Strengthening Gender Justice in a Just Transition: A Research Agenda Based on a Systematic Map of Gender in Coal Transitions

or climate change mitigation a rapid phase-out of fossil fuels such as coal is necessary. This has far-reaching gender-specific consequences. This paper presents a systematic map of the literature that examines the impact of historical coal phase-out processes on women and their role in these processes. The search process consists of screening 2,816 abstracts and reading 247 full-text studies. The analysis of the 73 publications ultimately included in the systematic map shows that past coal phase-outs meant, both opportunities (e.g. increased labour market participation) as well as burdens for women (e.g. double burden of job and household). It becomes clear that agency within coal transitions was also gendered. For example, it was difficult for women to gain access to union structures, which led them to organise themselves into grassroots movements. Our research shows that policies aiming for a just sustainability transition should always be explicitly gender-responsive. However, the impact of sustainability transitions on women's lives remains largely under-researched. Therefore, we propose a research agenda based on our findings containing six key issues that need to be addressed scientifically.

prices, will affect them particularly dramatically. 4 On the other hand, there is evidence that in the transport sector, the sustainability transition towards greater use of public transport and increased development of cycling and walking infrastructure benefits women; men use cars more often, while women are more likely to walk, cycle and use public transport (Spitzner et al. 2020). Such examples show that a better understanding of the gender dynamics of low-carbon transitions is very much needed (Lieu et al. 2020). The UN has also recognised the urgency of considering the gender dimension of climate impacts and has included gender equality as a goal in their Sustainable Development Goals. 5 However, research looking at the social outcome of low-carbon transitions is often limited to calculating net employment effects (García-García, Carpintero, and Buendía 2020). As most workers in the coal industry are male and direct job losses mainly affect them (e.g. in 2017, 78.4% of the workforce in the US coal industry was male; National Association of State Energy Officials and Energy Futures Initiative 2018), this emphasis on employment effects implies an indirect focus on the coal transition's effect on men. "While the ILO states that gender is an important variable for just energy transition analysis, as it could widen inequalities, gender worries are hardly present in the research" (García-García, Carpintero, and Buendía 2020, p. 12).
Our paper is an attempt to start filling in this research gap by systematically examining the impact of low-carbon transitions on women and the role that women played -referred to as agency -within these transitions using coal transitions as historical case studies. In this paper, we answer the research question: What evidence exists about the effects of coal transitions on women in contrast to men, and about female agency during coal transitions? Based on our research results, we develop a research agenda that shows where research is still lacking concerning the gendered aspects of low-carbon transitions and, on this basis, how researchers and policymakers can include gender aspects more properly in policies aimed to design a just transition.
We focus on the example of coal transitions, because they are essential building blocks on the way to a sustainable economy. Coal is still responsible for more than 40% of global CO2 emissions. 6 There will only be a chance to comply with the Paris Agreement if a global coal phase-out is achieved in a timely manner (Yanguas Parra, Hauenstein, and Oei 2021;IEA 2021). Therefore, there are many more coal transitions ahead of us and these must incorporate gender implications from the outset if a socially successful transition is to be achieved. At the same time, some countries, especially in the Global North, have already completed or almost completed their coal phase-out processes and are therefore very well suited as research subjects.
To collect what scientific work has been done on the topic so far, we conducted a systematic map ) of the existing research on the nexus between gender and coal transitions across a wide range of disciplines, such as anthropology, sociology and 4 Genanet. "Energiewende geschlechtergerecht": https://www.genanet.de/themen/energie (last accessed 27 April 2021) 5 United Nations. "Goal 5: Achieve Gender Equality and Empower all Women and Girls": https://sdgs.un.org/goals/goal5 (last accessed 15 June 2021) 6 International Energy Agency. "CO2 emissions by energy source, World 1990-2018 in percent": https://www.iea.org/data-andstatistics/data-browser?country=WORLD&fuel=CO2%20emissions&indicator=CO2BySource (last accessed 13 July 2021) economics. Our search strategy consisted of a database search in Scopus, EBSCO, Web of Science and ProQuest, a grey literature search, and a call for relevant literature in the scientific community; this was supplemented by snowballing the references of relevant publications. The literature search, and the subsequent full-text reading of 247 publications, yielded 73 relevant studies, mostly focusing on countries in the Global North (mainly the UK, the USA and Germany). We systematically extracted all information in the publications according to what impact historical coal transitions had on women, and what role women played in the transition. Based on our results, we show where more scientific knowledge is still needed for policymakers to be able to make just transition policies that adequately account for varying gendered outcomes.
As results, we found that across country contexts, the long-term employment effects of past coal transitions were that the (mostly male) miners in coal regions increasingly left the labour force due to early retirement or redundancy, while women increasingly sought employment -giving them a degree of financial independence. However, many of them had to take up precarious jobs which were mainly located in the service sector and provided little financial and planning security. Future research should ask about policies designed to improve working conditions in the service sector and create well-paid jobs for all genders in the Green Economy sectors. In addition, many studies reported an increased total workload for women, as their domestic responsibilities remained, but a new responsibility as wage earners was added. One important area of research to inform policymakers is the availability of care services in carbon-intensive regions and the extent to which missing support infrastructure prevents women from planning their own careers.
Turning towards women's agency, our analysis showed that women active in the political struggle around various historical coal transitions tended to organise themselves in less institutionalised ways compared to men (partly because institutions such as unions were difficult for women to enter). Their activism, for example in the miners' strike in the UK or in anti-coal activism opposing Mountain-Top Removal (MTR) in the USA, usually took place in self-organised grassroots movements or through involvement in community work. Future research should investigate the main interests, needs and concerns of the women affected by the transition, and the kinds of activities and forms of organisation they choose. First, this would make women's activism more visible, and second it would inform state institutions how to make the transition process more inclusive. Furthermore, it is important to investigate how women's access to resources, such as adequate funding or access to policymakers and institutions, could be improved so that they are able to better integrate their interests into the political process.
The remainder of the paper is structured as follows: Section 2 presents our methodological approachthe systematic map. Section 3 summarises the information we have extracted from the publications.
Based on those results, Section 4 presents our research agenda on how future research on just transition pathways can more comprehensively take into account gender dimensions. Section 5 concludes.

