Rebels, vigilantes and mavericks: heterodox actors in global health governance

COVID-19 has exposed profound governance challenges that demand more diverse and creative approaches to global health governance moving forward. This article works towards such a pluralization of the field by foregrounding the vital role played by heterodox actors during the pandemic. Heterodox global health actors are backgrounded actors who improve health in different parts of the world, but who remain politically marginalized – and epistemically invisibilized – because they depart in crucial respects from the liberal orthodoxy pervading the field of global health governance. The article analytically foregrounds those heterodox actors through an architectural inversion – a relational approach to the study of global health governance that builds upon recent methodological insights from postcolonial studies, infrastructure studies, and science and technology studies. The article then harnesses that methodological approach to empirically investigate the COVID-19 activities of three different heterodox actors: rebel groups providing public health in the borderlands of Myanmar, a women’s vigilante movement stitching face masks in the Czech Republic, and a maverick scientific platform for the international sharing of viral sequence data. Performing that architectural inversion begins to loosen the dominance of the liberal episteme within the practice and study of global health governance. It further visibilizes how that field is continually co-produced by the background activities of many such heterodox actors. It also lays conceptual foundations for a more heterodox future research agenda on global health governance – and arguably global governance more broadly – in response to the numerous unresolved challenges revealed by COVID-19.


Introduction 1
COVID-19 has been widely regarded as a protracted global health challenge (Agostinis et al., 2021;Davies and Wenham, 2020;Jones and Hameiri, 2022;Roberts, 2020), and as a key test for the institutional architecture of global health governance (Gostin et al., 2020;Phelan et al., 2020;Roemer-Mahler, 2021).The pandemic is also a salient reminder of the critical role that many non-state actors play within that rapidly evolving governance architecture.Amid the overwhelming pressures of the pandemic, a wide array of non-governmental organizations and charities sought to compensate for the overstretched capacities of national governments.A battery of pharmaceutical companies quickly entered the international race to develop life-saving diagnostics, vaccines and medicines for SARS-CoV-2.Concurrently, a plethora of philanthropic organizations funded wideranging pandemic responses in many countries throughout the world.None of this surprised scholars of global health governance, who have long considered non-state actors as central to the field.Most of those non-state actors -like international institutions, multilateral initiatives, non-governmental organizations, philanthropic organizations and public-private partnerships -are extensively studied in the literature; and their vital role in an increasingly stabilized and well-characterized institutional architecture of global health governance is widely recognized by now (Hanrieder, 2015;Kay and Williams, 2009;Lee, 2008;Patterson, 2018;Rushton and Williams, 2011).
One of the unanticipated legacies of COVID-19, however, has been to also reveal a plethora of more unusual and unconventional non-state actors contributing to the response.In various urban neighbourhoods and villages around the world, for example, self-help groups spontaneously formed in response to the overwhelming threat of the new coronavirus.In the corporate sphere, a swarm of companies not traditionally engaged in global health also started exploring innovative commercial and technological solutions to the unexpected crisis.All the while criminal gangs assumed responsibility for enforcing quarantine measures in crime-ridden cities, while insurgent groups undertook the distribution of personal protective equipment to vulnerable populations living in war-torn states.Echoing earlier experiences with HIV/AIDS, pandemic flu and Ebola, the international response to COVID-19 abounded with many unconventional health actors operating at the margins of the established institutional architecture of global health governance.How, then, can the study of global health governance be enriched and diversified to better account for the ways in which those unorthodox, but usually backgrounded, actors also contribute to the co-production of health across different parts of the world?
This article works towards broadening and pluralizing the study of global health governance by analytically foregrounding the vital role played by those heterodox actors.Heterodox global health actors are background actors who make significant contributions to the improvement of health in many parts of the world, but who usually remain politically marginalized -and epistemically invisibilized -because they depart in vital respects from the dominant liberal orthodoxy characterizing the field of global health governance.The article analytically foregrounds those heterodox actors by developing a relational approach to the study of global health governance -an architectural inversion -that builds upon recent methodological insights from postcolonial studies, infrastructure studies and science and technology studies (STS).The article subsequently harnesses that new methodological approach to empirically investigate the COVID-19 activities of three such heterodox actors: rebel groups providing public health in the borderlands of Myanmar, a women's vigilante movement stitching face masks in the Czech Republic, and a maverick scientific platform for the international sharing of viral sequence data.Performing that architectural inversion begins to loosen the dominance of the liberal episteme in global health governance.It further visibilizes how that field is continually co-produced by the background activities of many heterodox actors.It also, finally, lays down some conceptual foundations for a more heterodox research agenda on global health governance -and arguably global governance more broadly -in response to the numerous unresolved governance challenges exposed by COVID-19.

An architectural inversion of global health governance
The vital role of heterodox actors can be foregrounded through a novel methodological approach to the study of global health governance: an architectural inversion.The first part of this approach -the architectural -is invoked much like the existing literature on global health governance already mobilizes it, that is, to refer to the field's complex network of actors and institutions (see, for example, Fidler, 2007;Gostin et al., 2010;Hoffman et al., 2015;Lisk and Šehović, 2020;Roemer-Mahler and Elbe, 2016).Several foundational studies have mapped this architecture and point towards a fairly settled list of core actors (Clinton and Sridhar, 2017;Cooper, 2016;Harman, 2012;Kickbusch et al., 2013;McInnes et al., 2014;Youde, 2012Youde, , 2018)).Usually, those actors include international organizations like the United Nations (UN), the World Health Organization (WHO), the World Bank, the International Monetary Fund (IMF), the World Trade Organization (WTO) and the Joint United Nations Programme on HIV/AIDS (UNAIDS).Those studies further recognize several influential multilateral associations (like the G7/ G8 or the G20), as well as public-private partnerships (like the international vaccine alliance GAVI, or the Global Fund to Fight AIDS, Tuberculosis and Malaria).They finally also point to various non-governmental organizations, civil society organizations and philanthropic organizations active in global health (like the Gates Foundation).For scholars of global health governance the 'architecture' metaphor has long been attractive to capture this 'environment characterized . . .by unstructured plurality because the image of architecture communicates rational design, ordered stability, and functional vision' (Fidler, 2007: 14).The concept of the global health governance architecture thus signifies this assemblage of key institutions and actors participating in the quest to improve health globally, albeit in an international context that remains only loosely ordered and lacks a centralized political authority.
