Abstract
Intersectional analysis has been developing since its emergence from critical race feminism in the 1980s when it was used to conceptualize the inter-relationship of race and gender and, particularly, the experiences of discrimination and marginalization of black women in employment. While its contribution has been much debated within sociological and gender specific journals, its use still remains relatively limited within studies of work and employment relations. It is argued here that this field of study would benefit from greater engagement with and understanding of an intersectional approach to both the design and interpretation of research. Two lines of reasoning are put forward for this contention: firstly, that the intersectional approach contains an important caution against over-generalization that has been obscured; secondly, that separating the challenge for all academics to be more intersectionally sensitive from the methodological challenges of taking an intersectional approach brings the significance of intersectionality into sharper relief.
Introduction
While the term intersectionality was used first by Crenshaw (1989) to denote the experience of Black 1 women’s employment experiences, prior to this other critical race feminists had drawn attention to the fact that multiple axes of inequality (be they race, ethnicity, caste, class, gender) could not be considered in separate analytical spaces and that it was precisely at the point where multiple oppressions intersected that greater analytic focus was needed (Collins, 1989; Davis, 1981; hooks, 1981; Mies, 1982; Mies et al., 1986; Spelman, 1982, 1988). For example, as Spelman noted, much of the theorization around sexism and racism before this time ‘assumed that sexism is distinctly different from racism and classism, that whether and how one is subject to sexism is unaffected by whether or how one is subject to racism or classism’ (1988: 81). Within the study of work and employment relations, there has also been a long-standing tradition of research exploring intersections of class, race and gender (see Anthias and Yuval-Davis, 1992; Cavendish, 1982; Pollert, 1981; Westwood, 1984), so, why use Crenshaw’s 1989 and 1991 articles as our starting point? Crenshaw’s work was an explicit Black feminist critique of the manner in which a focus on the experiences of the ‘otherwise privileged’ within groups differentiated by gender and race led to the misunderstanding and marginalizing of the experiences of Black women and she gave a name to the phenomenon of intersectionality and set out its foundational propositions. It has also been so influential in current debates around intersectionality, with 4155 references to date 2 in academic articles.
Hailed as a ‘spectacular success’ within contemporary feminist scholarship (Davis, 2008), the concept of intersectionality has been re-interpreted since Crenshaw’s original usage. Rather than referring just to the intersection of racism and sexism (or race and gender) and thereby creating a focus on the experience of black and minority ethnic women, later interpretations have emphasized its potential to refer to the intersection of a broader range of oppressions (e.g. ageism, class) or social groupings (e.g. age, sexuality, disability).
Establishing an agreed focus for what is intersecting is only one of a number of debates in which academics have engaged over the years. These include questions such as whether intersectionality is a paradigm (Bilge, 2010; Hancock, 2007) or a theory of marginalized subjectivity or generalized identity (Nash, 2008), or if it should be considered a methodological approach (McCall, 2005; Yuval-Davis, 2006). More recently, Walby et al. (2012) have usefully identified (and responded to) six dilemmas of intersectionality. Since differences of its interpretation often reflect ontological and epistemological debates and controversies (Bilge, 2010), these discussions often assume specialist knowledge and engagement with earlier deliberations 3 (with a few notable exceptions, see Bradley and Healy, 2008: 42–60). The intensity and volume of these debates, however, has not been matched by the application of this concept to the field of work and employment relations. This, we argue, has led to missed opportunities in the understanding and theorization of the diversity of experiences in studies of work and employment relations.
The key contribution of this article is using the foundational principles of intersectionality to challenge all researchers to reflect on the extent (and acknowledgement) of generalization in their work. After identifying these foundational principles, we make a distinction between the methodological challenges for those wishing to be more intersectionally sensitive and for those wishing to take an explicitly intersectional approach to their research. Using this distinction, the remainder of this article illustrates how changes in methodology may facilitate greater insight in existing and future studies of work and employment relations. By endeavouring to provide working examples of how to address both needs, we hope to provide useful bridges from which to access the specialist intersectional knowledge.
