Policy Sciences 34: 1^33, 2001. ß 2001 Kluwer Academic Publishers. Printed in the Netherlands.

1

Science and the climate change regime RONALD D. BRUNNER Center for Public Policy Research, University of Colorado, Campus Box 333, Boulder, CO 80309-0333

Abstract. Given rapidly increasing losses from extreme climate events, the world community already has a common interest in action to mitigate and adapt to climate change. However, this common interest is not well served through continued promotion of either mandatory (legallybinding) policies or `do nothing' policies by various participants in the regime established by the U.N. Framework Convention on Climate Change. The common interest would be better served by a third way, comprised of voluntary `no regrets' policies that are commensurate with the limited political power of the regime and already have succeeded on small scales in reducing vulnerabilities to extreme climate events and in reducing greenhouse gas emissions. Both mandatory and `do nothing' policies, as well as the regime itself, have depended upon scientists for political support in the past. But scientists might better serve the common interest of the world community through support of a third way in the future.

Introduction The climate change regime was formally established in March 1994 when the ¢ftieth nation state rati¢ed the U.N. Framework Convention on Climate Change, which had been opened for signature at the Earth Summit in Rio in June 1992. The Framework Convention is primarily constitutive, in that it speci¢es who, acting how, should make decisions a¡ecting global climate change. For example, the governing body of the regime is the Conference of the Parties to the Framework Convention, which now includes more than 160 states. The Framework Convention also includes several policies that are outcomes of the constitutive process. The most widely known is Article 4 (2), a non-binding commitment of 36 industrialized countries speci¢ed in Annex I to reduce their emissions of greenhouse gases to 1990 levels by the year 2000. The Kyoto protocol negotiated by the Conference of the Parties in Japan in December 1997 is an incomplete e¡ort to prescribe more stringent, `legallybinding' targets and timetables for emissions reductions. The climate change regime is the world's principal response to the problem of global warming. The problem surfaced in the mid-1980s, when scientists advised governments that increasing concentrations of carbon dioxide (CO2) and other greenhouse gases in the atmosphere could raise the average surface temperature of the Earth, causing climate changes that would disrupt human and natural systems worldwide. They attributed the problem to human activities, especially the burning of fossil fuels, in the industrial age. In 1995, scienti¢c advisors to the Framework Convention projected that a doubling of atmos-

2 pheric concentrations over pre-industrial levels would increase the average surface temperature by 1.0 to 3.5 ³C within a century. Skeptical scientists projected the increase to be about 1.0 to 1.5 ³C. Scienti¢c uncertainties over the causes, extent, and impacts of global warming have impeded consensus and action on global warming policies. Nevertheless, there is some scienti¢c consensus that even a small increase in average surface temperature could trigger non-linear `climate surprises,' such as abrupt shifts in ocean currents, causing loss of life and property on a gigantic scale.1 Meanwhile, regardless of scienti¢c uncertainties, rapidly increasing losses from extreme climate events are already su¤cient to warrant action by the world community. According to estimates by the Worldwatch Institute and Munich Re, the reinsurance company, worldwide losses from storms, £oods, droughts and ¢res reached at least $89 billion for the ¢rst eleven months of 1998. This is well in excess of the $60 billion in losses for 1996, the previous one-year record, and more than the $82.7 billion in in£ation-adjusted losses over the entire decade of the 1980s. Natural disasters in the ¢rst eleven months of 1998 ^ including ice storms in North America, hurricane Mitch in Central America, and £ooding in China and Bangladesh ^ also killed an estimated 32,000 people and displaced more than 300 million.2 Such losses underscore present needs and opportunities for adaptation to extreme climate events, regardless of their causes, by reducing vulnerabilities to similar events. Reducing emissions of greenhouse gases for mitigation of climate change is also prudent insurance, if human activities are indeed causes of climate.3 In short, the world community already has a common interest in action on adaptation and mitigation policies to reduce losses from extreme climate events. This article contends that attempts by participants in the regime to prescribe more stringent, legally-binding targets and timetables for emissions reductions have not succeeded in clarifying and securing the common interest, and are unlikely to succeed unless or until the Conference of the Parties acquires much more power. Meanwhile, it is already prudent for the regime to support voluntary `no regrets' policies that have succeeded on a small scale in reducing vulnerabilities to extreme climate events and emissions of greenhouse gases. Harvesting experience from such policies ^ through evaluation, dissemination, and adaptation of the successes ^ would contribute directly toward the common interest in climate change without compromising other common interests. Voluntary policies like these constitute a third way, between the mandatory policies promoted by some participants in the regime and the `do nothing' policies promoted by others in the regime. Both mandatory and `do nothing' policies, as well as the regime itself, have depended upon scientists for political support in the past. But scientists might better serve the common interest of the world community through support of a third way in the future. The purpose of this article is to assist in opening up a third way by focusing on the past and future roles of scientists in the politics and policies of the regime. For this purpose, the ¢rst section provides an overview of the regime, which serves as an historical basis for an appraisal of the regime's policies in

3 the second section and for the consideration of policy alternatives, including a third way, in the third section. The conclusion highlights the need for scienti¢c leadership in clarifying and securing the common interest. 1. An overview The politics of climate change regime tend to be discounted or ignored in reassuring assessments of the regime's accomplishments and projected capabilities, but are nevertheless essential in understanding them. This section reviews in turn the origins of the Framework Convention, the Framework Convention as constitutive policy, and emissions-reduction targets and timetables as major policy outcomes of the constitutive process. Origins Scientists initiated the process that led to negotiation and rati¢cation of the Framework Convention on Climate Change, but in the process activated the latent political support and opposition of other interest groups. Among the scienti¢c developments that stimulated scientists' concern about global warming as a policy problem were direct measurements of atmospheric concentrations of CO2 at Mauna Loa in Hawaii, con¢rming the Revelle-Suess hypothesis that such concentrations were on the rise; more studies of the climatological record and development of general circulation models of the atmosphere, increasing con¢dence in model-based predictions of global warming; and recognition that other gases, in addition to CO2 , contribute to the greenhouse e¡ect. According to Bodansky (1994: p. 47), `these scienti¢c developments had combined to make the theory of greenhouse warming both more convincing and more urgent' by 1985, but they probably were not su¤cient to stimulate action outside the international research community. Among the factors that catalyzed public concern and transformed global warming into a political issue were `a number of scientists and non-governmental organizations (NGOs) [who] acted as entrepreneurs, promoting the climate change issue through conferences, reports and personal contacts' beginning in earnest in the early 1970s (Bodansky, 1994: p. 48). The conference held at Villach, Austria in October 1985 was among the most important. O¤cially called the International Conference on the Assessment of the Role of Carbon Dioxide and Other Greenhouse Gases in Climate Variations and Associated Impacts, it was supported by various scienti¢c institutions and sought to represent scienti¢c advice independent of governments. The conference concluded that `the understanding of the greenhouse question is su¤ciently developed that scientists and policy-makers should begin an active collaboration to explore the e¡ectiveness of alternative policies and adjustments' (quoted in Bodansky, 1994: p. 48). The Conference recommended action by the U.N. Environment Programme (UNEP), the World Meteorological

4 Organization (WMO), and the International Council of Scienti¢c Unions (ICSU) to `initiate, if deemed necessary, consideration of a global convention, and to advise on further mechanisms and actions required at the national and international levels' (quoted in Boehmer-Christiansen, 1994a: p. 157). The Conference also established an independent Advisory Group on Greenhouse Gases (AGGG) to advise the heads of these organizations for such purposes. The AGGG `remained a major in£uence and organizing force behind the dissemination of the climate threat after 1985' (Boehmer-Christiansen, 1994a: p. 157). Politically, the 1985 Villach conference revealed the extent to which the international network promoting the threat of global warming had expanded beyond climatologists. Several of the organizers of the Conference were deeply involved in energy and policy research and determined to initiate a dialogue with policy makers. In fact, the meeting can be seen as having been called by scientists from energy-poor, pro-nuclear European countries jointly with environmental activists from the U.S.A. and members of UN scienti¢c bureaucracies. (Boehmer-Christiansen, 1994a: p. 156) Many of the energy and policy research establishments had been enlarged or created after the oil price shocks of the 1970s to exploit improved prospects for low-carbon technologies, including nuclear and alternative energy sources. These establishments were attracted to global warming in the 1980s as an opportunity to sustain themselves amidst declining public funding for energy and policy research, and to push for governmental support of nuclear and alternative energy technologies. The conference held in Toronto in June 1988 was `the high-water mark of policy declarations on global warming' according to Bodansky (1994: p. 50). O¤cially known as the World Conference on the Changing Atmosphere: Implications for Global Security, it was organized with the assistance of several members of the AGGG. It had far more in£uence than other non-governmental conferences before or since for a variety of reasons: It was o¤cially sponsored by the government of Canada, it included high government o¤cials, and its timing was propitious. It occurred after the 1987 discovery of the ozone hole, which demonstrated the magnitude of the unintended e¡ects of human activities on the atmosphere; and in the midst of an extreme heat wave and drought in the U.S., which turned scienti¢c testimony on global warming into front-page news (Ungar, 1995: pp. 446^448). In addition, it was in£uential because the Toronto Conference Statement called for a 20% reduction in CO2 emissions, a provocative demand which `caused considerable unease among governments and industry' (Boehmer-Christiansen, 1994a: p. 157). Bodansky (1994: p. 49) observes that `because of its non-governmental character, the Conference Statement ^ including the 20 percent reduction target ^ was not a negotiated document and was not binding on anyone. It was drafted by a committee composed mostly of

