Intersectionality in Finnish adult education research : insights from the journal Aikuiskasvatus 2010 – 2016

The article studies intersectionality in Finnish research on adult education. Specifically, we investigate the kinds of discussions on differences and their relations that are going on in such research. To this end we seek to identify intersectional approaches in the articles published in the journal Aikuiskasvatus between 2010 and 2016, a period marked by an increase in multiculturalism and social division as well as in gendered and sexual diversity in Finnish society. We understand intersectional differences as performative processes, not stable essences. Our study indicates that only few articles analysed intersectional differences explicitly. Implicitly recognised differences were mostly seen as givens. Categories such as ethnicity and race were found to be lacking in the data, but age, gender, social class, education, occupation and learning difficulties were discussed.


Introduction
We approach the Finnish research on adult education from the perspective of intersectionality by analysing the articles published in Aikuiskasvatus (Adult education), the leading academic journal in the field, during the years 2010-2016.Our aim is to inspire theoretical and methodological discussion in adult education.We assume that focusing on transverse power relations helps to reveal hidden relations of subordination and thus inform identity politics.As researchers, we engage with the feminist adult education tradition and embrace the theoretical discussions in which the intersectional approach has a long history.
The journal "Aikuiskasvatus" is a window on the discussions going on in the Finnish adult education community and a forum having the power to create the canon in the field.The journal is a leading academic publication in its field in Finland.Its overarching aim is to advance adult education in what is a changing society and examine educational topics from the perspective of humanist educational ideas (Aikuiskasvatus, 2017).
Recent years have seen an increase in intersectional research in adult education.For example the "European Journal for Research on the Education and Learning of Adults" (2014) published a thematic issue on European migration with articles on topics such as whiteness and the multi-ethnic environments in which adult education takes place.A number of edited volumes have appeared as well, such as those by Greyerbiehl, Simmons & Mitchell (2014), Davis, Brunn-Bevel, Olive & Jones (2015), and Griffin & Museus (2011a;2011b).Despite this trend, Kalwant and Uvanney (2014, pp. 2-3) call for increasing intersectional research on race, class and gender in the areas of school and adult education.
The rationale for examining the years 2010 to 2016 lies in the changes seen in Finnish society during the period and the challenges these have posed to adult education.Three trends in particular can be cited.First, the country has become more multicultural.Second, the neoliberal policies of the state have diminished the country's status as a prominent Nordic welfare state, increasing social class divisions and gender inequality (Saari, 2015).Third, considerations of equality have drawn attention to the rights of gendered and sexual minorities and other marginalised groups, such as people with physical or other disabilities (e.g.Brunila, Hakala, Lahelma & Teittinen, 2013).
Our hypothesis is that despite the social changes noted above, mainstream adult education research in Finland has not embraced the intersectional viewpoint to any significant extent.We assume that categories and differences nevertheless inform research approaches and, accordingly, undertake to identify the influence of intersectional approaches in particular.
We first provide a brief review of intersectionality as a controversial concept.After describing the selection of data, we proceed to the analysis, which applies insights into the intersectional approach gained from the work of Lykke (2005Lykke ( , 2012) ) and McCall (2005).The results are described in terms of the explicit and implicit approaches to intersectionality, with the latter further divided into intercategorical, additive and intracategorical approaches.We conclude that while the articles studied do apply differences among and between identity categories in their analyses, intersectionality can be seen as only recently emerging as an explicit principle of research.
Intersectional approach to differences Crenshaw (1989;1991) has been acknowledged as the first scholar to use the term 'intersectionality'.She studied experiences of oppression among black women and faulted the feminist research of the time for excluding their experiences.The intersectional approach was seen as an answer to the question how multiple categories produce differences among women and men who were no longer seen simply as representatives of homogenous gender groups, but also as white and black, young and old, poor and rich, homo-and heterosexual.These axes of differences were seen as constituting new experiences of exclusion and suppression (Davis et al., 2015;Karkulehto, Saresma, Harjunen & Kantola, 2012).
