Situated dynamics of environmental governance in Swedish smart energy experimentation: Tentativeness, demonstration, upscaling

In this article unfolding processes of environmental governance are explored through the case of smart energy experimentation in Sweden. When experimentation merges with place-based dynamics, particular dispositions are shaped, with different implications for environmental politics. In accordance with other researchers’ findings it is illuminated how such dispositions tend to reinforce neoliberal environmental governance arrangements, whereas it is revealed how experimentation can also reconfigure governance. Attending to unfolding processes of accomplishing governance across a variety of spaces, this plasticity of experimentation is explored. The contribution consists of highlighting the situated dynamics playing out as environmental governance is produced across different places. This involves two analytical steps. First, disentangling some characteristics of experimentation as a situated disposition, through following how smart energy experimentation arranges environmental governance across the national Swedish energy and environmental policy landscape. The analysis highlights how particular characteristics of Swedish environmental governance molds experimentation into providing impetus for neoliberalization. Second, by zooming in two cases of experiments, elucidating the effects of the tentative and situated character of experimentation as governance is accomplished. Such analysis attends to the situated micro-politics of experiments, and how these can shape localized dispositions that refract broader neoliberal political rationality. Thereby the existence of a dominant ‘experimentation governmentality’, or disposition, is discovered, whereas situated experiments may always provide seedbeds of alternative forms of environmental governance due to experimentation’s plasticity.


Introduction
During a relatively short period of time, experimentation has become pervasive. Filtered out from the confines of laboratories, industrial trials and demonstration plants, testing is increasingly widespread across governance settings (Marres and Stark 2020). In public administrations, pilot projects have had a resounding impact as an organizational model (Hodgson et al., 2019). Few policy areas have become as emblematic of the breakthrough of experimentation as environmental politics (Bulkeley et al., 2015). In the past decades, cities and regions have emerged as central actors and arenas for such experimental governance, regularly framed as necessary to deal with the complexities of environmental issues (Bulkeley and Castán Broto 2013). As experimentation gains in popularity, a gradual shift has occurred within the EU where also traditional urban planning approaches are increasingly complemented with experimentalist approaches, leaning into promises of testing and innovation (Rangoni and Zeitlin 2021;Mukhtar-Landgren 2021). Environmental governance entails deep uncertainty in relation to the future, requiring governance approaches "that are more explorative, adaptive, learning-based, and evolutionary in nature" (Sharp and Raven 2021: 195). Coupled to ambitions of increased citizen participation, experimentation has been widely adopted as a mode of environmental governance (Eneqvist and Karvonen 2021).
Experimentation holds special salience within energy policy. Many pilot projects on so-called smart energyinterlacing energy systems with ICTare currently unfolding across the globe, with environmental ambitions as primary hallmark. In Swedena nation which has historically labeled itself a green frontrunnersmart energy has become intimately interconnected with environmental ambitions, and governmental effort has been devoted to foster smart energy experimentation (Envall 2021). Scholarly attention dedicated to experimentation in relation to environmental governance and energy transformation has mostly been concerned either with how experimentation reconfigures environmental governance through the prism of case studies of singular experiments (e.g. Levenda 2019), or with broad appraisals of the political implications of experimentation (e.g. Karvonen et al. 2020).
I will explore experimentation differently. Through following experimentation as a mode of governance across several places, I analyze what particular governance configurations are shaped. This move bridges analyzes of singular case studies and broader appraisals of the politics of experimentation.
The exploration hinges upon two analytical steps. First, an analysis of how experimentation is discussed and instituted at the national level in the Swedish smart energy policy landscape, to distinguish how experimentation merges with place-based dynamics of Swedish environmental politics. Thus it becomes possible to analyze how experimentation fuses with pre-existing discursive elements and governance arrangements. Second, by zooming in two cases of smart energy experiments. This allows for analyzing how experimentation as a mode of governance fuses with local place-based dynamics in the vicinity of experiments as governance is arranged in practice, or accomplished (Bulkeley 2016). The situated micro-politics that ensue can elucidate how this fusing engenders localized dispositions. Such analysis contributes a spatially sensitized conceptualization of the politics of arranging environmental governance through experimentation. The key questions animating the exploration center on what experimentation "does" to environmental governance and vice versawhich political dynamics and specific arrangements are produced?

Environmental governance experimentation and neoliberalism
Some central characteristics of experimentation in environmental governance settings include: tentativeness, as experiments revolve around testing; demonstration, e.g. of new technologies; and upscaling, as small-scale interventions with the aim of broader implementation of e.g. an organizational arrangement (cf. Bulkeley et al. 2015;Mukhtar-Landgren 2021). Scholars have uncovered how experiments, owing to such characteristics, tend to reconfigure governance arrangements, for instance through redefining the roles of urban planners and the relationship between public and private political domains (Kronsell and Mukhtar-Landgren 2018;Karvonen et al. 2020). It has been noted in particular how experimentation often resonates with neoliberalization processes (e.g. Levenda 2019). Maarten Hajer argues that there are obvious compatibilities between the increasing pervasiveness of experimentation and long-running neoliberalization processes, visible in particular in the public sector (Hajer 2016). For him, governance arranged through new public management (e.g. arranging public administration in line with market logic as a guiding principle and the proliferation of public-private partnership constellations), vindicated by neoliberal rationality of deregulation and privatization for economic efficiency, has provided fertile ground for experimentation. Austerity policy in particular, with calls for "doing more with less", has paved the way for experimentation and elucidates linkages to neoliberalization (Hajer 2016). In such contexts the appeal of experimentation lies in promises of quicker learning loops than traditional policy processes while involving more actors in "agile" projects, often engendering circumvention of conventional governance processes and resulting in depoliticization (Adkins and Ylöstalo 2018;cf. Brown 2015). Some scholars claim that experiments are at the forefront of neoliberal development, as the political effects of experimentation increasingly extend throughout social life (Peck and Theodore 2015). While experiments are in practice always complex, they tend to reinforce neoliberal governance arrangements (Adkins and Ylöstalo 2018;Mannevuo 2019). Specifically experimentation contributes to neoliberalization as it tends to be set up in favor of large corporations, shifting influence over political processes to these actors through various depoliticizing moves (Caprotti and Cowley 2017;Cowley and Caprotti 2019;Söderström et al. 2014).
