Architecture

The term architecture has been used in many ways to refer to the fundamental structural characteristics of an institution (e.g., Aldy, Barrett and Stavins Reference Aldy, Barrett and Stavins2003), a regime (e.g., Hare et al. Reference Hare, Stockwell, Flachsland and Oberthür2010) or a broader governance system (e.g., Young Reference Young2008). We define governance architectures in this book as the overarching system of public and private institutions, principles, norms, regulations, decision-making procedures and organizations that are valid or active in a given area of global governance (drawing on the definition by Biermann et al. Reference Biermann, Betsill and Gupta2010). Architecture can thus be described as the macro-level of governance, which can be regarded as a ‘bird’s-eye view’ on the global governance landscape. This macro-level is not to be understood as a static entity, but rather as a fluid, dynamic architecture that continuously evolves according to external and internal pressures and governance processes.

Three elements are key in this conceptualization.

First, our conceptualization denotes an overarching system, which is broader than a single institution but narrower than an all-encompassing global order (Biermann Reference Biermann2014: 81–2). Therefore, architecture specifically refers to structures at the overarching governance level, which in turn consist of numerous building blocks such as state and non-state actors, transnational networks, intergovernmental institutions and regime complexes.

Second, our conceptualization of architecture refers to institutional settings that shape decisions of actors and institutions which exist and interact in a given policy domain. Some degree of influence of architectures on agency is therefore assumed in terms of both the effectiveness of individual institutions and the overall governance system (see also Betsill, Benney and Gerlak Reference Betsill, Benney and Gerlak2019).

Third, the impact of an architecture extends to all levels of governance. While a key unit of analysis is the macrostructure at the global level, the level of analysis is not limited to the global. Examples include multilevel governance architectures in areas such as climate mitigation, ocean governance, biodiversity conservation and so on.

Importantly, the notion of governance architectures does not assume the existence of a single architect. Instead, in many cases, governance architectures emerge from incremental and tedious processes of institutionalization that are decentralized and hardly planned. In other words, instead of a single architect, many architects are involved in shaping an architecture, even though individually, most have little control over the overall design and performance. The resulting structural configuration then influences how institutions interact by limiting the choices and opportunities for actors or by creating and shaping opportunities (Betsill, Benney and Gerlak Reference Betsill, Benney and Gerlak2019). A governance architecture is, in other words, in constant flux, evolving through the interaction between individual institutions at the micro level and the dynamic structure at the macro level.

The notion of architectures of global governance has emerged over the last two decades. Twenty years ago, most institutional research focused on single institutions, such as the ozone regime or individual treaties on wildlife or fisheries. Comparative analyses of such institutions then led to a better understanding of the creation, maintenance and effectiveness of single international institutions and their relations with national policies (Brown Weiss and Jacobson Reference Brown Weiss and Jacobson1998; Young Reference Young1999). As more and more institutions were established for new, interconnected issues, the global governance landscape became increasingly crowded (Brown Weiss Reference Brown Weiss1993; Young Reference Young1996). Institutions started interacting and forming interlinkages, which has become a central analytical challenge (Young Reference Young2002; J. Kim Reference Kim2004; Oberthür and Gehring Reference Oberthür and Gehring2006; UNEP 2007, 2012, 2019; Chambers Reference Chambers2008; Pattberg Reference Pattberg2010; Oberthür and Stokke Reference Oberthür and Stokke2011; Pattberg et al. Reference Pattberg, Chan, Sanderink, Widerberg, Jordan, Huitema, van Asselt and Forster2018). Eventually, such interlinkages gave rise to larger complexes of interlocking institutions or actor configurations (Kanie, Andresen and Haas Reference Kanie, Andresen and Haas2013). While institutions in these complexes are only loosely coupled, they nonetheless often formed dense clusters that are sparsely connected to other clusters. As such, not every institution is connected to all others uniformly, just as some governance issues relate more closely to one than to others. Naturally, distinguishable clusters appear, and each of these inherently emergent, macro-scale modular structures is what we refer to as an architecture of global governance.

Research on governance architectures is marked by a remarkable co-evolution of analytical and normative debates, theory and practice, scholarly discourse and calls for political reform. Theoretical debates – for instance on the orchestration effects of intergovernmental organizations – go hand in hand with elaborate calls for institutional reforms, for example for the creation of a world environment organization or the amendment of the Charter of the United Nations. Debates in theory and practice often run in parallel; at times, they are interrelated, for instance when the urge of policy dilemmas and inconsistent governmental actions stimulates further advancement and research in academia.

The analytical research focuses here on studying the structures of global governance architectures and explaining their institutional variation and varying degrees of performance. One example are studies on the fragmentation of global governance architectures (see Chapter 8). These studies investigate the causes and consequences of governance fragmentation that results, for instance, from international agreements being negotiated by specialized government ministries in forums that are detached from negotiations of other agreements. Some scholars have argued that the current architecture of earth system governance is not conducive to the development of coordinated and synergistic approaches to collective problem-solving in the face of increasing global interdependencies (Young Reference Young2008 and Reference Young2019). Such research findings have had significant policy implications and impact, as evident in the emphasis on policy and institutional coherence in the United Nations agreement in 2015 on seventeen ‘Sustainable Development Goals’ (see Chapter 12; see also Nilsson, Griggs and Visbeck Reference Nilsson, Griggs and Visbeck2016).