Differences among groups had been studied earlier in feminist research using the concepts of "double-disadvantage", "multi jeopardy", and "triple oppression" (Gregoriou, 2013, p. 180).However, such categorisations may overlook the nature of power as a relation.If power is viewed as productive rather than as one-directional oppression, people in marginal positions gain agency that may lead to resistance (Prins, 2006, pp. 279-280).Moreover, when applying the concept of double or triple oppression, different positions are understood as being unconnected, potentially resulting in lists of differences in what is known as "additive" (Lykke, 2005) and "cumulative intersectionality" (Davis et al., 2015).These approaches embody the essentialist notion that people's identities are known and constant, not socially constructed.Analyses should address not only identities, but also power relations that position people in different social, economic and political orders (Wijeyesinghe & Jones, 2014).Focusing on supposedly stable categories may prevent perceiving differences as historical and cultural processes of "categorisation" (Christensen & Jensen, 2012, p. 111).More important than asking what the differences are is to ask how differentiation has taken place and what kind of power is responsible for it (Levine-Rasky, 2011, pp. 242-243).
Methodologically, it is difficult to consider many differences simultaneously, one consequence being a focus on the most obvious and visible differences, such as sexuality, class and ethnicity, with the rest subsumed under "etc."This "politically correct" list has been called 'the North American politically correct feminist litany' (Vuola, 2002, p. 182).Such lists do not help to avoid marginalisation, however, for those defined as "different" may be relegated to the position of "Others", who, although included, do not challenge the power of the majority (Cerwonka, 2008, pp. 818-920).Karkulehto et al. (2012) point out that if intersectionality is investigated only as a combination of subordinate positions affected by sexism, racism and social marginalisation, research in that vein may lose its power to see the breaks and exceptions in those categories.Moreover, not only subordinate positions but also "the other side of the power relations" -the inner relations of whiteness and upper classes -should be investigated.Levine-Rasky (2011, p. 251) observes that white and middle-class positions may both strengthen each other and clash.In any event, a researcher must remember that differences are hierarchical and are thus not valued equally.In some contexts and schools of thought, social class may be seen as the most important category; in others, ethnic, racial or educational privileges figure prominently (Naskali, 2013).In a comment on categories, Lykke (2012) points out that intersections of categories can be analysed without "explicitly" using the term "intersectionality", an approach she calls "implicit" intersectionality.As noted above, we adhere to this same distinction in analysing the articles.Crenshaw (1991) distinguishes structural, political and representational intersectionality.Structural intersectional research asks how racism, for example, intensifies sexism and when and how homophobia strengthens racism (Verloo, 2006, p. 213).For example, this approach has been used in analysing the equality policy of the European Union (Kantola & Nousiainen, 2009;Verloo, 2006).Political intersectionality addresses the question how intersectionality is constructed in political discourses and practices, and representational intersectionality analyses the inequality and suppression that are constituted in the matrixes of cultural representations of gendered and other identities (Crenshaw, 1991).In another approach, performative intersectionality proposes a knowledge interest that exceeds the traditional boundaries between structures, politics and representations as well as the boundaries between academic disciplines.Research informed by the idea of performativity (see Butler, 1990) investigates how 'norms, rules, systems, structures, institutions, as well as discourses and representations, reproduce differences and make them real and how they construct states of affairs and meanings' (Karkulehto et al., 2012, p. 25).We share this understanding of intersectionality as an approach that does not view categories and differences as essential identities or characteristics of individuals but as complex cultural and historical processes.
In both feminist and adult education research, intersectionality has been criticised as a 'common nodal point' (Carbin & Edenheim, 2013), a 'black box' (Lykke, 2012) or 'buzzword' (Davis, 2008;Jones, 2014), meaning that the term has become a rhetorical device that is an explanation in itself.If one is to respond to the challenges and questions prompted by the manifestations of phenomena in time and place, individual differences need to be analysed.Intersectionality does not guarantee the superiority of a piece of research, but an ethical undertaking requires that differences and differentiation be recognised.Intersectional research may reveal hidden power structures and thus support identity politics among adult education students from different backgrounds.We share the viewpoint of Davis (2008), who argues that precisely because of its open-endedness intersectional theorisation is valuable, for it keeps the discussion going and challenges researchers to explore what is a contradictory and complicated world.