The second dimension of this new methodological approach -the inversion -builds upon a notion from infrastructure studies.Social scientists working in that field are already accustomed to performing 'infrastructural inversions' to counter the scholarly oversight of infrastructure.Infrastructures often remain invisible and overlooked because they operate in the background of society.An infrastructural inversion counters this problem by proactively foregrounding the role of infrastructural systems when investigating a social field's constitution, practices and global dissemination (Bowker, 1994;Harvey et al., 2017).By way of extension, an architectural inversion encourages a similarly fundamental reversal of perspective; it too brings to the analytical fore constitutive elements that usually remain backgrounded and invisibilized.Instead of focusing on infrastructures, however, an architectural inversion directly explores the realm of global health governance -to deliberately foreground the vital role played by non-state actors who usually remain politically and epistemically backgrounded because they operate at the margins of (and potentially even in tension with) the established institutional architecture of global health governance: what we call heterodox global health actors.Thus, an architectural inversion shares broad similarities with some existing International Relations (IR) studies -particularly feminist global health scholarship highlighting the agency of those who are mostly invisibilized (Anderson, 2015;Davies, 2010;Harman, 2016Harman, , 2021;;Wenham, 2021), as well as scholars critiquing the widespread influence of neoliberalism on global health policies (Rushton and Williams, 2012;Schrecker, 2016;Sparke and Williams, 2022).Yet an architectural inversion ultimately also goes beyond any singular category of class, gender or race to expand this analytical sensitivity towards a much more diverse array of heterodox actors operating at the margins of the established global health architecture.
An architectural inversion is, then, a methodological approach principally designed to explore global health governance at the boundaries of the liberal episteme running through much of that governance architecture (Deudney and Ikenberry, 1999;Doyle, 1986;Keohane, 1998;Park, 2018).Most of the institutions foregrounded in the study of global health governance embody liberalism's belief in the value of international organizations -as platforms for addressing global (health) challenges in an increasingly interdependent world (Jahn, 2018: 48; see also Ikenberry, 2018).Much of that established global health architecture also operates within the broad economic parameters of liberalism -particularly in its reliance on free market and capitalist principles to procure global health goods, as well as the extensive financial support it receives from international economic institutions like the IMF, World Bank and WTO (Clinton and Sridhar, 2017;see Kay and Williams, 2009).Finally, many of those core global health actors also share a broad commitment to the international spread of liberal norms around individual freedoms, democracy and human rights -especially the right to health (Meier and Gostin, 2018;Van de Pas et al., 2017).Yet this pervasive liberalism raises a key question: what does global health governance also look like at the margins of -or even beyond -this liberal episteme?An architectural inversion is a methodological approach designed to explore this very question.It aims to capture, and bring to the analytical fore, the diverse substrate of more heterodox actors continually contributing to the production of health in many different parts of the world, but who usually remain backgrounded because they operate at the margins of that liberal episteme.Overall, such an architectural inversion represents a more relational approach to the study of global health governance that concurrently builds upon multiple interdisciplinary insights from the fields of postcolonial studies, infrastructure studies and STS.

Postcolonial studies and the liberal episteme of global health governance
Postcolonial critiques are becoming more prevalent across the fields of IR and Global Health (Abímbọ ´lá and Pai, 2020;Anderson, 2006;Hilberg, 2022;King, 2002;Richardson, 2020Richardson, , 2020;;Sekalala, 2017).Those critiques have significant ramifications for the study of non-state actors in world politics, as postcolonial scholars have already shown in relation to security studies.They illustrate how mainstream scholarship frequently misunderstands the role of non-state actors and governance arrangements because the field conceives them through an inherently liberal prism (Barkawi and Laffey, 2006;Sabaratnam, 2013).Even ostensibly more emancipatory security approaches -like human security -struggle to render intelligible the role of non-state actors that do not easily fit within this Western framework of global civil society, because human security remains fundamentally wedded to a Eurocentric idea of security as emancipation (Barkawi and Laffey, 2006).Thus, much of the existing scholarship foregrounds those types of actors who are more readily compatible with liberal assumptions around global civil society, while backgrounding many other types of actors.This leads to a second problem highlighted by postcolonial scholars.
Postcolonial critiques argue that mainstream approaches to security studies, and even some critical ones, also underestimate the extent to which those neglected local actors are themselves co-constituted by the more readily legible global governance structures -and vice versa.Postcolonial scholars have therefore further critiqued scholarship on hybrid, everyday and local peacebuilding interventions for privileging civil society-style non-state actors over other kinds of formations (Nadarajah and Rampton, 2015;Sabaratnam, 2013).Those common perspectives, postcolonial scholars argue, often depoliticize and romanticize non-state actors, essentializing them as 'local' actors who are juxtaposed as a force of emancipation vis-a-vis the state and the international -albeit without appreciating how those actors have themselves become co-constituted through formative histories of global imperialism and postcolonial state-formation (Nadarajah and Rampton, 2015;Sabaratnam, 2013).This leads scholars of hybrid peacebuilding to prioritize non-state actors that fit (or are fitted) within a liberal frame of civil society over others.In that process, identity-based projects, armed resistance to the international state system and generally ideologies that do not sit well with a liberal prescription fall off the analytical radar (Nadarajah and Rampton, 2015, p. 64).Postcolonial critiques have thus exposed a fundamental problem surrounding the constrained manner in which non-state actors are frequently studied in security studies, and how this field has overlooked the ways in which notions of the local and the global are historically co-constituted.This postcolonial critique is readily extendable to the field of global health governance.That field too is pervaded by a liberal episteme constraining which non-state actors are studied and how they are studied -with the established global health governance architecture commonly revolving around a comparatively limited set of actors and institutions identified above.In the first instance, an architectural inversion of global health governance therefore builds upon postcolonial critiques to open up a new analytical space for exploring the international health activities of more heterodox global health actors, that is, unconventional non-state actors who depart in key respects from this liberal episteme and who tend to remain backgrounded in the study of global health governance.Although many of these heterodox actors operate in the Global South, geographic considerations are not their primarily determining characteristic.These health actors are heterodox not simply because they are more indigenous actors operating in the Global South.Nor are they heterodox just because they represent more unofficial actors -like the many informal healthcare providers operating throughout the world.It is not even that these heterodox actors are necessarily 'newcomers' to global health governance.Although heterodox actors can encompass any (or even all) of these characteristics, their heterodoxy is principally a function of their epistemic positionality; and heterodox actors can therefore be found operating in most regions of the world.Heterodox actors are essentially non-state actors who operate at the margins of (and potentially in tension with) key precepts of this liberal episteme of global health governance.

