The multi-level perspective on sustainability transitions: Responses to seven criticisms

TL;DR: The multi-level perspective (MLP) has emerged as a fruitful middle-range framework for analysing socio-technical transitions to sustainability as discussed by the authors. But the MLP also received constructive criticisms.

Abstract: The multi-level perspective (MLP) has emerged as a fruitful middle-range framework for analysing socio-technical transitions to sustainability. The MLP also received constructive criticisms. This paper summarises seven criticisms, formulates responses to them, and translates these into suggestions for future research. The criticisms relate to: (1) lack of agency, (2) operationalization of regimes, (3) bias towards bottom-up change models, (4) epistemology and explanatory style, (5) methodology, (6) socio-technical landscape as residual category, and (7) flat ontologies versus hierarchical levels.

Summary

Introduction

• Contents lists available at ScienceDirect Environmental Innovation and Societal Transitions journa l homepage: www.e lsev ier .com/ locate /e is t Survey.
• (1) lack of agency, (2) operationalization of regimes, (3) bias towards bottom-up change models, (4) epistemology and explanatory style, (5) methodology, (6) socio-technical landscape as residual category, and (7) flat ontologies versus hierarchical levels, also known as The criticisms relate to.

1. Transitions to sustainability

• Contemporary environmental problems, such as climate change, loss of biodiversity, and resource depletion (clean water, oil, forests, fish stocks, etc.) present formidable societal challenges.
• Transitions towards sustainability have some special characteristics that make them different, in certain respects, from many (though not all) historical transitions.
• Their involvement might accelerate the breakthrough of environmental innovations if they support these innovations with their complementary assets and resources.
• With regard to structural change the problem is that many existing systems are stabilized through various lock-in mechanisms, such as scale economies, sunk investments in machines, infrastructures and competencies.
• Section “The multi-level perspective on socio-technical transitions” briefly describes the MLP.

2. The multi-level perspective on socio-technical transitions

• The multi-level perspective (MLP) is a middle-range theory that conceptualizes overall dynamic patterns in socio-technical transitions.
• The MLP views transitions as non-linear processes that results from the interplay of developments at three analytical levels: niches (the locus for radical innovations), socio-technical regimes (the locus of established practices and associated rules that stabilize existing systems), and an exogenous sociotechnical landscape (Rip and Kemp, 1998; Geels, 2002, 2005a).
• While early work characterized the MLP as a nested hierarchy, Section “Flat ontologies versus hierarchical levels” proposes to drop that qualification.
• MRT has the following characteristics (Geels, 2007): (a) MRT are not about broad, abstract entities such as ‘society’ or ‘social system’, but about concrete phenomena (such as socio-technical transitions), (b) MRT differs from grand theory, because it emphasizes interactions between theory and empirical research.
• It is only when such concepts are interrelated in the form of a scheme that a theory begins to emerge” (Merton, 1968:143).

2.1. Socio-technical regime

• The socio-technical regime forms the ‘deep structure’ that accounts for the stability of an existing socio-technical system (Geels, 2004).
• Following Giddens (1984), these regime rules are both medium and outcome of action (“duality of structure”).
• Because existing regimes are characterized by lock-in, innovation occurs incrementally, with small adjustments accumulating into stable trajectories.
• These trajectories occur not only in technology, but also in cultural, political, scientific, market and industrial dimensions.
• The socio-technical regime concept aims to capture the meta-coordination between different sub-regimes (Geels, 2004).

2.2. Niches

• Niches are ‘protected spaces’ such as R&D laboratories, subsidised demonstration projects, or small market niches where users have special demands and are willing to support emerging innovations.
• This is not easy, however, because the existing regime is stabilized by many lock-in mechanisms and because niche-innovations may have a mis-match with existing regime dimensions (e.g. lack of appropriate infrastructure, regulations or consumer practices).
• Niches are crucial for transitions, because they provide the seeds for systemic change.
• The articulation (and adjustment) of expectations or visions, which provide guidance to the innovation activities, and aim to attract attention and funding from external actors.
• The building of social networks and the enrolment of more actors, which expand the resource base of niche-innovations.

2.3. Socio-technical landscape

• The sociotechnical landscape is the wider context, which influences niche and regime dynamics (Rip and Kemp, 1998).
• The landscape level, which has similarities to the concept of longue durée proposed by the historian Braudel, highlights not only the technical and material backdrop that sustains society, but also includes demographical trends, political ideologies, societal values, and macro-economic patterns.
• The general dynamic pattern is characterized by transitions resulting from the interaction between processes at different levels: (a) niche-innovations build up internal momentum, (b) changes at the landscape level create pressure on the regime, and (c) destabilisation of the regime creates windows of opportunity for nicheinnovations.
• Each of these phases can be linked to particular mechanisms (Geels, 2005a).
• An important implication is that the MLP does away with simple causality in transitions.