Data and analysis
The journal "Aikuiskasvatus" has been published four times a year since 1981.It features scientific articles, book reviews, and survey articles dealing with education policy and practices and has regularly published special thematic issues.Our data include two such issues on 'gender' (3/2011 and 3/2016).Among the other themes featured are 'teacher as developer' (1/2011), 'age, education and work' (3/2012) 'expertise' (4/2012), 'lifelong learning' (2/2013) and 'liberal adult education' (3/2013).
In the period 2010 to 2016 (the last issue of 2016 was not available while writing the article) 27 issues have been published containing a total of 91 refereed articles.After familiarising ourselves with the material as a whole, we decided to omit articles on topics such as research methodology, pedagogical theories, leadership, adult educational policy, and the history and philosophy of adult education.We also excluded articles that described the groups studied using terms like "students", "adults", "citizens", "interviewees" and "workers" as uniform groups without mentioning the positions or identity categories of the people.Applying these criteria yielded a dataset of fourteen refereed and six survey articles addressing.
We subscribe to social constructivism as an epistemological approach.Accordingly, in studying cultural artefacts such as a scientific journal, the focus is on the use of language and construction of meanings, not the thinking or purposes of the authors.In this respect, our analysis adheres to the tradition of "discursive reading" (Fairclough, 2003) and the idea of performativity of language (Austin, 1962), according to which reality can only be reached via language; language is an act and has the power to make things happen.Moreover, we understand values as a part of meaning systems; valueneutral knowledge does not exist but we can pursue strong objectivity by making our starting points clear (Tanesini, 1999, pp. 12-13, 269;Harding, 2004, pp.132-136).This way of thinking emphasises the ethical value of describing the complexity of social life instead of seeking generalisations and sameness (McCall, 2005).The purpose of our analysis is affirmative: we want to open up theoretical and methodological discussions in the field of adult education research.
In practice, we read the articles carefully by asking questions about the assumptions made in them regarding differences and categories.This approach to reading focuses attention on expressions and details that seem self-evident because they are firmly established culturally.The reading is attuned to what is said, and how, but also to what is not said; that is, we try to determine what elements have been left out of the text (Reinharz & Kulick, 2007, p. 259).
To do justice to the articles, we did not decide beforehand whether an article used an intersectional approach.As analytic tools we draw on the frameworks applied by Lykke (2005;2012) and McCall (2005).Lykke distinguishes analyses that theorise intersections explicitly using the term "intersectionality" and analyses that do so implicitly, without using the term.She points out that referring to differences and their connections is possible using other concepts, such as a rhizome.McCall (2005) refers to an approach that presupposes existing social groups and inequality between them as "intercategorical".The categories may be used strategically for improving the position of suppressed people, but the approach minimises the complexity inside the groups studied.The wealth of differences within categories is taken into account in the "intracategorical" approach, which is often used to reveal the hidden power relations within groups.A third approach, the "anticategorical", refers to deconstructive research that endeavours to uncover the complexity of social life without defining any categories.The problem in such research is that language in itself creates categories and some degree of uniformity is needed for identity politics.
We emphasise that the boundaries between the approaches are not sharply defined; a given article belonging mostly to one approach may cross the boundaries (see McCall, 2005).First, we divided the articles into two groups: those applying implicit intersectionality and those applying explicit intersectionality.Within the first group we further distinguished articles using intercategorical, additive and intracategorical approaches.We then proceeded to analyse closely whether differences were understood in terms of already defined, essential characteristics or constructed by power relations, that is, performatively.The first group comprised twelve refereed and four survey articles.The second was quite small, containing two refereed and two survey articles.The decisions on the group to which an article was assigned were made in a joint discussion after we had grouped the texts independently.

