24.1 Introduction

Most scholars will agree that understanding and catalysing sustainability transitions require insights and engagement from across disciplines. Nevertheless, establishing an understanding or working across disciplinary boundaries is seldom easy. Even scholars who work within socio-technical traditions may quarrel over questions such as ‘when innovation is environmental’ or ‘what constitutes societal transition’ (Stirling, Reference Stirling2011: 82). One of the reasons for this is that different theoretical perspectives and research approaches are anchored in different ontologies, that is, ‘foundational assumptions about the nature of the (social) world, and its causal relationships’ (Geels, Reference Geels2010: 496). Further, scholars operate within different epistemologies, namely different ways of understanding ‘what constitutes valuable knowledge’ (Hazard et al., Reference Hazard, Cerf, Lamine, Magda and Steyaert2020). Spelled out differently, this means that how we understand reality, combined with our methods, theories and evaluations of normative questions, feeds into the production of knowledge, resulting in distinctly different styles of research that are not always easily compared, combined or integrated.

Within sustainability transitions research, such dynamics have been exacerbated by the research community growing rapidly over the last decade. With this growth, the focus of transition scholars has expanded. From its beginnings, sustainability transitions research studied patterns of technological change and stability, with the underlying question being how new and more sustainable technologies could replace existing, environmentally harmful ones (e.g. Kemp, Reference Kemp1994). This was done through conceptualising sectors like energy supply, water supply or transportation as socio-technical systems, alongside an understanding of socio-technical transitions as ‘a set of processes that lead to a fundamental shift in socio-technical systems’ (Markard, Raven and Truffer, Reference Markard, Raven and Truffer2012: 956). An example is the study of the processes that advance or impair a shift from fossil fuel energy systems to renewable energy systems. Today, such system-internal inquiry is accompanied by research that probes sustainability transitions but from different angles. Examples include studies of the justice and ethical implications of transitions, non-technological transitions (e.g. through sufficiency) or research on the interaction between multiple sectors and systems as a consequence of transitions (e.g. Köhler et al., Reference Köhler, Geels, Kern, Markard, Onsongo, Wieczorek and Wells2019; Khmara and Kronenberg, Reference Khmara and Kronenberg2020; Ryghaug and Skjølsvold, Reference Ryghaug and Skjølsvold2021; Solbu et al., Reference Solbu, Ryghaug, Skjølsvold, Heidenreich and Næss2024; Andersen et al., Reference Andersen, Geels, Coenen, Hanson, Korsnes, Linnerud and Wiebe2023). Hence, transition studies have experienced an expansion in what we can call their epistemological and ontological focus: what they recognise as important knowledge and what the key building blocks of that knowledge are.

Through exploring ontological and epistemological debates, this chapter seeks to advance reflections about some of the different ways that we can know sustainability transitions and about the relationships between these ways of knowing. The ambition is on the one hand to facilitate navigation of an increasingly complex field, but perhaps also to help make sense of some of the confusions and frustrations that may emerge as part of working in an interdisciplinary reality. I will argue that the epistemological and ontological diversity that is currently emerging within sustainability transitions research is a strength that should be embraced, rather than pursuing a strategy of consolidating a uniform style of research where disciplinary boundaries are tightly policed and competing understandings of what matters pushed out.

24.2 Epistemological and Ontological Debates in Sustainability Transitions Research

Ontological and epistemological debates within sustainability transitions research have typically emerged around what we can call the dominant frameworks in the tradition. Geels (Reference Geels2022) notes that a ‘first wave’ of transition studies emerged during the 1990s and early 2000s, and that this wave served to establish frameworks such as the multi-level perspective (MLP), strategic niche management (SNM) and technological innovation systems (TIS). With the addition of transition management (TM), these are what have become the four dominant frameworks within early sustainability transitions research (Markard, Raven and Truffer, Reference Markard, Raven and Truffer2012; Zolfagharian et al., Reference Zolfagharian, Walrave, Raven and Romme2019). For much of the research during this early phase, the MLP served either as a basis for the research conducted or as a reference point. Today, the MLP arguably serves as what sociologists of innovation have called an obligatory passage point (see, e.g. Callon, Reference Callon1986) within the transition community, that is, a prism that shapes debates, and that scholars often relate to whether they use the perspective or not (Zolfagharian et al., Reference Zolfagharian, Walrave, Raven and Romme2019). Given its centrality in the field, and because SNM, TM and the MLP all circle around the basic concepts of niche, regime and landscape, the MLP will serve as the starting point and basis for the following discussion.

24.3 Beyond Studies of Transitions as Innovation and Technology Journeys

In the discussions raised over the last pages, I have provided some pointers to ontological and epistemological debates within first-wave transition studies. Here, transitions were seen through processes where niches bump against regimes. With research questions along such lines, the MLP and associated frameworks provide a basic ‘plot for the study of transitions’ (Geels, Reference Geels2011; Sorrell, Reference Sorrell2018). Stirling (Reference Stirling2019) notes that there has been a tendency within this tradition of reifying concepts such as ‘incumbency’ or ‘regimes’, that is, treating these ideas as if they are solid material entities that exist in the world. This tendency, he notes, might result in blind spots, leading the community of transitions researchers to overlook crucial phenomena that lie beyond what concepts such as the ‘regime’ may explain. Through relying too much on a reified standard narrative, transition studies may end up in a paradoxical position, where they see possibilities for change within discrete socio-technical systems while rendering the rest of society and its power dynamics stable, or, as Haugland (Reference Haugland2022) noted, promoting ‘innovation for preservation’. Perhaps as a reaction to the reification of concepts and a basic transition plotline, sustainability transitions research has become populated with scholars that contest these ideas. Reviewing papers from the last decade published within the journal Environmental innovations and Societal Transitions, Truffer et al. (Reference Truffer, Rohracher, Kivimaa, Raven, Alkemade, Carvalho and Feola2022: 335) note that ‘Gender and different forms of power and conflict constellations are receiving increasing attention and, alongside with that, critical perspectives have gained increasing salience in the field’. In the following, I will discuss three of the many epistemological strands of research that coexist with the first-wave traditions.

While first-wave transition studies circled around technology, new work challenges the epistemic primacy of this focus, calling for epistemic and normative engagement with more fundamental traits of contemporary societies. Critique along these lines notes that the basic plotline of the MLP has made sustainability transitions scholars ‘blind’ to phenomena such as capitalism (Feola, Reference Feola2020). Feola argues that sustainability transitions research tends to treat capitalism as a landscape phenomenon, and therefore neglects to analyse it. On the one hand, this externalisation is problematic because there are a variety of capitalisms across different territories that might have different bearings on transitions (see also Loewen, Reference Loewen2022). Further, Feoloa argues that the blindness to capitalism makes it difficult to assess to what degree the transitions studied by sustainability transitions scholars are actually sustainable, because with ‘endless accumulation of capital and economic growth, gains in efficiency are likely to be accompanied by rebound effects’ (Feoloa, Reference Feola2020: 243). This line of critique, then, challenges both how we evaluate transitions (e.g. the diffusion rate of electric vehicles vs. the evaluation of social, economic and environmental consequences of a strategy that replaces combustion engines with batteries) and how we understand how transitions come about. It also comes with its own plotline, which emphasises that the central activity in social scientific inquiry is to expose how a capitalist mode of organisation leads to negative environmental, social and economic effects, as opposed to a plot that emphasises niche innovation.