5 environmentalists and discussed in less than a day.' 4 But environmentalists promoting concerns about global warming merely `had a head start; opponents in industry and government took longer to mobilize' (Bodansky, 1994: p. 50). The Toronto Conference Statement catalyzed opposition from o¤cial and non-o¤cial groups who perceived their interests to be threatened by the proposed 20% reduction target or by earlier calls for action on global warming. For example, By June 1987, the U.S. State Department had become unhappy about the AGGG as representing little more than `free wheeling academics.' Governmental bodies therefore began to wrest the policy initiative from the AGGG network by replacing parts of it and extending others to include governmental research bodies, especially those close to WMO. (Boehmer-Christiansen, 1994b: p. 187) Moreover, governmental agencies representing broader interests, including economic policies, increasingly became involved. Oil, coal, and other industries organized and lobbied against carbon or energy taxes proposed to reduce emissions of greenhouse gases. In the U.S., for example, the Global Change Coalition spun o¡ from the National Association of Manufacturers in 1989, and continues to lobby on behalf of a consortium of utility, coal, oil, and automobile companies (Fialka, 1997). By the late 1980s, some governments, including the U.S. and members of the Organization of Petroleum Exporting Countries (OPEC), formed an alliance to oppose carbon or energy taxes. For the U.S. at least, research was an acceptable substitute for action to change policies. Citing scienti¢c uncertainties, the head of the Environmental Protection Agency announced at an international conference in 1989 that `the United States is substantially increasing its budgets for scienti¢c research into the causes and consequences of climate change' (quoted in Boehmer-Christiansen, 1994b: p. 190). The chief bene¢ciary was the U.S. Global Change Research Program, a new interagency initiative (Pielke, 1995). Some other nations pressed for more stringent action, particularly in Europe, where global warming was still viewed rather narrowly as an environmental issue. In 1988, shortly before the Toronto Conference, the U.S. and allied governments responded to the expanding political issue by requesting that WMO and UNEP establish an organization that became the Intergovernmental Panel on Climate Change (IPCC), a scienti¢c advisory organization that e¡ectively replaced the AGGG. `Through WMO, governments gained the power to veto participants to the IPCC and in£uence its brief' (Boehmer-Christiansen, 1994b: p. 189). But `there was considerable bitterness in some quarters about the fate of the AGGG, which was disbanded under pressure from the U.S. State Department' (Boehmer-Christiansen, 1994b: p. 189). The IPCC was the ¢rst formal, intergovernmental response to concerns about global warming. It met for the ¢rst time in November 1988, and prepared its ¢rst assessment for the

6 Second World Climate Conference in November 1990. It updated that assessment for the U.N. Conference on Environment and Development, the Earth Summit in Rio in 1992. `This 1992 update resulted in a generally more guarded statement of the climate threat than that of 1990' (Boehmer-Christiansen, 1994b: p. 191). The Framework Convention on Climate Change was negotiated by diplomats through the International Negotiation Committee (INC). The Committee was established by the U.N. General Assembly in December 1990 as an alternative to the WMO and UNEP. `The INC was to take into account, inter alia , the work of the IPCC, though few direct links if any were actually formed' (Boehmer-Christiansen, 1994: p. 191). The formal mandate of the IPCC, as stated in U.N. General Assembly Res. 43/53 (1988), was to provide `internationally coordinated assessments of the magnitude, timing and potential environmental and socio-economic impact of climate change and realistic response strategies.' (Response strategies were e¡ectively eliminated when the IPCC revised the objectives of Working Group III prior to the second assessment.) But the informal mandate was `in part to reassert governmental control and supervision over what was becoming an increasingly prominent political issue' (Bodansky, 1994: p. 51). Thus the IPCC was caught up in political divisions within and among the governments of the U.S., European and other industrialized nations, and the less developed nations ^ divisions over energy, economic development, equity, and other issues that had become entangled with global warming. As O'Riordan and Ja«ger (1996: p. 346) put it, `The science of climatic change ... is increasingly being drawn into politically supported analytical structures, to the point where ``climate change science'' is not always separately identi¢able from the political process that shapes it.' Boehmer-Christiansen (1994b: p. 195) claims that scientists in the aggregate made the expedient adaptation to the political situation in which they found themselves, and for which they shared some responsibility. To protect its own interests, science had to respond to a new international context in which most governments needed time and therefore welcomed uncertainties. Governmental interest was in funding more research rather than enforcing changes in energy policy. Scienti¢c institutions nationally and globally could not reject this o¡er. If forced to choose, the interests of science cannot but lie with research rather than policy change. Policy neutrality was therefore becoming increasingly attractive. Moreover, Boehmer-Christiansen (1994b: p. 192) claims, scientists were e¡ectively neutralized in the political balance of power: `The policy outcome so far ... is a weak, research intensive framework treaty which re£ects a political balance of power rather than any ¢rm direction derived from science.' These claims are corroborated by a review of the Framework Convention as constitutive policy.

7 The framework convention Like its origins, the constitutive foundations of the climate change regime in the Framework Convention are essential in understanding its limited accomplishments and future prospects. This section reviews the Framework Convention as constitutive policy ^ in other words, policy regarding how policies should be made. It begins with participation in the regime.5 The Framework Convention authorizes participation in the Conference of Parties, the governing body, by national governments that haved rati¢ed the Framework Convention and thereby formally accepted certain responsibilities under it. As noted above, more than 160 national governments have done so. Thus participation is rather inclusive of the world's national governments, but they are not the only participants in the regime. For example, 13 intergovernmental organizations and 25 non-governmental organizations, as well as 116 governments, participated in the line-by-line approval of the ¢nal draft of the IPCC Second Assessment Synthesis at the IPCC's eleventh session in Rome in December 1995 (IPCC, 1995: section 1.1). It is not clear how well the people of the world generally are represented in the policy processes of the climate change regime, or to what extent the people generally accept any responsibility in connection with it. It is clear that the climate change regime cannot stabilize concentrations of greenhouse gases in the atmosphere, its formal objective, without persuading or coercing people around the world to participate in mitigation policies to reduce emissions of greenhouse gases.6 Similarly, public participation is a necessary condition for e¡ective adaptation policies to reduce vulnerabilities to extreme climate events, whether anthropogenic or natural. Thus it is important to ask, how do mitigation and adaptation policies play in the Peorias of the world? Answers are mixed where the question has been asked. For example, on the eve of the Conference of the Parties meeting in Kyoto in December 1997, 65% of Americans who responded to a New York Times poll agreed that `the U.S. [should] take steps now to cut emissions of greenhouse gases' rather than `wait for many countries to agree to take steps together' (Cushman, 1997). But the intensity of support is quite low. In the same poll, when asked what was the most important problem facing the country, just 1% of all respondents mentioned the environment; and air pollution, water pollution, and pollution in general were mentioned more often than global warming as the most important environmental problem. Moreover, words are not necessarily in harmony with deeds. Fifty percent of all respondents in the poll said they favored the imposition of energy e¤ciency standards on manufacturers of cars, appliances, and the like, in order to reduce emissions. But at the same time, Americans in record numbers were buying the gas-guzzling light trucks (including sports utility vehicles) that are expected to be the fastestgrowing source of greenhouse gases in the U.S. over the next decade (Bradsher, 1997). The Framework Convention on Climate Change does not represent a robust consensus on basic policies, although it does establish an institution that may

8 be used to develop such a consensus incrementally over a number of years.7 The Framework Convention does express multiple objectives and constraints, ambiguous and often incompatible, re£ecting the plurality of interests represented in the regime. Article 2 declares that `The ultimate objective of this Convention ... is to achieve ... stabilization of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system.' What is a `dangerous' level of concentrations is unspeci¢ed and unspeci¢able with any con¢dence, except perhaps in retrospect when manifest dangers indicate that the level already has been exceeded. Moreover, manifest dangers in some regions might or might not be o¡set by manifest bene¢ts in others. Article 2 continues: `Such a level should be achieved within a time frame su¤cient to allow ecosystems to adapt naturally to climate change, to ensure that food production is not threatened and to enable economic development to proceed in a sustainable manner.' `Development' is highly ambiguous and incompatible with the ultimate objective to the extent that it means larger populations, and higher net emissions of greenhouse gases per capita, as assumed in the IPCC's scenarios. Moreover, Article 4 (7) makes it clear that `economic and social development and poverty eradication are the ¢rst and overriding priorities of the developing country Parties.' Developing countries are exempt from `the aim of returning individually or jointly to their 1990 levels of these anthropogenic emissions' as speci¢ed in Article 4 (2), even though their emissions are expected to exceed the emissions of industrialized countries in a decade or two. To protect their own economic, political, and other interests, the industrialized countries speci¢ed in Annex I accepted Article 4 (2) only as an aim, a voluntary commitment, and retained the £exibility to pursue it by di¡erent national policies. The Framework Convention `is neutral regarding policy options. .. . As a result, states have maximum £exibility in designing response strategies' (Bodansky, 1995: pp. 429^430). Other functional and territorial interests to be given full consideration are cited in Article 4 (5)^(10), ranging from `Small island countries' and `Countries with low-lying coastal areas' to `Parties with economies that are highly dependent on income generated from the production, processing and export, and/or consumption of fossil fuels' ^ including members of OPEC. Evidently, identi¢cations with the emerging climate change regime are weak relative to national and other identi¢cations the Parties brought to the Framework Convention. Within these multiple objectives and constraints, the general commitment and responsibility of all Parties to the Framework Convention is to research and report on the situation, but not to change the situation.8 In Bodansky's (1995: pp. 435^436) summary of Articles 4 (1), 5, 6, and 12 (1), Each country must develop a national inventory of its anthropogenic emissions by sources and removals by sinks of [greenhouse gases]; formulate and implement national programs to mitigate and adapt to climate change; report to the FCCC's Conference of the Parties on its national inventories

9 and implementation steps; and promote and cooperate in scienti¢c research, exchange of information, education, training, and public awareness related to climate change. Eventually the general commitment might support action that results in signi¢cant material changes in the situation, such as reduced emissions or reduced vulnerability to climate impacts. But the general commitment so far has substituted symbols for action. The general commitment is research intensive at least until satisfactory national inventories and implementation steps are in place. But even research is an interest balanced by a complementary interest in action, as expressed in the precautionary principle in Article 3(3): `The Parties should take precautionary measures to anticipate, prevent or minimize the causes of climate change and mitigate its adverse e¡ects. Where there are threats of serious or irreversible damage, lack of full scienti¢c certainty should not be used as a reason for postponing such measures... .' Similarly, the preamble to the Convention acknowledges that `steps required to understand and address climate change will be...most e¡ective if they are based on relevant scienti¢c, technical and economic considerations and continually re-evaluated in the light of new ¢ndings in these areas... . ' This promotes continuous research. But the preamble also acknowledges that `various actions to address climate can be justi¢ed economically in their own right... . ' This promotes action on `no regrets' policies that make sense whether or not anthropogenic climate change is a signi¢cant problem. The second-best connotations of the term `no regrets' are unwarranted, because many mitigation and adaptation policies cannot be justi¢ed by climate change concerns alone. Thus Rayner and Malone (1997: p. 333) urge policymakers to `Incorporate climate change concerns into other, more immediate issues such as employment, defence, economic development and public health.' The Framework Convention establishes centralized structures to coordinate the speci¢c and general commitments of the Parties: The Conference of the Parties (Article 7), a Secretariat (Article 8), a Subsidiary Body for Scienti¢c and Technological Advice (Article 9), a Subsidiary Body for Implementation (Article 10), and a Financial Mechanism (Article 11). There was no agreement in the negotiations on a mechanism to address questions of implementation, including non-compliance with the voluntary commitments in Article 4 (2). `Instead, the FCCC [in Article 13] merely calls on the parties to ``consider'' establishing a ``multilateral consultative process'' for the ``resolution of questions'' concerning implementation' (Bodansky, 1995: pp. 446^447). Thus the regime has no authorized structures or procedures for adjudicating allegations of non-compliance, or for sanctioning non-compliance by any Party. The resources allocated to the Conference of the Parties for sanctioning purposes are quite limited. Self-reporting and international review of greenhouse gas emissions and removals by Annex I countries serve informally as the main sanctioning mechanism for the voluntary commitments in Article 4 (2). `Although [reporting and review] do not guarantee meaningful action, they put

