing regional body which will monitor progress towards reducing existing visibility-impairing pollution and preventing future pollution.

Figure 2 provides a picture of the typical pattern of consensus building in multiparty policy development and dispute resolution. The chart shows a progression from mutual education to identification of values and interests, identification of issues, development of criteria, generation of options, proposal development, and the final process of building consensus. While this process is presented as a logical sequence, the actual practice is almost always more complex and iterative. For instance, early in the process, the parties may educate each other regarding their perspectives. Later, in the final stages of negotiations, the parties may have learn from each other at a deeper level, in order to complete agreements. Similarly, the assessment of options does not occur simply at one point, but continues at varying levels of intensity throughout the latter half of the process. We will now discuss each step in the process briefly.

Mutual Education: The process of consensus building depends partly on making sure that the parties know each other and understand each others' perspectives. It is not important that they agree with those perspectives, but it is crucial that they learn to respect the valid experience of others and grasp the reasons why people assert certain views. A parallel part of the process relates to the issues themselves-the history, technical background, scope, etc. An element of this process is making sure that all participants have a relatively similar base of information regarding the issues.

Identification of Values and Interests: Each of the participants brings values grounded in their experience and reflected in the organizations they represent at the table. Each of the individuals and organizations also hold interests that they will seek to advance through the negotiation process. At this stage of consensus building, the facilitators assist a process of digging beneath participants' stated positions to discover their underlying interests. For instance, an environmental advocate might insist on complete clean up of a hazardous waste site so that it will become "clean enough for babies to play in." Embedded in this position an interest in promoting safe drinking water and preventing the potential negative health effects from residual toxics contained in soils. Other parties, although they may disagree with these interests or may support them less vociferously, can negotiate more effectively with concerns for health and safety than they can with the stated position of absolute clean up. There are multiple possible ways to protect health and safety-that may involve something other than clean up to point of safety for babies to play in. One key role for facilitators is to "translate" positions into interests through the process of "reframing."

Identification of Issues to Be Addressed: At the outset of complex negotiations among multiple parties, it is not always apparent which issues will be on the table, or, among the issues that are clearly on the table, which subissues or major concerns will have to be addressed. Early in the process, the facilitator works with the parties to identify their most urgent issues and then to make agreements about the scope of discussions. In addition, the parties may set priorities among the issues and/or a sequence of discussion.

Development of Criteria: Once the parties have identified the issues, the next step is to develop criteria to guide the choice of policies, agreements, resolution of problems, etc. Most often, these criteria are constructed from the combined interests of the parties and, to the extent possible, their values. An example of a criterion might be: "We need to find a way to protect the species and allow reasonable economic development at the same time."

Generation of Options: Based on the criteria and to meet as many interests as possible, the negotiating group engages in generation of multiple options that might solve the problem, prove an acceptable policy, or provide implementable guidelines for regulating an activity. Facilitators encourage generation of more than one or two options, in order to push for more creative solutions. An initial, quite broad set of options may be narrowed somewhat at this stage of group deliberations. The group then initiates a process for assessing the remaining options for legal and technical viability. Some options may be eliminated, changed or combined based on these assessments.

Proposal Development: Once options have been assessed, the group works to develop specific proposals through choosing among options, combining elements from several options, selecting some components and rejecting others. It is at this stage that the toughest negotiations may take place, as parties advocate strongly for their interests, making sure that their needs are met in the emerging proposals.

Final Agreement/Consensus: In the final bargaining process, facilitators ask participants to identify concerns they have with draft proposals. These concerns become the focus for further problem solving and refinement of text that captures agreements. This process of identifying concerns and attempting to build more and consensus in support of the continually refined proposals may be repeated several times and over many months, depending on the complexity of the issues and the depth of disagreement among parties. When this process is successful, the parties eventually arrive at full consensus agreement.

Earlier in the Commission process from 1992-1994, the PAC's role was unclear or confined to providing input from stakeholders into proposal development by other groups within the Commission structure. Later, the Commission gave the PAC the assignment of developing consensus recommendations for their consideration-a much more difficult and intense task. At this point, almost two years into its life, the leadership of the PAC decided to ask for the help of professional facilitators who would guide the consensus building process, working closely with the PAC Chair, Vice Chair, Secretary and the Steering Committee, a smaller fifteen-person groups with balanced representation from PAC members. Several firms bid on a contract to provide facilitation services, and CDR Associates of Boulder, Colorado was selected and worked intensively with the PAC and other Commission groups during the last year and a half of its existence.