The normative debate focuses on how architectures should be redesigned or reconstructed in a way that improves the performance of institutions in achieving their primary objectives. A key guiding consideration has been the ‘problem of fit’ of architectures and institutions to underlying problems (Young Reference Young2002; Folke et al. Reference Folke, Pritchard, Berkes, Colding and Svedin2007; Galaz et al. Reference Galaz, Olsson, Hahn, Folke, Svedin, Young, King and Schroeder2008; Ekstrom and Young Reference Ekstrom and Young2009; Ekstrom and Crona Reference Ekstrom and Crona2017). The logic behind this approach is that ‘[m]ore effective architectures of earth system governance may come about by better linking the study of nature with the study of governance’ (Biermann et al. Reference Biermann, Betsill and Gupta2009a: 82). Politically, the challenge here is how to shape governance architectures in a way that they better fit the structure and dynamics of the regulatory target, including at the level of the earth system (Kim and Mackey Reference Kim and Mackey2014). This needs interdisciplinary research that draws, for instance, on the insights of earth system scientists who study the structure of the earth system (Schellnhuber et al. Reference Schellnhuber, Crutzen, Clark, Claussen and Held2004; Steffen et al. Reference Steffen, Sanderson and Tyson2004; Reid et al. Reference Reid, Chen and Goldfarb2010) and how different ecosystems and natural processes are interacting (Rockström et al. Reference Rockström, Steffen and Noone2009; Steffen et al. Reference Steffen, Richardson and Rockström2015, Reference Steffen, Rockström and Richardson2018). Such research needs to inform the work of governance scholars who explore the institutional implications for the broader architectures of global governance.Footnote ³ However, the analytical and normative dimensions of research on architecture are not mutually exclusive. Instead, they cut across the interface between earth system analysis and governance theories (Biermann Reference Biermann2007).

Building Blocks, Structural Features and Policy Responses

Any global governance architecture consists of building blocks that share structural features and that are subject to a variety of policy interventions by governments and other international actors (see Figure 1.1). This book has been organized around these building blocks (Part I), structural features (Part II) and policy responses (Part III).

Figure 1.1: Architectures of global governance

Despite the rise of non-state actors and transnational institutions, the central actors in global governance are still the governments of sovereign states. The literature on the state as actor in global governance is endless; hence we did not include a separate chapter on state actors but refer instead to the comprehensive review by Compagnon and colleagues (Reference Compagnon, Chan, Mert, Biermann and Pattberg2012) in an earlier synthesis project. The first major building blocks that we are discussing in this book are international institutions that governments create to jointly tackle common problems. In world politics, such institutions with shared norms, principles and rules and decision-making procedures are often referred to as regimes (see Chapter 2). International institutions function through international bureaucracies that governments set up to administer and support such institutions. Such bureaucracies can be, for instance, treaty secretariats, such as the relatively large secretariat in support of the United Nations Framework Convention on Climate Change. Also the administrative apparatus of the World Bank or the International Maritime Organization fall into this broad category. While international bureaucracies formally function only to serve the interests of national governments, recent research has compellingly shown that international bureaucracies often enjoy large degrees of political freedom and play independent roles in global governance (Hickmann et al. Reference Hickmann, Widerberg, Lederer and Pattberg2019). They are hence important building blocks of global governance architectures with some degree of autonomy (see Chapter 3).

Furthermore, non-state actors have gained much influence in global governance, along with traditional state actors. In some areas of global governance, such as forest governance, non-state actors even seem to play central governing roles. Importantly, non-state actors often create their own transnational institutions, which we define as institutions set up by private actors with only a marginal role for governments or without any involvement of governments. These transnational institutions and networks are another major building block in any global governance architecture that will be discussed in this book in a separate chapter (see Chapter 4). Finally, vast areas of our planet are beyond the jurisdiction and effective control of governments, including the high seas and Antarctica, along with outer space. Other areas are diffuse, virtual and difficult to control by governments, such as the Internet. Thus, this book includes a chapter on institutions operating beyond the jurisdiction or control of national governments, as a final major building block of architectures of earth system governance (see Chapter 5).

These building blocks interact within broader global governance architectures in multiple ways. This has led in the last decade to a prolific research strand that we report on in Part II of this book. The first chapter in that part focuses on the interlinkages between two or only a few institutions (see Chapter 6). The following chapter reviews research on the emergence of larger sets of interrelated and interdependent institutions, often described as regime complexes (see Chapter 7). In studying an entire governance architecture in which institutions interact, the third chapter of Part II investigates variant degrees of fragmentation in global governance architectures (see Chapter 8).

Finally, these institutional interlinkages and the fragmentation of global governance architectures have forced governments, non-state actors, the decision-making bodies of international conventions and treaties as well as leaders of international bureaucracies to develop a variety of policy responses (see Part III of this book). These responses in turn have led to substantial research programmes, which again have often informed policy decisions and pushed forward reforms. Typical responses by state representatives, international bureaucrats and treaty bodies have been, for example, the push for the mutual integration of policies through the advancement of systematic policy integration or mainstreaming (see Chapter 9).

In situations of institutional interlinkages – especially when norm conflicts and weak governance performance were identified – governments and treaty bodies reacted by purposeful attempts at interplay management (see Chapter 10). This research is inevitably linked to the study of institutional interlinkages, which is discussed by Hickmann and colleagues earlier in this book (see Chapter 6). While research on institutional interlinkages analyzes the effects of such interactions – often with a focus on identifying and explaining synergistic or conflicting interactions – the research on interplay management investigates the effectiveness of policy interventions to minimize disruptive effects of conflictive interlinkages.