10 pressure on states by holding them up to domestic and international scrutiny' (Bodansky, 1995: p. 444). But any Party not particularly concerned about domestic or international pressure may report non-compliance at a cost easily ignored. Other Parties may minimize pressure by self-reporting of the `1990 level' or inventory data to give the appearance of more compliance. (Parties in fact have chosen both options, or failed to report altogether, as detailed below.) In any case, the Conference of the Parties lacks the power to force Parties to comply with the speci¢c and general commitments of the Framework Convention, even if the Conference had the political will to do so. Article 7(3) of the Framework Convention defers rules of procedure, including voting, within the Conference of the Parties. This unresolved issue is an old one. In March 1989, `The Hague Conference Declaration made the radical proposal that countries develop ``new institutional authority'' involving nonunanimous decision-making ^ in e¡ect, a partial renunciation of sovereignty' (Bodansky, 1994: p. 52; Ja«ger and O'Riordan, 1996: p. 26). This radical proposal was quickly buried and largely forgotten. Moreover, members of OPEC initially insisted on unanimity, which is equivalent to a veto for each Party on every issue. Until the voting issue is resolved, the default rule is broad consensus, which permits minorities to block majorities ^ even majorities attempting to clarify and secure the stabilization of concentrations of greenhouse gases in the atmosphere, the regime's objective. In general, the common interest of a community is better served by majority or super-majority rule (Goodin, 1996). The lack of majority or super-majority rule is yet another indication of the weakness of the climate change regime. In short, the Framework Convention does not appear to be a prescription, but at most a candidate for prescription. It becomes a prescription only if the Conference of the Parties acquires the political power necessary to enforce policies with severe sanctions against challengers. Until then, and regardless of its formal enactments, the regime is e¡ectively limited to voluntary policies that leave the choice of compliance or non-compliance to the individual Parties. The most signi¢cant voluntary policies are `no regrets' policies. Indeed, as Ungar (1995: p. 448) suggests, acceptance of the Framework Convention was predicated on a `no regrets' strategy. The weakness of the regime is in proportion to the limited commitments of the Parties, but not in proportion to the magnitude of the task implied by its formal objective, stabilization of concentrations of greenhouse gases in the atmosphere. Stabilization requires major changes in energy, land use, and other policies in the public and private sectors worldwide. Policy outcomes The principal policy outcomes of the climate change regime corroborate its weakness as a matter of deeds, not merely of words in the Framework Convention.9 Article 4 (2), as previously noted, is a policy to reduce greenhouse gas emissions to 1990 levels by 2000 in Annex I countries through voluntary and

11 £exible means at the national level. This policy acknowledges the risk of anthropogenic climate change, but is otherwise largely independent of research in the biological or geosciences. The 1990 levels imply reductions far short of the reductions of over 60% thought necessary for stabilization in IPCC reports at least since 1990 (Boehmer-Christiansen, 1994b: p. 191). Article 4 (2) re£ects a minimal political consensus reached at an unsustained peak of concern about climate change. Moreover, there is no de¢nite policy on adaptation, despite scienti¢c projections that a doubling of pre-industrial concentrations of greenhouses gases in the atmosphere is likely to be unavoidable under any plausible policy scenario.10 The exemption of developing countries and avoidance of binding commitments by industrial countries indicate the subordination of any common interest in mitigation and adaptation to other interests. Each of the 36 Annex I countries is supposed to comply with the speci¢c commitments of Article 4 (2) and the general reporting commitments of the Framework Convention. Compliance has been quite limited, however. In 1996 the U.S. General Accounting O¤ce (1996: p. 4) found that `The incomplete, unreliable, and inconsistent data on emissions prevent a complete assessment of Annex I countries' e¡orts to limit greenhouse gas emissions to 1990 levels by 2000.' Eight countries did not provide projections for methane or nitrous oxide. `Because these gases come from many sources and are nontoxic, little e¡ort has been given to measuring their emissions' (U.S. GAO, 1996: p. 5). Canada reported uncertainties of plus or minus 30% (at a 90% con¢dence level) for methane emissions, and plus or minus 40% (at an 85% con¢dence level) for nitrous oxide emissions. Inconsistencies sometimes arose from the self-reporting of adjusted rather than actual 1990 levels as an emissions target. For example, `Denmark adjusted its 1990 inventory level upward to show what emissions would have been if imported hydroelectric power had been generated domestically with fossil fuels' (U.S. GAO, 1996: p. 5). With this adjustment, Denmark may achieve formal compliance even if its emissions exceed the actual 1990 level. Similarly, `France and Japan [came] up with ``the guileful idea of stabilizing emissions per capita , which will allow them to increase emissions by the same amount as their population grows'' ' (Flavin and Tunali, 1995: p. 12). Using data on CO2 emissions, which are relatively reliable, GAO also found little voluntary compliance with Article 4 (2). `The projections by the Annex I countries themselves indicate that only 7 of the 24 countries that provided point estimates of carbon dioxide emissions in 2000 project that they can hold emissions near or below 1990 levels. For the remaining countries, the increases over the 1990 inventory levels ranged from 1.7 percent to 28.8 percent' (U.S. GAO, 1996: p. 6 and Table 1). Of the 7 countries, 4 are part of the former Soviet bloc that experienced deep and abrupt economic decline after the end of the Cold War. A ¢fth, Germany, reduced overall emissions by incorporating East Germany from the Soviet bloc. The other two are small countries, the Netherlands and Switzerland. The United Kingdom held the projected increase in emissions down to 1.7% for reasons unrelated to climate change ^ the

12 privatization of utilities that led to a shift from coal and oil to natural gas consumption. If international scrutiny of non-reporting or reported non-compliance in the other Annex I countries made any di¡erence in behavior, the di¡erence has not been widely reported. The Framework Convention called for an appraisal of the adequacy of the Article 4 (2) target and timetable at the ¢rst session of the Conference of the Parties, which met in Berlin in March and April 1995. In the Berlin Mandate, the Conference concluded that the existing target and timetable were inadequate, and agreed `to begin a process to enable it to take appropriate action for the period beyond 2000, including the strengthening of the commitments of Annex I Parties in Article 4, paragraphs 2(a) and (b), through the adoption of a protocol or another legal instrument.' 11 `Strengthening' was understood to include more stringent quanti¢ed emission reduction objectives within speci¢ed time frames, as well as legally-binding commitments to them. The diagnosis, evidently, was that the target and timetable had not been stringent enough, and the commitments were merely voluntary. The Berlin Mandate came less than three years after the Framework Convention was opened for signature, less than a year after it went into e¡ect, and nearly ¢ve years before the target year 2000. Nevertheless, the conclusion that the initial target and timetable were inadequate has been supported by subsequent evaluations. The diagnosis of the inadequacies did not go very far, however. The target and timetable had not been stringent enough, and the commitments were merely voluntary, because other interests prevailed over reductions in greenhouse gas emissions in the domestic politics of the Parties. In an independent appraisal, the U.S. GAO (1996: p. 2) found that factors such as economic growth, population growth, fuel prices, and energy e¤ciency a¡ect trends in energy use, thereby in£uencing trends in greenhouse gas emissions. For example, higher-than-expected economic and population growth and lower fuel prices resulting in higher energy use will probably prevent the United States and Canada from reaching the goal. In the U.S., the Clinton Administration was unable to raise fuel prices through the broad-based BTU tax proposed in its ¢rst budget in 1993, despite Democratic majorities in Congress, and was unwilling to curb economic and population growth for any reason, including emissions reductions. The Clinton Administration also was unable to avoid deep cuts in funding the U.S. Climate Change Action Plan, which was announced in October 1993 to meet U.S. commitments under Article 4 (2). Emissions reductions have been blocked for similar reasons in other countries. `The basic problem is that few countries have been willing to make di¤cult political decisions to limit emissions' (Bodansky, 1995: p. 438; see also Muller, 1996 and Paarlberg, 1996). Contrary to experience under Article 4 (2), the diagnosis in the Berlin Mandate simply presumed that the Parties would make the di¤cult domestic political decisions necessary to limit emissions, rather than protect their other interests. Without this presumption, it

13 would be di¤cult to sustain the e¡ort to prescribe such targets and timetables in the Kyoto protocol. The o¤cial source of intelligence for the Kyoto protocol was the IPCC's second assessment in 1995, especially the summaries for policymakers prepared by the IPCC's three working groups and its Synthesis, all of which were approved line-by-line with representatives of governmental and non-governmental interests.12 According to the Synthesis (section 1.10), `Decisions with respect to Article 2 [on the Objective] of the FCCC involve three distinct but interrelated choices: stabilization level, net emissions pathway and mitigation technologies and policies. [This] report presents available scienti¢c and technical information on these three choices.' The Synthesis takes a national and international perspective: It presumes that the Conference of the Parties will decide the stabilization level and the net emissions pathway to that level for the world as a whole, and that national governments will decide the technologies and policies necessary to travel that pathway. At the same time, however, the Synthesis recognizes that multiple factors, as they combine in particular circumstances, must be taken into account in choosing adaptation and mitigation measures and the stabilization level. (The stabilization level depends upon the vulnerability of particular systems, which is a function of their sensitivity to climate change and inability to adapt.) For example: à

à

à

à

`The vulnerability of human health and socioeconomic systems ^ and, to a lesser extent, ecological systems ^ depends upon economic circumstances and institutional infrastructure.' (section 3.3) `The e¤cacy and cost-e¡ective use of adaptation strategies will depend upon the availability of ¢nancial resources, technology transfer, and cultural, educational, managerial, institutional, legal, and regulatory practices, both domestic and international in scope.' (section 3.18) `The optimum mix of policies will vary from country to country, depending upon their energy markets, economic considerations, political structure and societal receptiveness.' (section 5.13) `Because costs [of di¡erent energy technologies] vary by location and application, the wide variety of circumstances creates initial opportunities for new technologies to enter the market. Deeper understanding of the opportunities for emissions reductions would require more detailed analysis of options, taking into account local conditions.' (section 5.10)

Notice that the particular circumstances are not only international and national but also local in scope. The relevant factors may combine di¡erently within and across countries, as well as between them. No country is entirely isolated or perfectly homogenous with respect to the relevant factors. If rational decisions are contingent upon particular circumstances, climate change is not `an irreducibly global problem' as claimed in the Synthesis (section 1.9). Instead, mitigation and adaptation must be pursued through modi¢cations in myriad existing policies and policy processes that are localized