Wendy Brown has argued that neoliberalism can be productively gauged as a particular political rationality. For her, a neoliberal power-knowledge regime has disseminated throughout the conduits of society to permeate social life, from statecraft and education to everyday life (Brown 2015). Hallmarks of neoliberal governance include fiscal discipline, deregulation, privatization, and tax cuts (Birch and Mykhnenko 2010). Governance arrangements emanating from this political rationality include best practice benchmarking, devolution, and technicized policymaking (Brown 2015). The resemblance with Hajer's argued convergence between neoliberalization and experimentation is striking; technicization is usually a hallmark of austerity policy whilst projectified governance setups for increased agility and quicker learning loops reflect the neoliberal tendency of devolution. Thus, a compatibility between neoliberalization and experimentation can be observed. Above all, argues Brown, neoliberalism as a political rationality is characterized by the marketization of virtually all aspects of social life (Brown 2015). While there is common ground between policy programs implemented in such a neoliberal vein across the globe, I will speak of neoliberalization processes (Peck and Tickell 2002). Thus, one avoids describing neoliberalism as an omnipotent monster, or a stable program or coherent end-state (Birch and Mykhnenko 2009). Neoliberal governance arrangements tend to be enacted as responses to crises engendered by contradictions inherent in neoliberal political rationalityindicating that neoliberalization comprises an amalgamation of policies, practices, and discourses that are always entwined with preexisting webs of relations (Ong 2006). This underlines the necessity of sensitizing analysis to placebased aspects by following governance as it is produced, unpacking experimentation as a situated disposition.
Projects as an organizational form is central for such experimental governance setups; some scholars speak of a broad projectification of e.g. the state apparatus (cf. Murray Li 2016). Projects are almost ubiquitous in environmental governance experimentation, not least in smart energy and smart city endeavors (Bulkeley et al. 2019). Smart energy experiments are often arranged as spatiotemporally confined public-private partnership projects, where concerns have been raised that such arrangements contribute further to neoliberalization of environmental governance by excluding structural analysis of social and environmental issues in favor of piecemeal project interventions (Späth and Knieling 2020).
Studies on experimentation's implications for environmental governance tends to consist either of broad appraisals of experimentation (Caprotti and Cowley 2017;Edwards and Bulkeley 2017) or case studies of singular experiments (Karvonen et al., 2019). This article contributes to such literature by charting a different path, through exploring how experimentation merges with placebased dynamics in unfolding governance processes both across the national Swedish environmental and energy policy landscape and through case studies of experiments. Thereby it becomes possible to discern situated dispositions shaped through experimentation, and the broader dynamics engendered as experimentation arranges environmental governance.
Theoretical approach: dynamics of producing governance The overarching theoretical approach in this article is rooted in the governmentality literature. Broadly residing within Foucauldian-inspired research, governmentality entails fixing the gaze upon "the ensemble formed by institutions, procedures, analyses and reflections, and tactics […]" (Foucault 2007: 108). This is often summarized in the term dispositif, or disposition, to highlight how systems of relations are continuously stitched together through processes of governingextending across different registers such as knowledge generation, discourses, and administration (cf. Braun 2014). Any disposition is underpinned by a rationale of governance, or a political rationality, underlining how governance processes are inherently dynamic; problems are constructed through the interplay between knowledge-making and specific devices (Rose and Miller, 2008). Consequently in such conceptualizations governance processes are not confined to the realms of formal government such as policy implementation; governmentality entails tracing processes extending for instance across the work of government agencies, international organizations, local public administrations, funding infrastructures, research strategies, regulations, scientific knowledge production, civil society organizations, and policy documents.
In this article a disposition is conceptualized as a bundle of discursive elements, such as the formulation of specific problems and solutions, and the knowledges actualized thereby, codified into different institutional arrangements such as regulations, guidelines, funding schemes, actor constellations, and policy instruments, forming a particular 'apparatus ' (cf. Hajer 1995: 61). Such a conceptualization makes it possible to trace how discursive elements gradually become stabilized through attending to governance processes as they unfold over timefor instance whether a particular problem that experimentation is intended to solve is written into policy documents and guidelines, thus influencing the setup of new experiments.
Hower dispositions are only ever temporarily stabilized, in need of constant rearranging. This necessitates engaging analytically with uncovering the "relational dynamics of governance that change over time and across space" (Colona and Jaffe 2016: 8). As an analytical term disposition risks implying an unwarranted sense of stability, and the interest in this article lies in how environmental governance is produced through experimentation by following unfolding governance processes across several places. This is the intended contribution; through this move it becomes possible to catch a glimpse of the broader implications of experimentation for environmental politics that goes beyond too rigid characterizations of experimentation as a stable entity and case studies of singular experiments. To realize this contribution I will lean into Harriet Bulkeley's conceptualization of accomplishing governance. This approach is rooted in governmentality but entails conceptualizing governance as practices of continuous socio-material assembling (Bulkeley 2016). Bulkeley argues that environmental governance can productively be analyzed as processes that are constantly expanding but always situated, and as an accomplishment always in the making. This necessitates attending to how interventions are accomplishedto how they are practiced and Envall contestedacross a variety of places (Bulkeley 2016). Bulkeley states that "[…] a critical engagement requires that we examine the dynamics through which governing is producedto the ways in which climate governance is accomplished.
[…] Understanding how governance is accomplished is then a matter of examining how it is performed, fulfilled and completed in relation to different desires and objectives through the acquisition and honing of particular skills and techniques" (Bulkeley 2016: 14). Analyzing experimentation as a disposition arranging environmental governance, then, requires analyzing how diverse socio-material elements are assembled into relationsrelations that are always prone to destabilization or reconfiguration (McGuirk et al. 2016). Rooted in insights from governmentality, the approach of accomplishing governance tells us that governance arrangements are never completely stabilized; stabilization requires work and might always be disrupted (Bulkeley 2016).