A new strand of research explores the role of international bureaucracies, but also other actors, as orchestrators in global governance. If a governance architecture can be likened to the ensemble of musicians in an orchestra, international bureaucracies – such as the United Nations Environment Programme – might then be the conductors, or the orchestrators, through non-hierarchical, non-compulsory soft ways of global institutional steering (see Chapter 11). A special case, much related to orchestration, is the policy of negotiating global goals as joint, widely accepted standards of conduct, with the expectation that public and private actors align their behaviour with such goals. The 17 Sustainable Development Goals, agreed by the United Nations General Assembly in 2015, are the most prominent example of such ‘governing through goals’ (see Chapter 12). In some cases, however, governments might choose to reform the broader institutional architectures beyond simple interplay management and orchestration by more structural change that establishes hierarchical orders among institutions and actors in global governance. We call this policy strategy hierarchization, acknowledging that such a strategy might evolve in numerous ways (see Chapter 13).

Figure 1.1 presents a schematic diagram of the architectures of global governance as we conceptualize them in this book. As mentioned above, the concept of governance architecture resembles the macro level structure and offers a bird’s-eye view on the global governance landscape. While the overarching global governance architecture comprises the entirety of actors and institutions, individual governance architectures within a given domain may consist of one or more regime complexes, which are meso-level structures. When only one regime complex operates within a global governance domain, it would equate with an architecture. Key elements of a regime complex in turn include, at the micro level, intergovernmental institutions such as international treaties and organizations, as well as transnational networks and non-governmental organizations. These elementary building blocks along with emergent meso- and macrostructures form complicated interlinkages with one another and interact both horizontally and vertically. The increasing diversity and multiplicity of the building blocks and structural features has made global governance more and more complex over time. Such a conceptualization of global governance architectures underscores the fact that any attempt at structural transformation must be informed by a deep understanding of the intricate system of actors, norms, institutions and their networks.

Performance

The first question that the Earth System Governance Project sought to address over the last decade was, how is the performance of individual environmental institutions – and of clusters of institutions – affected by the fact that institutions are embedded in larger architectures?

The performance of architectures can be measured in multiple ways. A central focus in earth system governance research has been the problem-solving effectiveness of architectures (Breitmeier, Underdal and Young Reference Breitmeier, Underdal and Young2011; Young Reference Young2011a, Reference Young2018). Other studies have also investigated other variables of governance performance, such as legitimacy, accountability and equity (Gupta and Lebel Reference Gupta and Lebel2010; Biermann and Gupta Reference Biermann and Gupta2011; Karlsson-Vinkhuyzen and McGee Reference Karlsson-Vinkhuyzen and McGee2013). Research on performance in terms of problem-solving (output legitimacy) and on performance in terms of democratic and transparent procedures (input legitimacy) are deeply related, however (Scharpf Reference Scharpf1999). As argued by Oran Young (Reference Young2011a: 19855), the ‘[m]aintenance of feelings of fairness and legitimacy is important to effectiveness, especially in cases where success requires active participation on the part of the members of the group over time’.

The performance of a global governance architecture is generally expected to be shaped by certain institutional qualities. One defining characteristic of global governance architectures has been described as fragmentation (see Chapter 8), and many studies have sought to address key questions around the causes and consequences of, and responses to, the fragmentation of governance architectures (e.g., Zelli Reference Zelli2011; Van de Graaf Reference Van de Graaf2013; Zelli and van Asselt Reference Zelli and van Asselt2013). Similar situations have been described from a different perspective as regime complexes (see Chapter 7), or as situations of institutional complexity or polycentricity (Jordan et al. Reference Jordan, Huitema, van Asselt and Forster2018b). As for these characteristics of an architecture, the debate about the consequences of different degrees of fragmentation or the complexity of broader governance architectures has been most controversial. The key question has been whether high degrees of fragmentation or complexity of an architecture affect its performance as well as key institutions operating therein. However, the research community is far from conclusive in the current state of debate. Essentially, there are four broad clusters of assessments.

1. (1) First, in situations of fragmented architectures or convoluted regime complexes, there is agreement that institutions or actors are not uniformly affected by this broader architecture. For example, effects on the performance of an international institution depend on where the institution is located in an institutional architecture. One may ask, is the institution in the centre or periphery or does it assume a bottleneck position or not? Recent research has shown, for instance, that the texts and decisions of multilateral agreements with central positions attract far more citations in later agreements, hence spreading their rules or expanding legal effects (Kim Reference Kim2013). Furthermore, the centrality of an international organization’s position in a network is positively associated with the output of that organization (Murdie Reference Murdie2013). These institutional networks serve like a public good, and central positions are associated with a higher level of social capital (Young Reference Young2017). This also implies that an institution’s performance depends not only on its position within an architecture, but also on the characteristics of that architecture. Social network analysis is one useful tool to study such complex arrangements (Kim Reference Kim2019; see also Chapters 6 and 14).