Infrastructure studies and the heterodox substrate of global health governance
Foregrounding those heterodox actors is not merely a matter of filling a gap in the expanding literature on global health governance.An architectural inversion additionally builds upon methodological insights from infrastructure studies to further explore how those heterodox actors also form a co-constitutive substrate within the wider architecture of global health governance.Infrastructure studies broadly conceptualizes infrastructures as socio-material arrangements that bring together people, things, and ideas in new ways and that stabilize new meanings.Infrastructures are vehicles of mobility and an 'architecture of circulation' residing at the heart of modernity, since they not only facilitate -but also organize -the flow of people, goods and capital, and thus create the environment of our everyday lives (Larkin, 2013: 328).More recently, Anna Leander has pointed to two further concepts for operationalising the study of infrastructure in International Relations.Infrastructuring involves enacting politics, reproducing infrastructures, but also the redesigning of infrastructures and thus also power relations; whereas infrapolitics captures 'the politics of intervening with infrastructures and infrastructuring processes.Infrapolitics takes place outside and away from the formal political sphere' (Leander, 2021: 209).Thus, the burgeoning study of infrastructure suggests that infrastructures are not just tools of governance; they co-constitute the political regimes through which society is governed.The materiality of infrastructure can itself engender, constrain, and contour complex new formations and practices -intimating that the 'social' is not produced by human practices alone.
An architectural inversion of global health governance builds upon this insight about the socially constitutive role of infrastructure -by teasing out how the background health infrastructures of heterodox actors similarly play a materially co-constitutive role within the global health architecture.Heterodox actors are therefore not just backgrounded actors within the liberal episteme of global health governance; they are also actors whose infrastructures productively contribute to the co-constitution of health globally.Like an infrastructural inversion, the architectural inversion of global health governance thus seeks to analytically foreground how a particular field of social practice becomes coconstituted by crucial elements that usually remain backgrounded, at the margins, or effaced in the existing scholarship.Like an infrastructural inversion, moreover, this architectural inversion also broadly conceptualizes those heterodox actors as a thick coconstitutive infrastructural substrate operating in the background of the contemporary architecture of global health governance.
As we will see below, the material infrastructures maintained by those heterodox actors may evince substantial empirical variability.Some heterodox infrastructures might, for instance, take a more institutionalized form, while others are far less fixed; and the degree of their heterodoxy can also vary considerably.Precisely because their heterogeneity defies the homogenizing tenets of liberal institutionalism, heterodox actors do not necessarily emerge as institutionalized organizations from within the established architecture of global health, and their heterodox infrastructures can therefore operate according to a much more diverse set of logics than liberal governance norms alone.Our perspective even allows for the possibility that their attributes can change over time.Indeed, we expect that this will frequently be the case as the uneasy relationship between heterodox actors and the dominant institutions of liberal global health evolves.Thus, some heterodox global health actors might only exist temporarily, while others might provide health mostly as a by-product of a completely different agenda.Others still might eventually become part of the institutional mainstream of global health as part of continuous transformations.Importantly, however, an architectural inversion points to the impact that these tensions have in shaping the architecture of global health beyond the liberal imaginary.

STS and the turn towards heterodox global health
An architectural inversion finally also incorporates insights pioneered in STS.One of the most prominent idioms within that field -the notion of co-production -contends that 'the ways in which we know and represent the world (both nature and society) are inseparable from the ways in which we choose to live in it' (Jasanoff, 2004: 2).Those scholars reject a view of science as being divorced from social context, just as they reject the opposite view of science being driven exclusively by social factors.Rather, science and society are co-produced; advances in science and technology both embed -and are embedded in -evolving societal practices, identities, norms, conventions, discourses, instruments and institutions.This deeper insight into the inseparability of the epistemological and the social has further ramifications still for the study of global health governance (Elbe and Buckland-Merrett, 2019;Lakoff, 2020).It suggests that our social scientific understanding of global health governance is itself bounded by the ways in which this field of practice has evolved around that liberal orthodoxy.The liberal episteme running through the institutional architecture of global health governance also epistemologically conditions what scholars tend to recognize as constituting a core 'actor' in that field.Consequently, existing scholarship on global health governance ends up implicitly foregrounding those types of actors who are most closely aligned with the field's embedded liberalism.
In the third instance, then, an architectural inversion of global health governance also builds upon STS insights to loosen those dominant epistemic frames of scholarship and to proactively pluralize the study of global health governance.Ultimately, an architectural inversion of global health governance is a methodological vehicle for encouraging a broader epistemic transformation of this entire scholarly field -not dissimilar to the rise of heterodox thought in the discipline of Economics.That discipline has recently witnessed concerted attempts to expand the field's epistemic confines beyond the dominance of neoclassical economics.Already those efforts have spurned a diverse subfield known as heterodox economics encompassing many diverse approaches -like post-Keynesianism, (old) institutionalism, feminist, social, Marxian, Austrian and social economics, as well as others (see also Kvangraven and Alves, 2019;Lawson, 2006: 484).That transformation was partly triggered by the discipline's epistemic narrowness, but also by a perceived disconnect between the academic study of Economics and the onset of the major economic crises like the financial crisis of 2008 (Colander et al., 2004;Dequech, 2007;Dow, 2011;Garnett, 2006;Hodgson, 2019;Lawson, 2006;Lee, 2012;Lee and Cronin, 2016).Similarly, an architectural inversion seeks to mobilize the study of heterodox actors as a way of pluralizing and diversifying the study of global health governance in the wake of the persistent governance shortcomings exposed by COVID-19.How, in other words, can the study of heterodox actors open up different ways of doing/thinking global health in the 21st century?