Methodology -Systematic Map
We answer our research question "what evidence exists about the effects of coal transitions on women in difference to men and about female agency within coal transitions?" using a systematic map. A systematic map is well suited to answer this question, since it is designed to collect, describe and catalogue existing evidence on a specific topic in a comprehensive and repeatable way. In doing so, it provides information about the current scientific state of the field of interest, such as the amount of existing literature as well as geographical foci, methodological approaches, and scientific disciplines. A systematic map can also help to identify research gaps or controversies in the existing literature James, Randall, and Haddaway 2016). The results of our systematic map therefore provide a good basis for developing a research agenda.
Starting our research process, we developed a systematic map protocol, which included the research idea, question and aim, inclusion and exclusion criteria, and search strategy. After having received feedback on it from experts in the field, it became the basis of our methodological procedure. In the following, we describe the four stages of our methodological approach.

Development of the Search String
Our research question includes the two basic terms: coal transition and agency. Before we present our methodological approach, it is important to provide a definition of these two terms and present our concept of gender, which is also essential to our work.
We define a coal transition as a (regional) decline in the extraction or combustion of coal, or a substantial loss of jobs in the coal industry. This definition allows us to include smaller, regional coal transitions and honours the social significance they bear. We define the concept of agency as "the property or capacity of actors to make things happen" (Coole 2010, 11). We conceptualise actors not merely as individuals, separate from their society, but as unique parts of a collective, who make choices based on the structurally presupposed rationales surrounding them, rather than simply autonomously (B. Davies 1991, 42-46).
In our study, we use gender as an analytical tool to make socioeconomic inequalities that affect women visible. 7 This is a very simplified approach to gender, but we consider it appropriate for our research interest. It would be beyond the scope of this paper to comprehensively present the findings of gender research (for a more elaborate and differentiated discussion of gender, see for example Butler 1999;Alaimo and Hekman 2008). Being assigned to the category "woman" generally goes along with gendered norms that lead to constraints, while being assigned to the category "man" commonly brings more privileges (Young 2002). One of the assumptions that goes along with the category "woman" is being a caregiver and socially oriented. Thus, women are expected to do most of the unpaid care work, both related to the physical as well as the emotional well-being of people. Care work is essential for the functioning of a society, however its value is barely recognized, neither in economic terms nor as an high-skilled and energy-intensive activity. The time and energy women invest in care work is no longer available for other activities, which could bring higher status, therefore limiting their own life chances in comparison to men. If a person does not fulfil gendered norms, this behaviour is accompanied by social sanctions. It is essential to note that women are not a homogeneous group, but have very different life situations depending e.g. on age, socioeconomic background, place of origin and race. Still, women have on average less access to resources (e.g. finances, influence on decision-making) than men and consequently different opportunities in life. Women's social positioning not only influences how transitions affect them, but also their agency within coal transitions, for example regarding their political articulation, their forms of organisation or their representation in decision-making bodies.
In an iterative process, we developed a search string based on these three central terms. A search string is designed to find as many relevant publications as possible when using it in databases. It consists of several categories including synonyms or similar terms. We used one category for coal (e.g. lignite, mining), one for gender (e.g. women, woman) and one for transition & agency (see Table 1). Transition and agency were grouped into one category to ensure that a transition moment is either directly covered or questions of agency are addressed, which might point towards a transition in process (e.g. strike, activis*). We collected a broad range of relevant search terms for all three categories, which allows us to detect records across disciplines and times of publication -these might differ starkly in their terminology while still addressing the same topics. A preliminary search was conducted to evaluate the relevance of the terms, and a benchmark test with five relevant publications was established (test list). Text mining software was used to analyse the results of the test search and identify further relevant search terms.
With this process, we came up with a comprehensive as well as consistent search string (see Table 1).

