Technological Forecasting & Social Change 78 (2011) 872–882

Contents lists available at ScienceDirect

Technological Forecasting & Social Change

Exploring social structures and agency in backcasting studies for

sustainable development
Josefin Wangel ⁎
Division of Environmental Strategies Research, Royal Institute of Technology, Stockholm, Sweden

a r t i c l e i n f o a b s t r a c t

Article history: This paper examines how social structures and agency have been included in backcasting
Received 22 December 2010 studies for sustainable development. For this purpose an analytical framework was developed,
Received in revised form 2 March 2011 based on what objects of change (whats), measures (hows) and change agents (whos) are
Accepted 3 March 2011
included in the scenario, and to which extent these are approached in an explorative way.
Available online 3 April 2011
Through reviewing a number of backcasting studies it was found that these typically are built
upon and elaborated with a predominant focus on the questions of what and how physical/
Keywords: technical aspects could change. Social objects of change and explicit representation or analysis
Backcasting of the question of who could change is rarely included in the analysis. This unbalance brings a
Socio-technical
number of implications. Firstly, not including social structures and agency obstructs developing
Social
socio-technically consistent and comprehensive scenarios. Secondly, through not addressing
Agency
Change agent the questions of how to change and change by whom in an explicit and explorative way, social
structures and agency become represented only implicitly and/or are maintained according to
the status quo.
© 2011 Elsevier Inc. All rights reserved.

1. Introduction

Throughout its history, the focal point of (environmental) sustainable development has shifted numerous times, both in terms
of problem formulation and of measures to be taken [1,2]. Starting in nature conservation, ecology, and end-of-pipe pollution
control, sustainable development then became recognised as a matter of cleaner production, a question for local action and
participation [3,4], and sustainable consumption [5–10]. Rather than being consecutive, these focal points have also led to an
understanding of sustainable development as characterised by complexity. This complexity arises from the perceived need to
integrate social, ecological and technical subsystems [11–13] and from these systems being understood as multi-levelled [16],
dynamic and characterised by different types of uncertainty [14]. The shift in focal points does not only imply a shift from end-
point abatement to more pro-active approaches but has also brought about recognition of the importance of also including social
structures when addressing sustainable development, preferably through a socio-technical approach [9,10,15].
Backcasting is a futures studies approach which has been recognised as a fruitful way for addressing sustainable development,
an object of study which is complex, calling for major changes, and for which dominant trends are parts of the problem [16]. One
main characteristic of backcasting studies is the development of one or more goal-fulfilling images of the future, answering the
question of how a certain target can be met when contemporary structures block the changes sought [17–19]. The images of the future
are then connected to the present through elaborating one or more pathways of transition, developed from the future looking back
[17,18,20].
Backcasting was first used in the 1970s as an approach for the analysis and planning of energy systems, but has since been used
to address a wider field of sustainability issues, such as land use, transport, buildings and food. In this way backcasting can be seen

⁎ Environmental Strategies Research, DKV 30, Royal Institute of Technology, SE-100 44 Stockholm, Sweden. Tel.: +46 8 790 85 88.
E-mail address: josefi[email protected].

0040-1625/$ – see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.techfore.2011.03.007
 J. Wangel / Technological Forecasting & Social Change 78 (2011) 872–882 873

as having followed the focal points of sustainable development quite closely. Also in the field of backcasting can be found a
recognition that far-reaching societal changes – such as sustainable development – require social structures to be included in the
process of change [18,21–26].
However, recognition in theory does not equate recognition in practice. This paper aims at exploring to which extent this
recognition has influenced the practice of backcasting; i.e. in which ways and to what extent do backcasting studies of sustainable
development include, analyse and represent social structures and agency.
This paper is structured as follows: Section 2 gives a brief introduction to how social structures and agency can be understood,
while the concept of backcasting is further introduced in Section 3. Section 4 describes the methodology used for the literature
review, including an analytical framework developed for this purpose. Section 5 presents the findings, which are discussed in
Section 6 and used as the basis for some concluding reflections which are presented in Section 7.

2. Social structures and agency

A variety of social and socio-technical theories and approaches exist through which social structures and agency can be
understood; each with their implications concerning what is found through a backcasting study and how these findings are
interpreted and translated into recommendations for policy and practice.
In contemporary strategies for sustainable development technological fixes and behavioural changes typically make the
standard pair of solution. To explore and explain opportunities and pitfalls of policies and behavioural interventions the social
sciences have typically been ascribed a role of ‘social engineering’ in which unwanted behaviours are seen as amendable through
the two magic wands of economic incentives and informational campaigns. However, this role has been criticised for being carved
out in a too rational and technocratic manner [16,27–32]. Critics question the understanding of people as individual decision-
makers driven by economic and moral rationality and argue that people rather are to be understood as social negotiators who
reflexively rationalise and change their activities in relation to other people [33]. Moreover, that the ‘social engineering’ approach
neglects the extent to which technological and physical structures influence the social dittos. Instead, critics propose a socio-
technical approach.
In its most basic sense a socio-technical starting-point implies an understanding of society as a seamless web in which
technological and social aspects are interwoven and mutually structuring. Thus, no part of a socio-technical configuration can be
changed without also taking account of the others. At the macro-level of society this means seeing policy at the same time being
dependent on and constituting techno-political paradigms and regimes, and socio-technical path-dependency [11,14,34]. At the
level of households or individuals this not only means acknowledging people as social negotiators but also taking into account the
enabling, restricting or even persuasive effect of technologies and physical structures [29,31,32].
In this paper the terms social structures and agency are used to denote ‘the social’ part of the socio-technical society. This
comprises both formal institutions, i.e. explicit and formalized structures such as policies, recommendations, taxes, and
organisations, and informal institutions such as regimes, norms, values, and social practices. Agent and agency are here used to
mean human actors and are used to refer to individuals and organisations with a factual or fictitious capacity to act. Besides this,
social structure and agency are not restricted to any specific theoretical starting-point concerning how ‘the social’ and ‘the
technical’ are understood, or how they are thought to interact.

3. Backcasting

Backcasting is not an unambiguous concept. There are a variety of interpretations, with one of the main dividing lines being the
relative emphasis on whether the image of the future needs to be developed as goal-fulfilling or not [35,36], and whether it is seen
as crucial or optional to outline pathways of transition too [18,21], or whether this is something that should be avoided altogether
[23]. Another dividing line noted is the degree of participation in the backcasting study, i.e. whether the scenarios are developed
mainly by experts or through stakeholder or citizen participation [19,37].