Building Relationships: The early days of the PAC were spent in people getting to know each other, mutual education about their perspectives, and learning about the issues involved with regional haze. While members did engage in general discussions about the issues, they made no attempt to craft agreements until the final six months (December 1995 - May 1996), after the major technical studies were completed. In addition to interactions during formal meetings, members also got to know each other through informal receptions and visits to the Grand Canyon (one of the sites to be protected through the regional haze control strategies) and the Navajo Reservation.

In order to prepare the PAC for their role in developing recommendations, in late 1994 the CDR facilitation team provided a workshop on conflict resolution and consensus decision making for PAC members.

Educational Activities: Because the regional haze issue involved complex chemistry, meteorology, economics (etcetera!), and because the members brought a wide range of technical skills and background, a high priority for the group was to provide a series of educational sessions about key issues. Technical experts, many of them members of some of the Commission's technical committees, served as resource people for workshops on air quality and visibility; the chemistry of haze; sources of haze-causing particles; meteorology in the Western United States; the role of fire and burning of biological materials in forest management and agriculture; pollution generated by mobile sources (cars, trucks, etc.); and the role of power-generation plants and potential effects due to expected changes in the utility industry. Tribal representatives also prepared presentations regarding the sovereign status of tribes and the tribal perspectives on development and environmental protection. Other members of the PAC made presentations regarding pollution prevention strategies.

In addition to receiving large amounts of information, and using processes designed by the facilitation team, PAC members discussed their reactions to the information and engaged in early discussion of the possible shape of recommendations, without pushing, at that point, for any agreement. These discussions did help, however, appraise the range of views among members.

Identification of Issues: For the PAC process, the list of issues was partly defined by the sources of pollution themselves, both close to and far from the national parks and wilderness areas. Each of the major sources required discussion and development of recommended strategies for reducing emissions on a regional basis. In addition, the Commission had been directed by Congress to address some specific matters, including protection of existing sources of clean air in the West. Finally, members of the PAC placed issues on the agenda, such as pollution prevention strategies and attention to impacts on tribal lands.

Development of Criteria: One of the first tasks undertaken by the PAC was development of a set of evaluation criteria to guide development of recommendations. While these principles were quite broad, they represented the major interests of all of the sectors represented on the PAC. The Commission developed six broad criteria for assessing different emission management strategies:

The purpose of these criteria was to ensure that all significant impacts of different emission management strategies would be considered in making recommendations for the future.

Generation and Evaluation of Options: The PAC undertook generation of options in several ways. One place where this occurred was in another Commission committee, the Alternatives Assessment Committee. Drawing on input from the PAC and other Commission bodies, this committee developed five main strategies for controlling and reducing haze-causing pollution. They then developed a Request for Proposals aimed at hiring a series of technical contractors to assess the options. In fact, some of the most difficult negotiations in the Commission's process surrounded the specifications of the RFP. The three-volume, approximately 300-page report from the technical study and its assessment of policy alternatives was called Options for Western Vistas. The report provided projections of visibility-impairing air pollution through the year 2040 and evaluated the costs and benefits (in terms of improved visibility) from the various proposed strategies.

Proposal Development: Following the publication of Options for Western Vistas in November, 1995, the PAC had a limited period in which to develop consensus recommendations, essentially until early May, 1996. Clearly, it would have been inefficient to craft proposals in the full PAC. In order to generate proposals in a timely manner, the PAC formed a ten-person Consensus Drafting Group. This subcommittee was carefully chosen to represent all of the major stakeholder groups: environmental concerns, industry, tribal governments, state environmental departments, and federal agencies, with additional resource people who provided technical input. The group met frequently both before and after PAC meetings, in person and on conference calls, to develop proposals for consideration by the full PAC.

