It's ordered chaos: what really makes polycentrism work

This forum reimagines polycentric governance. It develops ideas of “or-dered polycentrism” that can help international relations scholarship make fuller sense of contemporary governance of global affairs. How can we theorize the implicit bonding forces that bring deeper order to the surface disorganization of polycentric governance? We offer a key corrective to actor-focused institutionalist understandings by showing how polycentrism also involves deeper relations and structures. Six contributions offer various avenues to theorize deeper order in polycentric governance, each with reference to a substantive issue area. Jens Steffek draws upon constructivist theory of “norms” to argue that standards acquire autonomous ordering power in polycentric governance of global business. Maryam Deloffre adopts a “metagovernance” perspective to identify norms as aspirational visions structuring the regulation of humanitarian assistance. Next, Frank Gadinger explores polycentrism through the lens of “practices” that organize the everyday activities by multiple actors such as negotiating as well as the objects, technologies and expertise they use in these governance efforts. Zeynep Mencutek highlights “techniques” as micro-carriers of ordering practices in polycentric governance of irregular migration, stretching the limits of institutional rules. Maria Koinova discusses “informality” as a deeper structuring force in the governance of transit migration and diasporas, and how it is shaped

by state capacities, political regimes, and regional dynamics.Finally, Jan Aart Scholte adds "underlying order" through macro-frameworks and, with illustrations from Internet governance, suggests that polycentrism is structured through a threefold combination of norms, practices, and underlying orders.Together, the six commentaries offer a menu of ways that future research can explore order in what institutionalism has depicted as chaos.

Introduction
As reflected in several recent contributions to International Studies Review, contemporary international relations (IR) scholarship is much occupied with the messiness of current institutional arrangements for addressing global challenges (Kim 2020;Orsini et al. 2020).How can contemporary society treat planetary problems such as pandemics, ecological changes, migration, cybersecurity, and so on when governance is so "fragmented" (Biermann et al. 2009), "complex" (Pattberg and Zelli 2016), and "liquid" (Krisch 2017)?
In this ISR forum, we engage with the notion of "polycentrism" as a particularly promising concept in this genre.More than other options, the term "polycentrism" directly conveys that contemporary governing occurs through "many sites."In addition, "polycentrism" deftly blends connotations of diffusion ("poly") and order ("centrism"), as mirrors this forum's core argument that governance of global affairs is aptly understood as "ordered chaos."The more challenging question-and our key contribution here-is how to theorize that ordering.
Regarding issue areas, research to date has especially invoked polycentrism to study climate governance (Ostrom 2014;Ahlström and Cornell 2018;Jordan et al. 2018).This forum broadens the application to other issues of high relevance to IR, such as security, humanitarian assistance, international migration, standardization, and Internet governance.Polycentrism works in all these areas through the ordering power of social relationships.Indeed, comparative analysis of polycentrism across different policy fields presents a propitious avenue for future research.
As a general definition, "polycentric" governing occurs when "many centers" address a given policy concern.The diffuse decision points can be scattered across multiple scales (local, national, regional, and global) and various sectors (public, private, and hybrid).The participating organizations in a polycentric arrangement often have overlapping mandates, ambiguous hierarchies of authority, and no ultimate arbiter.Continual creation and reconstruction of institutions and relationships among them also tend to make polycentric governing processes quite fluid.
At the level of actors (i.e., the administrative organizations that govern), polycentrism looks quite disordered, and this feature is usually the headline story for institutionalist analysis in IR.1 From a narrow actor-focused perspective, the main challenges facing "chaotic" polycentric governance include accountability, compliance, coordination, efficiency, forum shopping, and gridlock (Black 2008;Hale, Held, and Young 2013;Murphy and Kellow 2013;Abbott et al. 2015).The typical institutionalist answer to these problems is "orchestration," "managing complexity," and other steering strategies to bring administrative "order" to the disarray (Oberthür and Pożarowska 2013; Abbott et al. 2015).
We suggest that focusing on relational aspects of polycentric systems, rather than on the actors themselves, shows that polycentrism may not be as disorganized as it seems at first blush.Polanyi (1951, 176) characterized polycentricity as a system of "self-coordination," albeit without greatly detailing the dynamics involved.The Ostroms asked how authority relationships and distributions of power could affect outcomes of polycentric processes, albeit likewise without much specifying these ordering patterns (Ostrom 2010a;2015).
In IR, institutionalist treatments of polycentrism have neglected these relational components and questions of deeper order.These accounts understand governance-i.e., societal rules and regulatory processes-to consist (only) of tangible directives and organizations.Institutionalist scholars, therefore, depict polycentrism in terms of immediately perceptible phenomena, that is, explicit regulations and concrete administrative frameworks.Institutionalists break a polycentric aggregate down into actor parts and presume that actor properties (such as intentions, perceptions, choices, initiatives, tactics, and energies) drive the process of governing (Axelrod 1997)."Order" in this rationalist perspective consists of formal agreements between governing bodies, as are indeed often absent in situations of polycentrism.Thus, a focus on actors-on behavioral units to the exclusion of behavioral patterns-can make polycentric governing appear chaotic.
Yet polycentric situations, like the ones addressed in this forum, can have order without formal central authority.While attention to actor attributes offers important insights into polycentric governing, the problem is that institutionalist analysis reduces governance to nothing more than explicit rules and concrete regulatory organizations.Our forum moves polycentrism from a focus on actors and their nested positions (which obscures the politics of how these fields work) to an examination of how social ties are built, maintained, and disrupted through norms, practices, techniques, informality, and underlying macro-orders.
Appreciating the need for this corrective, Kim's recent ISR article (2020) prioritizes the issue of "structure" for future research on decentered governance.Drawing on network theory, Kim distinguishes six ways to plot relationships between the institutions in a governance system.The article presents diagrams where nodes represent the various governance organizations and lines show the interactions between them.The pictographs suggest that this "structural" arrangement of the institutions matters, that is, how the actors are clustered, how far those clusters have hubs, and how much the institutions interact.Although not done in Kim's analysis, other network images adjust the length and breadth of the connecting lines in order to designate the intensity of contacts between the institutions (Murdie 2014).
While ideas of network structure mark an advance on institutionalist notions of "chaotic" polycentrism, these conceptions of order also fall short.In particular, as Kim himself acknowledges, this approach to "structure" leaves unanswered the crucial question of what gives the network its binding glue.What is the content of the lines that connect?Since the lines are what holds the parts together in a polycentric whole, the features and forces of these fasteners matter.
This forum picks up this challenge by asking how we can theorize the qualities that bring deeper organization to polycentric governance.What are the implicit and intangible bonding forces that bring greater coherence to the apparent incoherence?In a word, what really makes polycentrism work?
The following six contributions explore various ways to conceptualize deeper order in polycentric governance.We argue there is nothing inherent in polycentrism that produces coordination and order.The essays explore the ordering power of social relationships through norms, micro-patterns of practice, and macro-frameworks of social structure that generate governance effects, "organizing" chaos and making polycentricity work.In doing so, we bridge discussions of polycentricity with the broader IR literature and social theory.We start with explicit frameworks, in terms of norms; move to implicit micro-patterns of practice, techniques, and informality; and finish with macro-structures of social order.Each piece illustrates its approach to "ordering chaos" with reference to a different substantive issue area.
More specifically, in the first commentary, Jens Steffek draws upon constructivist theory of "norms" to argue that standards acquire autonomous ordering power in polycentric governance of global business.The second contribution, from Maryam Deloffre, develops the notion of "metagovernance" to identify patterns that structure the regulation of humanitarian assistance.Next, Frank Gadinger explores ordering of polycentrism through "practices," as seen in the field of maritime security.A fourth piece, from Zeynep Mencutek, highlights "techniques" as micro-carriers of ordering practices in polycentric governance of irregular migration.Maria Koinova discusses "informality" as a deeper structuring force in the governance of transit migration and diasporas.Finally, Jan Aart Scholte adds "underlying order" through macro-frameworks and, with illustrations from Internet governance, suggests that polycentrism is structured through a threefold combination of norms, practices, and underlying orders.
Together, the six commentaries offer a menu of ways that future IR research can explore order in what institutionalism has depicted as chaos.We offer a key corrective to actor-focused institutionalism by showing how "coordination" in world politics also involves deeper patterns and logics.To be sure, (ordered) polycentrism is more a metatheoretical heuristic than a full-fledged explanatory formula.Rather than aspiring to provide a new integrated paradigm, this forum introduces a range of propositions that subsequent research can explore further, singly or in combinations.