Overall, then, an architectural inversion forms quite a different methodological approach for studying global health governance that builds upon the more relational methods pioneered within postcolonial studies, infrastructure studies and STS.Below such an architectural inversion of global health governance is performed by empirically investigating the COVID-19 activities of three heterodox actors.Those three actors are all quite different, but each have all been selected because they decisively depart from one of the three key precepts of global health's liberal episteme.We start by looking to an actor who deviates from liberal assumptions about intergovernmentalism and who is located in southeast Asia: rebel groups providing public health in the borderlands of Myanmar.We then turn towards an actor whose efforts digress from liberal assumption about the value of free market economics and who emerged in central Europe: a women's vigilante movement for stitching face masks in the Czech Republic.We finally focus our attention on an actor who challenges liberal norms about unrestricted data sharing in the international scientific field: a maverick platform for the global exchange of viral genetic sequence data.Guided by our relational approach, we empirically explore these heterodox actors through the three analytical dimensions of an architectural inversion: First, how exactly do these actors stand -at least partially -outside the liberal episteme of global health governance?Second, how do the infrastructures of those actors materially interact with the more established institutional architecture of global health governance and contribute to the co-constitution of health around the world?Finally, what resistances and tensions are engendered by the workings of these heterodox actors -pointing towards alternative ways of contributing towards global health governance?

Rebels: public health in the borderlands of Myanmar
One of the cornerstones of the liberal episteme in global health governance is intergovernmentalism -the assumption about the political primacy of nation states and the concomitant conviction that international organizations can foster meaningful health cooperation between them (Ng and Ruger, 2011;Ruckert et al., 2016).An extensive literature thus highlights the influential role played by nation states in global health governance, describes how the global health agenda is funded by many liberal-democratic states, and explores their diplomatic activities at various international organizations (Cooper, 2016;Lee, 2008).When, moreover, this literature also acknowledges the contributions of non-state actors, it focuses predominantly on philanthropical organizations that fit the model of global civil society -like the Gates Foundation, Médecins Sans Frontières or Oxfam.During COVID-19, however, there were also far more heterodox non-state actors operating in the background of that established global health governance architecture.
Take Panghsang for instance.Panghsang is a prosperous town at Myanmar's border with China.It draws Chinese customers to its thriving nightlife scene, including casinos, strip clubs and karaoke bars.Yet on 26 January 2020, local authorities unexpectedly announced the closure of these lucrative industries, banned public gatherings and imposed strict immigration controls at the nearby border crossings with China (Sit Htet Aung, 2020).They did so in response to the unexpected outbreak of COVID-19.The lethal coronavirus, which was first confirmed in central China's city of Wuhan the month before, had reached China's southwestern border with Myanmar by late January.Myanmar's official government was still dragging its feet over the introduction of public health interventions; and the state would not implement formal measures to stop the spread of COVID-19 for another 2 months (Holpuch, 2020).Yet Panghsang's local authorities were not acting on behalf of Myanmar's state, because Panghsang is governed by the United Wa State Army (UWSA) -one of the many non-state armed groups that control large pockets of territory in Myanmar's restive borderlands.
As in other parts of the Global South, postcolonial state formation in Myanmar has not led to the consolidation of state authority over territory and population.In fact, the Myanmar state has never controlled most parts of the country's borderlands.Instead, these regions are governed by ethnonational rebel armies, who have struggled for more autonomy or outright secession from an ethnocratic central state for many decades (Brenner, 2019: 29-46).Throughout this protracted conflict, Myanmar's rebels have not only challenged the state on the battlefield; they also established quasi-states within the state -including elaborate governance infrastructures such as public health systems.Arguably, these rebel health systems are not just aimed at providing public health for populations under rebel control; they are also crucial for building non-state nations in competition with the existing nation-state (Brenner and Tazzioli, 2022), which is why reproductive health inhabits a particularly important role in these governance arrangements (Hedström, 2022).Myanmar's rebels thus train teachers and nurses, who in turn staff their schools and medical facilities.Their rebel health systems mostly provide basic healthcare, including maternal and child healthcare and vaccination programmes, as well as humanitarian aid to displaced people, including physical rehabilitation facilities for landmine victims.But they also engage in disease control and prevention measures, such as anti-malaria programmes and COVID-19 interventions (Davis and Jolliffe, 2016;Karen Peace Support Network, 2020).While some scholarship in conflict and security studies has recently discussed how rebels govern populations, including in the context of COVID-19 (Arjona et al., 2015;Furlan, 2020), this literature focuses mostly on the ways in which rebel groups generate public goods to legitimize themselves or control local populations -leaving little understanding as to how such rebel groups interact with, and ultimately co-produce, health infrastructures.
A closer look at Myanmar's rebel-controlled border areas suggest that they do so in tandem with state and non-state actors.Along Myanmar's border with China, for instance, rebel governance infrastructures are intertwined with the infrastructure of the Chinese state.The Kachin Independence Organization (KIO), one of the strongest rebel movements in Myanmar, trains its own nurses in a college located in its liberated capital of Laiza, a small town clung to the Chinese border.But most of the movement's other medical staff, including doctors, lab technicians and radiologists, are trained in Chinese universities.To prepare students to study in China, among other purposes, the KIO established a Chinese language school.Trained KIO health staff work in hospitals in Laiza and Maijayang -the second biggest town under KIO control -where they treat KIO personnel and civilians alike.Moreover, the KIO's health staff operate field clinics in the many camps for the tens of thousands of internally displaced people (IDPs) who have fled from counterinsurgency operations.Myanmar's state has tightly restricted international relief aid to KIO-controlled territories.Local brokers such as the Kachin churches managed to channel some medical material from international humanitarian agencies into KIO areas.Yet, most public goods, including health, are financed by KIO revenues, generated by taxing local business and border trade as well as natural resource extraction (Brenner, 2019: 77-90).