Transition and Agency
Coal Gender transition* coal gender* transformation* lignite* woman* change* mining women* Reform anthracite* (anthracites) female* closure* miner mother* decreas* miners femini* declin* coalfield* *wives collaps* coalmin* *wife crises coalface crisis AND NOT "data mining" strik* AND NOT "text mining" resistance protest* agency activis* oppos* At first, unsystematic pre-tests allowed us to grasp an idea of the scope of existing literature and the exact inclusion and exclusion criteria needed to identify relevant literature. Following the PICO (Population, Intervention, Comparator 8 , Outcome) format, we identified the key inclusion and exclusion criteria (O'Connor, Green, and Higgins 2008): Population encompasses people living in coal-intensive regions worldwide, Interventions includes coal transitions as defined above, and Outcome relates to gender-related social, political, economic and demographic effects, as well as gendered agency within these transitions. To gather a comprehensive literature base, books, reports, conference papers, theses and working papers as well as journal articles were included in the systematic map.
Combining this with the aforementioned PICO categories, we identified the following inclusion/exclusion criteria: Primarily, we retrieved full texts through access provided by Technische Universität Berlin (TU Berlin). In the next step, we drew on library subscriptions at Freie Universität Berlin and Humboldt-Universität zu Berlin. Where full texts were still missing, we requested articles directly from corresponding authors or made requests through Researchgate.org. 10 JStor would also be relevant, but could not be used due to its significant restrictions on the length of search strings. our research question; they were thus included in the systematic map. We complemented our database search with a three-step approach to searching publications not available in our databases: 1) The search for grey literature consisted of two parts (Mahood et al. 2014) and rendered nine publications for inclusion in the map: • The use of two grey literature indexes operated by the Institute for Scientific and Technical Information (France) and by the European Institute for Gender Equality.
• A search using Google Scholar using a shortened search string, screening the first 100 results. 12 2) To ask for additional literature recommendations, we established contact with the research community via email and social media. From this, eight additional records were added to the map.
3) Furthermore, we screened all bibliographies of previously included texts for potentially relevant publications (snowballing); this resulted in twelve additional publications.
A total of 25 relevant records were identified through the non-database searches, leading to total of 68 publications. Publications were split into multiple studies if they a) treated coal transitions in different nations or b) various chapters of one book were found to be relevant. Ultimately, 73 studies were captured systematically in the codebook. For our database search, we only used the above search string in English. We developed and tested a search string in German, but it did not yield any relevant results. We refrained from testing search strings 12 Additional results did not prove to be relevant. in other languages. However, the search in English and some literature recommendations included publications in German or Spanish which were included in our analysis. Our search string in English is certainly one reason why the publications in our systematic map predominantly deal with coal transitions in the Global North and especially in the UK and USA. It is possible that a search string in Spanish, for example, would have yielded more results on countries in the Global South (e.g. Chile has experienced a decline in coal production in recent years 13 ). Moreover, it is very likely that scientific institutions in the Global North have more resources to facilitate research on this topic.

Data-Coding Strategy
We developed a comprehensive codebook in a multi-staged process including a test-coding phase. Table 3 presents a list of all categories and subcategories of the codebook. In addition to containing general information about the studies and coal transitions at hand (metadata), the codebook is split into two main sections to reflect the twofold research question. The first section includes information on gendered outcomes of coal transitions in five subcategories: ranging from the individual level, to the household, to the community, as well as information on gendered effects on job markets and the political sphere. The second part of the codebook focuses on female agency in coal transitions, and is split into different groups of female stakeholders. Moreover, we collected information on climate and environmental aspects -which ultimately proved to be disregarded in the literature we read -as well as on concepts of gender, and reported intersectionality with class, race and more. The subcategories featured a mix of fixed dropdown menus and open answer fields, so that the collected data was both highly detailed and consistent among the coding team. After we had read 10% of the texts from the database search, we discussed inconsistencies in the coding strategies, and identified any missing or unclear categories. The codebook was adjusted slightly to ensure it included all relevant information.

Gender in Historical Coal Transitions
The 73 studies recorded in the systematic map covered gender aspects of coal transitions in eight countries: the USA, Colombia, Germany, Romania, Spain, Turkey, the UK and Japan. Figure 2 provides an overview of the historical coal-production developments in the respective countries and summarises the main gender-related findings detailed in the following sections. It must be noted that Turkey and Colombia did not undergo a national or lasting coal transition. In the case of Turkey, the reduction in coal production remained confined to the region of Zonguldak near the Black Sea, while the decline in coal production in Colombia was only a temporary phenomenon in 2014. Nevertheless, both countries experienced a substantial decline in coal production and thereby met our inclusion criteria. Before we present our results, we would like to point out that the coal transitions in these various countries have played an important role in explaining the developments we describe below. However, there have been other social, political and economic changes (e.g. increasing neoliberalisation in the 1980s and 1990s in the USA (Smith 2015) and the UK (Dalingwater 2018)) that have influenced the situation of women, and therefore there is not always a clear causality between the coal phase-outs and the developments described.

Figure 2: Transitions' overview and main reported gender aspects per country
Source: Own depiction, based on BP (2020) and information collected in the codebook.

Background Information on Studies and Coal Transitions
Most of the studies addressed transitions in the Global North (see Figure  Arango, Flórez, and Olarte-Delgado 2019) as well as a study by Botta (2019), which covers more than one country. In our research, countries of the Global North are disproportionately represented due to our focus on historical coal transitions. Most studies stem from sociology (20/73), followed by history, geography and political science (see Figure 3, b). The majority of studies conducted were qualitative (51/73, see Figure 3, c). 48 out of the 73 studies were published in journals. Another 15 were book chapters and the rest were grey literature. The vast majority of studies were written by female authors, 14 with 31 (42%) written by single female authors, 21 (29%) by multiple female authors, and only 16 from single or multiple male authors (22%). An additional five studies (7%) appeared to have mixed authorship.

Figure 3: Country distribution, scientific disciplines & methodologies of the studies
Source: Own depiction, global map from d-maps 15 The years of publication span from 1984 to 2020, covering transitions starting in Japan after the First World War (Mathias 1993) up to today. Studies on the effects of coal transitions in the UK and the USA, as well as on the UK's miners' strikes have been published throughout the whole period (see Figure 4, a).
Turning towards the focus of inquiry (see Figure 4, b), the miners' strikes in the USA and especially the which started to be published around 15 years ago (Barry 2008;2012;Bell and Braun 2010;McNeil 2011). Socioeconomic effects as well as outcomes on community level were also frequently discussed in the studies; this will be described in detail in Section 3.2.