This critique of sustainability transitions research is echoed in work that foregrounds the role of coloniality both in upholding unsustainable societies, and in influencing and shaping the ways that research understands transitions (e.g. Hopkins et al., Reference Hopkins, Kester, Meelen and Schwanen2020; Arora and Stirling, Reference Arora and Stirling2023). Such work, often done under the banner of de-colonialism, emphasises how coloniality as a deep power structure, together with other intersecting structures (e.g. patriarchy), constitutes broad developments such as capitalism and modernity. This critique carries with it the notion that core theories of sustainability transitions research have been developed within the context of western societies, and that they come with distinct packages of understandings that explain, for example the way firms and individuals behave. These packages of generalised knowledge, then, carry with them values that are not neutral or objective but which reflect historical priorities made in those societies. Arora and Stirling (Reference Arora and Stirling2023) discuss such dynamics with respect to the regime concept in the MLP, concluding that ‘the preoccupation with defining dominant rules can mean that plural and marginal voices are side-lined not only in the ‘objects’ of research, but also in subjective processes of analysis’ (11). Hence, this line of epistemological critique emphasises that the challenge is not only to expand on the themes and topics that sustainability transitions research engages with (e.g. by shifting from the study of regimes to the study of capitalism) but also to cultivate a more reflexive style of research that does not merely reflect dominant interests (see also Jasanoff, Reference Jasanoff2018). As Hopkins et al. (Reference Hopkins, Kester, Meelen and Schwanen2020) notes, this is particularly important when conducting research in non-western contexts.

Together with exposing and challenging power relations, de-colonial analysis tends to promote ethical evaluations rooted in concerns for justice in transitions. Around sustainability transitions, several strands of research have emerged, which provide a different plotline than that of the MLP, rooted in ideas of justice. Built on the social justice theories of John Rawls, the notion of energy justice proposes a set of principles or tenets by which one can assess the justice implications of transition activities. First is the notion of distributional justice, which concerns the distribution of burdens and benefits of transitions within and across societies, generations, social groups and so on. Second is the procedural justice, which points towards a focus on the ways that decisions are made in transition processes. Third is recognitional justice, which points towards a focus on who are recognised as legitimate stakeholders in such processes (see, e.g. Jenkins et al., Reference Jenkins, McCauley, Heffron, Stephan and Rehner2016). In terms of expansion, then, this style of research proposes a shift from asking how one can diffuse niche technology to asking who benefits, how the transition can be organised and who are recognised as legitimate participants. While there have been conceptual efforts to internalise energy justice concerns to the MLP (Jenkins, Sovacool and McCauley, Reference Jenkins, Sovacool and McCauley2018), crossovers between core transition theories and justice-oriented scholarship remain scarce. As Sareen and Haarstad (Reference Sareen and Haarstad2018) note, the explanation might be found in the fact that exploring such normative issues does not fit the basic ontological and epistemological underpinnings, or in other words, the main plotline of the MLP. This, Sareen and Haarstad note, brings with it a ‘silencing effect’ for concerns that reside outside that basic plot of activities across niche, regime and landscape.

24.4 Concluding Discussion

This chapter has done three things. First, it has recapped the epistemological and ontological building blocks of first-wave transition studies. Second, it has illustrated some of the ways that this way of knowing transitions has been challenged. Third, it has pointed to epistemological approaches that probe transitions through asking different types of questions and highlighting knowledge on aspects beyond innovation as central for understanding transitions. First, the discussion illustrates a basic aspect of all knowledge production. Knowledge is always produced from within an epistemological and ontological position, which serves to highlight certain aspects of reality over others or to describe reality in different ways. Second, the discussion shows that the way scholars engage with sustainability transitions is reflective of empirical developments. The epistemological and ontological expansions within the field follow an empirical development where sustainability transitions have shifted from being a marginal phenomenon to being one that increasingly affects all aspects of society (see, e.g. Markard, Reference Markard2018). This brings with it more adverse consequences and the need to have more and different types of academic conversations about transitions.

The positions discussed in this chapter are far from exhaustive when it comes to potential ontological and epistemological positions. As an example, the contributions discussed in this chapter all share a rejection of ‘the core assumptions of positivism’ (Zolfagharian et al., Reference Zolfagharian, Walrave, Raven and Romme2019: 1277), which is one of the key ontological positions in science broadly (e.g. Hazard et al., Reference Hazard, Cerf, Lamine, Magda and Steyaert2020). Positivism ‘assumes an independent objective world, with deterministic relations between variables’ (McDowall and Geels, Reference McDowall and Geels2017: 46). Analysis within such a perspective tends to mobilise statistical data or formal models in analyses. I will not dive into positivism here, but simply note that it is important for researchers within the transitions community to know about it, especially for researchers involved in interdisciplinary and collaborative work, because such perspectives can also have a silencing effect. Geels et al. (Reference Geels, Berkhout and Van Vuuren2016) note that in integrated assessment models, a central tool in climate science, positivism is assumed, also when assessing societal processes. Such a view, the authors note, means that most social scientific inquiry will not fit, and therefore cannot be easily integrated into such models. In turn, this means that social elements are typically silenced within such frameworks.

Hopkins et al. (Reference Hopkins, Kester, Meelen and Schwanen2020) argue that there should be room for what they call radical theoretical pluralism within sustainability transitions research. This is basically an argument for embracing plotlines beyond those promoted by the core transition frameworks. This, they argue, has benefits and risks for sustainability transitions research as a research field. First, radical theoretical pluralism might result in what they dub a conceptual Esperanto, reducing shared worldviews and problematisations, as well as a loss in the sense of community among sustainability transitions researchers. Externally, it could reduce the recognisability of sustainability transitions research as a field. On the other hand, it can open up new, diverse insights as well as greater reflexivity and debate about black-boxed assumptions. Externally, this might foster the entry of new forms of scholars into transition studies, and the broadening of transition studies repertoire of policy-relevant insights.

I am sympathetic to the plea for epistemological and ontological diversity in sustainability transitions research. The challenges contemporary societies face are a complex and interrelated mix of social, technological and environmental. They cannot be reduced to narratives about introducing new technologies. Understanding, and acting on these issues in a meaningful way, requires engaging with knowledge that transgresses innovation journey plotlines. This can probably not be achieved through the integration of an ever-expanding set of issues in core theories that were designed primarily to grasp innovation journeys. From my perspective, current dynamics pose a strong argument for the study and conceptualisation of transitions from a broad range of disciplines and an open approach to epistemology and ontology. This, however, is not an argument for a retreat to disciplinary boundary lines, because there is a strong need for interaction between scholars and to understand the epistemological and ontological differences that produce very different analyses of related phenomena. Achieving this will make our field more open to the diverse ways that societies transition, what such transitions might entail and the deep societal consequences of such transitions.

25.1 Introduction: Transitions Thinking – A Systemic Approach to Sustainability

Transitions research has gained traction in sustainability studies for its holistic, systemic, policy-operational approach to environmental challenges. A central tenet is that the persistence of these challenges can be attributed to ‘system failures’ (Rotmans, Reference Rotmans2005; Turnheim Reference Turnheim and Sovacool2020; Turnheim & Sovacool, 2020). Accordingly, system innovation is held to be needed, that is, a multitude of innovations that co-evolve into system-wide transformations (Chapter 2). So, transitions are transitions of systems. They are processes of change from one defined system state to another (Chapter 2). Recurring questions therefore arise concerning: Transition of which system? To where? For instance, is transitioning the ‘mobility system’ a matter of electrification or of unlearning car dependency? In transitioning the ‘energy system’, is importation of rare minerals a central part or an externality?

This chapter provides an overview of discussions around the ‘which system, to where’ issues. We approach it as a methodological challenge: When studying transitions, one needs to define and conceptualise the system under study (ontology), to think through how it can become known to us (epistemology; cf. Chapter 24), to determine what matters most in any given context (normativity), to specify system demarcations (analytical focus) and to choose methods for the empirical analysis of the systems under study (data gathering).