Technical University of Darmstadt, Germany
This contribution argues that knowledge of polycentric governance could profit from a more profound engagement with the structuring power of international norms.I draw on global standard-setting by the International Organization for Standardization (ISO) to illustrate the merits and limitations of polycentric governance as an analytical concept.
Standards are a type of international norm that is powerful in guiding the choices of private companies, technical experts, and national regulators.ISO standards create orientation in a complex world, shape expectations about the behavior of others, and help promote uniformity across borders.Standards govern through economic incentives, cognitive salience, and perceptions of appropriateness.As an institution, ISO is an important node in a polycentric governance network of global dimensions that formulates and promotes such standards.Once they have emerged from that polycentric network, however, transnational standards emancipate from the institutions that created them and spread (or fail to do so) through countless individual decisions.This mode of governing "at a distance" is diffuse and largely decentered as standards acquire autonomous ordering power.Standards, like other types of transnational norms, hover above the polycentric institutional landscape rather than being in any actor's possession.

Ordering Polycentrism through Norms
Polycentricity is a spatial metaphor that describes decentered governance in a certain issue area.The conceptual opposite would be centralized governance that is organized around a single institution where decision-making is concentrated.The polycentric landscape, by contrast, consists of several subcenters of governance that are connected through repeated and patterned interactions.Kim (2020) depicts polycentricity as a network with multiple nodes.These nodes have some centrality in the network, each with a cluster of other actors or institutions revolving around them.
The spatial metaphor of polycentricity and the analytical focus on patterned interactions direct our attention to the actors of governance (mostly bureaucratic organizations) and their exchanges with each other.The implicit assumption is that once we know which actors are more central to the network and which are at the margins, then we will understand how governance functions in the respective issue areas.We might even be able to judge if the polycentric governance arrangement is more or less functional, in the sense of both efficacy and efficiency, for performing its designated tasks (Ahlström and Cornell 2018).
Such models presume that (configurations of) institutional actors matter most for our understanding of polycentric governance.We can challenge this often implicit assumption by highlighting the role of norms in polycentric governance.Constructivists in IR have long studied the power of norms.Challenging neorealists and rational-choice institutionalists in the late twentieth century, constructivists argued that norms "mattered" in international affairs vis-à-vis considerations of state power and economic self-interest (Kratochwil 1984;Wendt 1995;Tannenwald 1999;Evangelista 2001).Importantly, constructivists also shifted the analytical focus in IR Downloaded from https://academic.oup.com/isr/advance-article/doi/10.1093/isr/viab030/6403480 by Universiteit Leiden / LUMC user on 04 November 2021 away from the actors in governance and toward the social structures that emerge as a result of continuing actor interactions and mutual perceptions (Wendt 1999).
The power of transnational norms is harder to measure than that of material resources such as nuclear arsenals or gross domestic product.Moreover, norms are moving targets.It is often not easy to trace how they spread across the world and where they actually "reside" (Cortell and Davis 1996;Florini 1996;Finnemore and Sikkink 1998).The constructivist answer, adopted from sociology, is that norms are intersubjectively shared social facts (Finnemore and Sikkink 2001, 393).Only when norms are widely shared can they shape collective expectations about how actors will behave (and ought to behave) in certain situations.Norms thus reside in a social community where they are known, supported, and sometimes enforced.The size of these social communities differs greatly.Some norms only affect certain professions.Others hold within a national society but hardly beyond it.Other norms have truly global reach, such as the customary norms of international law.
Constructivist insights on norms can be applied profitably to the discussion of polycentric governance and the problem of global order.Much of the governing in polycentric governance is done through norms, as shown also in Deloffre's contribution below.The problem with institutionalist conceptions of polycentric governance is that they seek to localize norms in specific actors.However, as social facts, norms hover above the polycentric actor landscape, rather than having a locale within it.Of course, norms need actors as carriers: to propagate them, to develop them further, or even to enforce them.Norms, as Thomas Risse once put it, "do not float freely," but travel through social networks (Risse-Kappen 1994, 195-96).Norms have patrons in society, in terms of individuals and institutions within a polycentric governance network.A discussion of global standards and their role in governance can help us explore this idea further.

Standards as Ordering Norms
Standards are a special type of norm that extensively (though not exclusively) targets the behavior of business actors.Standards govern economic activity in many fields: sometimes as product standards that describe the desired quality of a good and sometimes as process standards that determine how a production operation should function (Timmermans and Epstein 2010).Global standards are negotiated in networks of specialists who work in certain standard-setting institutions.Although the term "polycentric" is rarely used in this field, standard-producing networks can be described as such.Some analysts employ network approaches and mapping exercises to document the enormous variety of actors involved in the standards area (Perry and Noelke 2005;Richardson 2009).These complex polycentric networks include both public and private actors, in particular businesses to whom the standards apply.
In the global arena, ISO represents a node of some centrality within a polycentric network.Rather than being a monolithic institution, ISO brings together specialists from national standard-setting bodies (Murphy and Yates 2009).Like other organizations that produce standards, ISO's internal procedures are consensus-oriented and the resulting products are not formally binding.They require voluntary adherence.
While global standards emerge from a polycentric governance network of institutions, some features of governing through standards defy an actor-centered imagery.These structural qualities arise not during formulation but in the implementation phase of governance.Once transnational standards are codified, they work quite independently of the institutions that produced them, as those organizations cannot enforce them.
Take for instance ISO 9001, a prominent transnational standard that is widely applied and regularly updated (Heras-Saizarbitoria, Boiral, and Allur 2018, 6-7).This voluntary standard contains minimum requirements for quality management procedures and is designed for use by a broad range of private companies and public sector institutions.Adherence to ISO 9001 is audited through third-party certification.If certification is successful, institutions can advertise their compliance with the standard to customers, suppliers, and other stakeholders.
Whether or not businesses seek ISO 9001 certification largely depends on expected benefits in terms of efficiency and competitiveness, which are set against the administrative and financial costs of obtaining certification.Material cost-benefit considerations explain compliance with ISO 9001 to a large degree, at least in the private sector.In contrast, compliance with other standards, such as ISO 14001 on environmental impact assessment, involves at least partly altruistic motivations, specifically a commitment to the ideal of sustainability (Sartor et al. 2019).Thus, the overall governance capacity of ISO standards involves a myriad of decentralized decisions to (not) comply with them.
However, this multitude of decisions is not "chaotic."The level of adherence to a transnational standard within the peer group usually matters greatly.This general support signals-and here we come back to constructivist insights about social facts-the degree to which a norm is entrenched in the relevant constituency and shapes its expectations.Early adopters may still have ample discretion whether or not to follow a standard.However, once peers widely endorse it, compliance is expected and in practice becomes hard to avoid.This self-sustaining feature of standards makes them attractive as a governance tool, especially for liberals (who generally prefer voluntary compliance to enforcement) and especially in the transnational arena (where enforcement of norms is difficult anyway).The example of ISO standards shows that to understand governance through norms, we need to move beyond the polycentric network of specialists who negotiate and codify them.We need also to consider the diffusion of norms in a vast and "centerless" population of implementing actors.
Such dynamics not only arise around ISO standards and the business sector, but also figure more generally in what Finnemore and Sikkink have termed the "norm life cycle" (1998,.In the early phase of such a cycle, polycentric networks of norm entrepreneurs draw up and propagate a new norm.If this norm finds resonance, a "cascade" sets in that broadly diffuses the norm across a transnational community.In consequence, as Finnemore and Sikkink put it, "norms may become so widely accepted that they are internalized by actors and achieve a "taken-forgranted" quality that makes conformance with the norm almost automatic" (1998,904).At this point, actors tend to follow a well-established norm because of its cognitive salience and may not even consider alternative courses of action anymore.

Conclusion
The concept of polycentricity seems a useful heuristic to analyze the formulation of transnational standards through decentered networks of experts and institutions.The polycentric imagery helpfully directs our attention to the complex interplay of actors.However, the institutionalist focus on the (would-be) governors becomes less helpful if we want to understand how some standards become deeply ingrained while others do not.Such outcomes depend largely on the embedding of a standard in the relevant community.Once standards are widely shared, they can order the polycentric landscape with expectations of appropriateness and peer pressure.What is true for the diffusion of standards seems to hold for the spread of norms more generally.While processes of norm adoption and diffusion are substantially decentered-as they take place simultaneously in many quarters-they are also ordered through the relatively autonomous power of entrenched norms.This "centerless" but regularized process is a critical contribution that IR constructivism can make to the conception of polycentric governance.