This approach allows for understanding dispositions as always in the making, and thus more or less unstable. Such a conceptualization helps focusing on the merging of place-based dynamics and experimentation as a mode of governance, through focusing on governance as constantly (re-) assembled in socio-material practices; some relations are shaped while other possible configurations remain un-assembled. Through following processes of accomplishing governancein policy documents, interviews with experts at various organizations, and observations at experimentsit becomes possible to trace how dispositions are shaped. This includes broader dispositions on a national level as experimentation merges with place-based dynamics of Swedish environmental and energy politics as well as localized dispositions as experimentation fuses with place-based dynamics within the local geographies of experiments.
From literature exploring experimentation and environmental governance, we might expect that such merging engenders neoliberal configurations. The approach outlined here will allow for disentangling experimentation in a situated manner, looking across several spaces of governance as such processes unfold to explore the situated dynamics emerging when experimentation arranges environmental governance, in terms of the arrangements shaped and the relations enactedi.e. how experimentation intermingles with environmental governance in a concrete setting through connecting analysis of the broader policy landscape with local case studies, thus sensitizing analysis to place-based dynamics. Such an approach can help specify how characteristics of experimentation mentioned previouslytentativeness, demonstration, upscalingfeed into and play out in relation to place-based dynamics.
The approach of accomplishing governance is particularly suited to following micro-politics across cases of experiments. The concept of disposition helps fixing the gaze upon codified and institutionalized arrangements and how these are shaped by place-based dynamics across the Swedish environmental and energy policy landscape as experimentation arranges governance, whereas the concept also allows for distinguishing 'snapshots' of temporarily stable, localized dispositions in cases of experiments. These two analytical steps can bring new insights as to the implications of experimentation for environmental politics, contributing to the literatures on environmental governance and experimentation. Crucially, this conceptualization spatializes experimentationa move that can reflexively develop our understanding of the implications of experimentation for environmental politics.

Methodological approach
The analytical strategy is tailored to investigate and compare cases of experiments, to examine whether governance is accomplished differently, while also digging into how experimentation arranges governance on the national level. To disentangle some characteristics of situated dispositions and thereby unpack how experimentation arranges environmental governance, the analysis spans different places where governance is produced. These include the state apparatus and public administration, including the government offices, governmental agencies, and international organizations, and reach into places where experiments unfold. The analysis is based on 31 interviews with actors across these spaces, document analysis of a significant text materialincluding e.g. national policy documents and reports as well as policy documents from local public administrations and project documentsand observations in the vicinities of experiments (including site visits as well as public dialogue forums, hearings, presentations etc.). Interviewees include representatives of various state agencies, the government offices, energy companies, grid operators, civil servants employed in local public administrations, and international organizations. 1 Interviews lasted between 60-120 minutes. All fieldwork was conducted in iterations between 2017-2019. The material was subsequently thematically coded and analyzed using the coding software Atlas. ti.
During the empirical work there were many parallel processes of policy development ongoing across the national environmental and energy policy landscapesmeaning that networks of actors were being assembled, plans made, and strategies, policy documents, and pre-studies were published. Consequently the material is apt for elucidating unfolding governance processes and the ideas underpinning them. Following such processes across the various places just mentioned, and gathering a vast document material over time, made it possible to follow how particular actors posed particular problems and posited solutions in relation to these, as well as the actualization of knowledges in line with these statements that were increasingly codified into e.g. policy documents. The length of time of empirical work made it possible to follow such processes over a longer duration, whereas doing several rounds of interviews and observations during a stretch of years provided insight into processes of accomplishing governance. Thus the empirical material was thematically coded accordingly, to spot shifting problematizations, follow how particular knowledges became increasingly important and codified into policy documents, as well as how particular relations between actors, material infrastructures such as the power grid, and various devices shifted. During fieldwork and when coding, particular attention was devoted to whether and how place-based dynamics came into play in the unfolding governance processes, e.g. whether different problematizations were tied to particular geographies or influenced by discursive elements specific for Swedish environmental discourse. Through this research design it is possible to analyze how experimentation emerges in multi-scalar landscapes, by following how environmental governance is arranged in practice across different spaces of governance, constituting a link between particular experiments and broader governance arrangements (cf. Braun 2014).
The cases are located on the island of Gotland in the Baltic Sea and in the city of Västerås, west of Stockholm. On Gotland, a large governmental initiative is currently playing out, whereby the island has been appointed a pilot region for transforming to a smart and sustainable energy system as well as reaching environmental policy goals before the rest of Sweden. The Swedish Energy Agency initially conducted an investigation of how this could be achieved and has together with other governmental agencies remained involved in the initiative beyond the initial investigation.
In a new city district under construction in Västerås, the self-described climate-activist company ETC, backed through crowdfunding, attempts to build affordable wood-constructed 'plus-energy standard' houses and a local solar-powered DC nanogrid, with plans for engaging residents in energy transformation in their everyday life practices. The Västerås case is primarily based on one interview, besides publicly available documentation about the project collected online. The case thus stands out in comparison to the other empirical material considering its small sample size. Despite this it is still motivated to include since it highlights important contrasts of accomplishing governance. Analytically it will be treated as highlighting an actor attempting to experiment differently, marking a significant difference towards most other Swedish smart energy experiments.
Both cases encompass explicit ambitions of transforming the energy system by making it smart and sustainable, and include technical infrastructure, regulatory aspects, users' behavior, and actor constellations. Together with ambitions of testing, learning, and scaling up, this underlines their Envall experimental status. Such ambitions, coupled to the early phase that both experiments were in during the time of empirical work, makes them compelling cases that can grant insight into unfolding processes of accomplishing governance.

Situating smart energy experimentation in Swedish environmental and energy politics
Swedish energy and environmental policy are intimately interconnected. Contemporary energy policy rests upon three overarching pillars: security of supply, competitiveness, and ecological sustainability. There is a substantial emphasis on innovation as a means to govern the energy system in a manner coherent with these pillarsrevealing the ecomodern orientation of Swedish environmental politics (cf. Lundqvist 2004). Ecomodern environmental governance comes in different guises, but generally it is a problem-solving approach to governance emphasizing the necessity to unleash entrepreneurs in an effort of greening the capitalist political economy through new technological innovations, thus casting environmental issues as impetus for economic growth (Hajer 1995).