Figure 4: Year of publication & focus of inquiry
Source: Own depiction.
Gender interacts with other social statuses and identities, such as class, race, age or religion. Together, they strongly influence a person's position in society and, consequently, the discrimination they face or privileges they enjoy. Introduced by Kimberlé Crenshaw (1991) and further developed by many feminist scholars, the analytical framework of intersectionality has been widely used to describe the overlapping advantages and disadvantages that go in hand with the above-mentioned statuses and identities. Figure 5 shows which categories were considered together with gender in the studies. It also shows whether the categories were taken into account comprehensively as a core concept in the study and thus play an overriding role alongside gender, or whether they were treated secondarily to gender as a side concept or were only mentioned in passing. Together with gender, class was an analytical lens in 46 out of the 73 studies, making it a highly relevant intersectional category for coal transitions. Especially in the miners' strikes in the UK 1983-1984, class was essential to women's identities: "Class identity remains intact as a greater 'us', but women are now included within that" (Stephenson and Spence 2013, 230). Also beyond the UK, class was a relevant category in coalmining communities across national borders and emerged in studies about almost all countries represented in the map (e.g. Further relevant intersectional categories are race and age, both of which were addressed in nine out of the 73 studies (see Figure 5). Most studies were concerned with the experiences of "white working class women" both for participation in miners' strikes as well as the transition's outcomes (e.g. Beckwith 1996;Seitz 1998;Schell and Silva 2020). However, the situation for women of colour was also treated in some texts, with Joyce Barry noting that "race has been a major framing device for Environmental Justice thought and activism, and [in anti-MTR activism,] people of colour have been at the forefront of this movement" (Barry 2012, 47). Age was also a relevant factor in the coal transitions. A women's age entails different child care duties, working experiences and educational opportunities, and therefore leads to different coping strategies and possibilities in transition moments (e.g. Ali 1986;Jung 2006).

Figure 5: Intersectionality in the studies
Source: Own depiction.

Reported Transition Effects on Women
The following chapter reports the transition effects on women we found in the literature. The evidence was grouped into four areas of impact: gendered labour market effects, and effects at the community, household and individual levels.

Gendered Labour Market Effects
In wide-scale transition moments, the effects on local job markets hold particular relevance. In accordance with this, 40 studies reported outcomes within the primary job market -in our case the coal industry -of which 24 reported job losses in general (usually mentioning that these mining jobs were largely occupied by men, e.g. Dublin and Licht 2000), 8 specifically reported job losses for men and 2 reported reduced employment for women in the coal industry (Mathias 1993;Clemens and Rauhut 1999). Furthermore, various studies emphasise the stark gender segregation of occupations with men working in heavy industry while women predominantly held factory jobs or performed care-based activities (e.g. Dublin and Licht 2000;Oberhauser, Turnage, and Smith 1999). Before the transitions gender differences in tasks and skills were rigid, as some of the coal regions had little other industry and thus scarce alternative employment opportunities for women, leading them to mainly perform unpaid work (e.g. Barry 2001;Spence and Stephenson 2009). In sum, in the course of the transitions the predominantly male workers were subject to massive layoffs, while the impact on women occurred on the secondary and tertiary job markets. In order to compensate for the decrease in household income, women increasingly took up paid jobs (e.g. Beatty 2016). 35 studies reported outcomes on the secondary and tertiary job market. 20 of those 35 showed increased female employment rates in the aftermath of the transition (e.g. Maggard 1994a;Bennett 2004;Phillips 2018). Additionally, seven studies reported job gains in general, which -again due to gendered segregation of occupations -also meant an increase in female employment (e.g. Oberhauser 1993; Measham and Allen 1994;Miewald and McCann 2004). The conditions in these jobs are usually reported to be more precarious, low-paid and insecure compared to the relatively secure mining jobs (e.g. Oberhauser 1995; Spence 1998; Miewald and McCann 2004;Phillips 2018). In contrast to the above findings, one study for the UK reported that after the mines closed, women lost employment in the manufacturing service sector, due to male ex-mine workers pushing them out of the job market (Aragón et al. 2018; similar but less substantiated findings in Arango et al. 2019).

Effects on Community Level
Since mining was often a large part of community life, social changes from the transitions were vast. 56 of the 73 studies in the systematic map reported effects at the community level (see Figure 6, a).
The most prevalent effect of a coal transition was the deterioration of the corresponding locality due to outmigration. This was reported in 29% of the studies we examined and we found different gendered outmigration patterns. For example, in Appalachia (USA) in the 1980s and 1990s, men migrated in high numbers in search of employment in construction or factory work (Oberhauser 1995). In contrast, in Lusatia, former East Germany, it was mostly young women who left after the fall of the Iron Curtain (which was accompanied by a huge decline of the coal industry) to find employment outside their home region (Jacobsen and Winkler 2011). However, outmigration already affected some of the coal mining areas before the transitions took place. For example, Barry (2001) reported in a study focused on the USA that coalmining operations were very loud and environmentally destructive, causing many people to move away; this in turn left mostly empty villages with deteriorating social and economic infrastructure.