Taking a reflexive methodology approach (Alvesson & Sköldberg, Reference Alvesson and Sköldberg2017), this chapter addresses ‘systems’ as representations. Common notions of ‘energy systems’, ‘mobility systems’ or ‘innovation systems’ – and related more institutional ideas around socio-technical ‘regimes’ and ‘incumbent actors’ – are all representations of relevant aspects of transitions issues. So it is not self-evident what might count as the most ‘systematic system representation’ (Holtz et al., Reference Holtz, Alkemade, de Haan, Köhler and Trutnevyte2015). It is through a variety of corresponding ‘mental models’ of systems that individuals address transitions challenges (van Den Broek et al., Reference Van den Broek, Negro and Hekkert2024), and it is by developing sophisticated but understandable system models (MLP, transition management, TIS and many others) that transitions thinking has become attractive for policymakers and other practitioners (Voß, Reference Voß2014).

Our reflexive, constructivist approach serves two purposes. First, it provides a canvas for our methodological overview of systems analysis. Second, it brings out how transitions research has developed a fine sense of Critical Systems Thinking (CST) (Ulrich, Reference Ulrich2003; Midgley & Rajagopalan, Reference Midgley and Rajagopalan2020). CST underlines that system change revolves around the negotiation of different system understandings – and especially around the implicit assumptions underlying these system understandings. CST opens up dialogue on these assumptions through so-called boundary questions. Transforming ‘systems’: Which? How? Whose? Why? Whither? Whence?

Accordingly, the chapter will bring out how transitions research has engaged with such critical systems thinking through a range of methodological advances. On the other hand, we will also show the need for further methodological reflection. Many contrasting views of ‘systems’ and ‘systemic frameworks’ are seemingly always clamouring for attention, whilst it is not always clear what they’re referring to. Importantly, this ambiguity is sometimes strategic: Casual references to ‘systems’ often conveniently sidestep uncertainty and disagreement – don’t we all aim for sustainable mobility, energy and food ‘systems’? The chapter therefore highlights how system analysis has brought transitions research not only a common language for interdisciplinary collaboration but also a space within which to convene contending interests and perspectives. As such, systems thinking raises some particularly controversial methodological issues: Identification of particular ‘systems’ and ‘systemic problems’ informs proposals for particular systemic interventions.

The chapter continues with an overview of systems thinking in transitions research, sketching historical development and identifying the main research themes (Section 25.2). After providing some empirical illustrations (Section 25.3), we delve deeper into ongoing debates (Section 25.4) and reflect on recently emerging and future research (Section 25.5).

25.2 Historical and Thematic Development: ‘Systems’ in Transitions Research

Transitions thinking has made its mark in sustainability science/governance through its holistic, systemic policy-operational approach (Chapters 1 and 2). As it emerged around the turn of the millennium, its key argument was that persistent sustainability problems were of a deep-rooted, systemic nature – and that systemic ‘solution’ strategies should accordingly be devised. Rotmans (Reference Rotmans2005) underlined that sustainability governance should move beyond the ideological discussions of market and government failures and address systemic failures instead. Smith et al. (Reference Smith, Voß and Grin2010) have clarified how the associated proposals for system innovation marked a shift in sustainable innovation research. Compared to the still prevalent focus on sustainable technologies and innovation in products and processes, transitions thinking presented an attractive allure of comprehensiveness.

A key concept in this systems thinking is the socio-technical ‘regime’. Throughout transitions research, these are taken as more institutionally informed views of the ‘systems’ to be pushed towards sustainable development paths. Especially in early transitions research, much effort was expended in seeking to specify system components: ANT, SCOT and the history of technology have been invoked to clarify what socio-technical systems are made of. Regarding the system boundaries, transitions thinking is informed by insights on organisational fields, large technical systems and innovation systems. These concepts help to understand how socio-technical networks can solidify into more or less coherent systems. Regarding the system dynamics, transitions thinking draws on the history of technology and innovation studies to understand how social and technological system components cluster, form webs of relations and interdependencies, and solidify eventually in path-dependent and possibly locked-in (Unruh, Reference Unruh2000) systems. Combined with insights from sociology and especially institutional theory (Fuenfschilling & Truffer, Reference Fuenfschilling and Truffer2014), this makes for quite elaborate insights into system reproduction.

Beyond these ideas about system reproduction and systemic inertia, transitions research stands out for its exploration of dynamics of system change. Here, a key concept is the transitions S-curve, indicating shifting rates of change and innovation diffusion. Elaborating these system understandings further through insights from Complex Adaptive Systems, evolutionary economics, sociology, institutional theory and governance studies, transitions research has theorised the dynamic instability of the socio-technical ‘regimes’, the co-evolution between external pressures from system environments (e.g. ‘landscape’ pressures and emerging ‘niches’), internal dynamics (like ‘endogenous renewal’ of regimes) and system transformation ‘pathways’ that can be discerned across empirical contexts. Taken together, these insights have yielded a basic grammar of system transitions. Broadly shared by an interdisciplinary transitions research community and a wide range of practitioners as well, this systems vocabulary might be visualised as follows:

Figure 25.1 schematises some key features in this shared systems language. It shows how this system transformation vocabulary allows diverse actors to work together despite contrasting notions of system change (in multiple dimensions and modalities collapsed onto the horizontal axis) and of system aggregation (multiple dimensions and levels of aggregation collapsed onto the vertical axis). Frameworks such as the MLP, TIS and SNM have proposed more specific ideas about such features of systems ontologies. In particular, transition management (Rotmans, Reference Rotmans2005; Loorbach, Reference Loorbach2007) has been important in conceptualising transitions in terms of complex adaptive system dynamics (Grin et al., Reference Grin, Rotmans and Schot2010). Functioning implicitly as ‘boundary objects’ (Voß, Reference Voß2014), the elements represented in Figure 25.1 offer only a rough consensus. Throughout the past twenty-five years of transitions research, questions have arisen on all these elements:

• What are the key system dynamics and how to underpin accounts of these? This is a longstanding question in transitions research, and notably in transition modelling. Through its disciplinary roots in integrated sustainability assessment and Complex Adaptive Systems research (Peter & Swilling, Reference Peter and Swilling2014), transitions research has always relied strongly on systems modelling approaches (Holtz et al., Reference Holtz, Alkemade, de Haan, Köhler and Trutnevyte2015). Intensively researched issues of system dynamics are the co-evolution patterns underlying transition pathways (Chapter 2), the dynamics of niche formation (Chapter 5) or the barriers and drivers in innovation systems (Chapter 4). After this ‘first wave’ (Chapter 24) of research on the transition dynamics themselves, the underpinning of the theorised system dynamics has been debated more in recent years (Sorrell, Reference Sorrell2018). A prominent example is Geels (Reference Geels2022), discussing different kinds of complex causal mechanisms.

• How to transform the system? This line of questioning tends to accompany the aforementioned system dynamics questions (Grin et al., Reference Grin, Rotmans and Schot2010). This coupling of analysis and intervention is usual in systems thinking, operations research and management. The ‘how to transform’ issues have of course been considered extensively in transitions governance research (Chapter 3). Particularly noteworthy however are the specifically systems-analytical ideas on ‘tipping points’ and ‘intervention points’, in the tradition of Meadows’ (Reference Meadows1997) ‘leverage points’ in complex systems.