In contrast to the limited support from the kinds of international humanitarian agencies that are already well known to scholars of global health governance, the KIO coordinates its health interventions with the regional government in China's neighbouring Yunnan province.Chinese authorities, in fact, have a pragmatic working relationship with several rebel groups on their border with Myanmar.They seek coordination of a wide set of non-traditional security concerns from human trafficking and narcotics smuggling.Security concerns also include health issues surrounding the spread of disease, including one of the world's most dramatic HIV/AIDS crises, seasonal outbreaks of dengue fever and malaria, as well as COVID-19.The China Center for Disease Control and Prevention (China CDC) collaborates with the public health authorities of several ethnic rebel groups at the Myanmar-Chinese border in controlling the spread of diseases.The KIO Health Department has, for instance, met with China CDC on a biannual basis since the early 2000s.China CDC also arranged for KIO hospitals in Laiza and Maijayang to send polymerase chain reaction (PCR) tests to laboratories in Yunnan in the initial months of the COVID-19 outbreak in 2020. 2  Shifting our attention to the situation in Myanmar's eastern borderlands to Thailand illustrates how rebel health governance can co-produce public health as part of a wider global health assemblage surrounding international humanitarian agencies and development donors.Compared with the tightly restricted Chinese borderland with Myanmar, the Thai border with Myanmar has long hosted a donor-funded humanitarian complex of international organizations and international non-governmental organization (INGOs).This complex emerged after violent military crackdowns on civilian protesters and counterinsurgency campaigns in rebel-controlled areas sent large-scale refugee waves to Thailand since in the late 1980s and throughout the 1990s and 2000s.International agencies cooperated with ethnic rebel movements and affiliated community-based organizations in providing relief to both, the refugee populations in the sprawling camps along the Thai border as well as the many, often displaced, local communities that remained in inaccessible conflict zones (Brenner and Schulman, 2019).Contrary to the oft-heard 'refugee warrior' narrative -according to which rebels simply expropriate humanitarian relief for the perpetuation of war -many rebel rulers in Myanmar emerged as partners for providing health and welfare to vulnerable communities, who often viewed rebel movements like the Karen National Union (KNU) as legitimate local authorities (McConnachie, 2012).
To reach the most vulnerable communities in inaccessible conflict-zones, international relief efforts needed to cooperate with the health departments of ethnic Karen, Mon, Karenni and Shan rebel groups, as well as a range of community-based organizations.Today about 600,000 people in south-eastern Myanmar are dependent on this infrastructure employing around 3,000 healthcare workers in 139 clinics and 93 mobile medic teams (Davis and Jolliffe, 2016: 10).Since Western donors abandoned many border-based initiatives after Myanmar's generals initiated reforms in 2011, this infrastructure has, however, suffered significant setbacks.As Myanmar rapidly rehabilitated itself in the eyes of the international community, Western donors started to work with state channels in Myanmar instead.While donor support had strengthened rebel health infrastructure in the 1990s and 2000s, it now bolstered militarized state territorialization (Brenner and Schulman, 2019: 28-30).After the military coup in 2021, it comes to be seen how the assemblage of rebel and global health infrastructures realigns in southeastern Myanmar.In a meeting with British parliamentarians, the health minister of Myanmar's newly formed exile government, Zaw Wai Soe, asked for the resumption of international support to Myanmar's rebel health providers in what he described as the country's 'liberated areas '. 3  This vital provision of public health by armed rebel groups in Myanmar's borderlands marks a salient entry point for capturing the importance of heterodox actors in global health and shows how they stand in productive tension with the existing global health architecture.This rebel governance of health stands in direct competition with the existing nation-state -fundamentally challenging liberal assumptions around the primacy of nation states and international organization in global health governance.Furthermore, it illustrates the critical role that non-liberal political ideologies, like ethnic nationalism and separatism, can play in the provision of health across different parts of the world.At the same time, tracing the infrastructures of rebel health governance in Myanmar's borderlands also reveals how rebel groups end up co-producing global public health together with other actors -like foreign government institutions, international donors, and humanitarian agencies, who remain much more readily legible from a global governance perspective.Their tension with the ethos of liberal internationalism notwithstanding, Myanmar's armed resistance movements help to co-constitute the governance architecture of global health.

Vigilantes: the great stitch-up in the Czech Republic
The liberal episteme of global health governance also foregrounds many commercial actors -particularly pharmaceutical companies producing vital medical products (Elbe, 2014;Lakoff, 2010).Much of the world's pharmaceutical production is carried out by private companies operating according to free-market principles within liberal and capitalist economies.How to distribute medicines and vaccines more equitably around the world has therefore formed a long-standing challenge for global health governancefrom anti-retroviral therapies (ARVs) during the AIDS pandemic, and antivirals like Tamiflu against pandemic influenza, through to commercially produced COVID-19 vaccines (Elbe, 2018;Merson and Inrig, 2017).Yet an equally important, if more backgrounded, dimension of any public health system is access to medical devices.At the beginning of the COVID-19 pandemic, one such critical medical device quickly became unavailable in many countries around the world: face masks.Governments relying upon the commercial market provision of medical devices suddenly found themselves unable to procure sufficient face masks to protect their citizens.Prompted by this 'market failure', a more heterodox type of actor began operating at the margins of the conventional architecture of global health governance -in the Czech Republic.
There, an influential entrepreneur mobilized a new bottom-up movement that rapidly developed an alternative epistemic and material infrastructure for promoting the sewing of face masks by citizens themselves.This new movement prompted a wave of mass mask sewing (and donating) in local communities, mostly carried out by women (Gorčíková et al., 2020), and during which solidarity and volunteer labour came to temporarily replace the liberal market economy and wage labour as key organizing principles for producing this vital medical equipment.The obligation to wear face masks in populated spaces has since become widely accepted as a key element in the fight against the coronavirus.At the beginning of the pandemic, however, the situation was not very clear-cut.Initially, there was a lack of scientific consensus about the efficacy of masks as a protection against the spread of SARS-CoV-2, and there was also a global scarcity of masks (Kirubarajan et al., 2020).The Czech Republic was no exception in this regard, but the government still made it obligatory to wear a face mask when leaving one's home.This decision was made in a context when face masks were already sold out nationally, and were even unavailable to most medical professionals.This national policy on face masks could therefore only be made possible through the actions of another and more heterodox actor: citizen volunteers.
That striking story of the Czech home-made face mask-making began with a critical failure of the national government to assure its citizens that the country was prepared for COVID-19.Once the health minister admitted in March 2020 that the Czech healthcare system needed around 1 million respirators, as well as other medical equipment unavailable at that point, a powerful entrepreneur intervened into this market failure created by the pandemic with its unpredictable surge in demand for some medical devices.The Czech businessman, self-help author and social media personality Petr Ludwig posted a YouTube video calling for a bottom-up action aimed at protecting the nation against the spread of the pandemic.The video promoted a simple message: wear a face mask and, if necessary, make your own.Ludwig had drawn inspiration from Mongolia, which had made wearing face masks mandatory in January 2020 and had effectively prevented the spread of COVID-19.His video quickly went viral and set into motion a curious case of decentralized social mobilization -based upon volunteerism and radical self-reliance.People started stitching improvised masks from shirts, curtains, bed linen, and even vacuum cleaner bags.They usually made these first for their family members, but then also started distributing them to hospitals, elderly homes and among themselves.These self-made masks were mostly donated rather than being sold for profit, with many participants condemning any commercial profiteering during the national shortages.While the government struggled to coordinate emergency supplies from China, Czech media was already reporting on the success of 'the whole country sewing'.