Figure 6: Studies reporting transition effects on women
The second most affected area is community activity. Here we found nearly equal reports on increases and decreases in community activity due to the coal transition, with eight studies reporting an increase while seven studies investigated a decrease (see Figure 6, a). Often, an increase of community activity was reported during and after miners' strikes where women were involved (e.g. Sanz Hernández and López Rodríguez 2017). People developed a stronger sense of community as they collectively (especially women) made the strike possible, for example, by supporting each other financially or with housework (e.g. Miller 1985;Measham and Allen 1994;Spence and Stephenson 2009;Seitz 1998). As one study on the UK miners' strikes put it: "The discrepancies between activism and ideological support for the strike confirm that people were active in the dispute as much because they had the time, the material resources and the opportunities to allow the possibility of giving support and wanted to help friends and neighbours in the close-knit community, as because of their commitment to the strike itself." (Measham and Allen 1994, 38) Other studies argued that the necessity for women to become wage earners put heavy strains on community involvement, leading to reported decreases of community activity. This development was attributed to women being traditionally in charge of organising community events and maintaining social networks that they subsequently had no time for after entering the job market (e.g. Miewald and McCann 2004). Furthermore, de-unionisation throughout the transition processes and conflicts during the strike weakened the community bond among former miners and led to further defragmentation within communities (Kideckel 2004;Smith 2015;Spence 1998). Hence, the development of social life remained unique to each community, and there was no unifying trend in social life accompanying these transitions.
13% of the studies reporting effects at the community level mentioned direct effects on the mining communities' identities. Coal communities often followed an especially patriarchal pattern. According to Maggard (1994a), this was connected to the coal industry itself, as gender inequality was exacerbated by the arrival of the industry in Appalachia. A historical study of the extent to which the arrival of the coal industry exacerbated gender segregation and by what mechanisms would be interesting, but is beyond the scope of our research. Multiple studies reported that the transition away from coal weakened these traditional structures, as women broke out of their assigned sphere of the private (e.g. Spence and Stephenson 2007;2009;Miller 1985) (see below). In many cases, the image of mining communities became heavily romanticised after the coal transition, leading to the founding of institutions and clubs to preserve a long-gone mining identity and tradition (Dawson 2000). Spence and Stephenson (2009) argued that the strong patriarchal tradition also influenced the way women were portrayed in British strikes. They showed that women were presented as only becoming politically active with the miners' strike. However, they found that this misrepresented women's participation in society before the transition. Many of them were already active before, for example, in community work. Mining as an identity was also affected in other ways, however. In the case of Romania, the transition destroyed trust among community members, by pitting active miners that broke the strike (for various, often financial, reasons) against their former colleagues, who were then out of work (Kideckel 2004).

Effects at the Household Level
While the two levels of affectedness elaborated on above reflect wide-ranging effects, transition moments also change more small-scale structures, such as household dynamics. For an overview of the different effects found in 50 of our 73 studies, see Figure 6, b.
15 out of 50 (30%) studies reported a change in the gendered division of labour, making it the most prominent effect at the household level. Some studies on households in former coal communities in the UK reported that the new employment situation changed the division of labour, as men started to spend more time at home and women went out to work. As one study put it, "research suggest[s] subtle changes regarding the intersection of class and gender regimes in households that shape this former coalfield. [...] there was growing intolerance of men doing nothing in the way of childcare and housework" (Bennett 2015(Bennett , 1296. Four studies reported a general liberalisation of gender roles (Miewald and McCann 2004;Stephenson and Spence 2013;Shaw and Mundy 2005;Miller 1985 (Miewald andMcCann 2004, 1055-57). However there is no clear pattern as other studies also looking at the UK and the USA argued that the patriarchal gender division remained largely the same during and after the transition, often leaving the domestic work to women in addition to their paid jobs (Measham and Allen 1994;Schell and Silva 2020). Various studies stated that the total workload for women increased, as their domestic tasks remained, but a new responsibility as wage earner was added (e.g. Measham and Allen 1994;Dublin and Licht 2016;Williams 1999). In these cases domestic tasks did not seem to have decreased substantially, as men started to be unemployed.
Williams noted, "if a man did help out around the house, his assistance was given 'behind closed doors' and 'covertly in order to avoid embarrassment, and what might be interpreted as an affront to his manhood'." (Williams 1999, 69).
The second most reported outcome of coal transitions at the household level were domestic conflicts (e.g. Kideckel 2004;Barry 2001;Miller 1985). Four of the studies we evaluated even mentioned increases in domestic violence during the transition (Maggard 1994a;1994b;Kideckel 2004;Bennett 2004). While the causes of domestic conflicts are manifold, the juxtaposition between women's traditional responsibilities in households and the changing nature of women's social positions through employment or political engagement and the resulting erosion of the traditional gendered order has to be seen as one of them (Dublin and Licht 2000;Kideckel 2004). One study mentioned that after the strikes against the mine closures in the UK were over, women were expected to return to their domestic roles as carers, which many refused to do (Shaw and Mundy 2005).

