womens studies international forum 46 (2014)

133

Upload: rosana-fabrin-lindorfer

Post on 19-Feb-2016

93 views

Category:

Documents


1 download

DESCRIPTION

"Women's Studies International Forum (formerly Women's Studies International Quarterly, established in 1978) is a bimonthly journal to aid the distribution and exchange of feminist research in the multidisciplinary, international area of women's studies and in feminist research in other disciplines. The policy of the journal is to establish a feminist forum for discussion and debate." Editorial of the journal

TRANSCRIPT

Page 1: Womens Studies International Forum 46 (2014)
Page 2: Womens Studies International Forum 46 (2014)

WOMEN’S STUDIES INTERNATIONAL FORUMTHE EDITORS

Editor-in-ChiefKalwant BhopalDept. of Education University of Southampton Highfield, SouthamptonSO17 1BJUKEmail: [email protected]

Associate EditorsCigdem BalimDept. of NELCIndiana UniversityGoodbody Hall 102Bloomington, IN 47405 USAE-mail: [email protected]

Jane CareySchool of Humanities and Social Inquiry,University of Wollongong, NSW, AUSTRALIA, 2522E-mail: [email protected]

Shirlena HuangDepartment of GeographyNational University of Singapore (NUS) 1 Arts Link117570 SingaporeSingaporeFax: +(65) (0) 6777 3091Tel: +(65) (0) 6874 3851E-mail: [email protected]

Jane C. Park School of Philosophical and Historical InquiryFaculty of Arts and Social SciencesThe University of SydneyK.406 Level 4 Main Quadrangle A14, NSW 2006E-mail: jane.park @sydney.edu.au

Afroditi PinaDepartment of PsychologyUniversity of Kent Canterbury, Kent CT2 7NPUKE-mail: [email protected]

Elizabeth RibetColumbia University School of LawCenter on Intersectionality and Social Policy NY, 10027 USAE-mail: [email protected]

Lavinia StanDept. of Political ScienceSt. Francis Xavier UniversityPO Box 5000, AntigonishB2G 2W5Canada E-mail: [email protected]

Book Review EditorFin (F.) Cullen School of Health Sciences and Social CareBrunel University Mary Seacole Building Kingston LaneUB8 3PH, Uxbridge, EnglandUK E-mail: [email protected];[email protected]

Founding EditorDale Spender

Consulting EditorsCynthia EnloePhyllis HallRenate D. KleinCarole LeathwoodRonit LentinPat MahonyRobyn RowlandAilbhe Smyth

EDITORIAL ADVISORY BOARD

Advertising information: If you are interested in advertising or other commercial opportunities please e-mail [email protected] and your enquiry will be passed to the correct person who will respond to you within 48 hours.USA mailing notice: Women’s Studies International Forum (ISSN 0277-5395) is published bimonthly by Elsevier B.V.(P.O. Box 211, 1000 AE Amsterdam, The Netherlands). Periodicals postage paid at Jamaica, NY 11431 and additional mailing offices.USA POSTMASTER: Send change of address to Women’s Studies International Forum, Elsevier Customer Service Department, 3251 Riverport Lane, Maryland Heights, MO 63043, USA.

Airfreight and MAILING in USA by Air Business Ltd., c/o Worldnet Shipping Inc., 156-15, 146th Avenue, 2nd Floor, Jamaica, NY 11434, USA.Orders, claims, and journal enquiries: please contact the Elsevier Customer Service Department nearest you:St. Louis: Elsevier Customer Service Department, 3251 Riverport Lane, Maryland Heights, MO 63043, USA; phone: (877) 8397126 [toll free within the USA]; (+1) (314) 4478878 [outside the USA]; fax: (+1) (314) 4478077; e-mail: [email protected]: Elsevier Customer Service Department, The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK; phone: (+44) (1865) 843434; fax: (+44) (1865) 843970; e-mail: [email protected] Tokyo: Elsevier Customer Service Department, 4F Higashi-Azabu, 1-Chome Bldg, 1-9-15 Higashi-Azabu, Minato-ku, Tokyo 106-0044, Japan; phone: (+81) (3) 5561 5037; fax: (+81) (3) 5561 5047; e-mail: [email protected]: Elsevier Customer Service Department, 3 Killiney Road, #08-01 Winsland House I, Singapore 239519; phone: (+65) 63490222; fax: (+65) 67331510; e-mail: [email protected]

P.J. Burke, UK R.M. Calogero, USA M. Coy, UK P. Easteal, Australia K. Ghodsee, USA L. Gottschalk, AustraliaP.D. Gqola, South AfricaS. Hawthorne, AustraliaM. Horvath, UK M. Hunter, USA

S. Jackson, UKS. Jeffreys, Australia L. Kelly, UK R. Klein, AustraliaC. Kramarae, USA P. Lan, ROC C. Leathwood, UK G. Letherby, UK P. Mahony, UK S. Nadar, South Africa

V. Reddy, South Africa D. Roman, USA S. Roseneil, UK Z.S. Salhi, UK C.P. Shimizu, USAR.M. Silvey, Canada J. Smart, Australia C. Taylor AM, Australia S. Walby, UK E. Weiner, Canada

Page 3: Womens Studies International Forum 46 (2014)

Women's Studies International Forum 46 (2014) 1–4

Contents lists available at ScienceDirect

Women's Studies International Forum

j ourna l homepage: www.e lsev ie r .com/ locate /ws i f

Feminization of labor: Domestic work between regulation and

intimacy

Time seems to stand still as we flick idly through glossymagazines decorated by pictures of celebrity “hands on”mumsand dads. They smile at the camera with perfect hair andmake-up, holding their children as they pose in impeccablydecorated living rooms or well looked after designer gardens.Despite full agendas and demanding careers, they give theimpression that they still find time for their children and theirhouseholds. How do theymanage it? This is the question askedby the average reader of these magazines, who are mostlyfemale and less affluent. In their search for answers they willquestion their ownpersonal abilities and that of other householdmembers. Certainly the female reader will wonder about thedistribution of domestic work in their households and noticethat unfortunately this work remains their burden.

Broadly culturally perceived as women's terrain, thehistorically established correlation between domestic work andfemininity remains unbroken (Federici, 2004). Domestic workhas been naturalized as labor that women do because of theirinnate caring faculties. It is, thus, considered inferior to otherforms of work (Bock & Duden, 1977; Chaney & Garcia Castro,1993;Delphy, 1984;Duden, 2009, Federici, 2004;Hausen, 2000).Domestic workers continue to be subjected to poorly paid,precarious, abusive and exploitative working conditions (ETUC,2005, 2012; ILO, 2012; WIEGO1). Further, the contribution ofdomestic work to national and international economic growth issilenced in GNP calculations (Himmelweit, 1995; Melo,Pessanha, & Parreiras, 2005; Pérez Orozco, 2010; Waring,1999). This persistent social devaluation of physical, emotionaland intellectual capacities and practices which engage withwell-being and care of ourselves, others and our environment,is reinforced by the constraints produced by migration policies(see in this volume Anderson; Assis; Courtis & Pacecca; GilAraújo & Gonzalez; Gorban & Tizziani; Gutiérrez Rodríguez).

Due to the increased participation of women in the labormarket, which has led to a demand for domestic and careworkers (ILO, 2012), the affective and emotional dimension aswell as the persistent social devaluation of domestic work hascome to the fore of current public and academic debates. As thearticles here demonstrate, domestic work in private householdsremains a highly un-regulated work sector. Even in countriessuch as Spain, where domestic workers' organizations havesucceeded in including domestic work into the general socialsecurity system (see in this volume Gil Araújo & Gonzalez;

0277-5395/$ – see front matter © 2013 Published by Elsevier Ltd.http://dx.doi.org/10.1016/j.wsif.2013.07.009

Gutiérrez Rodríguez), these achievements are challengedthrough the economic crisis impacting on private households,resulting in lower pay and lack of social benefit coverage.

As some of the articles in this volume show with regard toSpain, Germany and the UK (Anderson; Gil Araújo & Gonzalez;Gutiérrez Rodríguez) the policy provisions addressing the sphereof domestic work are mainly centered on care work. Theseprograms rarely address domestic work as awhole and the Stateleaves domestic work to private households. Thus, domesticwork continues to be socially perceived as a private matter.Further, in times of austerity measures and public spending cutsin Western Europe, the policy provisions addressing care anddomestic work are being severely cut.

Unfortunately, very little has changed since the 1970swhen the women's movements highlighted domestic workas devalued labor (cf., Benería, 1979; Dalla Costa & James,1972; Molyneux, 1979; Saffioti, 1979). This debate led to aninternational scholarship interested in questions of multi-ple and simultaneous oppression (Azerêdo, 2002; Brito daMotta, 1977; Castro, 1993; Chaney & Garcia Castro, 1993;Phizacklea & Wolkowitz, 1995; Kofes, 1994; Rollins,1985; Romero, 2002) as well as global inequalities andsocial reproduction (Bakker & Gill, 2003; Benería, 1979;Constable, 1997; Gogna, 1981; Jelin, 1977; Melo, 2002;Molyneux, 1979; Saffioti, 1979; Zurita, 1983). Continuedengagement with these issues was conducive for more recentresearch on care-chains and the reshuffling of global inequalities(cf., Assis, 2011; Caixeta, Gutiérrez Rodríguez, Tate, & VegaSolís, 2004; Constable, 1997; Ehrenreich & Hochschild, 2002;Fleischer, 2002; Hondagneu-Sotelo, 2001; Lan, 2006; Lutz, 2008;Oso & Parella, 2012; Parreñas Salazar, 2001; Pedone, Agrela,& Gil Araujo, 2012), as well as the feminization of theinternational economy (cf., Barker & Feiner, 2010; Bedford &Rai, 2010).

This special issue draws on these debates by engaging withan intersectional approach (see in this volume Bernadino-Costa)and taking a comparative perspective in looking at theorganization of paid domestic work in Europe (Spain, Germanyand the UK) and South America (Argentina and Brazil). It doesso by, on the one hand, re-addressing the question of Stateregulation in the organization of care and domestic work inprivate households, and on the other, by exploring the dynamicsof intimacy in private households through the focus on the

Page 4: Womens Studies International Forum 46 (2014)

2 Editorial

interpersonal relationship between domestic workers and theiremployers.

Highlighting local particularities and global commonalities,this issue, thus, engages with the analysis of domestic work assites of the negotiation of social boundaries (cf., Anderson, 2000;Brites, 2001; Gutiérrez Rodríguez, 2010; Hondagneu-Sotelo,2001; Kofes, 1994; Lan, 2006; Saffioti, 1979) and the examina-tion of domestic work as emotional and affective labor (Brites,2001; Gutiérrez Rodríguez, 2010; Hochschild, 2008; Illouz, 2007;Vega Solís, 2009; Zelizer, 2009). This requires an analysis of thenegotiation of intimacy in private households as reflecting thenation's structural dynamics of inclusion and exclusion based on‘race’, ethnicity and migration (see in this volume Anderson;Assis; Bernadino-Costa; Courtis and Pacecca; Gil Araújo &Gonzalez; Gutiérrez Rodríguez), as well as an understandingof the affective and emotional interactions between employersand domestic workers (see in this volume Brites, GutiérrezRodríguez; Gorban & Tizziani).

While not identical with the European situation, the socialorganization of domestic work in South America is similarlyorganized as the articles on Brazil and Argentina in this volumediscuss (Assis; Bernadino-Costa; Courtis & Pacecca; Gorban &Tizziani). As is the case in Western Europe the organization ofdomestic work in private households relies on householdmembers' private arrangements, with women in charge oforganizing this work. In poor households, domestic work ismainly delivered by the women in the households or bytheir female relatives, friends and neighbors. More affluenthouseholds, (cf., Anderson, 2000; Chaney & Garcia Castro, 1993;Momsen, 1999), opt to employ another person,mainly awoman,to deliver this work. Frequently, these women belong to aneconomically, racially, ethnically and religious subordinatedsocial group.

As all the chapters in this volume show, the reshufflingof gender, racial and class divides on a global level arere-negotiated on a local level (Anderson, 2000; Azerêdo,2002; Cox, 2006; Davis, 1983; Hondagneu-Sotelo, 2001;Kofes, 1994; Lan, 2006; Parreñas Salazar, 2001; Rollins,1985; Romero, 2002; Sharpless, 2010).

The organization of domestic work is shaped by the localprocesses of marginalization and subordination articulated inthe gender, ‘race’ and class stratification of the labor market,which migration policies reactivate these divisions in newways (cf., Assis & Siquiera, 2009; Barbosa, 2000; GutiérrezRodríguez, 2010; Herrera & Ramírez, 2008; Herrera & Yepezde Castillo, 2007; Hondagneu-Sotelo, 2001; Jelin, 1977;Kofman, 2012). Thus, migration policies create different legalresidency categories or the position of “illegal residency”,introducing differing degrees of treatment and (lack of)protection for workers. As the articles on this topic in thisissue demonstrate (see in this volume Anderson; Assis; Courtis& Pacecca; Gil Araújo & Gonzalez; Gorban & Tizziani; GutiérrezRodríguez), migrant women, employed as domestic workers inprivate households very often earn less than the nationaldomestic workers, work under precarious conditions and areopen to physical, psychological, emotional and sexual abuses(ETUC, 2005; ETUC, 2012; FRA, 2011; Lalani, 2011).

In her article Nations, migration and domestic labor: thecase of the UK, Bridget Anderson discusses private house-holds as sites of negotiations of national belonging. Engagingwith themechanisms of immigration control, Anderson explores

the symbolic function of delineating the border between “Us”and “Them” in the family by focusing on the implications of twotypes of visa for domesticwork recruitment in the UK, “domesticworker accompanying an employer” and “au pair”, for theconstruction of ideas about the family,work, and Britishness. In asimilar vein, Sandra Gil Araújo and Tania Gonzalez argue in theirarticle, International migration, public policies and domesticwork: Latin American migrant women in the Spanish domesticwork sector, that the boundaries of exclusion inserted bymigration policies impact on the private households employingdomestic workers in Spain. Focusing on the case of LatinAmerican women employed in the care and domestic worksectors, Gil Araújo andGonzalez explore the connectionbetweenmigration policies, family policies, gender regimes and theforeignization of the domestic work sector. They thus relateprocesses of feminization to migratory policies and patterns ofmigration, the features of the welfare state and the organizationof work. A similar approach to the question of feminization,migration and domestic work is outlined in Corina Courtis andMaría Inés Pacecca's article onDomesticwork and internationalmigration in Latin America: Exploring trajectories of regionalmigrant women in domestic service in Argentina. Focusing onthe case of Argentina they examine theworking conditions andbackgrounds of migrant women mostly from Peru, Boliviaand Paraguay working in the Buenos Aires Metropolitanarea.

Continuing the analysis of the relationship between migra-tion policy and domestic work, Glauciá de Olivera Assis in herarticle Gender and migration from invisibility to agency: theroutes of Brazilian women in contemporary migrations onBrazilian migrant women working as cleaners in New England,United States, stresses these women's professionalization strat-egies. Despite suffering marginalization and exclusion throughbeing made “illegal”, these women find ways to build ontheir professional experiences prior to migration in orderto improve their opportunities in the country of arrival.They experience migration as an opportunity for entrepreneur-ial ventures, resulting from their experiences with various socialnetworks. Set in this context “housecleaning”becomes a strategythat enables these women to travel and insert themselves intothe project of transnational migration.

All these articles delve into an understanding of theinterlocking of domestic work with migration regimes andprocesses of racialization, a perspective which is also addressedby EncarnaciónGutiérrez Rodríguez's article onDomesticwork –affective labor: on feminization and the coloniality of labor.drawing on research conducted in private households on therelationship between domestic workers and their employers inWestern Europe (Gutiérrez Rodríguez, 2010), this paper arguesfor an understanding of domestic work as affective labor.Expanding on feminist debates ondomesticwork's constitutivevalue for societal reproduction, Gutiérrez Rodríguez considersthe affective dimension in which social boundaries relyingon the logic of feminization and the coloniality of power arenegotiated, enacted and performed. Drawing on interviewsconducted with domestic workers and their employersthe “dis-animating”, but also “animating” energies circulat-ing in the households are discussed and situated within thesocietal context of migration policies and the organization ofdomestic work. Déborah Gorbán and Ania Tizziani's article onInferiorisation and deference: the construction of social

Page 5: Womens Studies International Forum 46 (2014)

3Editorial

hierarchies in the context of paid domestic labor focuses on theenactment of social distinctions in the encounter betweendomestic workers and their employers. Based on qualita-tive research on affluent private households in BuenosAires, Tizziani and Gorbán explore how employers setsocial boundaries and construct the domestic worker associally inferior in order to legitimize their attempt tocontrol them, but also how the domestic workers underminethe employers' imposition of superiority.

Countering subordination and subverting the stereotypesof social inferiority imposed on domestic workers by theemployers, is also a topic that Jurema Brites raises in herarticle Domestic service, affection and inequality: elementsfor a Study of Subordination. Based on an ethnographic studyconducted in Brazilian middle and upper class households,Brites discusses the concept of the stratified complementary,related to Shelle Colen's (1995) notion of stratified repro-duction, by focusing on the affective relationship betweenemployers and domestic workers. On this basis, she statesthat the social hierarchies established in the relationshipbetween domestic workers and their employers are en-hanced by an emotional ambiguity. Processes of subordinationare thus unintentionally sustained by moments of affectionexpressed by the domestic workers and employers alike, butembedded in a matrix of power relations in which thedomestic worker is still treated as a subordinate even whenshe has become the surrogate mother of the household'schildren.

As the articles by Brites, Gutiérrez Rodríguez; Gorban andTizziani demonstrate, the intimate relationships establishedbetween domestic workers and their employers are driven byboth positive and negative affects. The social hierarchies in theprivate households, circumscribed by national policies ofbelonging (Anderson); migration policies (Courtis & Pacecca;Gil Araújo & Gonzalez; Gutiérrez Rodríguez) and the persistentgendered division of work, are enacted, negotiated and pro-duced in the encounter between domestic workers and theiremployers.

We conclude the special issue with some considerationsregarding the different strategies of claiming agency andownership of domestic workers and domestic workers'organizations. As Joaze Bernadino-Costa suggests in hisarticle Intersectionality and female domestic workers'unions in Brazil, scholarship on domestic work needs toshift the focus from an analysis of multiple oppression toone that departs from intersectionality as a vantage point forpolitical organizing. Taking the experience of domestic worktrade union organizers in Brazil, Bernadino-Costa argues for aconsideration of these women's strategies of professionalizingdomestic work. By so doing, Bernadino-Costa concentrates onthese women's political strategy to form a public agenda forfemale domestic workers' unions and their negotiations withother trade unions, as well as the feminist and Afro-descendentmovements in Brazil.

Attending to the various proposals for understanding theinterconnections of migration, domestic work and affect, thisspecial issue attempts to broaden the analytical framework ofproduction and reproduction by considering the affective,emotional and caring dimension of domestic work. It alsoaims to establish a dialog between Western European andLatin American scholarship enabling us to understand the

persistence of feminization and coloniality of labor expressedin the societal organization of domestic work and throughnew systems of social classification/categorization fuelled bymigration control policies.

Endnote

1 See http://wiego.org/informal-economy/occupational-groups/domestic-workers.

References

Anderson, B. (2000). Doing the dirty work? The global politics of domesticlabour. London: Zed Books.

Assis, G. de O. (2011). De Criciúma para o mundo: rearranjos familiares dosnovos migrantes brasileiros. : Florianópolis: Editora Mulheres.

Assis, G. de O., & Siquiera, S. (2009). Muhleres emigrantes e a configuraçaode redes sociais: construindo conexões entre o Brasil e os EstadosUnidos. REMHU (Brasilia), 16, 25–46.

Azerêdo, S. (2002). A ânsia, o sino e a transversalidade na relação entreempregadas e patroas, 19, : Cadernos Pagu [online] (Retrievedfromvhttp://www.scielo.br/scielo.php?script=sci_arttext&pid=S0104-83332002000200014&lng=en&nrm=iso).

Bakker, I., & Gill, S. (2003). Power, production and social reproduction.Basingstoke, New York: Palgrave Macmillan.

Barbosa, F. C. (2000). Trabalho e Residência: estudo das ocupações de empregadadoméstica e empregado de edifício a partir de migrantes nordestinos. Niterói:Editora da Universidade Federal Fluminense.

Barker, D. K., & Feiner, S. F. (2010). As theworld turns:Globalization, consumption,and the feminization of work. Rethinking Marxism, 22(2), 246–252.

Bedford, K., & Rai, M. S. (2010). Feminists theorize international politicaleconomy. Signs, 36(1), 1–18.

Benería, L. (1979). Reproduction, production and the sexual division oflabour. Cambridge Journal of Economics, 3(3), 203–225.

Bock, G., & Duden, B. (1977). Arbeit aus liebe — liebe als arbeit: zurentstehung der hausarbeit im kapitalismus. In Gruppe Berliner Dozentinnen(Ed.), I: Frauen und Wissenschaft (pp. 118–199). Berlin: Beiträge zur 1.Sommeruniversität für Frauen.

Brites, J. (2001). Afeto, desigualdades e rebeldia — bastidores do serviçodoméstico. PhD Thesis. Social Anthropology. Porto Alegre: UniversidadeFederal do Rio Grande do Sul.

Brito da Motta, A. (1977). Visão demundo da empregada doméstica— (um estudode caso). Salvador: Curso de Pós-Graduação em Ciências Humanas daUniversidade Federal da Bahia.

Caixeta, L., Gutiérrez Rodríguez, E., Tate, S., & Vega Solís, C. (2004).Homes, Care andBorders [Hogares, Cuidados y Fronteras].Madrid: Traficantes de Sueños.

Castro, M. G. (1993). ¿Qué se compra y qué se vende em el serviviodoméstico? El caso de Bogotá: una revisión crítica. In E. Chaney, & M.Castro García (Eds.),Muchacha/cachifa/criada/empleada/empregadinha/sirvienta y… más nada: trabajadoras domésticas en América Latina yCaribe (pp. 99–116). Caracas: Ed. EPU.

Chaney, E., & Garcia Castro, M. (Eds.). (1993). Muchacha/cachifa/criada/empleada/empregadinha/sirvienta y… más nada: trabajadoras domésticasen América Latina y Caribe. Caracas: Ed. EPU.

Colen, S. (1995). Like a mother to them: Stratified reproduction and WestIndian childcare workers and employers in New York. In F. Ginsburg, &R. Rapp (Eds.), Conceiving the new world order: The global politics atreproduction (pp. 78–102). Berkley: University California Press.

Constable, N. (1997). Maid to order in Hong Kong: An ethnography of Filipinaworkers. New York: Cornell University Press.

Cox, R. (2006). The servant problem. Domestic employment in a globaleconomy. London/New York: I.B. Tauris.

Dalla Costa, M., & James, S. (1972). The power of women and the subversion ofthe community. London: Butler and Tanner Ltd.

Davis, A. (1983). Women, race & class. New York: Vintage Books.Delphy, C. (1984). Close to home: A materialist analysis of women's oppression.

Translated by D. Leonard. London: Hutchinson.Duden, B. (2009). Arbeit aus liebe — liebe als arbeit. ein rückblick. Olympe,

30, 16–26.Ehrenreich, B., & Hochschild, A. R. (2002). Global woman: Nannies, maids and

sex workers in the new economy. London: Granta Books.ETUC (2005). Out of the shadow: Organising domestic workers. Towards a

protective regulatory framework for domestic work (Brussels, http://www.etuc.org/a/5827 (12.06.2013)).

ETUC (2012). 5th annual ETUC 8March survey 2012. (Brussels, http://www.etuc.org/IMG/pdf/8_March12_analysis_FINAL_EN-5.pdf (accessed 07.08.2013)).

Page 6: Womens Studies International Forum 46 (2014)

4 Editorial

Federici, S. (2004). Caliban and the witch. Women, the body and primitiveaccumulation. Brooklyn: Autonomedia.

Fleischer, S. R. (2002). Passando a América a limpo: o trabalho de housecleanersbrasileiras em Boston, Massachussets. São Paulo: Annablume.

FRA (European Union Agency for Fundamental Rights) (2011). Migrants in anirregular situation employed in domestic work. (Vienna: Austria, http://fra.europa.eu/sites/default/files/fra_uploads/1827-FRA_2011_Migrants_in_an_irregular_situation_EN.pdf (accessed 07.08.2013)).

Gogna, M. (1981). El servicio doméstico en Buenos Aires: características deempleo y relaciónlaboral. CEIL — CONICET, informes de investigación, no.5, Buenos Aires.

Gutiérrez Rodríguez, E. (2010). Migration, domestic work and affect. London/NewYork: Routledge.

Hausen, K. (2000). Arbeit und Geschlecht. In J. Kocka, C. Offe, & B. Redslob(Eds.), Geschichte und Zukunft der Arbeit (pp. 343–361). Frankfurt/NewYork: Campus.

Herrera, G., & Ramírez, J. (Eds.). (2008). América Latina migrante. Estado,família, identidades. Quito: FLACSO.

Nuevas migraciones latinoamericanas a Europa. Herrera, G., & Yepez deCastillo, I. (Eds.). (2007). Balances y desafios. Quito: FLACSO-GRIAL-OBREAL.

Himmelweit, S. (1995). The discovery of ‘unpaid work’. The socialconsequences of the expansion of ‘work’. Feminist Economics, 1(2),1–19.

Hochschild, A. R. (2008). La mercantilización de la vida íntima. Madrid: KatzEditores.

Hondagneu-Sotelo, P. (2001).Doméstica: Immigrantworkers cleaning and caringin the shadows of affluence. Berkeley: University of California Press.

Illouz, E. (2007). Cold Intimacies: The Making of Emotional Capitalism.Cambridge: Malden: Polity Press.

ILO (2012). Domestic workers across the world: Global and regional statisticsand the extent of legal protection. Geneva: ILO.

Jelin, E. (1977). Migration and labour force participation of Latin Americanwomen: The domestic servants in the cities. Sings, 3(1), 129–141.

Kofes, S. (1994). Entre nós mulheres, elas as patroas e elas as empregadas. InA. Arantes, & A. Augusto (Eds.), Colcha de retalhos: estudos sobre a famíliano Brasil (pp. 183–193). Campinas São Paulo: UNICAMP.

Kofman, E. (2012). Rethinking care through social reproduction: Articulatingcircuits of migration. Social Politics, 19(1), 142–162.

Lalani, M. (2011). Ending the abuse. Policies that work to protect migrantdomestic workers. London: Kalayaan.

Lan, P. C. (2006). Global cinderellas: Migrant domestic and newly rich employersin Taiwan. Durham: Duke University Press.

Lutz, H. (2008). Migration and domestic work. A European perspective on aglobal theme. Aldershot: Ashgate.

Melo, H. (2002). Trabalhadoras Domésticas. UNIFEM: Revista do Fundo deDesenvolvimento das Nações Unidas para a Mulher, 4. (pp. 2).

Melo, H., Pessanha, M. E., & Parreiras, L. (2005). A economia política doserviço doméstico remunerado: Rendimentos e luta sindical. Mulher eTrabalho (pp. 5) (Retrieved from http://revistas.fee.tche.br/index.php/mulheretrabalho/article/viewFile/2715/3038).

Molyneux, M. (1979). Beyond the domestic labour debate. New Left Review,115, 3–28.

Momsen, J. H. (1999). Gender, migration and domestic service. New York:Routledge.

Inmigración, género y Mercado de trabajo. Oso, L., & Parella, S. (Eds.). (2012).Cuadernos de Relaciones Laborales, 30(1).

Parreñas Salazar, R. (2001). Servants of globalization: Women, migration anddomestic work. Stanford, CA: Stanford University Press.

Políticas públicas, migración y familia. Una mirada desde el género. Pedone, A.,Agrela, B., & Gil Araujo, S. (Eds.). (2012). Papers: Revista de Sociología, 97(3).

Pérez Orozco, A. (2010). Global perspectives on the social organization ofcare in times of crisis: Assessing the situation. Gender, migration anddevelopmentWorking paper, 5, . : INSTRAW.

Phizacklea, A., & Wolkowitz, C. (1995). Homeworking women: Gender, racismand class at work. London: Sage.

Rollins, J. (1985). Between women: Domestics and their employers. Philadelphia:Temple University Press.

Romero, M. (2002). Maid in the USA. London/New York: Routledge.Saffioti, H. (1979). Emprego Doméstico e Capitalismo. Petrópolis: Vozes.Sharpless, R. (2010). Cooking in other women's kitchens: Domestic workers in

the South, 1865–1960. Chapel Hill: The University of North CarolinaPress.

Vega Solís, C. (2009). Culturas del cuidado en transición. Barcelona: EditorialUOC, Niberta.

Waring, M. (1999). Counting for nothing. What men value and what womenare worth. Toronto, Buffalo, London: University of Toronto Press.

Zelizer, V. (2009). La negociacion de la intimidad. México: Fondo deCultura.

Zurita, C. (1983). El servicio doméstico en Argentina. El caso de Santiago delEstero. Santiago del Estero: INCIC, Universidad Católica de Santiago delEstero.

Encarnación Gutiérrez RodríguezCorresponding author.

Jurema Brites

Page 7: Womens Studies International Forum 46 (2014)

Women's Studies International Forum 46 (2014) 5–12

Contents lists available at ScienceDirect

Women's Studies International Forum

j ourna l homepage: www.e lsev ie r .com/ locate /ws i f

Nations, migration and domestic labor: The case of the UK

Bridget AndersonCentre on Migration, Policy and Society, University of Oxford, 58 Banbury Road, Oxford, OX2 6QS, United Kingdom

a r t i c l e i n f o

0277-5395/$ – see front matter © 2014 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.wsif.2014.01.005

s y n o p s i s

Available online 4 February 2014

Immigration controls serve a crucial symbolic function of delineating the nation and thepeople, the boundaries of ‘Us’ and ‘Them’ and state legitimacy is, through immigration policy,linked to ideas of nation building and preservation (Honig, 2003). Labor migration policy is notsimply an instrumental response to the needs of employers but highly symbolic and politicallycontested terrain that assumes intense public significance. This is particularly evident in thecase of domestic workers, who are embedded in the family, the ‘heart of the nation’. This paperexplores the ways in which migration policies on domestic work not only produce asubordinated workforce, but reflect and construct ideas about family, work, and Britishness,with a particular focus on two visa types: domestic worker accompanying an employer and aupair visas.

© 2014 Elsevier Ltd. All rights reserved.

Introduction

In 2002 Wimmer and Glick Schiller famously alerted us to‘methodological nationalism,’ ‘the assumption that thenation/state/society is the natural social and political formof the modern world.’ They argued that mainstream socialscience was infused with methodological nationalism, andthat the study of migration in particular had been limited bythis. The significance of the study of ‘transnational commu-nities’ lay in this epistemic move away from methodologicalnationalism rather than in the identifying of new objects ofobservation (Wimmer & Glick Schiller, 2002). They wereabsolutely right to identify the importance of de-naturalizingthe nation state, and to point out the ways in whichintellectual work and research participates in the construc-tion of the categories of state, nation, migrant and citizen. Inthis paper, I want to consider the implications of this forresearch on migrant domestic workers, but I also want toargue that resisting methodological nationalism doesn'tmean we should not take the nation seriously. On thecontrary, we must acknowledge its importance in order tosee beyond it.