3.1. Backcasting for different purposes

This paper recognises that even though this diversity of interpretations might be confusing, there are good grounds for
retaining them. Backcasting studies are used for a variety of purposes and depending on the aim, each of the aforementioned
backcasting practices has its rationale. The potential gains of including social structures in a backcasting study are highly
dependent on the purpose of the study. Consequently, in accordance with the scenario typology developed by Börjeson et al. [20],
this paper distinguishes between different approaches based on the overarching aim of the backcasting study. Primarily it can be
useful to distinguish between backcasting as a result-orientated research approach, and backcasting as a participation-orientated
creative workshop technique [38]. When being used as a research approach methodological stringency is an important trait and
the question of participation is subordinate to the purpose of the study. When being used as a workshop technique the reverse is
true and the backcasting methodology can be adjusted to better suit the desired outcome of the participation. Result-orientated
backcasting can be further distinguished depending on the extent to which the study is target-orientated, pathway-orientated,
and/or action-orientated. These are not mutually exclusive, as a single study could include all three aspects, but depending on the
relative emphasis placed on these, the rationale for including social structures and agency will differ.
874 J. Wangel / Technological Forecasting & Social Change 78 (2011) 872–882

3.2. Result-orientated backcasting

3.2.1. Target-orientated backcasting

In target-orientated backcasting, the emphasis is on the need to develop and describe images of the future as goal-fulfilling.
This goal is typically expressed in a quantitative manner, such as a 60% decrease in energy use. The rationale for stressing goal
fulfilment is that this challenges the imagination to identify more solutions and to go beyond that which is considered probable or
feasible. Thus, target-orientated backcasting can be said to explore the question of what can change. Depending on the object of the
study, the need for a comprehensive socio-technical approach to this question differs. In backcasting studies the objects of change
is typically of a technical or physical character, but if including also social structures the objects of change could also be e.g. social
innovations, organisational changes, taxes, behavioural changes or changed norms and values.

3.2.2. Pathway-orientated backcasting

In pathway-orientated backcasting the formulation of an image of the future as (quantitatively) goal-fulfilling is seen as less
important. It has even been argued that setting strict goals is something that should be avoided, the argument being that
sustainable development is, and should be, a ‘moving target’ [35]. Instead, the emphasis is on exploring and bridging the gap
between the sustainable image of the future and today. With the aim of exploring this, the overarching aim of pathway-orientated
scenarios can be understood as exploring how change can take place. The focus on the process of change means that it is important
to include also non-technical measures, such as policy, taxes or behavioural changes, but could also mean including the agents
planning, implementing and managing the transition.

3.2.3. Action-orientated backcasting

In action-orientated backcasting the overarching aim is to develop some kind of action plan or strategy. Here the questions
what to change and how to change are both important, but the question who could make change happen also needs to be addressed.
This is typically delimited to a predetermined set of actors, in which case the scenario is limited to that which is within the scope of
influence of these actors, but could also make use of a more exploratory approach to the question of who, thereby working as a way
of stakeholder identification. Action-orientated backcasting studies typically speak of transition management, adaptive and
reflexive management, buy-in and the importance of social learning, rendering stakeholder participation in the process an
important trait. Such studies explore how a group of actors could achieve a desired future of some kind, or avoid an undesired one,
as when creating a strategic plan for a predefined set of stakeholders in scenario planning [39].

3.3. Participation-orientated backcasting

As described earlier, participation-orientated backcasting is used as a creative workshop technique. In this case the backcasting
methodology is subordinate to the intended outcome of the participation, such as empowerment, increase in social capital or
creation of other ‘soft values’. Indeed also result-orientated backcasting approaches can make use of participative techniques, the
difference being that in participation-orientated backcasting it is the participation per se that constitutes the overarching aim, and
is placed above other potential outcomes.

3.4. Backcasting and participation

One main difference between this division and other categorisations is that participatory backcasting is seen not as a separate
category but rather as one possible approach by many through which scenarios can be developed. Thus the notion ‘participatory’
can be assigned to any or none of the three aforementioned categories. The potential benefits with participatory scenario
approaches have been recognised by many, leading to an upheaval of studies in which different kinds of participants are actively
involved in the process of scenario development [19,36,40]. Commonly heard reasons for participation are to include a variety of
perspectives and kinds of knowledge in the scenarios, to create consensus or map conflicts, to generate community
empowerment, social learning and capacity-building, to increase stakeholder buy-in, and to increase the legitimacy of the
resulting scenarios. However, participation also comes with potential drawbacks concerning the resulting scenario and how the
social aspects are explored and represented. This will be returned to in the discussion in Section 6.

3.5. Social structures and agency in backcasting studies

There are a number of arguments to include social structures and agency in backcasting studies for sustainable development.
Firstly, this enables addressing change as socio-technical. Moreover, rather than adding social structures and agency on
afterwards, including these from the start facilitates retaining the socio-technical character of change throughout the study. Socio-
technical configurations are complex and, if taken seriously, not only a matter of placing a technical change in a social context.
Simultaneously addressing not only what should change, but also how to change and change by whom has the potential to
produce richer material. It has been shown that a non-negotiable iteration between what could change and who could change
triggers imagination and is beneficial when it comes to identifying and exploring social solutions and innovations too [41–43].
Secondly, explicit inclusion of social structures and agency facilitates a more structured reflection on the socio-politico-cultural
context in which the image of the future is embedded, and in which the pathway of transition is to take place. For instance, in this
 J. Wangel / Technological Forecasting & Social Change 78 (2011) 872–882 875

way it could transpire that existing institutions block the changes sought. Furthermore, Hillman and Sandén [44] point out that
even a small difference in assumptions about agents' attitudes and choice of measures can have major implications for the
resulting scenario. This is also true for assumptions or choices of which agents to include. Thirdly, through an explicit and open-
minded inclusion of social structures and agency in backcasting these too can be approached as objects of study and be explored in
an unprejudiced way.
The inclusion of social structures and agency in backcasting studies comes with two methodological implications. Firstly, in
respect of developing images of the future, this means that rather than only including physical/technical objects of change, social
structures should also be included and explored. In this respect social structures are included as objects of change, such as energy
use practices, lifestyles, organisational changes, and social innovations answering the question what could change. Secondly, in
respect of developing pathways of transition, this also means including and taking an explorative approach to the question of
agency and measures, i.e. by whom and how change could take place, rather than developing these as according to the status quo.
These questions of objects (what), measures (how) and agency (who) can be further defined as follows:

• What can change?

Does the study include only physical/technical or also social objects of change? What kinds of social structures are included: e.g.
social innovations, behaviour, norms and values, taxes, the temporal organisation of society? Does the study use a socio-
technical approach in which social and technical/physical structures are seen as interwoven or are these kept separate?
• How could change take place?
Does the study include any discussion or analysis of how change could take place, and is this approached in an explorative way,
or does it adhere to existing structures and status quo? Furthermore, through what kinds of measures is change assumed to take
place? Are measures top-down, bottom-up or multi-levelled? Are they technical/physical, social or socio-technical?
• Who could change?Does the study include the question of who could change and if so, is this done through an explorative
approach or are the agents determined in advance (as when developing an action plan for a determined set of agents) or in other
ways based on existing structures? Furthermore, how are the agents represented in the resulting scenario? Are they compiled
into dimensions such as public/collective? Or are they presented as agents of change in the scenario narrative? If so, are they
then seen as separate agents or are they related to each other through e.g. collaboration or conflict? Are they organised into
structures and processes of governance?