Effects at the Individual Level
Effects on women at an individual level were reported in 48 studies (see Figure 6, c). As already mentioned in the section on Effects on Community Level, restructuring due to a coal transition had significant effects on the inhabitants' identities. 15 out of the 48 studies addressing the impact on women at the individual level focused on how female identity changed during the transition (e.g. Measham and Allen 1994;Clemens and Rauhut 1999;R. Davies 2010). Since identity is a broad concept, the aspects of female identity treated in each study vary greatly. In the UK miners' strikes, for example, traditional female gender norms changed abruptly. This forced a change in women's social identity, and how they were seen within their community (Spence 1998). As Peter put it, "Instead of occupying a prominent position only within the female community hierarchy, their role in the strike was acknowledged by their recognition as important individuals by both women and men" (1988,182). Another study, focusing on coping strategies of women in the face of economic challenges in Appalachia, recorded a fundamental change in the way women saw themselves through "increased self-esteem, confidence, and [in] gaining some financial independence" (Oberhauser 1995, 64) by generating income, which in turn gave them a perception of themselves as role models for their daughters. Sometimes identity was not simply remodelled; one study from Spain reported a loss of identity for women based on class and their social status as "miners' wives", who due to the economic restructuring had to replace their role as the reproducing and caring counterpart to the male miners with new roles, based on a new sense of self.
Young women in particular, recognising that the return to coal would only work in the short term, started to demand a more future-oriented perspective, and a search for alternative ways to sustain their communities (Sanz Hernández and López Rodríguez 2017).
We can identify the clear pattern across studies and transitions that more women entered the labour market (see section on gendered labour market effects). Regarding the gendered division of labour there is not such a clear pattern -some studies reported an increasing liberalisation of gender roles after the coal transition while others reported that the paid additionally to the unpaid labour led to a double burden for many women (see section on the effects on household level). That the division of labour remained largely the same is likely to be due to the gender norms and expectations in the coal transition areas dealt with in 12 out of 48 studies (see Figure 6, c). The highly gendered societal demands placed on women as care-givers and community upholders often remained (e.g. Bell, Fitzgerald, and York 2019; Marshall 2008). These norms had been internalised over centuries, explaining cases in the UK where women remained content with traditional gender norms and avidly supported preserving the pretransition social order. "Many [women of the support groups in UK] were keen to disassociate themselves from the more militant feminist groups who also became involved in strike support, and preferred to be seen as ordinary women from mining communities. They did not want to change their status within their communities; they were fighting for the stability of the world they knew" (R. Davies 2010, 246-47).

Female Agency in Historical Coal Transitions
Just as women's social positioning influences outcomes, agency itself is also gendered within coal transitions. This is reflected in women's forms of organisation, their representation in decision-making bodies and their visibility within coal transitions. Based on the definition of the term "agency" laid out in Chapter 2, we systematically analysed the information on female agency presented in the studies. 26 out of the 73 studies treated agency as their main or sole focus, showing that female agency was a relevant aspect of coal transitions. As the following section shows, female agency both shapes, and is shaped by, transition processes.
Overall, 63% of the 26 studies reported that female actors were moderately or strongly against the coal transition in which they were involved, while 24% studies found women moderately or strongly in favour of it (see Figure 7). The other 13% reported on female agency without explicitly stating their opinion. Sutcliffe-Braithwaite and Thomlinson 2018). Their activities ranged from collecting strike money, picketing or establishing community kitchens, to more radical forms of action, such as occupation.
Miners' support groups in the UK maintained close contacts with the National Union of Miners (NUM), 16 Translation from Spanish: Saving the mine is saving the mining villages.
although the form and intensity of cooperation were contentious among the women (Sutcliffe-Braithwaite and Thomlinson 2018). Most of the women's groups distanced themselves from feminist movements, identifying themselves as miners' relatives or working-class women rather than as feminists (e.g. Beckwith 1996;Seitz 1998;Spence 1998). However, some texts mentioned ties to contemporary feminist movements or stated that the miners' wives were at least influenced by them (Miller 1985  that meet their multiple, yet overlapping, aspirations and responsibilities" (Miewald andMcCann 2004, 1060) or chose this form of organisation after the initial protests against mine closures were over (Phillips 2018).

Figure 8: Political spheres of female agency
Source: Own depiction.
Various authors elaborated on the question of how increased female visibility and activism during coal transitions affected gender relations. Concerning the transition in Appalachia, Dublin and Licht (2000) found that "gender roles were both reinforced and reconfigured during the collapse of the anthracite region's economy. The system of separate spheres for men and women persisted and even contributed to family survival, although certain role reversals occurred" (85-86) and Maggard (1994b, 20) noted that traditional understandings of gender were disrupted and women who had not worked previously took up education and job training after participating in the strikes. Sanz Hernández and López Rodríguez (2017) stated that in Spain, the women moved from the private to the public sphere and from a passive role to an active one during their participation in the coal strikes. This changed their overall position in their communities. In the UK, female strike participation was reported to have had only a temporary effect on gender relations (Beckwith 1996;Waddington et al. 1991 cited in Spence andStephenson 2007).
Phillips (2018,46) on the contrary stated that "[m]any women remained active politically after the strike […] and acquired formal educational qualifications", while still acknowledging that this was a slow, procedural process.