Nationalism is integrally related to the projects of raceand gender, which all are permanently ‘under construction.’

ll rights reserved.

How ideas of the nation are linked to states' naturalizationpolicies has received some attention, but much less has beenpaid to how these are linked to states' discourses about theirimmigration policies.1 In the UK, social relations are increas-ingly dismissed by social policy: those in receipt of housingbenefits are required to move from neighborhoods wherethey have lived for years if their rent is deemed tooexpensive, single parents are being required to look forwork and so on. What matters in an age of austerity, it seems,are questions of economic costs and benefits. Yet nationalsocial relations have increased prominence. So governmentrhetoric dictates that employers' principal considerationshould not be the most efficient, profitable worker, butwhether or not they are British — though of courseimmigration policy is more complicated because of EUcitizenship. In this paper I want to consider how policies onimmigration contribute to nation building in the UK. Moreparticularly, I'm interested in how policies on immigrationand domestic work reflect and reproduce ideas aboutBritishness, Britain, the family and work. This is a comple-mentary approach to the research that argues for theimportance of both acknowledging the UK's need fordomestic workers and understanding why it is that thisdemand is met so overwhelmingly by migrant women. I'll

Page 8: Womens Studies International Forum 46 (2014)

6 B. Anderson / Women's Studies International Forum 46 (2014) 5–12

begin by considering the relation between immigration,states and nations, and the production of ‘us and them,’looking at how this is manifest in UK policy documents. I'llthen examine the two UK visas that have been available fordomestic work, looking at how they encapsulate certainassumptions about domestic work in the UK, about therelation between family and work, and ideas of equality,slavery and freedom.

States, immigration and nation

Nations are imagined as groups of people who share acommon culture, language and history. They are, as describedby Benedict Anderson, ‘imagined communities’ because ‘themembers of even the smallest nation will never know mostof their fellow-members, meet them, or even hear of them,yet in the minds of each lives the image of their communion.’Because of their association with history and with kin, thereis a strong element of ethnic myth around nation-ness,whether the promotion of nation as nationalism is inopposition to colonialism, or in opposition to the perceivedintrusion of ethnic outsiders.

Despite its sometimes archaic feel, the nation continues tohave resonance in contemporary liberal democracies. Statesare represented as having to act in the ‘national interest,’which is imagined as different from the interests of the stateper se. It is important to recognize that when it comes toimmigration policies, states must be seen as prioritizing theinterests of the ‘nation’ and ‘the people’ in ways that gobeyond simply a response to the demands of capital.Notoriously, in 2009, then Prime Minister Gordon Browncoined the phrase ‘British jobs for British workers’. This wastaken up by a wide range of political actors, includingelements of the trade union movement and Far Right parties.Its logic underpins immigration policy and rhetoric, whichconstructs migrants as a residual labor force, to be tappedonly when states lack the skills or otherwise cannot fillparticular vacancies. Of course in practice it is not easy to saywhat is a ‘British job’ given the complex and multinationalrelations of global capital. ‘British worker’ is equally tricky,particularly as those with settlement status and EU nationalsmust not be discriminated against because of their country ofbirth. Nevertheless the call reveals that ‘national interest’ isbound up with ‘national identity’ as much as with GDP andbalance of payments. Immigration policy is not only func-tional but a highly symbolic and politically contested terrainthat assumes intense public significance.

Liberal theorists who are concerned with justifying immi-gration controls often point to the importance of borders fordelineating the ‘community.’ They argue, for instance, thatpolitical and legal institutions embody values that atomizedindividuals cannot generate themselves (Miller, 1995). Impor-tantly, modern liberal democratic states portray themselvesnot as arbitrary collections of people hung together by acommon legal status and/or simply by common descent but ascommunities of shared value. Nation states are therefore notjust legally constructed but are communities of value. Onepolitical challenge for liberal immigration states is that thewayinwhich people are legally constructed, (i.e. formal citizenry, asthose who cannot be refused entry or deported because theirclaim to belong is recognized) cannot be translated to fit the

powerfully imagined category of the community of value.There is tension between the idea of belonging to ‘the people’by birth and belonging as a result of upholding certain values.The Good Citizen is in part constructed by immigration andcitizenship law (Honig, 2003), but by a whole lot more as well.The details are clearly context dependent, but the Good Citizenis law-abiding, hardworking, white, and heterosexual. They areliberal sovereign selves that are independent but rooted in acommunity, a community that is, importantly, unchallenged bygender, race and class relations. Not all natural born citizens areGood Citizens.

Citizenship acquisition by migrants is symbolically im-portant because it is a moment when the state, acting in the‘national interest,’ is seen to actively influence the composi-tion of the population on the territory of the state and in theheart of the nation, when the borders between ‘us’ and ‘them’

are revealed as unstable and therefore all the more importantto fix. Thus one moment where the (purported) key values ofstate membership can be discerned lies in the requirementslaid down for admission to citizenship. What states require ofnaturalizing citizens (such as lack of a criminal record,knowledge of the state's history, commitment to certainvalues, use of the language, ethnicity) offers a picture of thenormative content of citizenship. The non-citizen who isallowed access to citizenship must be the right kind ofperson.

Good citizenship is not only asserted through naturalizationprocesses, but also through controls over entry and exit.Attention to the borders of immigration and citizenship revealshow ‘we’make sense of ‘ourselves’ (Gutiérrez-Rodríguez, 2012).The state must be seen as ensuring the entry of the right type ofperson and excluding others, most obviously the terrorist,but also low-skilled workers, those who cannot integrate, orwho do not make the right kind of contribution. As authorslike Sara Van Walsum (2008), Mae Ngai (2004) and EithneLuibhéid (2002) have brilliantly demonstrated, gender andsexuality are important components of this fixing. Only the‘right’ kind of women, mothers, daughters, and workerscan be allowed entry onto the territory and into citizenship.The myth of ‘common origin,’ the imagining of the nation asfamily writ large, the role of women as reproducers of thenation and as bearers of cultural authenticity, and their rolein the ‘backward look’ to tradition and the past, have beenthe subject of a sophisticated literature on the complex andmutually constitutive relations between gender, ethnicityand nationalism.

Care work and labor migration

In this way, states are constrained in their responses tothe demands of capital and capitalism. States cannot be seento straightforwardly deliver low-wage, disposable migrantlabor to employers, because of the demands of nationalism.Immigration controls aren't simply instrumental responses tothe needs of capital, but are important to state legitimacy,which is, through immigration policy, linked to ideas ofnation building and preservation. The call ‘British jobs forBritish workers’ is a normative demand, calling upon thestate to recognize immigration and nationality as the crucialand defining factors in determining who ought to have accessto employment. Importantly, it also acknowledges, even if

Page 9: Womens Studies International Forum 46 (2014)

7B. Anderson / Women's Studies International Forum 46 (2014) 5–12

only by implication, that labor markets are or can be sociallyconstructed and are not only the domain of liberal sovereignselves making rational choices.

The unraveling of the ways in which immigrationcontrols interrelate with other ‘national level’ factors toproduce certain types of labor demand, social and employ-ment relations has been an important contribution madeby the growing interest in migration, care and domesticwork since the late 1990s. From the migration studies'viewpoint, what has been particularly productive about thisstrand of work is the manner in which it has been integratedinto examinations of social and economic change in so-called‘receiving countries’. There has been considerable interest fromacademics about how demand for particular types of paid carehas been shaped by the feminization of labor markets, agingpopulations and changes in the welfare state (Cangiano,Shutes, Spencer, & Leeson, 2009; Cox, 2006; Orloff, 2006;Williams & Gavanas, 2008; Bettio, Simonazzi, & Villa, 2006; seeWilliams, 2010 for a review article).

The work of Williams and Gavanas (2008) has beenparticularly important in this regard, as they set out aframework that facilitates an analysis of how the intersec-tions between regimes of migration, welfare and employ-ment shape demand and discourses around care work. Theparticularities of this vary from state to state of course, but itdoes seem to generally be the case that, firstly, demand forlow-wage care workers is structurally embedded in manyEuropean states' organization of care provision; secondly,that very different policy regimes result in ostensibly thesame outcome: significant numbers of migrants working inthe eldercare sector in particular (though also childcare)2;and thirdly, that there is a disconnection between thispractice and states' increasingly restrictive migrationpolicies. The majority of these migrants, though not all,are women, and this is discussed in terms of the genderedideology of care and gender relations more generally(Parreñas, 2001; Yeates, 2009). The ideology of genderthat underpins the way that care is organized at thehousehold and national level has received some attention.There has also been some work on racialized ideologiesand care (Glenn, 1992; Anderson, 2000), however, lessattention has been paid to the ideology of nation, andhow this structures the migration regime that shapes theentry and the living and working conditions of migrantworkers, and especially of those who work in care. Whileviewed as ‘exceptional’ because it often does not fit theconventional models of ‘employment’ and contract thatunderpin the way waged labor is organized, care work, in themost general sense, has always been done, and in this respectisn't exceptional at all. Arguably, in fact, what is surprising isthat we tolerate an analysis that has such difficulty inaccommodating care relations. Domestic and care work arenot some leftover, but precede ‘normal’ employment (Fraser,2011). As wage labor has become normalized and regulated,and the wage earner constituted as the normalized subject, wehave forgotten just how human the relations of care are, andhow limiting our ideas of ‘the job.’ This should remind us that itis not only the domestic worker who is marginalized, butunpaid female labor, the unemployed, the informal, ‘suchfigures are spectres outside its political economy domain’(Denning, 2010).

Immigration, Britishness and domestic work

UK immigration practice is infused with references to‘Britishness.’ Consider the foreword to the Home Office 2007Enforcement Strategy document, Enforcing the Rules:

Britain is a country where people work hard, play by therules, speak English, and get on through merit. It has a proud,centuries-old record of integrating immigrants from aroundthe world and, many times down the years, it has becomehome to communities fleeing persecution… the fact thatmany immigrants, at the end of their journey, end up inshadowy jobs in the grey economy undermines the termsand working conditions of British workers. That's not fair. Itchips away at the social contract and fabric of our country.Resentment of it breeds discontent and racism. This isespecially keenly felt among those who believe they are notgetting the economic or social opportunities they shouldbecause others, who have flouted the rules and often the law,seem to be getting on ahead of them.That's not fair either.

The concept of the kind of place that Britain is, wherecertain values and an ill-defined ‘British way of life’ prevail,can be found throughout policy documents. We can see howBritain is presented as a place of justice and it's interestingto see the prefiguration of current debates on ‘fairness’,3

with a strong social fabric, sensitive to the needs of itsnational workers, intent on combating racism, thoughaccepting that there may be misperceptions (‘believe theyare not getting,’ ‘seem to be getting on’). The reproduction ofBritishness is also manifest in practice: the UK now has acitizenship test (which many UK born nationals might finddifficulty in passing) and citizenship changes will make‘integration into the British way of life’ a requirement fornew citizens.

Immigration controls are not then simply about allowingin or out individual units of labor, family members, peoplefleeing persecution and so on, they are about admitting ‘goodworkers’ (who are ‘skilled’, who are financially stable), ‘goodspouses’ (who speak English and are ‘able to participate inBritish life from the outset’ (HO Press release 26th July2010)), and ‘genuine asylum seekers.’Who counts as suitableentrants in this way reflects and (re)produces ideas of Britainand Britishness, ideas of what counts as skill, who is a netcontributor, what are British values, etc.

Given the centrality of ideas of home, family and genderto nationhood, it is interesting to consider the tensionsand contradictions of specific immigration policies in thisregard. The position of migrants who do domestic work inprivate households sheds light on how state legislationand rhetoric manage and reproduce contradictions aroundfamily, work, public and private, citizenship and member-ship, and how these are then experienced by individuals andhouseholds. Like soldiering, the work of making homes is akey site for the production of nationhood. Families aresupposed to be where nations are constructed, throughblood, affection and sacrifice, they underpin ‘national life,’yet interestingly (again, as with soldiering), work in privatehomes is not a sector that is considered to be unsuitable fornon-citizens.

Page 10: Womens Studies International Forum 46 (2014)

8 B. Anderson / Women's Studies International Forum 46 (2014) 5–12

In the UK, domestic work is not a standard category ofemployment. It is not incorporated into employment legisla-tion as it is in some other states (though even then it is oftensubject to discriminatory provision). Domestic labor isinformalized as a result of economic and social processes,and there is a delicate ‘equilibriumof transgression’ thatmeansthat the widespread informality and illegality in the sector isgenerally not considered a problem. As with, for example,marijuana use by middle-class teenagers, middle-class parentsmay be paying a visa overstayer cash in hand to clean theirhome and this kind of illegality is not really regarded as a causefor concern. It is endemic, and for this very reason, publicfigures have run into problems when they have beendiscovered: in the same way that Bill Clinton famously ‘didn'tinhale,’ Baroness Scotland ‘never knowingly employed anillegal immigrant’ (http://www.theguardian.com/uk/2009/sep/17/attorney-general-sacks-maid). ‘Domestic worker’ is not astandard category of immigration entry unlike in Canada,several states in Asia including Hong Kong and Singapore, andsome European states. Thismeans thatmanymigrants who areworking as domestic workers are working (often perfectlylegally, at least in terms of immigration status) on student anddependant visas— the latter if they can't find work designatedas skilled. There are also migrants working in breach ofconditions (students working more than 20 h a week, asylumseekers, etc.) and overstayers. Thus most of themigrants doingdomestic work in the UK are hidden in both employment andimmigration status terms and the relation between domesticwork and migration is largely invisibilized.

‘Migrants,’ by which I mean people who do not have theright of abode in the UK (this group includes EU nationals aswell as foreign nationals who are subject to immigrationcontrols), often work as live-in domestic workers, but formany years the only visas available for domestic work werefor au pairs and domestic workers accompanying theiremployers, i.e. jobs where the work was imagined aslive-in. The visas signaled an acknowledgement that live-indomestic work does happen in the UK, but is imagined asexceptional. Thus the two visas have been devised to dealwith quite specific situations. Nevertheless, they have beenroutinely reformulated and these shifting categories arehighly dynamic, suggesting both the difficulty of incorporatingdomestic labor into the standard model of immigrationcontrols and the difficulty of accommodating the idea of workwithin the ideal of the family.

Au pair visas

The au pair visa was used for many years, starting in thepost-war period until its abolition under the Points-BasedSystem in 2008. While there are no longer au pair visasavailable, there are still lots of young people, mainly women,who are au pairs, and the relation is still shaped by pastimmigration rules and a long history of European exchange ofyoung people. As a practice, the idea of young people movingto work in households as a transition to setting up their ownfamily has a long European history. In the early modernperiod of north western Europe, late marriage meant a longgap between puberty and marriage, and in the 16th to 18thcenturies, an estimated three-fifths of English young peoplewere sent to work in other households under a system of

‘life-cycle service,’ a period of protected transition from thefamily home to adulthood. For the majority of life-cycleservants, this was a transient phase, and while elites did notsend out asmany young people as they hired,muchmovementwas lateral or nearly lateral across classes. Young peoplemoved into the homes of others of the same, or perhaps slightlybetter, social status and relations were familiar but they wereregarded as ‘junior members,’ subordinate but integral (Cooper,2004). They performed a range of tasks including haulingwater,doing laundry, gatheringwood and looking after children. Thesetasks were gendered, with childcare being predominantly thejob of female life-cycle servants.

This bears a strong similarity to the way that the au pairregime is imagined: it is the movement of young people, inmodern times, mainly females who are childcarers, it istransient and imagined as middle-class. The notion of aupairing as ‘cultural exchange,’ I would suggest, echoes thesetypes of historical practices and is related to a life stage inimmigration practice and in the imaginations of the au pairand the host family. This is an arrangement whereby oneperson's stage in the life cycle, as young, without dependants,inexperienced, flexible, fits another household's life stage —

overstretched, in need of ‘help’ with demanding youngchildren. It draws on an idea of the home as a particularsafe space for girls before they settle down, offering them theopportunity for ‘social progress’ without exposing them tothe dangers of the market. The 1969 Strasbourg Conventionformalized this process by marking an agreement betweencertain European countries to facilitate cultural exchange foryoung people at the same time as providing help to familieswith young children. In the preamble, the Conventionemphasized that au pairs do not belong to the ‘workercategory’ and described the movement of au pairs asconstituting ‘an important social problem with legal, moral,cultural and economic implications,’ which required specialmoral protections (Preamble Strasbourg Convention).

In the UK (not in fact a signatory to the StrasbourgConvention), the immigration status of au pairs was limitedto people coming from named European states aged betweenseventeen and twenty-seven. Applicants had also to besingle, with no dependants, and coming for the purpose of‘cultural exchange.’ The relation between au pair and hostfamily was explicitly one of fictive kin, and they wereexpected to live as ‘part of the family.’ Host families providedaccommodation and board and in return, the young personcould be expected to ‘help out’ for a maximum of 24 h plustwice weekly babysitting, in exchange for ‘pocket money.’ Aupair visas could not be renewed and were valid for amaximum of two years (with a possible six month exten-sion).4 The only possible ‘switch’ was to a spouse visa.Immigration rules both reflected and constructed notions ofwhat an au pair ‘is’ and in the UK this meant young, European(often then coded as ‘white’ and Christian cf. Cox, 2007;Newcombe, 2004), and, until 1993 when the rules changed,female.

They also made requirements of the family that is suitablefor ‘cultural exchange’: they had to be English speaking, havea spare room and not need too much ‘help.’ If they had apre-school age child there must be another caretakeravailable, presumably either a parent/mother or a nanny.While the immigration rules described the contours of the au

Page 11: Womens Studies International Forum 46 (2014)

9B. Anderson / Women's Studies International Forum 46 (2014) 5–12

pair and host families, these were filled in by agencies whoseliterature illustrated the wholesome au pairs and happyfamilies they were to be welcomed into (Cox, 2007). It wasnevertheless generally the au pair who was monitored andregulated rather than the host family, and this was true forboth immigration controls and agencies. In practice it seemsthat there was considerable ‘abuse’ of the au pair system byhost families. Research conducted in 2004/5 found that manyhard-pressed families were very different from the au pairideal, using au pairs as the cheapest available form ofchildcare, e.g. for single mothers working shifts. Moreover,au pair agencies reported considerable difficulties in manag-ing the assumptions of whiteness of both host families and aupairs. The latter caused particular problems as host families'photographs, unlike au pairs', were generally not required foragency registration (Anderson, 2007).

Immigration rules not only regulated the type of peoplewho could be subject to the au pair arrangement, but alsodescribed the kind of relationship that would pertain. The aupair visa institutionalized three ideas around the role of youngnon-citizens in British family homes: firstly, that that theywere‘equal,’ secondly, that au pairingwas not work and thirdly, thatthey were temporary. While ‘au pair’ quite literally designates‘equal’ in French, the question of who, in practice, the au pairwas equal to, can be unclear. In the context of a history oflife-cycle service, a certain equality of familial status might bereferred to, that is, an arrangement between equal family types,one middle-class family might host an au pair and send theirdaughter as an au pair to the other's parents' family. However,internally, families are not equal institutions but intenselyhierarchical. Were au pairs equal to big brothers, sisters,aunties, mothers or fathers? Moreover, the ‘equality’ onlyextended to within the home. Unlike other young people livingin the household, the au pair was not allowed to take on anyform of paid employment, as doing so would breach immigra-tion conditions. The family is imagined then not only as a spaceof equals, but as the private sphere, completely separated fromthe market, and what happens in public does not impinge onthe relations within the private sphere.

Au pairs were not ‘migrant workers.’ There was animportant relation between ‘equality’ and not working. Thetasks performed by au pairs were not work because theywere performed as ‘part of the family.’ This was not ‘work’because au pairs were ‘equal.’ ‘Equality’ signified not gettingpaid: wives and daughters don't get paid, only servants, whoare not equal. Thus the invisibility of the economic basis ofthe household was maintained, even as non-family memberswere accommodated. Au pair agencies were extremely activein lobbying the Labour government in the processes up to theintroduction of the 1998 National Minimum Wage Act. Theysuccessfully secured an exemption for people who werebeing paid but living as ‘part of the family’ (the ‘familyworker exemption.’) The Immigration Directorate's Instruc-tions on au pairs at the height of the au pair policy suggestthat the amount of money allocated to an au pair was animportant indicator of the nature of this relationship.‘Reasonable allowance’ was defined as up to £55 a week, as‘[a]ny sum significantly in excess of this might suggest thatthe person is filling the position of domestic servant’ (IDIMarch 2004 Chapter 4 Section 1, Annex A paragraph 4). Workin households was equated with servitude rather than the

‘home making’ of the au pair and family members. In additionto equality and not working, immigration controls imposedtemporariness as a condition of the au pair visa. Temporarinessis in keeping with the practices of life-cycle service and it alsofacilitated the delicate imaginative and emotional balancing ofmaintaining the relations of fictive kin (Anderson, 2007).

The idea of the ‘au pair’ depended for many years on thearchitecture of immigration controls. This began to changewith the free movement of European citizens, but the schemeexpanded in the early 2000s to include citizens of Central andEastern European states who still required a visa to come to theUK. On the eve of EU Enlargement in 2004 most au pair visaholders came from the new accession states, particularly theCzech Republic. However, following the EU enlargement,nationals from the new accession states (A8 nationals) werefree to work in the UK wherever they wanted, which causedconsiderable anxiety for many host families (Anderson, 2007).Until May 2011 all A8 national workers, however, had toregister with the ‘Workers Registration Scheme.’ Interestingly,Workers Registration Scheme data puts ‘domestic staff’ in thetop five occupations for A8 nationals, but ‘au pairs’ who areliving with a family do not have to register. While this meansthat this dataset is likely to underestimate the importance ofdomestic work to new arrivals, it is also significant that somedomestic workers/au pairs did register with the WRS. Theimportance of au pairing not being constructed as work seemsto have diminished with the opening of EU borders, whenpeople attained (alleged) equality of citizenship status.

This is confirmed by the position of Romanian andBulgarians(A2 nationals). There has been a significant growth in thesenationals working as au pairs in the UK5 beginning in early2004, just before the EU Enlargement gave A8 nationalsunrestricted access to the labor market. At that time, au pairagencies reported host families specifying that they wantedRomanian, Bulgarian or Turkish au pairs because they can'tlegally ‘run off,’ and the agencies in turn were recommendingvisa nationals: ‘MyRomanian is going up… because Romanianscan't get other jobs’ (Anderson, 2007). Romanians andBulgarians, however, found that, until 2014 should they wishto be au pairs, it was now considered a ‘category ofemployment’ and, although they did not need a visa they didrequire an accession worker card. These ‘au pairs’ were thusconstructed as workers but were highly constrained. Inparticular, unlike previous au pair visa holders, they couldwork only for the named employer on their accession workercard, because they were ‘workers.’ However, unlike holders ofaccession worker cards working in other sectors, ‘au pairs’ didnot qualify for minimumwage. Thus theywere workers and, atthe same time, not workers.

The au pair visa, (with the exception of the accessionworkers card) was abolished in November 2008. The schemewas incorporated into the Youth Temporary Mobilityscheme, which is applicable to only eight states and isaimed at those who ‘wish to experience life in the UK.’ Thosewho enter under this scheme may both work and undertake‘au pair placements,’ but what precisely is meant by ‘au pairplacement’ is, unlike before, not defined. There is no need todistinguish au pairing from working in private households,because this group is allowed to work. Thus, the discursiveidea of the au pair has a strong hold and the disappearance ofthe visa has not meant a disappearance of the practice and

Page 12: Womens Studies International Forum 46 (2014)

10 B. Anderson / Women's Studies International Forum 46 (2014) 5–12

type of arrangement. There are still large numbers of ‘au pair’agencies and it continues to signify a type of arrangementthat is not straightforwardly an employment relation, oftenrelying on fictive kin, to distinguish itself from a servicerelation. Questions like ‘How much do you pay your au pair?’and ‘Can an au pair live out?’ continue to attract multipleInternet postings. The current dominance of non-EU 15 statessuggests a shift in the equality of familial status aspects of aupairing: British nationals registered with au pair agenciestend to be looking for families in Austria, Spain and Holland,or further afield in Canada, the USA or Australia. There arevery few looking for places in Czech or Poland.

The history of the au pair visa demonstrates the shifting butmutually dependent nature of citizenship, family, work andgender, and the attempts to accommodate this within liberaldiscourses of equality. While the au pair visa has, for themoment, largely disappeared from theUK (it is still an importantyet under-researched component of immigration controls inother European states such as Sweden and Denmark), thecontradictions and tensions that gave rise to the immigrationstatus continue to underpin domestic arrangements in unknownnumbers of British households.

Domestic worker visa holders

While the au pair visa was issued for non-citizens livingwith British families, the domestic worker visa was designedfor non-citizens entering the UK and living with non-citizenfamilies. It began as an immigration concession outside therules in 1977. This followed a general tightening in immigra-tion controls, which meant that ‘low-skilled’ workers werenot given visas to work in the UK. The domestic workconcession encompassed a wide range of visas. Some wouldbe given a visitor's visa with an ‘employment prohibited’stamp, others had permission to enter to work for theiremployer, with the name of the employer written on theirvisa, while others were given permission to enter as familymembers. The practice seemed to depend partly on theindividual visa-issuing officer but also it later transpired onthe point of origin of the family and worker. Moreparticularly, African domestic workers accompanying fami-lies from Africa were usually given family member visas,while Filipinos accompanying families from the Middle Eastwere given visas with the name of the employer on them.

The employers of these domestic workers were generallywealthy foreign nationals, though they did also includereturning British expats. The abuse and exploitation thatwas a consequence of this kind of employment relationgradually became public, and a lively campaign against theimmigration arrangements was expressed as the impositionof ‘slavery’ in modern times. I must confess that my firstpublication, Britain's Secret Slaves: migrant domestic workersin the United Kingdom (Anderson, 1993), contributed to thiscampaign. The campaign generated considerable publicityand engaged the popular imagination. While it was recog-nized, even at the time, that there was a problem with thepresentation of foreign employers (often Arab) ‘importingslavery’ and barbaric practices, public sympathy seized onimages of abusive (male) employers, and this also generatedconsiderable parliamentary support. These families, in starkcontrast to the host families of au pairs, were not imagined as

safe spaces for women, but as cruel and exploitative placeswhere the threat of physical and sexual abuse was everpresent. Moreover, the contrast represented the UK as a siteof free labor and space of equality that is (nearly) free from‘slavery,’ where any sign of the slave trade would beruthlessly stamped out. Following any high profile publicevent, the domestic worker support group, Kalayaan, wouldreceive calls from well intentioned members of the Britishpublic, offering opportunities for workers who had beenabused to come and work for them. These opportunitieswould often be unpaid, but offering rooms with ‘a lovelyview’ or the chance to be with ‘a really nice family.’ This idealof the UK as a place of free labor, of British families as havenswas a feature of responses to the campaign in the 1980s and90s that has continued to the present day. As seen in theprevious citation from Enforcing the Rules, Britain is not thekind of place where slavery takes place, British people (whoare, after all, ‘fair’) are not the kind of people who put up withslavery, let alone benefit from the labor of ‘slaves.’

By tying domestic workers to their employers, the Britishgovernment was caught between a very public falling shortof liberal democratic ideals and ideas of ‘Britishness,’ and thedemands of a very specific group of wealthy employers:

Looking at our national interest, if wealthy investors,skilled workers and others with the potential to benefitour economy were unable to be accompanied by theirdomestic staff they might not come here at all but taketheir money and skills to other countries only too keen towelcome them.

[Lord Reay, 1990]

The response to this campaigning was to print leaflets,require pre-clearance interviews and require that domesticworkers had been employed for twelve months previous tocoming to the UK. However, reports of abuse continued, andin 1998 the recently elected Labour government finallyagreed to give domestic workers accompanying their em-ployer the right to change employer in the case of abuse, theright to settlement and to be subject to employmentprotections. Since then, they must be paid minimum wage,have holiday pay, etc. They can also, in theory, accessemployment tribunals and join trade unions. But a crucialrequirement of this is that, unlike au pairs, they are not‘family workers,’ i.e. are not ‘employed6 to live as membersof the family’ (House of commons library research paper 99/18, 1999: 4). In contrast to au pairs, domestic workersentering the UK accompanying their employer are con-structed as workers. Work in this case does not signifyservitude and inequality with mothers, sisters and daughters,but is rather a means to acquire social dignity. Work signifies‘not slavery’ and free labor. While au pairs must be distin-guished from servants, this category of migrant must bedistinguished from slaves.

Rather than life-cycle service, this model is closer to themodel of the nineteenth-century English domestic worker,wherein those who were still resident were physically andpsychically more distanced from the families they lived with,where class (for some domestic workers now, race), ratherthan life stage was what determined one's role in the family.Middle and upper classes were no longer providing servants,

Page 13: Womens Studies International Forum 46 (2014)

11B. Anderson / Women's Studies International Forum 46 (2014) 5–12

even before the post World War 1 decline, and poorer housesincreasingly had no paid workers at all. These families wouldno more consider ‘exchanging’ their children with thehouseholds of those providing servants than British familieswould consider their sons and daughters going to work asdomestic workers for, say, Indian families in India. Impor-tantly however, this kind of commodified relation isclassified as work. Employers must demonstrate evidenceof written terms and conditions of employment, must agreeto provide a separate bedroom for their worker, or, if theyare not to live in, that the worker will have access toadequate accommodation. Those who had entered under theprevious regime were to be regularized as they had ‘becomeillegal through no fault of their own.’ The latter is an interestingdescription as it does intimate the state construction ofirregularity at the same time as circumscribing it — stronglyimplying that there are others who become ‘illegal’ throughtheir own fault. Thus, at the very moment of winning theircampaign, the power of the state to set borders is reinscribed(Nyers, 2010).

There is a tension revealed in the affirmation that domesticwork iswork: is the call for domesticwork to be formalized andregulated as other types of work, or does it call for specialmeasures? While the call of the workers and the campaignerswas for an end to their invisibility through the recognition ofdomestic work as work, in practice the demand was for aspecific regime. At the time of the campaign and visa change,there were very limited routes of entry for the purpose of‘low-skilled’ work and work permits were available only for‘high-skilled’ jobs. The gendered nature of ‘skill’ and its relianceon specific specialisms and measurable achievements makes itpeculiarly unsuitable for capturing the requirements ofdomestic labor. Indeed, the roots of the concept of ‘skill’ areprecisely that it is not the general requirements of repro-ductive labor and the management thereof (Cockburn &Ormrod, 1993). However, being cast as ‘low-skilled’ had itsadvantages at that time, since work permits tied workers toemployers. The tying of workers to employers did not anddoes not, have the connotations with slavery and forcedlabor that it does in the private household, even thoughthere is evidence that it can have a significant effect onemployment relations, and as Robert Miles has argued, callinto question whether this is really ‘free labor’ (Miles, 1987;Anderson, 2010).