The contribution of intersectionality
As indicated by a number of authors, the use of the term intersectionality is not unproblematic, with key differences of opinion emerging over time. Rather than reviewing these differences, the purpose of this section is to isolate what we believe to be the ontological assumptions of intersectionality that enable it to be a key resource for analysts of work and employment relations. Two foundational propositions emanate from Crenshaw’s work (1989, 1991): the intersectional process and the erasure or conflation of intragroup differences.
Crenshaw conceptualized the location of Black women at the intersection of race and gender and called this ‘structural intersectionality’. Within this specific location, Black women experience sexism differently to that experienced by white women and experience racism in a different way to that of Black men. Moreover, Crenshaw argued that the experience of Black women is ‘greater than the sum of racism and sexism’, such that the intersectional process within this location cannot simply be understood through the adding of the individual effects of sexism and racism, but needs to assume a different, maybe multiplicative, effect within these intersections. This takes us away from earlier conceptualizations of the ‘triple oppression’ of black women (Amos and Parmar, 1981, cited in Westwood, 1984) whereby it was assumed that ‘black women share a triple, not just a double, burden of oppression through class, gender and race’ (Westwood, 1984: 10). This aspect prompts a research agenda that seeks to explore and theorize the process of intersectionality itself.
Crenshaw argued that using the single axis of analysis (of either race or gender) ensures a concentration on the experience of the ‘otherwise privileged’ within either group such that:
conceptions of race and sex are ‘grounded in experiences that actually represent only a subset of a much more complex phenomenon’ (1989: 140); and
group-based remedies tend to be concentrated on the needs of the ‘privileged’.
If Black women are erased from the process of conceptualization and identification, then they are erased from remediation too. Crenshaw’s contribution arguably lies in conceptualizing and then operationalizing (using specific examples from employment law and domestic violence) how the single axis of analysis conflates or ignores intragroup differences. This aspect of Crenshaw’s work prompts questions about research strategy and requires us to question whether researchers fully appreciate and acknowledge the implications of voices that are both present and missing – something we have also explored elsewhere (Holgate et al., 2006).
In essence, we all need to be more aware of intragroup differences that exist within society. Subsequent writers have expanded the application of Crenshaw’s core arguments to indicate that there is also something distinct happening at other intersections (e.g. race and class; gender and disability). When abstracted from the context of Black women’s experiences as discussed by Crenshaw, it could be considered as a methodological caution against over-generalization (see also Hancock, 2007) which – as the following section argues – has become increasingly hidden at the very time that its pertinence is increasing.
Despite its success within contemporary feminist scholarship, intersectionality has little prominence in many other fields of study. Part of this may be related to the manner in which academics have enthusiastically embraced the study of ‘particular social groups at neglected points of intersection’ (McCall, 2005: 1773) and have published in specialist (mainly gender) journals perceived as most receptive to such studies. However, part of its isolation is also due to the complexity that its more inclusive application entails. Developing it from its original context of race and sex has raised a number of challenging questions about how to deal with the differences in the ontological construction of characteristics such as race, gender, class, sexuality and disability, and whether we should be studying the intersection of demographic characteristics or inequalities.
A number of recent articles have sought to bring clarity to this concept and identify the choices that authors are making, or need to make, when applying the concept of intersectionality (Bilge, 2010; Choo and Ferree, 2010; Dhamoon, 2011; Hancock, 2007; McCall, 2005; Tatli and Özbilgin, 2012; Walby et al., 2012). Despite their good intentions, however, such texts are geared towards those very familiar with the topic, rather than those wishing to dip their toe in the intersectional waters for the first time. This may explain its under-usage in the field of work and employment relations. In essence, it is difficult to engage with the core principles of a concept that is rapidly being developed, problematized and deconstructed using increasingly specialist terminology.
