Chair: J. Michael Hall

Rapporteur: William Clark

Members: Richard Ball, Wiley Barbour, Abram Chayes, Sheila Jasanoff, Robert Kates, Will Kellogg, Marybeth Long, Clark Miller, Tom Parris, Dan Schrag

The Working Group (WG) drew from the theme background paper by Miller et al. a view of assessment as a communication process. It focused on four groups of questions:

1. Who is communicating with whom through the assessment process? With what other channels of communication do they compete for attention?
2. What is being communicated in the assessment process? How is the selection of what is communicated made from among other possible messages?
3. How credible is the communication conveyed through the assessment process?
4. What is the effect of assessment communications on research agendas, policies and social capacity to cope with the problem of climate change?

The Working Group considered assessment as a process through which communication can occur between:

• scientific communities of different disciplines or nations (assessments as a means of scientific communication);

• scientific communities and communities of policy makers (assessments as direct and indirect advice to decision makers in the public and private sectors);

• scientific communities and the general public (assessments as means of general education).

The Working Group noted that numerous other channels are available for communication among these communities, and that these other channels may complement or compete with formal assessments for attention. These other channels include:

• personal contacts (e.g., coffee room chats; briefings by individual scientists, e.g., Albritton briefings, etc.). 
• professional publications (e.g., an article in Nature; Kellogg, William W. Climate Change and Society: Consequences of Increasing Atmospheric Carbon Dioxide. Westview special study. Boulder, Colorado: Westview Press, 1981). 
• popular publications (e.g., NOAA’s global change educational brochures; Schneider, Stephen. 1989. Global Warming: Are we Entering the Greenhouse Century? San Francisco: Sierra Club). 
• mass media (e.g., the National Academy of Science’s television series on Planet Earth; talk radio shows exposing the "climate conspiracy"; the New York Times).

The Working Group was unaware of any systematic study of the relative importance of the respective channels for communication among particular communities. It discussed its own qualitative impressions of the distributions of communications in the contemporary US dialog on global climate change, while emphasizing that real data on the topic would be more useful than the Working Group’s poorly informed judgment.

Three possible patterns of communication were discussed. First, the Working Group felt that the majority of the communication between science communities and the general public on the question of global climate change almost certainly occurs through the mass media. (Members of the media, of course, might refer to formal assessments in preparing their stories). Second, the WG guessed that most of the communication between science communities and policy making communities probably takes place through personal channels. (Again, the individuals to whom policy makers turn for advice may well have consulted formal assessments in preparing their remarks). Third, scientific communities seem likely to employ a variety of channels in communicating among themselves on issues of climate change. Assessment processes are certainly important for communication among scientific communities on complex issues such as global climate change. It was pointed out, however, that one consequence of the move towards very large, comprehensive assessments such as the IPCC may be a relative or even absolute reduction in the amount of cross-disciplinary communication that such assessments convey. A climate assessment such as that conducted by the National Research Council in 1983 would seem to have involved a significant amount of cross disciplinary communication, both among the dozen or so authors of the assessment, and for readers of the (relatively) small report who almost inevitably would have encountered a variety of disciplinary perspectives though even a cursory skimming. The sheer size of the IPCC documents makes it unlikely that most authors or readers encounter such a variety of disciplinary perspectives. 

 What today is generally referred to as the issue of global climate change has at various times been viewed from the perspective of "energy and climate" or the "CO₂ problem" or "global warming" or the "greenhouse effect." As Thomas Schelling long ago suggested, it could as plausibly have been called the "global water" problem instead. Which of these "frames" is prevalent at any given time within a segment of society seems likely to have a lot to do with how that society thinks about the problem, its possible causes, and the options for dealing with it. In particular, the dominant frame seems likely to influence what kinds of assessments a particular group undertakes and finds relevant. When different national groups see, or frame, the problem in different ways, the problems of building international consensus seem likely to intensify. It is therefore important to understand why one framing rather than another takes hold at particular places and times; what difference this makes for the nature and outcomes of assessments undertaken in those contexts; and the extent to which assessments can in fact shape common frames extending across multiple groups.