The fact that domestic work is not considered high-skilledmeant that, despite it being recognized that theywereworkers,they were ‘anomalous workers’ in that they were permitted toenter even though they were considered low-skilled, andconsequently were not tied to their employers. However,because it is considered unskilled, the hard-won right to betreated as workers came under pressure only ten years later in2008. The domestic worker visa could not be accommodatedwithin the new Points-Based Immigration system. Given thatthe UK was no longer supposed to be granting visas to‘low-skilled’ non-EU migrants, the domestic worker visarepresented an anomaly. Rather than a working visa, it wasproposed to issue a six-month ‘business visitor visa,’ whichwould not give domestic workers any employment rights andwould tie them to the employer with whom they entered. Thisvisa would not be renewable and would be issued for thepurpose of enabling employers to train an Eastern European

(who would not require a visa) to do the work and replace theoriginal domestic worker when she returned.

The government had not reckoned with the symbolicimportance of this particular group of migrants. Whileaccounting for a very small proportion of entrants and evensmaller proportion of long stayers, domestic worker visaholders continue to have a lot of support from those whofeared a return to ‘modern day slavery.’ Domestic workervisa holders and other campaigners protested that thecontinuing high rates of abuse and exploitation by firstemployers meant that tying them to abusive employers inthis way would mean, in effect, that the UK government wasfacilitating ‘trafficking.’ The Labour government's initialresponse was interesting: they claimed that traffickingprevention measures would be put in place that wouldensure all employers and workers would be interviewedbefore entry and at the border, and if the employer wassuspected of maltreating the worker they would be refusedentry. Trafficking and slavery would not be allowed onBritish soil. It is not hard to imagine the consequences for adomestic worker if she and her employing family arereturned to the state from which they came (often not thestate of citizenship of the domestic worker, but the state ofcitizenship of the employing family). While this proposalstops abuse from happening in the UK territory, it is at thecost for the worker of far greater risk and powerlessness onthe territory of another state. State intervention would notbe neutral and prevent harm, it would actively increase therisk of harm to a vulnerable person, but it would prevent theharm from occurring on British territory. The suggestedresponse to the problem reveals the importance of keepingBritain a national space of liberal freedoms, while simulta-neously disavowing interest in human rights outside ofBritish territory, even when the British state has had anactive role in creating vulnerability. It very explicitly tiesrights to the presence of the territory, even if this presence isnot legally endorsed.

The new Coalition government reopened the debate, intheir pursuit of reduced net migration. They reinvigoratedsome of Labour's original proposals, and announced that fromApril 2012 domestic workers accompanying their employerswould only be eligible for a six-month non-renewable visa.They would not be able to change employer and would have toleave the UKwith their employer. Theywould be given visas asworkers, but, like Romanian and Bulgarian accession cardholders, this did not mean that they would be paid minimumwage, as the UK Border Agency claimed ‘it is not for us to saywho is entitled to the National Minimum Wage’.7 It wasacknowledged by Government Ministers that employment inprivate households could be abusive, but the association of thiswith the foreignness of households was made explicit. HomeSecretary Teresa May, for example, stated: ‘We recognise thatthe ODW routes can at times result in the import of abusiveemployer/employee relationships to the UK’.8 Whereas previ-ously vulnerability to abuse had been used to demand laborrights, these two have now been separated, and protectionrather than rights were invoked as the appropriate response:‘We do not necessarily believe that a right to change employerwhatever the reason is the only way to provide protection’(Home Office Response to OSCE Country report on the UK,Justine Currell, 2012). The principle protection for migrant

Page 14: Womens Studies International Forum 46 (2014)

12 B. Anderson / Women's Studies International Forum 46 (2014) 5–12

domestic workers was now to be refusal of entry. ‘The biggestprotection for these workers will be delivered by limitingaccess to the UK through these routes’ (Home Office, 2012).Refugees may, theoretically at least, be protected by flight, butmigrant women, it seems, are protected by not being able tomove in the first place.

Conclusion

A focus on nation building and the ideological andproductive function of borders and immigration controlspoints to the disjuncture between the formal community ofthe state, comprised of legal citizenry, and the community ofvalue. It indicates that we need to look beyond formalcitizenship in order to understand the discrimination facedby migrant domestic workers. Attention to social constructionand embodiment – which is where domestic work leads us –moves us beyond wage differentials and the lump of laborfallacy to the nexus of gender, race and nationalism, thematerialities of bodies and the relations of affect, all of whichshape the intricacies of regimes of care, welfare, employmentand the labor markets for paid care.

Endnotes

1 I am using ‘state’ to refer to what Abrams (1988) called ‘the state idea’rather than ‘the state system’ i.e. ‘the mask which prevents us seeingpolitical practice as it really is,’ rather than the heterogeneous ‘state’institutions and actors.

2 There is of course nevertheless considerable heterogeneity masked bythis, depending on labor market segment, source state, employmentrelations and migration status, etc.

3 Particularly in the light of the application of a ‘points based system’ toincapacity benefit/jobseeker's allowance assessment — previously it hasprincipally been associated only with immigration

4 Across Europe, these regulations vary widely: in Norway, Netherlands,Sweden and Denmark for instance, the movement of young people forcultural exchange more obviously includes people that might be betterdesignated as ‘domestic workers’ from Thailand and the Philippines.

5 http://www.runnymedetrust.org/uploads/file/LivingTransnationally-2008.pdf.

6 The terminology is interesting here as it captures acute contradictions.Can one really be ‘employed’ to be in a fictive kin relation, i.e. not consideredas ‘working’?

7 P.c. Jenny Moss, Kalayaan.8 http://www.parliament.uk/documents/commons-vote-office/February_

2012/29-02-12/6.Home-Immigration.pdf.

References

Abrams, P. (1988). Notes on the difficulty of studying the state. Journal ofHistorical Sociology, 1(1), 58–89.

Anderson, B. (1993). Britain’s Secret Slaves: Migrant domestic workers in theUK. London: Kalayaan and Anti-Slavery International.

Anderson, B. (2000). Doing the dirty work? The global politics of domesticlabour. London: Zed Books.

Anderson, B. (2007). A very private business: Employment in privatehouseholds. European Journal of Women's Studies, 14(3).

Anderson, B. (2010). Migration, Immigration Controls and the Fashioning ofPrecarious Labour. Work, Employment and Society, 24, 300–317.

Bettio, F., Simonazzi, A., & Villa, P. (2006). Change in care regimes and femalemigration: The care drain in the Mediterranean. Journal of EuropeanSocial Policy, 16(3), 271–285.

Cangiano, A., Shutes, I., Spencer, S., & Leeson, G. (2009).Migrant care workers inageing societies: Research findings in the United Kingdom. Oxford: COMPAS.

Cockburn, C., & Ormrod, S. (1993). Gender and technology in the making.London: Sage.

Cooper, S. (2004). From family member to employee: Aspects of continuityand discontinuity in English domestic service 1600–2000. In A.Fauve-Chamoux (Ed.), Domestic Service and the Formation of Europeanindentity: Understanding the Globalization of Domestic Work, 16th–21stCenturies. Bern, Switzerland: Peter Lang.

Cox, R. (2006). The servant problem: Domestic employment in a globaleconomy. London: I.B.Tauris.

Cox, R. (2007). The au pair body: Appearance, sex object, sister or student?European Journal of Women's Studies, 14(3), 281–296.

Denning, M. (2010). Wageless life. New Left Review, 66, 79–97.Fraser, N. (2011). The wages of care: Reproductive labour as fictitious

commodity. Lecture, University of Cambridge, 9th March 2011. Citedin Williams, F. (2014, in press) Making Connections across theTransnational Political Economy of Care. In B. Anderson, & I. Shutes(Eds.), Migration and care labour: Theory, policy and politics. London:Palgrave.

Glenn, E. (1992). From servitude to service work: The historical continu-ities of women's paid and unpaid reproductive labour’. Signs, 18(1),1–44.

Gutiérrez-Rodríguez, E. (2012). 2012 affective value: On coloniality, feminizationand migration’, transversal: multilingual webjournal, Dossier: UnsettlingKnowledges. http://eipcp.net/transversal/0112/gutierrez-rodriguez/en

Home Office (2012). Immigration (employment-related settlement, overseasdomestic workers, Tier 5 of the points based system and visitors). Statementof intent: 29 Feb. 2012.

Honig, B. (2003). Democracy and the foreigner. Princeton: Princeton UniversityPress.

Luibhéid, E. (2002). Entry denied: Controlling sexuality at the border.Minnesota:University of Minnesota Press.

Miller, D. (1995). On nationality. Oxford: Clarendon Press.Miles, R. (1987). Capitalism and unfree labour: Anomaly or necessity? London:

Tavistock Publications.Newcombe, E. (2004). Temporary migration to the UK as an au pair: Cultural

exchange or reproductive labour? Sussex Migration Working Paper No.21. University of Sussex.

Orloff, A. (2006). From maternalism to ‘employment for all’: State policies topromote women's employment across the affluent democracies. In J. D.Levy (Ed.), The state after statism: New state activities in the age of liberalization.Cambridge, MA: Harvard University Press.

Ngai, M. (2004). Impossible subjects: Illegal aliens and the making of modernAmerica. New Jersey: Princeton University Press.

Nyers, P. (2010). Abject cosmopolitanism: The politics of protection in theanti-deportation movement. In N. de Genova, & N. Peutz (Eds.), TheDeportation Regime. Durham: Duke University Press.

Parreñas, R. (2001). Servants of globalisation: Women, migration and domesticwork. Stanford, CA: Stanford University Press.

Van Walsum, S. (2008). The family and the nation: Dutch family migrationpolitics in the context of changing family norms. Newcastle Upon Tyne:Cambridge Scholars Publishing.

Williams, F. (2010). Migration and care: Themes, concepts and challenges.Social Policy & Society, 9(3), 385–396.

Williams, F., & Gavanas, A. (2008). The intersection of childcare regimesand migration regimes: A three-country study. In H. Lutz (Ed.), Migrationand domestic work: A European perspective on a global theme. Aldershot:Ashgate.

Wimmer, A., & Glick Schiller, N. (2002). Methodological nationalism andbeyond: Nation-state building, migration and the social sciences. GlobalNetworks, 2(4), 301–334.

Yeates, N. (2009). Globalizing care economies and migrant workers. Basingstoke:Macmillan.

Page 15: Womens Studies International Forum 46 (2014)

Women's Studies International Forum 46 (2014) 13–23

Contents lists available at ScienceDirect

Women's Studies International Forum

j ourna l homepage: www.e lsev ie r .com/ locate /ws i f

International migration, public policies and domestic work

Latin American migrant women in the Spanish domestic work sector

Sandra Gil Araujo a,⁎, Tania González-Fernández b

a CONICET (National Council of Scientific and Technical Research), Gino Germani Research Institute, Faculty of Social Science, University of Buenos Aires, Argentinab Department of Social Anthropology, Stockholm University, Marie Curie ITN CoHaB (Diasporic Constructions of Home and Belonging), Sweden

a r t i c l e i n f o

⁎ Corresponding author at: Instituto de InvestigacionUriburu 959, 6º (1144), Ciudad Autónoma de Buenos

0277-5395/$ – see front matter © 2014 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.wsif.2014.01.007

s y n o p s i s

Available online 11 February 2014

Based on the Spanish case, in this article we explore the connection between migrationpolicies, family policies, gender regimes and the insertion of Latin American migrant womeninto the domestic work sector. Over the first decade of the twenty-first century, Latin Americabecame the main region of origin of migrants who had settled in Spain, being women the firstlink in these migration chains. The main factors that have affected the configuration of thisfeminization are linked to migration policies and patterns of migration, the features of thewelfare state, the characteristics of the labor market and the way in which gender organizesand stratifies migration and domestic work. The achievement of national middle classwomen's rights to conciliate their professional and family life through outsourcing domesticwork to “non-national” women also brings with it a deep inequality in terms of citizenship.

© 2014 Elsevier Ltd. All rights reserved.

Introduction

Over the first decade of the 21st century, Spain becamethe primary destination for Latin American immigration, par-ticularly from Ecuador, Colombia, Argentina, Bolivia, Peru andthe Dominican Republic. This growingmigration can bemainlyexplained through the financial crisis, impoverishment anddegradation of working conditions in their countries of origin,1

but also (and we should emphasize this) through the far-reaching social transformation of Spanish society since themid-1990s. The analysis of Latin American migration to Spain,thus, needs to be attentive to the conditions in which thisphenomenon is embedded: accelerated economic growth, thespecificities of the regional labor market, state policies, genderrelations, the welfare regime, the historical bonds betweenSpain and Latin American countries, the consolidation of migra-tion networks and increasing importance of family migration(Gil Araujo, 2008).

es Gino Germani, Pte.Aires, Argentina.

ll rights reserved.

Unlike traditional Latin American migration to the USA,headed foremost bymen, there is a substantial female presencein Latin American migration to Spain, especially in the earlieststages of the project and by the role of women as the first linkwithin themigration chains (Pedone, 2006). Aswewill show inthe following sections, the factors affecting the configurationof this feminization are linked to the history, policies and pat-terns of migration, the features of the welfare state, the socialand sexual division of reproductive labor, the characteristicsof the labor market and the way in which gender permeates,organizes and stratifies migration and work. As many scholarshave shown, migration policies, family policies, labor markets,racial and gender relations shape the transfer of productiveand reproductive labor from the South to the North in thecontext of international migration (Anderson, 2000; GutiérrezRodríguez, 2010; Hondagneu-Sotelo, 2001; Parreñas, 2001).

In this article we focus on the connection betweenSpanish migration policies, family policies, gender relations,the “de-nationalization” or “foreignization” of domestic work,and the feminization process of Latin American migrationtowards Spain, as well as the implications in terms of in-equalities based on gender, class, legal status and nationality.

Page 16: Womens Studies International Forum 46 (2014)

14 S. Gil Araujo, T. González-Fernández / Women's Studies International Forum 46 (2014) 13–23

In addition to the limitations on family life that the gruelingpace of domestic work implies (especially in the case oflive-in workers), migration legislation imposes requirementsfor housing, earnings and employment, which make officialfamily reunification very difficult. These limitations on theright to family life have deepened family transnationalism,encouraging various forms of long-distance motherhood,rarely freely chosen by the women themselves. The transferof inequality between women involved in domestic workbeing outsourced from ‘national’ to ‘non-national’ womenalso brings with it a deep exclusion in terms of the right tofamily life, care and affection.

We will draw on the data collected within differentresearch projects we have participated in over the last sixyears.2 Between 2006 and 2007, as part of the Spanish researchteam for the project Civic Stratification, Gender and FamilyMigration Policies in Europe, we took a gender perspective foranalyzing the diverse formal and informal regulations thataffect the “non-national” population established in Spain inrelation to access to the labor market, social rights and familylife. We also carried out sixteen interviews in Barcelona andMadrid, conducted with a lawyer, a social mediator, tworepresentatives of migrant associations along with nine femaleand three male migrants with different nationalities andmigration status. All the women interviewed worked or hadworked in the Spanish domestic labor sector. Between 2009and 2011, as part of the project Fundamental rights situation ofirregular immigrants in the EU carried out in different Europeancountries, we elaborated a Spanish case study about irregularmigrants employed in domestic work. Interviews were con-ducted in Madrid with both migrants in irregular situationemployed in domestic work and other stakeholders such asmembers of NGOs, migrant associations, women's organiza-tions, trade unions and experts in the field. Between 2010and 2012, we were part of the research team for the projectMigration Policies, Family Transnationalism and Civic Stratification:Latin American Migration to Spain. From a transnational andgender-based perspective, the aim of this project was to explorethe connections between Spanish migratory policies, the strat-ification of migrants' rights and the transnationalization offamily life among Latin American migrants who had settled inSpain. Within this framework we looked at migration, social,labor and family legislation and (formal and informal) reg-ulations that particularly affect the Latin American migrantpopulation. Transnational fieldwork was carried out betweenSpain (Barcelona andMadrid), Colombia and Ecuador, includingthirty in-depth interviews with members of migrant familiesin both the country of origin and destination. All the mi-grant women interviewed had worked in Spain as domesticemployees.

We have organized the text into four sections. Firstly,we will present a brief review of the research on femalemigration in Spain, with the aim of highlighting that from thevery beginning, the feminization process of migration wasrelated to the insertion of migrant women into the domesticwork sector. Secondly, we will summarize the particularfeatures of the Spanish migratory regime and present data onthe profiles of Latin American migrants who have settled inSpain, including information about the impact of the currenteconomic crisis on their unemployment rates and settlementpatterns. Thirdly, we will explore some of the features that

have historically characterized the Spanish welfare regime(such as the family as the main provider in times of need,the decentralization of social policies or the notion that carewas closely linked to and relied on families, households andwomen), which add to the implementation of recent familypolicies that urge private households to have their care needsmet by the market, thus generating a sustained and in-creasing demand for domestic workers. The data we presentin this section shows that to a large extent, Latin Americanmigrant women were the ones to meet this demand and thatthe increase in migration was accompanied by growingnumbers of migrant women in domestic work. As describedin the testimonies of our informants, the demand of womenfor domestic work is crucial in explaining the feminizationof the migration flows from some Latin American countries(theDominican Republic, Bolivia, and Paraguay, among others)and the role of many of these women as the first link in thesemigratory chains. In order to understand this process, we givean account of the immigration regulations (visa exemptionsand regularization processes) that facilitated both the entryinto Spain and the later regularization of Latin American mi-grant women through their insertion into the domestic worksector. Lastly, in section four, we will focus on the effectsof public policies (migration, labor) and the domestic workregime on the family life of Latin American domestic workers.This paper sheds light on how the difficulties in fulfilling therequirements for family reunification deepen and broaden theexercising of transnational motherhood, which the womenrarely freely choose. Within this framework, national middleclass women's rights to conciliate their professional and familylife have been achieved through denying “non-national” femaledomestic workers a family life of their own.

Female migration and domestic work in Spain: a briefstate of the art

In the Spanish context the first debates on the feminizationofmigrationwere related to studies on domesticwork from theearly 1990s. For instance, María E. Sánchez Martin's (1992)study Nuestras hermanas del Sur: La inmigraciónmarroquí y elservicio doméstico en Madrid that dealt with the situation ofMoroccan women in Spain, and Gina Gallardo Rivas's (1995)exploration on Dominican women's migration to Spain andtheir settlement patterns published in her book Buscandola vida: dominicanas en el servicio doméstico en Madrid.3

Both publications are crucial in studying the feminization ofmigration in Spain, because at that time there was alreadya noticeable presence of women among the nationals fromboth countries. Theseworks addressed the situation ofmigrantwomen in Spain and drew attention to the fact that domesticwork was one of themain entry points into the labor market, asituation that has hardly changed since.

In the late 1990s and the early twenty-first century, severaldissertations that focused on female migration and domesticwork were published,4 in addition to Colectivo IOE's (1991)groundbreaking study for the InternationalWork Organization.Although the data is outmoded, it remains a crucial frame ofreference. Later works, such as the Colectivo IOÉ (2001) on thesituation of migrant women and their entry into the labormarket, and Parella Rubio (2000, 2003) analyses of the

Page 17: Womens Studies International Forum 46 (2014)

15S. Gil Araujo, T. González-Fernández / Women's Studies International Forum 46 (2014) 13–23

domestic sector in Catalonia in relation to proximity services,have shed further light on this phenomenon.

This second wave of studies on gender and migrationhighlights the direct link between the increasing demand fordomestic work in Spain and outsourcing care and domestictasks. In order to cover those precarious, invisible and sociallyunvalued job positions, a high demand for a female labor forceis generated. The analysis of migration in terms of gender hasled scholars to the conclusion that female migration can nolonger be attributed to a traditional gender model in whichwomen are merely passive followers of their mobile husbands.These works demonstrate the very opposite, namely thatwomenmigrate alone and that their patterns differ from thoseof men and, in some cases, that they headed family migrationprojects (Oso, 1998). Consequently, in somemigrant groups inSpain, such as those from the Dominican Republic, Brazil, Peru,Colombia, Bolivia and Paraguay, there is a high proportionof women and, as previous research has shown, in the case ofEcuadorean women, they are the first link in the migratorychain (Gil Araujo & Pedone, 2013).

In this vein, changes in migration patterns in both Spainand other southern European countries can be attributed tothe following structural factors: 1) an aging population as aresult of a combination of increased life expectancy and afalling birth-rate; 2) growing participation of women in thelabor market and an increase in their skills; 3) a new man-agement of time and organization within families, which areincreasingly smaller and more geographically dispersed (thepossibility of creating support networks among the membersis diminished); 4) a tendency towards reducing state supportfor family matters and an inadequate supply of services(welfare regime based on a familistic model5), 5) a migrationpolicy designed in order to cover those needs; and 6) a lack ofreproductive work shared between men and women.

The contribution of international literature on these issues,and the intersection of the welfare, gender and migrationregimes are just as crucial for understanding the responsibilityof the Spanish state, since states have become key playersin these new global developments (Lutz, 2008;Williams, 2005;Williams & Gavanas, 2008). Therefore, this complex issuemustbe approached by considering women's contribution to socialreproductive work within households; the relation betweenthe state, the market, the family and the community in thesupply of services; and pigeonholing migrants in certainextremely feminized and racialized/ethnicized labor niches(Anderson, 2000; Gutiérrez Rodríguez, 2010; Hondagneu-Sotelo, 2001; Parreñas, 2001).

The internationalization of reproductive labor: a transfer ofinequalities among women

This situation takes place against a backdrop where cap-italism and patriarchy are strongly interlinked. This dyadleads to a sexual and ethnic division of labor on an inter-national scale and to a transfer of reproductive labor amongwomen. Regarding these specific issues, a key contribution isVega, García, and Monteros's (2004) report on Madrid, whichwas conducted within the framework of an European projecton domestic work, migrant women's rights and inequalities(Caixeta, Gutiérrez Rodríguez, Tate, & Vega Solis, 2004).It sheds light on how, in the context of a migration society,

women experience the provision, sharing and transfer ofthose tasks, which range from highly valued to the so-called“dirty” tasks. The national dimension is added to the sexualand ethnic division of work and becomes central in this trans-fer between women (from “nationals” to “non-nationals”). Itinvolves major changes, such as an alteration of domesticprivacy, and a perception of “complementariness” or “substi-tution” of one woman by another (Caixeta et al., 2004).Therefore the relationships or ties that may be establishedbetween women conceal the asymmetry of new expressionsof patriarchal relations experienced on a global scale (Vegaet al., 2004).

The employment of migrant women in this sector enablesus to identify a transfer of class and ethnic inequalities amongwomen, as Parella Rubio (2000, 2003) points out. Parellaconsiders that the myth of new and more equal gender rela-tions between heterosexual couples is masked, while patriar-chy remains embedded within domestic structures and paidemployment. Female migrant workers thus act as a peripheral“reserve army” that reduces the cost of services linked to socialreproduction for both capital and the state, by increasing socialinequalities between women.

The result of this transfer is the ethnicization of domesticand care work (Catarino & Oso, 2000; Parella Rubio, 2003).Although migrant women and certain profiles of nationalwomen share the same sector of activity, they are not presentin great numbers in the same modalities. Migrant women arenormally hired for the job positions least desired and valued,for instance they are mainly concentrated in the live-in do-mestic work sector. Parella Rubio (2003) points out thatthis process of ethnicization not only applies to the personalcharacteristics of migrant women and their strategies, but alsoto the stereotyped configurations of theworkforce demands. Inthis way, criteria of differentiation, discrimination and exclu-sion that rely on ethnic and cultural factors are produced. Onthe other hand,migrant status createsmarginal groupswith noaccess to the formal labormarket, ormakes their legal situationdirectly dependent on this type of work.

Lastly, more recent works provide a new approach thatlinks migration flows and care in a global context. Theirtheoretical and political aim is to bring the notion of care tothe fore and to acknowledge its complexity as an activityembedded in our everyday lives. By referring to conceptssuch as the “global care chain” (Hochschild, 2000), theseworks draw attention to the interaction between two house-holds whose developments are intertwined, simultaneousand interdependent. They seek to contribute in developing abroader reflection on the social organization of care fromwhich the premise of the present social model sustaining thereproduction of life may be questioned as a whole (DíazGorfinkiel & Orozco, 2009). In light of this, Díaz Gorfinkieland P. Orozco make use of the “right to care” in a twofoldmeaning: the right to receive and to provide care. However,according to the authors, this right should also recognize theoption to freely decide whether to provide care or not. Morerecent studies focusing on the intersection between the mi-gration regime, care regimes and gender relations include theVega Solís's (2009) work, Díaz Gorfinkiel's (2008a) doctoralthesis on childcare, Martínez Buján's (2008) doctoral thesison care for the elderly, and the special issue of the JournalAlternativas edited by Agrela Romero, Martín Palomo, and

Page 18: Womens Studies International Forum 46 (2014)

16 S. Gil Araujo, T. González-Fernández / Women's Studies International Forum 46 (2014) 13–23

Langa Rosado (2010), which is a compilation of the paperspresented at the First Annual Congress of the Spanish SocialPolicy Network in Oviedo in November 2009.

Nevertheless, while caring for dependents (children,elderly, sick people) has been the subject of public debatesfor the last decade, domestic work has been kept a privatematter, without assuming “its structural character as neces-sary societal labor” (Gutiérrez Rodríguez, 2010: 75). Follow-ing Anderson's (2000) argument, Gutiérrez Rodríguez (2010)writes that the demand for care workers is only one of thefactors that configure the labor market for migrant domesticworkers. Domestic work involves cleaning and caring. Thestate-led focus on care not only camouflages the fact that careworkers perform domestic work as well, but it also obscuresthe increasing transfers of domestic work to migrant women,many of whom are in an irregular situation and are hiredby private households, meaning they also have very limitedeconomic, social and political rights.

However, most of the researchers haven't paid adequateattention to the way in which policies of the receivingstates shape the family life of migrant workers. TakingLatin American migration to Spain as a case study, the aimof our analysis is to demonstrate how the combinationof migration policies (controls, residence, family reunifica-tion) and the precarious labor conditions of domestic work(informality, temporary, low salaries, long working days)creates obstacles for domestic workers living in migrationto reunite their families in Spain. In this way, Spain'spublic policies foster the extension and prolongation oftransnational family bonds in general, and more specifi-cally, transnational maternity. It seems that the geograph-ical dispersal of migrant family members is becoming adistinctive feature of North–South migration in the twenty-first century.

The migratory regime and Latin American migrationin Spain

In the 1980s, various processes coincided, which have hadan impact on Spain's current political and economic profile,including its democratic consolidation, integration into thetransnational economic sphere via a political and economicopening, and its alliance with advanced capitalist nationsafter joining the EU in 1986. These dynamics produced anexplosion of high-paid professional and technical jobs, butalso an increase in low-paid jobs and employment in theinformal sector. Within this framework, Spain came to bereferred to as a country of immigration, in a context that wasclearly different from postwar migration in Europe (Gil Araujo,2005, 2010a).

The development of social protection coincided with alimited deployment of the welfare state, the existence of highlevels of unemployment and the increasing incorporationof migrant labor, principally in the area of informal work. Theinformal economy continued to play an important role indifferent sectors, such as textiles, services, construction,and agriculture. The weight of the informal economy shouldnot be interpreted as a sign that Spain was less developedcompared to other EU countries, but as a specific form ofinsertion into the international market. The informal econo-my results in a minority of qualified workers with high

salaries and a majority of workers in precarious conditions. Itis this hierarchical structure that immigrant laborers enterinto in different ways.

The migrant population is heterogeneous, with notabledifferences between nationalities and in terms of sex, age,education levels, migration projects, and degrees of laborinsertion. A great number of migrants came from formerSpanish colonies, especially Morocco and Latin Americancountries. The 2001 Census estimates the foreign populationin Spain at 1.5 million, which was 3.7% of the total populationof 40,847,371. At that time, the largest migrant communitieswere from Morocco, Ecuador, the United Kingdom, Germanyand France (Gil Araujo, 2005). In January 2013, municipalityregistrations showed that migrants (documented and un-documented) represent 11.7% of the total population. Themain countries of origin are Rumania, Morocco, the UnitedKingdom, Ecuador, Colombia, Italy, Germany, China, Boliviaand Bulgaria (INE, 2013).

Registered migrant population and percentage of women by nationality(regions), (January 2013).

Region of nationality

Numbers % of total migrantpopulation

% ofwomen

UE-27

2,352,978 42.6 48.2 Europe 247,441 4.5 57 Africa 1,096,392 19.9 38.8 Asia 376,285 6.8 41.2 North America 57,340 1 55.8 Central America and Caribbean 222,767 4.0 61.5 South America 1,163,705 21.1 55.9 Others 3225 0.1 –

Total

5,520,133 100 48.5

Source: Padrón de Habitantes. INE, 2013.

The insertion of migrants into the labor market is limitedto only a few sectors—commerce and the hotel industry;domestic work; agriculture; construction—and displays arapidly developing tendency towards segmentation andethnicization (Pedreño, 2005). In particular, non-EUmigrantshold low-qualified jobs in the secondary labor markets, andthey work in positions that are often temporary or precarious,with very poor working conditions and frequently poorly paid(Arango, 2004; Cachón, 2002; Pedreño, 2005; Calavita, 2005).

The connection between access to rights and work permitspermeates all the norms that regulate migrant residence inSpain, whichmakes evident that non-EUmigrants are basicallyperceived as part of the labor market. Legislation establishesthat foreigners over sixteen years of age must obtain a workpermit aswell as a residence permit in order towork in Spain. Ajob offer must be submitted along with the application for afirst work permit. This initial permit is valid for one year andmay be limited to a particular sector of activity and geograph-ical area. In order to renew the work permit, the applicantsmust have contributed to the social security system and havepaid their taxes. Since 14 January 2002 it is no longer possiblefor persons already on Spanish territory to apply for a workpermit. These are only granted to those who submitan application in their country of origin, and quotas exist foragriculture, domestic work, construction and service sectors.This has meant that the hundreds of thousands of non-EU

Page 19: Womens Studies International Forum 46 (2014)

17S. Gil Araujo, T. González-Fernández / Women's Studies International Forum 46 (2014) 13–23

migrants already settled in Spain are constricted to a status ofirregularity; although they may have a job offer, they cannotobtain the necessary permit.

As Arango (2004) points out, the high proportion ofundocumented migrants is the most distinctive structuraland constant characteristic of immigration patterns in Spain.Until recently, Spanish immigration policy focused on theentry of undocumented migrants, who would potentiallyreceive “reparation” through being offered the opportunityof regularization (Cachón, 2002). Domestic work has alwaysbeen one of the main channels for obtaining regularizationduring different periods of Spanish government legislation,in 1986, 1991, 1996, 2000, and 2001. The last regularizationwave took place in 2005, during which 31.6% of the ap-plications were for the domestic sector, 20.7% for construc-tion, 14.2% for agriculture and 10.4% for the hotel industry.In the domestic sector, 83.4% applicants were women. Inthe construction and agriculture sectors, the situation isthe other way around, as respectively 94.9% and 83.1% of theapplicants were male (SEIE, 2006). Nonetheless, domesticwork was defined as an exceptional channel of regularization(vía excepcional de regularización).