Intercategorical approach
Gender is an often-used background variable in much survey research in the social and educational sciences, and it has been examined in qualitative research as an important social category.Discussion of gender in "Aikuiskasvatus" has increased in recent years.Vuorikoski and Ojala (2006) analysed scientific articles published in the journal between the years 1981 and 2005, investigating the incidence of gender in them.Their research shows that only nine articles discussed gender issues (Vuorikoski et al., 2006, pp. 318-319).In recent years, gender has figured more prominently as an analytical tool; our data contain thirteen articles that discuss gender.Eight of these are refereed and five survey articles.
Identifying the educational opportunities of women and men is important for advancing equality.However, gender can be discussed without subscribing wholeheartedly to contemporary feminist approaches and thus succumbing to reproducing gendered stereotypes and dichotomist thinking.For example, the article 'The Story of Chief Shop Steward Aliisa' (3/2014) seeks to challenge gendered and generationoriented ways to act but the results of the study are broken down in terms of masculine and feminine stories: 'The masculine story plays out on the stages of formal and informal meetings, the feminine in quiet corners where women share their experiences, concerns and feelings' (p.166).The gender dichotomy is not questioned, and the impossibility of breaking out of the feminine story seems to be attributed specifically to the personal characteristics of Aliisa.
Much as the story of Aliisa does not consider differences other than gender, neither does the article 'Female Researchers in a Gendered University Organisation' (1/2016), which analyses women's position in the university context.The article is based on a survey carried out in a Finnish university and analyses the qualitative data collected.The questionnaire was sent to female employees only, underscoring the assumption that women's position in universities is worse than men's.The study does not consider differences among female employees and thus possible oppression caused by prejudice against transgender people and sexual minorities may be ignored.The age distribution of the respondents is described, but the article does not examine its significance for a person's position in the university.While it is occasionally necessary to concentrate on a single category to bring suppressive power relations to light, many studies have shown that gendered relations in universities (e.g.Lempiäinen, 2014) are complicated and not restricted to a single category.Focusing on women only may emphasise the idea that "female" is tantamount to "gender".This mind-set harkens back to the long-refuted view that a man is representative of mankind and humanity at large, and a woman of the Other, as Simone de Beauvoir (1949Beauvoir ( /1987) ) has described the position.
The image of a feminine citizenship -one that can be distinguished from "real" citizenship that is gender neutral but male -is also found in special issue 3/2016, which examines women's vocational education and care as topics representative of "female citizenship".The articles in this issue are linked together, forming a narrative of an average girl and woman and her role in transforming Finnish society.The description of how an imaginary obedient girl becomes a caring woman constitutes a story that reproduces women as a uniform category.
One refereed article and several of the survey articles take up issues of masculinity.The articles seem to view men and masculinities as coherent groups and fail to problematise them as culturally constructed categories.This interpretation resonates with the findings of Hearn (2012) that studies of men have neglected intersectional analysis.For example, one of the articles in our data, 'Is Gender Decisive in Education?Male Academics' Experiences of their Studies and of the Labour Market' (3/2011), discusses the experiences of male media education students and claims that the male gender identity is undergoing a transformation because of the changing demands of the labour markets.Male students' experiences are contrasted with those of female students, in which respect the article sustains the gender dichotomy and adheres to an intercategorical approach.
Recent years have seen increased attention paid to age in social studies.One special issue on age was published during the period studied.Entitled 'Age, education and work' (3/2012), the articles consider a range of topics spanning age as an economic problem, conflicts between generations, educating elderly persons in working life, age and the educability of elderly women, ageing policy and lifelong learning.These themes attest to the fact that the population in the Nordic countries is ageing rapidly and that this development is associated with economic regression and viewed as a challenge to the Nordic welfare model (Keskitalo-Foley & Naskali, 2016).