2. (2) Second, a cluster of studies expects a strongly fragmented governance architecture to be less effective than a more integrated architecture (M. Young Reference Young2012; Biermann Reference Biermann2014; van Asselt Reference van Asselt2014). Here, the argument is that disconnects between international institutions operating in silos result in limited cooperation or coordination among them. Most institutional interlinkages are believed to be disruptive or conflictive in character rather than synergistic or cooperative (Biermann et al. Reference Biermann, Pattberg, van Asselt and Zelli2009b). More specifically, institutional fragmentation is seen as hindering progress in negotiating stringent targets and actions; this limits incentives of subnational actors to take urgently needed action and reduces the overall credibility, stability and coherence of the entire architecture of global governance (Biermann Reference Biermann2014: 83–97). Some researchers observe that fragmented governance architectures are intentionally created by states that benefit from them and maintained by power asymmetry, and in turn unevenly privilege more powerful countries (Benvenisti and Downs Reference Benvenisti and Downs2007; Alter and Meunier Reference Alter and Meunier2009). International ‘forum-shopping’ by astute national governments for pursuing their parochial interests is a case in point (Murphy and Kellow Reference Murphy and Kellow2013).

The fragmentation of governance architectures is also seen as causing, or as being unable to prevent, negative spillovers and externalities (Ostrom Reference Ostrom2010a; Truelove et al. Reference Truelove, Carrico, Weber, Raimi and Vandenbergh2014). For example, because global climate governance is fragmented, carbon leakages may occur if businesses transfer production to countries with lower emission restrictions (Carlarne Reference Carlarne2008; Eichner and Pethig Reference Eichner and Pethig2011). An emerging concept here is environmental problem-shifting, referring to instances where policies to solve one environmental problem transfer damage elsewhere or transform one type of pollution into another (Yang et al. Reference Yang, Bae, Kim and Suh2012; van den Bergh et al. Reference Van den Bergh, Folke, Polasky, Scheffer and Steffen2015). Problem-shifting is increasingly observed between international environmental regimes, presenting a risk to the overall performance of earth system governance (Kim and Bosselmann Reference Kim and Bosselmann2013; Kim and van Asselt Reference Kim, van Asselt, Morgera and Kulovesi2016).

1. (3) Third, an opposing line of studies offers a more positive assessment of the performance of governance architectures that are highly fragmentated or complex. For instance, scholars have claimed that fragmentation is ubiquitous and inherent to any governance architecture (Zelli Reference Zelli2011; Kim Reference Kim2013; Pauwelyn Reference Pauwelyn2014) and that in most systems everything is not connected (directly) to everything else, while most actors and institutions can be reached in a small number of steps. There is some evidence that the density of most systems is low, which can make a governance architecture appear as if it is excessively fragmented yet without significantly affecting its functioning.

Furthermore, contrary to the assumption that institutional proliferation leads to further fragmentation, some authors suggest the presence of defragmentation as a self-organized counteracting process that allows for keeping some cohesiveness in governance architectures despite the increasing number of institutions (Kim Reference Kim2013). This process is also responsible for creating modular structures that are internally densely connected but externally sparsely connected through ‘weak ties’ (Granovetter Reference Granovetter1973). A seemingly sparse and fragmented architecture could hence also be a well-functioning system (Stevens Reference Stevens2018). In this view, institutional fragmentation does not mean anarchy (Galaz et al. Reference Galaz, Crona, Österblom, Olsson and Folke2012) and there could be also ‘healthy’ fragmentation (Kanie Reference Kanie, Swart and Perry2007). For example, regime complexes consisting of loosely coupled, partially overlapping institutions are found in some studies as being more adaptable over time and flexible across issues when compared to other more integrated international regimes (Keohane and Victor Reference Keohane and Victor2011).

1. (4) Fourth, a more recent position emerged in the middle ground between these opposing poles of the debate. According to these scholars, the problem is ‘not fragmentation per se, but rather the coordination (or lack of it) of fragmented or differentiated institutions’ (Zürn and Faude Reference Zürn and Faude2013: 120). What matters then is the character or nature of the institutional interplay and interplay management, which is ‘just as likely to produce positive or even synergistic results as it is to lead to interference between or among regimes’ (Young Reference Young2011a: 19856; see also Oberthür and Pożarowska Reference Oberthür and Pożarowska2013). The key question is to what extent conflicts between regimes can be resolved through a better division of labour, or through negotiations that lead to mutual accommodation (see Chapter 10). Often, this line of research draws on an overtly optimistic view of governments that carry a general interest in complementarity within global governance architectures (Gehring and Faude Reference Gehring and Faude2014), especially when international cooperation in one regime impedes international cooperation in another (Johnson and Urpelainen Reference Johnson and Urpelainen2012).

The debate on the implications of institutional fragmentation for global governance is far from resolved. Attempts are being made to test various theories, for example through a comparison across architectures or over time, and explain their performance in relation to the degree of fragmentation or other structural features (see Chapter 8). But we do not expect a broad agreement on how best to improve the performance of governance architectures soon.

Multilevel Governance

The second question emphasized by the Earth System Governance Project in 2009 was, what is the relative performance of different types of multilevel governance architectures? This question entails the comparative analysis of governance systems as multilevel structures where many interacting authorities are at work, including local, national, regional and global (Zürn Reference Zürn and Levi-Faur2012; also Peel, Golden and Keenan Reference Peel, Godden and Keenan2012). The analytical focus has been on vertical institutional interaction and variability in the performance of institutions within such multilevel governance systems (Biermann Reference Biermann2014: 83–97). Regime complexes of various international institutions could, for example, be conceptualized and analyzed as multilevel governance architectures to the extent that there is a vertical dimension between institutions (Abbott Reference Abbott2014; Hickmann Reference Hickmann2017). The same holds of course for the interaction when regional integration organizations – like the European Union – provide a further level of governance between states and international institutions.