Metagovernance Norms and Polycentricity in
Global Humanitarian Governance George Washington University, USA Scholarship on polycentric governance, particularly the work of Elinor and Vincent Ostrom, assumes an "organized chaos" where actors recognize and are aware of each other, coordinate their activities, and operate under a set of overarching rules (Ostrom 2010a, b;2014).Drawing on the US federal system, their original formulation viewed these overarching rules as legal frameworks and formal governance systems at national, state, and local levels (Ostrom, Tiebout, and Warren 1961).However, when polycentricity is applied to global policy areas-which lack a formal central authority-it is unclear which shared understandings, rules, or laws coordinate actors.I examine the role of metagovernance norms in generating shared obligations, relationships, and regulatory effects in polycentric fields.I argue that norms of metagovernance-defined as the governance of governance (Jessop 2003)-are constitutive of actor interests and behavior, and thereby produce regulatory effects in polycentric contexts.Metagovernance norms specify an aspirational vision for the governance of a policy area and generate shared expectations for actor behavior.For example, in development policy, sustainability is a metagovernance norm that prescribes what desirable development policy looks like.The metagovernance norm produces the expectation that development actors will strive for sustainability and coordinate their activities around sustainable practices.
The global humanitarian field is polycentric in that a vast array of state and nonstate actors work across subnational, national, regional, and global scales to supply emergency relief to millions of people each year.In 2019 alone, 61.3 million people across twenty-two countries received humanitarian assistance provided by thousands of organizations (OCHA 2019).Until the late 1990s, little agreement existed on principles, norms, and standards for humanitarian action.With no central authority serving as a gatekeeper, a constant proliferation of inexperienced actors wanting to "do good" hindered collective action, increased inefficiency, and lowered the quality of humanitarian assistance.The humanitarian system was polycentric and chaotic.
During the 1990s, humanitarian actors developed normative and operational coordination mechanisms to improve the quality and effectiveness of humanitarian action.To this end, the United Nations (UN) founded the Office for the Coordination of Humanitarian Assistance (OCHA) in 1991 and the European Union (EU) founded a Directorate-General for European Civil Protection and Humanitarian Aid Operations (ECHO) in 1992.Yet neither OCHA nor ECHO had the capacity to coordinate and monitor the multitude of actors-ranging from community organizations to church groups to local and international NGOs-who are the first responders in humanitarian crises.This lack of coordination was particularly noticeable in the international response to the refugee crisis following the genocide in Rwanda, which prompted deep introspection in the sector (Deloffre 2016).In debating what constitutes "good" humanitarian action, humanitarians generated agreement around metagovernance norms to providing humanitarian assistance in ways that do no harm (DNH) and are accountable to affected populations (AAP).

Metagovernance Norms
As stressed throughout this forum, a focus on actors does not tell us whether and how order is constituted in polycentric policy fields.Like Steffek in this forum, Downloaded from https://academic.oup.com/isr/advance-article/doi/10.1093/isr/viab030/6403480 by Universiteit Leiden / LUMC user on 04 November 2021 I examine how norms constitute actor identities, relationships, activities, and practices.However, whereas Steffek considers the ordering effects of the ISO standard, I demonstrate the regulatory effects of metagovernance norms, such as DNH and AAP.
Metagovernance norms are overarching rules that govern a policy area.They develop through reflexive and iterative processes during which actors discuss, formulate, and implement the values and principles of governance (Kooiman and Jentoft 2009;Sørensen and Torfing 2009).Individuals and organizations define good governance and develop rules, standards, and procedures to regulate their policy arena (Sørensen and Torfing 2009;Holzscheiter, Bahr, and Pantzerhielm 2016).These interorganizational interactions generate shared interests and identities, as well as convergence around field-level metagovernance norms.These norms in turn produce patterns of stable relationships and define what activities should be governed, by whom, and how (Deloffre 2016;Holzscheiter, Bahr, and Pantzerhielm 2016).Actors do not have predetermined preferences in terms of what these norms are; the norms are produced through social interactions and discussions (Deloffre 2016).

Regulating Chaos in the Humanitarian Sector
The following brief sketch of the development of collective accountability standards shows how defining and operationalizing metagovernance norms generate regulatory effects (See also Deloffre 2010Deloffre , 2016, n.d.;, n.d.;Krause 2014;Kennedy 2019).Accountability includes the processes and practices through which an actor reports on and answers for its conduct to those parties whom it affects (Scholte 2011).In the humanitarian context, accountability relationships are multifaceted and nested, given the many actors with overlapping jurisdictions.For example, an international NGO is legally and contractually accountable to home and host governments, principals, partners, and staff; financially accountable to donors, governance boards, and members; and socially accountable to peers and affected populations.Populations receiving humanitarian assistance are most affected by NGO actions but have the least power to hold them accountable.
In the 1990s, humanitarian NGOs established shared transnational principles and standards to improve their performance and accountability.Prior to these efforts, standard-setting in the humanitarian sector occurred primarily at the organizational or associational level (Buchanan-Smith 2003;Walker and Purdin 2004).These initiatives included the Red Cross Code of Conduct, the Humanitarian Accountability Partnership-International (HAP-I), COMPAS Qualité, and Sphere.All reflected a shift from financial and legal accountability to powerful donors to dialogical or social accountability to those affected by humanitarian action (Deloffre n.d.; Crack 2018).
The process of developing these initiatives involved defining both accountable for what and to whom.The debates on accountable for what centered on two conflicting views of moral obligation, one drawing on duty-based ethics (good intentions) and another on consequentialist ethics (DNH) (Barnett and Weiss 2008;Deloffre 2010).The DNH principle, which focuses attention on impact rather than intentions (Anderson 1999), ultimately prevailed as a core metagovernance norm for humanitarian action.DNH has attracted wide adoption, including by the United Nations High Commissioner for Refugees (UNHCR), the Organization for Economic Cooperation and Development (OECD), and the EU (OECD 2001(OECD , 2010;;EU 2008;UNHCR 2020).
NGOs also considered their accountability relationships, which tended to emphasize financial and contractual accountability to donors and states, and shifted focus to being accountable to crisis-and conflict-affected populations.AAP thus emerged as a second core metagovernance norm in the field.Focusing on accountability to affected people does not mean that humanitarians neglect other key stakeholders Downloaded from https://academic.oup.com/isr/advance-article/doi/10.1093/isr/viab030/6403480 by Universiteit Leiden / LUMC user on 04 November 2021 such as donors or staff, but they try to center their policy formulation and execution on the recipients of assistance.
Debating how to implement DNH and AAP resulted in numerous collective standards projects.Humanitarian NGOs working on Sphere defined effective humanitarian action in rights-based terms that uphold a human right to humanitarian assistance.Sphere's Humanitarian Charter sets forth a series of common principles (humanity, humanitarian imperative), rights (to life with dignity, to receive humanitarian assistance, to protection), and duties derived from international law (Sphere 2018).By 2000, the UN, the OECD, and donor governments adopted and incorporated Sphere standards and indicators in their appeals, funding, and assessments processes.At the subnational level, local NGOs cite the Sphere standards as important both for designing humanitarian programming and for creating a humanitarian space in which to work (De Geoffroy and Grunewald 2017; Howe, Munive, and Rosenstock 2019).
NGO-led efforts on DNH and AAP encouraged other humanitarian actors to develop related standards.For example, in 2003, states founded the Good Humanitarian Donorship (GHD) initiative to define principles of good humanitarian action.GHD adopted some of the same language on the right to humanitarian assistance and the ways that donors can contribute to improving sector accountability.In 2007, the European Consensus on Humanitarian Aid referenced the Red Cross Code of Conduct and Sphere in its discussion of how to ensure the quality and accountability of humanitarian assistance.Finally, in 2015, several initiatives merged to form the Core Humanitarian Standard (CHS) Alliance, which developed a meta-standard for the sector and works with the Humanitarian Quality Assurance Initiative to certify organizations against the meta-standard.A recent analysis of policy trends from 1996 to 2020 shows that accountability has received consistently high attention during this period (Alexander and Parker 2020).
The preceding sketch shows how metagovernance norms of DNH and AAP have steered and shaped interorganizational relationships to coordinate actor behavior in a polycentric setting.Whereas the sector was earlier termed an "aid circus" which lacked explicit standards beyond basic humanitarian principles, the regulatory effects of these metagovernance norms now coordinate humanitarian activity around shared obligations.