In Sweden, the roots of such ecomodernism have branched out across the state apparatus. Already in the mid-1990s the largest political party, the Social Democrats, adopted ecomodernism as a grand societal project (Anshelm 2002). Since then, ecomodernism has become increasingly embedded within the state apparatus to the extent that it can be argued to provide overarching direction for environmental and economic policytoday epitomized in the ambition of becoming "the world's first fossil-free welfare state" through green technology exports, firmly rooted in the ecomodern conviction that environmental action can boost national economic growth and Swedish firms. Swedish energy politics in particular has undergone a significant shift during ecomodernism's ascendancy. Historically, energy politics were on the one hand characterized by intense conflict dynamics within and outside of the parliament (particularly related to nuclear power), and on the other hand by corporatist governance setups emphasizing consensus-seeking and technocratic governance ideals, up until a contemporary inclination to arrange governance in line with neoliberal principles (Anshelm 2000;Högselius and Kaijser 2010).
As in other traditionally social democratic countries, Sweden has gradually adopted neoliberal governance principles (cf. Mudge 2018). During the early 2000s, a new right-wing government was elected, breaking a long chain of predominantly social democratic reign. Erixon speaks of a new policy regime, comprised e.g. of "deregulation of financial, product and labour markets and rulebased macroeconomic policy" as "paramount features" (Erixon 2011: 3). In environmental politics, this inclination has led to hesitance to invest or proactively govern by the state, for instance related to the mining sector (Anshelm et al. 2018). Instead, instituting governance arrangements aligned with market principles has been favored, leaning into technicized forms of governance, visible e.g. in a widespread projectification and public-private partnership constellations pursuing commercial endeavors (cf. Hodgson et al. 2019).
The constitution of the energy system illustrates convergence between ecomodernism and neoliberal governance arrangements. The electricity system is interlinked to other Nordic countries as well as the European power grid, and the sector was marketized during the mid-1990s. There is an integrated Nordic electricity trading market, Nord Pool, whereas Sweden is divided into four bidding areas. Further, the Swedish energy system is a large-scale system with centralized ownership, where a few large energy companies make up the main players. Electricity generation primarily stems from large-scale nuclear power, hydropower, wind power, and combined heat and power, transferred from generation plants to energy users through large-scale grid infrastructure. Since the marketization, neoliberal governance arrangements have become increasingly pervasive across energy policy; the state has taken on a rather passive role, preferring to emphasize innovation by energy entrepreneurs and customers reacting to price signals to achieve the sought-after green growth (Envall 2021;Wallsten 2017).
Smart energy experimentation as arbiter of neoliberal environmental governance?
The following sections will explore how experimentation arranges governance across the national Swedish smart energy policy landscape. The analysis investigates how some central characteristics of experimentation merges with place-based dynamics of Swedish environmental and energy governance to discern a broader, situated disposition of experimentation, and the implications for environmental politics.

Smart energy experimentation as a situated disposition
Smart energythe promise of greening the energy system by interlacing it with ICTis ascribed the possibility to solve a range of issues interconnected with the goal of Sweden becoming the world's first fossil-free welfare state. Primarily, making power grids green in the Swedish context is interlinked with capacity limitations in the existing grid and the risk of power shortages in an increasingly intermittent system. Power and capacity limitations arise for different reasons, which can however mainly be connected to increased growthexpansion of cities and perpetual digitalization requires more electricity, while construction of electricity-devouring data centers contributes to such issues. Through integrating ICT into the energy system, so the idea goes, it becomes possible to shave and shift peak loads. Reaping such benefits primarily revolves around instituting market solutions (SOU 2014:84). This includes casting electricity users as customers and encouraging them to act rationally on price signals relayed to smartphones by technologies such as smart meters and algorithmic setups collecting data from the Nord Pool spot market. In the long run, the ambition of the Swedish government is for smart energy to become a significant export industry, sustaining the green frontrunner status Sweden wishes to uphold in the international arena (Interviewee A). This ambition is articulated as necessarily driven by entrepreneurs through marketized arrangements, visible across policy documents where it is e.g. underlined that a primary rationale of interweaving the energy system and ICT is to steer consumers to adapt their electricity use to price signals from the market (SOU 2014:84), or that state funding of energy research should be geared towards assisting businesses commercialize new technologies and reap global market shares (SOU 2017:2).
It is notable that the primary governmental body responsible for coordinating smart energy, the now dissolved Swedish Smart Grid Forum, were incorporated explicitly as an element of the government's broader export strategy and significant parts of their work revolved around promoting smart energy technology exports. For instance, one of the main outputs from the Forum was the publication of two strategies, one of which exclusively focused on commercialization of research results by exporting new products (Forum för smarta elnät 2017). This export orientation resounded in documents and interviews across the policy landscape in relation to experiments. Two government officials kept referring to "having something to show" during an interview, since otherwise "you can take up no position at all" in an international context, and this was described as a central value of smart energy experimentsthey are considered an avenue for marketing Swedish competence as well as Swedish firms as spearheading green technology development (Interviewee A; Interviewee B). Interviewees across governmental agencies expressed similar sentiments, whereas energy experts engaged in international organizations stated that the first large smart energy experiments in Sweden had served to raise attention and spread international awareness of Swedish smart energy ambitions, contributing to the goal of increasing exports (Interviewee C; Interviewee D). A committee member of the Smart Grid Forum argued that there was a need for deepened Envall collaboration between industry and academia; researchers should engage as a core part of publicprivate ensembles travelling to international business conferences to promote Swedish excellence in smart energy (Interviewee E). Having researchers accompany firms involved in experiments to international conferences was necessary to create legitimacy for the Swedish export ambitions, thus assigning research an instrumental role for fulfilling such ambitions. The quotes show how the export orientation aligns with a specific way of perceiving and framing experiments as something to showcase, and thus not only legitimizes experiments but also configures them in a particular way as a persuasion device mainly for private firms.