 The concept of framing is employed in the social sciences as a means of drawing attention to the processes of selection, emphasis and presentation through which a particular view of an issue or problem comes to dominate other possible ones over particular periods and for particular groups. The framing of an issue in a particular way -- for example as "the CO₂ problem" -- tends to carry with it an implicit choice of what matters (e.g., the release of CO₂ to the atmosphere, rather than "climate change"). Likewise, it points to where solutions are to be sought (e.g., in reducing CO₂ emissions, rather than in reducing vulnerability to climate change). Frames often entrain analogies, as when an issue is viewed as a sporting contest (with winners and losers), or as an evolutionary process (with progress through adaptation and selection of competitive strategies). Alternatively, frames may entail a specific cause-effect model, such as that implied when one speaks of a "drunk driving problem". Frames are social constructions, derived not from nature alone but from which of the many perspectives on a complex nature people choose to adopt. Similarly, frames are stabilized by social institutions that lend them credibility and authority, and that propagate them throughout larger segments of society.

 Two kinds of variation in framing are particularly relevant to the understanding of science / policy relationships in the global environmental arena. One is shifts through time in the frames that shape the thinking of a particular group or nation. Possible explanations for such shifts may appeal to failure of the dominant frame to accommodate contemporary concerns, and replacement by one that does; or to changes in the distribution of power among those groups that define the dominant frame; or to bargaining and negotiation among contesting frame-holders. A second important variation in framing reflects the fact that different groups or nations look at the an issue such as climate change or acid rain in different ways at any particular time. A hope often expressed in discussions of the role of science in managing global environmental risks is that it can bring about convergence among such differing views, and thus help to coordinate the responses of different countries to the problems. Possible explanations for initial differences in frames among countries or other groups include differences in domestic politics; in history; or in culture. The most common explanations for convergence among previously differing national frames appeal to international science, politics, or law as universalizing agents. The WG focused its discussions on the first type of variation, looking at changes in the framing of the climate issue in the United States, and the ways in which assessments were shaped by and in turn shaped such changes.

 The WG explored the science/policy discourse on climate as it was carried out in America’s elite media and national science and government institutions during the period from the mid 1960s to present. (For further documentation see the Overview paper and the Background Paper by Miller et al.)

 It identified three major frame shifts that occurred over the period:

 In seeking to explain these changes in the framing of the science/policy discourse, the WG found it helpful to explore concurrent changes in ongoing "streams" of climate change politics, scientific problem definition, and policy. (These "streams" are consciously, if tentatively, associated with those used by Kingdon and other political scientists to explain the development of political issues and agendas). Salient developments in the "politics" stream were seen to include the active interest in the CO₂ problem by nuclear energy community in the early ‘80s; the embrace of the climate issue by the NGOs in the mid ‘80s; the shift to an international frame of reference for climate discussions after 1987; and the emergence of stakeholders, as opposed to only experts, as legitimate participants in the debate in the mid 90s. The "solution" stream, though less explored by the WG, entailed shifts from thinking about climate change as a pollution problem to one of energy efficiency and, more recently, the beginnings of a shift from prevention thinking to adaptation thinking. The "problem definition" stream, dominated by science through most of the early period of concern with the climate issue, involved one thrust growing from work on weather modification, another from concerns with the global biogeochemical cycles, and a third with long term (natural) climate changes. Only in the mid 80’s did these streams begin to coalesce in the broader conceptualization of "global (environmental) change". In addition, the WG reviewed the record, quoted in the Overview paper to this Workshop, of climate related assessments produced during this period. The WG’s highly tentative account of the relation between these macro changes in the framing of the science / policy discourse and underlying social developments is summarized below: 

1. weather modification / climate variability à domestic energy choices / CO₂: Throughout the 1960s and early 1970s, the connections between science and policy in what would later become the global climate issue were discussed largely within the context of problems of climate variability (e.g., drought) and weather modification. Both of these issues had entrained significant policy debate of their own as a result of technological developments and natural droughts, and provided a modestly hospitable home to the early proponents of CO₂-induced climate change. Science/policy discussions on the CO₂-climate connection languished, however, until the issue was coupled with the much more vigorous and high profile energy policy debates that developed in the wake of "energy crisis" thinking beginning in the mid-1970s. Under the energy impetus, the problem frame within which policy makers encountered the climate issue shifted to energy policy, with a focus on the CO₂ emissions of alternative energy technologies. Assessments followed, reflected, and institutionalized this frame shift. 

The WG postulates that scientific assessments generally take on the most powerful relevant frames current in the policy debate at the time of their inception. Conversely, assessments rarely play a "frame-breaking" role. Frame changes in the science / policy debate on climate occurred largely in response to externally driven changes in the problem, policy, and politics "streams." Selections from these streams were combined into new frames through the work of all manner of entrepreneurs -- scientists, NGOs, private sector interests, and government program managers. This entrepreneurial work is especially influential when more than one frame is available for selection. This may have been the case in the 1985 contest between the old CO₂/climate frame adopted in DoE’s 1985 "CO₂" assessment and the new international global climate frames then being pushed by various NGOs and international institutions (see Franz, 1997).