14 in villages, towns, cities, or regions around the world, or specialized to public health, insurance, transportation or other functions that often bridge geographic areas. These localized and specialized policies and policy processes are best understood by participants in them. The participants are likely to resist national or international climate policies that make little or no sense from their localized or specialized perspectives, even if they are favorably predisposed toward mitigation or adaptation goals (Taubes, 1997). Who should decide under these circumstances? The Synthesis avoids a direct answer to this constitutive question, despite its relevance to the Objective in Article 2 of the Framework Convention. But consider this hint of technocratic hubris in the Synthesis : `The degree to which technical potential and cost-e¡ectiveness are realized is dependent on initiatives to counter lack of information and overcome cultural, institutional, legal, ¢nancial and economic barriers which can hinder di¡usion of technology or behavioural changes' (section 5.3). This is perhaps the most direct recognition of politics, the exercise of power, in the Synthesis. It is not very direct at all. The Synthesis and reports like it serve primarily the promotional needs of the climate change regime, not the intelligence needs of localized and specialized policymakers who control the relevant policies. Such reports promote the regime itself, in competition with localized and specialized policymakers, when the reports presume that the important decisions will be made at the international and national levels. Such reports promote the investment of limited resources in climate change concerns, in competition with other human concerns, when they predict signi¢cant and adverse impacts from anthropogenic climate change.13 Such reports promote predictive research, when they reinforce the expectation that the reduction of scienti¢c uncertainties will pay o¡ for policy purposes, without appraising previous predictive research according to policy purposes.14 Finally, such reports promote in action when they deemphasize opportunities for `no regrets' actions that already make sense, regardless of present uncertainties about climate change. The Synthesis notes that `signi¢cant ``no regrets'' opportunities are available in most countries' but emphasizes `rationales for actions beyond ``no regrets'' ' (section 8.2). Actions beyond `no regrets' presume what national and international authorities have avoided, the exercise of power. In contrast, `no regrets' actions may be taken voluntarily by localized or specialized policymakers. Recent intelligence and promotional activities have done little to strengthen targets and timetables for emissions reductions. Even after subsequent meetings of the Conference of the Parties, the Kyoto protocol is still only a candidate for prescription: It defers `appropriate and e¡ective' sanctioning mechanisms to deal with non-compliance by the Parties, and it defers formal requirements to comply with targets and timetables for another decade at least. Thirty-eight industrial countries are formally required to reduce the CO2-equivalent of emissions of six greenhouse gases by 5% to 8% below 1990 levels between 2008 and 2012. Once again these formal requirements re£ect political, not scienti¢c, considerations. From a scienti¢c perspective, reductions of this magnitude

15 would merely slow the accumulation of greenhouse gases in the atmosphere. According to one scientist, ` ``it might take another 30 Kyotos over the next century'' to cut global warming down to size' (quoted in Malako¡, 1997). Moreover, even if the reductions in the Kyoto protocol were achieved, they probably would be overwhelmed by increases from developing countries. The protocol will not enter into force until rati¢cation by enough Annex I countries to account for at least 55% of their combined emissions in 1990. The U.S. alone accounts for about 38% of these emissions, but leaders of the Senate declared the Kyoto protocol `dead on arrival' (Dewar and Sullivan, 1997). The protocol is not expected to be submitted to the Senate until after the elections in 2000 at the earliest (Cushman, 1998). Even if a completed Kyoto protocol is eventually rati¢ed, an international reporting system to verify net changes in greenhouse gas emissions would be necessary to enforce and appraise any mandatory policies included within it. National compliance with legally-binding targets and timetables is not the only veri¢cation problem. An emissions trading system would depend upon veri¢cation of compliance with the permits traded; a clean development mechanism would depend upon veri¢cation of emissions reductions claimed by certi¢ed joint-implementation projects.15 According to the former Chairman of the IPCC, `The protocol refers to work by the IPCC to resolve this issue before the next conference of parties. It is, however, not clear how to devise satisfactory methods to achieve what is envisaged in the protocol.' 16 It is clear that the need for an international reporting system has prompted announcement of a major overhaul of the U.S. Global Change Research Program to focus on the major scienti¢c uncertainty, how much carbon is stored and released by the world's forests (Schmidt, 1998; Kaiser, 1998a; 1998b; Lawler, 1998). In summary, by establishing the IPCC in 1988, governments gained control over independent atmospheric scientists who had begun to demand action on climate change. Then in negotiating provisions of the Framework Convention, governments obtained authority for various economic, political, and other priority interests, and protected them from control by the Conference of the Parties. Finally, in implementing emissions-reductions policies in Article 4 (2) and negotiating the Kyoto protocol, nearly all Parties to the Framework Convention have subordinated climate change to their other interests. Over the last decade, the pattern has been to substitute symbols for action on climate change. The symbols include scienti¢c assessments by the IPCC, international negotiations by the Conference of the Parties, and national reports by individual Parties. The neglect of action leaves the world with little material progress on climate change, as gauged even by the regime's modest policy goals. 2. An appraisal The signi¢cance of these events depends upon the outcomes selected and the standpoint applied to appraise them. For example, one scholar selected the

16 integration of scienti¢c research through comprehensive computer models as the principal outcome of the regime, and appraised that as a success from the standpoint of building an epistemic community ^ one in which members share basic assumptions about the problem. In particular, Edwards (1996: p. 150), contends that `the emergence of such [an epistemic] community is one major reason why global change has reached the political agenda of governments, and thus that comprehensive model-building serves an all-important political purpose even if it does not and perhaps cannot serve the immediate needs of policymakers.' This section appraises the quest for legally-binding targets and timetables and related policies of the regime as failures from a common interest standpoint, and emphasizes the political factors that account for them. Before turning to the policy failures, it must be acknowledged that the regime has succeeded in promoting climate change to a place on the political agendas of governments. This is consistent with the common interest of the world community, despite scienti¢c uncertainties, so long as climate change remains a credible and signi¢cant threat to human and natural systems. International negotiations and national reports remind the world that it has a policy problem, even if it is not a globally-irreducible problem to be addressed exclusively from the global level. Scienti¢c predictions and assessments remind the world community that more extreme climate events are possible, even if scienti¢c uncertainties are misused as justi¢cation for inaction. Such reminders appear to help integrate and intensify the perspectives of those who are favorably predisposed to accept them, and therefore contribute toward solidarity and cooperation within the regime's epistemic core and peripheral communities over the longterm. However, such reminders appear to have little e¡ect on others outside who must be e¡ectively included if the regime is to achieve its policy and ultimate objectives eventually. Furthermore, some of the non-climate interests authorized and protected under the Framework Convention cannot be easily dismissed as special interests. For developing countries, the regime authorizes the priority of economic and social development and poverty eradication goals. For industrial countries, the regime authorizes £exibility in responding to the demands of diverse constituencies. Politicians are supposed to count votes, if not campaign contributions, where they are accountable to the public through periodic elections. Moreover, the opposition to major changes in policy is not limited to corporate interests. In the U.S., for example, the opposition may be led by the Global Change Coalition, but the opposition also includes many Americans who would resist large increases in the cost of gasoline or sports utility vehicles ^ if the federal gasoline tax were sharply increased, or if sports utility vehicles were reclassi¢ed as passenger vehicles, subjecting their manufacturers to steep ¢nes for failing to meet more stringent fuel e¤ciency standards. Major changes in policy typically cannot and normally should not go far beyond the consent of the governed in democracies. But the common interest of the world community also includes prudent action to reduce losses from extreme climate events through adaptation and mitiga-

17 tion. This is the legitimate justi¢cation of the regime. From this standpoint there are at least three principal policy failures: à

à

à

The regime has not realized its own very modest mitigation goals, despite its major commitment of resources to the quest for legally-binding targets and timetables and other mandatory policies. The regime has largely neglected adaptation as a policy problem, despite recent increases in the vulnerability of many communities to extreme climate events and IPCC projections of further increases in vulnerability. The regime has failed to open the policy process to policy alternatives, despite the failed quest for mandatory policies, acknowledged needs for adaptation policies, and acknowledged opportunities for voluntary `no regrets' policies.

`No regrets' mitigation and adaptation policies are consistent with many interests formulated in the Framework Convention and sought in policy outcomes under the Convention. Indeed, such policies would integrate both climate and non-climate common interests. There are other good reasons to take `no regrets' policies seriously as policy alternatives. As argued below, action on `no regrets' policies is feasible and e¡ective: They already have been implemented voluntarily and successfully on a small scale. Action on `no regrets' policies is prudent if not necessary, at least until the regime acquires power su¤cient to prescribe and enforce the mandatory or legally-binding alternatives. Finally, action on `no regrets' policies is a responsibility of the principal interests represented in the regime, including the opposition, if their professed aims are to be taken seriously. The professed aims may be exposed as rationalizations for special interests through persistent indi¡erence or opposition to `no regrets' policies that are economically more e¤cient, environmentally more e¡ective, and politically more feasible than either the mandatory policies promoted by some participants in the regime or the `do nothing' policies promoted by others in the regime. Explanation of these failures is important for purposes of projecting the probable consequences of reliance on current or alternative policies in the future. The failure of the quest for legally-binding targets and timetables or other mandatory policies may be explained most directly and obviously by the fact that governments withheld from the Conference of the Parties the power necessary to impose changes in energy, land use, and other relevant policies worldwide ^ including both the public policies of the governments themselves and the private policies of their constituents. The governments limited the power of the Conference of the Parties to authority and other assets to research climate change through the IPCC and associated research programs, and to report on climate change through o¤cial and non-o¤cial channels. Under these circumstances, the quest for legally-binding or mandatory policies is primarily an exercise in pretended power. The quest may help build the core epistemic community of the regime. However, the quest may also falsely reassure