Systematic comparisons of multilevel governance architectures require typologies of such architectures. Often the vertical dimension is unpacked along two axes: levels of political jurisdiction, from global institutions to local administrations, and spatial scale (Van Doren et al. Reference Van Doren, Driessen, Runhaar and Giezen2018) (with scale generally seen as the spatial, temporal, quantitative or analytical dimensions to measure and study any phenomenon, and levels as units of analysis located at different positions on a scale, see Gibson et al. Reference Gibson, Ostrom and Ahn2000; Cash et al. Reference Cash, Adger and Berkes2006: 2). Scale and level determine the frame within which architectures are designed, contested and evaluated. A core question in the study of multilevel governance architectures is, therefore, how problems are framed in terms of scale and level and what implications the framing has for the performance of institutions.

For example, research has sought to identify the appropriate scale at which to address a problem such as climate change or ocean acidification (Kim Reference Kim2012; Billé et al. Reference Billé, Kelly and Biastoch2013). As for climate change, for instance, it has been concluded that ‘local thinking must be coupled with global and national scales of action to achieve the levels of carbon dioxide reductions needed to avoid dangerous climate impacts’ (Sovacool and Brown Reference Sovacool and Brown2009). Some analysts have employed the neologism of ‘glocal’ to refer to characteristics of both local and global considerations, for example in multilevel water governance (Gupta, Pahl-Wostl and Zondervan Reference Gupta, Pahl-Wostl and Zondervan2013). Furthermore, scholars focusing on polycentric governance argue that climate change should not be framed and approached exclusively as a global issue. They argue that actors and authorities at all levels should take initiatives for, among others, policy experimentation (Hoffmann Reference Hoffmann2011; see also Ostrom Reference Ostrom2010b; Galaz et al. Reference Galaz, Crona, Österblom, Olsson and Folke2012; Betsill et al. Reference Betsill, Dubash, Paterson, van Asselt, Vihma and Winkler2015; Dorsch and Flachsland Reference Dorsch and Flachsland2017). In fact, such a polycentric arrangement, where top-down and bottom-up processes are simultaneously at play, is becoming widespread. In global climate governance, for instance, we can observe both governance through legally binding, multilateral agreements (Hare et al. Reference Hare, Stockwell, Flachsland and Oberthür2010) and governance through voluntary commitments and non-state action (Chan et al. Reference Chan, van Asselt and Hale2015).

Successful policy measures can be translated to other levels or contexts through scaling (Hoffmann Reference Hoffmann2011). But under what circumstances can policies be scaled up and down? Context-specific barriers exist that hamper scaling of successful initiatives, and it is important that these are identified and addressed (Van Doren et al. Reference Van Doren, Giezen, Driessen and Runhaar2016; Fuhr, Hickmann and Kern Reference Fuhr, Hickmann and Kern2018). Furthermore, scaling processes change how policy measures and instruments operate and may also render them less effective. Likewise, it is important to understand whether there can be consistent architectures if policies cannot be easily scaled up and down administrative levels (Biermann et al. Reference Biermann, Betsill and Gupta2009a). Key conditions identified in the literature include the presence of boundary organizations, such as the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services, which act as brokers connecting institutions across scales or levels (Gupta, Pistorius and Vijge Reference Gupta, Pistorius and Vijge2016; Holzscheiter Reference Holzscheiter2017; Morin et al. Reference Morin, Pauwelyn and Hollway2017).

In this regard, the role of subnational or transnational regions has received new attention in multilevel governance (Balsiger and VanDeveer Reference Balsiger and VanDeveer2012). Regional agreements such as the Convention for the Protection of the Marine Environment of the North-East Atlantic make up two-thirds of all international environmental agreements. Some of these institutions form regional governance architectures that have been mapped and analyzed (Fidelman and Ekstrom Reference Fidelman and Ekstrom2012). While the significance of regional governance has often been overshadowed by global counterparts such as the United Nations Convention on the Law of the Sea, researchers have rediscovered the important role that regional agreements play in vertically and horizontally linking governance levels, especially as intermediary institutions bridging the national and the global (Balsiger and Prys Reference Balsiger and Prys2016).

Fundamental Norms

The third architecture question that the Earth System Governance Project formulated in 2009 was, what are overarching and crosscutting norms of earth system governance?

Not all norms are equal in their status, scope or strength. A relatively small number of norms run through most, if not all, institutions and architectures of global governance. One prominent example is the widely accepted norm of the precautionary approach, which requires states to take measures to prevent serious or irreversible environmental harm despite scientific uncertainty. Such overarching or crosscutting norms have formed the normative basis on which many multilateral institutions are negotiated, agreed and implemented. Research in this field has sought: (1) to show such overarching or crosscutting norms that are common to many institutions in the field of earth system governance; (2) to analyze and understand key functions they perform; and (3) to explore ways in which we can bring about progressive development of these norms. These three research foci are discussed below in turn.