What started as a bottom-up societal response to the state's failure eventually become co-opted by the government authorities.On 17 March ministers started wearing face masks in public, and from 19 March onwards covering one's mouth and nostrils in public spaces became mandatory.As proper masks were not universally available, however, those measures were declared in extremely vague terms: the government made compulsory the use of 'any protection of mouth and nose' -even mere 'head scarfs, shawls, and other means' should do (Government of the Czech Republic, 2020b).As the interior minister, Jan Hamáček put it, 'anything is perhaps better than nothing'(Government of the Czech Republic, 2020a).When the Prime Minister addressed the Czech nation on 23 March, the protective mask on his face signified how to defend oneself against the virus and sent a clear signal to the public: the government had not only listened but also appreciated and thanked citizens for their activity.At this point, masks had become the 'new normal' in Czechia.
These Czech volunteers stood in a heterodox positionality to the more established field of global health governance.First, those mask-sewing volunteers operated outside the traditional institutional architecture of global health governance: a loose network -of predominantly women -who started sewing face masks on their own, and without being associated with any formal organization or institution.Second, the volunteers also pushed a hygienic measure that was, at the time, based on intuitive knowledge rather than on clear scientific evidence, thus contradicting the democratic ideal of evidencebased policymaking.In fact, other states and international institutions like WHO only encouraged the use masks under specific circumstances at that time (Feng et al., 2020).When the Czech government declared the policy of obligatory cloth face coverings on 19 March 2020, it was the first non-Asian country to react to the COVID-19 crisis in this way.While the shortage of mask stockpiles in other countries led their political elites to discourage mask wearing, Czech politicians instead encouraged citizens to stitch masks without citing any evidence for this measure.
Over time this heterodox 'hygiene infrastructure' also began to interact with wider socio-material arrangements and more established public health infrastructures.In terms of social organization, two types of engagement could be observed: individual mask sewing and community collaboration, typically related with some distribution of tasks.These communities were created bottom-up, and rarely with any help of local municipalities.This swiftly organized social solidarity was striking in terms of how it cut across classes, generations and localized global North/South divides: it involved celebrities as well actors usually at the margins of the liberal global health order -like Roma, Vietnamese, refugees, people in care houses, prisoners, etc.The infrastructure for mask distribution also relied heavily upon social media.Numerous initiatives sprung up, like the Facebook group Česko šije roušky (Czechia stiches masks), in which people shared tips for mask stitching and offered masks, or an online community map Dáme roušky (Let's have masks; cesko.digital,2022), which helped distribute more than 600,000 masks.Beyond such online platforms, there were local coordination centres (e.g.organized by theatres or cafés) which coordinated the distribution of masks to care homes, hospitals and so on.Some public institutions assisted the working of the rebel infrastructure too, as for instance Czech Post made it possible to ship masks for free.
Eventually this citizen infrastructure also became more closely integrated into government infrastructures, helping the state -or other 'recognized' political actors -to perform their functions (cf.Helmke and Levitsky, 2004).The state would not only come to acknowledge the citizen infrastructure; its policy of obligatory masks effectively relied upon the responsibilization of people and their ability to get home-made masks.On one hand, the establishment and good functioning of the alternative hygiene infrastructure was thus interpreted as an insult to the state, as ordinary citizens were more effective in protecting themselves in the time of crisis than the government.On the other hand, the social mobilization did not in any way translate into political action.The movement never attempted any form of formal organization and its activity and visibility gradually declined, as people eventually stopped sewing the face masks when surgical masks became commercially available.For instance, the relevant Facebook groups mostly turned into community fora about home sewing and home improvement.A contributing factor to this nature of the movement is that mask-sewing was perceived as an apolitical activity -an act of care (Gorčíková et al., 2020).
In the end, the vigilante movement's key impact was thus to expose the unpreparedness of the Czech government to a major public health crisis, as well as the limits of (neo) liberal governance that relied upon the logic of free trade.The volunteer mask-sewers operated in the context of a comparatively wealthy and liberal European country with an open economy and well-developed state institutions.Still, no private company -nationally or globally -was able to produce and distribute the protective measures as quickly as efficiently as the citizens themselves.In this time of crisis, a grass-roots movement, organized virtually overnight and with no formal governance structure, co-produced vital knowledge and best practices on how to manage public health crisis, while taking pride in doing this without any reward and as a voluntary activity.Having built an alternative health infrastructure operating under the logic of care -rather than financial profit -and relying on citizen solidarity, unpaid labour, and radical self-reliance, this well-meaning heterodox actor thus filled the vacuum of pandemic (un)preparedness left by the liberal state and market.

Mavericks: the international sharing of pathogen sequence data
The life sciences -encompassing scientists, universities, research funders, and biomedical companies -are another prominent set of actors foregrounded within the liberal episteme of global health governance, principally because of their critical role in developing new medicines and vaccines (Gostin et al., 2014;Long, 2021).International organizations like WHO have therefore long championed liberal principles of international scientific cooperation as a mechanism for improving health globally (WHO, 2016).During COVID-19, however, a third type of heterodox actor was operating in the background of this scientific field, where it directly challenged established liberal norms around the unrestricted international sharing of scientific data.Scientific and technological advances in sequencing technology make it possible now to decode the genetic sequence data of new viruses in near real time (Long, 2021;Rizk et al., 2020).Much like face masks, digital pathogen sequence data has thus become an increasingly visible feature of the international response to COVID-19.Throughout the pandemic, the media reported how scientists harnessed those sequence data to identify the new virus causing the outbreak, track its geographic spread, identify emerging variants, and develop new biomedical interventions (Elbe, 2021).
Until very recently, the world's leading repository for sharing such pathogen sequence data around the world was GenBank -a public database hosted by the US government through the National Institutes of Health (Benson et al., 2013).GenBank is historically closely associated with the open science movement and operates a public domain regime of data-sharing based upon liberal norms of unrestricted international data sharing.This means that once a sequence is deposited in GenBank, anyone around the world with Internet access can anonymously download those data and use them freely.GenBank's data sharing operations also demonstrate a broad commitment to the liberal principles of laissez-faire and minimal government interference because GenBank places no additional restrictions on how this sequence data is subsequently used or distributed.This liberal approach allows sequence data to flow quite seamlessly into scientific research around the world, and to inform the commercial development of new medical products.