18 others who are sympathetic to the regime's objectives and falsely threaten others who are not sympathetic. Neither of these promotional outcomes is conducive to prescription and action. But the weak power position of the regime does not explain the regime's neglect of adaptation policies or voluntary `no regrets' policies. The most direct explanation for these failures is that the principal interests represented in the regime are better served by the quest for mandatory policies alone. The quest gives national politicians, beleaguered by demands from environmentalists, `a world stage on which to indulge in global green rhetoric without ... having to face issues of domestic implementation' (Boehmer-Christiansen, 1994c: p. 401). The quest gives opposition leaders a target to rally the public against action. It is argued, for example, that the Kyoto protocol is imprudent because of scienti¢c uncertainties, unfair because developing countries are exempt, and a violation of national sovereignty if rati¢ed. Finally, the quest helps environmental leaders rally ¢nancial and political support from the environmental rank-and-¢le. In particular, the grand scale of the quest ^ to impose changes in energy, land use, and other relevant policies worldwide ^ is commensurate with and rea¤rms an apocalyptic vision of anthropogenic environmental disasters worldwide. Historically, support for environmental organizations tends to decline whenever expectations of environmental disaster recede into the background (Dunlap and Mertig, 1991). If the principal interests considered inaction or support for their separate organizations to be consistent with the common interest, they would use them in public as justi¢cations for the regime or its policies. With less attention in the news media, the regime's quest also allows research agencies closely linked to the natural sciences to maintain their expanded in£uence and funding. For example, agencies of the U.S. government quickly expanded their research budgets through the interagency U.S. Global Change Research Program (USGCRP), which has spent more than $15 billion over the last decade (Lawler, 1998; p. 1682). Its annual budget of about $1.8 billion per year is more than ten times the annual budget of the U.S. Climate Change Action Plan.17 (The Action Plan was introduced by the Clinton Administration in October 1993 to ful¢ll the voluntary emissions-reductions responsibilities of the U.S. under Article 4 (2) of the Framework Convention.) USGCRP supports about half the world's scienti¢c research aimed at models for predicting and assessing the impacts of climate change. But it supports little or no policy research aimed at clarifying and recommending speci¢c policies that might be taken by speci¢c policymakers in particular circumstances.18 Similarly, the Chairman of the IPCC, as it began its second assessment, rea¤rmed that `Our task is to assess knowledge rather than to recommend measures to be taken' (Bolin, 1994: p. 94). Basic assumptions tend to rationalize these research interests, and to obscure their actual role in the regime.19 Scientists and others typically assume that predictions and assessments of climate impacts from improved scienti¢c models will be accurate, necessary for rational policy decisions, and independent of any political position. However, models of open systems cannot be validated or

19 veri¢ed even in principle (Oreskes et al., 1994). In practice the assumptions of such models are undermined by learning and creativity in human choices and decisions, among other factors; and the accuracy of the models' predictions decays as a function of the time horizon ^ in other words, the length of time the predictions are exposed to learning, creativity, and other factors omitted from the models. Furthermore, accurate predictions of costs and bene¢ts are neither necessary nor su¤cient for rational policy decisions. They are not su¤cient because rational decisions also depend upon clarifying policy goals, evaluating policy alternatives, and reconciling the inevitable policy di¡erences through politics. They are not necessary because it is procedurally rational to act on a promising policy alternative despite uncertain costs and bene¢ts ^ provided the alternative is modest enough to assess in an appropriate time frame, to fail gracefully if it does fail, and to learn from the experience. Finally, predictions and assessments of climate impacts are not politically independent, even if the research is carried out as objectively as possible. As noted above, the IPCC's Synthesis assumes the policies of policymakers in the Conference of the Parties and supports their power position, to the disadvantage of localized and specialized policymakers. Whether intended or not, the actual role of the research interests is to subordinate the intelligence process to sustaining policies that have failed to achieve their own stated goals. These policies express the principal interests represented in the regime.20 However, from a common interest standpoint, the intelligence process is insu¤ciently comprehensive insofar as it discounts or ignores the exercise of power, the major factor that stands in the way of mandatory policy proposals.21 It is insu¤ciently creative insofar as it defends old policies that have not succeeded according to their own goals and ignores newer and more realistic alternatives. Preoccupation with mitigation through a legally-binding, international convention was apparent in the 1988 U.N. General Assembly resolution that provided initial guidance for the IPCC, if not before.22 It is inappropriately selective insofar as it ignores the need for adaptation. Adaptation has been `an unacceptable, even politically incorrect, idea' within the regime, where it is perceived to detract from negotiations on targets and timetables for mitigation, and to depend upon reduction of scienti¢c uncertainties about the regional impacts of climate change (Pielke, 1998: p. 162). The intelligence process is insu¤ciently dependable to the extent that it construes the reduction of scienti¢c uncertainties as a prerequisite for action on all mitigation or adaptation measures. `No regrets' actions already have been taken in the U.S. Climate Change Action Plan and its equivalents in other countries for several years at least, despite persistent scienti¢c uncertainties. Finally, the intelligence process is insu¤ciently open to the extent that it focuses on the concerns of international and national policymakers, and fails to address in the gathering and dissemination of information the concerns of localized and specialized policymakers who have the power to accept or reject changes in their own policies that are relevant to climate change. In summary, the regime has failed to realize its own very modest mitigation

20 goals because it lacks the power to impose legally-binding targets and timetables or other mandatory policies. In support of its quest for mandatory policies, the regime has failed to give serious consideration to adaptation policies or voluntary `no regrets' policies despite acknowledged needs and opportunities for them. In short, the regime's quest for mandatory policies serves as a substitute for the common interest in reducing losses from extreme climate events through action on prudent adaptation and mitigation policies. This pattern is explained and sustained by the principal interests in the regime, including signi¢cantly the research interests that dominate the intelligence process. 3. Alternatives What might done by the climate change regime on behalf of the common interest? Among the alternatives worth considering are persistence in the quest for mandatory policies, on the expectation that the regime might acquire the power necessary to prescribe and enforce them; harvesting experience from `no regrets' policies; and reconsidering the basic assumptions underlying the regime and its approach to policy. The reduction of scienti¢c uncertainty is sometimes expected to provide the power necessary for the regime to prescribe mandatory policies. In particular, scientists are expected to reach a consensus on the causes, extent, and impacts of global warming, producing a political consensus and prescription and action on mandatory policies devised by the regime. However, achievement of a scienti¢c consensus cannot be assumed. It will be di¤cult because scientists bring diverse scienti¢c and non-scienti¢c perspectives to their work, and because there is no way to verify or validate their models of open systems completely. If a scienti¢c consensus is achieved, it is likely to be achieved after the fact, on the basis of observed, not predicted, climate changes and impacts. Moreover, a political consensus cannot be assumed to follow from a scienti¢c consensus. A scienti¢c consensus will be perceived as propaganda by political interests predisposed against the regime's mandatory policies, and will not weaken their opposition in the absence of non-propaganda factors such as extreme climate events.23 It is also possible to accept a scienti¢c consensus on the problem of global warming and still to oppose policy solutions on normative grounds, including the costs of action. Finally, prescription and action cannot be assumed to follow from a political consensus to do something about climate change. There are still the tasks of devising speci¢c policies that might be taken by speci¢c policymakers in particular circumstances, and of devising strategies to persuade or coerce them to accept the policies. As noted above, these tasks have been avoided in the o¤cial emphasis on scienti¢c predictions and assessments in the intelligence process. Climate surprises or extreme but lesser climate events are more likely than scienti¢c developments to support the regime's quest for mandatory policies.

21 The heat wave, drought, and other extreme climate events that a¥icted North America in 1988 are an important precedent. Because of them, Congressional testimony by climate scientists that was largely ignored in previous years became front-page news, and the in£uence of the Toronto Conference Statement was multiplied. According to climate scientist, Stephen Schneider, `In 1988, nature did more for the notoriety of global warming in ¢fteen weeks than any of us [scientists] or sympathetic journalists and politicians were able to do in the previous ¢fteen years' (quoted in Ungar, 1995: p. 446). While nature intensi¢ed public concerns, public o¤cials allayed them through support for scienti¢c research, including establishment of the USGCRP and the IPCC. Social scares like the one in 1988 are `acute episodes of collective fear that lead to accelerated demands on the political arena' (Ungar, 1995: p. 445). In the future, extreme climate events including climate surprises could abruptly intensify the political will and the practical necessity to take action on various scales. If so, policymakers will not need scienti¢c predictions and assessments of climate impacts, or a legally-binding international agreement, or reports of greenhouse gas inventories. Policymakers will need ¢eld-tested action alternatives that can be implemented quickly and concurrently in the a¡ected areas, in response to unavoidable public demands to `Do something!' Without ¢eld-tested alternatives already available, the human costs of any extreme climate events will be unnecessarily exacerbated; the regime will miss an opportunity to broaden and deepen political support for itself and for policies that go beyond `no regrets'; and the principal interests represented in the regime, including science, will become visible and vulnerable politically. Visibility and vulnerability stem from expectations created by participants in the regime itself. Whenever they attempt to justify another increment of funding, authority, or some other resource, they reinforce expectations among attentive members of the world community that the regime will somehow attend to the common interest in climate change. The situation of science in the regime calls to mind what Lasswell (1970: p. 119) wrote nearly thirty years ago: ...science has grown strong enough to acquire visibility, and therefore to become eligible as a potential scapegoat for whatever disenchantment there may be with the earlier promises of a science-based technology. .. . The verdict may be that whom the historical process would destroy it ¢rst must make strong enough to achieve a visibility su¤cient to arouse false hopes, while remaining weak enough to acquiesce and connive in the frustrations of their potential ^ thus for science and scientists. Ironically, sustaining the climate change regime in its present form depends upon the falsi¢cation of the regime's projections of anthropogenic climate disasters worldwide, including the possibility of climate surprises. The con¢rmation of those projections by future events would expose the regime's neglect of policy research and prudent action. Noting the regime's lack of power, some analysts have concluded that the

22 substantive approach taken in the Berlin Mandate ^ toward more stringent, legally-binding targets and timetables ^ is not likely to succeed in the absence of major changes in circumstances. They have recommended instead a procedural approach at the international level that emphasizes, in one version, `strengthening the reporting and review process, building the FCCC's institutions, and developing an e¡ective noncompliance procedure, rather than attempting immediately to negotiate additional targets and timetables' (Bodansky, 1995: p. 447; Sebenius, 1994; Victor and Salt, 1995; Flavin and Tunali, 1995; Rayner and Malone, 1997; 1998). A procedural approach would moderate the expectations created by the regime and therefore reduce its vulnerability. If successful, a procedural approach also would put the regime in a better position to enforce compliance with mandatory policies, when and if they become politically feasible. Meanwhile, however, the procedural approach entails the expense of constructing a worldwide system for monitoring net changes in greenhouse gas emissions, and risks reinforcement of the regime's tendency to substitute symbols for action and material progress in mitigation and adaptation. Already the unresolved issue of assigning credit for emissions reductions under future binding commitments appears to be a deterrent to present action (Coppock, 1998). Nevertheless, the procedural approach is preferable to the substantive approach taken in the Berlin Mandate, and supportable so long as it does not preclude or marginalize a concurrent emphasis on action alternatives that are already prudent. Such alternatives can be found by looking down from the aggregate failure of Article 4 (2) at the international level to small-scale `no regrets' policies that already have realized the objectives of localized and specialized policymakers, and contributed toward mitigation of climate change or adaptation to extreme climate events.24 These policies do not require a legally-binding international agreement, a worldwide monitoring system, or other enforcement mechanisms, and, as noted above, they are already authorized in the Framework Convention. They are voluntary policies, capitalizing upon existing motivations not limited to prevention of dangerous interference with the climate system. Opportunities for `no regrets' adaptation policies exist whenever and wherever storms, £oods, droughts, or ¢res devastate human systems, making it necessary to reconstruct those systems and possible to reconstruct them to reduce future vulnerabilities. Opportunities for `no regrets' mitigation policies exist whenever and wherever emission-producing equipment becomes obsolete or worn-out, making it necessary to replace the equipment and possible to replace it with cost-e¡ective technologies producing fewer emissions.25 A case of `no regrets' mitigation is Johnson & Johnson, the manufacturer of medical supplies.26 Under the leadership of its corporate energy director, Harry Kau¡man, the company in 1991 made a voluntary commitment to the Green Lights program to upgrade lighting in 90% of its building space in ¢ve years. In return, the company received technical assistance from the Green Lights program, one of more than forty existing or new programs folded into the U.S. Climate Change Action Plan by the time the Clinton Administration