1. (1) First, regarding norm identification, two different approaches have been employed. For identifying overarching norms, researchers have assessed the legal status of candidate norms. Here, one would ask whether a norm has reached a certain status in international law, for example, as customary international law or a general principle of international law. This approach is typically taken up by legal scholars who assume a hierarchy among norms, with some norms being superior or more widely applicable than others (Shelton Reference Shelton2006). For identifying crosscutting norms, researchers have looked for norms that appear as central across a range of global governance architectures. Here, the analytical concern is less on the actual legal status of a norm, but rather on the extent to which it appears as a constitutional principle or a basic norm. An example of such a crosscutting norm is the duty to maintain or restore the integrity of the earth’s ecosystem (Kim and Bosselmann Reference Kim and Bosselmann2013; Bridgewater, Kim and Bosselmann Reference Bridgewater, Kim and Bosselmann2014). Here the argument is that, although there is no explicit recognition among governments that global ecological integrity is a higher order norm than others, it is nonetheless an emergent ‘common denominator’ that all international soft law instruments such as the 1992 United Nations Declaration on Environment and Development have in their core (Kim and Bosselmann Reference Kim and Bosselmann2013).

It should be noted that this norm identification, sometimes referred to as codification, is done not only by academics but also by policymakers and other practitioners. Probably the most important organization with a codification mandate is the International Law Commission of the United Nations. In earth system governance, the International Union for Conservation of Nature has played an unofficial but significant role through the drafting and updating of an International Covenant on Environment and Development since 1995. Most recently, the United Nations General Assembly has initiated an intergovernmental process for developing a framework agreement for international environmental law, the Global Pact for the Environment. This pact is intended to entrench key fundamental principles that could serve to integrate the fragmented field of earth system governance (see Chapter 13; see also Kotzé and French Reference Kotzé and French2018).

1. (2) Second, researchers have been interested in determining key functions that these overarching or crosscutting norms perform. One key function is the steering or adjudication of international institutions (Hickmann Reference Hickmann2017). There is an emerging consensus in the literature that basic norms are key ingredients for maintaining some degree of polycentric order in global governance architectures (Jordan et al. Reference Jordan, Huitema, van Asselt and Forster2018b). Here, overarching or crosscutting norms may ‘provide a means to settle disputes and reduce the level of discord between units to a manageable level’ (Jordan et al. Reference Jordan, Huitema, Schoenefeld, van Asselt, Forster, Jordan, Huitema, van Asselt and Forster2018a: 19). Therefore, these overarching or crosscutting norms may also contribute to orchestration efforts of international organizations and coordinate the plethora of international institutions (see Chapter 11). In this light, some scholars argue that the absence of a clearly defined goal of international environmental law, for example, is conducive to conflicts and competition between international environmental regimes. In other words, the apparent ineffectiveness of earth system governance could be attributed in significant part to the weak constitutional character of international environmental law (Kim and Bosselmann Reference Kim and Bosselmann2013).

Although the absence of an overarching norm is likely to be a key barrier for international institutions to function in a mutually beneficial manner, this theoretical proposition has yet to be empirically tested and supported. One area in which empirical evidence can be gathered are the Sustainable Development Goals (see Chapter 12). These goals are an outcome of political processes that codified overarching and crosscutting norms of earth system governance (see Chapter 13; see also Biermann, Kanie and Kim Reference Biermann, Kanie and Kim2017). Will these global goals have any measurable effects on integrating and harmonizing policies that point in different directions into a more coherent whole? If so, what are the key mechanisms of governance through goals? Some see the Sustainable Development Goals as a well-integrated network of targets that will help coordinate competing policy objectives (Le Blanc Reference Le Blanc2015). Others are more cautious and argue that the goals merely reflect the fragmented set of existing commitments (Kim Reference Kim2016; see also Kim and Bosselmann Reference Kim and Bosselmann2015; Underdal and Kim Reference Underdal, Kim, Kanie and Biermann2017; Young et al. Reference Young, Underdal, Kanie, Kim, Kanie and Biermann2017). They argue that it is imperative to agree on one overarching goal for the Sustainable Development Goals (Costanza et al. Reference Costanza, McGlade, Lovins and Kubiszewski2015).

1. (3) Third, earth system governance research has considered the more normative question of how the current set of overarching or crosscutting norms should be reconfigured or reinterpreted in light of rapid global change. For example, researchers have begun to engage cautiously in prescribing which norms should be recognized as most fundamental, especially in the face of challenges of the Anthropocene (Robinson Reference Robinson2014). Here, governance scholars often draw on scientific understandings of the earth system. The concept of planetary boundaries (Rockström et al. Reference Rockström, Steffen and Noone2009; Steffen et al. Reference Steffen, Richardson and Rockström2015) has been particularly useful for legal and political theorizing regarding limitations of the sovereignty of nations (Biermann Reference Biermann2012; Kim and Bosselmann Reference Kim and Bosselmann2013; Ebbesson Reference Ebbesson2014; Kotzé and Kim Reference Kotzé and Kim2019). For example, certain standards that are fundamental to protecting planetary boundaries could be accepted by the international community as jus cogens norms, from which no derogation is permitted. The restrictions on emissions of ozone-depleting substances, in accordance to the provisions of the Montreal Protocol, is a good example of what could be recognized as such a peremptory norm in the Anthropocene (Biermann Reference Biermann2014: 112–8).