• What is the system environment? In transitions theory, the system in focus is by default a ‘regime’, in interaction with various external ‘niche’ and ‘landscape’ processes. By contrast, analyses in terms of TIS or social practices zoom in on configurationally smaller systems (with socio-technical ‘regimes’ as the system’s environment). For its part, ‘deep transitions’ research (Chapter 7) rather zooms out. More generally, there is an intensifying reflection on the socio-material, political-economic, cultural, ideological and social-ecological system environments through which transition processes are shaped.

• What are the system’s boundaries? Queries about system environments logically imply questions on the boundaries that separate systems from their environment. System demarcation is in fact a prominent line of questioning in its own right: Is it to be regarded as a ‘system innovation’, or is it just system optimization that innovates at the level of a subsystem? Such issues have been addressed in terms of ‘subsystems’ (Pel & Boons, Reference Pel and Boons2010), ‘whole system reconfiguration’ (McMeekin et al., Reference McMeekin, Geels and Hodson2019) and ‘multi-system interactions’ (Andersen & Geels, Reference Andersen and Geels2023).

• What are the systems composed of? Figure 25.1 only sketches the general assumptions of heterogeneous, socio-technical ‘regimes’. Basic questions on system components tend to be taken for granted, however. When they are addressed, this typically occurs through a dialogue between sector-specific and more general transitions-theoretical insights. For example, Gaitán-Cremaschi et al. (Reference Gaitán-Cremaschi, Klerkx, Duncan, Trienekens, Huenchuleo, Dogliotti and Rossing2019) distinguish ‘agricultural production system’, ‘value chain’ and ‘support structures’ to specify the components of food systems. Holtz et al. (Reference Holtz, Brugnach and Pahl-Wostl2008) presented a more fundamental consideration of ‘regime’ components, and Svensson & Nikoleris (Reference Svensson and Nikoleris2018) initiated further discussion: How to develop causal explanations if we conceive of transitions as changes in rules and institutions? The relevance of material system components has been argued in the account of social-metabolic transitions by Fischer-Kowalski (Reference Fischer-Kowalski2011), a view on system components that is particularly relevant in the context of ‘circular economy’ transitions.

• What is system reproduction? Questions on system components may appear rather formal, analytical matters. However, as Svensson & Nikoleris (Reference Svensson and Nikoleris2018) stressed, questions about system constitution are crucial if we want to distinguish system reproduction from system transition. Even if seldom undertaken through explicit systems thinking, questions on system reproduction form the core of research on transitions governance, politics and institutionalisation dynamics (Fuenfschilling & Truffer, Reference Fuenfschilling and Truffer2014). Highlighting the paradoxes, grey zones and ‘niche–regime dynamics’ (Chapter 9) of transition processes, these analyses contain urgent questions about what changes and what stays the same.

This overview summarises intensive explorations of the systems to be transformed. Yet beyond the above ‘what’ and ‘how’ questions, a further line of queries emerges. As underlined in second-order cybernetics, representations in the spirit of Figure 25.1 provide ‘first-order’ systems pictures: Meant to show a categorically objectified system ‘out there’, they downplay how such representations are shaped and coloured by the context of their respective system observers. (Krippendorff, Reference Krippendorff1996: 311) considered this difference essential: ‘To me, the shift from a first-order to a second-order cybernetics signaled a shift in scientific attitude toward reality, from privileging the perspectives of detached observers, spectators or engineers of a world outside of themselves to acknowledging our own participation in the world we observe and construct as its constituents’.

Figure 25.1

Transforming systems, a common systems vocabulary in transitions research

The systems that transitions thinking is dealing with are composed of people, Krippendorf reminds us. Yet many forms of system understanding are rather distant, objectifying people and speaking in the abstract of ‘them’. The conditions bearing on the kinds of representations people choose typically remain under-attended to. To remedy this, system analyses might more systematically disclose how different observers (and their different contexts of observation) tend to yield contrasting representations and framings of the same systems (Krippendorff, Reference Krippendorff1996: 314). System representations might thereby become more ‘plural and conditional’ (Stirling, Reference Stirling2008) – paying more balanced attention not just to a plurality of quite disparate system representations, but also to the particular political, cultural or institutional conditions that each embodies. This central importance of recognising how systems are ‘framed’ forms a key part of the ‘pathways approach’ to sustainability transformations (cf: figure 3.1 in Leach et al., Reference Leach, Stirling and Scoones2010).

As emphasised in participatory approaches to transition modelling (Köhler et al., Reference Köhler, Turnheim and Hodson2020), system analysis can take the form of a dialogue between transitions researchers and practitioners about the (mobility, energy, food or other) systems in which they are implicated. But the point here is not simply to aggregate into some supposedly transcendent ‘consensus view’ but to systematically elicit how contrasting equally valid political perspectives yield different system ontologies, epistemologies and normativities. Transitions analysis – no matter how authoritative it may seem – should never be treated as if it could stand definitively on its own (Saltelli et al., Reference Saltelli and Di Fiore2023), or as if it were exempt from thorough democratic deliberation (Stirling, Reference Stirling2024).

In keeping with this broader and deeper ‘second order’ systems view, Figure 25.2 displays three different framings of Figure 25.1. Different actors associated with diverse subjectivities across the ‘system context’ can make different assumptions about system components, system reproduction, transformation dynamics and system boundaries. This may lead them to widely different framings of transitioning systems – hence the visualisation in different colours. Likewise, as indicated through the arrows that define axes for each system view, they may analyse the systems along different epistemic, ontological and normative dimensions. As displayed at the bottom, this multiplicity of system framings raises a range of ‘boundary questions’ – the queries used in Critical Systems Thinking to disclose the different assumptions underlying system understandings (Ulrich, Reference Ulrich2003):

• Which systems? The shift from first-order (Figure 25.1) to second-order (Figure 25.2) systems thinking implies a move away from questions about ‘what the system components, boundaries, dynamics are’. Rather than categorically assuming objective systems ‘out there’, it raises relational ‘which systems’ questions that acknowledge the existence of multiple valid representations. This way of speaking about systems is not uncommon in transitions research. It is one way to avoid the ‘reification trap’ (Ison, Reference Ison2016) of mistaking system representations for reality itself. An MLP analysis, for instance, can identify the main features of mobility systems, but its results do not show us ‘the mobility system’. ‘Which systems’ questions help to avoid ‘voyeuristic’, managerial (Walker & Shove, Reference Walker and Shove2007:219) systems thinking that are overconfident regarding system interventions.

• Whose system representations count? Whose interests are reflected in system representations? The notion of ‘system failures’ is suggestive of objective states of affairs and agreed-upon problem. The same applies to common references to ‘the energy regime’ and ‘the incumbent actors’. This objectivism has been challenged particularly extensively in Smith & Stirling (Reference Smith and Stirling2010). Applying critical systems thinking to an empirical study on traffic management, Pel & Boons (Reference Pel and Boons2010) showed how transitions governance revolves around the re-definition of systems. This is in fact a key element of transition management (Loorbach, Reference Loorbach2007): Transition arenas and visioning activities are meant to instigate ‘second-loop learning’ and dialogue on systemic problems.

• Why is system transformation needed? These ‘why’ questions tend especially to open up normative dimensions of transitioning systems. These normative dimensions often remain implicit, or hidden under global consensus on ‘sustainability’, ‘system lock-in’ and ‘institutional inertia’. ‘Why’ questions have been asked in various early critical interventions (Walker & Shove, Reference Walker and Shove2007; Bening et al., Reference Bening, Blum and Schmidt2015; Schlaile et al., Reference Schlaile, Urmetzer, Blok, Andersen, Timmermans, Mueller and Pyka2017), but these normative aspects of systems thinking are becoming particularly central issues in analyses of ‘just’ transitions. Especially relevant readings are ‘justice, equity, inclusion, and fair transitions’ (Chapter 16) and studies of the ‘dark sides’ of transitions (Pel et al., 2023).