The group commonly referred to as "adults" is diverse, and looking at a range of different social and cultural factors would help in giving a valid picture of the complexity of adulthood.In our data, adulthood is often discussed as a background factor that is not explicitly identified.Age, signifying adulthood, is subsumed in terms such as "lifelong learning", "citizenship", "adult pedagogy", "labour markets", "generation", "liberal adult education" and "expertise".For the most part, age is not defined in the data, because working life, for example, is delimited as being between youth and retirement age."Youth", meaning "young adults", are the focus of articles discussing the risk of exclusion and voluntary work."Young adult" is defined explicitly, but in two different ways: in one article, appearing in 4/2012, the age range is 17 to 30 years; in the other, published in1/2013, it is 21 to 36 years.
Age and gender are also mentioned in studies where the data have been collected using surveys or quantified from data collected by interviews.However, qualitative approaches dominate, as only four articles in the data are based on quantitative methodologies.In those articles, categories of people are considered as background variables which, it is assumed, will have an explanatory power with regard to the phenomena examined.For example, the article titled 'The Portrait of an Open-University Student in the 2010s' (4/2014) uses a survey to explain the reasons and motives for studying in the Open University.Differences in motivation are explored by correlating them with typical background variables, for example, age, gender, occupation and education.
Examining background variables as factors helps to determine differences between groups of people, but such variables are given by the researcher; the groups subsumed under them are considered uniform, and differences within the groups become blurred (McCall, 2005).Moreover, correlations between variables are readily interpreted as causal relations, a tendency that runs counter to the intersectional idea that emphasises differences as the products of multiple power relations.
Yet, survey-based research does not prevent critical analysis that avoids essentialist or deterministic interpretations.The article 'Adults, Too, Need Help with Learning Difficulties' (2/2011) highlights learning difficulties as a difference discussed in the adult education context, albeit rarely.Gender is approached as a background variable, as are age, education and occupation.However, the article analyses critically what it sees as a self-evident supposition made in previous research: that men have more learning difficulties than women.The article goes on to suggest that this shortcoming may result from the underdiagnosing of women, who as a result received far fewer services than men during their school years (pp.129-130).
Despite the limitations of this kind of intercategorical approach, it may be an important tool in critical research because it highlights inequalities within a given population.Statistics and quantitative research are needed to illustrate the economic and social conditions of different population groups; in doing so they help to construct an identity category that has strategic meaning in the political struggle for social and educational rights.

Additive Intersectionality
The articles exemplifying additive intersectionality show an awareness of differences as important factors in constructing identities but do not consider them as being connected to each other.Differences are either mentioned as starting points of the research or listed as important identity positions.This interpretation typifies additive intersectionality, which according to Lykke (2005), incorporates the essentialist thought that people's positions are already known (see also Davis et al., 2015).
The use of additive intersectionality may also be seen as reflecting the methodological challenges of addressing the differences among many categories.Sometimes the challenge to pay attention to differences comes from the data.This is the case in the article 'How Ageing Employees View Change in Retail Trade' (2/2011), which analyses the experiences ageing employees have had of the changes in the business sector.Age is seen as the sole relevant factor even though the sector is one of the most gendered in the labour market.The article does not comment on this in any way, but mentions that six of the managers in the data were men and thirty-eight of the workers women.Gender comes up in the article through the interviews.The employees describe age discrimination combined with gender discrimination, with the latter reflected in the importance given to appearance.The article calls this a 'double standard': both age and gender contribute to discrimination against women in working life.This conclusion in the text reflects the view that differences are cumulative sources of oppression.
In a similar fashion, the article 'Hooked On Feelings -the Therapeutic Ethos in Education for Young Adults at Risk of Social Exclusion' (4/2012) lists the categories age, gender and social situation in examining groups of people who are at risk of being excluded from society and are therefore targets of educational interventions.The focus of the article is on educational projects as a facet of the culture of therapisation.Although it adopts a constructivist point of view, the article fails to analyse the relations between different categories: it introduces and applies them in an additive manner.