In the context of the current international crisis, whichis also affecting Spain, legislation imposing new limitationson migrants' rights has been passed. In 2009, the govern-ment leaders of the Spanish Socialist Workers' Party (PSOE)passed a new reform to the Immigration Law. The mainmodification limits the reunification of the parents and in-laws to non-EU migrants who hold long-term residencepermit (Pedone, Agrela, & Gil Araujo, 2012). In 2013, thegovernment leaders of the Popular Party (PP) repealedundocumented migrants' rights to public health care. Giventhe precarious legal and labor status of migrant domesticworkers, both reforms pose restrictions on their rights tohealth and to family life.

Figures and new trends

The year 2000 was a turning point for several reasons.With consolidation of migration networks, there has beenan increase in and diversification of the migration flows, aswell as a change in migrant demographics through familyreunification and the children born while living in migration.At the same time, immigration has become a matter of publicdebate. The beginning of the twenty-first century ushered ina phase of migration that was characterized by the domi-nance of migrants from Latin American countries (Actis,2009; Gil Araujo, 2008; Martínez Buján, 2003). Municipalityregistrations in Spain show a 663% increase in Latin Americanmigration between 2000 and 2005. During this period, thepercentage of Peruvians registered rose by 140%, that ofColombians tripled, Ecuadorians almost quadrupled, Argen-tines quintupled and Bolivians increased fifteen fold. If weadd to this the number of people from Latin America withSpanish nationality, we can conclude that the Latin Americanpopulation residing in Spain has multiplied six times inthe period between 1998 and 2005 (Vicente Torrado, 2006).It is important to stress that there is a greater number ofmigrants from Latin America than shown in the statistics,because (a) it is relatively easy for Latin American migrantsto obtain Spanish nationality (two years of regular residence

vs. ten years for other nationalities); (b) there are bilateralagreements between Spain and these countries on dualcitizenship; and (c) there are also Latin American migrantswho are descendants of Spanish emigrants (mainly Argenti-na, Cuba, Venezuela and Uruguay).

The growing visibility of this population has been, in part, aresult of state policies, such as the two regularization processesin 2000 and 2001, and a greater percentage of approvals ofapplications submitted by Latin American migrants (MartínezBuján, 2003: 17).Moreover, bilateral agreementswith Ecuador,Colombia and theDominican Republic prioritize hiringworkersfrom these countries, based on a quota system.Other importantfactors have been less strict (in comparison to the USA, forexample) controls of incoming migrants from most LatinAmerican countries until the early 2000s. In the politicaldebate, favoring Latin Americans in migration policy has beenjustified by claiming that an ‘ethnic affinity’ (Cook-Martín &Viladrich, 2009) exists between Latin American and Spanishpeople, based on an (imagined) commonhistorical and cultural(colonial) background (Gil Araujo, 2010b).

Since the mid-1980s, Latin American migrants in Spainwere predominately women, mainly from the DominicanRepublic, and continued to grow until the period between1992 and 1996. At that time, 62% of Latin American migrantswith resident permits were women. From 2002 to 2006,the preponderance of women fell to a more moderate 53%,according to the municipality registry (Actis, 2009). Later, theprocesses of family reunification and the demand for LatinAmerican workers in construction and agriculture generateda greater balance between the sexes, with major differencesbeing the countries of origin. The highest degree of feminiza-tion among groups are, on the one hand, those initiating theirmigration process to Spain from Honduras, Paraguay, Braziland Bolivia, along with some already stabilized groups such astheDominicanRepublic, Colombia, Venezuela and Cuba. On theother hand, Peru, Ecuador, Argentina and Uruguay show moreof a balance between the sexes.

Latin American migrants with residence permit by country of nationality andsex.

Country

No. of migrants withresidence permits

% women

Argentina

94,461 49.9 Bolivia 146,723 60.7 Brazil 62,429 68 Colombia 276,342 56.5 Cuba 53,393 55.1 Dominican Republic 90,947 58.5 Ecuador 406,330 51.3 Paraguay 39,984 71 Peru 140,157 52.6 Uruguay 33,351 49.5 Venezuela 41,614 59.2

Source: Ministerio de Trabajo e Inmigración. Date: 30 September 2011.

Currently, of all the migrants (with and without papers)in the municipal register in January 2013, 1,386,472 areLatin American migrants, which make for 25.1% of themigrant population, 56.8% of whom are women (INE, 2013).The increase in the proportion of women in Latin Americanmigration flows in recent years could be related to (a)

Page 20: Womens Studies International Forum 46 (2014)

18 S. Gil Araujo, T. González-Fernández / Women's Studies International Forum 46 (2014) 13–23

the consistency and/or increase in female migrationcoming from Cuba, the Dominican Republic and Honduras;and (b) the return migration of groups with a greaterbalance between the sexes, such as Argentineans andEcuadorians. The data indicates a greater impact of unem-ployment onmalemigrants compared withmigrant women inthe first cycle of the crisis (Colectivo IOÉ, 2012) which couldgenerate a significant increase in the return migration amongmen.

In spite of the diversity that characterizes Latin Americanimmigration in Spain in terms of country of origin, sex andtime of arrival, Walter Actis (2009) identifies some dominantcharacteristics: 1) increase in the flow during the first yearsof the 21st century, 2) initial importance of female flows,3) considerable importance of family chains at the begin-ning of emigration and for remittances in the present, and4) process of downward labor mobility in emigration, par-tially compensated by higher wages and access to publicservices (health, education, etc.).

This situation developed throughout an expansive eco-nomic cycle that ended in the middle of the year 2008. Theunemployment rate increased from 8.3% in 2007 to 26%at the end of 2012, with a rate of 36.5% in the overallmigrant population. In terms of origin, Africans have beenthe most affected by unemployment, followed by Euro-peans from non-EU countries and Latin Americans. Amongthe Latin American population men have a higher unem-ployment rate (31.1% in 2011) than women (26.3% in 2011)(Colectivo IOÉ, 2012: 193, based on Encuesta de PoblacionActiva).

As Actis (2009) stresses, the crisis raises various questionsregarding the future of migrants residing in Spain, as wellas the continuation of the migration flows. As a reflection ofthis difficult situation, 2011 data shows a negative balanceof South American migration to Spain and, to a lesser extent,African migration, while other contingents continue to in-crease (Colectivo IOÉ, 2012: 5). Although we cannot speak ofa widespread exodus, an increase in departures has beennoted—from 109,000 to 305,000 between 2007 and 2011—as well as a decrease in the number of registered migrants,mainly from countries in South America. Latin Americanswho have continued coming to Spain are chiefly migrantsfrom the Caribbean and Central America (Dominican Repub-lic, Mexico, Honduras and Cuba) and, to a lesser degree,from Venezuela, Peru and Paraguay. In Spain, there has beena decrease in the presence of migrants from the SouthAmerican countries of Bolivia, Ecuador, Colombia, Argentinaand Brazil (Colectivo IOÉ, 2012). Many migrants from thesecountries have returned to their home country, but others,mostly (but not exclusively) those who have EU citizen-ship, have re-migrated to other countries within the EU,like the United Kingdom and Belgium (Pedone, Echeverri,& Gil Araujo, in press). Our field work with Colombianand Ecuadorian migrants indicates that both new emi-grations and returns have been selective. If a decade agothe member of the household who was in the bestposition for migration was chosen, today families thinkabout the best strategy for combining the return of somemembers and the continuation of others who can keepgenerating economic resources at their destination (Gil Araujo& Pedone, 2013).

Welfare state, domestic work and migration: LatinAmerican migrant women in Spain

Taking into account the relationship between the state,family, market and non-governmental resources (Kofman &Raghuram, 2009) the Spanish welfare state fits the patternof Mediterranean welfare regimes. One of its distinctivefeatures is that the (nuclear hetero-patriarchal) family is themain provider in times of need. This trend towards the familyhas been reinforced by the decentralization of policies andthe state's withdrawal from the provision of social services—returning tasks of social reproduction to the family, andwithin it to women. Another important factor contributing tothe configuration of the welfare regime is the genealogy andculture of care (Vega Solís, 2009). In the case of Spain, thenotion of care has been historically closely linked to family,home and women. According to surveys conducted by theSociological Center Research in 2002 and 2004, 72% of thoseinterviewed would prefer to receive care in their home, while16.8% would prefer a nursing home. Regarding who theywould like to have to provide them care, 78% prefer a relative(45% spouse, 33% other relatives). These figures show homeand family as the fundamental places for care (Díaz Gorfinkiel& Orozco, 2009). In this regard, Díaz Gorfinkiel's (2008a)doctoral thesis on childcare shows thatmothers prefer to leavetheir sons and daughters up to two years old in a familiarenvironment, such as their own house, with someone provid-ing them personalized care. This person should preferably notbe a relative to avoid interference in the rearing of children.Care collectives are valued at a later stage, when theyencounter educational and socialization needs.

The link between the increase in Spanish women's par-ticipation in the labor market, the aging population,6 andthe lack of public institutions taking on responsibility forthe care of dependents has posed a major challenge to thewelfare regime in Spain (Martínez Buján, 2008; Williams,2005; Williams & Gavanas, 2008). In this context, manyfamilies have decided to hire domestic workers, oftenmigrants and mostly women, although some of them are inirregular situation (González Fernández, 2013). While thecommodification of domestic work has meant a reorganiza-tion of reproductive labor in the family sphere, so far it hasnot brought a better balance of responsibilities betweensexes and generations. Domestic work and family careremains a “woman's concern” (Martín Palomo, 2008).7

EU states have responded to the increasing needs ofcare work with privatization (Gutiérrez Rodríguez, 2010). InSpain, the transfer of care work for children and the elderlyto non-EU migrant women has been promoted by the statethrough social and immigration policies. As Fiona Williams(2005: 7) has observed: “care policies themselves maydirectly or indirectly facilitate the employment of migrantwomen as care workers”. Social policies of the Spanishgovernment that aim to address the care of dependents aresupported by a weak network of social services, economichandovers and the delegation of care to the family. Monetarytransfers to families that require care encourage the hiring ofdomestic workers to perform these tasks (Díaz Gorfinkiel,2008a,b; Martínez Buján, 2011). An example of this is theDependency Act (2006), conceived to cover the historicalabsence of institutional support for the care of dependents,

Page 21: Womens Studies International Forum 46 (2014)

19S. Gil Araujo, T. González-Fernández / Women's Studies International Forum 46 (2014) 13–23

including the consolidation of paid care work within privatehomes. Themajority of autonomous communities (the regionalgovernments are responsible for implementing the law) havetaken on this commitment by transferring money to familiesseeking care. According to recent research, most of these fundshave been used to subsidize the work of a relative or to pay aprivately hired care worker (Martínez Buján, 2011; AgrelaRomero et al., 2010) who takes care of the dependents and alsocarries out domestic work (Martínez Buján, 2010). In terms ofthe Law of Work and Family Life Conciliation, conciliation isusually a women's issue that is resolved individually, with thehelp of other family members or by hiring women to dodomestic work with very low incomes and limited rights.8

The passage of Latin American migrant women into thedomestic work sector

Coinciding with the increase in migration flows, thenumber of female workers who claim to work full time inthe domestic work sector has grown dramatically in recentdecades, from about 320,000 in 1994 to close to 800,000 in2007 (Díaz Gorfinkiel & Orozco, 2009). Although a feature ofdomestic work is its high degree of informality, the data foraffiliation to the Home Employment Regime give an ideaof the growth of the sector and its degree of feminization,“foreignization” and Latin Americanization. According toSocial Security data, the number of workers coming underthe Home Employment Regime rose from 156,019 in 2001 to296,293 in 2011 (Ministerio de Empleo y Seguridad Social,2012). In December 2011, of the 182,695 foreign affiliates inthat sector, 77% were women from non-EU countries. Themain countries of origin were Bolivia, Ecuador, Colombia andParaguay. The data shows that Bolivian migrants “specialize”in the sector (24% of the affiliations) given the recent arrivalof this flow in Spain, compared with other countries likeEcuador, Colombia, Peru and the Dominican Republic. Figuresalso show a strong concentration of Bolivian and Paraguayanmigrants in the domestic work sector, with 45% of Boliviansand 49% of Paraguayans affiliated with the Home Employ-ment Regime (Ministerio de Trabajo e Inmigración, 2011).Based on data from the Ministry of Employment and SocialSecurity, in April 2013 there were 183,659 non-EU migrantsaffiliated with the “Home Employment Special Regime”(Régimen Especial de Empleo de Hogar) at least 60% of themwere from Latin American countries (Ministerio de Empleo ySeguridad Social, 2013).

But the reality far exceeds these figures. Many domesticworkers' associations put the number of migrant womenworking in the sector at 700,000 (González Fernández, 2013).Domestic employment remains the niche of the irregular andinformal economy that allows many migrant women to enterthe labor market regardless of their skills or administrativestatus. Data presented by the Colectivo IOÉ and Fernández(2010), based on the National Survey of Migrants in 2007,indicates that for 74% of Bolivian women the first job in Spainwas in domestic service; and 50% of those from Ecuador, Peru,the Dominican Republic and Colombia had their first job in thissector. On the other hand, based on the survey Encuesta dePoblación Activa, of the National Institute of Statistics (INE),Actis has stated that in 2012, 34.4% of the (documented andundocumented) Latin American women employed in Spain

were working in the domestic sector; 69.9% of themwere fromBolivia (personal communication).

The passage of Latin American migrant women intodomestic work positions has nothing to do with their skillsor any “trend.” It is the result of the lack of regulation of thesector, its chronic informality and the implementation ofcertain policies, which regulate migration flows and the entryof migrant population into the Spanish labor market, andespecially into domestic work. Since the mid-1980s, whenthe first Immigration Law was passed in Spain and migrationbecame a field of state regulation, domestic work has beenthe main way to get access to a work and residence permitfor migrant women in general and Latin Americans in par-ticular, due to the lack of a “national” workforce to cover thedemand in the sector.9 For that reason it is one of the sectorsincluded in the quota system, which in the first decadefunctioned as a tool for regularizing migrants who werealready in Spain. That was the case of María who came fromthe Dominican Republic to Madrid in 1998. She receivedsupport from relatives who were already in the country.They found her a job looking after an elderly lady as a live-indomestic worker. Through that first job contract she managedto obtain both a work and a residence permit, therefore shecould regularize her situation in Spain.

I came on 20August 98. (…) The idea cameupbecause Iwasgoing to a house, she was a cousin of my mother's, then shehad daughters who were here. (…) and one day mymothersaid, why don't you tell her to tell the daughters to helpyou. And in that respect I was very shy, I still am a bit. Thenafterwards Imade upmymind, she toldme shewas going totalk to the daughters. At that time they had a contract, andthey said yes. I asked them what the conditions were like,they told me to come over here, and we were talking, theygave me a price for everything (…).

Then you paid them for the contract and for helping withthe papers (…) but the contract was for a job here or justto fix the papers?

No, once I was here I was in their house. They worked at ahairdresser's and that lady needed somebody to look afterthe grandparents. So they took me into the house. Andonce I was in the house they arranged the papers for me.(…) Then, while I was in that house, I took out the secondresidence permit. The first one for a year, followed by theothers. (Madrid, 2006)

Over the last decade, in interviews and informal conversa-tions with Latin American migrant women within the frame-work of different research projects (see note 2) we haveconfirmed that domestic work is the key sector for migrantwomen's entry into the Spanish labor market, whether theyare documented, undocumented or even if they hold Spanishcitizenship. In many cases, these women initiated family mi-gration projects because it would be easier for them to get a joband regularize their situation. It is in this sense that we speak offeminization of Latin American migration flow (Pedone et al.,2012). This is illustrated by the story of Laura, who came fromEcuador and firstly migrated to Belgium (1994), then to Spain(1995) and after a brief and failed return to her place of origin,

Page 22: Womens Studies International Forum 46 (2014)

20 S. Gil Araujo, T. González-Fernández / Women's Studies International Forum 46 (2014) 13–23

in 1999 shemigrated back to Spain again and brought her family(husband and three children) to Barcelona. In the beginningshe worked looking after a child. Later on she was able to starther own “cleaning business”with her sister.

And when we were in Ecuador things didn't work out aswe'd planned, and we came back again, I took the decisionto travel again myself, not to Madrid this time but straightto Barcelona. Well, I was looking, Murcia, Madrid…looking for contacts to have a job and be able to bringmy family because on the second journey I came with theidea of bringingmy family. Andwell, then, through contactsI came to Barcelona and when I was in Barcelona I alreadyhad a job, I made the contacts for the apartment and… myhusband and three children came.

And you, what were you working at?

I was looking after a child. In domestic service as well. InBelgium I worked looking after a child, of that lady, thatcouple, and I came to Madrid with her here, and I wasthere all the time with her. Looking after the child. Andthe second time I came too, because there was a bit ofcontact between my old bosses, and they got me a joblooking after a child here, in Barcelona. Then after that itwas a lot of hours, from 9 in the morning till 9 in theevening, and I didn't see my children, so I quit the job andstarted to look for a job per hour. And I started workingper hour. And then I started to meet people, to meet morepeople, and I started to have a lot of work, and I started totake people to help me because I couldn't cope and wehad to open a cleaning business between my sister andme. (Barcelona, 2006)

The historical relations between Spain and Latin Americaare also important when it comes to exploring the role ofmigratory regulations in the concentration of female workersfrom some Latin American countries in domestic work. Anexample is the Visa Waiver Agreement for stays of less thanthree months, signed at the time of the emigration of Spanishpeople to South America. The visa waiver enabled manymigrants from Latin America to come as tourists and staylonger to work in the informal economy until they couldregularize their status. Since the early 1990s, due to thegrowing flows from Latin America, Spain imposed a visa forseveral Latin American countries: Peru 1992, the DominicanRepublic 1993, Cuba 1999, Colombia 2001, Ecuador 2003 andBolivia 2007. In any case, the majority of Latin Americancountries do not need a visa to stay less than three months,among them Argentina, Brazil, Honduras, Paraguay, Uruguayand Venezuela. But the last regularization legislation was in2005. Women who came to Spain after 2005, mainly fromBolivia and Paraguay, have also mostly found jobs as domesticworkers, but they had to return to their countries of origin inorder to obtain a work and residence permit, therefore manyremain undocumented. That is what Mabel, from Cochabamba,Bolivia, who arrived in Barcelona in 2006, recounts:

That until a job comes up that can get me a work contract—that would be wonderful. (…) The requirements mean[for the papers] going to the country of origin, doing the

formalities there, waiting the time for the visa. If one ofthem offers me a job right now, they give me the okay, Ihave to travel to Bolivia… and thenwhat do I do? Do I leavemy daughters here? Who with? To travel there, at bestthey've taken two months, the fastest. I have to considerleaving enough to pay for two months' expenses here, myreturn ticket, and suffer for two months there for them togiveme the visa to comeback here, in fact… if they could dothe papers without my travelling. Because there's been thatamnesty, my sister got them with that amnesty, she's gotthe papers and didn't need to travel [referring to the lastregularization in 2005]. (Barcelona, 2007)

As well as affecting the direction, composition andfeminization of the migratory flows from Latin America andpromoting the entry of many of the Latin American migrantwomen into domestic work, migratory regulations configurea labor force stripped of all kinds of rights. The interconnec-tion of social policies, labor regulation, migration rules andgender relations at the core of the Spanish domestic workorder encourages the return of the patronage (patronazgo)relations in many of the Spanish households in postcolonialtimes, which were characteristic features of the colonial era(Gutiérrez Rodríguez, 2010). We focus on some of thoseimplications in the next section.

Migrant domestic workers and the right to family life: anabsent subject in political and academic debates

The everyday life of domestic workers consists of lookingafter children, the elderly or the sick, doing household tasks,mediating in family conflicts, and learning to cope with laborrelations in between affection and power. In short, it consistsof everything that the invisible and undervalued domesticand care work conceals, as Marisa, a 51 year old Colombianwoman who came to Madrid in 2006, reflects:

We provide a wonderful service … They should beconscious that we are offering them good work. Whatmore than taking care of their parents, older people, theirchildren, so that they can work and get their own thingsdone … The responsibility of caring for an older person,that is a really big responsibility … All that we want is forthem to value our work a bit more, because we, many ofthe domestic workers, we put a lot of life, soul and heartinto our work, and we do things with a lot of love. So itshould be appreciated. (Madrid, 2010)

The main problem regarding their employment situa-tion and working conditions comes up repeatedly in all theinterviews: that this is an under-regulated, informal andsubmerged sector. Despite the integration of domestic workersinto the General Employment Regime due to a new regulationthat came into effect in January 2012, the legislation itselfcontinues to place them at a disadvantage within a regimethat still fails to recognize basic rights such as unemploy-ment benefits and prioritizes personal and family aspectsover labor ones.10 Policies claiming to safeguard the privacy ofhomes alsomake work inspections less likely. Inferior workingconditions are compounded by domestic work regulations and

Page 23: Womens Studies International Forum 46 (2014)

21S. Gil Araujo, T. González-Fernández / Women's Studies International Forum 46 (2014) 13–23

a lack of control, which puts domestic workers in a position ofheightened vulnerability.

As Gutiérrez Rodríguez has pointed out: “Domestic work isnot only badly paid because it is signified as non-productive,but because those doing this work are feminized and racializedsubjects considered ‘inferior’ to the hegemonic normativesubject” (Gutiérrez Rodríguez, 2010: 15). Despite the vitalimportance of the work migrant women provide for the re-production of Spanish society, the Spanish government keepsall routes to regularized migration closed, with the exceptionof specific quotas, which entail figures that are a far cry frommeeting the demands of the sector. All the women who havemigrated outside the channels organized by the state are forcedto work in a situation of irregularity and to trust in thediscretion of their employers. They have no contract and nowork permit. Theyhaveno labor rights.Most of them live in thehouses where they work. The ones who work per hour oftenhave difficulty in renting a home, even though they haveenough income, which they cannot prove. Nor do they haveaccess to family reunification through formal paths. That iswhy in the case of the women from countries where the visaexemption agreements are in effect the reunifications areusually done independently (Pedone et al., 2012). As we haveanalyzed on other occasions (Gil Araujo & Pedone, 2013;Pedone & Gil Araujo, 2008), even for Latin American migrantwomen with work and residence permits it is difficult to meetthe economic and housing requirements that family reunifica-tion regulations demand. In other cases, even though theycan meet the legal requirements, the long working days inthe domestic work sector make care for their own childrenunviable in the context of migration. These obstacles to familybecome more complex and far-reaching when enacting trans-national motherhood, which the women rarely freely choose.In this regard it is essential to focus more attention on howimmigration policies impact andperpetuate transnational formsof family organization. As Bernhard, Landolt and Goldring pointout: “family dispersal is becoming part of the ‘new normal’rather than an exception” (2008: 25).

The so often mentioned conciliation of work and familylife is not on the agendas of policy-makers, political partiesor in academic research concerning migrant families (Caixetaet al., 2004; Pedone et al., 2012). This omission confirms thepersistent short-sightedness in relation to migrant womenwho are imagined and constructed as reunited and depen-dent “housewives.” Perhaps this is why the latest Immigra-tion Law reform (2009) excludes relatives in the ascendinglines of the non-EU workers from reunification, accusingthem of being unproductive and looking to cash in on welfarebenefits. These imaginaries (which generate quite specificeffects) revive old and still unresolved questions concerningthe value of reproductive labor (which is allotted to women)and the weakness of a system that links rights to havinga salaried status. But they also show the power of whatAbdelmalek Sayad (1999) called Penseé d'État, which struc-tures our thinking about the world, distinguishing nationalsand non-nationals, as if such a distinction were part of humannature and not a social construction. This discrimination be-tween ‘nationals’ and ‘non-nationals’ also organizes access torights, and it works as the latest (and natural) justification toexclude migrant women from citizenship. Within this frame-work, the achievement of civil rights formiddle classwomen to

conciliate their professional and family life is contingent ondenying “non-national” female domestic workers' right to afamily life of their own. That is what Hondagneu-Sotelo (2006)refers asNewWordDomestic Order. A representative of SEDOAC(an organization of migrant domestic workers) reflects as wellon this issue:

They talk about the work–family balance, if families havea bit of purchasing power, a little more, they try to getthis, the work–family balance, but at the cost of bringingin another woman. It isn't achieved by dividing the tasksbetween sexes but another woman takes her place, andit's even worse because it isn't just that she will do thehousehold chores like she did in her [own] home butthat she will be cancelled out as a person. That family orthat other woman is going to appropriate, take over, 100percent of her time and her life. So that is even moreregrettable. In other words, we've achieved the life–workbalance at the expense of completely cancelling out anotherwoman. (SEDOAC, Madrid, 2010)

Conclusions

In this article we have explored the connection betweenSpanish migration policies, family policies, gender relations,the “de-nationalization” or “foreignization” of the domesticwork sector and feminization process of Latin Americanmigration to Spain, as well as the implications in terms ofinequalities based on gender, class, legal status and nationality.As we have shown, in addition to the limitations on family lifethat the grueling pace of domestic work implies (especially inthe case of live-in workers), migration legislation imposesrequirements for housing, earnings and employment thatmake official family reunification very difficult for them. Theselimitations on the right to family life have deepened familytransnationalism, encouraging various forms of long-distancemotherhood, rarely freely chosen by the women themselves.

Any approach to the particular forms adopted by contem-porary South–North migrations must take into account thebroader context of (i) unequal relations between populationgroups, countries and regions and (ii) the international, sexualand racial division of productive and reproductive labor. In theSpanish case, the political, social and economic changes thatbegan in the early 1980s generated a higher participation ofwomen in the labor market. However, this access of womento the field of production was not accompanied by a sharingof tasks in the field of reproduction. This was not assumed as aresponsibility of the social state either. On the contrary, alldomestic and care work continued (and still continues) to beconsidered a private matter to be managed by families and,more specifically, by women. Many of these families have‘solved’ the conflict of sharing domestic work by transferringthem to other women, generally migrants, many of whom arein an irregular situation and come from peripheral countriesthat were once colonies within the Spanish Empire. In Spain,as in other countries in Southern Europe, reproductive laborhas traditionally gone to families, and within them to women.The connection between the provision of domestic work andwomen's international migration is the basis of the currentsolution of many European states, which use migration policy

Page 24: Womens Studies International Forum 46 (2014)

22 S. Gil Araujo, T. González-Fernández / Women's Studies International Forum 46 (2014) 13–23

to satisfy the demand for domestic workers. The privatedemand for migrant workers for domestic work has beenregulated and guaranteed by the Spanish state through itsmigration policies. In all the regularization processes, domesticwork has operated as the main route for migrant women ingeneral, and for Latin American women in particular, to gainaccess to regularization. Domestic work is also one of thesectors included in the quota system currently in force. In thecurrent economic crisis and with unemployment rising, thissector again provides Latin Americanwomenmigrants away toprovide subsistence for their families, in both their countries oforigin and destination.

The disengagement of the state from the welfare of thepopulation fueled by neoliberal social reforms since the mid-1980s, a limited welfare state, a culture of care centered on thefamily, lessmen taking on domestic duties, the aging population,and an increase in the presence of women in the job market;all these processes have led to the emergence of strategies ofoutsourcing and to the foreignization of reproductive socialtasks. The reorganization of reproductive labor offered on themarket has neither resulted in a division of labor between menand women, nor in improved social protection. Instead, it hasresulted in the transfer of inequalities from “national”women to“non-national”women from the global South. Nevertheless, themajority of these women have not been recognized as subjectswith families of their own andwhomight need care themselves.They are the absent key actors in this “debate.”

The New World Domestic Order has strong implicationsin terms of perpetuating inequality based on gender, class,ethnicity and nationality. The transfer of inequality betweenwomen involved in the outsourcing of domestic work alsobrings with it a deep stratification of the right to family life,care and affection. One of its conditions of possibility is theexpropriation of civil rights from a sector of the population.This expropriation is legitimized by the logic of the Penseéd'État (Sayad, 1999) that divides ‘nationals’ and ‘non-nationals’and thus naturalizes differentiated access to rights—includingfamily life—stratified to national origin. In this scenario, mi-grant families have transformed their structures, redefinedtheir roles, and constructed strategies for managing everydaylife in transnational contexts. From our point of view transna-tional motherhood should be seen as a substantial componentof the strategies of resistance and survival of many LatinAmerican families.

Endnotes

1 The changes in the composition of Latin American immigration in Spainexpress the social, economic and political conditions experienced in the countriesof origin. The Cuban flow arriving in the 1960s after the Revolution wasmade upof peoplewith a high economic standing. The first Argentine immigrationwas anoutcome of the exile during the 1970s. Andean immigration began in the 1980s,with the arrival of students and professionals. As for the Dominicans, the pioneerfemalemigrantswere students. In themid-1980s, especially inMadrid, therewasan increase in the number of women from rural origins working in domesticservice. Then, the second phase of Cuban immigration was motivated by thedesolate living conditions in Cuba. In the early 1990s, the growth andpreponderance of the Argentine presence revealed that the emigration of themiddle classes and professionals was a result of the hyper-inflationary crisisduring the late 1980s. In the late 1990s, therewas a notable rise in the number ofmigrants from Ecuador and Colombia, propelled by the crisis following thedollarization of the Ecuadorian economy and the spread of violence in Colombia.The new rise in registered Argentine migrants, which grew over 90% between

2002 and 2003, was a result of the economic and political collapse of Argentina inDecember 2001. Between the years 2004 and 2006, the number of Ecuadoriansarriving decreased. Theywere replacedbymigration fromBolivia,which lessenedwhen the entry visa requirement for the Schengen area came into effect in April2007 (Gil Araujo, 2008).

2 Migration Policies, Family Transnationalism and Civic Stratification: LatinAmerican Migration to Spain, funded by the Spanish Ministry of Science andInnovation (2010–2012). Fundamental rights situation of irregular migrants inthe EU (2009–2011), funded by the European Union Agency for FundamentalRights (FRA); Civic Stratification, Gender and Family Migration Policies in Europe,supported by the “Node”—New Orientations for Democracy in EuropeResearch Program, Ministry of Education and Culture, Austria (2006–2008).

3 The authorwas one of the founders of Asociación deMujeres Dominicanasen España (AMDE), and so this pioneering text—the first in Spain withtransnational aspirations—does not come from the academic sphere but fromthe field of migrant women associations.

4 Cf. for the case of Peruvian women Escrivá Chordá, 1999 for the case ofDominican women Gregorio Gil, 1996 and Cerón Ripoll, 1999; for the case ofEcuadorian women Cortina Nido, 2000.

5 This concept alludes to the existence of family and kinship supportnetworks to cover care tasks and guarantee the support, cohesion and welfareof themembers—tasks that have been traditionally consideredwomen's duties.