4. Examining the literature for social structures and agency

4.1. Literature review

To explore how social structures and agency is included and represented in backcasting studies, a number of backcasting
studies was identified through a literature search in the scientific publications data base SCOPUS and their content was reviewed.
The review was limited to scientific articles reporting on backcasting studies aimed at sustainable development focusing on
transport, the built environment, and energy in general.1 These areas were chosen due to their high dependency on social
structures and agency for environmental performance and processes of change. This way of selecting studies draws on case study
methodology's recognition that there can be more to learn from the unique and extreme than from the average [45].
However, not all backcasting studies are reported in the form of scientific papers, and scientific papers do not always report all
parts of a study. Thus the results of this literature review should be seen as a preliminary pointer and not as a comprehensive and
final account of the field of backcasting studies.
The publications were assessed through a two-step approach. First, the abstract was read to exclude all publications that did
not report on a backcasting study or backcasting methodology. Meta-studies such as literature reviews or comparisons were also
excluded, since these typically constitute second-hand information on the studies included. A selection of the publications passing
this first assessment was then fully read and analysed using the framework in Table 1.
During the literature review, note was also made of where in the paper/study ‘the social’ appeared; i.e. in the theory,
methodology, results, discussion, or conclusions section. When analysing the studies based on the aforementioned questions, the
main approach was to base this on the methodology and results, rather than on the theoretical framing. Having a theoretical
understanding of the need to connect the backcasting study (or research in general) to policy, institutions and other parts of the
social structures is not the same as actually making this connection. It was also noted whether the studies' main focus was
perceived as being target-orientated, pathway-orientated, action-orientated, or participation-orientated.

4.2. Framework for analysis and reflection

In Table 1 the aforementioned questions of what, how, and who are arranged in to a framework. The aim of this framework is
not to provide a typology, but to facilitate a more structured analysis and reflection on how social structures and agency have been
and can be included in backcasting studies. This framework distinguishes between backcasting studies according to what they
include in the scenario (objects, measures and agents) and also in how these are approached, i.e. whether they are included

1
To increase the quality of hits in relation to the conditions for publications to include in the literature review, the search words (backcast* OR back-cast*)
AND (sustainab*) AND (transport* OR living OR building* OR housing* OR urban* OR energy) were searched only in the title, abstract, or keywords.
876 J. Wangel / Technological Forecasting & Social Change 78 (2011) 872–882

Table 1
Framework for analysis and reflection.

Approach to object of Approach to measures for Approach to actors of Summarising description of the ideal types
change (what) change (how) change (who)

Explorative None None 1. Doesn't mind how or who?

Backcasting studies elaborating on WHAT in an explorative way.
Explorative Status quo None 2. Knows how and doesn't mind who?
Backcasting studies elaborating also on how, but not who, and based on existing
structures
Explorative Status quo Status quo 3. Knowing both how and who?
Backcasting studies elaborating on how and who, but based on existing structures
Explorative Explorative Status quo 4. Knowing who and wanna know how?
Backcasting studies elaborating on how in an explorative way and who based on
existing structures
Explorative Explorative Explorative 5. Wanna know both how and who?
Backcasting studies elaborating on how and who in an explorative way

according to status quo or elaborated upon in a more explorative way. In this table an explorative elaboration of what can change is
seen as an inherent trait of backcasting and is understood as the basis for all types of backcasting studies, i.e. the question of what is
not only present in the first type.

4.3. What is a what, a how and a who?

It can be difficult to distinguish between objects of change (what), measures (how), and actors of change (who) in a scenario.
For instance, a social innovation in terms of a transport demand management (TDM) institution can be seen as answering the
question ‘What could change?’, but could also be the answer to the question ‘How could change take place’? In the latter case the
social innovation (e.g. the TDM institution) is a measure to make something else happen (e.g. decreased use of private cars), i.e. the
question of ‘how’ comprises an underlying ‘what’. When used as a “how” the scenario could be narrated as: “Through establishing
a TDM institution a decrease in the use of private cars was accomplished.”
Furthermore, a TDM institution could also be the answer to the question ‘Who could make the changes happen?’, with the
underlying question of ‘how’ (e.g. through TDM measures) and ‘what’ (e.g. decreased use of private cars). In this case, the scenario
could be narrated as: “Through implementing a variety of TDM measures, the newly founded TDM institution succeeded in
decreasing the use of private cars.”
This understanding of scenarios as comprising multiple tiers of objects (what), measures (how) and agents (who) can also be
used to develop more comprehensive scenarios. With the target of the scenario study as the starting-point, one or more objects of
change (what) are identified, which are then complemented with measures (how) and actors (who). In the next step these
measures and actors are turned into new objects of change, which in turn are associated with measures and actors, and so on. This
kind of iterative approach supports the development of a socio-technically comprehensive and consistent scenario and also
facilitates a more deliberate consideration of the scenario's context.

5. Results

The results from the literature review are presented in the following subsections. A list of the studies is given in Table 2,
where they are arranged according to their approach to the questions what, how and who, and whether the author of this
paper categorised their main focus as being target-orientated, pathway-orientated, action-orientated or participation-
orientated.

5.1. Doesn't mind how or who? Backcasting studies elaborating on what in an explorative way

This section presents backcasting studies perceived as exploring what could change. The question of how or by whom change
could take place is thus not obviously included in the scenario study. Drawing on the studies included, this is not always an explicit,
or deliberate, delimitation.
One clear-cut example of this type of study is Krewitt et al. [46], which explores the potential for a 60% decrease in global CO2
emissions by 2050. The study includes both the demand and supply sides of the energy system and is explicitly delimited to
exploring only the technical potential, while still acknowledging the importance of “Significant innovations, not only in the
technical field, but in particular in the social and institutional context” [46, p. 4969]. Åkerman's study of sustainable air travel in
2050 [47] and the study by Van der Graaf et al. [48] on sustainable urban water cycles do not comprise any measures or agents, but
are less explicit whether this is the result of deliberate consideration. Yet another example of this type is the study by McDowell
and Eames [49] on the environmental, social and economic sustainability of hydrogen futures for the UK. The aim of the study was
to “open up debate on the relative sustainability of different options” [49, p. 4612] rather than identifying the most sustainable
hydrogen future. In total, six visions of hydrogen futures were developed based on ideas of what could change in terms of
 J. Wangel / Technological Forecasting & Social Change 78 (2011) 872–882 877

Table 2
Studies included in literature review.

1. Doesn't mind how or who? Backcasting studies elaborating on WHAT in an explorative way

Target-orientated
Krewitt et al. [46] The 2 °C scenario — A sustainable world energy perspective
Åkerman [47] Sustainable air transport — On track 2050.
Van Der Graaf et al. [48] Sustainable technological development for urban-water cycles.
McDowell and Eames [49] Towards a sustainable hydrogen economy: a multi-criteria sustainability appraisal of competing hydrogen futures.