The broad export and market inclinations of smart energy experimentation outlined here chimes with neoliberal environmental governance. Across the Swedish smart energy policy landscape, experiments are assigned the role of greasing the wheels of the export machine in the shape of collaborative public-private partnership projects, be it through showcasing competence on the home ground through demonstrative experiments or through globe-travelling constellations forging contacts and marketing Swedish competence to attract capital. Ambitions of testing usually associated with experimentation was not articulated as a primary purposeexperimentation in the guise of showroom, substantiating the entrepreneur-led green growth ambitions, seemed more important. Crucially, this setup appears to tilt smart energy experimentation in favor of large corporations and depoliticizes energy politics by emphasizing consensual techno-managerialism in collaborative commercial experiments. This highlights two place-based dynamics which contribute to shaping experimentation in a way which sustains neoliberal governance arrangements. First, the deep ecomodern Swedish legacy, with promises of green growth through development and export of new technology, extended through smart energy experimentation. Second, the significant market emphasis that reverberates across policy documents and interviews, which underlines the gradual neoliberalization of Swedish environmental politics. The arrangements facilitated through smart energy experimentation illuminates how a broader neoliberal political rationality increasingly enveloping Swedish environmental politics is further extended throughout smart energy experimentation.

Characteristics of experimentation as a situated disposition
The outlined examples highlight how three distinctive characteristics of experimentation merges with place-based dynamics to shape a situated disposition that provides a vehicle for neoliberalization. First is the cross-fertilization of experiments' demonstrative character and the increasingly neoliberalized Swedish ecomodern ambition of becoming the world's first fossil-free welfare state. Experiments in a societal context, outside laboratory settings, can to an extent be argued as inherently revolving around convincing publicswhether the experimenters consider the intended public as in need of convincing of a particular technology or the superiority of a policy program (Möllers 2016). As reflected in the empirical snippets above, experiments always include an aspect of demonstration. As shown, this characteristic fuses with the significant export-driven and market-inclined focus of neoliberal environmental governance. This was manifested e.g. in the Smart Grid Forum, who were explicitly formed as not only a coordinating body for smart energy but also part of the government's export strategywith facilitating demonstrations to solidify the competence of large Swedish tech-firms as a central goal. This can be considered a translation of the problem formulation of Sweden becoming a green frontrunner through tech exports into an actual institutional arrangement, i.e. part of a situated disposition of experimentation.
Another central characteristic of experimentation as a situated disposition is its tentative character, owing to the ambition of testing inherent in experimentation. This notion is recurrent across policy documents and in interviews, where it is emphasized that experiments can contribute to energy transition with new knowledge about what works in practice through open testing. An interviewee at the Smart Grid Forum, for instance, stated that experiments can guide policy by providing an evidence-base extracted from experiences gained and knowledge generated in experiments (Interviewee F). Other interviewees underlined the importance of testing in practice before implementing new technologies or business models (Interviewee G; Interviewee H), i.e. experiments as testing technologies and social arrangements in real-life contexts. Across interviews, experiments are ascribed the ability to reduce uncertainty through iteratively testing what works in practice. However despite emphasizing openly testing and learning in theory, smart energy experimentation appears less open in practice. Rather, dominant large incumbent energy companies are poised to take advantage of experiments' purported openness. This became visible in an interview where a government official reasoned that it does not matter if major funders of energy research would like to see smart energy experiments formulated; if the business community is "not ready to do it or not interested enough to formulate a pilot […], it isn't possible to launch one" (Interviewee B). The interviewee continued by stating that it is not possible for the Energy Agency to plan for pilot projects, since launching such endeavors necessitates private firms who are ready "to come together" (Interviewee B).
The line of reasoning displays an inversed version of project funding. Rather than public funders allocating funds based on calls, in practice it is rather the other way around; actors who wishes to launch an experiment reach out to policymakers to voice their desire for funds when they are ready. Crucially, the actors referred to by the interviewee tends to be the large incumbents in the energy sector, with established access to policymakers. This is intimately connected to how tightly woven the rationale of smart energy experimentation is around the Swedish green frontrunner ambitions, based upon developing and exporting new energy technology. This ambition becomes easier to sustain with the help of large, established firms with strong brands and significant marketing reach. Thus, the setup of experimentation is tilted in favor of these actors, shifting political influence to such large corporations by depoliticizing environmental politics. This example underlines how the tentativeness of experiments also provides impetus for neoliberalization, when merged with the particularities of Swedish environmental discourse. Testing through smart energy experiments appears to sustain the notion that environmental issues are best handled through market arrangements. The interviewee's insistence that it is impossible for state actors to plan for pilot projects elucidates how experimentation becomes a vehicle for neoliberalization; private firms are put in the driver's seat whereas state actors are cast as passive passengers. Further, it can be viewed in light of the legacy of the Swedish energy system with its centralized ownership structures. This can also be interpreted as aligned with neoliberal environmental governance since neoliberalism can be argued in practice to revolve around protecting corporate power rather than ensuring free markets (Crouch 2011).
This brings us to the final characteristic of experimentation as a disposition that will be highlighteda particular notion of 'scaling up'. Smart energy experiments were described across the policy landscape as small-scale tests, with the intention of scaling up after lessons had been learned. This notion echoes across policy contexts, seemingly taken for granted, as well as reproduced in academic literatures as diverse as industrial innovation and degrowth (Hellsmark et al. 2016;Kallis 2018). In the empirical material, the notion of achieving transformative change through scaling up experiments was often crystallized in the technology readiness level (TRL) scale. This scale is disseminated across the EU through innovation policy and research funding platforms, instituting arrangements to support technological innovation represented as unfolding in a sequential manner stretching from basic research up an imagined staircase to implementation of new products and servicessignificantly emphasizing commercialization. Across the Swedish energy policy landscape, interviewees regularly referred to technology development in scales or levels, where experiments constitute an important instrument to move a technology along this scale until mature and economically efficient enough to become commercializedoften explicitly referencing Envall the TRL scale. Through TRL, experimentation contributes to casting the problem of environmental governance as a deficit of technological innovation, where projectified sequential development of technologies by private firms or in public-private partnership constellations constitute primary means to develop and export new products and services (cf. Pfotenhauer et al. 2019). This setup is premised on the underlying notion of socio-technical change as achieved through scaling up smallscale tests. It contributes to neoliberal governance arrangements since it depoliticizes experimentation by making it largely into a matter of commercial technology development in projectified form, where large incumbent private companies are placed in prominent positions with their inhouse research and development departments while project funding is tilted in their favor. Consequently, experimentation merges with Sweden's ecomodern heritage, engendering a need to develop ever more new products and services to export for the purposes of greening economic growth, while propelling neoliberalization.