That laissez-fair model of unrestricted sequence data sharing is becoming progressively challenged, however.The need for an alternative data sharing model arose, in part, because of stark and persistent global health inequalities (Chatham House, 2017).From a postcolonial perspective, many low-and middle-income countries (LMICs) are concerned that this liberal model of unrestricted data-sharing perpetuates the extractive tendencies of colonialism, and allows scientific resources to be moved from the Global South for the primary benefit of populations in high-income countries.There is particular apprehension about how this liberal approach to data-sharing culminates in the highly unequal international access to new vaccines witnessed once again during COVID-19.LMIC scientists have voiced further unease about being excluded from authorship arrangements when scholarly analyses using those shared data are presented at prestigious conferences and published in leading journals (Sedyaningsih et al., 2008).What is more, LMIC governments remain aware that sequence data can potentially be of immense commercial value, and will not wish to freely give away something valuable by sharing it in a public database (Hilberg, 2022).From a postcolonial perspective, there are thus profound political tensions inherent in the liberal model of laissez-faire data sharing; and those tensions have already prompted some governments and scientists to withhold pathogen sequence data during recent outbreaks (Elbe, 2021; IOL, The issues at stake in this field of sequence data are ultimately not just scientific; they also pertain to broader questions of global equity and national sovereignty (Noorden, 2021).
In response to such international sensitivities, an alternative data-sharing mechanism called GISAID has emerged more recently.GISAID sidesteps the liberal norm of unrestricted data sharing in favour of an alternative sharing mechanism based on the principle of data licencing.In contrast to GenBank, any sequence data submitted through GISAID is governed by its unique database access agreement (DAA) that provides data contributors with greater protections about how data will be used.This binding DAA places legal restrictions on the distribution of data by allowing GISAID users to share data between them, but not with other non-GISAID users.The DAA further entails that scientists accessing data through GISAID will credit the use of others' data in publications, will make best efforts to collaborate with the originating laboratory in their analyses and research, and will maintain common access to technology derived from the data so that it can also be used for development of products such as diagnostics and vaccines (GISAID, 2011).Crucially, this alternative data licencing mechanism also means that any sequence data submitted through GISAID do not fall under the legal definition of 'public domain' since GISAID does not remove or waive any potential pre-existing rights to the data (GISAID, 2011).Overall, then, GISAID retains the principle of having a publicly accessible sequence database because any natural person can obtain access credentials to data in GISAID.Unlike GenBank, however, it pioneers a licencing mechanism to place meaningful restrictions (and obligations) upon how such licenced data is subsequently used.Along with several other aspects, this has made GISAID particularly attractive to low-and middle-income countries (Maxmen, 2021); and GISAID therefore now forms a direct challenge to the liberal laissez-faire approach of unrestricted sharing promoted by GenBank.
For many years after its initial inception, GISAID remained a backgrounded actor in global health governance.It's alternative licencing model was not widely endorsed within the field of global health governance; nor was its licencing model as well established as GenBank's public domain model.GISAID was, furthermore, developed by the influenza community and so initially was only used to share influenza sequence datamaking the scope of data shared through GISAID much narrower than GenBank's.Outside of the influenza community, GISAID was therefore not a particularly wellknown organization prior to COVID-19 -even if it occupied a crucial niche in global health governance by serving as the world's leading source of influenza virus sequence information.COVID-19 has dramatically elevated GISAID's international profile, however.In response to the pandemic, GISAID expanded its platform to also host sequence data for SARS-CoV-2 (Fearnley, 2020).Over the course of the pandemic GISAID then managed to amass more than 13 million SARS-CoV-2 sequences at the time of writing.This means that genomic sequences for SARS-CoV-2 are being shared through GISAID now at a historically unprecedented rate (WHO, 2021: 1); and GISAID is also managing to attract much more such sequence data now than older public domain databases like GenBank (Wadman, 2021).Already GISAID is therefore making quite a constitutive contribution to global health governance -by facilitating the rapid global exchange of high-quality sequence data about SARS-CoV-2.
Yet this growing international success of GISAID's more heterodox model is also engendering significant tensions within the existing field of global health governance.First, the major driving force behind this new initiative came from outside that established institutional architecture of global health governance.In a way that echoes the Czech case study above, GISAID's genesis is strongly associated with Peter Bogner -a former studio executive at Time Warner with a background in media and philanthropic work, but without prior global health experience (Elbe and Buckland Merrett, 2017).Although the initiative began to receive substantial support from the Ministry of Food and Agriculture in Germany from 2010 onwards, it is not managed through a major international organization like WHO, or a national government in the way that GenBank is.Instead, GISAID's activities are coordinated through a small and little-known association called Freunde von GISAID e.V. [Friends of GISAID] registered in Germany (GISAID, 2021).From its inception, GISAID's 'outsider' positionality has provoked questions about the status of this initiative, and the extent to which it can be trusted with such data (Wadman, 2021).In the run up to GISAID's inception in 2008, for example, WHO unsuccessfully attempted to create its own public domain system of virus data sharing, and reportedly prevented the release of funds intended to aid the development of GISAID (IHT, 2008).GISAID, in turn, has openly raised concerns about the secretariat of the Pandemic Influenza Preparedness (PIP) Framework operating within WHO (Saez, 2016).Other actors within WHO have also been more supportive, however, with WHO's Chief Scientist acknowledging GISAID as 'a game changer' (Swaminathan, 2020).Nevertheless, GISAID has for many years experienced recurring tensions with long established global health governance actors like WHO.
Second, GISAID also remains in tension with the many scientists who would still prefer the liberal norms of unrestricted sequence data sharing (Noorden, 2021).Some of those scientists argue that a system with no additional restrictions would allow data to flow more seamlessly and deliver even faster results (Maxmen, 2021).Many scientists even published a prominent open letter, garnering more than 800 signatures at the time of writing, calling for such SARS-CoV-2 data to be submitted to the databases like GenBank that are part the International Nucleotide Sequence Database Collaboration (INSDC) -though notably the overwhelming majority of those signatories are based in Europe, the United States and Canada (Maxmen, 2021).During COVID-19 there have also been reports of sporadic attempts to undermine the GISAID model (IOL, 2021).Like the rebel armed groups and vigilante citizens' movements analysed above, then, GISAID too stands in partial tension with established liberal architecture of global health governance, while its data-sharing infrastructure nevertheless contributes substantially towards the informational co-production of global health governance in the 21st century.