Action-orientated
Eames and Egmose [50] Community foresight for urban sustainability: insights from the Citizens Science for Sustainability (SuScit) project.

2. Knows how and doesn't mind who? Backcasting studies elaborating also on HOW, but not WHO, and based on existing structures

Target-orientated
Giurco et al. [51] Backcasting energy futures using industrial ecology.
Robèrt et al. [52] Biofuels in the energy transition beyond peak oil. A macroscopic study of energy demand in the Stockholm transport system 2030.
Geurs and Van Wee [26] Backcasting as a tool to develop a sustainable transport scenario assuming emission reductions of 80–90%.

Participation-orientated
Carlsson-Kanyama et al. [53] Participative backcasting: a tool for involving stakeholders in local sustainability planning.

3. Knowing who and wanna know how? Backcasting studies elaborating on HOW in an explorative way and WHO based on existing structures

Action-orientated
a
Truffer et al. [54] Mapping expectations for system transformations: lessons from sustainability foresight in German utility sectors.
de Graaf et al. [55] Roadmap for interactive exploration of sustainable development opportunities: the use of simple instruments in the complex setting of
bottom-up processes in rural areas.
a
Holmberg and Robèrt [56] Backcasting — a framework for strategic planning.
Robèrt [57] Backcasting and econometrics for sustainability planning: information technology and individual preferences of travel.

Target-orientated
Gomi et al. [58] Scenario study for a regional low-carbon society.
Wallgren and Höjer [59] Eating energy — identifying possibilities for reduced energy use in the future food system.
Åkerman and Höjer [23] How much energy can the climate stand? — Sweden on a sustainable path in 2050.
Anderson et al. [24] The Tyndall decarbonisation scenarios — Part II: Scenarios for a 60% CO2 reduction in the UK.
Mander et al. [25] The Tyndall decarbonisation scenarios — Part I: development of a backcasting methodology with stakeholder participation.

4. Wanna know both how and who? Backcasting studies elaborating on HOW and WHO in an explorative way

Pathway-orientated
a
Green and Vergragt [22] Towards sustainable households: a methodology for developing sustainable technological and social innovations.
Vergragt and Brown [60] Sustainable mobility: from technological innovation to sustainable learning.
Hisschemöller and Bode [61] Institutionalized knowledge conflicts in assessing the possible contributions of H2 to a sustainable energy system for the Netherlands.
a
Svenfelt et al. [62] Decreasing energy use in buildings by 50% by 2050 — a backcasting study using stakeholder groups.

Table 2 shows a list of studies included in the literature review. Altogether, 38 publications were identified through the literature search. However, a large
proportion of these were not backcasting studies, or were meta-studies. These were not included in the review. In this table category 3 is missing, the reason for
this being that no studies of this type were found in the review.
a
Studies not identified through the search in SCOPUS but through previous research.

technologies and infrastructures. With the main aim of the study being the sustainability appraisal rather than the visions per se, it
might not seem overtly relevant to also include agents in the vision. However, without a representation of agents it becomes more
difficult to assess the impact of the vision on social and economic sustainability in terms of e.g. social justice and who controls the
energy.
Eames and Egmose [50] report on a study aimed at letting citizens formulate their own sustainable futures. Those authors
regard research and policy as complex processes, in terms of being characterised by networks, participation, vested interests,
powers and agencies. This ‘Community Foresight’ methodology is used to outline a research agenda, so that instead of scenarios
and pathways the result is a number of identified fields of future research for sustainable urban development. The visioning
exercise comprises two parts: the making of fictitious positive news stories and a picture of a sustainable house, a sustainable
street/estate and a sustainable city in 2028. It is not unlikely that the news stories included some agents (it is hard to tell a story
without them) but if so these did not appear in the resulting research fields. Even though these are socio-technical in one sense
(e.g. “What are the specific barriers to promoting walking and cycling in deprived urban areas?”), there are no questions posed
about who should realise the research agenda or how this could be done. At the end of the paper the authors return to the question
of policy, but rather than viewing policy as an object of study they conclude that the ‘Community Foresight’ methodology could
be more action-orientated, through facilitating network formation and capacity building, but only “given the appropriate
institutional and political context and support”.
878 J. Wangel / Technological Forecasting & Social Change 78 (2011) 872–882

5.2. Knows how and doesn't mind who? Backcasting studies elaborating also on how, but not who, and based on existing structures

This section describes backcasting studies that explore what could change and how change could take place. The question of who
could change is not included in the studies. This could be seen as the typical pathway-orientated backcasting study in which an
image of the future is elaborated, supplemented with a pathway answering the question of how change could take place.
Giurco et al. [51] examined how to develop backcasting scenarios through industrial ecology. Although this study is not aimed
at exploring stakeholders or other agents in the scenario, the focus on industrial ecology renders the question of potential
industrial collaborations crucial. The study does not use any quantitative goal, but with a strict outlining of the scenario according
to the principles of industrial ecology, this can still be considered a target-orientated approach. Other examples of this type of
study are that by Robèrt et al. [52] on the potential for domestically produced biofuels in the Stockholm County transport system in
2030, and that by Geurs and Van Wee [26] on sustainable transport scenarios.
Another example of this type of study is the participation-orientated backcasting study of Carlsson-Kanyama et al. [53], which
mainly aimed to develop ways to include local stakeholders in discussions on sustainable development, but also resulted in a
number of images of the future based on scenario dimensions developed back-office. In most respects these dimensions were
based on what to change, but one of the dimensions, ‘the regime about organisation of society’, concerns the question of who could
change, but not in a detailed way.

5.3. Knowing who and wanna know how? Backcasting studies elaborating on how and who, but with who only based on existing
structures