3. Criticisms, responses, and suggestions for future research

• The structure of the subsequent sections is that I describe the criticism, provide a response or qualification, and try to make productive suggestions for future research.
• Not all of the criticisms can be entirely resolved, because some of them relate to differences in assumptions or academic styles.
• In those instances, I aim to clarify the choices and styles that are related to the MLP.

3.1. Lack of agency

• The MLP has been criticized for underplaying the role of agency in transitions.
• Given their interest in the governance of socio-technical transitions, they especially argue for more attention for the role of power and politics.
• Thus, social orders, institutional orders and artifacts are both the medium and outcome of human endeavors”.
• While detailed case studies show some of these types of agency, the MLP can be theoretically enriched by mobilizing insights from other theories.
• The literature on strategic alliances can offer relevant insights in collaborations between incumbents and new entrants in the development of niche-innovations (Rothaermel, 4 The suggestion by Genus and Coles to incorporate SCOT and CTA is somewhat strange, since these approaches were an input to the MLP from the start.

3.2. Operationalization and specification of regimes

• Several criticisms concern the operationalization and specification of regimes.
• The MLP does not prescribe how broad or narrow the empirical topic should be delineated.
• So ‘regime’ is an interpretive analytical concept that invites the analyst to investigate what lies underneath the activities of actors who reproduce system elements.
• That is probably true, especially for studies that focus on niche-innovations and how they struggle against existing regimes.
• A criticism with regard to transition processes is that most work seems to focus on a single regime, which faces pressures from niche-innovations and landscape developments.

3.3. Bias towards bottom-up change models

• The MLP has been criticized for a bias towards bottom-up change models.
• Competitive niche-innovations are well developed when landscape developments exert pressure on regimes.
• (2b) The ‘structural opportunity’ model suggests that broader secular changes create favorable conditions for revolutionary uprisings.
• This profile has similarities to the transformation pathway in which the regime adjusts to external pressures.

3.4. Heuristics, epistemology and explanatory style

• With regard to epistemological status Genus and Coles (2008) suggest that the “potential contribution of the MLP/transitions framework could be limited to offering a heuristic device” (p. 1442).
• The frameworks, however, seek to help the analyst to better think through the problem” (Porter, 1991:98).
• There is scope for scholars to conceptualize dynamic mechanisms with auxiliary theories.
• Process-oriented scholars have therefore turned towards narrative explanation (Abell, 2004; Pentland, 1999) because narratives can capture complex interactions 8 I include this representation to illustrate the logic of event chains and conjunctures, not to give an in-depth historical account of the Industrial Revolution.
• ‘Knock-on effects’ and ‘innovation cascades’ mean that new technologies may be adopted in a system for certain reasons.

3.5. Methodology

• With regard to the historical case studies that illustrate the MLP, Genus and Coles (2008) criticise the “flawed use of secondary data sources”, especially the use of “uncritically ‘accepted’ accounts of the historical significance of certain socio-technical developments” (p. 1441).
• But the criticism remains under-specified, because Genus and Coles do not indicate which empirical mistakes were made in particular case studies.
• Because process theories tend to emphasise complex dynamics such as path dependence, interaction effects, tipping points, thresholds, bifurcations, and conjunctures, they may be at odds with assumptions required for standard regression techniques and conventional comparative methods.
• Process theories, in contrast, tend to have a more complex ontology, but less developed methodology.
• This general dilemma also affects the MLP, with most empirical research using case study methodologies.

3.6. Socio-technical landscape as residual category

• The landscape level has been criticized for being a residual analytical category, a kind of ‘garbage can’ concept that accounts for many kinds of contextual influences.
• This is a fair criticism, which can be made productive by reformulating it as a need for more theorization.
• A subsequent puzzle then becomes how to determine the relative influence of stabilizing and destabilizing landscape developments.
• Along similar lines, Berkhout et al. (2009:225) suggest that macroeconomic development can be studied as the outcome of “the emergence of new socio-technical systems, replacing or radically altering traditional and early modern systems in key sectors, including energy, transport, agriculture and food, water and urban development”.

3.7. Flat ontologies versus hierarchical levels

• From a relationist ontology, which assumes a ‘flat’ world, the idea of ‘levels’ can be criticized.
• This is again in contrast to the more hierarchical aspect of the multi-level perspective” (p. 472).
• To explain transitions, practice theory distinguishes between: (a) more “enduring and relatively stable practices” (p. 475), which are routinely reproduced and characterized by predictable trajectories, and (b) new practices, which are more fluid and unstable.
• This is a deviation from earlier MLP-work, which used the notion of ‘nested hierarchy’.
• While this is an attractive metaphor, most niches do not emerge within regimes, but often outside them (although niche actors are usually aware of regime structures).

4. Concluding comments

• Socio-technical transitions (to sustainability) are a special research topic, because they are about relatively rare, long-term macro-changes.
• Other types of theories and methodologies are therefore needed.
• In response to criticisms, this paper has tried to clarify the choices and styles that underlie the multilevel perspective on socio-technical transitions.
• Still, I have tried to argue that the MLP is well suited to address the special characteristics of the transitions topic.
• Disagreements may lead to critical debates which are the lifeblood of academic practice and may lead to refinement of theories and frameworks.