• How to intervene? Observed from where? As mentioned, Figure 25.2 highlights the position of the system observer. Jørgensen (Reference Jørgensen2012) pointed out similarly how societal actors can have widely divergent interpretations of the MLP. So: From which vantage point do we know and engage with the system? To help address this, Smith & Stirling (Reference Smith and Stirling2007) elaborated a distinction between ‘inside’ and ‘outside’ perspectives. Quite literal approaches to the ‘observing from where’ question were offered by Stirling (Reference Stirling2019) and Sovacool et al. (Reference Sovacool, Hess, Amir, Geels, Hirsh, Medina and Yearley2020), who argued for ‘worm eye’ modes of systems analysis. Similar arguments against transcendent, ‘ideal-observer’, ‘cockpit’ and ‘bird’s eye’ perspectives have been made in calls for ‘humility’, ‘situated analysis’, reflexive research (Chapter 27) and in (often participatory) transitions research that accounts for positionality.

• Whence the system history? There is agreement that transitions are processes of change that take place over time. Figure 25.1 expresses this temporal aspect through the notions of system reproduction and system dynamics. Guided by ideas about transition processes between present and future system states, the transitions-theoretical systems thinking revolves around evolutionary mechanisms, such as niche development, selection pressures, emergence and co-evolution. Many studies work with a rather pragmatically chosen t = 0, but there are questions to ask about the start of system evolution: The ‘deep transitions’ framework (Chapter 7) proposes to start from ‘industrial modernity’, for example, but there are also more fundamental moves towards a plural understanding of system histories: Key examples are the analyses of temporality (Garud & Gehman, Reference Garud and Gehman2012) and engagements with coloniality (Ghosh et al., Reference Ghosh, Ramos-Mejía, Machado, Yuana and Schiller2021). Importantly, different histories lead to different system understandings (Balanzó-Guzmán & Ramos-Mejía, Reference Balanzó-Guzmán and Ramos-Mejía2023).

• Whither the system direction? There are many ways of knowing systems: as a basis for future predictions or as a diagnosis of past transformations. Where the former tends to disclose possibilities and initiatives towards sustainability transitions, the latter tends to elicit lock-in, debt and injustice. Informed by historical case studies, transitions thinking has developed a distinctly prospective mode of systems thinking. A key element of transition management is indeed to evoke reflection on ‘whither the system direction’, through activities of problem structuring and visioning in transition arenas (Loorbach, Reference Loorbach2007). Scholarship on transitions directionality has underlined that this reflection on possible and desirable futures is indeed crucial for the transitions-theoretical systems thinking. It has long been observed that actors in system transitions can influence the direction for change by successfully articulating expectations (Brown & Michael, Reference Brown and Michael2002). Importantly, transition scholars’ own representations also raise expectations (Stirling, Reference Stirling2019). The ‘whither’ question is thus fundamental. It should not be reduced to a matter of ‘directing the transition towards the right direction’ (Stirling, Reference Stirling2024): Transitions research deals with systems that are not only complex as objects but also as performative representations (they create expectations).

Figure 25.2

Critical systems thinking: boundary questions in systems thinking

This overview already reveals how systems thinking forms a quite diverse methodological family. It gives rise to debates and controversies. Before going into these in Section 25.4, we briefly discuss empirical applications.

25.3 Empirical Application: System Transitions across Contexts

Transitions research has developed a systems language that helps to understand transformative change across contexts. It helps to compare ‘niche–regime’ dynamics (Chapter 9) or ‘strategic niche management’ (Chapter 5) across empirical domains. Featuring volumes on transitions in mobility, agriculture, energy and health systems, the KSI book series starting from Grin et al. (Reference Grin, Rotmans and Schot2010) illustrates this very well. Inversely, context-transcending models such as the MLP, TIS, SNM and transition management have been developed and validated through the comparison and formalisation of various historical case studies. Empirical application of the systems thinking has proven to be easy: This speaks to the proliferating multiplicity of adapted MLP or TIS diagrams, or schemes of transition management phases, that have been crafted to express case specifics.

Transitioning-systems language has been applied to a very wide range of empirical contexts: Beyond the many studies on functional systems of energy, mobility, agriculture and health, other remarkable examples are practices of burying and cremation, rock ‘n’ roll, animal welfare or earth-space engineering. Kanger (Reference Kanger2020) stresses how the study of such ‘neglected systems’ – even if less interesting in terms of environmental impacts – is likely to support theoretical advances. Also, the ‘geographical turn’ in transitions research (Chapters 21–23) has questioned how widely our systems language can be applied: Does it make sense to conceive of transformations in national systems, as if they were isolated processes? Is the idea of coherent, identifiable ‘regimes’ applicable beyond the (northwestern-) European context?

Throughout empirical domains and across the different strands of transitions research, we see many applications of critical systems thinking. Researchers and practitioners illuminate in many ways how the conceptualisation and demarcation of ‘systems’ tends to be neither obvious nor innocent: The ‘mobility system’, with its unsustainable fossil fuel dependency, is often represented as a sub-system of the energy system. Meanwhile, the mobility system has also been characterised by the regime of car dependence. Importantly, the latter system understanding refers to other mechanisms of system reproduction and other key system components than the former. In turn, these different ‘mobility system’ understandings lead to quite different evaluations of system innovation and system reproduction: Electric vehicles can be considered as radical ‘niches’ challenging the dominant design of the internal combustion motor, but they can also be considered as ‘techno fixes’ that essentially reproduce the car-dependency system. Meanwhile, it is not even obvious to characterise energy transition as a matter only of decarbonisation. Again, this underscores how greater reflection on the diversity of possible attributed system purposes – whether in terms of ‘energy democracy’ (Stirling, Reference Stirling2014) or wider ‘innovation democracy’ – supports accountable practices concerning system representations.

Studying transitions processes through systems analysis directs attention to very basic aspects of empirical situations. A most instructive application is the comparative analysis by Papachristos et al. (Reference Papachristos, Sofianos and Adamides2013). Starting from the common assumption that transitions are often triggered by particular ‘outsiders’, they revisited several of the well-known empirical cases that underpinned typologies of transition pathways. Considering concrete examples of ‘outsiders’ (e.g. medical doctors or the taxi sector), they explored a series of fundamental theoretical issues: The outsiders, in which sense are they external to the system under study? Should they perhaps be rather considered as parts of niches, spawned by the internal tensions of the ‘regime’, and as such as parts of the system under study? And if they are indeed to be considered as external to the system under study, are they then to be considered as niches challenging other, surrounding ‘regimes’?

Considering further that radical niches can come from within and from without, and that ‘regime’ shifts are partly resulting from the dynamics of other, surrounding systems – is it then not logical that we see empirical examples of newly emerging systems, such as the functional foods ‘niche’, or the computer regime (ibidem.: 62)? And these newly emerging systems, how can we distinguish them from their ‘parent systems’? Rigorously examining empirical cases for their systemic properties, this study laid important foundations for later explorations of ‘multi-system interactions’ (Andersen & Geels, Reference Andersen and Geels2023). Importantly, it also provided a good demonstration of CST: The analysis of ‘multi-system interactions’ presupposes a series of assumptions about insiders and outsiders, about system boundaries, about exogenous and endogenous ‘drivers’ of change, and about the supposed ‘functions’ of systems and technologies. An instructive example from Papachristos et al. (Reference Papachristos, Sofianos and Adamides2013) was the ‘functional foods’: Should they be analysed as derivative of foods or of pharmaceutical products, as a hybrid between these two, or as something entirely different altogether?