Differences such as age, gender, social background and occupation easily form a list that is repeated and applied in different studies without the researcher stopping to ask what differences are missing.The present data contained such a case, with the exception that the overlooked differences were internationally crucial ones, such as race and ethnicity.Geographical location is another difference that is seldom taken into account in intersectional research and this shortcoming has evidently led to researchers making generalisations and assumptions that nations are coherent entities.From a feminist point of view, knowledge is connected to its material basis, and the situatedness of knowledge and place shape agency (Naskali, 2013;Tanesini, 1999).Our data contain one article, 'The Values in Young Adults' Volunteering', (1/2013) that took into account the importance of place and thus of geographical differences.The data on which the article is based are explicitly located in metropolitan Helsinki.However, although the area is highly fragmented into wealthy and poor districts (Keränen, Vaattovaara, Kortteinen, Koski, Ratvio & Rantala, 2013), the differences within the capital area are not discussed.Moreover, at the end of the article, the results are generalised as applying to the country using terms such as 'Finnish young adults': 'Finnish young adults thus appear […] to find volunteer work interesting particularly because it involves helping others' (p.34).
As noted above, it is difficult to draw a clear line of demarcation between analytical groups.This became evident in two articles in particular.The article 'Attention Must Be Given to the Masculine Occupational Identity in Education and Counselling' (3/2014) analyses male-dominated professions from the viewpoint of the masculine culture that should be taken into account when providing counselling on a healthy way of life.The article assumes that belonging to a masculine community, with its longing for freedom and resistance to control, is an important part of the identity of working-class employees and that this will be transmitted from one generation to the next.The focus is on how the identities of the men studied should be taken into account in education.The article considers both gender and social class and in this respect can be seen as using an intracategorical approach.However, gender and social class are taken as given in the article.As the text does not reflect on these categories or analyse the interaction between them, we view the approach used as additive intersectionality.
The article 'Reflexivity and Gender in Career Counselling' (3/2011) seems to have ingredients allowing it to be assigned to either the additive or intercategorical group.Fundamentally, it can be seen as addressing the need to recognise the intracategorical differences between men and women.For example, it quotes the following statement from Hyvärinen, Peltonen & Vilkko (1998, p. 185): 'The difference between genders should be conceptualised such that it provides an opportunity to thematise the differences among women, for example, as well as the differences within one and the same woman'.Despite embracing this principle, the article converts the differences into background variables when analysing the data, as the following statement indicates: 'Acquiring selfreflection skills is heavily dependent on a person's age, generation, social class, educational background, profession, and gender ' (p. 186).This approach constructs quite stereotypical gendered categories, that is, 'personal women' and 'selfish men'.Men are seen as analysing their positions on the general level, rationally, whereas women's reflection is more personal and emotional.The conclusion underscores the traditional gender roles: 'The men and women in the study behaved largely in keeping with the gender roles characteristic of their generation ' (p. 191).In this light, we assigned the article to the group applying additive intersectionality, not to that using intracategorical approach, the form of implicit intersectionality which we will take up in what follows.

Intracategorical approach
In the group of implicit intracategorical research, the articles demonstrate an awareness of the importance of differences in striving to understand the complexity of life.The analyses take into account the intersection of many differences, but the term 'intersectionality' is not used.The articles in this category also reflect on the differences and take a critical stance towards readymade categorisations.
In the article 'Age and the Dilemmas of Educability in the Narratives of Women in Their Fifties and Sixties' (3/2012), age is connected to other differences.The article starts from the principle that not only age but also gender and social class must be taken into account in analysing narratives on educability: 'Age, gender and social class are defined in this article as lived social relations, not abstract or static positions, and they involve tensions and negotiation ' (p. 193).We interpret this expression as the text defining differences in accordance with constructionist theory; that is, differences are not essential characteristics of individuals but processes that are negotiated in social interactions.The analysis is critical of the categorisation of students according to their cognitive achievements and of social age norms (e.g.matriculation) that define educability.The research analyses differences between the age groups comprising people in their fifties and in their sixties: for the former, education can be interpreted as a resource in the labour market; for the latter, a completed qualification is seen as part of self-development, with the group defined as being on the margin of the labour market (pp. 197-198).