23 announced it in 1993. The company exceeded the initial commitment by upgrading lighting in 95% of its building space. Over the ¢ve-year period it avoided 33,500 tons of CO2 emissions. According to an internal tracking system, Johnson & Johnson's lighting program saves the company $3.55 million annually in energy costs avoided. Recognizing and building upon that success, the company signed up with the Energy Star Buildings program, another Action Plan program initiated by the Environmental Protection Agency in 1995. In the ¢rst year and a half, Johnson & Johnson avoided another 25,200 tons of CO2 emissions. A case of `no regrets' adaptation is Fargo ND, which was £ooded in the spring of 1997 along with other cities on the Red River of the North.27 In Grand Forks ND, the Red River crested at a record level of 54 feet ^ 5 feet higher than city o¤cials had expected and 2 feet higher than they had prepared for. Grand Forks and its neighbor just across the river su¡ered most of the losses from the 1997 £ood, estimated at about $1.2 billion. In Fargo in contrast, a public works o¤cial, Dennis Walaker, recognized uncertainties in the predicted crest and took the lead in preparing the city to minimize damage by building dikes on higher ground farther from the river. Some 20,000 properties were e¡ectively protected by exposing about 50 to £ooding. Within a year after the £ood, Fargo with federal assistance bought most of the £ooded properties near the river and thereby reduced its vulnerability to an equivalent £ood in the future. The di¡erence in outcomes between Grand Forks and Fargo lies not in the accuracy of forecasts from the U.S. Weather Service, but in the civic capacity necessary to integrate such forecasts with relevant local knowledge and information to make local policy decisions. Harvesting experience from such `no regrets' adaptation and mitigation policies is an obvious priority for global change policy research. The aim is to provide information readily usable by specialized and localized policymakers, based on the experience of leaders like those in Johnson & Johnson and Fargo. As developed elsewhere (Brunner and Klein, 1999) harvesting experience is a strategy that can be broken down into three principal tasks: à

à

à

The ¢rst task is to identify and describe policies provisionally appraised as successful, to verify that they have in fact succeeded according to the mitigation or adaptation criteria of national policymakers and the `no regrets' criteria of localized or specialized policymakers, and to explain formal and e¡ective responsibility for those successes. The second task is to disseminate the policies as intelligence for other policymakers in other localized or specialized policy processes who might consider and adapt them to their own circumstances on a voluntary basis. Dissemination typically includes, but is not limited to, a clearinghouse to match speci¢c requests with speci¢c sets of policies. The third task is to stimulate the innovation and ¢eld-testing of new policies in promising but neglected areas, such as transportation or impoverished places. Some resources might be freed up by terminating

24 once-promising programs that have failed. Termination is an important part of a strategy that depends upon bold, persistent experimentation in the ¢eld, rather than the re¢nement of predictions and assessments. Notice that the strategy focuses attention, a limited resource, on successful policies and the leaders responsible for them. Identi¢cation of the leaders rewards and reinforces them with recognition if nothing else. Information on the policies de¢nes de facto standards of best practice to motivate others, and clari¢es the speci¢c means by which others might meet the standards. Thus by consolidating successful policy solutions and reinforcing successful leaders at each stage, the strategy builds the civic capacity needed to sustain e¡ort over the long-term. Harvesting experience is also a process that tends to be initiated more or less spontaneously whenever policymakers who face signi¢cant problems or opportunities begin looking for others who already have met similar challenges. For example, independent pro¢t centers within Johnson & Johnson's operations worldwide have been clarifying, di¡using, and adapting best practices for the e¤cient use of fossil fuels (Kau¡man, 1997). Johnson & Johnson is itself a source of best practices for other companies through the Green Lights program within the U.S. Climate Change Action Plan (Brunner and Klein, 1999). More than a dozen European and North American cities several years ago began clarifying, di¡using, and adapting best practices for the e¤cient use of fossil fuels through the Urban CO2 Project (Harvey, 1993). In 1998 the Federal Emergency Management Agency set up a clearinghouse for `Mitigation Success Stories' on the internet.28 The Climate Institute (1998) also serves as a clearinghouse, among other functions. Furthermore, harvesting experience is not limited to environmental policy or national arenas: By 1980 more than 30 countries had voluntarily adapted the practice of environmental impact assessments after it was ¢rst introduced in the U.S. in 1969 (Paarlberg, 1996), and examples of harvesting experience can be found in a variety of policy areas, including education, economic, and welfare policy (e.g., Smith, 1995; Schorr, 1997). However, the process of harvesting experience is subject to various malfunctions. One of the most important is the tendency of policy innovators to in£ate claims of their own success, and in doing so to mislead potential di¡users or adaptors of the policy innovation. The obvious correction is third-party veri¢cation of policies provisionally appraised as successful. The identi¢cation and correction of such malfunctions are priorities for policy research. Such corrections may be implemented at any appropriate level, such as Green Lights at the program level, the U.S. Climate Change Action Plan at the national policy level, and the Framework Convention at the international policy level. Notice that harvesting experience would rede¢ne the role of higher-level policymakers ^ from attempting to negotiate and enforce compliance with top-down policies that go beyond `no regrets,' to facilitating the innovation, di¡usion, and adaptation of `no regrets' policies that have already proven to be successful in

25 particular circumstances. The strategy of harvesting experience, including the correction of malfunctions, provides a basis for institutional arrangements of the kind suggested by Rayner and Malone (1997: p. 332). Noting the diversity and unpredictability of change in society, they suggest a `focus on building responsive institutional arrangements that monitor change and maximize the £exibility of populations to respond creatively and constructively to it.' If the point is to prepare the world community for extreme climate events, including climate surprises, the strategy of harvesting experience will succeed to the extent that it builds up an inventory of ¢eld-tested policies and institutional arrangements by which the policies can be adapted quickly, concurrently, and voluntarily by policymakers in many localized and specialized contexts ^ whenever and wherever there is the political will or the practical necessity to act for any reason. Any short-term contribution to the present emissions-reductions objectives of the climate change regime may be considered a bonus. In any case, it is not possible to predict the aggregate contribution of the strategy with any con¢dence or accuracy, because it depends upon human learning and creativity in many particular circumstances. It is not necessary to predict the aggregate contribution, because the relevant projections are those made by Harry Kau¡man, Dennis Walaker, and other policymakers in particular circumstances. The task is less to predict the aggregate future than to shape it through modest actions on prudent policies, informed by the best available science and technology, wherever and whenever opportunities exist. This is probably the most that can be done, given the climate change regime's inability to impose stringent, `legally-binding' targets and timetables for emissions reductions from the top down. Finally, it would be helpful for all concerned to surface for evaluation the basic assumptions underlying the climate change regime. The regime's core epistemic community tends to construe the Earth and its component systems as complex mechanisms that obey scienti¢c laws, which, when understood and integrated into computer models, will enable scientists and allied policymakers to predict and control the systems' behavior. Some scientists in the formative years of the regime went so far as to liken their role to the construction of an operator's manual for an automobile or even for Spaceship Earth.29 This implies a command center controlled by those few with the necessary technical expertise; the rest of us, presumably, are passengers along for the ride. These assumptions are expressed in the top-down orientation of the regime, and most spectacularly in global circulation models, the hierarchical structure of the formal regime, and the expected decisions on a stabilization level and net emissions pathway for the world as a whole. These assumptions are called into question by persistent if not expanding scienti¢c uncertainties, the political power withheld from the Conference of the Parties, and the failure of the quest for legally-binding targets and timetables. Alternative basic assumptions might construe the Earth and its component systems as adaptive complex systems comprised of myriad living forms (Holland, 1992a; 1992b; Holling; 1995). From any initial condition, such systems may

26 take many possible trajectories because living forms respond to diverse external environments according to diverse internalized predispositions, instinctive or acquired, that are subject to change. Human behavior and the behavior of the Earth system and its components become unpredictable as people exploit their capacity for learning and creativity and their freedom to choose, and as the time horizon extends into the future. Nevertheless, complex problems have been solved as diverse groups and communities, proceeding in parallel, try diverse solutions to variants of common problems in concrete circumstances, and clarify, disseminate, and adapt what works. These distributed policies and policy processes turn diversity, learning, and creativity into assets, preserve the freedom to choose, and encourage widespread participation. Policy scientists might conceive their role as applying knowledge to improve particular policies and policy processes, as well as improving knowledge for this purpose. Support for these basic assumptions can be found in the IPCC's ¢ndings, as expressed in the Synthesis, that rational decisions depend upon combinations of multiple factors in particular circumstances ^ circumstances far too numerous and far too complex to understand and control from the top down. As a point of departure for the reassessment of basic assumptions, consider the re£ections of the Director of Research for the Royal Netherlands Meteorological Institute. In 1990 he noted the title of a special issue of Scienti¢c American , `Managing the Planet,' and the name of a program launched by the U.S. Environmental Protection Agency, `Stabilizing the Climate System.' Then he confessed that I am terri¢ed by the hubris, the conceit, the arrogance implied by words like these. Who are we to claim that we can manage the planet? We cannot even manage ourselves. Who are we to claim that we can run the planetary ecosystem? In an ecosystem no one is boss, virtually by de¢nition. Why are we, with our magni¢cent brains, so easily seduced by technocratic totalitarianism? (Tennekes, 1990) `Technocratic totalitarianism' refers to a con£uence of scienti¢c and management pretenses that ignore the simple feedback suggested by another question: `But how does it play in Peoria?'30 Technocratic hubris if not totalitarianism has also been explicitly rejected by some policymakers (e.g., Havel, 1992) as well as by some scientists. 4. Conclusion The basic justi¢cation of the U.S. Global Change Research Program and its funding request has been rea¤rmed since 1989: `In the coming decades, global change may well represent the most signi¢cant societal, environmental, and economic challenges facing this nation and the world. The national goal of developing a predictive understanding of global change is, in its truest sense, science in the