Conclusion
Polycentric governance involves multiple actors working across scales and sectors to address public policy challenges.An essential, yet understudied, component of polycentric arrangements are the overarching rules that coordinate this governance activity.As this forum argues, the focus of earlier IR literature on formal institutions as authorities misses critical aspects of how mutual awareness and coordination occurs in polycentric contexts.
In the humanitarian sector, top-down, centralized coordination mechanisms have been unsuccessful in coordinating the activity of actors with different mandates and stakeholders (Stephenson 2006;Knox Clarke and Campbell 2018).Nevertheless, coordination does occur through normative commitments that are more flexible and adaptive to crisis settings (Knox Clarke and Campbell 2018).For example, organizations verified by the CHS Alliance consistently score highest in the area of "coordination and complementarity." My brief examination of the humanitarian field has shown how two metagovernance norms-do no harm and accountability to affected populations-have constituted humanitarian obligations, responsibilities, and principles.These norms have diffused beyond NGOs to other humanitarian actors and improved coordination of the field as a whole.While these metagovernance norms have not erased Downloaded from https://academic.oup.com/isr/advance-article/doi/10.1093/isr/viab030/6403480 by Universiteit Leiden / LUMC user on 04 November 2021 coordination problems or homogenized the sector, they have created a focal point around which humanitarian actors organize their activities.

University of Duisburg-Essen, Germany
There is a growing interest among IR scholars to study world politics through the lens of practice (Adler and Pouliot 2011;Bueger and Gadinger 2015).This trend has impacted recent debates on global governance, generating new research about how to handle the increasingly complex regulatory dynamics of today's rapidly changing world (e.g., Bueger 2018;Pouliot and Thérien 2018).Practice-oriented scholars suggest that we examine governance processes by focusing on practices rather than actors and their interests.Such a view understands governing as an activity and foregrounds the relational (i.e., continuously emerging) nature of governance arrangements.
In this contribution, I argue that a practice-oriented perspective is particularly useful to understand major features of polycentric governance processes, such as fluid authority and the blurring of the public and the private.Practice scholars use different conceptual vocabularies such as "assemblage" (Abrahamsen and Williams 2009), "actor-network" (Berger and Esguerra 2018), and "governmentality" (Joseph 2012).Nevertheless, despite following different research methodologies than institutionalist IR scholars, practice theorists speak to similar problems as those who use concepts such as "regime-complex" (Raustiala and Victor 2004), "fragmented architectures" (Biermann et al. 2009), and "multilevel governance" (Hooghe and Marks 2001).Thus institutionalist and practice analyses can richly complement each other.
A focus on polycentric governing practices has three major benefits.First, this approach shows how governance efforts materialize in and through practice.It highlights the relationships that multiple actors (re-)produce through their everyday activities (e.g., negotiating) and through the tools that they use (e.g., objects, technology, expertise).Second, a practice-oriented perspective subscribes to a relational notion of authority.This conception takes into account the emergence of new institutions and governance arrangements, their potential erosion, and the rise and fall of a political actor's authority.Third, the practice theory premise that "the mundane matters" acknowledges that polycentric governing is often driven by informal processes and ad hoc initiatives, which take place in a political and judicial grey zone and often have regulatory effects outside legal frameworks.
The argument of this commentary follows two steps.First, I briefly explain how practice-oriented perspectives bring novel contributions to understanding polycentric governance.Then I illustrate the conceptual argument by examining the case of a counter-piracy assemblage off the coast of Somalia (Bueger 2018).This example shows how management and demarcation practices stabilized cooperation efforts, organized state-industry relations, and provided temporary security in an area of extremely fragmented governance.

Polycentrism and the Promise of the Practice Turn
As outlined in the forum's introduction, institutionalist research and network analysis put more analytical emphasis on the nodes than the ties of governing arrangements-and are rather silent about the binding glue of polycentric apparatuses.This shortcoming provides the key entry point for practice-based perspectives to put life into rather static notions of relations among actors and institutions.In this approach, practices involve "embodied, materially mediated arrays of human activity centrally organized around shared practical understandings" (Schatzki 2001, 2).Practices not only organize our everyday life (including how we consume, work, and exercise), but also affect processes of governing in ways that other traditions of IR might overlook as trivial, such as benchmarking or filling out a bureaucratic form.
Analyzing practices implies observing, learning, and understanding the distinct everyday activities within a field (Bueger and Gadinger 2018).Researchers must sensitize themselves to the activities and relationships that organize governance efforts, such as the role of inanimate objects, expertise, and technologies.Material devices such as databases, indicators, and statistics are thus considered vital for governing (Freistein 2016).Practice theory focuses on everyday activities: for example, the use of debit cards by asylum seekers in refugee camps (Tazzioli 2019), security checks at airports by private guards (Berndtsson and Stern 2011), monitoring of commercial transactions in counterterrorism measures (de Goede 2018), and the use of trackchange in negotiating diplomatic documents (Adler-Nissen and Drieschova 2019).Such studies underline how material activities and technologies reorder publicprivate distinctions and reveal ambivalent features of polycentric governing in terms of hidden rule-making and elusive accountability.
Practice theory also analyzes power relations differently.For instance, Pouliot and Thérien (2018) show how practices such as NGO accreditation, expert groups, and multistakeholder partnerships have underlying power dynamics in social hierarchies and their effects of inclusion/exclusion.Stratification practices such as showing esprit de corps explain how pecking orders operate as informal governance structures (Pouliot 2016).In polycentric fields such as global health governance, a practice lens elucidates the loss of authority for the WHO as a single organization in competition with new private actors as well as its successful repositioning after adapting to the new (economic) rules of the game (Sending 2017, 320).Meanwhile, practice research in the vein of Pierre Bourdieu foregrounds the implicit dimension around symbolic power struggles (Kuus 2015).
Practice theory also sheds light on new modes of governance driven by experimentation, informality, and "low-cost institutions" (Abbott and Faude 2020; see also Koinova's contribution below).In moments of crisis, diverse actors are more concerned to devise collaborative solutions than to ensure compliance with formalized rules (see also Mencutek below on irregular migration).In my own example here, the fight against piracy demonstrates this experimental way of governing and aptly illustrates practice theory themes around relational authority and materiality.

The Making of the Counter-Piracy Assemblage
The resurgence of piracy off the coast of Somalia in the 2010s produced an urgent maritime security problem for numerous state authorities and private actors.After a series of incidents, the fight against piracy was organized around an informal cooperation initiative.This "assemblage" of disparate actors and measures around an overarching practice succeeded temporarily, as no major further incident was reported from 2012 to 2016.
As Bueger (2018) shows in a practice-oriented study that zooms in on the negotiation process and the following controversy, the main glue for this remarkable cooperation was the joint activity of working on a common informal framework, labeled as "Best Management Practices" (BMP).This tool organized state-industry relations and stabilized the "chaotic" polycentric complex of the cooperation efforts.Through BMP, actors with little prior experience of working together (e.g., navies Downloaded from https://academic.oup.com/isr/advance-article/doi/10.1093/isr/viab030/6403480 by Universiteit Leiden / LUMC user on 04 November 2021 from China, Russia, Japan, and NATO, as well as the shipping industry) aligned closely with established agencies like the International Maritime Organization.
A key site in this "counter-piracy assemblage" was the Contact Group on Piracy off the Coast of Somalia (CGPCS), which was established in 2009 as an informal governance mechanism after the UN Security Council provided the primary authorization for counter-piracy.Initially, only a limited forum, participation in the CGPCS quickly grew to over sixty states, international organizations, and NGOs.During the negotiation process, the so-called High-Risk Area was introduced in the BMP to signify sites with major liability to piracy attack.The BMP provided guidance documents for the shipping industry and prescribed a range of measures to limit the risk of attack.As Bueger (2018,625) explains, several practices maintained and strengthened the relations of the BMP.These practices were inscribed into material artifacts and formalized through endorsements and a multilateral declaration.Even physical artifacts were created, such as print copies of the BMP in a portable format, so that "every mariner in the world would carry the BMP in their pocket" (Bueger 2018, 626).Through these artifacts, the BMP could be circulated globally and the assemblage extended its reach.In such contexts, practice-oriented concepts such as "inscriptions" (Walters 2002) and "translations" (Berger and Esguerra 2018) enable innovative analysis of subject-object relations and the performativity of material artifacts.
The establishment of the BMP as a unifying regulatory tool for a polycentric complex was a success story.Yet, paradoxically, its success led to a controversy.The assumed apolitical nature of the practice sidelined some unresolved political and legal questions and led to the erosion of the counter-piracy assemblage when the problem became less urgent.Coastal states like Egypt and India complained about the definition of the High-Risk Area as they struggled with negative consequences (e.g., undermining the reputation of their ports).For its part, the shipping industry opposed any revision and claimed exclusive authorship over the BMP.
These politics around authority and legitimacy came to the fore as the BMP operates in a political and judicial grey zone.The industry group argued that states do not have the authority to request a revision, given that the BMP is a means of industry self-regulation (Bueger 2018, 629).The controversy revealed the hidden politics that often lurks in polycentric governance arrangements.Once the core object of governance was no longer the actual threat of piracy, but only the potential risk, it became difficult to sustain compliance (Bueger 2018, 631).While the BMP had initially stabilized cooperation, it became politicized around struggles over who governs maritime space and who provides the expertise for evaluating risk.
Zooming further out of the specific case of the BMP, private companies have rising authority in many fields of polycentric governing, blurring distinctions between public and private.The shipping industry is a particularly interesting case, as it involves a long history of seeking autonomy from the state.As far back as the medieval Hanseatic League, city-states and shipping industry guilds in Northwestern and Central Europe combined their interests of maritime security and independence to form a trading bloc.Hence, struggles about informal authority in polycentric governing networks are far from new.