 Do assessments, then, make a difference in framing? The WG believes they do, and in at least two ways. First, assessments surely play a role in formalizing and institutionalizing frame changes that are well under way for other, largely external, reasons. Assessment reports are frequently cited or tabled in science / policy debates to impart authority and lineage to frame changes that have already occurred. This ceremonial / rhetorical relationship of assessments to frames may complement a second way that assessments seem to matter: through disseminating to a wider audience frame changes that have already occurred. The assessment process is thus usefully viewed as a communication channel through which exogenously-driven frame changes are spread over a larger community. The crucial question of whether, and to what extent, this dissemination function helped to homogenize initially differing frames of various national or sectoral groups was not explored by the Working Group.

3. HOW CREDIBLE IS THE COMMUNICATION CONVEYED THROUGH THE ASSESSMENT PROCESS?

 Choosing to view assessment as a communication process leads naturally to questions of credibility. What makes the communication attempted by an assessment credible to the parties involved? How do these determinants of credibility vary across groups or countries? How important is credibility in determining the effects of assessments on societal responses to issues of global environmental issues?

 The Working Group readily agreed that "credibility" was not was not just about the "facts" of the case and the technical uncertainty associated with them. Rather, credibility is bound up with a host of related questions involving power, authority, and legitimacy.

 Within this broader context, credibility and related properties of assessments are clearly shaped by decisions regarding participation in, and conduct of, the assessment process as well as the content of assessment itself. Moreover, the mode of communication of assessment products – in particular the role and authorship of executive summaries - seems likely to bear on credibility as well. Little solid research on the relative importance of, or interactions among, these determinants of credibility was known to the Working Group.

 It was generally agreed that decisions on who participates in assessments, and how they participate, are among the most immediate ways in which the designers of assessments affect credibility. Interestingly, however, some experience in climate change and other fields suggests that the substantive influence of participants on assessment content may not be the most important pathways through which participation affects credibility. Rather, consistent with the view of assessments as a communication process, much of the significance of participation may lie in the participation itself, and the relationship of respect and concern it establishes among participants. If true, this could have substantial implications for the question of "stakeholder" participation in global and regional assessments of environmental change.

The Working Group also explored evidence that in other fields seeking to link uncertain knowledge to policy, procedures that accord a formal and honorable place for "dissenting opinions" can play a major role in enhancing credibility. Examples given included American experience with national intelligence estimates and Supreme Court rulings. Some also felt that the unusual published dissent of some contributors to the NRC 1991 assessment of policy alternatives for dealing with climate change actually had the effect of enhancing that study’s credibility. In light of this and other experience, it may be appropriate to reexamine the general presumption in the design of most global environmental assessments that they must at all costs avoid dissent.

Finally, based on scholarship in other fields and the Working Group’s collective experience, it seemed almost certain that the factors most important in shaping the credibility of assessment processes differed depending on which communication role an assessment was playing. In other words, decisions on process and participation taken to enhance the credibility of an assessment serving as scientist-to-scientist communication are unlikely to be the same ones that would be taken to enhance the credibility of an assessment serving as scientist-to-policy maker communication. (The potential tradeoffs between scientific credibility and political utility were further explored in Working Group 1). Furthermore, both the literature on risk communication and practical experience suggest that steps taken to enhance the credibility of an assessment process to scientists or high level policy makers are almost certainly neither necessary nor sufficient to ensure credibility to the general public. Finally, scholarship in other fields strongly suggests that there will be significant differences among countries in what makes an assessment process credible in any of its communication modes.

These observations led the Working Group to speculate on the importance of designing explicit credibility strategies to accompany assessment efforts. Such strategies would be a means to avoid implicit assumptions that measures taken to enhance the credibility of assessments as a means of scientist-to-scientist communication would automatically serve to enhance credibility in the minds of policy makers or the general public as well. Elements of a credibility strategy might include: decisions on what to "black-box" versus what to render transparent; the role of standardization of methods or processes; and, as noted above, suitable decisions on participation. Some Working Group members further speculated that the credibility of an assessment process communicating to policy makers and publics would need to be finely tuned to local circumstances and institutions. In a similar vein, some in the WG suspected that the credibility criteria for assessments communicating within particular science communities (e.g., those contributing to IPCC WGI) could be quite different from those applicable to communications among different scientific communities (e.g., between WGI and WGIII). To the extent that these speculations turn out to be true, they raise serious challenges to the credibility of big, integrated "one-size-fits-all" assessments such as the IPCC. Indeed, they suggest that an explicit consideration of credibility strategies might well lead away from single comprehensive assessments towards multiple assessments tailored to the needs – and credibility criteria – of specific "users," clients or participants in the assessment process.