27 service of mankind' (Subcommittee on Global Change Research, 1997: p. 76; emphasis in the original). Rea¤rmation of `science in the service of mankind' for funding and related purposes creates public expectations and a professional responsibility on the part of scientists to meet those expectations. In the U.S. at least, the attentive public and some of their elected representatives expect ^ according to the Global Change Research Act of 1990 (Public Law 101^606) ^ `information readily usable by policymakers attempting to formulate e¡ective strategies for preventing, mitigating, and adapting to the e¡ects of global climate change.' A predictive understanding of global change was not necessary for localized and specialized policy makers to implement e¡ective `no regrets' policies on a voluntary basis. Some `no regrets' policies have succeeded on small scales in mitigating or adapting to climate change without compromising economic, democratic, and other aspects of the common interest. Harvesting experience from these policies can provide information readily usable by other localized and specialized policy makers, as well as the Conference of the Parties, to help realize the common interest more broadly. Moreover, the scienti¢c focus on a predictive understanding of global change has political consequences, whether intended or not: It supports the quest for mandatory global policies, even though the regime lacks the power to prescribe and enforce such policies. It legitimizes demands to do nothing, as if reductions in scienti¢c uncertainty were a prerequisite for rational actions. At the same time, it diverts attention and other resources from additional policy alternatives. Complacency with this state of a¡airs ^ infeasible mandatory policies, inappropriate `do nothing' policies, and neglect of alternatives ^ leaves the world community unnecessarily vulnerable to extreme climate events, especially climate surprises. It also leaves scientists and their political allies unnecessarily vulnerable to an outraged public that will expect more than predictions when such events occur. However, complacency is not inevitable. Scienti¢c leaders do have opportunities to clarify and secure the common interest of science and the world community. Insofar as the governments represented in the Conference of the Parties depend upon science for political support, they must take scienti¢c leaders into account in the formulation of policies. Scienti¢c leaders might begin by seeing through the formal role of science and beyond the next few budget cycles to larger realities. They might acknowledge the political origins and consequences of global change research programs, consider whether those programs are consistent with public expectations and professional responsibilities, and rally scienti¢c and political support for any program adjustments that may be warranted. They might conclude that the common interest in climate change is better served by distinguishing between predictive research and policy research; by reauthorizing predictive research as basic research further removed from the immediate pressures of the political arena; and by supporting policy research to provide information readily usable by policy makers, including those in the Peorias of the world. In any case, opening up the climate change regime to a wider range of action alternatives for a wider range of policy

28 makers is the key to science in the service of humankind. And as Lasswell (1970: p. 120) observed, `It must not be supposed that all men [and women] of knowledge, and notably scientists, are happily reconciled to the contemporary situation.' Acknowledgements This is a revision of a paper presented at the 17th Policy Sciences Annual Institute, Yale Law School, New Haven, CT, October 2^4, 1998. For comments and suggestions on earlier versions the author is grateful to Matthew R. Auer, Radford Byerly Jr., Jonathan I. Charney, Roger A. Pielke Jr., W. Michael Reisman, and Andrew R. Willard. This material is based on work supported by the National Science Foundation under Grant No. SBR-95-12026. Any opinions, ¢ndings, and conclusions or recommendations are those of the author and do not necessarily re£ect the views of the National Science Foundation.

Notes 1. See the Intergovernmental Panel on Climate Change (IPCC, 1995) for the mainstream conclusion on temperature projections (section 2.7) and climate surprises (section 2.12): `Future unexpected, large and rapid climate system changes (as have occurred in the past) are, by their nature di¤cult to predict. This implies that future climate changes may also involve ``surprises.'' ' On the skeptical scientists see Pearce (1997: p. 43), which quotes one of the skeptics, Pat Michaels: `You can't make a case for a global apocalypse out of a 1.5 ³C warming.' However, another skeptic does not rule out the possibility that `the climate system may one day jump to a new, very much warmer equilibrium, or some cloud feedback or biological response may exacerbate global warming' (Balling, 1992: p. 134). See also Kerr (1998). 2. Abramovitz and Dunn (1998). The International Federation of Red Cross and Red Crescent Societies (1997) corroborates the general pattern. 3. But climate change should not be equated with losses. Pielke and Landsea (1998) show that increases in population and property along the Gulf and Atlantic coasts of the U.S. are more important factors accounting for increases in losses from hurricanes in those areas. 4. In addition to the emissions reduction target, other demands in the Toronto Conference Statement also in£uenced the policy agenda over the next decade. The Statement called for `taxes on CO2 emissions, an environmental trust fund to aid Third World countries, and a supranational institution with enforcement powers' (Ungar, 1995: p. 447). 5. The text of the Framework Convention is conveniently reproduced as Appendix I in O'Riordan and Ja«ger (1996). This review follows McDougal, Lasswell, and Reisman (1981), Lasswell and McDougal (1992: pp. 1137^1138), and similar sources on constitutive inquiry in the policy sciences. In addition to identifying participants, constitutive inquiry takes into account their perspectives (including basic policies and interests), and the situations in which they interact; the resources (including authority and control) and strategies available to shape policy outcomes ; and responsibility for the intended and unintended e¡ects of policy outcomes. See also Batt and Short (1992^1993). 6. Cf. Cooper (1998: p. 72): `An e¡ective treaty to inhibit greenhouse gas emissions needs the cooperation of consuming publics.' 7. Cf. Bodansky (1995: p. 432): `Although it by no means guarantees action, [the Framework

29

8.

9. 10.

11. 12. 13.

14. 15.

16.

17.

18.

19. 20.

21.

Convention] helps generate the scienti¢c and normative consensus necessary to adopt more substantive obligations in the future.' OECD countries also have a speci¢c commitment to provide ¢nancial and technological assistance to developing countries according to Article 4 (3)^(5), but there are no mandatory provisions for raising or appropriating revenues for this purpose. Adequate ¢nancial assistance nevertheless is recognized as a condition for implementation of the general commitments of developing countries. The following description of the outcomes employs the conceptual model of decision process in Lasswell (1971) and McDougal, Lasswell, and Reisman (1981). The IPCC (1996: p. 188) has concluded that `even with the most ambitious abatement policy, some climate change is likely to occur.' See also Kauppi (1995). More recently, Malako¡ (1997: p. 2048) reports that `climate models ... suggest that concentrations will at least double by 2100 unless developing nations hold their emissions steady and industrial nations progressively reduce emissions by roughly half.' The Berlin Mandate is reproduced as Appendix II in O'Riordan and Ja«ger (1996). The Synthesis, retrieved from the IPCC's website, was prepared by a team of lead authors selected and chaired by Bert Bolin, then Chairman of the IPCC. Cf. White (1992): `The media's current attention to the mounting burden of greenhouse gases may be unfortunate insofar as it neglects other issues. Far more signi¢cant are questions of how population and development policies may be designed to assure an equitable quality of life that degrades neither society nor the environment.' The concluding section of the Synthesis, on The Road Forward, asserts that unspeci¢ed `improvement of knowledge can reduce the risks posed by climate change' (section 8.4). The negotiation of rules for crediting claimed reductions are complicated by the advantages and disadvantages of alternative baselines for di¡erent parties, and by `additionality': Article 12 of the Kyoto Protocol requires that emissions reductions `are additional to any that would occur in the absence of the certi¢ed project activity.' Bert Bolin (1998: p. 331). He also observes that `The IPCC has developed guidelines to establish a common base for determination of changes in sources and sinks, but these were not designed to serve as a legal basis for compliance.' The 1997 draft of the o¤cial U.S. Climate Action Report to the Framework Convention notes (chapter 4, p. 7) that the budget for the Action Plan was $184 million in FY 95, $158 million in FY 96, and $163 million in FY 97. According to the Committee on Earth and Environmental Sciences (1992 p. 3), `Policy analysis, the short-term evaluation of speci¢c policy proposals, is excluded from consideration' under the economics component of USGCRP. The Subcommittee on Global Change Research (1997: p. 76) recently rea¤rmed USGCRP's original emphasis on a predictive understanding of global change. Pielke (1995) documents and explains the neglect of policy research under USGCRP through 1994. For further documentation and development of the following points on basic assumptions, see Brunner (1993; 1996). Cf. Edwards (1996: p. 152): `Its ability to stake an authoritative claim to knowledge is what gives an epistemic community its power.' The criteria applied in this paragraph ^ comprehensiveness, creativity, selectivity, dependability, and openness ^ were formulated in Lasswell (1971: chapter 5) for appraisal of intelligence processes from a common interest standpoint. For example, Swift (1998, p. 81) advocates an international emissions trading system, and acknowledges that `such a system is unlikely to be fully mapped out at the Buenos Aires meeting [of the Conference of the Parties] in November [1998],' but discounts the political factors that stand in the way. Cooper (1998) proposes an international carbon emissions tax, but does not identify sanctions for national non-compliance with it. Schneider (1998) proposes a `modest solution,' but does not identify what can be done under present political circumstances to implement it: `I realize that it is easy for an academic to say, ``do the right thing for history,'' not having to confront daily hostile political interests' (p. 18).

30 22. See Bodansky (1994: p. 53). Ja«ger and O'Riordan (1996: p. 2) observe that `The fresh evidence from the natural science community is not producing much that is really new.' 23. On the theoretical basis for this projection, see Lasswell and Kaplan (1950: p. 113). 24. O'Riordan and Ja«ger (1996) draw attention to small-scale successes in Europe. For the U.S., see Brunner and Klein (1999). This focus is compatible with Rayner and Malone (1998). 25. Such opportunities are acknowledged in o¤cial sources (e.g., IPCC 1995: section 5.4) and are prominent in technological approaches to mitigation in climate change policy (e.g., Coppock, 1998). 26. This case is developed and documented more extensively in Kau¡man (1997), and in Brunner and Klein (1999). 27. This case was considered at a workshop on `Prediction in the Earth Sciences: Uses and Misuses in Policymaking,' Estes Park, CO, September 10^12, 1998. For a report on the workshop, see Stevens (1998). See also Pielke (1997) on the Grand Forks case, and Anon. (1998) on the technical aspects of the Fargo case. 28. The clearinghouse can be observed at http://www.fema.gov/mit/sstory/. 29. Pielke (1995: p. 43) documents use of the owner's manual by global change scientists and policymakers to clarify the role of USGCRP. For a critique of Spaceship Earth as a metaphor, see Adams (1998: p. 79). On the blinders that frequently arise from such metaphors, see Scho«n (1979) and Limerick (1992). 30. The author is grateful to Dr. Tennekes for this elaboration of his reference to `technocratic totalitarianism' in a letter dated July 21, 1997.