One way forward is to separate out the fundamental need for all researchers to be intersectionally sensitive (or aware) from the needs of a smaller number who may wish to do intersectional research, as arguably it is the needs of the latter that have obscured the importance of the former. It is inevitable that research cannot capture every nuance of work and employment relations but are we sufficiently cautious in the generalizations that flow from our methodologies? Are we sufficiently aware of the manner in which (sometimes relatively small) changes in methodology could open up important new lines of enquiry? A useful example of such sensitivity can be seen in the Cathles et al. (2010) study of the regulation of funeral directors. Suspecting that embalming regulations may have a disproportionate effect on the numbers of women who become funeral directors, Cathles et al. compare the experiences of male and female funeral directors. The authors acknowledge that further data is required to determine whether and how regulation may have an impact on the experience of black female funeral directors. While not using the terminology of intersectionality, this acknowledgement indicates their sensitivity to diversity within the category of women and their desire not to generalize the female experience. This level of sensitivity, however, is arguably the exception, rather than the norm, thereby raising the need for more of us to be intersectionally sensitive.
The challenge to be intersectionally sensitive
It is useful to start with McCall (2005) – as many writers do – as her seminal work in the feminist journal, Signs, identifies three distinct approaches to intersectionality and provides working examples of each. McCall reminds us that underpinning the three approaches are differing means by which we try and understand complexity in any field of enquiry. First, we might develop categories as a means of comparing experiences (intercategorical). Second, we might study experiences within a given category (intracategorical). Third, we might reject categorization and study experiences with no preconceptions of what characteristics individuals might share (anticategorical).
In the study of work and employment relations, feminist arguments against generalizations based on male ‘norms’ have ensured that most studies of workers include a gendered categorization and comparison of male and female experiences (Holgate et al., 2006). The additional questions that intersectionality introduces are twofold. First, the concept reminds us that there will be diversity within each category such that, again, at the very least we should acknowledge, for example, the limits of generalizability of the male or the female experience. Second, the concept reminds us that individuals within an intersectional space (i.e. of two overlapping categories) may be experiencing something significantly different to those occupying one of the categories. Thus, irrespective of our particular interests, arguably these foundational propositions challenge us all to be intersectionally sensitive both in our research practice and in accounts of our research.
The following intersectionally sensitive studies provide useful illustrations of insights waiting to be captured with more intersectional sensitivity. Browne and Misra (2003) indicate how race and gender intersect under certain conditions in labour markets. Rakoviski and Price-Glynn (2010) analyse a large-scale survey of nursing assistants to differentiate the experiences of this cohort of workers in relation to race, citizenship and gender. McDonald et al.’s (2011) work provides a perspective on how the intersection of gender and class affects young people’s aspirations in Australia. Jrkinen and McKie (2012) provide examples of gendered ageism and Wilton (2011) and Rafferty (2012) present useful evidence of graduate level over-education, under-employment and wages by ethnicity and gender. Given the potential for intersectional experiences to be erased in single category analysis, arguably there could be more acknowledgement in research accounts (as indicated earlier in Cathles et al., 2010) of what might be missing from single category analysis (even in a comparison of male and female experiences) and why this might be the case. For example, if such data does not exist, how important and feasible is its future collection? Could focus groups facilitate the collection and analysis of data from black women workers, for example, when survey responses are too small for statistical analysis?
The same questions apply to qualitative work. Baines and Cunningham’s (2011) contribution on gendered violence at work illustrates how a fascinating and well researched study on gender provides fertile ground for additional research. They use labour process theory to understand why violence is tolerated in the voluntary sector and show how gendered notions of women as natural caregivers are both propagated by management and invested in by female care workers themselves. How does being intersectionally sensitive add to an already revealing study of gendered violence at work? The only way to tackle this is to think about what is not evident. For example, the verbal abuse reported by care workers focused on vulnerability such as build, weight or sexual orientation (Baines and Cunningham, 2011: 767). This suggests that intersections of gender and sexuality may have shaped the particular forms of violence and abuse experienced by particular groups of women. Racial abuse was not reported nor is it evident but why not? An intersectionally sensitive approach would engage with this by discussing the research approach: was this because the care workers in both Scotland and Canada were white, or was it because race and ethnicity were irrelevant in the experience of violence? Baines and Cunningham are reflexive about the limitations of their sample in its focus on voluntary sector workplaces, but what would be interesting is to uncover whether voluntary workplaces are predominantly white and whether other forms of care work are not.