25.4 Ongoing Debates: Adequacy, Politics and Plural Perspectives

Regarding ongoing debates, a first observation is perhaps that many accounts of ‘system transitions’ lack specific reference to any concrete, bounded, empirical system. A certain common understanding has developed around socio-technical ‘regimes’, their components and the co-evolution dynamics of system transitions, but the systems vocabulary is seldom operationalised in detail. This is reflected in a tendency towards relatively casual references to ‘outsiders’ or ‘endogenous change’ (Section 25.3).

This relative absence of systematic debate is noteworthy, but we have also seen a quite thorough exploration of systems and system understandings (Section 25.2). The engagements with the boundary questions have opened up scientific debates and indeed controversies. As expressed through the Figures 25.1 and 25.2, one can raise different kinds of questions about ‘systems’ in transitions. The questions we raise reflect fundamental ontological and epistemological assumptions, as well as contrasting patterns of normative interest in systems: Systems can be taken as analytical constructs for modelling, as coding devices for case comparison, as representations of governance issues, or as resources for policy advocacy. As Krippendorff (Reference Krippendorff1996) stressed, first-order and second-order systems thinking really raise different questions and perspectives. It makes a difference whether one takes an ‘eagle eye’ or a ‘worm eye’ perspective.

Overall, we can see three main debates about the systems-in-transitions thinking. These debates are connected in terms of theoretical and practical implications, yet scholars tend to engage in them out of different knowledge interests. The debates unfold to a certain extent in parallel.

1. (1) Ontological adequacy. In first-order systems thinking, ‘what’ and ‘how’ questions predominate. This involves intensive debates about the social-technical-ecological constitution of the systems to transform (‘how about economic sectors, socio-economic classes, geographic regions, or geo-morphological layers?’), and about the underpinnings of theorized transition dynamics (‘how useful is the focus on rules/institutions for transition analysis, what about material impacts?’). These debates have yielded a rich systems-theoretical body of knowledge on units and levels of transitions analysis, systems-of-systems, whole-systems analysis, subsystem innovation and multi-system interactions. These debates are driven mostly by scholars engaged in theory development, and by the transitions modelling community. While the discussions on transitions ontologies involve an increasingly diversified transitions research field, the modelling community plays a key role in the formalisation of the ‘systems’ under study.

2. (2) Normative-political implications. The ‘which’, ‘why’ and ‘whose’ systems questions have gained attention, especially in research on transitions governance and politics. A recurring issue is whether prevailing system representations are giving appropriate, balanced and fair accounts of systemic problems and solutions. Are the representations of systems challenging or reproducing dominant actor constellations, interests and discourses? Whose system definitions count? In the politics-focused approaches and in ‘just’ transitions scholarship, a central issue is who gets to define the system. This marks to a certain extent the shift from Figure 25.1 (representing the singular system ‘out there’) to Figure 25.2 (representing plurality of system understandings). On the other hand, some of these debates stay within first-order systems thinking. This is the case for example when dominant system representations are dismissed as ‘false’ system representations.

3. (3) Plural perspectives. Finally, there is the dispersed but ongoing engagement with the ‘whence’, ‘whither’ and ‘observed from where’ lines of questioning. These are the typical second-order systems thinking questions of Figure 25.2, starting from the diversity of system understandings. A key issue is the position of the actor defining the system, and related to this, the many subtle ways in which system definitions introduce assumptions of expertise, overview and control. The system’s language itself carries a certain technocratic temptation, a mindset of overview and control (Stirling, Reference Stirling2019). An increasingly prominent question is therefore (see also Chapter 23 on the Global South and North): How are system representations articulating or suppressing epistemic diversity? The debates on epistemic diversity involve engagements with entirely different approaches to sustainable development, such as political ecology. This deepens transitions research through questions such as the following: Is the ‘systems’ notion at all compatible with the handling of complex transformation processes, and with the heterogeneity of social-ecological networks? Should we perhaps understand transitions in terms of networks, arenas, processes, flows, webs or fields? It is worth noting how this line of questioning links back into the debates on ontological adequacy.

25.5 Emerging Research and Further Needs

Systems thinking is fundamental for transitions research – the cross-references to other chapters have underlined this. The vocabulary of transitioning systems has helped to develop our highly interdisciplinary transitions research community. The notion of ‘the system’ is easily taken for granted, however. This chapter has shown how casual references to ‘outsiders’, ‘endogenous change’ and ‘system transformations’ call for greater clarity about systems and system representations. Invoking Critical Systems Thinking, we have also shown how some transitions research has developed quite rich debates around various ‘boundary questions’ about the ‘systems’ in transition (Section 25.2). This systems thinking tradition provides useful methodological resources for the investigation and comparison of diverse transition cases (Section 25.3).

Regarding the emergence of promising lines of research and further needs, it is important to recognise the notable advances that are already being made. There is some shared critical awareness of the ‘reification trap’, that is, of assuming system understandings to be obvious and beyond discussion. First-order systems thinking still prevails as a norm for its capacity to make complexity tractable, but it is generally accompanied with at least some second-order systems thinking awareness. System models and theories tend to be presented cautiously as ‘frameworks’ or ‘perspectives’, for example, and there have been many efforts to articulate the variety of system understandings. We have indicated various newly emerging themes, but the debates over ontological adequacy, normative-political implications and epistemic diversity also indicate an already longstanding tradition of systems thinking (Section 25.4).

In more operational terms, it is noteworthy to see how researchers in transition modelling and scenario development are developing iterative, participatory approaches (Köhler et al., Reference Köhler, Turnheim and Hodson2020). This shows in a nutshell how reflection on first-order and second-order systems thinking can be applied and valorised. Also encouraging is the emergence of methodologies for addressing epistemic diversity, such as multi-criteria analysis, ‘quantitative storytelling’ (Kovacic et al., Reference Kovacic, Strand and Völker2019) or critical innovation studies (Godin & Vinck, Reference Godin and Vinck2017). Finally, there is also transitions research that is clearly in the second-order systems spirit of Krippendorff (Reference Krippendorff1996), highlighting how appraisal itself forms part of the transitioning systems being studied. This includes studies on the performativity of the ‘transition management’ governance model (Voß, Reference Voß2014) and its susceptibility to inadvertent influence by incumbent interests (Stirling, Reference Stirling2019). Such second-order systems thinking approaches remind us of the fundamental democratisation issues that transitions research itself is raising – through the circulation of system representations (Stirling, Reference Stirling2024).

Regarding ‘further needs’, it must be stressed in critical systems thinking mode that no ‘advance’ can be considered definitive, self-evident or unambiguous. Figures 25.1 and 25.2 indicate an important epistemological move towards second-order systems thinking. The awareness of this inescapable subjectivity is a fundamental advance in systems thinking. Still, these first and second orders should not be mistaken for hierarchical orders: as shown, they indicate different modes of asking questions about systems. The three debates distinguished in Section 25.4 should similarly not be taken as progressive ‘stages of reflection’. This would neglect how the various ‘what’, ‘which’ and ‘whose’ questions evoke each other. As already elaborated under the banners of ‘double-loop learning’ (Argyris, Reference Argyris1977), Soft Systems Methodology (Checkland & Scholes, Reference Checkland and Scholes1999) and indeed Critical Systems Thinking (Ulrich, Reference Ulrich2003), any of these questions can be the start of a dialogue about the ‘systems’ to be transitioned. This chapter has provided concepts, themes and search terms that support this.