The article 'Justification for Studying at an Older Age' (2/2013) can also be placed into the intracategorical group.It recognises differences within the group of elderly people studying at the University of the Third Age.The article asserts that if elderly people are expected to be studying only at the University of the Third Age, they will be marginalised vis-a-vis other students: 'Even as the policy of lifelong learning and Universities of the Third Age increase social inclusion, they create exclusion as well' (p.92).This 'double-edged sword' is said to lead to a situation where 'people of retirement age, who end up in the margins in terms of the policy on lifelong learning, enter the educational spotlight in Universities of the Third Age and represent a retired elite to people of the same age who are excluded from the University of the Third Age' (p.92).The marginal position of elderly students is produced in relation to younger, "proper" students, but this becomes an elite position in relation to retired people who lack the resources to attend a University of the Third Age.In other words, the article recognises age and class positions within a group of elderly persons that is usually represented as homogeneous.

Explicit Intersectionality
Our data set contains only two refereed and two survey articles that explicitly locate their theoretical starting point in intersectional theorisation.The interdisciplinary background of these articles lies in feminist adult education, and the themes relate to teaching in the university and feminist pedagogy.Also identifiable in this group, is the challenge of analysing the multiple interplays of differences and the power structures that underlie them.According to our analysis, one article adheres to the principle of performative intersectionality in demonstrating how multiple differences do not exist as essential characteristics but are constructed in power structures and reciprocal interaction.
The article 'Feminist Pedagogy in Gender Equality Training: From Theory to Practice' (2/2013) analyses teaching episodes from the perspective of multiple power relations.The article could also be seen as applying an anticategorical approach (McCall, 2005) given its openly stated purpose of deconstructing the dichotomous and essentialist thinking in gender equality.Indeed, it emphasises the connection between gender and age, class, ethnic background and sexual orientation, as well as differences and multiple power relations within these categories (p.120).The article gives the following episode as an example of analysing multiple power relations: 'In equality education, power relations in the workplace affect the dynamics among participants [...] At one session, an older man who belonged to the senior management said at the very outset that inequality in the workplace could be explained by the fact that women do not want to become managers.In such a situation it is difficult for women working on the lowest levels of the hierarchy to start arguing ' (p. 125).
The article points out that the intertwining of gender, social hierarchy and age produce different power positions and spaces.In the example, power is highlighted as reproducing traditional gendered positions between women and men.Sexuality and race are also recognised as constructing power positions in teaching situations: 'When a white male professor, for example, questions his own identity and emphasises the rights of women and minorities, he is a hero, whereas a lesbian who emphasises the acknowledgement of differences is easily labelled a fanatic ' (p. 123).The position of a teacher as a representative of the "neutral" dominating majority is deconstructed and situated in the categories of whiteness and heteronormativity, which are not accepted as positions of the universal knowing subject.
The second article in this category, 'Teaching and Gender in the Neoliberal University', (3/2011) analyses the gendered effects of the changed university politics and working culture.It proceeds from an assumption that gender is a fundamental category in the hierarchical academic organisation and stresses that 'it is important to remember that gender becomes intertwined with other divisions, such as age, socio-economic status and social background' (p.165).Moreover, intracategorical (McCall, 2005) differences are emphasised: 'Women (like men) differ; they all have different experiences and they interpret their experiences differently.The power relations associated with academic hierarchies have impacts within the categories "men" and "women" as well' (p.166).However, the article chooses to focus on the position of women because previous research has shown that women still have difficulties being recognised as distinguished researchers, receive less encouragement during their careers and are expected to take more responsibility in teaching and administrative tasks.Our interpretation is that the article is very much aware of the differences within and between categories.However, implementing this principle in methodological practice is challenging, and the text does not take the opportunity to make the differences and the power visible.
In the Nordic countries, especially Sweden, intersectionality has had a particularly strong position in feminist studies, where the concept was introduced in the early 2000s in order to criticise 'hegemonic white/Nordic feminism and its blindness to race and ethnicity' (Carbin & Edenheim, 2013, p. 235).The article 'Nordic Research Co-operation on Gender Issues in Education', published in the section of survey articles (3/2011), uses the term "intersectionality" explicitly in discussing the change that has taken place in Nordic research on education and gender: '[…] the ways in which gender is understood have changed over the years and in the 2000s intersectionality in particular -the examination of gender alongside other distinctions such as social class, ethnicity and sexuality -has become common' (p.207).