6 Spanish population forecasts indicate a clear increase in the number ofpeople over 60: 16.6% in 2007, 21% in 2021 (Díaz Gorfinkiel & Orozco, 2009).

7 Data on time spent working in the household show that 93% of womendo household chores and childcare, in contrast to 70% of men. Women spend5 h a day on these tasks (two more than men) and have less spare time.However, in recent years, men have been spending more time caring forchildren. Age is also an important point of differentiation in terms of timespent on housework: those under 25 spend just 1 h a day on these tasks(Díaz Gorfinkiel & Orozco, 2009, based on Encuesta del Empleo del tiempo2002–2003, Instituto Nacional de Estadística).

8 Women have taken 98.4% of maternity leaves and have applied for 66%of time off to care for dependent family members. Women also account for95.8% of those who left the work market for family reasons, 96.5% of thosewho are not seeking employment for family reasons and 97.2% of those whoclaim to work part time on family responsibilities (Díaz Gorfinkiel & Orozco,2009, based on Statics of Instituto de la Mujer).

9 One regulation that differentiates non-EU migrants' access to the jobmarket is the reference to the national employment situation (NationalPriority). This establishes that ‘insufficiency or scarcity of Spanish labor in theactivity or profession and geographical zones in which work is sought’ will betaken into account for granting or renewing awork permit. Thismeans that therefusal of a work permit can be justified if there are applicants who are Spanishcitizens, EU members or legal residents.

10 A new regulation on domestic work was approved in November 2011.It has been in effect since January 2012. Royal Decree 1620/2011, of 14November, regulating the special labor relation of service in the family home:http://www.boe.es/boe/dias/2011/11/17/pdfs/BOE-A-2011-17975.pdf. However,owing to thehighdegree of irregularity of LatinAmerican femaleworkers and thefact that the jobs are mostly in private homes, the effect this change in the lawmay have on the working and living conditions of the female migrant workers islimited.

References

Actis, Walter (2009). Inmigrantes latinoamericanos en España. Una visión deconjunto.Migraçoes, 4, . Lisboa: Observatorio da Immigraçao (Retrieved from:http://es.scribd.com/doc/31857955/ART-inmigracion-latinoamericana)

Agrela Romero, Belén,Martín, PalomoMaria Teresa, & Langa, RosadoDelia (2010).Modelos de provisión de cuidados: género familia y migraciones. Nuevosretos y configuraciones para las políticas públicas,. Alternativas, 17, 9–17.

Anderson, Bridget (2000). Doing the dirty work? The global politics of domesticlabor. Londres: Zed Publishers.

Arango, Joaquín (2004). Inmigración, cambio demográfico y cambio social.Información Comercial Española, 815, 31–44.

Bernhard, Judith, Landolt, Patricia, & Goldring, Luin (2008). Transnationalizingfamilies. Canadian immigration policy and the spatial fragmentation ofcare-giving among Latin American newcomers. International Migration,47(2), 3–31.

Cachón, Lorenzo (2002). La formación de la ‘España inmigrante’: mercado yciudadanía. Revista Española de Sociología, 97, 95–126.

Caixeta, Luzenir, Gutiérrez Rodríguez, Encarnación, Tate, Shirley, & Vega Solis,Cristina (2004). Hogares, cuidados y fronteras. Derechos de las mujeresinmigrantes y conciliación. Madrid: Traficantes de Sueños.

Page 25: Womens Studies International Forum 46 (2014)

23S. Gil Araujo, T. González-Fernández / Women's Studies International Forum 46 (2014) 13–23

Calavita, Kitty (2005). Immigrants at the margins: Law, race, and exclusion insouthern Europe. Nueva York: Cambridge University Press.

Catarino, Christine, & Oso, Laura (2000). La inmigración femenina en Madridy Lisboa. Hacia una etnización del Servicio doméstico y de las empresasde limpieza. Papers Revista de sociología, 60, 183–207.

Cerón Ripoll, Paloma (1999). La migración de dominicanas en Madrid. (PhDDissertation). Madrid: Universidad Complutense de Madrid.

Colectivo I.O.É., & Fernández,Mercedes (2010). EncuestaNacional de Inmigrantes2007: mercado de trabajo y las redes sociales de las migrantes. Madrid:Ministerio de Trabajo e Inmigración.

Colectivo IOÉ (1991). Trabajadoras extranjeras de servicio doméstico enMadrid. Ginebra: Organización Internacional del Trabajo.

Colectivo IOÉ (2001).Mujer, inmigración y trabajo.Madrid: IMSERSO-Ministeriode Trabajo y Asuntos Sociales.

Colectivo IOÉ (2012). Impactos de la crisis en la población inmigrante. Madrid:Organización Internacional para las Migraciones (Retrieved from: http://www.colectivoioe.org/uploads/0bae582aa3b0842a9eaf50cde16f4f97d9527bcb.pd)

Cook-Martín, David, & Viladrich, Anahí (2009). The problem with similarity:Ethnic affinity migrants in Spain. Journal of Ethnic and Migration Studies,35(1), 151–170.

Cortina Nido, Alejandro (2000). ¿Quién trabaja en nuestras casas? Mujeresecuatorianas en el servicio doméstico en Valencia. (MA Thesis). Madrid:Universidad Autónoma de Madrid.

Díaz Gorfinkiel, Magdalena (2008). El sistema internacional de cuidados: laexperiencia de Madrid en la desnacionalización del trabajo reproductivo.(PhD Dissertation). Madrid: Universidad Carlos III.

Díaz Gorfinkiel, Magdalena (2008). El Mercado de trabajo de los cuidados yla creación de las cadenas globales de cuidado: ¿cómo concilian lascuidadoras? Cuadernos de Relaciones Laborales, 26(2), 71–89.

Díaz Gorfinkiel, Magdalena, & P. Orozco, Amaia (2009). La organización social de loscuidados y vulneraciónde los derechos enEspaña. SantoDomingo:ONUWomen.

Escrivá Chordá María Ángeles (1999). Mujeres peruanas del serviciodoméstico en Barcelona: Trayectorias Sociolaborales. (PhD Dissertation).Barcelona: Universidad Autónoma de Barcelona.

Gallardo Rivas, Gina (1995). Buscando la vida: dominicanas en el serviciodoméstico en Madrid. Santo Domingo: CIPAF–IEPALA.

Gil Araujo, Sandra (2005). Nation-state building process and cultural diversity:Spain. In Jochen Blaschke (Ed.), Nation-state building processes and culturaldiversity (pp. 325–348). Berlin: Edition Parabolis.

Gil Araujo, Sandra (2008). Migraciones latinoamericanas hacia el Estadoespañol. La reactivación del sistema migratorio transatlántico. In IleanaRodríguez, & Josebe Martínez (Eds.), Postcolonialidades históricas:In/visibilidades hispanoamericanas/colonianismos ibéricos (pp. 189–220).Barcelona: Anthropos.

Gil Araujo, Sandra (2010). Las argucias de la integración. Políticas migratorias,construcción nacional y cuestión Social. Madrid: IEPALA.

Gil Araujo, Sandra (2010). The coloniality of power andethnic affinity inmigrationpolicy. The Spanish case. In Encarnación Gutiérrez, Manolea Boatca, & SergioCosta (Eds.), Decolonizing European sociology: Transdisciplinary approaches(pp. 179–194). Surrey: Ed. Ashgate.

Gil Araujo, Sandra, & Pedone, Claudia (2013). Migration policies and family lifeacross the Atlantic. Latin American migrant families in Spain. In ThomasGeisen, Tobias Studer, & Erol Yildiz (Eds.), Migration and family. Theory,research and fields of action. Wiesbaden: VS-Verlag.

González Fernández, Tania (2013). Globally interdependent households:Irregular migrants employed in domestic and care work in Spain. InAnna Triandafyllidou (Ed.), Irregular migrant domestic workers in Europe:Who cares? (pp. 187–208). Aldershot: Ashgate.

Gregorio Gil, Carmen (1996). Sistemas de género y migración internacional: laemigración dominicana en la Comunidad de Madrid. (PhD Dissertation).Madrid: Universidad Autónoma de Madrid.

Gutiérrez Rodríguez, Encarnación (2010). Migration, domestic work andaffect: A decolonial approach on value and the feminization of labor. NewYork and Abingdon: Routledge.

Hochschild, Arlie R. (2000). Global care chains and emotional surplus value.In Will Hutton, & Anthony Giddens (Eds.), On the edge: Living with globalcapitalism (pp. 130–146). London: Jonathan Cape.

Hondagneu-Sotelo, Pierrette (2001). Doméstica: Immigrant workers cleaning andcaring in the shadows of affluence. Berkeley: University of California Press.

Hondagneu-Sotelo, Pierrette (2006). Narrating the New World Domesticorder. IIIS discussion paper no 150. Available at SSRN: http://ssrn.com/abstract=925656, http://dx.doi.org/10.2139/ssrn.925656

Instituto Nacional de Estadística (2013). Avance de la Estadística del PadrónContinuo a 1 de enero de 2013. Datos provisionales. Notas de prensa, April22nd, Madrid (Retrieved from: http://www.ine.es/prensa/np776.pdf)

Kofman, Eleonore, & Raghuram, Paravati (2009). The implication of migration forgender andcare regimes in the south.UnitedNationsResearch Institute for SocialDevelopment, Social Policy and Development. Programme Paper Number 41.

Lutz, Helma (2008). Introduction. In Helma Lutz (Ed.), Migration and domesticwork; a European perspective on a global theme. Aldershot: Ashgate.

Martín Palomo, María Teresa (2008). Los cuidados y las mujeres en lasfamilias. Política y Sociedad, 45(2), 29–47.

Martínez Buján, Raquel (2003). La reciente inmigración latinoamericana aEspaña. Serie Población y Desarrollo, 40. Santiago de Chile: CELADE-CEPAL.

Martínez Buján, Raquel (2008). Bienestar y Cuidados: el oficio del cariño.(PhD Dissertation). Mujeres inmigrantes y mayores nativos. A Coruña:Universidad de A Coruña.

Martínez Buján, Raquel (2010). Servicio doméstico y trabajo de cuidados:Hacia la privatización del cuidado familiar. Alternativas, 17, 157–179.

Martínez Buján, Raquel (2011). La reorganización de los cuidados familiares enun contexto demigración internacional. Cuadernos de Relaciones Laborales,29(1), 93–123.

Ministerio de Empleo y Seguridad Social (2012). Afiliados ocupados a laSeguridad Social, Diciembre 2012. Retrieved from: http://www1.seg-social.es/ActivaInternet/groups/public/documents/rev_anexo/rev_031356.pdf

Ministerio de Empleo y Seguridad Social (2013). Afiliación de extranjeros a laSeguridad Social, Abril 2013. Retrieved from: http://www1.seg-social.es/ActivaInternet/groups/public/documents/rev_anexo/rev_032024.pdf

Ministerio de Trabajo e Inmigración (2011). Afiliación media de extranjeros a laSeguridad Social junio 2011. Madrid: Ministerio de Trabajo e Inmigración(Retrieved from: http://www.slideshare.net/IntegraLocal/afiliacin-media-de-extranjeros-a-la-seguridad-social-junio-2011#btnNext)

Oso, Laura (1998). La migración hacia España de las mujeres jefas de hogar.Madrid: Instituto de la Mujer del Ministerio de Trabajo y Asuntos Sociales.

Parella Rubio, Sonia (2000). El trasvase de desigualdades de clase y etnia entremujeres: los servicios deproximidad. Papers Revista de sociología,60, 275–289.

Parella Rubio, Sonia (2003).Mujer, inmigrante y trabajadora: la triple discriminación.Barcelona: Anthropos.

Pedone, Claudia (2006). Estrategias migratorias y poder. ‘Tú siempre jalas a lastuyos’. Quito: Editorial Abya Yala.

Pedone, Claudia, Agrela, Belén, & Gil Araujo, Sandra (2012). Políticas públicas,migración y familia. Una mirada desde el género. Papers Revista deSociología, 97(3), 541–568.

Pedone, Claudia, Echeverri, María Margarita, & Gil Araujo, Sandra (2014).Entre dos orillas. Cambios en las formas de organización de las familiasmigrantes latinoamericanas en España en tiempos de crisis global. InMaría Eugenia Zavala Castelo, & Virginie Rosee (Eds.), Las migracionesfemeninas en América Latina y las transformaciones en las relaciones degénero. México: Colegio de México (in press).

Pedone, Claudia, & Gil Araujo, Sandra (2008). Maternidades transnacionalesentre América Latina y el Estado español. El impacto de las políticasmigratorias en las estrategias de reagrupación familiar. In Carlota Solé,Sonia Parella, & Leonardo Cavalcanti (Eds.), Nuevos retos deltransnacionalismo en el estudio de las migraciones (pp. 149–176). Madrid:Observatorio Permanente de la Inmigración.

Pedreño, Andrés (2005). Sociedades etnofragmentadas. In Andres PedreñoCanovas, & Manuel Hernández Pedreño (Eds.), La condición inmigrante.Exploraciones e investigaciones desde la Región de Murcia (pp. 75–103).Murcia: Universidad de Murcia.

Parreñas, Rhacel Salazar (2001). Servants of globalization. Women, migrationand domestic work. Stanford: Stanford University Press.

Sánchez Martín, María E. (Coord.) (1992). Nuestras hermanas del Sur. Lainmigraciónmarroquí y el servicio doméstico enMadrid,Madrid, DirecciónGeneral de Migraciones. Unpublished.

Sayad, Abdelmalek (1999). Immigration et pensée d'État. Actes de la rechercheen sciences sociales, 129(1), 5–14.

SEIE-Secretaría de Estadode Inmigración y Emigración (2006).Anuario Estadísticode Inmigración. Año 2005.Madrid: Ministerio de Trabajo y Asuntos Sociales.

Vega, Cristina, García, Mar & Monteros, Silvina (2004). Trabajo doméstico y decuidados. Estrategias de conciliación y condiciones de empleo en distintasunidades familiares. Desigualdades de género, clase y etnicidad/origen.Unpublished Report.

Vega Solís, Cristina (2009). Culturas del cuidado en transición. Espacios, sujetose imaginarios en una sociedad de Migración. Barcelona: Editorial UOC.

Vicente Torrado, Trinidad L. (2006). La inmigración latinoamericanaen España, expert group meeting on International Migration andDevelopment in Latin America and the Caribbean, Population Division,Department of Economic and Social Affairs, United Nations Secretariat,Mexico City, 30 November – 2 December 2005.

Williams, Fiona (2005). Intersecting issues of gender, ‘race’ and migration inthe changing care regime of UK, Sweden and Spain. Paper presented toAnnual Conference of International Sociological, Conference Research Com-mittee 19, September 8–10. Chicago: Northwestern University.

Williams, Fiona, & Gavanas, Anna (2008). The intersection of childcareregimes and migration regimes: A three country study. In Helma Lutz(Ed.), Migration and domestic work. A European perspective on a globaltheme (pp. 13–28). Aldershot: Ashgate.

Page 26: Womens Studies International Forum 46 (2014)

Women's Studies International Forum 46 (2014) 24–32

Contents lists available at ScienceDirect

Women's Studies International Forum

j ourna l homepage: www.e lsev ie r .com/ locate /ws i f

Domestic work and international migration in Latin America:Exploring trajectories of regional migrant women in domesticservice in Argentina

Corina Courtis a,⁎, María Inés Pacecca b

a Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Instituto de Ciencias Antropológicas, Universidad de Buenos Aires, Puan480, (1406) Ciudad Autónoma de Buenos Aires, Argentinab Instituto de Ciencias Antropológicas, Universidad de Buenos Aires, Puan 480, (1406) Ciudad Autónoma de Buenos Aires, Argentina

a r t i c l e i n f o

⁎ Corresponding author.

0277-5395/$ – see front matter © 2014 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.wsif.2014.01.002

s y n o p s i s

Available online 22 February 2014

Besides emigration towards developed countries, Latin America has a regional migration dynamicsof its own— one inwhich the presence ofwomen, aswell as their employment in domestic service,has proved decisive. Combining a macro perspective with a case-based socio-anthropologicalapproach, this paper examines international migration and domestic service at an intra-regionallevel. Drawing on statistical information, we first present an outline of the regional migrationcontext and the insertion of migrant women as domestic workers in destination countries of theregion. The core section of the article centers on the particular case of Argentina, and illuminates theexperience of migrant domestic workers in the Buenos Aires Metropolitan Area. The paper closeswith a series of reflections on the operation of gender as an organizing principle of relations andopportunities involved in international migration.

© 2014 Elsevier Ltd. All rights reserved.

Introduction

In recent decades, a significant increase in internationalmigration (CEPAL, 2007) – chiefly to developed countries –

has brought migration into the limelight and to the center ofdiscussion in numerous forums. The growing share of womenin migration trends – conceptualized, at least qualitatively, asthe “feminization” of migration flows – has become a matterof growing academic interest, and has made apparent theneed to introduce a gender perspective in migration studies.

As a region,1 Latin America has a substantial share in themigration processes that presently raise concerns among“First World” States, as most Latin American countries havestrengthened their role as labor exporters and emigrants'destinations have diversified beyond the United States, toEuropean countries — particularly Spain, Japan, Canada,Australia and Israel. There are a considerable number ofLatin Americans living outside their countries of birth: it isestimated that at least 4% of the region's population lives in

ll rights reserved.

extra-regional countries and that this figure accounts forapproximately 13% of international migrants worldwide.International migration from the region, which stands outfor its increase in women migrants and their widespreadentry into the service market of host countries, particularlythe domestic service sector2 — the emigration of SouthAmericanwomen to Spain, the US, and Canada being especiallyrelevant, has also contributed to the quantitative relevance ofwomen in contemporary international migrations.

But besides emigration to developed countries, Latin Americaalso shows amigration dynamics taking place within the region,one inwhichwomen play a decisive role, with domestic work asa sector of paramount significance. By combining a macro-perspective with a case-based, socio-anthropological perspec-tive, this paper approaches the relationship between interna-tional migration and domestic work at the Latin American level.

Domestic work in Latin America may be traced back tocolonial times (Kuznesof, 1989), and there is a long-standinglink between female migration and employment in domesticservice in the region. Since the 1960s, urban middle-classhouseholds in several Latin American countries have relied

Page 27: Womens Studies International Forum 46 (2014)

Table 1Latin American-born migrant women employed in domestic service in mainregional destination countries (00 and % on total stock of economicallyactive regional migrant women) circa 2000.Centro Latinoamericano y Caribeño de Demografía (CELADE), ECLAC'sPopulation Division. Data and special tabulations provided by IMILA Project(Investigación de la Migración Internacional en Latinoamérica, www.eclac.cl/celade).

Country of residence Total (00) % of total stock of economicallyactive regional migrant women

Argentina 81,194 29.3%Chile 13,149 42.6%Costa Rica 15,978 35.5%Venezuela 49,863 28.2%Total Latin America 177,004 27.1%

25C. Courtis, M.I. Pacecca / Women's Studies International Forum 46 (2014) 24–32

on migrant women (both internal and international) forhousework and childcare. This demand has set off and fueledthe movement of hundreds of thousands of young girls andwomen, giving rise to an early, albeit thriving feminizationprocess supported on “family” logic: the migrant woman'sfamily of origin, confident that household work is much moreprotective than factory or office work; and the employingfamily at the destination, willing to hire (and house) girls andwomen through work agreements where labor rights areeasily blurred in favor of (fictional) kinship practices. Asshown by a general overview of regional statistical informa-tion,3 this process has been particularly active in Argentina,Venezuela and Costa Rica in the past decade. The analysis ofcontemporary Argentine cases, focusing on the experience ofBolivian, Paraguayan and Peruvian migrant women taking onjobs as domestic workers in the Buenos Aires MetropolitanArea (BAMA),4 allows us to review different stages of themigration process (deciding on migration, organizing thetrip, arriving at the destination, finding lodgings and work,saving or sending remittances, family reunification, etc.) inorder to highlight the relevance of gender and gender-basedpractices as organizing principles of many of the relation-ships and opportunities involved in international migrations(see Grieco and Boyd, 1998; De Jon, 2000; Pessar, 1982;Piper, 2007, among other authors who have explored therelationship between gender and migration).

Context: international migration, female migration anddomestic work in Latin America

International migration has been an integral componentof Latin American history. Vis-à-vis the drop in overseasmigration inflows5 (which were remarkably high at the turnof the twentieth century and played an influential demo-graphic and cultural role in some countries), recent decadeshave seen a rebound in intra-regional migration — a long-standing trend characterized by the ever-changing economicand political conditions of the countries in the region.Argentina, Costa Rica and Venezuela have historically beenthe main “receptors” of regional migrants. According toMartínez Pizarro (2008):2, in 2000 “the number of regionalimmigrants stabilized in the leading immigration countries(i.e., Argentina and Venezuela), and increased considerablyin Costa Rica and Chile.” Also, during the 1990s and 2000s,intra-regional migration found new impulses, particularlytowards border or neighboring countries, bringing the topicof humanmobility into the integration agendas, as can be seenthroughout the proceedings of the Conferencia Sudamericana deMigraciones (since 2000), the Foro Especializado de Migracionesdel Mercosur (since 2003), and the Foro Andino de Migraciones(since 2008).

The growth in the share of women among the migrantpopulation is a tendency observed in several intra-regional flowsas well. Although the extent of feminization varies depend-ing on the origin and destination of migration flows, it cangenerally be asserted that there has been a rise in femalemigration, from 44.5% of total intra-regional migrants in 1960to 50.3% in 2005.6

Several factors are at stake when it comes to determiningthe shares of men and women in migration flows, namelythe nature of labor markets both in origin and destination

countries, specific labor demands, family reunification decisionsand the existence of networks. In the case of female mobilityin the region, migration flows have become increasingly labormarket-related, and to a large extent cater to demands in theservice sector, particularly for domestic workers and carersfor elderly or disabled persons.7 International female migrantshired as domestic workers in the region amount to 27% of thetotal female migrant workforce, a striking figure, which isnevertheless lower than the sizeable 40% accounted for by LatinAmerican and Caribbean women working in the same nichein Spain. The transnationalization of domestic and personalservices in both the developed world and Latin America – thedifferences between which would be worth exploring8 – is butan expression of workforce flexiblilization and cost reductionachieved by labor markets through the articulation of gender,ethnic and class relations. A close look at the stocks of femalemigrants employed as domestic workers by country of originand country of residence yields a more detailed picture of theregional situation. Information from the 2000 census revealedthat the major destination countries in the region also hadhigher stocks of migrant women hired as domestic workers:Argentina (81,000 women) stands out side-by-side withVenezuela (50,000 women), followed by significantly lowernumbers in Costa Rica (16,000) and Chile (13,000). However,the highest percentages of domestic workers over the totaleconomically active regional female migrants could be ob-served in Chile (42%) and Costa Rica (35.5%) (Table 1).

Regarding source countries, the 2000 census showed thatcertain countries stand out as “exporters” of women findingjobs in this niche in the region: Colombia was the sourcecountry of 46,000 women employed in the domestic servicesector, chiefly in Venezuela; Paraguay hadnearly 35,000womenresiding in this capacity,mainly inArgentina; andPeru had beenthe country of origin of 33,000 domestic workers distributedbetween Argentina and Chile. Yet, when considering theproportion of migrant domestic workers from each countryover the total number of economically active migrant womenof the same origin, it is evident that Peruvian women exceedthe regional average (48%), followed by those from Nicaragua(38%, with Costa Rica as their principal destination), Paraguay(36%), Guatemala (32%, mostly concentrated in Mexico) andColombia (30%).

Regarding the demographic profile of cross-border femaledomestic workers in the region, it should be noted thatschooling is either high or intermediate in several migrationflows – particularly as far as Peruvianmigrants are concerned –

Page 28: Womens Studies International Forum 46 (2014)

26 C. Courtis, M.I. Pacecca / Women's Studies International Forum 46 (2014) 24–32

and that most of them are mothers and household heads. Tocomplete this scenario, it should be added that the vulnera-bilities created by the precarious nature of domestic work canbe – and usually are – tied in with the vulnerabilities arisingfrom migration. In recent years, the concerns voiced by theinternational community on the vulnerabilities affecting mi-grant women have largely emerged from the realities resultingfrom the combination of international migration and domesticwork (see, for example, ONU, 1993; Staab, 2003; Cortés, 2005).

Let us now examine some of the textures this intersectioncan bring about in the experiences of Latin American migrantwomen working in the domestic service sector in Argentina,which concentrates, by large, the highest absolute number ofLatin American migrants employed as paid domestic workersin the region.

In Argentina, the link between domestic work and femalemigration deserves special attention. Internal and cross-bordermigration to large cities kept domestic service costs low enoughtomake themaffordable to themiddle class (Jelin, 1976). Studiessuch as Berger (1986), Chaney and García Castro (1989),Szretter (1985), Zurita (1983, 1997) and Zurutuza andBercovich (1986a,b, 1987) further develop this bind. Someinclude a brief diagnosis as well as the occupational profiles ofthe population working in this niche (Goren, 2000; Torres &Mazzino, 1996; Mezzatesta and Raimundo, 2001), and intro-duce the migration variable. In addition, migration studiesemphasizing femalemobility usually consider domestic service(Cacopardo, 2002, 2004; Cortés & Groisman, 2004; Cacopardoand Maguid, 2001; CECYM et al., 2005; Pacecca, 2000, 2010).However, specific studies – particularly from a qualitativeperspective – on contemporary international migrant do-mestic workers are still scarce (Buccafusca & Serulnicoff,2004; Canevaro, 2012).

Rethinking migration and domestic work in Argentina

During the late nineteenth and early twentieth centuries,through the influx of over 2,000,000 European immigrants,Argentina clearly became an immigration country. Since the1850s, the country has also received small albeit steady inflowsfrom neighboring and nearby countries such as Bolivia, Brazil,Chile, Paraguay, Uruguay and – more recently – Peru. Untilthe mid-twentieth century, these Latin American immigrants(always less than 3.5% of the total population) remained in theArgentine provinces closest to their own home country. Overthe past few decades, however, the Buenos Aires MetropolitanArea9 has become a key destination for regional migrants,especially for those coming from Bolivia, Paraguay and Peru —

currently the largest and most active inflows. These immi-grants enter the labor market as an ancillary workforce to thenative population, and take on unstable and poorly paid jobsthat rarely abide by labor law requirements (see Benencia andKarasik, 1995; Cerrutti, 2005; Maguid, 1990, 1995 and 1997;Marshal, 1983; Marshall and Orlansky, 1983).

Migration policies have had a bearing on regional migrants'limited access to secure jobs, as well as on their generally poorhousing, education and health situations. Unlike the welcominglegislation originally designed to target the historical massiveimmigration form Europe, Argentine migration law and regu-lations became notably more restrictive vis-à-vis the increasedregional composition of immigrant inflows. Since the 1960s,

a combination of lax terrestrial border control, by which peoplefrom neighboring countries could easily enter as tourists,and progressively stricter – even unfulfillable – in-countryrequirements for obtaining a residence permit generated a vastpopulation of irregular migrants, and created a delegitimizingframework for them. The restrictive nature of Migration Act No.22.439, sanctioned in 1981, during the last military regime, washardened by several regulations passed in democratic times,and until 2004, when a new Migration Act was approved, theonly alternatives that regional migrants had in order to seekresident status were periodical amnesties and limited bilateralagreements.

Now, the current composition and geographic distributionof immigration described above are linked to the presence ofwomen immigrants seeking jobs as domestic workers in theMetropolitan Area.10 34% of the 240,000 women from Bolivia,Paraguay and Peru that where censed in Argentina in 2001worked in domestic service, following the occupational path ofearlier internal migrants employed in the same unprotectedniche — a niche where, according to a 2004 Ministry of Labor,Employment and Social Security report, over 96% of theworkers were unregistered.11

The overriding presence of migrant women from neigh-boring countries has been analyzed in terms of feminizationof contemporary migrations to Argentina. Even though thenotion of feminization holds true from a quantitative stand-point, it requires specification. First, feminization patterns varyaccording to country of origin: the Paraguayan migrationunderwent an early feminization and urbanization process,and Paraguayan women in the Buenos AiresMetropolitan Areawere noticeable as early as the 1960s— i.e. themasculinity rateof Paraguayan migration decreased from 101 in 1960 to 79 in2010. Bolivian migration to the Metropolitan Area did notbegin to grow until the 1980s; however, due to still attractiverural and construction jobs,12 the feminization process isslow-paced: masculinity rates went from 125 in 1980 to 99 in2010. Furthermore, the economic conditions of the 1990s13

spurred the hitherto sparse Peruvian migration with highlyfeminized inflows, so that the masculinity rate plummetedfrom 198 to 81 (see Tables 2 and 3).

Secondly, considering feminization in a qualitative sense –

alluding to the autonomy of female migrants, or at least tomigration patterns unrelated to those ofmales – , the scenariosdiffer between Paraguayan and Peruvian women on the onehand, and their Bolivian counterparts on the other. Womenfrom Paraguay and Peru find job opportunities mainly asdomestic workers in the Buenos Aires Metropolitan Area and –

to a lesser degree – in trade and communal services. Unlikethem, Bolivian women tend to find domestic service jobs atother destinations, aswell as employment in other sectors suchas agriculture,manufacturing and trade. The smaller number ofBolivian women in domestic service and the broader spectrumof options available to them may be accounted for by thefamily nature of Bolivian migration and by the role of ethnicnetworks in the accumulation of capital, used to start upethnic enterprises (such as horticulture since the 1980s andclothes manufacturing since the 1990s). Such niches havebeen consolidated due to the additional workforce providedby new arrivals (Benencia, 2006; Aranda, 1997, 2005). Theslower drop in the masculinity rate of Bolivian immigrants,along with the relevance of family-based enterprises must

Page 29: Womens Studies International Forum 46 (2014)

Table 2Evolution of Latin American migration to Argentina and evolution of masculinity rates by country of origin. Selected countries, 1991–2010.INDEC 1996 and 2010.

Masculinity rate

Country of birth 1980 1991 2001 2010 1980 1991 2001 2010

Bolivia 118,141 143,569 233,464 345,272 125.4 107.3 101.3 98.7Brazil 42,757 33,476 34,712 41,330 85.5 77.3 71.8 72.9Chile 215,623 244,410 212,429 191,147 114.6 99.9 91.7 87.1Paraguay 262,799 250,450 325,046 550,713 85.6 78.7 73.5 79.7Uruguay 114,108 133,453 117,564 116,592 95.2 95.2 92.5 90.8Peru 8561 15,939 87,546 157,514 197.9 146 68.5 81.9Total selected countries 753,428 857,636⁎ 1,010,761 1,402,568 117 100.7 83 86.0

⁎ Includes migrants from neighboring areas whose country of origin is unknown.

27C. Courtis, M.I. Pacecca / Women's Studies International Forum 46 (2014) 24–32

be taken into account when discussing feminization inqualitative terms. What role do adult women play infamily-based enterprises? Are both gender-based roles(men and women) equally relevant for the success offamily-based enterprises? Current ethnographic research isbeing carried out in order to provide at least preliminaryanswers to these questions.