References Abramovitz, J. N. and S. Dunn (1998). `Record year for weather-related disasters,' Worldwatch Institute Vital Signs Brief 5 (November 27). Adams, B. (1998). Timescapes of Modernity. New York and London: Routledge. Anon. (1998). `Flood preparedness in Fargo,'A/E/C: The Magazine of Computer Solutions (Winter): 9^13. Balling, R. (1992). The Heated Debate: Greenhouse Predictions Versus Climate Reality. San Francisco: Paci¢c Research Institute for Public Policy. Batt, J. and D. C. Short (1992^1993). `The jurisprudence of the 1992 Rio Declaration on Environment and Development: A law, science and policy explication of certain aspects of the United Nations Conference on Environment and Development,' Natural Resources and Environmental Law 8: 229^292. Bodansky, D. M. (1994). `Prologue to the climate change convention,' in I. M. Mintzer and J. A. Leonard, eds., Negotiating Climate Change: The Inside Story of the Rio Convention. Cambridge: Cambridge University Press, pp. 45^74. Bodansky, D. M. (1995). `The emerging climate change regime,' Annual Review of Energy and Environment 20: 425^461. Boehmer-Christiansen, S. A. (1994a). `Global climate protection policy: The limits of scienti¢c advice,' part I. Global Environmental Change 4: 140^159. Boehmer-Christiansen, S. A. (1994b). `Global climate protection policy: The limits of scienti¢c advice,' part II. Global Environmental Change 4: 185^200. Boehmer-Christiansen, S. A. (1994c). `A scienti¢c agenda for climate policy?' Nature 372 (1 December): 400^402. Bolin, B. (1994). `Next step for climate change analysis,' Nature 368 (10 March): 94. Bolin, B. (1998). `The Kyoto negotiations on climate change: A science perspective,' Science 279 (16 January): 330^331. Bradsher, K. (1997). `Trucks, darlings of drivers, are favored by the law, too,' New York Times (November 30): A1.

31 Brunner, R. D. (1993). `Myths, scienti¢c and political,' in The Objectivity Crisis: Rethinking the Role of Science in Society, Chairman's Report to the Committee on Science, Space and Technology, House of Representatives, 103rd Congress, 1st Session, Serial D, 5-12. Brunner, R. D. (1996). `Policy and Global Change Research: A Modest Proposal,' Climatic Change 32: 121^147. Brunner, R. D. and R. Klein (1999). `Harvesting experience: A reappraisal of the U.S. Climate Change Action Plan,' Policy Sciences 32: 133^161. Climate Institute (1998). `Public and private groups give vigorous boost to states and localities in enterprising green energy programs,' Climate Alert 11: 1f. Committee on Earth and Environmental Sciences (1992). Economics and Global Change: A Supplement to the U.S. President's FY 1993 Budget. Cooper, R. N. (1998). `Toward a real global warming treaty,' Foreign A¡airs 77 (March/April): 66^79. Coppock, R. (1998). `Implementing the Kyoto Protocol,' Issues in Science and Technology XIV (Spring): 66^74. Cushman, J. H. Jr. (1997). `Global warming ¢ght wins wide support,' New York Times (November 28): A14. Cushman, J. H. Jr. (1998). `U.S. signs a pact to reduce gases tied to warming,' New York Times (November 13): A1. Dewar, H. and K. Sullivan (1997). `Senate GOP declares Kyoto pact dead,' Washington Post (December 11): A37. Dunlap, R. E. and A. G. Mertig (1991). `The evolution of the U.S. environmental movement from 1970 to 1990: An overview,' Society and Natural Resources 4: 209^218. Edwards, P. (1996). `Global comprehensive models in politics and policymaking,' Climatic Change 32: 149^161. Fialka, J. J. (1997). `Clinton's e¡ort to curb global warming draws some business support, but it may be too late,' Wall Street Journal (October 22): A24. Flavin, C. and O. Tunali (1995). `Getting warmer: Looking for a way out of the climate impasse,' World Watch (March/April): 10^19. Goodin, R. E. (1996). `Institutionalizing the public interest: The defense of deadlock and beyond,' American Political Science Review 90 (June): 331^343. Harvey, L. D. D. (1993). `Tackling urban CO2 emissions in Toronto,' Environment 35 (September): 16^20, 38^44. Havel, V. (1992). `The end of the modern era,' New York Times (March 1): E15. Holland, J. H. (1992a). `Adaptive complex systems,' Daedalus 121 (Winter): 17^30. Holland, J. H. (1992b). `Genetic algorithms,' Scienti¢c American 267 (July): 66^72. Holling, C. S. (1995). `What barriers? What bridges?' in L. H. Gunderson, C. S. Holling and S. S. Light, eds., Barriers and Bridges to the Renewal of Ecosystems and Institutions. New York: Columbia University Press, pp. 3^34. Intergovernmental Panel on Climate Change (1995). IPCC Second Assessment Synthesis of Scienti¢ctechnical Information Relevant to Interpreting Article 2 of the U.N. Framework Convention on Climate Change. Intergovernmental Panel on Climate Change (1996). Economic and Social Dimensions of Climate Change: Contribution of Working Group III to the Second Assessment Report of the IPCC. Cambridge: Cambridge University Press. International Federation of Red Cross and Red Crescent Societies (1997). World Disasters Report 1997. Ja«ger, J. and T. O'Riordan (1996). `The history of climate change science and politics,' in T. O'Riordan and J. Ja«ger, eds., Politics of Climate Change: A European Perspective. London and New York: Routledge, pp. 1^31. Kaiser, J. (1998a). `New network aims to take the world's CO2 pulse,' Science 281 (24 July): 506^ 507. Kaiser, J. (1998b). `Possibly vast greenhouse gas sponge ignites controversy,' Science 282 (16 October): 386^387.

32 Kau¡man, H. (1997). `Change in the business climate: Case studies in greenhouse gas reduction.' Manuscript (June 19). Kauppi, P. (1995). `The United Nations climate convention: Unattainable or irrelevant,' Science 270 (1 December): 1454. Kerr, R. A. (1998). `Warming's unpleasant surprise: Shivering in the greenhouse? Science 281 (10 July): 156^158. Lasswell, H. D. (1970). Must science serve political power? American Psychologist 25 (1970): 117^123. Lasswell, H. D. (1971). A Pre-view of Policy Sciences. New York: Elsevier. Lasswell, H. D. and A. Kaplan (1950). Power and Society. New Haven: Yale University Press. Lasswell, H. D. and M S. McDougal (1992). Jurisprudence for a Free Society. New Haven and Dordrecht: New Haven Press and Martinus Nijho¡. Lawler, A. (1998). `Global change ¢ghts o¡ a chill,' Science 280 (12 June): 1682^1684. Limerick, P. N. (1992). `Imagined frontiers: Westward expansion and the future of the space program,' in R. Byerly Jr., ed., Space Policy Alternatives. Boulder, CO: Westview Press, pp. 249^ 273. Malako¡, D. (1997). `Thirty Kyotos needed to control warming,' Science 278 (19 December): 2048. McDougal, M. S., H. D. Lasswell and W. M. Reisman (1981). `The world constitutive process of authoritative decision,' in M. S. McDougal and W. M. Reisman, eds., International Law Essays: A Supplement to International Law in Contemporary Perspective. Mineola, NY: Foundation Press, pp. 191^286. Muller, F. (1996). `Mitigating climate change: The case for energy taxes,' Environment 38 (March): 13^20, 36^43. Oreskes, N., K. Schrader-Frechette and K. Belitz (1994). `Veri¢cation, validation, and con¢rmation of numerical models in the earth sciences,' Science 263 (4 February): 641^646. O'Riordan, T. and J. Ja«ger, eds. (1996). Politics of Climate Change: A European Perspective. London and New York: Routledge. Paarlberg, R. L. (1996). `A domestic dispute: Clinton, Congress, and international environmental policy,' Environment 38 (October): 16^20, 28^33. Pearce, F. (1997). `Greenhouse wars,' New Scientist (19 July): 38^43. Pielke, R. A. Jr. (1995). `Usable information for policy: An appraisal of the U.S. Global Change Research Program,' Policy Sciences 28: 39^77. Pielke, R. A. Jr. (1997). `Meeting the promise of £ood forecasting,' Natural Hazards Observer (September): 8^9. Pielke, R. A. Jr. (1998). `Rethinking the role of adaptation in climate policy,' Global Environmental Change 8: 159^170. Pielke, R. A. Jr. and C.W. Landsea (1998). `Normalized hurricane damages in the United States: 1925-95,' Weather and Forecasting 13: 621^631. Rayner, S. and E. L. Malone (1997). `Zen and the art of climate maintenance,' Nature 390 (27 November): 332^334. Rayner, S. and E. L. Malone (1998). `Ten suggestions for policymakers,' in Human Choice and Climate Change, Vol. 4: What Have We Learned? Columbus, OH: Battelle Press, pp. 109^138. Schmidt, K. (1998). `Coming to grips with the world's greenhouse gases,' Science 281 (24 July): 504^506. Sebenius, J. K. (1994). `Toward a winning climate coalition,' in I. M. Mintzer and J. A Leonard, eds., Negotiating Climate Change: The Inside Story of the Rio Convention , Cambridge: Cambridge University Press, pp. 277^320. Schneider, S. H. (1998). `Kyoto Protocol: The un¢nished agenda,' Climatic Change 38: 1^21. Scho«n, D. A. (1979). `Generative metaphor: A perspective on problem-setting in social policy,' in A. Ortony, ed., Metaphors and Thought. Cambridge: Cambridge University Press, pp. 254^283. Schorr, L. B. (1997). Common Purpose: Strengthening Families and Neighborhoods to Rebuild America. New York: Anchor Books. Smith, H. (1995). Rethinking America: A New Game Plan from the American Innovators: Schools, Business, People,Work. New York: Random House.

33 Stevens, W. K. (1998). `When scienti¢c predictions are so good, they're bad,' New York Times (September 29): B11. Swift, B. (1998). `A low-cost way to control climate change,' Issues in Science and Technology XIV (Spring): 75^81. Subcommittee on Global Change Research (1997). Our Changing Planet: The FY 1998 U.S. Global Change Research Program. A Supplement to the President's Fiscal Year 1998 Budget. Taubes, G. (1997). `Apocalypse not,' Science 278 (7 November): 1004^1006. Tennekes, H. (1990). `A sideways look at climate research,' Weather 45: 67^68. Ungar, S. (1995). `Social scares and global warming: Beyond the Rio convention,' Society and Natural Resources 8: 443^456. U.S. General Accounting O¤ce (1996). Global Warming: Di¤culties Assessing Countries' Progress Stabilizing Emissions of Greenhouse Gases. GAO/RCED-96-188 (November). Victor, D. G. and J. E. Salt (1995). `Keeping the climate treaty relevant,' Nature 373 (26 January): 280^282. White, G. F. (1992). `Vital links in environmental systems,' Environment 34 (May): i.