Discussion and Research Agenda
While there has been extensive academic research on coal transitions in general, there has only been a limited amount of academic literature from a gender perspective. Nevertheless, the existing research shows that women have been affected differently by historical coal phase-outs than men, and that women have chosen their own forms of participation in the transition processes.
A central question is if these results can be transferred to today's coal phase-out processes. Many socioeconomic conditions have changed. In general, climate change considerations, as prevalent today, were not the reason for the historical coal transitions analysed in this paper; these coal phase-outs were mainly caused by economic factors. Moreover, the transitions were hardly cushioned by social policy.
This could be one of the reasons why these transition processes were mainly opposed by the female stakeholders we analysed and why the tenor of the studies was mostly negative. Today, the climate crisis makes the coal phase-out urgent. Furthermore, there is a growing recognition of the need to support coal regions to develop alternative industries (e.g. through the European Just Transitions Fund 17 ).
Our analysis of historical coal transitions shows that gender had a major impact on how a person was affected by the transition and to what extent she was involved in decision making. Even though many initial conditions have changed, we can conclude from our analysis and from the fact that gender is still very powerful in assigning men and women different positions in most areas of life, that gender is also relevant for today's sustainability transitions (Spitzner et al. 2020). We see a considerable lack of both quantitative and qualitative gender-aggregated data, and a need for more research to fully understand the gender dimension of today's sustainability transitions. Furthermore, novel policy instruments are needed to cushion the impacts of the transition on women, and consideration should be given to offering them more forms of participation besides their involvement in social dialogue 18 .
Only one of the ILO's just transition guidelines directly addresses the aspect of gender equality.
However, we argue that gender should not be treated as just one aspect of the transition, but that the gender dimension runs through all aspects of transitions and should therefore be seen as an integral part of all the guidelines. Based on our findings on women in historical coal transitions, we have therefore developed a research agenda addressing all of the ILO's guidelines. This research agenda should 17 European Commission. "Just Transition funding sources": https://ec.europa.eu/info/strategy/priorities-2019-2024/europeangreen-deal/actions-being-taken-eu/just-transition-mechanism/just-transition-funding-sources_en (last accessed 06 July 2021) 18 "Social dialogues include all types of negotiation, consultation or simply exchange of information between, or among, representatives of governments, employers and workers, on issues of common interest relating to economic and social policy" In: International Labour Organization. "Social dialogue": https://www.ilo.org/ifpdial/areas-of-work/social-dialogue/lang-en/index.htm)%20%20a (last accessed 01 July 2021) facilitate a systematic scientific exploration of the gender perspective on today's just transitions away from carbon-intensive industries. Such research would contribute significantly to shaping gendersensitive transition policies as well as gender-aware forms of participation. We recognize that meeting the ILO criteria alone would not lead to a just transition in the broader sense, as the ILO perspective merely focuses on carbon-intensive regions, and fails to address other important aspects of just transitions, such as energy and climate justice (Carley and Konisky 2020). Nonetheless, meeting these criteria would be a first step towards a gender just transition.

a) Participation
The ILO's first principle of a just transition refers to the participation of all relevant stakeholders to ensure social consensus on the goals and pathways to a sustainability transition. This procedural dimension of justice is indeed key when it comes to the question of distributing the costs and benefits of transition processes in a fair way. Our research shows that women in the coal regions have borne a relevant share of the transition costs and had, for example, economic self-interests in transitions. However, due to their limited access to institutionalised decision-making processes and resources, their interests have not fully been taken into account.
To fulfil this ILO principle in future transition processes, greater efforts should be made to include the interests of women (and other marginalised groups). However, to fully integrate gender perspectives into transition processes, it is not enough to involve more women in decision-making processes. To develop gender-sensitive policies, the involvement of actors who are knowledgeable about gender aspects is necessary (Kronsell 2013). Future research on gender and sustainability transitions will be crucial for generating the required knowledge. This includes questions such as the following: What are the main interests, needs and concerns of women affected by transitions? Which type of activities and form of organisation do politically active women choose? What issues do they prioritize? Do they have access to resources (e.g. financial means or access to policymakers)? Has there been a shift in women's political activity in recent decades away from grassroots structures towards more formalised types of organisation because of improved access to such structures (e.g. unions, parties)? If not, how can the work of informally organised women be better transmitted to political bodies that shape transformation processes? Are there still barriers for women to join formal political organisations, such as unions or parties?

b) Rights at Work
Policies must respect, promote and bring about fundamental rights at work. This is the ILO's second principle. From a gender perspective, this principle is of great interest. Our research has shown that during and after the transition, more women entered the labour market, mainly taking up precarious and low-paid jobs in the service sector. Such a shift from industrial jobs to a more service-oriented regional economy can be expected in many future sustainability transitions. A fundamental question in this context is how to improve working conditions in the service sector. Another question concerns how women can be involved in the planning process at an early stage, before the transformation process is in full swing (see Guideline a) above), to ensure that they have time to participate in training programmes, for example.

c) Gender Equality
The third ILO principle directly addresses gender within sustainability transitions, and states that policies and programmes must take into account the significant gender dimension of many environmental challenges and opportunities. We hope our research will help broaden the focus of this important point: not only environmental degradation, but the transitions itself and its consequences affect women differently than men. Historical coal transitions illustrate that the double burden of (usually unpaid) care

d) Coherence
Energy policy and policymaking processes are dominated by a technocratic perspective. Economic perspectives are limited to energy prices, jobs and labour, while social aspects are often neglected entirely (Lieu et al. 2020;García-García, Carpintero, and Buendía 2020). This one-sidedness is also reflected in the ILO's mention of coherence, which states that policies in all areas should focus on the creation of an enabling environment for "enterprises, workers, investors and consumers to embrace and drive the transition towards environmentally sustainable and inclusive economies and societies." (ILO 2015, 6). What about the citizens, the community members, those who have left the workforce, or future generations? Coal phase-out processes have multiple social as well as economic impacts on these groups. Existing research mentions, for example, increased intra-household conflict, domestic violence, social tensions in communities, and outmigration. However, the scientific evidence is particularly thin concerning these aspects, and much more research is needed to understand what impacts (in addition to economic ones) low-carbon transitions have on communities and families, as well as how these impacts could be prevented or compensated for. On the other hand, research also provides examples of positive effects of coal transitions on women (intra-household empowerment, shifting traditional role models). The loss of (regional) identity is another aspect that is often discussed in the context of coal transitions. However, we did not find a significant amount of literature addressing this aspect from a female perspective. Future research could explore the question of whether women and men see the fossil-fuel industry as equally important for a given region. From studies on the attitudes of men and women to the climate crisis, it can generally be said that women on average perceive the climate crisis as a greater threat (Spitzner et al. 2020). Therefore, one research hypothesis would be that they also rate the coal phase-out as more urgent in order to overcome the climate crisis. Further research questions could be: Is the decline of the fossil-fuel industry accompanied by a loss of identity? Do women and men view it differently? It would also be of interest to better understand what empowering moments coal transitions have on women, and how this could be considered in policymaking processes.