Conclusion
Under conditions of increasing complexity in world politics, the relational perspective of practice theory provides helpful analytical entry points into the seeming disorder of polycentric governing.The core methodological move is to start the analysis with practices rather than actors.Practice-oriented perspectives provide new insights and fresh empirical results for the study of polycentrism.Everyday activities, material objects, and governing tools get more attention than in institutionalist governance research, and notions of relational authority come into sharper focus.Taken together, a practice-oriented analysis explains how an informal governance mechanism (such as the BMP) creates order in polycentrism, at least temporarily.A practice-based view sheds new light on the emergence and erosion of polycentric governance arrangements-often a blind spot in existing governance research.

Bonn International Center for Conversion, Germany and Ryerson University, Canada
This contribution focuses on techniques in polycentric governing.Building on policy studies and practice theories, I define techniques both as the operational components of practices and the means for translating policy ideas in discursive, material, or institutional forms (Bueger and Gadinger 2018;Simons and Voß 2018).2Techniques are a repertoire of possible action, flowing from policies, strategies, and practices.I argue that an emphasis on techniques-the micro-carriers of ordering practices-is useful to understand underlying structures of governance and "what really makes polycentrism work."As an empirical illustration, I focus on border control techniques in the governance of irregular migration.These tools aim to control, restrict, and deter the "movement of persons that take place outside the laws, regulations or international agreements governing the entry into or exits from the state of origin, transit or destination" (IOM 2019, 113).In migration governance, the multiplication of actors and regulations are repeatedly identified with descriptors such as "complex" (Scholten 2020), "fragmented" (King 2019), "multi-level" (Panizzon and van Riemsdijk 2019), and "multi-scalar" (Glick-Schiller 2015).There are repeated calls for decentering and pluralizing the understanding of international migration governance with reference to multiple actors involved (Triandafyllidou 2020;Kutz and Wolff 2021).Irregular migration increasingly becomes subject to polycentric governance as diverse authorities and centers of decision-making operate across various scales and sectors (Breugel and Scholten 2020).From a practice perspective, the governing of border controls by many centers is frequently studied with concepts of "securitization," "governmentality," "actor-network," and "assemblage" (Bigo 2002; Pallister-Wilkins 2015; Walters 2018).So far the lens of "polycentrism" has not been applied, but this approach can help to discern deeper patterns and logics of order as well as coordination mechanisms in the otherwise seemingly confused governance of irregular migration.An emphasis on techniques highlights the micro-carriers of these mechanisms.

Techniques in Polycentric Governing
Recent work in IR examines techniques within the global governmentality debate as a means to revisit mentalities, rationalities, and practices of international rule with reference to discipline, control, and power (Neumann and Sending 2010;Bigo 2011).Practice research in IR relates techniques to technology, materiality, and knowledge (Ruppert and Scheel 2019).Meanwhile, actor-network theory emphasizes the material aspect of techniques to explore power relations, as Gadinger discusses in this forum.Global governance scholarship on policy formulation also focuses on techniques as policy instruments (Beland and Howlett 2016;Simons and Voß 2018).
Techniques make governance practices work.To illustrate, agricultural management policies are implemented through techniques such as incentive programs or market instruments (Higgins, Dibden, and Cocklin 2012).Environmental governance works with evaluation techniques like the calculation of ecological footprints (Collins, Cowell, and Flynn 2009).In security governance, algorithmic techniques gather intelligence data for policing activities (Amoore and Raley 2017).
While policies are generally formulated through formal institutions, the techniques that implement policy can arise from a polycentric situation in which actors adapt to each other.In contrast to policies, techniques are often intentionally meant to be ambiguous, flexible, or ad hoc.They arise from judicial and administrative grey zones and are often kept deliberately vague.They may blur geographical scales and institutional hierarchies (Berndtsson and Stern 2011).
Techniques become especially relevant in responding to crises such as large irregular migration, wars, and other security problems.Such contexts more readily allow powerful actors to introduce governance innovations and new forms of relationships, including "informalization," to test what works for filling in "governance gaps" marked by coordination failures and gridlock (Peterson 2010).

Governance of Irregular Migration
The field of irregular migration well exemplifies polycentric governing.The diverse decision points are scattered across local, national, regional, and global scales.The latest global initiative on migration governance, the United Nations Global Compact for Safe, Orderly, and Regular Migration (GCM) illustrates this polycentrism (UN 2019).Preparation of the GCM involved dense interactions among multiple diffuse players including states, intergovernmental organizations, civil society associations, scientific institutions, parliaments, local authorities, the private sector, and migrants themselves.
Behind these disparate actors, however, the GCM reflects deeper order and logic in the governance of irregular migration, as manifested through techniques.The GCM invokes principles of national sovereignty, the rule of law, international law, and human rights as overarching ordering rules (UN 2019, 7).These principles are among "the prominent examples of embedded norms in contemporary governance" (Scholte contribution below).While these norms aim to protect (older) international order, the crisis perception by multiple actors encourages the construction of new norms.For example, the GCM introduces a novel metagovernance norm, namely "safe migration," which implies to "manage borders in an integrated, secure and coordinated manner" (UN 2019, 6).To achieve this objective, the GCM suggests cooperation around "border crossing procedures" and prescribes specific techniques such as "pre-screening of arriving persons, pre-reporting by carriers of passengers, the use of information and communications technology" (UN 2019, 20).The GCM envisions developing cooperation agreements that enable states to request and offer each other technical assistance to strengthen border control.
Techniques may take discursive, material, or institutional forms, which complement each other and contribute to the creation of order in migration governance.For example, discursive techniques reiterate the "illegality" of people on the move, as well as the "risk" brought by migration (Bigo 2011).The material techniques range from constructing fingerprint databases, biometric digital registration systems, border surveillance operations, border installations, automated border control systems, remote control tools, electronic phone tapping, risk profiling algorithms, and others (Broeders 2007; Amoore and Raley 2017).Tailor-made technical cooperation agreements, known as migration deals, widely deploy these techniques.In these ways, migration governance operates beyond formal institutions, negotiations, and explicit rules.
Border controls are polycentric, operating across many sites, at national and supranational scales, and through multiple practices and techniques.Moreover, border controls are increasingly operated through the outsourcing of some services to contractors, thereby bringing the private sector into governance (Davitti 2019).Three prevailing practices-namely, border surveillance, pre-screening, and forced returns-are strategically put into action through ambiguous techniques.
Border surveillance is the first step to prevent irregular crossing by deterring or returning migrants.Technology companies supply a network of vast databases for distinguishing between legal and irregular migrants during visa procedures, creating "digital borders" (Broeders 2007).Countries like Turkey, Libya, and Morocco build walls or offshore monitoring systems on their borders with funding from the EU and technical support from private firms.Security companies provide ships, aircraft, drones, and sensors to control crossings at land and sea (Nieto-Gomez 2014; Davitti 2019).The European Border and Coast Guard Agency (Frontex) is ambiguously neither a humanitarian nor a law enforcement agency.Frontex assists the border policing practices of national authorities, supports them in trainings and supplying surveillance material, all wrapped in a discourse of humanitarianism and protecting Europe (Pallister-Wilkins 2015, 65).To intercept migrant boats in so-called high-risk areas, Frontex uses the European Border Surveillance system, undertaking "real-time surveillance" that relies on data from aircraft, drones, and aerostatic balloons (Jeandesboz 2017).
If border surveillance techniques fail to prevent irregular migrants from crossing borders, then pre-screening techniques are implemented across a polycentric governance apparatus to impede the arrivals from staying and applying for asylum.Pre-screening involves registration, identification, and fingerprinting.These practices are similarly not centralized but governed by multiple actors, including UN and EU agencies, national border guards, local police and asylum units, reception centers, detention institutions, social services, NGOs, courts, companies, and individuals.When irregular arrivals were narrated as a migrant "crisis" in 2015-2016, the many governance actors increasingly experimented with novel techniques to stem migration, given limitations of the existing legal and institutional framework of the EU.
Since then, the European Commission (EC) has implemented the "hotspot" approach, a new policy that triages irregular migrants who are in need of international protection from those to be forcibly returned.Non-state actors, such as banks, software companies, and NGOs, working with IGOs and states, track migrants through case assistance programs and debit cards in Greece and Turkey (Tazzioli 2019).If pre-screening results in a decision for forced return, logistics need to be coordinated with the country of return.Frontex or national authorities make contracts with companies for return flights or ferries, although the legal regulations are unclear regarding the obligations of privately contracted actors toward migrants (DM 2018).As existing human rights and EU asylum legislation do not allow forced returns of migrants, a technical cooperation deal (the EU-Turkey Statement of March 18, 2016) provided the means (Mencutek 2018).
The EU considers pre-screening in hotspots and returns through deals to be "successful" techniques for ordering borders, even if these tools are not formally institutionalized, at least initially.The EU aims further to institutionalize these techniques through a new pact on Migration and Asylum, currently under discussion (Nielsen 2020).It involves a polycentric complex of EU agencies, national authorities, private actors (e.g., banks, security and transportation companies, smugglers).and migrants, all of them using or abiding by specific techniques.
Techniques are not necessarily anti-regulatory.Rather, they tend to deceive or stretch the limits of norms and institutional rules.They make notions of authority more fluid between national and supranational as well as public and private actors.The multiplication of governing centers through authority sharing and privatization Downloaded from https://academic.oup.com/isr/advance-article/doi/10.1093/isr/viab030/6403480 by Universiteit Leiden / LUMC user on 04 November 2021 allows states to avoid their "responsibility" of protection derived from the official international refugee and human rights regime.If found effective, techniques may take on a life of their own and eventually be accepted by formal institutions and turned into official regulations and norms.