There has been a considerable amount of political and academic effort to facilitate and formalize the recognition of, and respect for, planetary boundaries as a fundamental norm of earth system governance. Examples include the lobbying by scientists in the run-up to the 2012 United Nations Conference on Sustainable Development (‘Rio+20’) to insert planetary boundaries into the 2030 Agenda for Sustainable Development as well as a proposal for a framework convention on planetary boundaries (Fernández and Malwé Reference Fernández and Malwé2018). However, the recognition of planetary boundaries has been perceived as conflicting with some other fundamental norms including global social equity (Steffen and Stafford-Smith Reference Steffen and Stafford-Smith2013). Consequently, the effort did not succeed, and the 2030 Agenda does not include any reference to planetary boundaries. Nonetheless, the Sustainable Development Goals reflect a novel normative development of global governance, and some scholars have put forward the argument that some of the existing but outdated norms and principles of international law should be reinterpreted in light of the new political consensus reached by the international community (Kim Reference Kim2017).

Interaction between Policy Domains

The fourth question emphasized by the Earth System Governance Project was, what are the environmental consequences of non-environmental governance systems? Earth system governance is not just about protecting the environment but more broadly about achieving sustainable development. Therefore, institutional interactions between environmental institutions and those in other policy domains such as trade, energy, transport and agriculture become critical. Mapping of the interaction between the Sustainable Development Goals and a corresponding call for institutional ‘championing of one or a subset of the goals is a recent example (Nilsson, Griggs and Visbeck Reference Nilsson, Griggs and Visbeck2016). Research has been organized around analytical and normative dimensions: first, to understand the nature of interaction between institutions in different policy domains (Gehring and Oberthür Reference Gehring and Oberthür2009) and second, to strike a balance between environmental and non-environmental policy objectives and to address trade-offs (Boyle Reference Boyle, Bodansky, Brunnée and Hey2007; Pauwelyn Reference Pauwelyn2007).

One of the most researched cross-domain interactions is between trade and environment. Traditionally, trade liberalization was viewed as undermining global environmental governance (Conca Reference Conca2000). These scholars condemn free trade agreements that favour intensive production and long-distance transportation for causing additional pollution. They also fear that trade agreements can limit the regulatory capacity of political leaders to enact environmental regulations that frustrate the interest of exporters and investors (Morin and Bialais Reference Morin and Bialais2018). In particular, researchers have found potential conflicts between climate measures and the law of the World Trade Organization and examined to what extent the principle of integration might address trade-offs (Voigt Reference Voigt2009; Kim Reference Kim2016).

More recently, however, the nexus between trade and environment has changed through a new generation of preferential trade agreements from the 2000s, such as the 2009 United States–Peru Trade Promotion Agreement (Morin, Pauwelyn and Hollway Reference Morin, Pauwelyn and Hollway2017). These agreements have created strategic linkages between trade and environment (Jinnah and Morgera Reference Jinnah and Morgera2013) through including environmental provisions (Morin, Dür and Lechner Reference Morin, Dür and Lechner2018). Contrary to traditional wisdom, research on the new trade–environment nexus suggests the possibility of trade agreements having positive implications for consolidating the effectiveness of environmental agreements (Morin and Bialais Reference Morin and Bialais2018). The dynamism of the trade regime, in particular, has been identified as a leverage point for strengthening environmental governance (Morin and Jinnah Reference Jinnah2018).

Other non-environmental governance systems studied for their impact on the environment include food and energy (Boas, Biermann and Kanie Reference Boas, Biermann and Kanie2016). As for food, the research community has investigated, for example, how threats to food security and associated norms of the right to food influence negotiations on the architectures of earth system governance (Ziervogel and Ericksen Reference Ziervogel and Ericksen2010). The governance of the nitrogen and phosphorus cycles has also emerged as an important topic for addressing the food and environment nexus (Ahlström and Cornell Reference Ahlström and Cornell2018; Morseletto Reference Morseletto2019). Furthermore, policies around controversial energy sources such as biofuel and hydropower have come under scrutiny for their negative consequences for environmental concerns (Pittock Reference Pittock2010; Scott et al. Reference Scott, Hitchner, Maclin and Dammert2014).

Although research has so far pointed to some positive trends, the general finding still is that more policy integration or coordination is needed across policy domains. Then, the question arises how to respond to negative consequences of non-environmental governance systems on the environment. One key political response is environmental policy integration (see Chapter 9), which refers to the incorporation of environmental goals into non-environmental policy sectors with the aim of targeting the underlying driving forces rather than merely symptoms of environmental degradation (Persson et al. Reference Persson, Runhaar, Karlsson-Vinkhuyzen, Mullally, Russel and Widmer2018; see also Jordan and Lenschow Reference Jordan and Lenschow2010). More recently, the concept of mainstreaming has become popular in research as well, for example regarding the mainstreaming of climate and biodiversity policies (e.g., Nunan, Campell and Foster Reference Nunan, Campbell and Foster2012; Runhaar, Driessen and Uitenbroek Reference Runhaar, Driessen and Uittenbroek2014; Karlsson-Vinkhuyzen et al. Reference Karlsson-Vinkhuyzen, Kok, Visseren-Hamakers and Termeer2017; Runhaar et al. Reference Runhaar, Wilk, Persson, Uittenbroek and Wamsler2018).