Final Agreement/Consensus: The PAC met four times between January and late March to discuss proposals developed by the Consensus Drafting Group. For significant portions of each meeting, PAC members divided into smaller groups organized around the main issues: pollution prevention, stationary sources, mobile sources, area sources (home heating, fire, road dust, etc.), clean air corridors, emissions in and near parks, and emissions from Mexico. Each small group identified problems members had with the proposals and worked to resolve difficulties in fundamental concepts or language.

Between meetings of the PAC, members of the Consensus Drafting Group also formed small subcommittees (mostly 2-3 people) on the main issue areas to take feedback from PAC deliberations and other commentary and continuously refine the proposed recommendations, with assistance from the facilitation team.

Some of these subcommittees were the locus of intense negotiations, especially the group dealing with stationary sources. In that group, environmentalists and utility industry representatives struggled hard to develop agreements regarding the rate of reductions of sulfur dioxide emissions due to additional emissions control measures and the retirement of older power plants in the transport region. This negotiation process continued at CDG and PAC meetings, on conference calls and sidebar meetings right up until the last day of the last meeting of the PAC.

During this period, the facilitation team worked constantly with all of the groups and subcommittees, collecting suggested changes in language and substance, reissuing the text of the recommendations numerous times. There were five major editions of the recommendations, and, for certain sections, many interim versions distributed to subcommittee members or the CDG.

In late March, an interim draft of the recommendations was released for public comment. Over a two-week period, seventeen public meetings were held throughout the West to educate members of the public about haze issues and listen to their comments on the proposed recommendations. In addition, many people provided written feedback. At the end of this process, a group met in Denver to collate all of the input from the public meetings and written input, resulting in a 150-page summary document which was distributed to all PAC members prior to their final meeting at the beginning of May.

At the final meeting in May, subcommittees sifted through all of this feedback to identify major themes and important proposed changes to the recommendations-in addition to continuing to articulate their own concerns. Text changes were made and brought to the full PAC for consideration. At several points in the discussion, stakeholder groups met separately in caucus to discuss outstanding points of disagreement and consider how to resolve those issues. When the groups came back together, the facilitation team continued to prod the full group towards consensus by focusing on concerns, encouraging proposals for changes, and organizing problem solving. During this intense meeting, many members of the PAC worked hard to help each other discover the basis for agreement. By this time the group had developed strong collaborative skills, so that many members of the group helped facilitate the consensus building process on an informal basis.

At several points certain stakeholder groups declared that they had "blocking objections" to the recommendations-unless specific concerns were met. Other groups exhibited remarkable flexibility by adopting requested changes in recommendations and text-but not without struggle. For instance, an important issue concerned the ability of tribal participants to state, in the report, specific challenges about air pollution, while expressing overall support for the recommendations. Ultimately, the PAC agreed to include the tribal perspectives, even though not all members were agreed with the sentiments expressed.

By the end of the May meeting, the PAC had achieved consensus agreement on all of the recommendations. As people rushed to the airport, they shared a profound sense of accomplishment.

The final recommendations from the PAC covered a range of strategies for preserving and improving visibility in the Transport Region. The following is a summary of the conclusions:

The PAC also recommended the creation of an ongoing entity like the Commission to oversee, promote, and support many of the recommendations.

On June 10, 1996, the Grand Canyon Visibility Transport Commission met at the Grand Canyon and approved the recommendations, essentially as produced by the PAC, with only a few minor changes. Following the Commission meeting, members of the PAC and many of the other Commission committees joined together in a catered lunch along the rim of the canyon. On that particular day, visibility was reasonably good, about ninety miles.

The process of the GCVTC, especially within its Public Advisory Committee, represents an example of the application of consensus building to resolution of a complex set of policy issues among a diverse group of stakeholders. As our democracy continues to evolve, we can expect such processes to improve and expand into many other areas of public life.

While negotiations among stakeholders will not displace traditional representative democracy, consensus building processes constitute increasingly sophisticated systems for reconciling competing interests within our diverse society. Winner-take-all voting allows us to ignore the needs of up to 49% of our fellow citizens. Consensus procedures force us to deal with the interests of all groups, even, in some cases, small but significant minorities. We are challenged, in this era, to cope creatively with many forms of diversity. Consensus represents one hopeful means for forging new bonds of unity, while encouraging groups and individuals to assert their unique interests and identities.