Conclusion
Polycentrism is a useful heuristic to study extensive techniques of migration control across multiple scales and sectors and to analyze deeper patterns and logics in governance mechanisms.Techniques serve as a specific type of glue for polycentric arrangements, along with practices (Gadinger contribution above) and informal relations (Koinova contribution below).An emphasis on techniques enables us to identify micro "organizing frames" in the "ordered chaos" of polycentrism, enlarging our understanding of the dynamics of (mis)management of particular issues, such as migration as illustrated here.

University of Warwick, UK
Polycentric governance theories are rarely studied in the international politics of migration, refugees, and diasporas (with the exception of Breugel and Scholten 2020).This omission is not surprising, given that migration governance is usually associated with explicit policies of nation-states and intergovernmental organizations.Yet the field entails much more, with an abundance of policies across various scales and sectors, creating a seemingly chaotic picture.Such apparent disarray is especially visible in fields where institutional policies are ineffective, as in transit migration, or little institutionalized, as in diaspora engagement.Here governance often occurs in informal ways, filling the void of missing policies or challenging explicit rules.
The following piece enriches theories of polycentric governance with attention to informality.Polycentric governance works through repeated informal relationships that structure expectations in actor behaviors.This informality operates on different scales and is shaped contextually by state capacities, political regimes, and regional dynamics.While institutionalist accounts view informality as a disordering force for official governance arrangements, I argue on the contrary that informal relationships are constitutive of polycentric ordering.I illustrate these points with evidence from the Balkans and the Middle East, where transit migration and diaspora engagement are especially pronounced.

Beyond Institutionalist Perspectives
Migration governance theories focus predominantly on the formal politics of states and intergovernmental organizations."Formal" politics denotes exchanges based on explicit rules in legislation, policy documents, and bureaucratic processes."Informal" politics entails implicit exchanges based on "conventions and codes [of] behavior" (Yu 2007, 417).Informal relationships manifest themselves when unwritten agreements fill in for absent formal rules, when formal rules are ambiguous, when actual practices depart from formal rules (Kleine 2018, 876), and when bureaucracies circumvent political limitations on their autonomy (Roger 2020).While informality can occur in well-institutionalized environments, I focus here on policy Downloaded from https://academic.oup.com/isr/advance-article/doi/10.1093/isr/viab030/6403480 by Universiteit Leiden / LUMC user on 04 November 2021 areas where formal institutions are ineffective (transit migration) or underdeveloped (diaspora engagement).
Transit migration is broadly defined as "migrants having the intention to move onwards to a third country" (Wissink, Düvell, andVan Wedewijk 2013, 1087).It narrowly concerns persons holding a transit visa, but in practice involves much more.For example, under the EU Dublin Regulations, refugees can acquire asylum in the first receiving state, but many choose to travel further with reliance on smugglers.Transit migration is also pronounced among labor migrants who lose their contracts or work informally from the start.While states and international organizations govern transit migration through formal policies, enhanced border controls, and "externalization" (Collyer, Düvell, and de Haas 2012), such steps often turn largely ineffective (Perkowski and Squire 2019) and informality steps into the gap.
Diaspora governance is even less institutionalized.The World Bank measures remittance flows and the International Organization of Migration (IOM) creates ad hoc programs for transfer of expertise.Yet much extraterritorial diaspora governance rests with countries of origin.In a growing trend, diaspora ministries reach out for remittances, financial investments, cultural reproduction, and brain gain (Gamlen 2014).However, many original home-states encourage diaspora selfreliance through governmentality (Ragazzi 2009).Occasionally diasporas themselves become the agents of institutional reforms (Brinkerhoff 2016).A lot depends on how diaspora entrepreneurs become idiosyncratically involved in policy processes, especially in developing countries and in conflict and post-conflict contexts (Koinova 2017).

A Relational Perspective on Informality in Polycentric Governance
Informality is much more prominent in polycentric governance of transit migration and diaspora engagement than institutionalist theories in IR tend to think.Informality is vital to what Ostrom (2010b, 553) calls "cumulatively additive" polycentric relationships among states, international organizations, and private and non-state actors.In the migration settings studied here, repeated informal relationships create durable interaction modes, interdependencies, adaptation, and implicit organization among actors at different scales, thereby establishing governance that is not written in explicit rules.Underpinned by informal linkages, polycentric complexes could successfully govern with experimentation, entrepreneurship, and diffusion of knowledge and practices across different scales (Kim 2020).
Polycentric governance can suitably be called relational since it revolves around not just the attributes of actors involved, but also the relations among them.Interactions between social agents, when repeated regularly, form more or less durable social structures in polycentric governance.Related IR theories have analyzed relationships between states and non-state actors (Nexon 2009), transactional approaches to social movements (Stroschein 2012), international networks (Hafner-Burton and Montgomery 2006), and conflict-related path-dependencies (Goddard 2012;Koinova 2013).Durable interactions can establish informal frameworks as "socially shared rules, usually unwritten, that are created, communicated, and enforced outside of officially sanctioned channels" (Helmke and Levitsky 2004, 727).Such frameworks are context-specific.
Informal relationships among actors seeking to solve a real-world migration problem often arise in situations of weak state capacity.In fragile states, many in the Global South, where legislation is difficult to enforce and governments may not fully control a state's territory, informality can take a life of its own.In transit states, governmental and non-state actors may be subject to stringent policies in principle, but corruption can undermine implementation in practice.As a result, transit migrants, border agents, smugglers, and other non-state actors become interdependent and adapt to each other.They learn how far a formal policy would Downloaded from https://academic.oup.com/isr/advance-article/doi/10.1093/isr/viab030/6403480 by Universiteit Leiden / LUMC user on 04 November 2021 be implemented or could be informally circumvented and undermined.Similarly, regarding diasporas, countries of origin may have institutionalized policies but struggle to implement them if they have minimal state capacities.Diaspora may further prefer to avoid governmental influence and participate in homeland affairs on their own terms.Then, diaspora governance takes place informally by way of various actors in different global locations interacting durably beyond official rules.
In addition to state capacities, political regimes also shape informal relationships in migration governance.For example, in democratic regimes, relationships among actors are supposed to be relatively consensual, NGOs to be more autonomous, and all actors to be considerate of migrants' rights.In authoritarian regimes, by contrast, consensual deliberations are minimal and crude power politics prevails.Power relationships entice actors to cooperate, coerce, coopt, or simply tolerate others depending on the specific political regime.Although political regime types are not restricted to specific world regions, they often cluster regionally.Moreover, regional amities and enmities among states and non-state actors, as well as established regional interactions in trade, security, and migration, can shape how informal relationships evolve.