This section presents backcasting studies exploring what could change, how change could take place and who could change, but
with who being addressed only as based on existing structures. In this type of study the agents thought to realise the scenario
are seen as determined in advance and are taken as input to the scenario exercise, leaving only the question of how open for
exploring.
Many of the action-orientated backcasting studies fall into this category. One example is Truffer et al.'s ‘Sustainability Foresight
method’ for developing sustainable utility sector futures in Germany [54]. Another is the ‘Roadmap method’ as reported by de
Graaf et al. [55], in which it is worth noting that in spite of a stated stakeholder focus, the identified characteristics required to
achieve the goals are only represented in terms of what is needed; agents and other social structures are missing completely [55,
p. 302]. The Natural Step approach to backcasting in strategic planning for sustainable business [56] can also be seen as belonging
to this category, as the study by Robèrt [57] on backcasting and econometric modelling to explore the potential of mobility
management.
Other examples of this type are studies on integrated scenario assessment [13,63], although it is debatable whether these
approaches can be considered backcasting studies per se, or rather should be seen as participation-orientated backcasting
approaches. Indeed, it should be noted that these transition approaches do not always make use of scenarios, and even when they
do, they do not use them to explore the agents or governance of transition, but rather adopt a learning-by-doing approach.
There are also studies of this kind without the expressed aim of developing the findings into any kind of action plan.
Consequently, there is less need for including stakeholders or other participants in the backcasting exercise.
In creating scenarios for a low-carbon future in the Shiga region in Japan, Gomi et al. [58] not only reflected upon what could be
changed, and through which policies and measures, but also identified local government as playing an important role to “to set up
an arena of dialogue and partnership and, on the other hand, to integrate policies on land-use reform, transport and urban
planning, industrial policy providing incentives to introduce low-CO2 technologies, and housing guidance policy.” [58, p. 129]
However, in the table of “Main policies, measures and technologies considered in scenarios” there are many ideas on measures and
technologies, but no policies or agents. Measures are divided into the sectoral categories “energy conversion, industrial, household,
commercial, passenger and freight sectors” but this does not correlate to any kind of assigned agency or responsibility. This
indicates that the reflections on agency and policy were made ex post the scenario creation, rather than being included in the
scenario per se.
In their study on ‘how much transport the climate can stand’, Åkerman and Höjer [23] discuss the kind of policies that could
contribute to the scenario and conclude that the uncertainty of the future calls for flexible policies, as well as that there is a need of
ensuring that implementing the outcome of the policies in practice does not lead to any lock-ins with unsustainable outcomes.
They also conclude that in order for the changes in the scenario to come true, “substantial changes regarding both spatial and
institutional structures are necessary.” [23, p. 1955]. However, it can be worth reflecting upon whether it is the policies that need to
be made more flexible, or whether it is the process of policy-making that needs to be more reflexive.
Wallgren and Höjer [59] explore the possibilities of reducing energy use for eating to a sustainable level, defined as 60% less
than today. A total of 14 changes are identified. These are not explicitly combined with agents, but some examples are given in the
descriptions of the changes: “Energy use for cooking can also be decreased by cooking larger amounts of food on each occasion,
which could be achieved by residents cooking food together in common neighborhood kitchens and having meals together in local
groups” [59, p. 5808]. In the narrative of the scenario, agents and social innovations are also present to some extent: “Various
cooperative forms of interaction between producers and consumers have become common. Consumers are often part-owners in
farms and employed on the farm during sowing, weeding and harvest or in processing of products. Schools and working life is
organised so as to allow people to take leave to work in the fields or in their own allotment gardens.” [59. p. 5810].
 J. Wangel / Technological Forecasting & Social Change 78 (2011) 872–882 879

The Tyndall decarbonisation scenarios [24,25] outline pathways to a 2050 UK energy system with a 60% reduction in CO2
emissions. In the study, six different images of the future are defined and pathways of transition are developed. The scenarios are
based upon a number of key factors clustered in a number of demand side and supply side themes. These themes are divided into a
number of sectors, such as household, passenger transport, role of nuclear power and localised versus centralised generation of
energy, but do not explicitly display any key actors (to deal with the key factors). Factors included are defined as “a level of change
in technologies, values, behaviors, infrastructure, or other physical or social variables, excluding policy instruments, necessary to
bring about an end-point scenario” [24 p. 3759]. The resulting scenarios are mainly made up of objects and measures of change,
but some examples of agents can be found, mostly in quite coarse terms such as government, local government and sectors,
although homeowners and landlords are also mentioned. Even though not being an explicit aim one of the scenarios also includes
an innovated agent: “ESCOs aim to achieve long-term improvements in the energy performance and carbon-reduction targets and
are regulated by an independent regulator whose remit includes social and environmental, as well as economic, criteria.” [24]
(p. 3766, emphasis by this author) and “(P) public-private partnerships are established between research groups and the energy
industry to develop a series of pilot carbon capture experiments.” (24, p. 3768).

5.4. Wanna know both how and who? Backcasting studies elaborating on both how and who in an explorative way

This section presents backcasting studies that explicitly and openly explore both how and by whom changes could take place.
Although agents are not included to any great extent in the resulting scenario, the SusHouse methodology has the explicit aim
of identifying “future stakeholders, potential actors in future scenarios” [22]. The SusHouse methodology is aimed at exploring
technological and social innovations for more sustainable household functions (clothing care, shelter and food). One reason why
the SusHouse methodology does not discuss the agency of change might be that it makes use of design-orientated scenarios (DOS)
instead of policy-orientated scenarios (POS). The difference is a matter of scale. DOS are developed at the ‘micro’ level of household
functions, while POS typically are directed at the societal level [22].
As far as can be judged from the initial reporting, another study of this type is the Tellus institute's scenarios for a more
sustainable greater Boston area in 2050 [60]. In the last part of the ‘narrative for sustainable mobility’, agents are clearly present:
“From an institutional perspective, public-private partnerships have realized innovative solutions by experimenting with new
technologies. By including universities and technical institutes in research and development, new private high-tech enterprises
developing innovative mobility solutions are thriving. Advanced information and communications technology (ICT) is widely used
for road pricing, congestion pricing, fare payment, trip reservation, information and communication services, tele-working and
tele-shopping, combining trips, and vehicles sharing. These organizations and businesses are an important component of a
thriving regional economy.” [60, p. 1112].
Another example is the study on hydrogen futures reported by Hisschemöller and Bode [61]. That study makes use of stakeholder
dialogue to address how hydrogen can contribute to a future sustainable energy system in the Netherlands, including the question
“To what extent does technological change require institutional change and vice versa?” [61, p. 2]. This question is clearly mirrored
in the resulting scenarios, which present a variety of agents and measures combined into processes of change such as “Lease
companies and public transport could act as launching customers, provided that a market will emerge for used H2FC vehicles. Here,
the group saw a task for government. Another government task would be to provide an adequate hydrogen infrastructure to supply
the transport sector.” [61, p.3–4]. In this way of linking technology and institutions in the scenario the socio-technical character of
change is explored in its entirety, which also renders ideas and assumptions about agency explicit and an object open to discussion.
In the backcasting study by Svenfelt et al. [62], focus groups were used to elaborate on both what is needed to achieve a 50%
reduction in energy use in buildings by 2050, and who could be seen as the agents of change for the measures identified. The
measures (what) and stakeholders (who) are presented together in a list, but the study does not incorporate the ‘agents of change’
into any scenario narrative.