Intersectionality as a principle is also mentioned in the survey article 'Older Women in the University of the Third Age' (2/2011).It discusses active ageing and stresses that studying and other activities of elderly persons must be analysed in the context of the relevant social and institutional practices if we are to understand how class, age and gender construct differences among elderly people.We submit that such statements can be seen as a strategy on the part of the researcher to locate her-/himself in the theoretical field even though he/she does not apply intersectionality to any significant extent in the analysis.

Concluding remarks
On the basis of our analysis, we suggest that intersectionality is coming to the Finnish adult education discussion but has not yet worked its way into the research frameworks to any noteworthy extent.To avoid amplifying the use of what is a topical concept as a "buzzword", we sought to analyse the implicit ways of examining differences, which we consider an important ethical principle if we are to capture and comprehend the complexity of adult education.For the most part, the articles examined here represent implicit intersectionality; that is, they focus on identity in terms of already defined categories rather than defining categories as socially constructed by power relations.Differences are also seen as background variables and separate or additive categories that are comparable; little consideration is given to intracategorical differences, and performative intersectionality was found only in one article.
Differences such as age, gender, geographical location, social class, education, occupation, and learning difficulties are recognised as salient factors in discussing adult education issues.By contrast, ethnicity and disability are lacking among the differences studied, and sexual and gender minorities are not discussed either.In this respect, the journal paints a rather coherent picture of the people who are involved in Finnish adult education.
One particularly interesting finding is that ethnicity and race are not discussed in the issues analysed here.Finland has seen an increased number of people coming from different cultures as immigrant workers or refugees and multiculturalism has been a focus of research in education (Souto, 2011) and the social and cultural sciences (Tuori, Mulinari, Keskinen & Irni, 2009).In an editorial in the issue 3/2011, the chief editor points out that the situation in Finland should be seen in relation to that in the rest of the world.He describes how indigenous people, poor people in rural areas, other minorities, the disabled and, in particular, girls suffer from being in an unequal position (Silvennoinen, 2011, p. 163).Yet this awareness cannot be seen in the journal articles, and the opinion of the chief editor can be interpreted as detaching these marginal groups from the Finnish context: talk about diversity as such presupposes a "we" as representatives of "same"; the problems are "out there" (see Ahmed, 2012).
The articles that had their starting point in feminist theory were based on poststructural epistemology, which assumes that differences are not determined by biological characteristics or social structures but are constructed in ongoing processes.In this understanding, the boundaries between gender, class and age are not fixed but flexible, and are constructed in social processes.Yet, in many articles, the analyses failed to take the multiple power structures into account; it would seem that intersectionality remained more of a declaration of "political correctness" demonstrating that the author was in touch with, and wanted to be recognised as adhering to, the canon.
In feminist research, intersectionality has been a subject of spirited discussion and also an object of intense criticism.It has been pointed out that taking minorities into account does not mean that the gaze is directed to the majority still in power, and talk about categories may at the same time keep the categories alive (Carbin & Edenheim, 2013).As Ahmed (2012, 182) has remarked, attention to a group defined as different in terms of ethnic or racial definitions may hide whiteness.She also points out that it is not categories as such that are a problem but the way they are used.Categories are a part of our social existence and do not disappear because 'they should not matter'.
Intersectionality has been criticised for becoming a 'catch-all phrase' and 'allinclusive' travelling theory.Nevertheless, it may offer the 'promise of complexity', of being critical and of overcoming divisions (Carbin & Edenheim, 2013, pp. 237-238).Like all theoretical approaches and concepts, it must be applied reflectively, taking into account the context and purposes of the research; it is only then that we will find in it an important vehicle for analysing both the intercategorical and intracategorical complexities of the phenomena we encounter in adult education.