e) Employment
The ILO's employment dimension emphasises the need to create a just transition for all. So far, however, transition policies have focused on compensating workers and communities directly affected by transitions away from fossil fuels (Piggot et al. 2019

Conclusion
The low-carbon transition that needs to be implemented in all sectors of the economy to mitigate the catastrophic effects of unlimited global warming affects different social groups differently and creates unequal chances for these groups to voice their opinions depending on their position in society. It is the task of science to examine the different needs and interests of these various groups -for example women, who unfortunately still have different starting conditions and resource endowments than menand thus develop a basis for fair structural policies for all social groups.
The conviction that low-carbon transitions should be implemented as fairly as possible for all parties involved (i.e. a just transition) has gained ground in international discourse in recent years. The gender dimension of climate and environmental policies is also recognised by major international political organisations such as the UN in its SDGs (Sustainable Development Goals). However, there is little scientific basis so far for gender-sensitive transition policies, because scholars looking at the consequences of low-carbon transitions often merely analyse broader economic effects (e.g. net employment effects) on entire regions, rather than the effects that transitions have on actor groups other than predominantly male miners. Addressing this issue, we developed a research agenda which leads to a better understanding of the interrelationship between gender and low-carbon transitions.
To develop such a research agenda, it was essential to gain an overview of all existing empirical information on this topic. To this end, we have conducted a systematic map that was as comprehensive as possible, containing publications dealing with our research question -the impact of historical coal transitions on women and their role in it. In the process of compiling the systematic map, we read over 3100 abstracts of potentially relevant publications obtained through a systematic literature search. After title and abstract screening, we identified 247 texts that could be relevant to our research question and read them in full text. 73 studies contained relevant results, which we then analysed. Based on these studies, we identified the most important links between coal transitions and gender, and developed a research agenda corresponding to the ILO guidelines for just transitions with essential questions that need to be investigated to make structural policies more gender-sensitive in the future.
We found various effects that historical coal phase-out processes had on women; these very much depended on the country context and the living reality of each woman. However, there are still some general statements we can make.
Previous coal transitions have often led to women entering the labour market, as many men in the coal industry lost their jobs and thus income. These women found employment especially in the service sector where, on average, the pay was significantly lower and the working conditions worse than in male-dominated sectors. Even though the economic situation for many families in former coal-mining regions worsened in general, women in those families gained some financial independence due to their increased employment; this strengthened their self-esteem and confidence. However, the lack of childcare services also led to an increased double burden and limited women's career opportunities.
The coal phase-out process additionally changed community life within coal regions. Often whole communities participated, for example, in miners' strikes, which strengthened their sense of community.
Women contributed significantly to these activities and networked the community members. On the other hand, with the advancing coal phase-out, women in former coal regions had less and less time to do community work as they traditionally had done due to their increased double work burden. In addition, structural change in the context of the coal phase-out had an impact on household cohabitation. In some cases, men and women were able to reorganise the division of labour in the household, as women increasingly took up employment. In other cases, however, the increasing change in the male breadwinner model had negative effects and led to intra-family conflicts and even violence.
Just as women's social positioning influences the effects that coal transitions have on them, agency within coal transitions is also gendered. Women have been active in both pro-coal and anti-coal movements. Regarding pro-coal movements, we found information on women's roles in miners' strikes in the UK, the USA and Spain. It was difficult for women to gain access to union structures, which led them to organise themselves into grassroots movements. They took up a variety of actions, from fundraising and setting up community kitchens to more radical actions, such as occupation. Women also played a large role in anti-coal movements, such as in the fight against MTR (Mountain Top Removal) in the USA. These grassroots groups were largely established by women. One reason for this is that men had closer ties to the coal industry. Activism in both pro-coal and anti-coal movements -even though this activism has been restricted by patriarchal structures in society -has often empowered women and increased their self-confidence.
Based on these findings, we identified the following key issues that need to be addressed scientifically in order to achieve a gender-just transition.
Furthermore, we found only a relatively small number of relevant studies in spite of a complex and extensive research process, which shows that the topic is still underresearched. In general, much more gender-aggregated qualitative and quantitative data is needed. In sum, we agree with Garcia-Garcia et al. (2020) that failing to consider "the gender implications of the energy transition is an analytical bias we can no longer afford" (13) since women, due to their social positioning in society, are very differently affected by low-carbon transitions and have different opportunities to shape it politically compared to men. Understanding their needs and positions more thoroughly is key to ensuring an economically and socially successful transition.    (transition*+OR+transformation*+OR+change*+OR+reform+ OR+closure*+OR+decreasing+OR+decrease*+OR+declin*+ OR+collaps*+OR+cris*s+OR+strik*+OR+resistance+OR+pr otest*+OR+agency+OR+activis*+OR+oppos*)+AND+(minin g+OR+lignite*+OR+anthracite*+OR+coal+OR+miner+OR+ miners+OR+coalfield*+OR+coalface+OR+((NOT+"data+min ing"))+OR+((NOT+"text+mining")))+AND+(gender*+OR+wo man*+OR+women*+OR+female*+OR+mother*+OR+femini* +OR+*wives+OR+*wife)&type=1&searchMode=And&site=e host-live no limitations applied 3 June 2020 122