6. Discussion

6.1. Some potential explanations to the lack of social structures and agency

There are many possible explanations for the lack of explicit elaboration of social structures and agency in backcasting studies
for sustainable development. Firstly, as all sciences, also backcasting is characterized by path-dependency enforced by e.g.
education and research funding [64]. Backcasting can also be understood as a social practice in which, at least up to now, the
exploration of social structures and agency has not been a part.
One plausible explanation is that the changes sought are regarded not as the outcome of social pragmatism, but as relying on
the behaviour of economically and morally rational individuals [30–32], best influenced through information campaigns and
economic incentives [65]. However, this standpoint receives little support from empirical studies of how information affects
behaviour [10,28,29,66]. In connection to this it could also be the case that policy processes and governance are understood from a
predominately technocratic ‘rational’ perspective, seeing these as linear and straight-forwardly implementable; not much need for
analysing the governance of change in that case… Furthermore, since many of the backcasting studies are policy-orientated
scenarios (POS) rather than design-orientated scenarios (DOS) [22], the need and reasons to include agents in terms of end-users
in the resulting scenario become less pressing. However, in a policy-orientated scenario it could still be worthwhile to include
agents in other roles.
880 J. Wangel / Technological Forecasting & Social Change 78 (2011) 872–882

It could also be argued that it is offensive to elaborate on future actions of real-life stakeholders without their participation or
approval. But is a scenario in which a municipality teams up with a car-pool association and the public transport provider to build a
car-free city district more improbable (or offensive) than the same scenario but without explicit agents of change? That scenario
also comprises stakeholders, but implicit and thus hidden from analysis. This reluctance goes for social structures as well. It is not
unlikely that it is perceived as less problematic to elaborate on technological changes since technologies typically are understood
free from values and norms — in difference to the value-laden social. However, technology is policy; technology is social and can
only be understood in association to the social context in which it takes part.
Lastly, it could also be so that the absence of agents (i.e. the subjects in a narrative) in scenarios of backcasting studies is partly
an effect of the traditional ideal of scientific writing in a ‘neutral tense’ (without any subject), and that this practice persists in
writing scenario narratives.

6.2. Asking who?

The what–who approach used by Svenfelt et al. [62] is also used in an ongoing research project2 in which the author is involved.
In this project the questions of what and who could change, together with how much and when, form the basis for developing
scenarios of more sustainable Stockholm City Districts. Here an explicit aim is to use the agents of change identified in the scenario
to elaborate also on the governance of change. This approach of iteratively identifying and elaborating upon objects, measures and
agents, and developing this into a process of governance has so far been used in two scenario studies in the project, both with an
overarching aim of exploring potentials for decreasing energy use. The what–who approach has also been successfully tested in
education [41,42]. A fuller description of the what–who approach and the outcome of these first scenario studies is given in a
research report-in-progress [43] and will be analysed in detail in forthcoming papers.
As has been found, apart from the aforementioned what–who approach one has to go outside the field of backcasting to find a
futures studies methodology in which the identification of actors is given an explicit and prominent place in the methodology and
approached from an explorative perspective. The first example of this is the European Union's futures study methodology
Fleximodo [67]. This Fleximodo methodology which is not a backcasting methodology has been included here as an example of a
study that explicitly builds upon the questions how and who? The latter is described thus: “WHO will take care of solutions for
sustainable urban living? Will solutions and actions mainly come from users and local residents, or from investments and
innovations by the private sector, or will solutions come from authorities, regulations and large infrastructures?” [67, p.25] The
second example is the methodology called “Shaping Actors–Shaping Factors” [67]. In this methodology, scenarios are built around
“shaping factors, processes and constitutive elements” and “key shaping actors affecting these elements and thus driving the
development (…)” [68, p. 628].

6.3. Stakeholder participation in scenario development

Not many of the reviewed papers comprised any reflection or accounts on how the choice of stakeholders or other participants
to involve in the scenario development might have influenced the outcomes of this. Hisschemöller and Bode [61] claim that
stakeholder participation can in fact contribute to a retained path-dependency due to the participants safeguarding their vested
interests. Svenfelt et al. [62] found that participants were reluctant to identify themselves as the stakeholders responsible. While
the former may have more impact on the scenario's objects and measures of change, the latter has clear implications for how
agents are represented in the scenario. On the other hand, Bohunovsky et al. [63] found nothing of this phenomenon, but then it
should be noted that their study was of an action-orientated type in which stakeholders had already accepted their responsibility
by joining in and that they also were empowered and allowed to write their ‘own’ future.
Should the participants be stakeholders or non-stakeholders? Should they be experts or lay people? Should the group be
homogeneous, to facilitate group dynamics, or heterogeneous, so as to let different perspectives meet? The decision on whether to
use a participatory approach and which stakeholders or other participants to include must be based on the purpose of the study. If
the primary purpose is to affect policy or practice, then inclusion of stakeholders might be worthwhile due to the increased
potential for buy-in, social learning and empowerment, and the risks of safeguarded interests obstructing innovation can be seen
as less important. On the other hand, if the purpose is to explore alternative, innovative or even radical images of the future,
stakeholder participation can be more of a hindrance than a help, at least in the initial phases of scenario creation. Then it can be
considered whether stakeholders should be given a more reactive rather than constructive role in the scenario study, e.g. in terms
of evaluating the feasibility of the scenario or being involved in other types of scenario assessments. However, and once again
depending on the purpose of the study, this too can be questioned – ironic as it might seem – from the perspective of feasibility —
because what is the rationale, what is the intended outcome, of asking stakeholders of today what they think about the feasibility
of a future placed half a century from now and that is deliberately aimed at being different from today? Drawing on Sagoff [69], one
answer to this question is that the needs of future generations will be determined by what we leave them, culture and nature
included. Furthermore, the uncertainty of the future makes ethics, rather than knowledge in terms of rational reasoning, the basis
for assessing feasibility.

2
Learn more of the research project SitCit at http://www.kth.se/abe/forskning/sitcit.
 J. Wangel / Technological Forecasting & Social Change 78 (2011) 872–882 881

7. Concluding reflections

Through examining a selection of backcasting studies for sustainable development it was found that these typically are built
upon and elaborated around the questions of what and how something could change. Explicit representation or analysis of the
question of who could change is rarely included in the analysis. Furthermore, contrary to the proclaimed benefit of backcasting to
break free of existing structures blocking the changes sought, the questions of how to change? and change by whom? are
predominantly elaborated in terms of existing social structures and agency and according to status quo.
An analysis of how and by whom the changes could take place is admittedly not always necessary or relevant in scenario
development. The point made here is not that all backcasting studies need to include a comprehensive and explorative elaboration
of social structures and agency, but rather that an increased awareness about whether and how measures and agents are identified
and included can contribute to making assumptions about these explicit.
It can also be argued that from a social science or socio-technical perspective, scenarios always comprise social structures and
agency. In one sense this is true; realisation of technical potentials, behavioural changes, reconfigurations of energy systems, all
depend on human activities to come about. However, not being explicit about these activities and agents obstructs analysis and
discussion of e.g. the power structures and political landscape in which the image of the future is embedded and through which
the pathway of transition is to take place.
Furthermore, ideas of social structures and agents in the scenario can appear without being explicitly sought. However, one
general finding from the literature review is that there are very few publications in which the questions around which the
scenarios are developed are presented in a precise way. This renders it difficult to assess the extent to which the presence or
absence of social structures and agency in the scenario developed is a result of these not being included in the methodology, or
whether social structures were sought but not identified. It could also be the case that social structures, although not included in
the process of scenario development, are raised in the discussion section. This lack of clarity reveals a rather non-reflective
approach to including social structures in backcasting studies as well as concerning the implications this might have.
This paper is one first step in exploring the inclusion of social structures and agency in backcasting studies for sustainable
development. In order to get a comprehensive understanding of the entire backcasting field, more research needs to be done. This
could include reviewing a wider field of backcasting studies but also a more in-depth examining of the socio-technical
assumptions made. The author is currently working on developing a methodological framework for the considerations needed
when adding agents and a dimension of governance to futures studies. This will be presented in a forthcoming paper.