 The assessment process can affect not only how societies think about global environmental change, but also what they do about it. The Working Group discussed three aspects of social action that might be influenced by assessment: research agendas, policy choices, and response capacity. For each, it was easy to find particular instances of influenced action -- at least once the Working Group looked outside the relatively narrow confines of climate. The assessments of the Montreal process, for instance, pushed NOAA to shift resources into the chemistry of CFC substitutes. In the policy domain, a good case can be made that the WMO 1985 ozone assessment (the "Blue Books") proved valuable to policy advocates seeking to put the Vienna Convention into place. And IPCC has certainly been pushing the development of capacity to produce subsequent assessments. More generally, the assessment process frequently serves as a "site" or occasion for managers and decision makers to reexamine their priorities and preferences. If the assessments seldom are in and of themselves sufficient to bring about radical transformations of existing policy agendas, they are almost certainly a necessary contributor to them.

 Moving beyond individually selected cases of "success" towards a predictive theory of assessment effectiveness is nonetheless complex. In part, this is because the Working Group was not able to explore in any systematic fashion the "null hypothesis" cases of big environmental problem areas where few or no assessments exist. In part, it is because any effort to sort out multiple contributing factors to a behavioral change is bound to end ambiguously. Faced with these difficulties for the general case of assessment effects, turned its attention to the specific case of regional climate impact assessments.

 Work on the impacts of climate variability at regional scales is prominent on the contemporary agenda of both scientists and policy makers. As noted earlier, many scientists and policy advisors -- including members of this Working Group -- suspect that regional variability may become central to the next reframing of the climate debate. If so, it is likely to be a highly politicized frame. The need of political leaders for constituents who understand what climate change will mean to them will push for a growing emphasis on regional impact assessments. But the growing consensus on the basic physics and chemistry of climate change means that political attacks on climate science will be more likely than ever to focus on the uncertainties inherent in those impact studies. Are the science and assessment of impacts up to this dual political and technical challenge?

 In a background paper for the 1997 GEA Workshop, Long and Iles (1997) argue that research, policy and social capacity to address climate impacts have remained relatively undeveloped in the American and international history of the climate change issue. Climate impact assessments have been relatively rare, relatively controversial, and relatively undeveloped methodologically over the same period.

Previous efforts to advance regional impact studies – notably DoE’s initial design of a carbon dioxide and climate assessment program in the late 1970s, and SCOPE’s work on the topic in the mid-1980s – both fizzled. Some members of the Group felt that the most recent round of interest was likely to go the same way. Others, however, pointed out that the context for regional impact assessments had improved significantly since those early failures. In particular: 

• Many natural scientists initially resisted the inclusion of impact studies and other "human dimensions" research in climate assessments on disciplinary grounds, or because they feared that controversy over the regional impact studies would implicate and undermine good science elsewhere in the assessment. Today, with improved observations, models, and confidence scientists are much more receptive to regional impact studies. 
• Government program managers in the United States, politicized by the anti-regulatory fervor of the early Reagan Administration, resisted the regional impact research and assessment for fear that these activities would strengthen the case for action on the climate issue. Today, the situation is reversed.  
• NGOs were generally not engaged in the climate debate until the mid-1980s. Those that were often opposed studies or assessments that dealt with adaptation options for fear that if such options were judged to be effective, they would undermine support for emission reduction plans. Today, as part of a general recognition that some climate change is likely to occur no matter what actions are taken, the anti-adaptation bias has somewhat diminished.
• The close collaboration between the assessment and policy communities so essential to good regional impact assessments was rare in the 70s and 80s; it is much more common today. 
• The overall trend towards regional perspectives in work on global climate change work parallels comparable reframings of the debate in other fields of inquiry. 

Despite these positive indications, however, the Working Group still felt that regional impact studies would need both luck and originality to prosper in the late years of the 20^th century. Three considerations were given special attention: 

Having consumed more than its allotted time, said more than it thought it knew, and both amazed and enjoyed itself in the process, the Working Group at this point resurfaced to see what had been going on elsewhere at the workshop.