Informality in Polycentric Governance in the Balkans and the Middle East
We can illustrate the above points with reference to the large-scale refugee wave that occurred in 2014-2016 due to the war in Syria.On this occasion, an official EU corridor was opened in the Balkans for refugees to transit to Western Europe.The EU added to its earlier enlargement conditionality further pressures to contain illicit migration, notably toward Serbia and North Macedonia.
However, migration governance did not stop with these formal arrangements.Having gained more strength and autonomy with democratization, NGOs in the Balkans often informally supported transit migrants across borders, with or without collaboration from their governments.Moreover, even if Balkan states formally fought against migrant smuggling, with rampant corruption they often informally tolerated it.For example, in 2015, authorities in Belgrade district did little to remove visibly operating human traffickers (Beznec, Speer, and Mitrovic 2016).People smuggling became a booming business, producing annual revenues of €2 billion (BIEPAG 2016).Some border officials turned into gatekeepers, determining on their own terms whether and how migrants could move across borders (Perkowski and Squire 2019).At a grassroots level, informal relationships created mutual understandings among migrants, corrupt officials, and smugglers about the limitations of formal rules and the opportunities to operate beyond them.
Regarding diaspora engagement, Croatia, North Macedonia, Serbia, and Kosovo all created diaspora-related ministries.As democratizing countries, they established formal voting procedures for diaspora citizens, as well as policies for financial investment from and cultural reproduction in diaspora communities.
However, many aspects of diaspora governance remain informal.For example, diasporas informally supported fighting factions during the 1990s wars in former Yugoslavia.Later, during post-conflict reconstruction, they sent remittances, built housing, and made financial investments.Informal yet durable interactions emerged between diasporas and businesses, political parties, non-state actors, governmental agencies (with little capacity to sponsor diaspora initiatives), and international organizations (with no particular diaspora programs).Such interdependencies across different scales and global locations rendered diasporas important in polycentric governing.Examples of influential diasporas include Croatians in the United States, Macedonians in Australia, and Kosovo Albanians in Germany, Switzerland, and the United States.
In the Middle East, power politics dominates relationships among various actors involved in transit migration governance.Special deals with the EU prevail, introduced as tit-for-tat financial incentives.These measures include the 2016 EU-Turkey deal to stop irregular migration toward Europe and "migration compacts" with Jordan and Lebanon to host Syrian refugees.In contrast to the Balkans, authoritarian states of the Middle East refused to sign the 1951 Refugee Convention, thereby creating a permanent state of informality for refugees and transit migrants.Authoritarian states also left little space for NGOs to acquire autonomy and contest government practices.Many NGOs, including international ones, were informally pressured to follow the official government line in migration governance.As in the Balkans, governments officially resisted migrant smuggling but tacitly tolerated it.For example, a human smuggler argued: "The state turns a blind eye to the flow of migrants ... [because] migrants are a great source of income for Turkey" (Cicek 2018).In further contrast with the Balkans, non-state actors dealing with transit migration in the Middle East include radical formations, classed as terrorists by outside parties, such as Hamas in Palestine and Hezbollah in Lebanon.Many relationships of these groups remain covert or are conducted through proxies.
Regarding diaspora engagement, countries of origin in the Middle East have regularly considered diasporas in their calculations, but rarely through formal laws and institutions.Diasporas are courted primarily for financial remittances and investments, while diaspora voting is largely restricted.Much of the diaspora engagement-including support for conflicts and post-conflict reconstructionoccurs informally, not least because diasporas fear that the state in the country of origin might be monitoring them.Indeed, sending states such as Egypt, Iran, and Syria practice silencing of refugees and migrants abroad (Glasius 2018).

Conclusion
Informality plays an important part in the polycentric governance of migration, especially in areas where formal institutions are ineffective (transit migration) or underdeveloped (diaspora engagement).This informality manifests itself in various ways: for example, when non-state actors fill the gaps of dysfunctional policies; when government officials tacitly tolerate migrant smuggling; and when diasporas support their original homelands in ways that they themselves deem important, beyond the policies of sending states or international organizations.As examples above illustrate, informal relationships are constitutive of polycentric governance and vary by context, shaped by state capacities, political regimes, and world regions.
Informality is an important type of ordering force in polycentric governance, in addition to the norms, metagovernance, practices, and techniques discussed earlier in this forum.Repeated informal relationships create durable frameworks of expectations among participating actors about what is and is not possible in migration governance.In this way, too, polycentric governance has a deeper order underlying its surface chaos.stewardship transition, a reconstruction of the management of key technical functions that make possible a single global Internet.Protagonists from hundreds of associations representing innumerable interests swirled around us in near panic, as eighteen months of painstaking deliberations verged on implosion.Still, the serene veteran remained confident that a less perceptible order would see things through: it is organized chaos.Indeed, a year later, the IANA transition concluded without leaving so much as a ripple on the waters of world politics (Becker 2019;Palladino and Santaniello 2021).Perhaps polycentrism was not so disordered after all?Like preceding contributions to this forum, the following piece looks for structure in polycentrism, in this case taking its empirical cue from the Internet sphere.Governance of this global digital communications network is paradigmatically polycentric (Nye 2014;Carr 2015;Scholte 2017;DeNardis et al. 2020).The regulation of associated hardware, software, data, and content involves a multitude of interconnected sites: national and subnational governments, formal regional and global intergovernmental organizations, informal transgovernmental networks, multistakeholder arrangements, commercial enterprises, academic institutions, local community associations, and more.The myriad of regulatory measures includes laws from states, directives from the EU, resolutions from the UN, requests for comments (RFCs) from the Internet Engineering Task Force (IETF), number allocations from Internet registries, Unicode standards for text and emoji, take-down decisions from Facebook and WeChat, and much more.

Structuring Polycentrism: Norms, Practices and Underlying Orders in Internet Governance
This decentered governance apparatus bears all the hallmarks of polycentrism.It is trans-scalar (intertwining local, national, regional, and global institutions) and trans-sectoral (moving across public, private, and hybrid organizations).It involves diffuse decision points and fluid policy processes.The participating agencies have many overlapping mandates, ambiguous hierarchies of authority, and no ultimate arbiter.
So is there deeper order behind the surface disarray in polycentric Internet governance?If so, what kind of ordering dynamics operate in this context?The following discussion distinguishes three different kinds, or layers, of structure: namely, norms, practices, and underlying orders.In part, then, this analysis builds on previous contributions to this forum: by Deloffre and Steffek on norms, by Gadinger and Mencutek on practices, and by Koinova on informality.However, I also add a third dimension of structure with the notion of underlying orders: that is, deeper structures such as a hegemonic state, capitalism, and embedded social hierarchies.My argument is that polycentric governance encompasses all three layers of structure, as well as interconnections between them.Thus, whereas we might have expected that attention to structure would simplify our analysis, the multiplicity of norms, practices, and underlying orders actually makes polycentrism still more complex.

Norms
Norms are general articulated principles that inform the process of governing.This type of structure figures elsewhere in the present forum when Steffek speaks of "norms" and Deloffre speaks of "metagovernance."The notion is that certain guiding ideas of the good and the correct become embedded in the conduct of world politics, such that they acquire a force of their own, separate from the actors who enact them.This insight, previously developed in liberal regime theory and constructivist IR (Rittberg 1993;Bernstein 2001), can readily transfer to investigations of polycentric governing, although institutionalist research to date has not systematically done so.
Prominent examples of embedded norms in contemporary governance include democracy, economic growth, gender equality, human rights, peace, rule of law, sovereignty, and sustainable development.Governance institutions generally align their regulatory measures with such reigning norms as a way to attract legitimacy Downloaded from https://academic.oup.com/isr/advance-article/doi/10.1093/isr/viab030/6403480 by Universiteit Leiden / LUMC user on 04 November 2021 and compliance.When the various institutions in a polycentric complex draw upon the same norms, the organizationally decentered apparatus structurally converges on common principles.In this way, norms provide an interlinking thread for the whole: a "glue" that holds together the formally dispersed institutional parts.
Internet governance well illustrates this dynamic.For example, pretty well all of the manifold institutions involved prominently invoke norms of transparency and accountability to describe and justify their operations.In addition, the various private institutions in Internet governance regularly summon a shared norm of socalled corporate social responsibility.Likewise, players across the many quarters of polycentric Internet governance have initiated measures in the name of a gender equality norm.Even "technical" institutions responsible for global Internet infrastructure have (rather grudgingly) adopted the norm of human rights, implicitly recognizing that their previous denials of concept's relevance weakened their legitimacy (Ten Oever 2018; Cath 2021).
As these examples show, established norms exert structural power in polycentric governing.The various institutions are substantially constrained to adopt common principles such as transparency, gender equality, and human rights.The organizations do not embrace the norms wholly of their own accord, but substantially because they (implicitly) recognize that their legitimacy depends on conforming to generally accepted criteria of good governance.