Propelling neoliberalization through experimentation
By disentangling how experimentation as a mode of governance fuses with place-based characteristics of Swedish environmental politics, it can be conceptualized as a situated disposition across the national smart energy policy landscape. Experimentation fuses with the mentioned pre-existing arrangements and discursive elements developed over time in Swedish environmental and energy politics, resulting in a particular situated disposition that seems to propel neoliberalization processes. The experimental setup is tilted in favor of dominant, large corporations in the Swedish context, and through various depoliticizing moves such actors gain increased political influence through experimentation. For instance, as shown the demonstrative character of experiments goes hand in hand with the significant export-driven focus of the increasingly neoliberal Swedish ecomodern ambitions, facilitating marketing ambitions across several levels. Crucially, experimentation becomes a neoliberalization vehicle as the state largely leaves such marketing and export ambitions to private firms; as the reasoning on project funding by interviewees at the government offices highlights, state actors cast themselves as unable to act proactively, clearly favoring market arrangements. Understanding experimentation as a situated disposition has underlined how ambitions of testing tends to be overshadowed by such marketing ambitions, owing to Sweden's ecomodern heritage while providing impetus for neoliberalization.

Experiments as political arenas: refracting political rationality in Gotland and Västerås
Attending to experimentation across the national policy landscape has illuminated that it contributes to sustaining a broader neoliberal political rationality, in line with other studies. Closely attending to unfolding processes of accomplishing governance on Gotland and in Västerås, however, tells a somewhat different story.

The case of Gotland: twists and turns of accomplishing governance
In early summer 2017, the Swedish transmission systems operator announced that a previously planned infrastructure project would be cancelledthe construction of a new transmission line connecting the regional grid on Gotland to the Swedish mainland grid, to alleviate a strained capacity situation on the island. The decision caused strong reactions on Gotland, partly since regional climate and energy policy was predicated on the planned new transmission line, and since new arrangements for the energy system on the island might now have to be implemented urgently. It certainly showcases a neoliberal environmental politics, leaving the island to fend for itself in the shadow of a distant state without additional funding, despite the abandoning of a large, state-funded infrastructure project. Soon afterwards Gotland was designated by the government as a pilot region for a 100% renewable, smart energy system and achieving environmental policy goals before the rest of Sweden, generating lessons and scalable solutions for the rest of the country. The Energy Agency was afforded the central role in this still-ongoing endeavor. While clearly an instance of experimentationwith ambitions of collaborative public-private partnerships testing, learning, and subsequently scaling up technologies and transferring knowledge to the rest of Swedenthe initiative is broad in scope, comprising an ecology of interventions rather than an individual pilot project. Over time various initiatives have been couched under the 'energy pilot' umbrella on Gotland, including projects trying to establish smart solar energy communities, balancing power demand through smart meters, and utilization of hydrogen as energy storage. I followed processes as they unfolded for a few years from the inception of the initiative, where much focus was devoted to establishing networks of actors, with the Energy Agency emphasizing collaboration and coordination. Related to a previous experiment, Smart Grid Gotland, studies have shown how there is deep engagement in energy and environmental issues on the island, both among the broader public and various grassroots actors such as micro-producers of renewable electricity (Wallsten 2017). At the launch of the energy pilot initiative in 2017, hopes of fundamentally changing the energy systemaway from the large-scale, growth-focused configuration of today towards small-scale, decentralized configurationswere expressed by some of these actors, visible both in local news media and in observations at public dialogue forums.
At the outset of the initiative, then, a struggle to define the future energy system ensued. This was enabled precisely by a key characteristic of experimentationthe inherent tentativeness embodied in the ambition of testing. The governmental initiative was from its inception experimental in character, since there were no clear outer boundaries set. Rather, the onus was on uncertain testing. While the initiative was imprinted by the neoliberal political rationality, e.g. by depoliticization through centering commercialization by firms and attempts by the Energy Agency to forge a consensus and quell the contestations flaring up over the configuration of the future energy system, the tentativeness of experimentation opened the door to alternative modes of governing when merged with place-based dynamics on Gotland. Central here were both the island's grid infrastructure, with the increasingly unstable connection between the regional and mainland grids, and the relation between Gotland as a region and the state, as well as the already existing grassroots engagement in environmental and energy issues on the island. Whereas such grassroots actors greeted the uncertainty of the energy pilot initiative with excitement, others expressed worry. Several interviewees at the local grid company and at Vattenfall exclaimed that this new situation might render them losing control over the energy system, painting the picture of a possibly chaotic development. One interviewee stated that "this is a society for god's sake, not a laboratory" (Interviewee I). Similarly, during a presentation on Gotland an employee of the Smart Grid Forum depicted the German energy transition as a disorganized process of muddling through, casting it as a cautionary tale. Instead, the employee called for unity, consensus, and cooperation, for actors engaged in energy transition on the island and beyond to pull in the same direction (Interviewee F).