Conclusion
COVID-19 has revealed many unresolved global governance challenges (see Legge, 2020).The countries deemed best prepared for a global health crisis did not always deal very successfully with the pandemic, while surprisingly effective solutions sometimes emanated from unexpected actors not usually recognized within the established governance architecture.Many working in the field of global health governance are responding to these shortcomings by attempting to further improve the institutional workings of the global health architecture (e.g.Gostin et al., 2020).This article, by contrast, has viewed those copious governance shortcomings as an invitation to re-think seemingly unquestionable global health practices.The article has therefore proposed and developed a new approach to the study of global health governance -an architectural inversion -that analytically foregrounds the role of more heterodox global health actors working beyond the architecture of liberal international order in many parts of the world.
Exploring those heterodox actors is not the same as politically endorsing or normatively commending them; but it can make three wider contributions to the study of global health governance.First, performing such an architectural inversion begins to challenge the dominance of the liberal episteme within that field, and works towards a more pluralistic understanding of global health governance -as something that is broader and more diverse than the organizations prominently headquartered in metropolitan centres like Geneva, Washington, London and Berlin (Hoffman et al., 2015).Thus, the article has highlighted a group of much more heterodox health actors operating quite diverse epistemic, social, and material infrastructures: Myanmar's rebels, the Czech self-help movement and GISAID.Their vital differences notwithstanding, each of those non-state actors departs from one key dimension of the liberal episteme of global health governance: as armed rebels to the postcolonial state who produce public health in their controlled territories, as citizen-led vigilantes that deliver emergency protection in the context of state and market failure, or as maverick initiatives that develop alternatives to the liberal principles of unrestricted data sharing.Foregrounding those actors has not just revealed how messy and complex the management of the pandemic has been in different parts the world, but also how such heterodox actors can engender a diverse array of different health governance logics -like ethno-national rather than cosmopolitan ideologies (rebels), logics of care rather than commercial profit (vigilantes), or different principles of international data sharing (mavericks).In the first instance, greater attentiveness to heterodox actors thus contributes towards an epistemological pluralization of the scholarly study of global health governance.
Second, an architectural inversion has also revealed, analytically, how that field of global health governance is always already made possible by the activities of many such heterodox actors operating all around the world.Through its empirical case studies, the article has explored how those heterodox actors form a thick and co-constitutive -if backgrounded -substrate within the field.Despite their heterodox positionality in relation to the liberal episteme of global health governance, all three actors operate material infrastructures that productively co-produce global health in different ways.This is perhaps most obvious in the case of GISAID.Despite its tension with established norms and institutions in global health governance, the maverick platform makes a major co-productive contribution to global health, by amassing the world's largest pool of high-quality SARS-CoV-2 sequence data, which has become central to pandemic response.On a domestic scale, the Czech citizen-led movement stitching masks in a global pandemic demonstrates how commonly backgrounded actors are central in coproducing public health in contexts of emergencies and failing state and market responses, increasingly also in the neoliberal 'global north'.Yet even in contexts of pronounced conflict between the state and non-state actors, such as in the case of Myanmar's rebel movements, heterodox actors are crucial to creating and maintaining significant health infrastructures within the territories under their control and thus co-produce key public health infrastructures, locally and internationally.
Beyond those epistemological and analytical contributions to the study of global health governance, the architectural inversion performed here finally also contributes to the theoretical development of this wider scholarly field -by systematically incorporating interdisciplinary insights from more relational social sciences like post-colonial studies, infrastructure studies and STS.Drawing upon those methodological approaches shows global health governance to also be a field co-produced by a plethora of informal practices, contested knowledges, unrecognized networks, and often unpaid labour of heterodox actors.That opens a vital space to study many more (and even more diverse) types of heterodox actors moving forward.Beyond the three case studies explored here, there are evidently many other types of rebel movements and non-state armed groups co-producing public health despite their conflict with state actors worldwide (Breslawski, 2022;Furlan, 2020;e.g. Sahay et al., 2016).There is also no doubt that vigilante movements co-producing public health in contexts of increasingly hollowed out state capacities exist elsewhere (Low-Beer and Sempala, 2010;Restoy and Elbe, 2021).Indeed, the mushrooming of self-help groups across Europe during COVID-19 suggests that coproducing health between the state and heterodox grassroots actors is not a phenomenon confined to countries of Global South.Similarly, other examples of maverick health actors that challenge established institutions also come to mind -such as ProMED-mail, which was established in 1994 as new way of using email lists as an early warning system for several outbreaks (Carrion and Madoff, 2017;Yu and Madoff, 2004).Nor are rebels, vigilantes, and mavericks even the only conceivable categories of heterodox actors.Here, then, an architectural inversion generates substantial scope to expand this kind of analysis in future to also include a much broader range of heterodox actors and logics that have hitherto been backgrounded in the study of global health governance.
In the end, those contributions do not only lay new conceptual foundations for a more heterodox research agenda on global health governance, but arguably also on global governance more broadly.That is because the crisis in global health governance revealed by COVID-19 is increasingly mirrored by a similar sense of crisis engulfing global governance more widely.In that field, too, the authority and capacity of liberal institutions to respond to global challenges appears to be decreasing, and liberal orthodoxy is being challenged from multiple sites (e.g.Lake et al., 2021).In that context, there could be benefits for broader scholarship in IR to move towards a more heterodox research agenda -to better understand the multiple and decentralized nature of those alternative logics and practices of governance.That could potentially allow alternative conceptualizations and theorizations of global governance to emerge in future -as has already begun to happen with the rise of heterodox economics.Of course, the political implications for rethinking these fields through such heterodox actors remains a significant challengebecause recognizing the importance of actors who operate beyond liberal international institutions, liberal economy, and liberal norms, would mean to acknowledge the significance of other than orthodox liberal frameworks and policies.Given the many protracted fissures, inadequacies and global injustices that permeate the existing system of global governance, however, perhaps the COVID-19 pandemic needs to be seen as a call for such an urgent rethinking of the fields' broader parameters, including its political and ideological pluralization.The study of heterodox actors in global governance can make one contribution towards that wider effort.