26.1 Introduction

Sustainability transitions research focuses on understanding and fostering fundamental change processes in our human systems and making our societies more sustainable. It is urgent that these system transitions need to increase in scope, scale and speed, as highlighted by the IPCC’s 2018 Special Report on limiting warming to 1.5°C (de Coninck et al., Reference De Coninck, Revi, Babiker, Bertoldi, Buckeridge, Cartwright, Dong, Ford, Fuss, Hourcade, Ley, Mechler, Newman, Revokatova, Schultz, Steg, Sugiyama, Araos, Bakker and Bazaz2018). Interventions in our human systems intended to bring about the necessary transitions have to come from a systems-based view that accounts for interactions within and across systems (Andersen and Geels, Reference Andersen and Geels2023; Köhler et al., Reference Köhler, Geels, Kern, Markard, Wieczorek, Alkemade, Avelino, Bergek, Boons, Fünfschilling, Hess, Holtz, Hyysalo, Jenkins, Kivimaa, Martiskainen, McMeekin, Mühlemeier, Nykvist, Onsongo, Pel, Raven, Rohracher, Sandén, Schot, Sovacool, Turnheim, Welch and Wells2019; Papachristos et al., Reference Papachristos, Sofianos and Adamides2013). While the sustainability transitions field has developed its own theoretical frameworks and related approaches, such as the multi-level perspective (MLP, see Chapter 2) and technological innovation systems (TIS, see Chapter 4), it has also benefited significantly from insights drawn from complexity science and systems thinking. This chapter examines how complexity science and systems thinking approaches can enhance our understanding of sustainability transitions. These approaches are particularly valuable because they provide frameworks and tools for the analysis of interactions, feedback loops and emergent behaviours that characterise transition processes.

This chapter provides a cursory review of how the sustainability transitions literature builds on other systems-oriented approaches, specifically complex systems approaches and system dynamics. It outlines their link to the main theoretical sustainability transitions approaches and describes the extent to which insights from different systems approaches have been applied and have informed transition research. The chapter first explores how systems thinking has informed our conceptualisation of socio-technical systems and innovation systems. We then examine how complexity science and system dynamics approaches help us understand transition dynamics, particularly through concepts like path dependence, feedback loops and emergent behaviour. Finally, we discuss and illustrate through specific case studies the practical application of these approaches in transition research and policymaking. For early career researchers, this chapter provides conceptual tools for transitions analysis through systems thinking. By understanding systems approaches, researchers can better grasp how different elements of transitions interact and how these interactions influence transition processes.

The unit of analysis in transitions research is the socio-technical system and the way it undergoes fundamental change towards a more sustainable direction. Transitions literature builds on insights from systems science, science and technology studies (Bijker, Reference Bijker1997; Bijker et al., Reference Bijker, Hughes and Pinch2012) and evolutionary economics (Nelson and Winter, Reference Nelson and Winter1982), as well as frameworks such as Large Technical Systems (Hughes, Reference Hughes1993). In systems science (Boulding, Reference Boulding1956; Forrester, Reference Forrester1968; Von Bertalanffy, Reference Von Bertalanffy1968), a system is conceptualised as consisting of diverse and interacting elements that adhere to a set of rules and form part of system mechanisms that serve a particular function. The interdependence of system elements leads to understanding socio-technical systems as complex adaptive systems (Miller and Page, Reference Miller and Page2007), where system behaviour emerges out of system elements interactions and the operation of several mechanisms.

In line with this systems-based view, a socio-technical system is defined as the actors, technologies and infrastructures that provide key societal functions like energy or mobility provision as well as the formal and informal institutions that govern their interaction (Geels, Reference Geels2002). A defining characteristic of socio-technical systems is the high degree of interdependence between their components and groups of actors, which develop and evolve over time through multiple feedback processes (Geels, Reference Geels2005; Smith et al., Reference Smith, Voss and Grin2010). The strong interdependence of all system elements requires a high level of coordination and alignment between them to enable the system to function. This has two implications. First, the tight coordination and alignment between elements make the system resistant to change. Second, a high level of interdependence between actors, technologies, infrastructures and institutions makes understanding systems difficult by looking at the behaviour of individual elements, for example, actors or technologies. The implication is that system change is difficult even when change is desirable and urgent. This is a notion that transition research takes into account in theory development and case analysis, as discussed in Section 26.2.

26.2 Researching System Interactions and Transitions

Understanding how socio-technical systems change requires examining the complex interactions and feedback mechanisms and understanding how they can either drive or inhibit transitions (Geels, Reference Geels2022; Papachristos, Reference Papachristos2014; Reference Papachristos2018; Reference Papachristos and Adamides2019; Sorrell, Reference Sorrell2018). This is the aim of two established frameworks for transitions research discussed in the following paragraphs. This section then focuses on how different types of interactions can shape transition dynamics and non-linear system behaviour.

26.3 Examples of Application of Complexity and System Dynamics Approaches

26.4 Conclusions and Future Research Directions

While insights from complexity science on path dependence and uncertainty have been embraced by historical studies of transitions, this is much less so for future-oriented studies. Especially TIS approaches often operate with an implicit assumption that the system under study is desirable and should be stimulated. The approaches outlined in this chapter offer well-tested methods to incorporate some of the complexity arising from interactions between systems in our analyses. The systems approaches discussed in this chapter create opportunities for better connection with other communities investigating transition pathways. For example, in climate science, where modelling approaches are much more common, these methods can help bridge different analytical traditions. This integration is particularly valuable as transitions research increasingly engages with urgent climate challenges. It also makes it easier to connect to other communities that investigate transition pathways such as climate science where modelling approaches are much more common.

A crucial methodological challenge in transitions research is connecting models that operate on very different timescales and aggregation levels in a meaningful way. Current approaches often struggle to integrate analyses of short-term dynamics (such as policy implementation or market responses) with longer-term evolutionary processes (like infrastructure development or institutional change). This challenge becomes particularly pertinent in studies of structural coupling, systems coupling, tipping dynamics and phase-out processes, where changes at different scales interact in complex ways. For example, while system dynamics models can capture feedback loops within specific subsystems, they often struggle to incorporate broader institutional changes that emerge from MLP analyses. Similarly, agent-based models that excel at representing individual actor behaviours may not adequately capture slower-moving landscape developments.

To address these challenges, promising methodological developments are emerging at the intersection of different modelling approaches. Hybrid modelling frameworks that combine qualitative case studies with quantitative simulation models offer one pathway forward. Another promising direction is the development of multi-scale modelling architectures that can maintain consistency across different levels of analysis while allowing for appropriate methodological approaches at each level. These developments could help bridge the gap between detailed technological transition pathways and broader socio-technical system change, particularly in areas like energy system transformation where changes in infrastructure, behaviour and institutions need to be analysed simultaneously. Such methodological advances would be particularly valuable for studying how different transition processes might interact and potentially accelerate or hinder each other, an understanding that is crucial for steering sustainability transitions.

The development of more sophisticated methodological approaches to study sustainability transitions through a systems lens is not merely an academic exercise – it is essential for understanding and steering the complex societal changes that are necessary to address our urgent sustainability challenges. The combination of insights from complexity science, system dynamics and transitions research and the development of new tools that bridge different analytical scales and approaches, will place us in a better position to understand and influence the profound systems changes that sustainability transitions require.

At a time of unprecedented climate pressures, rising inequality and increasingly militant demands for societal transformation, academic researchers grapple with the question of engagement. Should they, as detached observers, strictly adhere to the adage of presenting the scientific facts and leave acting upon them to others? Or does their deeper understanding of the issues bestow upon them a greater obligation to actively bring about change? Such one-sided views on scientific engagement are rooted in an outdated conception of knowledge as a neutral mirror of reality. In this chapter, we argue that, whether you are a theoretical physicist, archaeologist or transition researcher, you are always inherently engaged. As a producer of knowledge about how the world works, you inevitably – and often implicitly – address the question of what the world should look like (Jasanoff, Reference Jasanoff2004). The issue, therefore, is not whether you engage as a researcher, but how you shape that engagement. Addressing this issue asks for reflexivity – an introspective process in which the researcher turns their own engagement into a subject of research, reflection and discussion.