In order to highlight certain aspects of the individual andhousehold processes underlying the trends revealed bycensus data, the following pages analyze a set of in-depthinterviews conducted with migrant women working in thedomestic service sector in the Buenos Aires MetropolitanArea. Eighteen women who had migrated to Argentina fromBolivia, Paraguay and Peru between 1996 and 2003 wereinterviewed during August and September 2005. Theinterviews were conducted out of the women's workplaces,mostly in their homes or at locations that they themselvessuggested. The design of the intended sample followedseveral criteria. First, cut-off dates were set to homogenizethe interviewees' opportunities for accessing formal resi-dent status made available by existing regulations.14 Sincethe selected period includes the dramatic social, politicaland economic crisis that struck Argentina in December 2001(which also ended the currency exchange parity, thusaffecting migrants' remittances), the interview sample includ-ed one woman of each nationality entering the country after2001. Respondents were required to be mothers of childrenborn in their home countries;15 it was required that at least oneof them had been under fourteen – fully dependant – at thetime of their respective mothers' migration to Argentina. Thefollowing pages present the information collected through theinterviews and discuss how gender dimensions and practicesplay relevant and intertwined parts in the networks thatsupport several of the different stages of themigration process.

Table 3Women born in neighboring countries residing in BAMA by country of origin(2001 and 2010) (00 and %).Argentine National Population Censuses 2001 and 2010.

2001 2010

Country Total stock residingin Argentina

% residingin BAMA

Total stockresiding inArgentina

% residingin BAMA

Bolivia 116,002 52 173,779 56Paraguay 187,323 75 306,434 76Peru 52,389 73 86,615 73

Table 4 below summarizes a few key socio-demographictraits of all eighteen interviewees: age, age at migration, ageat childbirth, number of children and educational advancement.These dimensions trace core responsibilities and resourcescritical to the development (and understanding) of migrationtrajectories.

All interviewees came from large families (from four totwelve siblings) dwelling in rural areas, small villages ormajor cities. Their formal education ranged from incompleteprimary schooling to complete college studies. Regardless ofeducational achievements, the interviews revealed discon-tinuous schooling experiences related to early entrance intothe labor market. The low number of children per woman isworth noting, as well as the fact that only three of them hadArgentine-born offspring. The average age of the children(the eldest in the cases of siblings) was seven when themothersmigrated to Argentina. As of the date of the interviews,seven respondents had brought some or all of the children toArgentina.

Five respondents raised their children as single parents.Of the remaining interviewees, six were separated from thefather of their children and one was a widow. Thus, prior tomigration, divorce and widowhood had turned these womeninto heads of households, a status they singled out as one ofthe major reasons for moving to Buenos Aires. Only six of allinterviewed women lived in Buenos Aires with their spousesand fathers of their children. In half of the cases, the womenmigrated first, and their husbands and children followed whenthey were already employed and fit to ensure their families'upkeep in Buenos Aires.

Regarding the work experience of the respondents beforecoming to Buenos Aires, none of them had been unemployedin the source country. All had entered the labor marketbefore motherhood, and although some had given up work toraise their children, most were the sole person responsible fortheir children'swelfare beforemigrating toArgentina. Their jobsas nurses, petty shopkeepers, street sellers, cooks or domesticworkers provided barely enough to support their children andclearly forfeited their access to quality education — the mainlegacy that they wished to pass on to their offspring.

All interviewees pointed out they had expected to migratetemporarily (no longer than two years), just long enough tosave the money that would allow them to significantlyimprove their life in their home country by buying a house,starting a business or paying for their children's education.All of them entered the country as tourists and stayed aftertheir visa had expired. Of eighteen respondents, ten were

Page 30: Womens Studies International Forum 46 (2014)

Table 4Socio-demographic characteristics of respondents.

Case 1 Case 2 Case 3 Case 4 Case 5 Case 6 Average

Paraguayan womenYear of birth 1978 1978 1976 1979 1959 1977 –

Age at childbirth 15 17 18 19 24 17 18Number of children 1 2 1 1 3 3 1.8a

Age at migration 23 21 23 24 38 21 25Education Incomplete

primaryeducation

Incompleteprimaryeducation

Completeprimaryeducation

Completeprimaryeducation

Completeprimaryeducation

Incompleteprimaryeducation

Peruvian womenYear of birth 1968 1974 1966 1966 1969 1970 –

Age at childbirth 23 18 20 19 23 21 20Number of children 2 1 1 5 3 1 2.1b

Age at migration 29 21 34 34 26 33 29Education Complete

collegeeducation

Incompletesecondaryeducation

Incompletesecondaryeducation

Incompletesecondaryeducation

Completecollegeeducation

Completecollegeeducation

Bolivian womenYear of birth 1971 1970 1963 1966 1972 1972 –

Age at childbirth 20 17 19 23 18 18 19Number of children 1 1 2 2 5 3 2.3c

Age at migration 22 20 33 39 21 28 27Education Complete

secondaryeducation

Incompleteprimaryeducation

Incompleteprimaryeducation

Completesecondaryeducation

Completesecondaryeducation

Completesecondaryeducation

a,b and c. are not “final” averages because many of these women, especially those of Paraguayan origin, are young enough to bear more children.

28 C. Courtis, M.I. Pacecca / Women's Studies International Forum 46 (2014) 24–32

in irregular migratory conditions when the interview wasconducted. To justify this situation, they mentioned the highcost of regularization procedures and their conviction that,employment-wise, having proper resident documentationdid not entail any comparative advantages.

Even as independent or autonomous migrants, women'smigration is usually linked to family decisions, and thefamily is where gender subordination is most heavily felt.As household heads, the interviewees decided to migratemainly on account of the constraints that economic factorsimposed on their children's upbringing and education. Albeita single-handed resolution, migration was decided in theimpoverished context of marital separation and, as such, itrequired no negotiation in a subordinate husband–wife rela-tionship. However, family involvement was relevant and mosthousehold members discussed advantages, drawbacks andschemes for re-organizing the daily life of those left behind.These issues show that a woman's chance to migrate frequentlydepends on another woman stepping in and taking over herhousehold and childcare duties; whichwomenmay go dependson which women stay.

Knowing that they would probably find jobs as domesticworkers and work long hours, all of them migrated alone,leaving their children with their relatives. For some of thewomen, migration meant relying on another woman (usuallyher mother or her sister) to raise her children. Initially, thedivorced women had arranged for their ex-husbands tolook after the children; however, these arrangements wereshort-lived, either because the fathers couldn't cope andhanded the children over to the maternal grandmothers orbecause the mothers were uneasy about how their childrenwere being cared for. With regard to the women whose

husbands later joined them in Argentina, the childrenwere leftin charge of the fathers, who received considerable supportfrom their extended families.

Concerning the organization of the movement, regardlessof the source country three facts shared across all respon-dents are worth noting:

- the City of Buenos Aires was the direct destination of therespondents' journeys, i.e., there was no stage-migration;

- except for the Bolivian interviewees, who migrated withtheir husbands, the other women stated that their migra-tion was made possible by the support and assistance ofanother woman, who had migrated before them and wasalready residing in Buenos Aires;

- fifteen women had to borrow money to move to BuenosAires; however, none of them resorted to bank loans norborrowed from strangers: rather, their trip was financed bythis previous female migrant, who paid for all or most ofthe travel costs.

The figure of the previous migrant woman varies accordingto the interviewees' source country, the most significantdifference being their generation. The Paraguayan respondentswere assisted by their aunts (by blood or marriage), i.e., bywomen belonging to their mothers' generation, who hadmigrated several years and even decades before. By contrast,the Peruvian interviewees received assistance from their sistersor cousins, i.e., from women of their own generation who hadreached Buenos Aires much later than the Paraguayan aunts.These preceding migrant women are a fundamental source ofsupport for newcomers; in general, they have already set upjobs for the new arrivals even before their departure from their

Page 31: Womens Studies International Forum 46 (2014)

29C. Courtis, M.I. Pacecca / Women's Studies International Forum 46 (2014) 24–32

hometown, and provided accommodation in their housesduring their first months of their stay in the host country.

With regard to entry into the labor market, the absence ofsignificant periods of unemployment is quite striking: almostall the respondents got jobs within fifteen days of theirarrival in Buenos Aires. Besides, during the initial phase oftheir residence in the host country, they took jobs as live-inmaids and sent remittances back home while their childrenstayed behind in the country of origin.

Due to the characteristics of their own domestic units, theinterviewees had to save considerably to send remittances totheir children and pay off the loans that they had taken fortheir journey. During the 1990s, when the Argentine pesowas pegged to the US dollar, live-in maids were easily able toremit between USD100 and USD200 per month, plus theycould save enough money to pay for their children's trips toArgentina or their own trips back home. After the 2001 crisis,the dollar value of remittances dropped to such an extentthat these women were deprived of the opportunity to saveany money. Despite this scenario, they always managed tosend no less than 50% of their income, whichmeans that theirliving standards and purchasing power in Argentina weresignificantly affected (a demographic outlook on the impactof the 2001 crisis on immigration can be found in Maguid andArruñada, 2005). All the respondents mentioned theirpunctuality in the sending of remittances (“it was the firstthing I always did”), and their trust in the remittancebeneficiaries, as they never made any reference to improperlyallocated funds.

In migration contexts, the opportunities for saving andsending remittances back home are tightly connected todwelling cost reduction strategies.16 Thus, in the case of thissurvey's respondents, they initially lived in the house of thewoman that aided them in their act of migrating: theyresided there either for free or in an expense-sharing fashion.In the cases in which the children rejoined the respondents inBuenos Aires at a later stage, the move to a more adequatedwelling was funded with the discontinuation of remittances.Apart from this, the arrival of the interviewees' children ledthem to switch jobs from the “live-in”mode to its “by the hour”counterpart.

The respondents brought their children to Buenos Aires atdifferent times. Several factors seem to account for this: theirjob stability, the opportunities for accessing a dwelling withsufficient privacy for their families and continued schoolingfor their children. Besides leading these women to makeadjustments to their jobs as well as arrangements for a properhome, this decision implied that they would have to care forand spend time with their children, who were in turn facedwith the need of readapting to their relationship with theirmothers, and adapting to the new location. Added to theactualization of affection and desires, this led the intervieweesto narrate this particular experience with mixed emotions,between happiness about reunification and uncertainty aboutthe forthcoming change in their lives.

For these workers, employment conditions have alwaysbeen irregular, even in the few cases where resident statuswas regular. Still, their employers – often women, likethemselves – sometimes chose to give them holiday pay aswell as a year-end bonus. Not only did the interviewees viewthese as acts of generosity on their employers' side: but they

also failed to perceive non-standard employment as a violationof their rights as workers or as an act of abuse. This calls for anevaluation of how the respondents describe their relationshipwith their employers.

The interviewees have worked as domestic workers infamily residences, occasionally taking jobs in elderly care.Withonly one exception, all of them had been hired by women ofArgentine origin. The initial interview with their employersusually involved agreeing onworking hours, the tasks expectedof them (cleaning, washing and ironing, etc.), their salary andmode of payment. However, at this point, agency on thepotential employee's side appears blurred: conditions are setby la patrona (“the mistress”, as the interviewed womenreferred to their employer), and the candidate either accepts orrejects them, but does not negotiate. For example, a Peruvianwoman once took a job by the hour and, after having workedfor 8 h the first day – without having discussed her wages – ,shewas paid the equivalent of just 2 h, an amount the potentialemployer justified by saying it was what she usually paid allher domestic workers. These situations tend to occur becausethe patrona–employee relationship is agreed upon and takesplace outside a regulatory framework; this is due to ignoranceon both sides of the legal instruments that might preventinformality and precarious work, and to a general lack ofagreement on exactly what tasks are included in domesticwork and what they are worth.

The blurriness of this situation is comparable to anotherissue, which became evident when the interviewees evalu-ated their work conditions. In general, when they describedthem as “good” or “passable” they were not alluding to anemployment/contractual relationship, but rather to a per-sonal relationship experienced and expressed in terms offamily and kinship: “they treated me as just another familymember.” Here the respondents made reference to moneyloans, authorizations for long holidays in their home countries,assistance in obtaining medical appointments, medicines,textbooks for their children, and other actions seen as favorsor liberalities from the patrona. This mode of treatment is theprerogative of the employer, whoplays the role of “head of kin”,distributing resources and gifts in this fiction of kinship.Whenever the respondents expressed dissatisfaction, they didso not in connection with excessive job demands but onaccount of being treated in such a way that established adistinct boundary between “the family” and “the employee”, away which defined their relationship with the household interms of market and exchange rather than on the basis ofreciprocity and kinship.

Closing reflections: Women working for women

The household, conceived as a space for domestic life andas a workplace (this being true for both the employer and theemployee, though in differing degrees), becomes a place ofsolidarity and hierarchy, of reciprocity and exchange, of genderand class codes. The patrona and her employee appear incounterpoint, and the differences between them can be eitherexacerbated or softened by gender: the employer may erasethis common denominator and directly exercise the powerafforded to her by class (and by her position as a “purchaser ofdomestic work”) or may use gender to bridge the class gap andbuild the kinship and reciprocity fiction. This in turn can be

Page 32: Womens Studies International Forum 46 (2014)

30 C. Courtis, M.I. Pacecca / Women's Studies International Forum 46 (2014) 24–32

added to generational differences that may shape dependencebeyond economic considerations and transform the employeeinto either her employer's protégée or her victim. Unregisteredemployment figures in domestic work prove how difficult itis to earn respect for rights and benefits that are both long-established and regulated by market and exchange logic inother sectors. The work contract that might have offered thesemigrants access to legal status in the country and would haveled to other rights and better social integration, was thusheld back at the crossroads entailed by the above-describedcircumstances.

However, the interviewees' discourse does not suggestany influence of the employer's condition as a local and theemployee's condition as a foreigner on the relationship betweenboth. Even though interviewees agreed that discrimination onthe grounds of nationality or irregular residence status mayoccur in the street or in institutional settings (such as schools orhospitals), no such episodes were reported by them within thehousehold, and nationality does not seem to be pivotal to theiremployer–employee relationship. There are no elements in ourcorpus that might lead one to think of nationality as a specific oradditional source of ethnic derision, racism or abuse from theemployer. The fact that native (Argentine) women employed inhouseholds are unregistered workers just as often as migrantwomenmight support the interviewees' perception that being amigrant has no special bearing at the interactional level betweenthe patrona and the domestic worker. In this sense, the kinshipfiction and gender/age/ethnicity/class structures seem to operatejust the same for Argentine as for migrant domestic employees.

These interviews have served to unveil the experiences ofwomen who decide to migrate with the support of otherwomen, in the knowledge that their most likely job opportu-nities will be in domestic service, performing “women chores”for other women. Female networks support migrant womenbefore, during and after migration: taking care of their childrenin the source country, helping out with the journey, aiding withaccommodation and employment at destination, and providingjobs. As suggested by Grieco and Boyd (1998), in the casesunder study, gender becomes a core organizing principle ofrelationships and opportunities in the context of internationalmigration. The respondents' decision to migrate was directlylinked to their maternal responsibility to support their childrenand give theman education. The act ofmigratingwas supportedand enacted on the basis of ties of solidarity and reciprocitywithother women in their families. Finally, at their destination,employment as unregistered domestic workers reshapes gen-der structures and makes it difficult for these migrants toposition themselves as workers endowed with rights andobligations established by a third-party (i.e. the State) withwhich no kinship or reciprocity relationships can be established.

In addition, the cases under study reveal gender as acondition for essentially dual and conflicting insertion, be-tween kinship solidarity/reciprocity and market exchange. Inthe same way as female work in their own households has noexchange value, but is indispensable because it subsidizesreproduction costs unmet by salaries, the gendered bonds ofsolidarity and reciprocity subsidize the movement of women/mothers to another labor market (at the destination), wherethey take jobs in an entirely feminized and gendered niche.

Although the interviewees consider migration to havebeen a source of learning and opportunities, and an experience

modifying their perception of their own ability to manageuncertainty (see also Aranda, 2003; Stefoni, 2002; Maher andStaab, 2005), from the perspective of this study, it is evidentthat the influence of gender on the migration context tendsto reinforce complex subordination structures that tie socialstanding with membership in a certain class, gender, age groupand geographic origin.

Endnotes

1 Following the UN's terminology, we refer to Latin America and theCaribbean as a region. The term “regional migration” here refers to migrationprocesses taking place from or to Latin American countries. More specifically,we speak of intra-regional migrations for dynamics in which both origin anddestination countries belong to Latin America, while the term extra-regional isused for processes inwhich either origin or destination countries lie outside theLatin American region. The term “international migrations” implies processesinvolving international border-crossing, as opposed to internal or innermigration.

2 According to the UN definition, “a domestic worker is a personemployed part-time or full-time in a household or private residence, in anyof the following duties: cook, servant or waitress, butler, nurse, childminder,carer for elderly or disabled persons, personal servant, barman or barmaid,chauffeur, porter, gardener, washerman or washerwoman, guard.” (UNFPA,2006: 51).

3 Statistical data on migration and domestic service have been drawnfrom the IMILA Project (Investigación de la Migración Internacional enLatinoamérica) run by Centro Latinoamericano y Caribeño de Demografía(CELADE), ECLAC's Population Division www.eclac.cl/celade. Parcial inter-pretations of these data can be found in Martínez Pizarro 2003, 2005, 2008and 2009; and Kösters, 2008.

4 This section revisits the work conducted by the authors as part of theproject Migración y trabajo doméstico: una aproximación interdisciplinaria, alarger study requested by ILO Chile in 2005. Fragments by Courtis andPacecca (2008, 2010), and by Ceriani et al. (2009 in Valenzuela and Mora,2009) are reproduced here.

5 Martínez Pizarro, Jorge (2003) points out that in 1970, over 75% ofmigrants in Latin America and the Caribbean were extra-regional, and thatthis figure dropped to less than 50% in 2000.

6 UN Population Division: http://esa.un.org/migration/index.asp?panel=1.7 There is also a share of migrant women hired at more qualified jobs

both in the region and away from it.8 Although we will not expand on the issue in this paper, specifics

such as the historical development of domestic work in different countries,traditions of household organization, forms and range of access to domesticservices, women's participation in the non-domestic labor market and thecomposition and dynamics of the domestic work labor market (for instance,migrant women still share this niche with internal migrant women in LatinAmerica) would be worth exploring across contexts in order to graspnuances in the work conditions and experiences of international migrantdomestic workers (Chant, 1992, Hondagneu-Sotelo, 2007, Momsen, 1999).

9 The Buenos Aires Metropolitan Area comprises the City of Buenos Airesand its surrounding boroughs, and concentrates 32% of the total populationof Argentina.

10 The 2001 census reported 700,000 domestic workers: 88.4% of themwere Argentine, whereas 11.6% came from other Latin American countries,chiefly Paraguay, Peru, Chile and Bolivia. 45% of the total domestic workersresided in the BAMA.

11 It is important to emphasize the deficit of public statistical informationabout the employment conditions of domestic workers, particularly due to thehigh informality levels of this labor group. It is alsoworthmentioning that, until2013, domestic workwas regulated by a special regimewhich dated from1956(Decree 386) and granted domestic workers fewer rights than to workers ingeneral. In March 2013, a new law was enacted for domestic workers (Act No.26844/13) that incorporates rights such as maternity leave and other securitybenefits. How this new law will affect domestic workers, and internationalmigrant domestic workers in particular, is yet to be seen.

12 In the Bolivian case, feminization relates both to the increase ofindependent female migrants as well as to men moving with their wives.

13 In the 1990s, two combined processes affected migration from neigh-boring countries, particularly Peru: on the one hand, the collapse of regional

Page 33: Womens Studies International Forum 46 (2014)

31C. Courtis, M.I. Pacecca / Women's Studies International Forum 46 (2014) 24–32

economies, and on the other, a currency exchange system in Argentina thatpegged the Argentine Peso to the US dollar.

14 Between 1996 and 2003, access to temporary or permanent residentstatus was mainly determined by Executive Order 1023, which containedthe regulations for Migration Act No. 22439/81. This act set the following keyconditions: holding an employment contract (which was restricted later, in1998), being married to an Argentine citizen, or being the parent of anArgentine citizen. Another way of getting resident status was through thebilateral Migration Agreements signed in 1999 with Bolivia and Peru. Finally,a new migration law was enacted in January 2004, which has opened upnew alternatives for migrants seeking resident status, especially from 2005onwards. Since the analyzed interviews were carried out in 2005, they donot fully reflect the impact of this legal shift, its reach and limitations.

15 (Not) having Argentine-born children was not relevant to the sample,neither were the nature of domestic work (by the hour, live-in, formallyemployed, etc.), the migration status of the interviewees (regular or irregular/“undocumented”) or the presence or absence of a spouse in the source countryor in Argentina. These dimensions were later analyzed on a case-by-case basis.

16 The Buenos Aires Metropolitan Area offers three types of dwellings todisadvantaged migrants: shanty towns either in Buenos Aires City or itssurroundings; squatted houses; and hotels and boarding houses. All threedwelling types involve seriously deteriorated infrastructures and buildingconditions. Shantytowns are usually located in areas devoid of vehicle accessor public lighting. Squatted houses tend to be concentrated in specificBuenos Aires City neighborhoods enabled with transportation, publiclighting, schools and stores. Shantytowns and squatted houses are the mostaffordable dwelling types, but access to them is enabled only throughcontacts: conversely, boarding houses, though more expensive, are open toanyone who can pay the fee.

References

Aranda, Elizabeth M. (1997). De peones a patrones quinteros. Movilidad socialde familias bolivianas en la periferia bonaerense. Estudios MigratoriosLatinoamericanos, 12(35).

Aranda, Elizabeth M. (2003). Global care work and gendered constraints:The case of Puerto Rican transmigrants.Gender and society, 17(4), 609–626.

Aranda, Elizabeth M. (2005). Migración limítrofe y mercado de trabajo ruralen la Argentina. Estrategias de familias bolivianas en la conformación decomunidades transnacionales. Revista Latinoamericana de Estudios delTrabajo (RELET), 10(17).

Benencia, Roberto (2006). Bolivianización de la horticultura en Argentina.Procesos de migración transnacional y construcción de territoriosproductivos. In E. Jelin, & A. Grimson (Eds.), Migraciones regionales hacia laArgentina. Diferencia, desigualdad y derechos. Buenos Aires: Prometeo Libros.

Benencia, Roberto, &Karasik, Gabriela (1995). Inmigración limítrofe: los bolivianosen Buenos Aires. Buenos Aires: CEAL.

Berger, Silvia (1986). Inserción de la trabajadora doméstica en el mercadourbano. Buenos Aires: Centro de Estudios de la Mujer (CEM).

Buccafusca, Sandra, & Serulnicoff, Myriam (2004). Las trabajadoras extranjerasen el servicio doméstico en la Argentina. Un caso de la feminización de lasmigraciones (draft).

Cacopardo, María Cristina (2002). Mujeres migrantes y trabajadoras endistintos contextos regionales urbanos. Papeles de Población No. 34,CIEAP/UAEM.

Cacopardo, María Cristina (2004). Crisis y mujeres migrantes en laArgentina. Working paper submitted to the II Seminario de la Red deEstudios de Població, Centre d'Estudis Demografics, Feb. 2004.

Canevaro, Santiago (2012). Migración y permanencia de la comunidadperuana. Trayectorias de mujeres peruanas en el servicio doméstico deBuenos Aires (1990–2008). In R. Benencia, A. Pedreño, & G. Quaranta(Eds.), Proyectos y trayectorias migratorias, mercados laborales y políticasen perspectiva comparada (pp. 45–62). Buenos Aires: CICCUS.

CECYM, Cotidiano Mujer, Fundación Instituto de la Mujer, MEMCH, & REPEM(2005). Migraciones, globalización y género en Argentina y Chile. (BuenosAires).

CEPAL (2007). Migración internacional, derechos humanos y desarrollo enAmérica Latina y el Caribe. Síntesis y conclusiones. Santiago de Chile:CEPAL.

Ceriani, Pablo, Courtis, Corina, Pacecca, María Inés, Asa, Pablo, & Pautassi,Laura (2009). Migración y trabajo doméstico en Argentina: lasprecariedades en el marco global. In M. E. Valenzuela, & C. Mora (Eds.),Trabajo doméstico: un largo camino hacia el trabajo decente (pp. 147–189).Santiago de Chile: OIT.

Cerrutti, Marcela (2005). La migración peruana a la ciudad de Buenos Aires:su evolución y características. Población de Buenos Aires, 2(2).

Chaney, Elsa M., & García Castro, Mary (Eds.). (1989). Muchachas no more:Household workers in Latin America and the Caribbean. Philadelphia:Temple University Press.

Chant, Sylvia (Ed.). (1992). Gender and migration in developing countries. London:Belhaven Press.

Cortés, Patricia (2005). Mujeres migrantes de América Latina y el Caribe:derechos humanos, mitos y duras realidades. Serie Población y Desarrollo,61. Santiago de Chile: CELADE.

Cortés, Rosalía, & Groisman, Fernando (2004). Migraciones, mercado detrabajo y pobreza en el Gran Buenos Aires. Revista de la CEPAL, 82, (April2004).

Courtis, Corina, & Pacecca, María Inés (2008). La operatoria de género en lamigración: mujeres migrantes y trabajo doméstico en el AMBA. In MabelGrimberg, Martínez María Josefina, & María Inés Fernández Álvarez (Eds.),Investigaciones en Antropología Social (pp. 157–174). Buenos Aires: Facultadde Filosofía y Letras — UBA/Ed. Antropofagia.

Courtis, Corina, & Pacecca, María Inés (2010). Género y trayectoria migratoria:mujeresmigrantes y trabajo doméstico en el ÁreaMetropolitana deBuenosAires. Papeles de Población, 16(63), 155–185.

Cacopardo, María Cristina, & Maguid, Alicia (2001). International migrantsand gender inequality in the labor market. Paper delivered at the XXIVGeneral Population Conference. San Salvador de Bahia, Brasil: IUSSP.

De Jon, Gordon (2000). Expectations, gender, and norms in migrationdecision-making. Population studies, 54(3), 307–319.

Goren, Nora (2000). La mujer joven y su situación laboral. Master's Thesis.Caracterización y análisis. Buenos Aires: Centro de Estudios Avanzados,Universidad de Buenos Aires.

Grieco, Elizabeth, & Boyd, Mónica (1998). Women and migration: Incorpo-rating gender into migration theory.Working paper. Center for the Studyof Population, Florida State University, College of Social Sciences.

Hondagneu-Sotelo, Pierrette (2007). Domestica: Immigrant workers cleaning andcaring in the shadows of affluence. California: University of California Press.

INDEC — Instituto Nacional de Estadísticas y Censos (1996). Serie Análisisdemográfico. La población no nativa de la Argentina. 1869-1991.

INDEC — Instituto Nacional de Estadísticas y Censos (2010). NationalPopulation Census of 1960, 1970, 1980, 1991, 2001 and 2010. Availableat www.indec.gov.ar

Jelin, Elizabeth (1976).Migration and labor force participation of Latin Americanwomen: The domestic servants in the cities. Buenos Aires: CEDES.

Kösters, Johanna (2008). Migración y el trabajo doméstico en América Latina(draft).

Kuznesof, Elizabeth (1989). A history of domestic service in Spanish America,1492–1980. In Elsa M. Chaney, & Mary García Castro (Eds.), Muchachas nomore: Household workers in Latin America and the Caribbean. Philadelphia:Temple University Press.

Maguid, Alicia (1990). Migrantes limítrofes en la Argentina: perfil socio-demográfico y ocupacional en 1980. Total del país. Buenos Aires:Ministerio del Interior — Fondo de las Naciones Unidas para actividadesde población.

Maguid, Alicia (1995). Migrantes limítrofes en la Argentina: su inserción eimpacto en el mercado de trabajo. Estudios del Trabajo, 10.

Maguid, Alicia (1997). Migrantes limítrofes en el mercado de trabajo delÁrea Metropolitana de Buenos Aires. 1980–1996. Estudios MigratoriosLatinoamericanos, 12(35), 31–62.

Maguid, Alicia, & Arruñada, Verónica (2005). El impacto de la crisis en lainmigración limítrofe y del Perú hacia el Área Metropolitana de BuenosAires. Estudios del Trabajo, 30.

Maher, Kristen Hill, & Staab, Silke (2005). Nanny politics. The dilemma ofworking women's empowerment in Santiago, Chile. International FeministJournal of Politics, 7(1), 71–88.

Marshall, Adriana (1983). Las migraciones de países limítrofes en la Argentina.Memorias del Congreso Latinoamericano de Población y Desarrollo. El Colegiode México México: UNAM, PISPAL.

Marshall, Adriana, & Orlansky, Dora (1983). Inmigración de países limítrofesy demanda de mano de obra en la Argentina, 1940–1980. DesarrolloEconómico, 23(89) (April–June 1983).

Martínez Pizarro, Jorge (2003). El mapa migratorio de América Latina y elCaribe, las mujeres y el género. Serie Población y Desarrollo, 44. Santiago:CEPAL/CELADE.

Martínez Pizarro, Jorge (2005). Tendencias y rasgos salientes de la migracióninternacional en América Latina y el Caribe. Presentation at the Experts'Meeting on Migration and Development. México, D. F., Nov. 2005.

Martínez Pizarro, Jorge (2008). Cambios y Desafíos en la Migración InternacionalLatinoamericana: El Ejemplo de Iberoamérica. Ecuador: PYDLOS.

Martínez Pizarro, Jorge (2009). Notas sobre las características de la fuerzalaboral migrante en las Américas. Paper presented at the Workshop onTemporary Migrant Workers Programs, Regional Conference on Migration,El Salvador, April 2009.

Mezzatesta, Susana, & Raimundo, Nancy (2001). Documento: Mujer y trabajo.Buenos Aires: Consejo Nacional de la Mujer.

Page 34: Womens Studies International Forum 46 (2014)

32 C. Courtis, M.I. Pacecca / Women's Studies International Forum 46 (2014) 24–32

Momsen, Janet Henshall (Ed.). (1999). Gender, migration and domestic service.Londres: Routledge.

ONU (1993). Internalmigration ofwomen in developing countries. Proceedingsof the United Nations Experts' Meeting on the feminization of internal migration.New York: UNO.

Pacecca, María Inés (2000). Género, situación familiar y trayectoria laboralenmujeres migrantes. Proceedings of the V Jornadas Historia de la Mujeres yEstudios de Género, Septiembre de 1998. Santa Rosa: Universidad Nacionalde La Pampa.

Pacecca, María Inés (2010). Vivir y trabajar en Buenos Aires: los migrantesperuanos en el Área Metropolitana. In E. Oteiza, & A. Lattes (Eds.), Patronesmigratorios internacionales en América Latina. Buenos Aires: Eudeba.

Pessar, Patricia (1982). The role of households in international migrationand the case of U.S.-bound migration from the Dominican Republic.International Migration Review, 16(2), 342–364 (Special Issue: Theoryand Methods in Migration and Ethnic Research).