References

[1] J.S. Dryzek, D. Schlosberg (Eds.), Debating the Earth. The Environmental Politics Reader, Oxford University Press, Oxford, 2005.
[2] E. Störmer, Greening as strategic development in industrial change — Why companies participate in eco-networks, Geoforum 39 (2008) 32–47.
[3] R. Shipley, J.L. Michela, Can vision motivate planning action? Planning Practice & Research 21 (2) (2006) 223–244.
[4] United Nations, Rio Declaration on Environment and Development, A/CONF.151/26, Vol. I, 1992, including Agenda 21.
[5] Worldwatch Institute, State of the World 2010: Transforming Cultures: From Consumerism to Sustainability, 2010.
[6] Worldwatch Institute, State of the World 2004: Special Focus: The Consumer Society, 2004.
[7] European Commission, EU consumer policy strategy 2007–2013. Empowering consumers, enhancing their welfare, effectively protecting them, COM (2007)
998 final.
[8] European Commission, Sustainable Consumption and Production in the European Union, Office for Official Publications of the European Communities,
Luxembourg, 2004.
[9] O. Mont, A. Plepys, Sustainable consumption progress: should we be proud or alarmed? Journal of Cleaner Production Volume 16 (Issue 4) (March 2008)
531–537.
[10] A. Tukker, S. Emmert, M. Charter, C. Vezzoli, E. Sto, M. Munch Andersen, T. Geerken, U. Tischner, S. Lahlou, Fostering change to sustainable consumption and
production: an evidence based view, Journal of Cleaner Production Volume 16 (Issue 11) (July 2008) 1218–1225.
[11] F. Berkes, C. Folke, A systems perspective on the interrelations between natural, human-made and cultural capital, Ecological Economics Vol 5 (No 1) (1993)
1–8.
[12] H. Sverdrup, M. Svensson, Defining the concept of sustainability — a matter of systems thinking and applied systems analysis, in: Olsson, Sjöstedt (Eds.),
Systems Approaches and their Application — Examples from Sweden, Kluwer Academic Publishers, 2004.
[13] P.M. Weaver, J. Rotmans, Integrated sustainability assessment: what is it, why do it and how? International Journal of Innovation and Sustainable
Development Vol. 1 (No. 4) (2006) 284–303.
[14] R. Kemp, J. Schot, R. Hoogma, Regime shifts to sustainability through processes of niche formation: the approach of strategic niche management, Technology
Analysis and Strategic Management 10 (1998) 175–196.
[15] Å. Svenfelt, Two strategies for dealing with uncertainty in social-ecological systems. (2010) Doctoral Thesis in Infrastructure with specialization in
Environmental Strategic Analysis. TRITA-SoM 2010–12. KTH Architecture and Built Environment, Royal Institute of Technology.
[16] S. Owens, L. Driffill, How to change attitudes and behaviours in the context of energy, Energy Policy Volume 36 (Issue 12) (December 2008) 4412–4418.
[17] K.H. Dreborg, Essence of backcasting, Futures 28 (1996) 813–828.
[18] J. Robinson, Futures under glass: a recipe for people who hate to predict, Futures 22 (1990) 820–843.
[19] J. Quist, P. Vergragt, Past and future of backcasting: the shift to stakeholder participation and a proposal for a methodological framework, Futures 38 (2006)
1027–1045.
[20] L. Börjesson, M. Höjer, K.-H. Dreborg, T. Ekvall, G. Finnveden, Scenario types and techniques: towards a user's guide, Futures 38 (2006) 723–739.
[21] M. Höjer, L.-G. Mattsson, Determinism and backcasting in future studies, Futures 32 (2000) 613–634.
[22] K. Green, P. Vergragt, Towards sustainable households: a methodology for developing sustainable technological and social innovations, Futures 34 (2002)
381–400.
[23] J. Åkerman, M. Höjer, How much transport can the climate stand? — Sweden on a sustainable path in 2050, Energy Policy 34 (14) (2006) 1944–1957.
[24] K.L. Anderson, S.L. Mander, A. Bows, S. Shackley, P. Agnolucci, P. Ekins, The Tyndall decarbonisation scenarios — Part II: scenarios for a 60% CO2 reduction in
the UK, Energy Policy 36 (10) (2008) 3764–3773.
[25] S. Mander, A. Bows, K. Anderson, S. Shackley, P. Agnolucci, P. Ekins, The Tyndall decarbonisation scenarios — Part I: Development of a backcasting
methodology with stakeholder participation, Energy Policy 36 (10) (2008), pp. 3754–3763.
882 J. Wangel / Technological Forecasting & Social Change 78 (2011) 872–882