In this chapter, we explore the concept of reflexivity specifically in sustainability transitions research (STR) and aim to make it tangible for individual researchers. While reflexivity is relevant across all scientific disciplines, it is particularly critical in STR because the field explicitly aligns itself with a normative vision of how the world should be – namely ‘sustainable’. Given that sustainability is a widely embraced guiding principle, at least in our Western colonial context, there is a risk that unexamined engagement may too easily reinforce established power relations. In other words, the escape route of so-called neutrality is simply not an option for transition researchers.

The chapter unfolds as follows: In Section 27.1, we introduce the concept of reflexivity through the lens of Science and Technology Studies (STS), emphasising the social embeddedness and inherent non-neutrality of science and technology. Section 27.2 elucidates the importance of reflexivity for STR, particularly in relation to both its normative and descriptive categories. In Section 27.3, we focus on concrete instances of STR where mobilising reflexivity makes a difference in opening creative pathways of transformative research towards sustainability. This exploration culminates in the conclusion that reflexivity presents transition researchers with an inextricable and inescapable responsibility.

27.1 Reflexivity and the Non-neutrality of Technology and Knowledge

Technological innovation and knowledge production are socially embedded practices. It is always specific individuals who, within particular social, cultural and material contexts, develop new technology and knowledge, which in turn further shape these contexts. In STS, it has been convincingly demonstrated how this contextual embeddedness co-shapes research interests, methods and concepts. In other words, knowledge and technology are not neutral.

In this context, reflexivity is commonly cited as a means of addressing this non-neutrality. Reflexivity involves becoming aware of the embeddedness of the technology and knowledge that we, as human beings, develop. However, questions remain as to the extent to which one can become fully aware of their own embeddedness, what form this ‘knowledge about one’s own knowledge’ can take, and whether it can actually remedy the non-neutrality of technology and knowledge.

But what do we mean when we assert that technology and knowledge are (or are not) neutral? While STS offers many nuanced perspectives on these questions, this contribution focuses on three ideal-typical responses. In addition to and beyond the illusory claim that technology and knowledge are neutral, we distinguish two fundamentally different ways of understanding non-neutrality: a first-order and a second-order claim (Goeminne, Reference Goeminne2012). The first-order claim views non-neutrality as the result of – though in principle avoidable – influences from social, cultural and material contexts on technology and knowledge. This claim characterises non-neutrality negatively, as it represents a deviation from an ideal of neutrality. In contrast, the second-order claim acknowledges embeddedness affirmatively, seeing it as a necessary condition for, and even constitutive of, the production of meaningful knowledge and usable technology.

In what follows, we link these three ideal-typical responses to corresponding approaches for addressing (non-)neutrality in the application of technology and knowledge. We draw inspiration from Stirling’s (Reference Stirling, Voss, Bauknecht and Kemp2006) incremental distinction between ‘unreflectiveness’ (neutrality), ‘reflectiveness’ (first-order non-neutrality) and ‘reflexivity’ (second-order non-neutrality). This ideal-typical framework will guide our exploration of how transition researchers, both individually and collectively, can address this non-neutrality as conscious producers of transformative knowledge for shaping sustainability transitions.

27.2 Why Reflexivity is Key for Transition Researchers

Building on the preceding discussion regarding the non-neutrality of technology and knowledge, this section argues why STR should turn its gaze inward to the non-neutrality of its own research practices. As the STR community approaches technology and knowledge from an STS perspective – emphasising their co-evolution with society – numerous scholars have engaged with the concept of reflexivity (Beers & Van Mierlo, Reference Beers and Van Mierlo2017; Chilvers & Longhurst, Reference Chilvers and Longhurst2016; Smith & Stirling, Reference Smith and Stirling2007). However, the primary focus has been on reflexivity in the governance of transitions, particularly regarding the unintended consequences of transition interventions and strategies to anticipate and address them.

Rather than advocating for a governance approach aimed at optimising transitions by addressing as many societal implications as possible, this chapter concentrates on reflexivity in STR as an inherently engaged practice.Footnote ² As mentioned in the introduction, reflexivity is of particular significance to STR. Although the non-neutrality of sustainability as a normative guiding principle may seem self-evident, how this principle is given shape and direction within the situated practice of STR is far from straightforward.

Following Beck et al. (Reference Beck, Jasanoff, Stirling and Polzin2021), who recently argued that STR needs to step ‘back from the assumption that descriptive and normative categories in which people make sense of transitions and sustainability are somehow equally self-evident to everybody involved’ (p. 144), this section examines STR’s normative (e.g. sustainable, transformative, deep, radical) and descriptive (e.g. transition, system, regime, practice) categories. Although discussed separately later, these categories are intertwined and overlap. As situated concepts, both categories encompass inextricable epistemological, ontological and ethical dimensions. In combination with the third section of this chapter, the objective is to elucidate what reflexivity might entail for a transition researcher as an engaged individual.

27.3 Mobilising Reflexivity in Sustainability Transitions Research

The preceding section illuminated the non-neutrality of STR in terms of its normative and descriptive categories. As a result, transition researchers bear the responsibility to shape their engagement in a reflexive way, thereby probing the myriad ways in which their research, explicitly as well as implicitly, touches upon the question of what the world should look like. This section explores what this call for reflexivityFootnote ³ may imply by posing three questions: Where am I going (i.e. research orientation)? How can I act (i.e. research roles)? And where do I come from (i.e. positionality)? These questions are addressed through a few fictionalised experiences.

27.4 Conclusion

In this chapter, we explored the question of engagement within STR through the lens of reflexivity. In the first section, we introduced reflexivity from an STS perspective on the situatedness of technology and knowledge. Distinguishing between unreflectiveness, reflectiveness and reflexivity, we advocated for an incrementally introspective attitude towards the non-neutrality of technology and knowledge. In the second section, we outlined why reflexivity is important to STR. We demonstrated that reflexivity extends beyond acknowledging the non-neutrality concerning sustainability and the direction of transitions. It also encompasses the more implicit non-neutrality inherent in the concepts and frameworks STR mobilises. In the third and final section, we addressed how transition researchers can mobilise reflexivity. Through some fictionalised experiences, we illustrated how a reflexive attitude continually prompts researchers to confront new and often challenging questions about their own research practice and how this, in turn, invites creative, critical and relevant interventions.

We close by stating that an ethos of reflexivity confronts transition researchers with an inescapable responsibility. This responsibility complicates the research process, making it far from straightforward, but it also makes it considerably more challenging and, at times, existential. There is an essential portion of reflexivity that is indispensably personal, in that it cannot be outsourced or delegated. Reflexivity is not a subroutine of a research procedure that can be addressed with a toolbox, principles, a separate reflexivity workshop or work package. Rather, reflexivity is the systematic and cultivated wonder about the social, societal and political status of one’s own research practice, and, as such, it accompanies every move you make. This is something everyone has to do, by themselves, very personally. It is, after all, you, raising or not raising objections, seeing or not seeing the bigger picture, caring or not caring to understand what is at stake in what you are (or are not) doing. Much of reflexivity, therefore, lies in the question of how, and with what kind of awareness, you are personally doing what you are doing. If you are a transition researcher, you are in a comparatively powerful societal position so that your choices do make a difference in the world.