Figures

Fig. 3. Conjunctural sequence explanation of English industrialization (Mahoney, 2000:534).

Fig. 2. Multi-level perspective on transitions.

Fig. 1. Alignment of ongoing processes in a socio-technical regime.

Table 1 Three layers of power (Grin et al., 2010:283).

Frequently asked questions

Q1. What have the authors contributed in "The multi-level perspective on sustainability transitions: responses to seven criticisms" ?

This paper summarises seven criticisms, formulates responses to them, and translates these into suggestions for future research. This paper is about one particular approach, namely the multi-level perspective ( MLP ). While there are various approaches to transitions, with different strengths and weaknesses, the remainder of this paper focuses on the MLP. This paper aims to respond to these criticisms, qualify them, and, where possible, translate them into productive suggestions for future research. The paper is organized as follows. The niche and landscape levels can be seen as ‘ derived concepts ’, because they are defined in relation to the regime, namely as practices or technologies that deviate substantially from 1 Merton ( 1968 ) introduced the notion of middle range theory ( MRT ) to navigate between the extremes of grand theory ( such as Parson ’ s structural functionalism ) and abstracted empiricism, which focuses only on data-collection and -analysis. MRT has the following characteristics ( Geels, 2007 ): ( a ) MRT are not about broad, abstract entities such as ‘ society ’ or ‘ social system ’, but about concrete phenomena ( such as socio-technical transitions ), ( b ) MRT differs from grand theory, because it emphasizes interactions between theory and empirical research. Section “ Criticisms, responses, and suggestions for future research ” addresses seven types of criticisms. 

Q2. What future works have the authors mentioned in the paper "The multi-level perspective on sustainability transitions: responses to seven criticisms" ?

The structure of the subsequent sections is that I describe the criticism, provide a response or qualification, and try to make productive suggestions for future research. Not all of the criticisms can be entirely resolved, because some of them relate to differences in assumptions or academic styles. 

Q3. What are some examples from my work?

Some examples from my work are hype-disappointment cycles, niche-accumulation pattern, fit-stretch pattern, knock-on effects and innovation cascades, add-on and hybridization pattern (Geels, 2005b). 

Q4. What methods could be used to improve the transition research?

transition research could probably benefit from the application of other methods such as comparative or nested case studies, event-sequence analysis, network analysis, even-history methods, and agent-based modelling. 

Q5. Why do process theories tend to emphasize complex dynamics?

Because process theories tend to emphasise complex dynamics such as path dependence, interaction effects, tipping points, thresholds, bifurcations, and conjunctures, they may be at odds with assumptions required for standard regression techniques and conventional comparative methods. 

Q6. What are the trade-offs with regard to the ontology?

But because the ontology is (even more) complex than the MLP, there may be trade-offs with regard to accuracy, generalization, and empirical operationalization. 

Q7. What is the criticism of the landscape level?

The landscape level has been criticized for being a residual analytical category, a kind of ‘garbage can’ concept that accounts for many kinds of contextual influences. 

Q8. What is the importance of ideology and cultural framing in making people aware of their conditions?

Whereas the traditional colcanic model emphasized objective conditions of hardship, Goldstone’s fourth generation of revolutionary theory (Goldstone, 2001) highlights theimportance of ideology and cultural framing in making people become aware of their conditions and in constructing feelings that existing socio-economic or political conditions are unacceptable. 

Q9. What is the criticism of transition processes?

A criticism with regard to transition processes is that most work seems to focus on a single regime, which faces pressures from niche-innovations and landscape developments. 

Q10. What do Genus and Coles (2008) suggest the MLP should incorporate?

Genus and Coles (2008) repeat this point and suggest that the MLP should incorporate constructivist approaches, such as social construction of technology (SCOT), actornetwork theory (ANT) and constructive technology assessment (CTA), in order to “show concern for actors and alternative representations that could otherwise remain silent” (p. 1441). 

Q11. What is the definition of the multi-level perspective?

The multi-level perspective (MLP) is a middle-range theory that conceptualizes overall dynamic patterns in socio-technical transitions. 

Q12. What is the idea central to historical ways of thinking?

If there is any one idea central to historical ways of thinking, it is that the order of things makes a difference, that reality occurs not as time-bounded snapshots within which “causes” affect one another (. . .), but as stories, cascades of events. 

Q13. Why is the MLP more descriptive than explanatory?

Because the practice concepts are more descriptive than explanatory, it remains somewhat unclear how they can be used to analyse transition dynamics in a way that goes beyond the empirical mapping of individual cases. 

Q14. What is the difference between grand theory and MRT?

MRT has the following characteristics (Geels, 2007): (a) MRT are not about broad, abstract entities such as ‘society’ or ‘social system’, but about concrete phenomena (such as socio-technical transitions), (b) MRT differs from grand theory, because it emphasizes interactions between theory and empirical research.