The analysis of the sacrifices made by care workers and their altruistic approach reveals the complexity of how power operates in the workplace and why it is difficult to dismantle. Such caring identities may also be shaped by other power dynamics. For example, studies have shown that care workers from different ethnicities invest in different caring identities (Macdonald and Merrill, 2009). An intersectionally sensitive approach would ask whether racial or class identities may have led care workers to tolerate violence. The hiring practices of management may also have been racialized. The only reason we can begin to ask these questions about the research sample and data analysis of this particular study is because the study is an outstanding example of a piece of in-depth research that illuminates how gendered power relations operate in concrete ways at the level of management practices and worker identities. This depth provides fruitful avenues to take forward the research agenda on gendered violence at work to include an analysis of how the diversity and difference within the category of female care workers can reveal the ways other structures of inequality shape violence at work. Furthermore, looking at diversity and complexity within categories helps us think about how this is experienced by male care workers as well as women. It is difficult to research such complexity all together without sacrificing the depth that allows the insight into the processes that facilitate and perpetuate violence at work. However, this is an area of study that is starting to grow and therefore is now fertile for analysing the complexity of how power operates, while still (hopefully) being able to make some generalizations about the gendered power relations that perpetuate it.
The challenge of taking an intersectional approach
We have put much of the specialist literature to one side while we have encouraged research(ers) to be more intersectionally sensitive, but this growing body of literature exists precisely because of the methodological challenges of taking an explicitly intersectional approach to research. Such an approach requires going beyond problematizing the relationships within categories of difference (an intersectionally sensitive approach) to one that engages in the theoretical challenges of problematizing the relationship between categories of difference. One challenge arises from what McCall (2005: 1782) identifies as the intracategorical approach that recognizes intragroup diversity and seeks to ‘uncover the differences and complexities of experience’ embodied at the intersection of multiple categories. This approach often focuses on the identity of the individual or groups marked as ‘different’ and can be pursued through case studies or narratives (examples include Anitha et al., 2012; Foster and Fosh, 2010; Hebson, 2009; Williams-Whitt and Taras, 2010). While all these studies are intersectionally sensitive and provide us with new insights into the experience of hitherto neglected groups of workers, a number of writers warn against the dangers of essentialism, such that these insights are taken as representative of all workers who might fall within the same intersectional categories (Choo and Ferree, 2010; Dhamoon, 2011; Yuval-Davis, 2006). In line with these critiques, we would like to see greater acknowledgement of the implications of such methodologies.
A second challenge arises from what McCall calls the intercategorical approach that focuses on the analysis of categories of difference. McCall (2005) provides an example of conducting secondary analysis of wages in relation to race, gender and class across four US cities. Similarly, Macdonald and Merrill (2009: 121) indicate how their intercategorical analysis of labour market statistics leads them to argue that the ranking and sorting of available workers by employers ‘reflects regionally specific cultural logics that can only be revealed through an intersectional lens’. Again, however, while such studies provide us with new insights, opportunities are missed to engage with a vigorous and on-going parallel debate that problematizes the relationship between categories of difference.
In particular, Hancock (2007: 71) argues that the uniqueness of intersectionality lies in ‘the ways in which it conceptualizes the constitution of, relationship between, and multi-level analysis of categories of difference’. When categories are static and assumed to matter equally in a predetermined relationship to each other then Hancock would deem that a ‘multiple’ approach to difference, rather than an intersectional approach. We indicated above that intersectionality is a caution against over-generalization. Hancock’s (2007: 74) perspective adds an extra dimension. Her argument that ‘intersectionality serves as an important corrective for imprudent overemphasis on generalizability that overlooks the priority of producing valid knowledge claims’ (our emphasis), reminds us of the need to think about (and acknowledge) the relationship we assume exists at the intersection of categories. As indicated by Durbin and Conley (2010: 197), the questions of how categories intersect are ‘big theoretical challenges’ for feminist theory, labour process theory and social policy.