Ultimately, any response to interactions between policy domains entails a normative decision. Sustainable development is widely seen as the overarching objective of the international society; how to reconcile the three pillars of sustainable development – society, economy and environment – remains far from clear. Some argue that environmental concerns must be prioritized because the earth and its natural resources are finite; others argue that socio-economic concerns are as important. The latest version of this long-standing debate is the tension between the ‘ecological ceiling’ and the ‘social foundation’ (Raworth Reference Raworth2017). A group of scholars sought to ease the stalemate by proposing the protection of the earth’s life-support systems and poverty reduction as the twin priorities (Griggs et al. Reference Griggs, Stafford-Smith and Gaffney2013). Yet, critics argue that the twin priorities simply reflect the dichotomy between (environmental) sustainability and development, which has always been the core of the problem with the concept of sustainable development (Kim and Bosselmann Reference Kim and Bosselmann2015).

Architectural Voids

Finally, in 2009 the Earth System Governance Project called for more research on how to explain instances of non-governance, or the occurrence of ‘voids’ in the governance architecture.

This line of research started from the observation that in earth system governance we often do not have specialized institutions in areas or on issues that are widely identified as being problematic. For example, deforestation proceeds with tremendous speed, yet no global treaty exists to tackle this issue. While governments agreed in 1992 on a global convention on halting climate change and on protecting biological diversity, no treaty on deforestation was negotiated, even though this issue was also on the agenda at that time. How can we explain such instances of ‘non-governance’ or ‘architectural voids’? While the puzzle of the lacking agreement on global deforestation has attracted some attention by researchers and found compelling explanations (Dimitrov Reference Dimitrov2005), numerous other issues of non-governance are still insufficiently understood. Therefore, the study of governance architecture must also explore the gaps and empty spaces in or between them (Dimitrov, Sprinz and DiGiusto Reference Dimitrov, Sprinz and DiGiusto2007). In comparative research designs, studies on non-governance will also help better explain the emergence and performance of those institutions that have been agreed. Moreover, research on why there are no global institutions in some areas will further our understanding of the consequences of governance architectures. For example, the absence of a global biofuel regime has created a risky North-South allocation pattern (Lima and Gupta Reference Lima and Gupta2013).

Some studies in this direction have improved understanding on why we do not have a world environment organization despite 40 years of political struggles (Vijge Reference Vijge2013), why there is no comprehensive law of the atmosphere (Sand Reference Sand2017) and why we have no comprehensive treaty that covers the Arctic region (Young Reference Young2011b). However, there are still areas where our understanding is lacking. For instance, natural scientists have identified nine areas where they believe that our global system is facing planetary boundaries with fundamental value for human survival. While some of these boundaries are heavily institutionalized – for example by the climate convention of 1992 – others are hardly subject to global policy, notably the planetary boundaries on nitrogen and phosphorus (Ahlström and Cornell Reference Ahlström and Cornell2018; Morseletto Reference Morseletto2019) and ocean acidification (Kim Reference Kim2012). In addition, issues such as the global spread of plastics, especially into the marine environment, have been addressed in earth system governance only recently (Dauvergne Reference Dauvergne2018; Nielsen, Holmberg and Stripple Reference Nielsen, Holmberg and Stripple2019). In short, the research programme on global non-governance is by far not concluded.

Two reasons, however, might explain why non-governance is less prominent and possibly also less problematic. First, institutions are becoming more adaptive. New and emerging problems do keep surfacing, potentially creating governance gaps. But in most cases, these gaps do not last for very long, with existing institutions quickly adapting to fill them. For example, ocean acidification has emerged as a new global issue. While there is no specialized treaty on this issue, the problem has nonetheless been taken up by the climate regime (Campbell et al. Reference Campbell, Gray and Fairbanks2016). Similarly, the issue of deforestation has been taken up by the international climate convention since the introduction of Reducing Emissions from Deforestation and forest Degradation (REDD+) in 2005 (Gupta, Pistorius and Vijge Reference Gupta, Pistorius and Vijge2016). Another example is the debate on geoengineering, that is, artificially cooling the planet by for instance spraying aerosols into the atmosphere to reflect parts of the sunlight or various other technologies. One such technology – the ‘fertilization’ of oceans to increase the uptake of carbon dioxide from the atmosphere – quickly came under scrutiny because of the unknown environmental risks and side effects. In 2008, parties to the Convention on Biological Diversity agreed on a moratorium on ocean fertilization, and the issue has remained a priority topic as well for the London Convention on dumping of wastes at sea (Pasztor, Scharf and Schmidt Reference Pasztor, Scharf and Schmidt2017), both halting global non-governance at least in the area of ocean fertilization.

Second, in areas of seeming non-governance by states, we often see informal and unconventional forms of steering. The Sustainable Development Goals, as a novel and broad governing mechanism, are one example that leaves little space unaffected by the goals (see Chapter 12; see also Kanie and Biermann Reference Kanie and Biermann2017). Some also argue that, contrary to the popular claim that geoengineering is still a largely ungoverned space (Nicholson, Jinnah and Gillespie Reference Nicholson, Jinnah and Gillespie2017; Jinnah Reference Jinnah2018), certain types of geoengineering are governed through ‘unacknowledged sources of steering’ such as high-level authoritative assessments (Gupta and Möller Reference Gupta and Möller2018: 1). This point also reflects the emergence of new non-governmental actors, mechanisms and institutions, which go beyond traditional forms of state-led treaty-based regimes and that fill in areas marked by governmental non-action (Biermann and Pattberg Reference Biermann and Pattberg2008; Biermann Reference Biermann2010).