Such emphasis recurred across interviews. Especially the Energy Agency underlined the necessity of cooperation during several interviews and in documents. Mirroring the increasingly neoliberal character of Swedish environmental governance, an interviewee at the agency laconically ascertained that "private firms have special circumstances, they need to deliver black figures, not red figures" in their accounting (Interviewee J). The interviewee stated that these special circumstances means that private firms cannot funnel infinite amounts of money into research and development, thus making public funding of corporate R&D through experiments vital for energy transition. Combined with the marketing ambitions that favor large corporations, this underlines that experimentation as a situated disposition positions private firms as central actors of governance, while Envall state agencies such as the Energy Agency primarily become facilitators with the objective to entice such firms to commit to experiments. The interviewee firmly underlined the necessity of cooperation, a possibility that experiments conducted in public-private partnership constellations provides, and further that such arrangements are "important to offload risk" for private firms (Interviewee J). The recurrence of this emphasis on consensus further indicates how forging a consensus required labor, to quell the contestations that flared up and threatened to destabilize it (cf. Bulkeley 2016). This was evident at another public dialogue forum, in 2018, where local actors engaged in micro-production of electricity on Gotland voiced that the energy pilot's success hinged upon the Energy Agency being able to create channels of participation for the local population. The initiative had to become meaningful to Gotlanders, it was firmly stated. During this event, it became clear that there were ongoing attempts to mold the energy pilot initiative into an opportunity for Gotland, constituting a concrete mobilization for energy transformation. It was precisely the tentativeness of experimentation that opened for these visions of a different energy future. As the intervention was assembled, actors took up different positions and molded the initiative in their own preferred image of the future, sustainable energy systemthus the tentative character of experimentation not only lends itself to neoliberalization, but in combination with place-based dynamics on Gotland also facilitated the possibility of reconfiguring environmental governance.
Similar processes ensued when the Energy Agency drew the Gotland County Administrative Board into the initiative. An interviewee explained that the Agency viewed the County Administrative Board as a key actor; with their reach and connections on the island they would be vital to ensure that the initiative would become meaningful to Gotland's inhabitants, while mitigating risks of the Energy Agency being perceived as top-down state interventionists (Interviewee J). However the County Administrative Board resourcefully reciprocated the favor, drawing the Energy Agency into their new climate and energy strategy by making them principally responsible for the task of creating a sustainable energy system on the island (Interviewee K). During interviews with representatives of both organizations, as well as observations and informal conversations at various events on the island, it became clear that negotiations related to the roles of the respective actors in the energy pilot initiative were dynamically playing out. This underlines how experimentation, through its tentativeness, opens up for reconfiguring environmental governance.
These examples of situated micro-politics show how the neoliberal political rationality, seemingly very strong in the context of Swedish energy and environmental politics, became refracted on Gotland, in no small part due to the merging of experimentation's tentativeness and placebased dynamics.

The case of Västerås: politicizing energy transition
This section highlights how a smart energy experiment in the city of Västerås entails explicit ambitions of politicizing energy transition, pointing towards drastically reconfigured environmental governance. The experiment is mainly carried out by ETC, a company describing itself as climateactivist, including several branches such as renewable electricity production in solar energy parks. The experiment is located in the vicinity of a new city district, and it revolves around the ambition of wood-constructed affordable housing with self-generated and shared electricity through a local solar-powered DC grid. Various ICT components are necessary for such a setup, including smart data infrastructure to coordinate local electricity production, energy use in the buildings, EV charging, and the regional power grid. Before any construction commenced the national regulatory agency for energy markets clarified that it would be illegal under current regulations to physically transfer electricity between buildingsa privilege reserved for concession rights holders (i.e. licensed grid operators). During empirical work, this regulatory clarification had not yet been made, and afterwards the company immediately planned for alternative approaches. Additionally, the experiment was financed through crowdfunding, and there was an explicit ambition of openly sharing all documentation (Interviewee L).
Already through this setup, it is possible to glimpse how the notion of scaling up is put to work differently in Västerås compared to other experiments. In contrast to the overarching tendencies of smart energy experimentation, the Västerås experiment is resolutely not commercial (Interviewee L). An interviewee at the company explained how upscaling tends to be thought of in commercial terms, but from the interviewee's perspective patents and the world of venture capital "only slows it down" (Interviewee L). The experiment in Västerås aim for grassroots mobilization of citizens, rather than packaging experiences aimed at experts, with the long-term target to change politicswhich could only be done if enough people find energy transformation meaningful and become engaged (Interviewee L). According to the interviewee, testing in experiments and being transparent, rather than aiming for proprietary knowledge generation with the aim of commercialization: "has a practical value […] aimed at these people who are themselves going to try to transition […]" (Interviewee L). The interviewee problematized how other smart energy experiments were directed primarily towards experts, rather stating that upscaling would be about clever municipal companies who could learn from the experiences in Västerås as well as grassroots mobilization to "force a democratic grid" (Interviewee L). The interviewee stated: "smart grids are not, today it's not a way of increasing consumers' power, it's rather a way of protecting the lock-in", connected to a critique of neoliberal environmental politics (Interviewee L). Consequently, the characteristic of experimentation pertaining to socio-technical changeof small-scale tests to be scaled upis articulated very differently in the Västerås experiment. The interlinkage with critiquing neoliberal environmental governance form the backbone of the experiment, illuminating how scaling up was invoked as an essential element of accomplishing governance, to politicize energy transition through grassroots mobilization. This runs counter to how such notions of testing and upscaling through experiments served to depoliticize smart energy across the national policy landscape.
Intertwined with this notion of change, the demonstrative characteristic of experimentation was also expressed differently in Västerås. The interviewee explained that experiments are at their core about expanding imagination, broadening the horizon of what is perceived as possible: "I mean use it to say that another world is possible, another energy world is possible" (Interviewee L). Clearly this pertains to experiments' demonstrative characterbut in this case focus shifts from convincing publics of particular technological innovations or marketing Swedish firms to forge technomanagerial consensus, toward expanding the horizon of imagination. Further, through experiments it becomes possible to challenge incumbent energy companies according to the interviewee: "To frighten them a bit. They get scared, you know" (Interviewee L). This also emanates from the demonstrative character of experimentation; to show that the door is open to different futures through experiments entails a risk for incumbent companies of shifting power relations in the energy sector (cf. Brisbois 2020).
In Västerås, experimentation's tentativeness provides a means to repoliticize environmental politics. Drawing testing as an element into processes of accomplishing governance enables the company to enact a counter-model of smart energy experimentationone which is entirely facilitated by characteristics of experimentation, but which draws otherwise taken for granted assumptions infused by the neoliberal political rationality into the open to challenge them. It is experimentation as a mode of governance that enables these processes of accomplishing governance, politicizing energy transition through evoking alternative future visions and challenging dominant actors.