The work by Choo and Ferree (2010) is a useful piece that provides a bridge between these theoretical debates and the application of intersectionality. Choo and Ferree (2010) achieve this through the critique of four studies. Their critique of Lamont’s (2000) study, The Dignity of Working Men, is of most relevance to our audience. Using examples from Lamont’s text, Choo and Ferree (2010: 143) argue that ‘the dominant racial category in each country’ is seen as the norm against which ‘black men’s stories follow and their differences are pointed out in “addition” to the shared features of class and nation, represented by whites’. By including their critique of Lamont, Choo and Ferree make it easier to understand their (2010: 147) argument for a ‘more dynamic, process-oriented, nonhegemonic intersectional analysis’ and we find arguments for this too in Dhamoon (2011). In essence, the call is for researchers to broaden the intersectional analysis beyond the ‘target’ subjects and to include other actors with power who have the ability to influence discrimination and exploitation at work. However, this is no easy feat (see Walby et al., 2012). Arguably, even writers such as Healy et al. (2011) who have explicitly engaged with the challenge of moving beyond the study of static categories of difference would still need to do more to address Choo and Ferree’s (2010) and Dhamoon’s (2011) call to study actors with power as a means of understanding systems of domination. While Healy et al.’s study of Caribbean, Bangladeshi and Pakistani women uses Acker’s (2006) conceptual framework of inequality regimes to ‘investigate how inequality regimes are sustained in the public sector’, the primary focus is on the experiences of women. This raises the importance of establishing what motivates a study – is it a desire to understand the lived experience of workers at the intersection, or is it to better understand the dynamics of power at work? While the former has often dominated in intersectional accounts, writers such as Choo and Ferree (2010) and Dhamoon (2011), along with Walby et al. (2012) and Acker (2006), show us this is not the only way to do intersectional analysis.
Concluding remarks
It is our contention that the contribution of intersectionality has been obscured. Our intention, therefore, in this article has been to distil some of the complex debates about the interpretation and use of intersectionality, such that we can understand how these ideas and concepts can be applied to studies of work and employment. Two challenges have been identified, along with illustrative cases and additional readings. The first challenge is to all researchers to reflect on the extent to which one’s own methodological approaches may be erasing or conflating intersectional experiences. As Crenshaw quite rightly asks, are we only being exposed to the experiences of the ‘otherwise privileged’ and what happens to those missing voices and experiences? In essence, are we, as researchers, sufficiently problematizing relationships within categories of difference? In some senses it is hard to know as often we have little knowledge of those who are missing from research accounts and the difference their experiences might make to interpretations, conclusions and future research. However, an intersectionally sensitive approach and its methodological preoccupation of problematizing relationships within categories engages with these missing voices directly by either attempting to integrate them or acknowledging their absence, and what implications this absence has for research findings and agendas. The second challenge is to those already engaged in intersectionally sensitive work. Even when data is produced on specific groups or individuals residing in an intersectional space, some authors are failing to acknowledge that there is a difference between looking at the intersection of static categories and questioning the relationship between them, or the process and systems of domination that marginalize or discriminate (Hancock, 2007). In essence, are researchers sufficiently problematizing relationships between categories of difference?
Our argument is that it is important to studies of work and employment relations that we appreciate that without intersectional analysis, conceptions may become grounded in ‘experiences that actually represent only a subset of a much more complex phenomenon’ (Crenshaw, 1989: 140). We recognize that we cannot investigate everything in the same depth and it is necessary to continue to research and study exploitation, marginalization and discrimination in all its forms. This will inevitably require a focus on specific social groupings, but we cannot get any nuanced understanding of, for example, youth unemployment in today’s economic environment without understanding intragroup and intergroup differences among this sample population. There are no easy answers to the methodological challenges raised by intersectionality, but hopefully this article will prompt a broader audience to enter these hitherto specialist debates.
Footnotes
Acknowledgements
The authors would like to thank the anonymous reviewers for their constructive comments. Thanks also to Paul Edwards, Geraldine Healy and Ed Heery for their comments on a much earlier version of this work. This article also benefitted from delegate comments on an interim version presented to the 2012 Gender, Work and Organization Conference at Keele University. The search work of Catherine Deering, information specialist, is also acknowledged.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