Practices
A second type of structure in polycentric governing falls under the rubric of "practice."Whereas norms refer to what people believe in, practices relate to what people do.While norms are explicitly articulated, practices are often tacit and even unconscious.With the so-called practice turn of the past decade, IR theory has drawn inspiration from sociological ideas of "actor-network," "assemblage," and "field" (Adler and Pouliot 2011;Bueger and Gadinger 2018).These innovations, also reflected in contributions to this forum from Gadinger and Mencutek, highlight the regulating effects of everyday routines.Hence "the way things are done" lends an important, if generally subtle, degree of unifying organization to polycentric governing processes, also of the Internet (Galloway 2004).
One might distinguish four aspects of practices.First, discursive dimensions of practices take verbal form, such as routine words, phrases, and narratives.Across polycentric governance of the Internet, for example, one hears continual refrains of a technical discourse around "security, stability, and resiliency" and a democratic discourse around "bottom-up multistakeholder participation."Also regularly rehearsed across the many Internet governance venues are stories about the founding pioneers of cyberspace, lending a sense of common history, or bonding visions of an "open interoperable Internet."Other everyday verbal glue for the polycentric governance network comes from shared insider jokes and allusions, as well as a plethora of acronyms that unite those "in the know" (e.g., DNS, IPv6, IXP, RFC, etc.).
Second, behavioral dimensions of practices relate to routine ways of bodily interaction.For example, polycentric governance of the Internet has its particular dress codes (i.e., casual) and other bodily self-presentations (e.g., the hallmark male techie ponytail) that subtly shape the tone and direction of discussion.In addition, Internet governance arenas have customary ways of deliberating (e.g., how to perform at the microphone, around the committee table, in online conference calls) as well as socializing (e.g., corridor conversations, gala dinners).Such behaviors are transferable across the different nodes of the polycentric Internet governance network.Common behavioral routines thereby forge informal unity across formally scattered governance institutions.Third, material dimensions of practices involve objects as common reference points for a polycentric governing complex.For instance, the various institutions in Internet governance tend to use common technologies such as WhatsApp telecommunications and the Adobe Connect online meeting platform.In addition, Internet regulatory bodies brand themselves similarly with t-shirts, tote bags, stickers, pins, and other freebies distributed at their meetings.The Internet sector also has its archetypal building designs, office plans, and conference layouts that transfer across the regulatory institutions.These material objects, too, lend subtle order to institutional decenteredness, making it relatively easy for a participant to move between one governance venue and the next.
Finally, institutional dimensions of practice encompass ways that organizations construct and run their policy processes.For example, the multiple nodes of Internet governance all tend to have executive boards, secretariats (with similar departmental divisions), and constituency groups (academic, business, civil society, government, and technical).Common bureaucratic layouts and similar "multistakeholder" policymaking processes facilitate cross-institutional communication and coordination.Indeed, board members, senior staff, and constituency leaders of one Internet governance institution regularly attend meetings of other nodes in the polycentric regime, exchanging experiences and reinforcing their similar institutional modus operandi.
Like norms, practices (with these four qualities) exert structural power on a polycentric governance network.In other words, participants in the regime are considerably compelled to speak, behave, materialize, and institutionalize in certain ways in order to take part in the game.Marginalization awaits those who do not understand the lingo, who attend in unconventional dress, who cannot operate the technology, or who do not slot into standard constituency categories.Thus, practices have ordering effects, providing further substance for the lines in the pictogram of a polycentric network.

Underlying Orders
While norms are explicit and directly accessible, and while practices are largely implicit but still directly observable, a third layer of structure in polycentric governance lies still deeper below surface appearances, being mostly unspoken and only indirectly visible.The phrase "underlying orders" refers here to macro-structures that underpin-and manifest themselves through-norms and practices, as well as actor motivations and decisions.Underlying orders are systemic: they permeateand integrate-all locations and connections in a polycentric regime.
Different macro-theories of IR advance different conceptions of underlying world order.For example, neorealist theory depicts the primary structure in terms of the systemic distribution of power among states (e.g., whether it is unipolar, bipolar, or multipolar).For Marxists, the underlying order lies in the prevailing mode of production, currently capitalism.Weberian sociology identifies a basic structure of modernity.Feminists with a structuralist bent posit an underlying order of patriarchy, often intersectionally combined with racism, heterosexism, and other social hierarchies.Postcolonial theory identifies a deeper structure of Western imperialism; poststructuralists hone in on neoliberal governmentality; and posthumanist political ecologists highlight a systemic pattern of anthropocentrism and extractivism.Thus, while macro-theories agree that structure takes shape as an underlying order, they diverge in their accounts of the character of that deeper architecture.
Yet it is perhaps not necessary to affirm that one or the other underlying order is the "ultimate" structure in polycentric governance.Indeed, several deeper patterns would seem to figure in respect of the Internet.For example, unipolarity has shaped this policy field inasmuch as the US Government has exerted hegemonic leadership, particularly during the early years of global Internet governance.In addition, and concurrently, the dispersed rules and regulatory institutions for the Internet have systemically conformed to deeper ordering by capitalism.After all, most parts of the polycentric governance apparatus consistently enable commodification and surplus accumulation around digital communications.A deeper structure of modernity is apparent with the pervasive power across polycentric governance of the Internet of a techno-rationalism that prioritizes instrumentalist problem-solving through engineering fixes.Moreover, structurally embedded hierarchies of influence pervade the many institutions of polycentric Internet governance: inter alia in respect of age, ethnicity/race, gender, North-South geopolitics, and English language (Jongen and Scholte 2021).These various underlying ordering principles can operate concurrently on the Internet, sometimes in relations of mutual reinforcement and sometimes in contradiction (Scholte 2020).
Like norms and practices, underlying orders have structural power in polycentric arrangements.The various institutions in Internet governance do not fall in line with a hegemonic state, capitalism, modernity, or social stratifications of their own accord or coincidentally.Underlying orders are social forces in their own right that to some degree impose a deeper organization on the surface disarray of polycentric governing.

Conclusion: Complex Structure
The foregoing discussion has suggested that structure in polycentric governance involves three types of deeper layers: norms, practices, and underlying orders.Taking the example of Internet governance, we have seen that what appears chaotic in terms of tangible institutions is ordered in terms of less palpable organization.Indeed, in my experience as an advisor in the IANA transition (mentioned at the start of this commentary) I witnessed that all of the norms, practices, and underlying orders discussed here guided the rocky process to its smooth conclusion.Only when institutional analysis is supplemented with a threefold structural analysis do we obtain a fuller understanding of polycentrism.
The word "threefold" wants underlining here.IR theorists who address questions of structure tend to examine either norms or practices or macro-orders: scrutinizing one aspect, but neglecting the other two.Structuring polycentrism offers a chance to combine these three dimensions of social order in a single unified analysis.One can then examine how norms, practices, and underlying patterns interrelate.Often these three aspects of structure co-constitute and reinforce each other; however, in other combinations, they can contradict and undermine each other.In Internet governance, for instance, norms of liberal democracy, practices of multistakeholder policymaking, and underlying orders of US hegemony and capitalism can have largely co-creating logics.From another angle, however, norms of gender equality can be in tension with prevailing masculinist practices of deliberation and an underlying order of patriarchy.So, when transcending institutionalism with a structural analysis of polycentrism, the challenge is not only to identify which norms, practices, and underlying orders are relevant, but also to establish their multidirectional interconnections.
Hence, while this forum may have embarked on structural analysis of polycentrism in hopes of simplifying our understanding of institutional messiness, we end up with theory that is still more complex.Polycentric governance becomes a matter not only of multiple institutions in complicated networked interactions, but also of multiple structures in intricate mutual entanglements.Plus there are manifold interrelations between those intersecting structures and the interacting institutions.So, yes, there is order in the polycentric chaos, but that order is not simple.