Piper, Nicola (Ed.). (2007). New perspectives on gender and migration — Rights,entitlements and livelihoods. London: Routledge.

Staab, Silke (2003). En búsqueda de trabajo. Migración internacional de lasmujeres latinoamericanas y caribeñas. Serie Mujer y Desarrollo, 51. Santiago:CEPAL.

Stefoni, Carolina (2002). Mujeres inmigrantes peruanas en Chile. Papeles dePoblación, 8(33), 117–144.

Szretter, Héctor (1985). La terciarización del empleo en la Argentina. El sectordel servicio doméstico. Proyecto ARG/84/029, PNUD-OIT.

Torres, Alejandra, &Mazzino, Pablo (1996).Mujeres trabajadoras en la Argentinabrechas en participación, remuneración y política pública. Revista de Trabajoy seguridad Social, 3(10).

UNFPA (2006). Estado de la población mundial. Hacia la esperanza: Lasmujeres y la migración internacional. New York: UNFPA.

Valenzuela, María Elena, & Mora, Claudia (Eds.). (2009). Trabajo doméstico:un largo camino hacia el trabajo decente. Santiago: OIT.

Zurita, Carlos (1983). El servicio doméstico en Argentina. El caso de Santiagodel Estero. Santiago del estero: INCIC, Universidad Católica de Santiagodel Estero.

Zurutuza, Cristina, & Bercovich, Clelia (1986). Yo trabajo en casa de familia.Buenos Aires: Centro de Estudios de la Mujer (CEM).

Zurutuza, Cristina, & Bercovich, Clelia (1986). Las sirvientas: ellas, las otras,nosotras (Research project: Servicio doméstico remunerado y problemas desindicalización). Buenos Aires: Centro de Estudios de la Mujer (CEM).

Zurita, Carlos (1997). Trabajo, servidumbre y situaciones de género. Algunasacotaciones sobre el servicio doméstico en Santiago del Estero, Argentina.Paper submitted at LASA 1997.

Zurutuza, Cristina, & Bercovich, Clelia (1987).Muchacha se necesita: situaciónde la empleada doméstica en la Argentina. Buenos Aires: Centro de Estudiosde la Mujer (CEM).

Page 35: Womens Studies International Forum 46 (2014)

Women's Studies International Forum 46 (2014) 33–44

Contents lists available at ScienceDirect

Women's Studies International Forum

j ourna l homepage: www.e lsev ie r .com/ locate /ws i f

Gender and migration from invisibility to agency: The routes ofBrazilian women from transnational towns to the United States

Gláucia de Oliveira AssisState University of Santa Catarina UDESC, Department of Human Sciences, Av. Madre Benvenuta, 2007, Itacorubi, Florianópolis, SC 88035-001, Brazil

a r t i c l e i n f o

0277-5395/$ – see front matter © 2014 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.wsif.2014.01.003

s y n o p s i s

Available online 10 February 2014

In the late twentieth century, thousands of Brazilians left for the United States to “make it inAmerica,” inserting Brazilians into the new international labor flows. Brazilian women, likeother Latin American immigrants, became concentrated in housecleaning, a labor market thatis segmented by gender, class and ethnicity. Housecleaning became a female emigrationstrategy that allowed women to circulate through the globalized world and insert themselvesin transnational migration. This article analyzes how the configuration of “the housecleaningbusiness” and the organization of domestic labor redefined or problematized gender identities.The data comes from an ethnographic study conducted in Brazil and the New Englandregion of the United States. As housecleaners in the United States, men and women areconfronted with redefinitions of identities that may or may not imply changes in genderrelations.

© 2014 Elsevier Ltd. All rights reserved.

Introduction

This article demonstrates how Brazilian women insertthemselves into international labor flows, into a labor marketof domestic service, which also includes other immigrantwomen. The increased participation of women in migratoryflows has raised significant questions for research andtheories of international migrations, because it reveals thatwomen are active subjects in the migratory process whomake decisions and provoke changes within family andgender relations. Moreover, they initiate and undertake theirown migratory projects and do not only accompany theirhusbands or sons, as is often portrayed in migratory studies.

Immigrant women are inserted into the domestic servicessector and use informal social networks, the so-called ethnicenclaves of immigrants, working as caretakers for the elderly,nannys, part-time cleaning ladies or full-time maids (Anthias,2000; Assis, 2004, 2007; Fleischer, 2002; Foner, 2000;Morokvasic, 1984), and in the sex market (Maia, 2009,Margolis, 1994; Piscitelli, 2007). In this context of feminiza-tion of migratory flows, women participate in networks ofcare and sex, in a labor market that is segmented by gender,class and race. Women participate in the transnational flows

ll rights reserved.

that occur in the globalized world as flexible, precarious andoften invisible sources of labor, because their work takesplace in the private realm. As Saskia Sassen (2003) observes,the feminization of transborder migratory flows must beunderstood in the context of the expansion of the informaleconomy that favors the flexibilization and deregulation ofthe labor force and creates the conditions for absorbingfeminine and foreign labor.

Anthias (2000), upon analyzing migrations to southwest-ern Europe in the late twentieth century, emphasizes that it isnot a question of recognizing the proportional importance ofwomen or their economic and social contributions, but ofconsidering the role of the processes, discourse, and genderidentities within the processes of migration and establishmentin the destination society. This perspective reveals that agendered approach is important for understanding contem-porary migrations, although it does not involve a question ofthe presence of women in the flows, because even when theywere numerically significant, they were not considered in theclassic studies of migration. As demonstrated by Houstoun,Kramer, and Barret (1984), since the 1930s, women haveconstituted the majority of legal flows to the United States, buthave nevertheless remained invisible in studies on migration, a

Page 36: Womens Studies International Forum 46 (2014)

34 G. de Oliveira Assis / Women's Studies International Forum 46 (2014) 33–44

situation that only began to change in the 1970s. Studies byMorokvasic (1984), Gregorio Gil (1996, 2007), Menjivar (2000),Pessar (1999), Foner (2000), Anthias (2000) and Chant andRadcliffe (1992) revealed the theoretical transition that involvedusing the category of gender to consider migratory processes.

Until the early 1970s, women were not found in theempiric analyses or in the studies conducted, as highlightedby Patricia Pessar (1999), Chant and Radcliffe (1992),because many authors were influenced by neoclassicalmigration theories. There was a presumption that men weremore apt to run risks, while women were the guardians ofthe community and stability. This image, favored by thepush–pull theory, understood migration as a result of rationaland individual calculations and relegated women to asecondary place, without recognizing their work asimmigrants.

The increased female participation since the 1970s tookplace in a context of growth of international migrations sincethe second half of the twentieth century. Contemporarymigrants, unlike their antecessors, have access to lower-costcommunication and transportation, which has shorteneddistances and made contacts more frequent between thesocieties of origin and destination.

An important question that remains is the experiences ofwomen of different national origins in the contemporaryflows. As contemporary migration studies demonstrate, thereis a significant migration of Latin, Asian, African and EasternEuropean women that adds new ingredients to the under-standings of these movements. These women arrive withdifferent human capital – many of them with a bettereducational level and higher qualifications than the womenwho arrived in the late nineteenth and early twentiethcenturies. Contemporary immigrants benefit from the ex-pansion of educational and employment opportunities, andfrom more liberal legislation concerning divorce and genderdiscrimination. Although these differences are significant,there are more similarities than differences in the lives ofmigrant women from different national origins.

What they have in common is that, like the women whoarrived more than 100 years ago, contemporary immigrantsare found in labor markets segmented by gender and, despitebetter schooling and qualification, still steered towardscertain traditionally female occupations, which causes afield such as domestic employment for example, which haddecreased in the United States and Europe, only to rise onceagain in the late twentieth century.

Brazilian women, as well as Mexican, Peruvian, Philippineand other immigrant women, have left their countries to dohousecleaning in the United States. Hondagneu-Sotelo(2007) observed that theoreticians of globalization have notstopped to consider the importance of domestic labor—neither those who celebrate it nor those who criticize it. Inthe same way that modernization theories that forecast thedisappearance of domestic service in modern society werewrong, because this form of labor has grown and continues togrow, Hodagneu-Sotelo affirms that neither theoreticians ofglobalization nor those of modernization were able to foreseeor understand the role of domestic labor in the era ofglobalization and post-industrial society. The author observesthat a few feminist scholars and theoreticians began to payattention to the recent expansion and dispersion of these

transnational migrant domestic woman workers in variouscountries.

Another important factor to be revealed is that thisinsertion is not distinguished only by gender, but also bynational origin. Upon analyzing the representations aboutwomen immigrants of different nationalities in recent decadesto Europe, Anthias (2000) revealed how they are categorizeddifferently, according to racial standards and national origin.Some would be classified as victims (such as Sri Lankanwomen), others would be desired for their supposed submis-sion (such as Philippine women), others would be desired fortheir beauty that conforms to Western standards (such asEastern European women).

Therefore, the greater visibility of women in recentinternational migrations has contributed to problematizingthe crystallized visions about the insertion of men andwomen migrants in this process. From the start, the choiceof who will migrate, the reasons for migration, a permanenceor return, takes place within a network of relations thatshape the opportunities of men and women migrants.

This article reconstructs the emigration trajectories ofBrazilian women who left for the United States during thesecond half of the twentieth century. The stories of thesewomen allow us to reconstruct trajectories that began in the1960s and, in the following decades, established transna-tional ties between the locations of origin in Brazil – the cityof Governador Valadares, in Minas Gerais State and the city ofCriciúma, in the southern state of Santa Catarina – and theregion of Boston, Massachusetts in the United States. Theethnographic research that provided the data for this studywas conducted in two moments between 2001 and 2004(Assis, 2004) and in 2008 (Assis & Siqueira, 2009) and wasconducted in two cities in Brazil and in the region of Boston.In these Brazilian cities, emigration became part of the dailylife of many of their residents and is among the lifeexpectations of many youth—to migrate to America. Thetrajectories of women are presented here to demonstratehow migration became a female strategy for social mobilityor a search for a better life, as the migrants themselves say. Inthis scenario, housecleaning, work that has a low status inBrazil, becomes resignified and offers a possibility to “make itin America” and realize the migratory project.

Following Pisicitelli (2008) in her studies of immigrantBrazilian women, I also suggest in this article that “thesemigrants are affected by the imbrication between notions ofsexuality, gender, race, ethnicity and nationality.” Accordingto Piscitelli, these notions imply that whether they are whiteor brown-skinned, in international flows, they are racializedin the countries of the North as mestizas. She adds that thisracialization is sexualized.

The accounts of the women whose trajectories will bereported reveal how this racialization and sexualization ofBrazilian women is intertwined and how the women usethese markers of difference, which at times generateprejudice and discrimination, to have an advantage in thelabor or marriage markets. As we demonstrate in this article,based on the experiences of some of these migrant women,they use attributes of Brazilianness to negotiate their lovingrelationships.

We emphasize the trajectories of migrant women todemonstrate that, although they have emigrated somewhat

Page 37: Womens Studies International Forum 46 (2014)

35G. de Oliveira Assis / Women's Studies International Forum 46 (2014) 33–44

later than the men, and are confronted with different concernsthan men, they also undertook the adventure of emigrating. Inthe process, they wrote their own histories, not only byaccompanying their brothers, boyfriends or husbands, but alsoby establishing their own migration routes and networks.

Between the local and the global: Governador Valadares andCriciúma as the starting point of international migration

Brazil, which until the middle of the past centuryattracted thousands of immigrants in search of a better life,has experienced a significant movement of Brazilians head-ing abroad since the 1980s. Given this movement, the firstacademic works classified these emigrants as exiles from thecrisis – an allusion to the political exiles of the 1970s – who,to escape the Brazilian economic crisis, were obliged tomigrate, as demonstrated by Sales (1991, 1992), Goza (1992)and Margolis (1992, 1994). These articles emphasize emi-gration as an option taken by certain sectors of the Brazilianmiddle class to confront the lack of labor opportunities oropportunity for social mobility generated by the economiccrisis that shook the country in the 1980s – the so-called “lostdecade.”1 The main destinations of this initial flow were theUnited States, Paraguay and Japan.

In the case of Brazilians emigrating to the United States, thestudies began by following the route of the migratory flows. Thefirst studies by Margolis (1989, 1992, 1994), Assis (1995, 1999),Soares (1995) and Sales (1992, 1995) trace a profile of thepopulation and pointed to the city of Governador Valadares(MG) as the starting point for many who reached the Bostonregion. We note that in these studies, as in classic migrationstudies, little attention is paid to gender issues.

During the 1990s, as demonstrated by the works of Martes(2000), Ribeiro (1999), Sales (1999a), Reis and Sales (1999),the flow of Brazilians to the United States remained continu-ous, making the characteristics of the population in terms ofclass, gender and ethnicity more complex, as well as revealingother starting points for the emigration. The studies reveal thatthere was growth in the participation of women and that theywere constructing a niche through their work within domesticservice in the Boston region (Fleischer, 2002; Martes, 2000;Scudeler, 1999), while men were mainly employed in civilconstruction and in the restaurant sector. The studies also tryto problematize the changes in family and gender relations(Assis, 2003; Debiaggi, 2003; Debiaggi, 2002; Fusco, 2001).

In the 1990s, the city of Criciúma emerged as an importantstarting point in southern Brazil for emigrants headed to theUnited States and Europe. What I would like to highlight is thatduring this time migratory networks were constructed andconsolidated in both cities, where one immigrant wouldencourage another to emigrate, bringing friends, relatives andcompatriots, as studies by Massey (1997), Massey et al (1987),Levitt (2001), Gramusck and Pessar (1991) and Hagan (1998)demonstrate, thus transforming these cities into transnationalcommunities, as Fusco (2005) and Assis (2004) observe,establishing transnational ties, and making a significant impacton the daily life of the cities.

In these Brazilian cities, the most visible impact ofmigration is the increase in civil construction and theopening of small businesses with the resources sent byemigrants (Siqueira, 2006; Soares, 1995). When they leave,

emigrants generally think of returning to Brazil. Thus, formen and women, the migratory project usually consists of“building a home andmounting a business,”whichmakes themoney sent home an important mechanism for maintainingties with the country of origin and demonstrating the“success” of the migratory project.

In addition to this more visible aspect, the multiple family,emotional and religious relationships established among theemigrants and those who remain are expressed throughpresents sent by mail, containers with products from theUnited States, trips to Brazil to participate in weddings,birthdays and parties, or having relatives, usually parents,come for visits or to help care for the children, or helping otherfriends and relatives to emigrate. This demonstrates that, inaddition to monetary resources, there is a circulation ofemotional ties that establishes a field of transnational relations.News from abroad reaches these cities by letters and phonecalls, and more recently by email or Skype, with reports like“there in the United States, cleaning is not like it is in Brazil,”which causes many women to migrate knowing they will bedoing this kind of work. I met the women whose trajectories Iwill narrate in this article during my field work. I use fictitiousnames to guarantee their privacy, given that many are stillundocumented, and others preferred not to be identified.

The flow of Brazilians to the United States is predomi-nantly constituted of undocumented immigrants, eitherbecause they traveled with a tourist visa and overstayed thetime they were granted, or because they illegally entered theU.S. through the Mexican border. Some, as is the case ofvarious immigrants from Criciuma, who are descendants ofItalian immigrants (who came to Brazil in the nineteenthcentury) have Italian passports (and thus do not need a visato enter the United States), which creates an advantage whenpassing through immigration (Assis, 2004, 2011).

Nevertheless, the period of stay is the same regardless ifthe person holds a Brazilian or Italian passport and after thisperiod, the “tourists” become undocumented immigrants.Sales (1999a, 1999b) observed that during the 1980s, Brazilianimmigrants constructed a sense of legitimacy for theirclandestine situation. Although undocumented, they wereable to work under false documents, put their children inpublic school, receive hospital care, open bank accounts, etc.Therefore, in daily life, they did not sense a lack of legalization(a need for “papers”). Nevertheless, after the attacks onSeptember 11, 2001, life became more difficult for Brazilianimmigrants and those of other nationalities. The increase inmeasures restricting entrance to the country, granting visas,the greater rigor at airports, increased prejudice, and crimi-nalization of migrants forced them to question the legitimacyof the clandestine situation. Those who were undocumentedfelt more vulnerable to “blitzes” by immigration authorities attheir workplaces, problems when passing through immigra-tion at airports and greater difficulties when faced withsituations where they had not previously experienced prob-lems, even though they were undocumented, for example,renewing a drivers' license or registering children in school.After 2001, obtaining “papers” came to occupy another placein the life and concerns of Brazilian immigrants, who began toseek strategies for legalization.

Since it involves a contingent of mostly undocumentedmigrants, data about Brazilians abroad consists of estimates.

Page 38: Womens Studies International Forum 46 (2014)

36 G. de Oliveira Assis / Women's Studies International Forum 46 (2014) 33–44

According to the Ministry of Foreign Relations, based oninformation from Brazilian embassies and consulates, it isestimated that about 2.5–3 million Brazilians lived in othercountries between 2008 and 2011. The 2010 census2 was thefirst to include a question about international migration andprovide a more precise estimate of the number of Braziliansresiding abroad; reaching 491,645 in 193 countries, with264,743 women (53.8%) and 226,743 men (46.,1%); 60% ofthe emigrants were between twenty and thirty-four yearsold. Recognizing the large difference in estimates and thedifficulties in data collection, one thing the numbers doreveal is a significant presence of women emigrants. The2010 Census also indicates the United States (23.8%) as theleading destination. Nevertheless, when we look at the totalpercentage of emigrants who go to Europe – Portugal(13.4%), Spain (9.4%), Italy (7.0%) and England (6.2%), – wecan see that 36% of all Brazilian emigrants head for thatcontinent, which demonstrates an increased flow to Europein the early twenty-first century.

Marcella emigrated to the Boston region in 1988 andwitnessed the growing flow of Brazilians during the1990s. She arrived with a tourist visa, did not speakEnglish, and was taken in by relatives. Her plan was tosave enough money to return to Brazil and get married.Like other immigrant women she became a domesticmaid and, over time, began her own “cleaning business.”The years passed and Marcella returned to Brazil anumber of times, but was not able to stay, due to whatshe said was the economic instability in the country in the1990s. With her earnings in “America,” she was able tocome and go frequently between the U.S. and Brazil. Eventhough she did not have documents, as Marcellaestablished herself, she helped family and friends in theirmigratory trajectories, revealing the importance of socialnetworks for the consolidation of the flow of Brazilians inthe United States. In her statements, Marcella emphasizesthat “women here feel more secure, independent, there iswork here, you have opportunity. You can go anywhere,any shopping center and they don't want to know if youare a housecleaner or whatever.”3

Marcella, a woman who emigrated to try to make a life in“America,” does work common among immigrant women—cleaning while always avoiding the police, because she has nopapers; she is undocumented. This experience, which iscommon among women from Criciuma and GovernadorValadares that we accompanied during the ethnographicresearch, reveals to us that when the women leave, theyinsert themselves in a market that is segmented ethnicallyand by gender attributes. Domestic work becomes a femaleemigration strategy and allows the women to circulatethroughout the globalized world, cross borders and partici-pate in transnational migration.

Gender and domestic labor—the creation of a labormarket niche for Brazilian women

When they leave for the United States, the migrants takewith them the dream of a better life, they want to “make it inAmerica,” meaning they plan to work and save money to buy

a house and a car, or mount a business and return to Brazil. Inorder to realize this project, however, they must work asundocumented migrants in jobs that they would not performin Brazil. In the case of the group analyzed, the womenconsider themselves to be white and from the middle orlower middle class, with a high school education, and somehave taken classes at college. They worked in Brazil incommerce, as teachers, in banks or as independent profes-sionals. By emigrating to work in cleaning in the United States,they are clearly lowering their social position and occupationalstatus. In this sense, the emigrants interviewed are signifi-cantly different from the profile of women who are employedin domestic labor in Brazil, as demonstrated by Brites (2013)and Brites et al (2013). In Brazil, it is foremost that women whohave less schooling, are poor and describe themselves as blackor brownwho perform these low-pay tasks in Brazil. Thus, thesewomen, many of whom had their own cleaning ladies orfull-time maids before emigrating, try to redefine the loweringof their status, highlighting the differences between houseclean-ing in the United States and Brazil, as we will see below.

Brazilian migrants and those of other nationalities areemployed in the secondary labor market,4 which, as Scudeler(1999), Martes (2000) and Martes (2000) have pointed out,is characterized by low salaries, precarious working condi-tions, little security and high turnover rates. This market isgenerally aimed at women, adolescents, migrants andimmigrants, because it requires little qualification or knowl-edge of the language.

Like other Latin American and Asian immigrants, mostBrazilian immigrants are at the base of the occupationalstructure. Men usually look for work in the constructionsector, in restaurants, supermarkets and large fast-foodchains, while women seek employment through networksof domestic care, housecleaning and baby-sitting, most ofwhich are poorly paid, require few skills and attractundocumented workers who speak little English withoutproviding them basic social security. Although many Brazil-ian immigrants, both men and women, are from the middleclass and worked, for instance, as bank tellers, teachers orsales clerks in Brazil, within the migration of they insertthemselves into a different market and thereby accept adecline in occupational status, which, as we will see in thecase of female immigrants, is resignified through the earningsthey obtain from long hours of work in the “cleaningbusiness.” We can also say that, in addition to being insertedinto an ethnically segregated labor market, these immigrantsrespond differently to the “macro-structural forces” influ-enced by gender relations (Hondagneu-Sotelo, 1994). Listen-ing to the reports about the trajectories of migration, I foundthat many women had already been thinking of doing thiskind of work when they first migrated, because it isconsidered the most profitable and safest for those withoutdocumentation and limited knowledge of the English lan-guage. Thus, many began working for an immigrant who hadbeen there for some time and after they “learned the ropes,”they sought to set up their “own business,” even if it was acleaning business commonly consisting of the immigrantherself, a helper who may be a recently arrived immigrant orrelative, and a car full of cleaning supplies.

According to Fleischer (2002), domestic labor is a themethat has not been very visible in academic studies. The author

Page 39: Womens Studies International Forum 46 (2014)

37G. de Oliveira Assis / Women's Studies International Forum 46 (2014) 33–44

attributes this invisibility in part to the nature of domesticlabor, as it is very common work that is considered to lackcomplexity, that is performed within homes by women whoare usually from classes and ethnic groups that areundervalued, and that deals with elements related to pollution(such as garbage, human waste, dirt). Nevertheless, despitethis invisibility, domestic work has been increasing incountries like the United States. Mary Romero (1997) andothers have emphasized that domestic work is not perceivedas a job, because it is conducted in domestic spaces and relatedto family and leisure activities, as opposed to the world ofwork that takes place in factories or offices. In this context,women are identified as “natural” workers because washing,cleaning and caring for children are gendered tasks, which arecommonly considered to be female. Upon analyzing therelationship between bosses and cleaning women in Brazil,Kofes (2001) notes that while there has been an increase indomestic employment in European countries such as France,England and Spain, more extensive knowledge is in circulationabout entering into this type of employment in the UnitedStates. Kofes emphasizes that:

domestic employment involves the circulation of peoplebetween different social and cultural worlds (betweenclasses, many times ethnicities, people or racializedgroups; between rural and urban; between urban neigh-borhoods, the circuit is feminine and nearly always, thecirculation is of women). This circuit, in contexts oftransnational migration, is not only geographically ex-panded, but also qualified in a distinct manner, althougheverything indicates that it is predominantly feminine(Kofes, 2001, p. 23).

This draws attention to a connection between whatdomestic employees do in various parts of the world andbetween domestic work in Brazil and the United States, aconnection which, during the entire time of the field work,the immigrants try to hold separate by affirming thathousecleaning in the United States is different than in Brazil.In relation to this issue, there are reports that describe theexperiences of immigrants during the 1960s.

Women immigrants in the 1960s – living at work

In the 1960s, with the first flow of Brazilian women to theUnited States, single immigrant women found domestic laborto be a form of conducting the migratory project, whichprovided a certain amount of security, given their status assingle when they migrated, and living at their workplaceoffered them protection from the Immigration and Natural-ization Service (INS) and lowered the risks of living aloneabroad.

Maria, Miriam and Martha emigrated from GovernadorValadares to the United States. Maria emigrated in 1969 atthe age of 24, Miriam in 1980 at 20 and Martha in 1981 at 30.They were all single and worked in retail in the city ofGovernador Valadares. Maria had completed elementaryschool and the two younger ones were high school graduates.Their emigration was primarily motivated by the stories theyheard about the life of emigrants from Valadares in the

United States, tales of modernity, the lifestyle and thepossibilities for a better life.

In these women's imaginary, the idea of “America” is aland of great opportunities. “I thought that here [the USA] wassomething from another world (…). Maria sent photos [of theUSA] and we wanted to go, to leave that little life.” (Martha).

They belonged to the urban lower middle class and livedwith a certain financial difficulty. Maria decided to emigrateto improve her life. She told her parents of her decision.

“I got home, and all choked up, I told my parents that Iwanted to go to the United States. My father said that I mustbe joking and left the room. Then I said to my mother: I wantto go with your consent, but if you don't agree, I will goanyway.”

Maria said that her mother spoke with her father, whoagreed and supported the decision by helping her organizethe papers to get a visa and with some of the travel costs.Upon departure, her father gave her a letter, saying sheshould read it when she was in the plane. In the letter, herecommended she take the precautions that a single womanmust have in a strange land and that she maintain herdignity. Her parents were greatly concerned because sheplanned to emigrate as a single woman.

Maria worked as a maid living in the houses of U.S.families, because she thought it was safer. This type of work,“living at the job,” allowed her greater moral protection, alsofrom the Immigration and Naturalization Service, becauseshe emigrated with a tourist visa and did not have permissionto work, which made her situation different from that ofthose who first emigrated and were able to get work visas. Atthat time, domestic workers did not have any possibility toobtain legalization.

In 1973, Maria married a Brazilian of Portuguese descentwith U.S. citizenship. This allowed her to become document-ed. As we show below, this is the most common form oflegalization for women.

When her sisters decided to emigrate, Maria supportedthem and provided all the necessary means. She gotdocuments for her mother, so her sisters could get GreenCards, except Miriam, who was already living in the U.S. andhad married a Greek man who was a naturalized U.S. citizen.

Miriam and Martha began working as domestic cleaningladies. “We worked live-in jobs,”5 an expression that desig-nates this type of work, which Miriam still does until this day.The work involved dedicating a few hours a day to lookingafter the house and living at work. It allows the women toclean other houses for extra pay.

The three sisters are a family, each has two children, andMaria and Miriam now have grandchildren. They frequentlyreturn to Brazil, Miriam has an “American-style” house inGovernador Valadares where she spends her annual holidays.They all say that they intend to return to live in Brazil at somepoint in their lives.

As Hagan (1998), Hondagneu-Sotelo (1994) and Glenn(1986) observed, when they reach the society of theirdestination, many single women's first jobs are a type ofarrangement known as live-in, which means they work asdomestic maids and live at their workplace. This type of workhas historically been considered as a way of incorporating

Page 40: Womens Studies International Forum 46 (2014)

38 G. de Oliveira Assis / Women's Studies International Forum 46 (2014) 33–44

immigrant women into the new society, because it means thatrecent arrivals will not have to spend money on food, housingor transportation. These women are inserted into a precariousform of labor that is not regulated, and in a position where it isdifficult to become legalized. Living at the job means longwork hours, inseparability of work and leisure, and makes itmore difficult to develop contact networks outside thehousehold. Despite this situation of inequality and exploita-tion, from the perspective of these women, such arrangementsdo allow them to migrate and remain in the United States.

The first fieldwork I conducted on immigrants fromValadares in the Boston region included some women in thissituation (Assis, 1995). In general, they were middle-agedwomen (older than 40) with little schooling, and who haddifficulties finding other kinds of work. However, this was notthe predominant arrangement among Brazilian women whoemigrated in the late 1980s and early 1990s. This is not to saythat this strategy of living at work does not take place, butrather that it is less frequent, and doesmake a difference in thethese women's trajectories.

Although living at work has advantages, as mentionedabove, the women are also subjected to greater laborexploitation because limitations to their working hours arenot respected and the pay is much lower than for those whodo independent cleaning jobs. In addition, the limits betweenworking hours and times of rest are not clearly laid out(Glenn, 1986). Nevertheless, the greatest problem, and thegreatest difference to Brazilians who have immigrated inmore recent decades is that living at work denies thesewomen a regular social life and interactions with friends andrelatives. Instead, they often wind up assuming a relationshipof loyalty and dependence towards their employers.

The above-mentioned trajectories of the immigrants in the1960s corroborate Hondagneu-Sotelo's (1994) reports ofyoung single Mexicans with an “illegal” status, which affirmedthat this type of work arrangement offers less risk of exposingthe women as undocumented workers. Despite the disadvan-tages, according to those interviewed, living at work serves asprotection for recently arrived women, preserving themmorally and preventing them from being “pushed” into sexwork.6 Finally, the author highlights that after some time,these women transitioned to daily work and began participat-ing in the regular life of their migrant community.

Living at work was therefore employed as a strategy, tofacilitate one's migratory project and guarantee a certainamount of security in the early days of their migration. Aftersome time, they were able to become legalized throughmarriage with naturalized foreigners, and could free them-selves from this type of arrangement.

Brazilian emigrants in the 1980s

In the 1990s, the flow of Brazilians became continuous,and housecleaning became an occupational niche of Brazilianemigrants. Here, the stories of Marcella Lanza, who emigrat-ed in 1988, of Luisa and Roberto Ramella,7 who emigrated in1998, and of Cláudio and Leila,8 will help us understand theemergence of this labor niche for Brazilian emigrants in theBoston region.

According to Martes (2000), “the Brazilian immigrantsentered the cleaning sector to create advantages over other

migrant groups.” Nevertheless, it is interesting to note that asMartes (2000), Fleischer (2002), Melo (2003) and Assis(2004) have observed, it is not the male Brazilian migrantswho “dominate housecleaning” but the women migrantsmade cleaning “businesses.” In this sense, the authors citedshow the configuration of a labor niche for Brazilianemigrants. The women who migrated to the United Statesfound cleaning to be a type of work that could guarantee theywould “make it in America.” When men work in houseclean-ing, they are usually subordinate to women. The women arethe “bosses” who make the decisions about what to clean,about the service and pricing, and the men must obey. Thisarrangement is usually conducted by couples, as alsoobserved by Martes (2000) who studied housecleaning andaffirmed that men migrate to work in this service in order toset up a business for the couple.

What I would like to call attention to here is that inaddition to the different involvement of women and men, thetasks are also categorized by gender. The workplace (theoffice or factory), which is represented as a place of formalemployment, becomes masculine—it is not exclusive to men,because women also work in offices, but it is charged withattributes of the male gender: competitiveness, efficiency,rigid norms, and contracts. Domestic work, which is “done forlove”—and in which the employee is treated as someone inthe family, receives feminine gender attributes: cordiality,care, informality, and the absence of clear rules.