[26] K. Geurs, B. Van Wee, Backcasting as a tool to develop a sustainable transport scenario assuming emission reductions of 80–90 %, Innovation 13 (1) (2000)
47–62.
[27] S. Rayner, E. Malone, Human Choice and Climate Change. Volume 2: Resources and Technology, Battelle Press, Columbus, Ohio, 1998.
[28] E. Shove, L. Lutzenhiser, S. Guy, B. Hackett, H. Wilhite, Energy and Social Systems, in: S. Rayner, E. Malone (Eds.), Human Choice and Climate Change, Battelle
Press, Columbus, Ohio, 1998.
[29] J.B. Robinson, The proof of the pudding, Making energy efficiency work. Energy policy 19 (7) (1991) 631–645.
[30] W.N. Kaghan, G.C. Bowker, Out of machine age?: complexity, sociotechnical systems and actor network theory, Journal of Engineering and Technology
Management 18 (2001) 253–269.
[31] K. Gram-Hanssen, Consuming technologies e developing routines, Journal of Cleaner Production 16 (2008) 1181–1189.
[32] E. Shove, Comfort, Cleanliness and Convenience: the Social Organization of Normality, Berg, Oxford, 2003.
[33] A. Giddens, The Constitution of Society, California Press, 1984.
[34] P. Pierson, Increasing returns, path dependency, and the study of politics, American Political Science Review Vol. 94 (No.2) (June 2000).
[35] A. Bagheri, P. Hjorth, Planning for sustainable development: a paradigm shift towards a process-based approach, Sustainable Development 15 (2) (2007)
83–96.
[36] K. Larsen, U. Gunnarsson-Östling, Climate change scenarios and citizen-participation: mitigation and adaptation perspectives in constructing sustainable
futures, Habitat International Vol. 33 (Issue 3) (2009) 260–266.
[37] M. Höjer, A. Gullberg, R. Pettersson, Backcasting images of the future city — time and space for sustainable development in Stockholm, Technological
Forecasting and Social Change 78 (5) (2011) 819–834 (this issue).
[38] A. Wiek, C. Binder, R.W. Scholz, Functions of scenarios in transition processes, Futures 38 (2006) 740–766.
[39] K. van der Heijden, Scenarios: the Art of Strategic Conversation, Wiley, Chichester, 2007.
[40] P. Street, Scenario workshops: a participatory approach to sustainable urban living? Futures 29 (1997) 139–158.
[41] M. Sølgaard Bang, B. Lewakowski, Ö. Svane, Car Sharing and Low-Fossil Mobility. Master Students' Scenarios of a Situation of Opportunity in the Urban
Sustainable Development of Bromma, Stockholm. Report. TRITA-INFRA-FMS 2010:2, US AB, Stockholm, 2010.
[42] B. Lewakowski, M. Sølgaard Bang, Ö. Svane, ICT as a motor of transition. Master Students' Scenarios of a Situation of Opportunity in the Urban Sustainable
Development of Södermalm, Stockholm. Report. TRITA-INFRA-FMS 2010:3, US AB, Stockholm, 2010.
[43] J. Wangel, A meta-methodology for goal-based socio-technical scenarios — developing and testing methodologies. Forthcoming research report.
[44] K.M. Hillman, B.A. Sandén, Exploring technology paths: the development of alternative transport fuels in Sweden 2007–2020, Technological Forecasting and
Social Change 75 (8) (2007) 1279–1302.
[45] B. Flyvbjerg, Five misunderstandings about case-study research, Qualitative Enquiry vol 12 (2) (2004) 219–245.
[46] W. Krewitt, S. Simon, W. Graus, S. Teske, A. Zervos, O. Schäfer, The 2 °C scenario — a sustainable world energy perspective, Energy Policy 35 (10) (2007)
4969–4980.
[47] J. Åkerman, Sustainable air transport — on track in 2050, Transportation Research Part D: Transport and Environment 10 (2) (2005) 111–126.
[48] J.H.J.M. van der Graaf, H.A. Meester-Broertjes, W.A. Bruggeman, E.J. Vles, Sustainable technological development for urban-water cycles, Water Science and
Technology Vol. 35 (10) (1997) 213–220.
[49] W. McDowall, M. Eames, Towards a sustainable hydrogen economy: a multi-criteria sustainability appraisal of competing hydrogen futures, International
Journal of Hydrogen Energy 32 (18) (2007) 4611–4626.
[50] M. Eames, J. Egmose, Community foresight for urban sustainability: insights from the citizens science for sustainability (SuScit) project, Technological
Forecasting and Social Change 78 (5) (2011) 769–784 (this issue).
[51] D. Giurco, B. Cohen, E. Langham, M. Warnken, Backcasting energy futures using industrial ecology, (2011) Technological Forecasting & Social Change. (2011-this issue).
[52] M. Robèrt, P. Hultén, B. Frostell, Biofuels in the energy transition beyond peak oil. A macroscopic study of energy demand in the Stockholm transport system
2030, Energy 32 (11) (2007) 2089–2098.
[53] A. Carlsson-Kanyama, K.-H. Dreborg, H.C. Moll, D. Padovan, Participative backcasting: a tool for involving stakeholders in local sustainability planning, Futures
40 (1) (2008) 34–46.
[54] B. Truffer, J. Voss, K. Konrad, Mapping expectations for system transformations: lessons from sustainability foresight in German utility sectors, Technological
Forecasting and Social Change 75 (2008) 1360–1372.
[55] H.J. de Graaf, M.A.W. Noordervliet, C.J.M. Musters, G.R. de Snoo, Roadmap for interactive exploration of sustainable development opportunities: the use of
simple instruments in the complex setting of bottom-up processes in rural areas, Land Use Policy 26 (2) (2009) 295–307.
[56] J. Holmberg, K.-H. Robèrt, Backcasting — a framework for strategic planning, International Journal of Sustainable Development and World Ecology 7 (4)
(2000) 291–308.
[57] M. Robèrt, Backcasting and econometrics for sustainable planning: information technology and individual preferences of travel, Journal of Cleaner Production
13 (8) (2005) 841–851.
[58] K. Gomi, K. Shimada, Y. Matsuoka, M. Naito, Scenario study for a regional low-carbon society, Sustainability Science 2 (1) (2007) 121–131.
[59] C. Wallgren, M. Höjer, Eating energy — identifying possibilities for reduced energy use in the future food supply system, Energy Policy 37 (12) (2009)
5803–5813.
[60] P.J. Vergragt, H.S. Brown, Sustainable mobility: from technological innovation to societal learning, Journal of Cleaner Production 15 (11–12) (2007)
1104–1115.
[61] M. Hisschemöller, R. Bode, Institutionalized knowledge conflict in assessing the possible contributions of H2 to a sustainable energy system for the
Netherlands, International Journal of Hydrogen Energy 36 (1) (2011) 14–24.
[62] Å. Svenfelt, R. Engström, Ö. Svane, Decreasing energy use in buildings by 50% by 2050 — a backcasting study using stakeholder groups, Technological
Forecasting and Social Change 78 (5) (2011) 785–796 (this issue).
[63] L. Bohunovsky, J. Jäger, I. Omann, Participatory scenario development for integrated sustainability assessment. Regional Environmental Change (2010).
In press.
[64] B. Latour, Science in Action, Harvard University Press, 1987.
[65] M. Bucchi, F. Neresini, Science and Public Participation, in: E.J. Hackett, O. Amsterdamska, M. Lynchm, J. Wajcman (Eds.), The Handbook of Science and
Technology Studies, Third Edition, The MIT Press, Cambridge, Massachusetts, 2008.
[66] F. Bartiaux, Does environmental information overcome practice compartmentalisation and change consumers' behaviours? Journal of Cleaner Production 16
(11) (2008) 1170–1180.
[67] Fleximodo, Manual on urban ecology, urban mobility, urban information and communication and urban regeneration’, European Awareness Scenario
Workshop, European Commission DG XIII/, 1998.
[68] G. Bertrand, A. Michalski, L. Pench, Scenarios Europe 2010, European Commission Forward Studies Unit, Brussels, 1999.
[69] M. Sagoff, The allocation and distribution of resources” (1998) in J.S. Dryzek, D. Schlosberg, (Eds.) Debating the Earth. The Environmental Politics Reader.
(2005) Oxford University Press, Oxford.

Josefin Wangel is a PhD student at the Division of Environmental Strategies Research — fms, Royal Institute of Technology, Stockholm, Sweden. Her doctoral
studies focus on strategies, tools, and techniques for socio-technical transitions in the context of sustainable urban development. She holds BSc and MSc degrees
from Stockholm University.