Conclusions
Zooming in how governance is accomplished across different placesacross the national Swedish smart energy policy landscape and ongoing experimentswe have observed how different relations are assembled. In practice, experimentation contains a plasticity that engenders tensions e.g. between conflicting ambitions of open, uncertain testing, and marketing Sweden and Swedish firms as green technology prodigies. While experimentation has a depoliticizing potential aligned with neoliberal inclinations, experiments also provide arenas for politics, extending far beyond the confines of singular experiments. This is relevant since "the jury is still out" on whether experimentation contributes to further neoliberalization or genuine opportunities for more inclusive governance (Evans et al. 2019).
On the national level, central characteristics of experimentation merged with place-based dynamics of Swedish environmental and energy policy to shape a broader situated disposition, or 'experimentation governmentality'. This disposition, through the fusing of distinct characteristics of experimentationtentativeness, demonstration, upscalingand particularities of Swedish environmental discourse and energy policy contributes to neoliberalization. Such place-based dynamics include a techno-managerial framing of energy issues, strives for consensus, projectification, an energy system with centralized ownership structures, and the ecomodern heritage manifested in the strive to become the world's first fossil-free welfare state. Particularly the ecomodern ambitions, as these fused with experimentation to position large incumbents as primary actors to demonstrate and market their competence, in parallel with emphasizing consensus to quell potential contestations. The previously social democratic ecomodern ambitions of nurturing a strong Swedish environmental technology export industry has increasingly been dressed in neoliberal clothes since the mid-2000s. In smart energy experimentation, Sweden's green frontrunner ambitions are to be realized through voluntary actions, primarily by large, incumbent private firms, encouraged by measures within the experimentalist governance prism, e.g. public funding of corporate R&D and coordination by state agencies to arrive at consensual public-private partnership projects. This move depoliticizes environmental governance, turning it primarily into a series of technical questions.
Further, we saw how the same characteristics harbored by experimentation simultaneously accommodates and enable pathways toward reconfiguring environmental governanceeven within a setting predisposed to aligning experiments with neoliberal principlesas they fuse with placebased dynamics to form localized dispositions in Västerås and on Gotland. By following unfolding processes of accomplishing governance in these experiments, the shaping of particular, unstable localized dispositions became visible. On Gotland, for example, the precarious geography of the islandboth in terms of grid infrastructure and the neoliberal "hands-off" approach from the statefused with the tentativeness of experimentation as local, regional, and national actors attempted to mold the initiative in particular directions, with micro-politics ensuing. Approaching governance as an accomplishment in the making, we saw how experimentation provided opportunities for largely peripheral and marginalized actors on Gotland and in Västerås to channel their voices, whereas the County Administrative Board on Gotland skillfully utilized the tentativeness of experimentation to draw the Energy Agency into regional climate and energy policies. Thus, experimentation also provides an arena for politics to play out, whether in the shape of resistance as in Västerås or as reconfiguration in the case of Gotland, forming localized dispositions with the potential for distinctly non-neoliberal environmental politics.
This highlights that while there is a broader 'experimentation governmentality', or situated disposition on the national level in Sweden which is aligned with neoliberal political rationality, there are embryos of different rationalities practiced within experiments in Gotland and Västerås. Consequently, experimentation opens up for arranging environmental governance differently, even in a deeply ecomodern context geared towards neoliberal environmental governance.
Experimentation has the potential to reconfigure environmental governance as it shifts shape depending on the places it becomes embedded andcruciallypracticed in.
Previous research on environmental politics and experimentation has found that experimentation reconfigures governance, tending to reinforce neoliberal governance and depoliticization (Caprotti and Cowley 2017;Levenda 2019). Through exploring how experimentation arranges governance across different spaces, we have seen how place-based dynamics merges with experimentation in processes of accomplishing governance to shape situated dispositions that can both provide a vehicle for further neoliberalization as well as subversion of such processes. Conceptualizing experimentation as accomplishing governance underlines how environmental governance is always an accomplishment in the making, particularly visible in experimentation's plasticity. As Bulkeley states in an analysis of UK climate governance: "there is no one climate order, no predetermined climate to be governed, rather an ongoing (political) process through which what climate government entails is ordered, contested and reconfigured" (Bulkeley 2016: 99).
Consequently, the main finding of this article is that while the situated characteristics of experimentation feed into dynamics of neoliberalization and ecomodern environmental governance, the same characteristics also enable reconfiguring environmental governance. Thereby there is an inherent tension as experimentation arranges governance in practice. This finding was possible through the approach of following unfolding processes across different spaces of governance. The article contributes to scholarly debates about the implications of experimentation for environmental politics, with deepened knowledge about such implications through unravelling the merging of characteristics of experimentation with place-based dynamics.
Through the analysis some tentative conclusions on what shapes experimentation can be highlighted. The framing of experimentation in line with a broader 'innovation narrative' and its connection to EU innovation platforms appears an excellent fit with ecomodern, depoliticized environmental governance where political issues are redressed as a deficit of innovation (Pfotenhauer et al. 2019). The insistence on inclusion of large firms in experimentation, as illuminated across the policy landscape and on Gotland, further seems to mold experimentation in a manner aligned with neoliberal and ecomodern environmental governance. As we could see on Gotland, however, experimentation also opened up for reconfiguring governance. This is an indication of the importance of local places with their histories and entanglements, where experimentation may crystallize differently. On Gotland such aspects include the precarious grid infrastructure situation, the unique location on an island, and already existing grassroots actors engaged in the energy field when experimentation lands on the island. The core argument is that, somewhat paradoxically, the three characteristics of experimentation as a situated dispositiontentativeness, demonstration, upscalingboth propels neoliberalized ecomodern governance and opens for reconfiguring such arrangements.
However, grassroots actions are never enough to challenge a dominant political rationality, even though different rationalities can be nurtured in such settings. The impact of broader, pre-existing governance structures became apparent in the case of Västerås, where the regulatory agency's interpretation of contemporary concession legislation shut down attempts to test a new sociotechnical configuration. Still, representatives of the company explained how there were already ideas of getting around this obstacle, underlining how the tentative character of experimentation ensures the door remains open to different modes of environmental governance.

Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Energimyndigheten, grant no. 50969-1.