In this sense, the fact that Brazilian women are concen-trated in cleaning, as are other Latin American migrants,reveals something about gender relations, but also about theinterweaving of class and ethnicity. The increased insertionof migrant Latin American women in U.S. homes signified agradual substitution of African–American women for womenfrom Latin America, the Caribbean and Asia, etc. Thesewomen from less developed places in the world cannotonly be contracted for less money and are willing to worklonger and harder, but, due to the precarious economic andsocial conditions, are also more docile (Rollins, 1990).

As Parreñas (2000) observed when analyzing Philippinemigrants inserted into domestic work around the world, theglobalization of economic markets has extended the policiesof reproductive labor to an international level. Parrenas usesthe term “the international transfer of caretaking services” torefer to three levels at which women from the countries ofdeparture and arrival are interconnected in the division oflabor. According to the author, while women of the privilegedclasses pay lowwages to Philippine maids, Philippine womenpay the same low salaries to employees who care for theirhouseholds. Hochschild (2000) calls this the global carechain, which is “a series of personal links between peopleacross the globe based on the paid or unpaid work of caring.”In the case of immigrant Brazilian women inserted intohousecleaning, many of them had their own housecleaners inBrazil, because they were part of the Brazilian middle class,which is also part of this international care network. At thesame time, these women in migration have other women—generally maternal aunts or grandmothers—care for theirfamilies and children in Brazil.

Brazilian women are inserted into a market segmented bygender, class and ethnicity, working for U.S. families in whichthe women work and pay for other women to perform

Page 41: Womens Studies International Forum 46 (2014)

39G. de Oliveira Assis / Women's Studies International Forum 46 (2014) 33–44

domestic services. For recently arrived migrants, this work,which does not require much knowledge of the language orspecific abilities, is a good opportunity, because, in additionto everything else, it pays well. They charge about US$50 perhouse, which takes about two hours to clean. The desire ofmany immigrants is to work within a market that, in additionto being considered well paid, is also autonomous.

It is interesting to note that housecleaning acquires thecharacter of a business, since it is considered well paid, thusendowing women autonomy and prestige in the Braziliancommunity, a status which is very different from houseclean-ing in Brazil. This distinction in relation to housecleaning inBrazil is emphasized not only by the couples Roberto andLuisa, and Cláudio and Leila, but also by Marcella, who workswith helpers, who are generally recently arrived migrants.

The immigrants make a point of emphasizing thatcleaning in the United States is not as heavy as it is in Brazil.They say that the cleaning products are better, U.S. homesmore practical and that it is not necessary to throw water oneverything like they do in Brazil. Nevertheless, they have toclean kitchens and bathrooms, which is why I consider theserepresentations as attempts to resignify the character of hardwork and emphasize the gains in relation to the parametersof cleaning work in Brazil.

The work opportunities can be shaped by gender, asaffirmed by Hondagneu-Sotelo (1994), but in this and othercases in which men clean with women, the activities areredefined. The men must learn to execute tasks dictated bywomen who understand and dominate these services. AsRoberto reports: “Luisa is the boss. She speaks with the client,who only calls her [he said smiling], she understands more.”

The tasks are divided as follows: those that are consideredheavier go to Roberto: he does the vacuuming, dusts andhelps change the bed sheets. When there is laundry to put inthe machine, he does that as well. Luisa cleans the bathroomsand the kitchen, and mops the floors. The two work veryquickly and do the cleaning in two hours. Luisa is also the onewho reads messages when the owners leave instructions,arranges the service with them or listens to any complaintsand, for this reason, she is considered the boss. This samedivision of tasks was observed in the couple Cláudio andLeila. When the cleaning is done by two women, in general,the owner of the business does the lighter tasks, such ascleaning the bedrooms, the kitchen and the bathroom.

Luisa decides where to begin and what to do, and evenwhere they will skimp on the service that day. Robertomigrated to clean houses and had to learn to perform work hedid not do in Brazil. It is interesting to note that in this context,domestic work in the gender attribution becomes moredifferentiated, as vacuuming is then considered a male task,as is doing laundry. Even dusting, which is considered moredelicate work and therefore feminine, is taught to men, whohave to learn to be careful with fragile items. Thus, the tasksand the utensils are classified according to the relations themigrants establish with them, and the masculine and femininegender attributes are presented with new semantics.

In exchange, in the case of the couples analyzed, it is themen who are responsible for driving the car. Luisa andRoberto live in Lowell, a city close to Boston. They leave atabout 7 a.m. to do a cleaning job in Cambridge, then one inNewton, another in Lawrence, and then they drive 30 miles

to Boston. They then have to go back to Lowell where theylive and have “part-time work”; they say that the houses inthe city they live in are all taken by other migrants. Thus, theyspend work hours traveling from one city to another.

In general, they do not complain about their bosses,whom they refer to as “clients,” but, when I accompaniedLuisa and Roberto on a job, I realized some tension when oneof the U.S. bosses complained about the “vacuuming” donethe week before. The complaints were made to Luisa becauseRoberto doesn't speak English. The emigrants avoidedcomplaining about their bosses, but when I accompaniedthem on cleaning jobs during the field work, I realized thatthey did not like the dirtier homes and the more demandingbosses, and thought better of those who gave tips andpresents—good clothes and domestic utensils they no longerneeded. Leila showed me coats that she got from one of herclients. She said: “They think that we are poor, so they treatus like we treat our maids in Brazil, they give us what theydon't use anymore.”

When they get home, these men and women still have tocare for their own houses. Here, the men also share thedomestic tasks, which most men in Brazil do not do. Thus,cleaning and its earnings cause men to recognize domestictasks as work, which was invisible to them before. Therefore,they are more apt to help their wives because they realizethat this work is indeed tiring. I observed this among the menwho I accompanied.

In the case of the couples interviewed, the fact that theywork together and share domestic tasks at work helps them toperceive the difficulty of the domestic work in their ownhomes. Thus, when they arrive at home, Roberto, who did nothelp with household chores before they migrated, helpedprepare dinner, take clothes to be washed, and clean the house.The same can be said of Cláudio and Leila, who decide togetherwhat to do with earnings, which, in addition to paying theirdebts in Brazil, also allowed them to send their children to theuniversity (one of them attended a public university andanother a private university). The money was also used to helpclose family members in Brazil to pay for their studies, or toassist relatives who wanted to migrate. Decisions aboutinvestments, paying debts and how to help family memberswere taken together. These practices observed in the daily lifeof the couples' relationships revealed more egalitarian and lesshierarchical relations, which point to changes in genderrelations in the context of migration. This is not to say that itis possible to generalize the situation among immigrants,because in many cases there are conflicts and separationsbetween migrant couples who do not negotiate these newsituations, it does however indicate another form of looking attasks previously considered “women's work.” Another factorfrequently discussed in relation to cleaning is its businesscharacter, or one immigrant “selling” a route or “schedule” forhousecleaning to another. Studies about Brazilian immigrantsin Boston show that when they refer to housecleaning and themarket niche that Brazilians are creating, they address howdifficult it is to set up a schedule and the different ways ofacquiring it, highlighting a form that is considered “unchristian”within the community: “selling” ones cleaning route orschedule.

The report below discusses the context in which this saletakes place and how the people who experience this process

Page 42: Womens Studies International Forum 46 (2014)

40 G. de Oliveira Assis / Women's Studies International Forum 46 (2014) 33–44

analyze it. Here, I turn to statements by Leila, who, togetherwith Cláudio, purchased a schedule and, after working forthree and a half years, decided to return to Brazil once theyhad reached their goals, and sold the schedule to theirdaughter Vera. When I was in the United States, Vera was stilladjusting to the business with a friend of hers from Criciumawho had become her partner. They were both inspired andthere was no tension or complaints from the daughter aboutthe fact that her parents sold the business to her. Vera evenleft a job at a U.S. company where she had been working in anoffice (unlike her parents, she spoke English well) to take onthe cleaning business.

According to Leila, they sold the schedule for US$24,000.00,which is three times what they earned in one month of work.Leila justified the price by affirming that when they took overthe schedule it was worth US$ 4000 per month and when theyleft it, it was paying about US$8000 per month. From theirperspective, they had dedicated themselves to the business,and had invested not only in the work, but also in relationswith the bosses (they continued to receive news and postcardsfrom their American bosses, even though they had been backin Brazil for more than a year). They expanded the business,and due to this investment,9 were cleaning houses that weremuch better than the ones they received. Thus, when Leila'smother questioned why she had sold to her own daughter, sheshowed her mother what the cleaning represented in reals,how much they earned when they began and when theydecided to sell. After this explanation, Leila's mother under-stood the decision. In addition, there was no conflict or dissentbetween the parents and daughter, because she had alreadybeen getting help in various moments of her stay in the UnitedStates.

Therefore, from the couple's perspective, they were givingtheir daughter an opportunity: a “business” that was builtthrough lots of work and, if they had simply given it to theirdaughter, “she would not recognize its value.” For this reason,they charged the price they thought was fair. The daughterpurchased the schedule together with a friend and began towork. At first, they lost some homes, and according to hermother, it was because they did not do a good cleaning job.The daughter then decided to buy the schedule from thefriend, because she was not doing her work well, and thistime the parents helped out and loaned her money. Veracontinues to clean most of the same houses that her parentsdid, and the bosses still send pictures of their children andpost cards to the parents who returned to Brazil.

This example is perhaps interesting because, in combina-tion with information from people who got cleaning routesfrom friends, it demonstrates the role work ethic playsamong immigrants. This work ethic justifies the sale of workpoints, yet often, even with this logic, some people do notsell. This demonstrates that the cleaning business does notcontradict the notion of solidarity in migrant communities,but triggers different meanings in a single event, revealingthe ambiguities found in the constitution of a work nicheamong Brazilian immigrants in Boston.

Sales (1999a) demonstrated that Brazilian immigrantsconstruct themselves, and are portrayed in the U.S. press, as ahard-working people; and are thus welcome as workers. Inthe services that they provide (cleaning in general, house-cleaning and work with the elderly), they are considered

careful, polite and known for providing good services. Theimmigrants use this reputation to add value to the work theydo in the United States and establish a distinction in relationto other immigrant groups.

Therefore, although it is a business, as observed in studiesof Brazilian housecleaners by Martes (2000), Fleischer(2002), and Brazilian shoe shiners by Margolis (1994), andinvolves competition, it does not exclude reciprocity, andthere is substantive gain for those who purchase a schedule—the trust of future bosses, the work in homes and theopportunity to have their own business. The sellers believethat they are offering solidarity because they select who willbuy the schedule. In the reports I collected, I found that theydo not sell to just anyone, a circuit is triggered that involvesnetworks of friendship and a common origin, and the sellers,although they earn a considerable profit from the deal(which, in the immigrant community, is at times a source ofcomplaints and accusations of lack of solidarity), they stillbelieve that they are acting in solidarity with those who seeka schedule.

This, of course, does not mean that there are no conflicts,complaints or deceptions within this exchange. For the buyers,there is no guarantee that the house-owners will like the workof the “friend” and staywith the service (the buyers are alwayspresented as friends, with the exception of cases of relatives, asseen above). For those who buy the cleaning “points,” there isalso the risk that the people who sold the point decide toreturn from Brazil and try to get the houses back, as reportedbyMartes (2000) and Fleischer (2002). Conflicts or deceptionsconcerning the schedule, however, are merely a possibleoutcome, and are not necessarily the rule within the“business.” In the cases reported above, those involved didnot experience it in this way. The elements are frequently amix, alternating solidarity and conflict in a “business” thatrepresents one of the few opportunities to earn money as anundocumented migrant with little knowledge of English, andfor this reason, it is a source of great contention.

Housecleaning jobs are thus shared, negotiated, donatedand contested. In this migration context, work that isconsidered a subaltern service with feminine attributes inBrazil redefines gender roles and the positions of men andwomen when they work together. It is not very common formen and women to work together because, as the couplesinterviewed emphasized, a couple has to be in real harmony todo so. New gender attributes are established for domesticactivities, some tasks become more masculine and othersmore feminine. The women clean the bedrooms and at timesthe men help make the bed; cleaning the bathroom and thekitchen also usually stays with the women. The men take theclothes that need to be washed, do the vacuuming, which isconsidered a heavy service, but at the same time they mustlearn to dust without breaking things. The women teach themen the tasks and inspect to see if everything is cleanedaccording to the established standards of cleanliness.

The subaltern status of housecleaning is resignified amongBrazilian immigrants in the United States, because the womenare seen as business owners who earn well, have a car andcontribute substantially to the family budget. They can travelto Brazil or bring their relatives to the U.S. for a visit. Thesegains compensate for the manual nature of the work, whichis heavy and attributed a low status. This aspect was also

Page 43: Womens Studies International Forum 46 (2014)

41G. de Oliveira Assis / Women's Studies International Forum 46 (2014) 33–44

observed by Martes (2000) and Fleischer (2002). Thus, in lightof the conceptions of many immigrants cleaning in Boston, wecan say that housecleaning in the United States “empowers”Brazilian women to enter this competitive business and raisetheir status in work and employment, contrary to the findingsof the authors mentioned in the beginning of this text, whohad analyzed domestic employment among Latin and Asianimmigrants.

The experiences of the immigrants are similar to thosefound by Carpenedo and Nardi (2013) in their analysis ofBrazilian women in domestic service working in Paris.According to these authors, despite the fact that thesewomen are undocumented and subject to precarious work-ing conditions, they are involved in more than simplyoppressive, discriminatory and exploitative regimes thatcharacterize transnational reproductive work. To a certaindegree, they are able to create spaces of resistance, which, insome cases, can signify social repositioning in their context oforigin, because migration helps to interrupt the reproductionof the cycle of poverty in the trajectories of their own livesand those of their families. It is important to observe that thewomen feel empowered, but among the men who clean, I didnot observe this sentiment. For this reason, housecleaning inthe United States becomes a project for women. Although attimes they come home at night very tired and with sorebacks, allergies from the cleaning products, and feelingdepressed about performing tasks they paid others to do inBrazil, the guarantee of continuous income compensates forthe loss of status.

To help disguise this demotion, Brazilian immigrants calltheir bosses “clients” or “customers,” which seems to be anattempt to escape the view that they themselves have ofdomestic work and of housecleaning in Brazil. Housecleaninggains another meaning, instead of being devalued femininework that is not recognized as employment, it becomes abusiness that helps men and women realize the project of“making it in America.” Roberto and Luisa, Cláudio and Leila,as well as Marcella, used the money they earned fromcleaning to buy real estate in Brazil, through which they arematerially concretizing their migratory success and positivelyredefining a type of work that has low status and pay inBrazil. The local materialization of the gains from working inthe United States, leads other emigrants in Brazil to dream of“making it in America.”

Finally, the experiences of these women immigrants help usconsider how they not only resignify the place of domestic work,but also representations of “Brazilianwomen” in their emotionalrelationships as demonstrated in Piscitelli, Assis and Olivar(2011). In the case of Brazilian women in Europe, particularlyin Portugal, is an example of this process. Here, the representa-tions of “Brazilian women” produce an association betweengender and nationality. These representations place certainessentialized and exoticized characteristics – sensuality, joyful-ness, congeniality – in relation with the insertion into the sexmarket, which leads to the discrimination of to Brazilianimmigrants in Portugal (Padilla, 2007; Pontes, 2004). In thecase of the Brazilian immigrants interviewed, categorizationsthat articulate gender and sensuality do not produce the sameeffects. The image of sensuality is aggregated to the represen-tation of a caring woman, in contrast to Brazilian men who arenot represented as good partners, because they are seen as

chauvinist, commanding, unwilling to share domestic tasks,and thus as exhibiting models of masculinity that are notvalued in the migration context.

This is the case of Marcella Lanza, who had been in theUnited States for fourteen years when she was interviewed.She said that after various romantic relationships, she hadchanged her expectations about marriage and the idea ofreturning to Brazil. Marcella sought to construct otherromantic relationships in which she could find emotionalfulfillment and well-being, but also security regarding herimmigration status. As being undocumented became moredifficult following the attacks of September 11, she began tolook for a U.S. boyfriend. That's how Marcella met James, anAmerican, with whom she began a relationship and whomshe tried to win over by using attributes of Brazilianfemininity.

What did Marcella like about James? Firstly, he was not ajealous man and respected her work, her Brazilian friends,and her leisure time with them. In comparison, sheconsidered the masculine characteristics of American mento be positive attributes. Unlike the Brazilian men from herpast relationships, James gave her the space she felt sheneeded to live her life. On the other hand, Marcella did whatshe considered important for the relationship and for herboyfriend. According to her, this is something Brazilianwomen do well, even better than American women, forexample, offering a good meal, going out at times to talk withhis friends and maintaining a good sexual–romantic rela-tionship. Marcella constructed James' male attributes aspositive in comparison to her previous boyfriends, andhighlighted the security he provided her. In 2003, Marcellabecame pregnant from James. The pregnancy made her veryhappy because in her account, she could now begin her ownfamily. They got married on Valentine's Day in a civilceremony, and were still married when I met them in Bostonin 2008. The marriage to attain “papers” also represented astable relationship, based on respect and security, combiningaffection and interests, as observed by Piscitelli's (2011)analysis of relationships between Brazilians and their part-ners in Spain.

Marcella's account reveals a recurring sentiment amongthe Brazilian women I spoke with. The sense of security,autonomy, driving one's own car, and running one's ownbusiness made these women feel more autonomous, free tomake choices regarding their work, leisure time and romanticlife. It is important to emphasize that although Marcellamarried James, regularized her status by receiving a GreenCard, she continued her own business called “Marcella'scleaning service.”

Final considerations

Classic migration studies describe women as merelyaccompanying or waiting for their husbands or sons, andthe importance of their contributions to family income wasnot taken into consideration. Therefore, such analyses notonly often hid women's participation, but they also failed torecognize that long-distance migration takes place within acomplex network of social relations, in which women play animportant role.

Page 44: Womens Studies International Forum 46 (2014)

42 G. de Oliveira Assis / Women's Studies International Forum 46 (2014) 33–44

Thus, although women have predominated in legalmigration to the United States since the 1930s, thisperspective remained outside the scope of scholarly researchand, as a consequence, important aspects of migration andthe establishment of immigrants remained blind spots, aswere the differences among immigrants in terms of class,gender and ethnicity. Thus, gender-differentiated reasons forand characteristics of a mobility remained outside theresearch focus: migrant subjects were assumed to be maleand there was no visibility of female participation. However,since the mid-1970s gender and feminist studies havequestioned the crystallized images of women immigrants asthose who wait, or as passive dependents. This interrogationhas led migratory studies to cast another glance at themigratory process and to question theoretical presumptionsabout migration.

The article sought to demonstrate that Brazilian migrantwomen are agents in the migratory process and do notsimply migrate to accompany their husbands, children orrelatives. In this process, although they are inserted into asecondary labor market through working as housecleaners,they construct cleaning work and its significance in adifferent way that sets it apart from domestic work in Brazil.Thus, although they work every day of the week, doing heavywork, the fact that they have a schedule of houses where theywork by the hour, which pay well in comparison to what ahousecleaner earns in Brazil, and the fact that they transformthis work into a “business” resignifies domestic work. Thisprocess allows many of these women to become moreautonomous, and to become active agents in migration inthe United States. These portrayals do not seek to present thecomplete spectrum of Brazilian immigrant experiences, butrather to demonstrate how domestic cleaning – although ithas a subaltern character in Brazil – allows the womeninterviewed to realize their migratory projects and contrib-ute to redefining gender positions. By transforming cleaninginto an informal business, these women resignify this workthereby evading the creation of a link between their workand its subaltern character. Although the sale of cleaningschedules apparently goes against notions of solidarity and iscriticized by some immigrant community and religiousleaders, it is also presented as “proof” of the entrepreneurialcapacity of the Brazilian immigrants who construct anspecific labor niche for immigrant women. In the view ofthese women, the subaltern context of domestic work ispresented in a new semantic in comparison with theparameters relating to the same type of work in Brazil.

In the case of the Brazilian immigrants interviewed, theyall emphasize the sensation of greater autonomy; “Thewomen here have power and are more respected,” somesay. Nevertheless, the difficulties with legalization experi-enced by the majority demonstrate that this process ofagency does not take place in the same way for everyone orwith the same intensity. Therefore, unlike the Chineseenclaves in Zhou's study (1992) and Zhou and Logan(1989), in which the networks of their compatriots offeredwomen fewer economic advantages than the men, theBrazilian women interviewed appear to have establishednetworks offering entrance into and assistance within thelabor market, particularly in the informal cleaning business,through which they obtain work opportunities and economic

advantages, which, in some cases were better than the jobsavailable to men.

By incorporating the category “gender” into the analysisof migratory flows, migration is no longer seen as a rationalchoice of isolated individuals, but emerges as a processinvolving networks of social relations, which include strate-gies of groups of family members, friends or people from thesame community. In this context, women and men emerge atdifferent times– as the links connecting social networks – inBrazil and abroad—that help out in the days of migration inthe society of destination and with the maintenance of tieswith the place of origin.

These social networks are informed by kinship and gendernorms. Thus, women turn much more help from theirrelatives and those who articulate the networks amongother households. Men also find support within thesenetworks, but the information gathered indicates that theyrely much more on the help of friends to find work andhousing than of relatives. Within this process genderrelations are redefined: in general, women experiencegreater autonomy and agency in the destination society, notonly because they earn better, despite conducting work oflow status, but also because Brazilian femininity is attributeda specific value on the U.S. matrimonial market, creatingpossibilities for romantic relationships, which may lead tolegalization—as this is difficult for women to obtain solelythrough their work in domestic service. In this way, womennegotiate the attributes of Brazilianity and mobilize them toaffirm themselves positively in the United States, as demon-strated in Marcella's statements. Male immigrants, on theother hand, feel a loss of status even more, because they areasked to share their authority, negotiate it, and areconfronted with expectations that call for more egalitarianrelations. As Adriana Piscitelli (2011) observed when ana-lyzing Brazilians working in the sex market in Europe, thenotion of agency can only be understood when economic,political and social factors are also taken into consideration.Thus, Brazilian women who set up a business and supportthemselves in the United States discover that they arecapable and have autonomy, which gives them a sense ofpower and recognition in the society of emigration, andespecially in their cities of origin, where they are able to sendpresents and make investments.

Long-distance migration brings about great transforma-tions for the subjects who live this experience; instead ofthinking of it only as a factor that leads to ties being broken, Isought to conduct a more complex analysis and to demon-strate the constitution of a field of transnationalized relationsthat also allows new family and gender arrangements.

Therefore, in contemporary flows, women tend to migratealone or as the first in their families, being pioneers in findingwork in the United States, breaking with the image of womenas those who wait or follow in the men's footsteps. Finally,women become integrated into networks of migration andare key agents through their articulations of the connectionsbetween “here” and “there.”

Endnotes

1 According to Sales (1992:60), emigration would be the most bitterfruit of our “lost decade” – the denomination economists gave to the 1980s

Page 45: Womens Studies International Forum 46 (2014)

43G. de Oliveira Assis / Women's Studies International Forum 46 (2014) 33–44

due to the worsening economic indicators – because we entered the worldeconomy through the back door, providing illegal immigrant workers whoescaped the economic crisis.

2 Data from IBGE. Available at: http://saladeimprensa.ibge.gov.br/noticias?view=noticia&id=1&busca=1&idnoticia=2017.

3 Marcella Lanza 41 years old, high school graduate, interviewed inJanuary 2002, in the Boston region.

4 For a discussion about insertion in the U.S. labor market see: Sales(1999a, 1999b), Scudeler (1999), Martes (2000).

5 “to live at work” means to live at the employer's house. Thisarrangement was considered safer by those interviewed in the 1960s,mainly by young and single immigrant women, because they live in thehouse of the family.

6 Work that is quite stigmatized in the migrant group. The same type ofstigma is observed among Brazilian immigrants in relation to women whowork as go-go girls (Assis, 1995; Margolis, 1994).

7 Roberto and Luisa Ramella are from Criciúma and granted me aninterview in January 2002, in a city close to Boston.

8 Claudio and Leila, are from Criciúma and granted me an interview inFeburary 2002. They were friends of the couple Roberto and Luisa Ramella.

9 Fleischer (2002, p. 121) also highlighted this aspect of compensationthat the immigrants attributed to the sale of the cleaning business and in thiscontext relatives also treat the exchange as a business deal, but get specialpayment conditions.

References

Anthias, Floya (2000). Metaphors of home: Gendering new migrations insouthern Europe. In Floya Anthias, & Gabriela Lazaridis (Eds.),Gender and migration in Southern Europe (pp. 17–47). Oxford, NewYork: Berg.

Assis, Gláucia de O. (1995). Estar aqui… estar lá… uma cartografia da vidaentre dois lugares. Dissertação (Mestrado em Antropologia Social).Florianópolis, SC: Programa de Pós-Graduação em Antropologia Social,UFSC.

Assis, Gláucia de O. (1999). Estar aqui… estar lá… uma cartografia daemigração valadarense para os EUA. In Rossana R. Reis, & Teresa Sales(Eds.), Cenas do Brasil migrante (pp. 125–167). São Paulo: Boitempo.

Assis, Gláucia de O. (2004). De Criciúma para o mundo: os novos fluxos dapopulação brasileira e os rearranjos familiares e de gênero. Tese(Doutorado em Ciências Sociais). Programa de Pós-Graduação emCiências Sociais. Campinas, SP: Instituto de Filosofia e Ciências Humanas,Unicamp.

Assis, Gláucia de Oliveira (2007). Mulheres migrantes no passado e nopresente: gênero, redes sociais e migração internacional. Revista.Estudos. Feministas. (Florianópolis), 15(3), 745–772 (online).

Assis, Gláucia de O. (2011). De Criciúma para o mundo: rearranjos familiaresdos novos migrantes brasileiros. Florianópolis: Editora Mulheres.

Assis, Gláucia de O., & Siqueira, Sueli (2009). Mulheres emigrantes e aconfiguração de redes sociais: construindo conexões entre o Brasil e osEstados Unidos. Revista Interdisciplinar da Mobilidade Humana - REMHU(Brasília), 16, 25–46.

Assis, Gláucia de O (2003). De Criciúma para o mundo os novos fluxos dapopulação brasileira: gênero e rearranjos familiares. Martes, Ana Cristina B.e Fleischer, Soraya (Org.). Fronteiras Cruzadas: etnicidade, gênero e redessociais, (pp. 199–230). Paz e Terra: São Paulo.

Brites, Jurema (2013, enero-junio). Trabajo doméstico en Brasil: transformacionesy continuidades de La precariedad. Trayectorias, 15(36), 3–19.

Brites, Jurema, Tizziani, Ania, & Gorban, Débora (2013). Trabajo domésticoremunerado: espacios y desafíos de la visibilidad social. Revista deEstudios Sociales, 45, 226–228 (online).

Carpenedo, Manoela, & Nardi, Henrique Caetano (2013). Mulheres Brasileiras nadivisão internacional do trabalho reprodutivo: construindo subjetividade(s).Revista de Estudios Sociales, 45, 96–109 (online).

Chant, Sylvia, & Radcliffe, Sarah (1992). Migration and development: Theimportance of gender. In Sylvia Chant (Ed.), Gender and migration indeveloping countries (pp. 1–29). London and New York: BelhavenPress.

Debiaggi, Sylvia D. (2002). Changing gender roles: Brazilian immigrant familiesin the U.S. New York: LFB Scholarly Publishing, LLC.

Debiaggi, Sylvia D. (2003). Famílias brasileiras em um novo contextocultural. In Ana Cristina B. Martes, & Soraya Fleischer (Eds.), FronteirasCruzadas: etnicidade, gênero e redes sociais (pp. 175–199). São Paulo: Paze Terra.

Fleischer, Soraya R. (2002). Passando a América a limpo: o trabalho dehousecleaners brasileiras em Boston, Massachussets. São Paulo: Annablume.

Foner, Nancy (2000). From Ellis Island to JFK: New York's two waves ofimmigration. New York: Russell Sage Foundation.

Fusco, Wilson (2001). Redes sociais na migração internacional: o caso deGovernador Valadares. Campinas: Núcleo de Estudos de População,UNICAMP.

Fusco, Wilson (2005, jul./set). A formação de comunidades-filhas no fluxo debrasileiros para os Estados Unidos. São Paulo em Perspectiva, 19(3),55–63.

Glenn, Evelyn N. (1986). Issei, Nisei War bride: three generations of JapaneseAmerican women in domestic service. Philadelphia: Temple UniversityPress.

Goza, Franklin (1992, jan.jun). A imigração brasileira para a América doNorte. Revista Brasileira de Estudos de População, 9(1), 65–82.

Gramusck, Sherri, & Pessar, Patrícia (1991). Between two islands: Dominicaninternational migration. Berkeley, Los Angeles and London: CaliforniaUniversity Press.

Gregorio Gil, Carmen (1996). Sistemas de género y migracióninternacional. La emigración dominicana en la Comunidad de Madrid.Tese (Doctorado en Antropología Social). Madrid: UniversidadAutónoma de Madrid.

Gregorio Gil, Carmen (2007). Trabajando honestamente en casa de familia:entre la domesticidad y la hipersexualización. Revista Estudos Feministas,15(3), 699–716 (online).

Hagan, Jacqueline M. (1998). Social networks, gender and immigrantincorporation: Resources and constraints. American Sociological Review,63(1), 55–68.

Hochschild, Arlie Russell (2000). The nanny chain. American Prospect, 11(4),32–36.

Hondagneu-Sotelo, Pierette (1994). Gendered transitions: Mexican experi-ences of immigration. Berkley: University of California Press.

Hondagneu-Sotelo, Pierette (2007). Doméstica: Immigrant workerscleaning & caring in the shadows of affluence. Berkley: University ofCalifornia Press.

Houstoun, Marion F., Kramer, Roger G., & Barret, Joan M. (1984). Femalepredominance of immigration to the United States since 1930: A firstlook. International Migration Review, 18(4), 908–963.

Kofes, Suely (2001). Mulher, Mulheres: identidade, diferença e desigualdade narelação entre patroas e empregadas domésticas. Campinas: Editora daUnicamp.

Levitt, Peggy (2001). The transnational villagers. Berkley and Los Angeles,California: University of California Press.

Maia, Susanas (2009). Sedução e Identidade Nacional: mulheres brasileirasque trabalham como dançarinas eróticas em NY. Revista EstudosFeministas, 17(3), 769–797 (set./dez).

Margolis, Maxime L. (1989). An new ingredient in the melting pot: Braziliansin New York City. City and Society, 3(2), 179–187.

Margolis, Maxime L. (1992). From mistress to servant: Downwardmobility among Brazilians in New York City. Urban Anthropology,19(3), 215–231.

Margolis, Maxime L. (1994). Little Brazil: An ethnografhy of Brazilianimmigrants in New York City. New Jersey: Princeton University Press.

Martes, Ana Cristina B. (2000). Brasileiros nos Estados Unidos: um estudosobre imigrantes em Massachusetts. São Paulo: Paz e Terra.

Mary, Romero (1997). Domestic service in the transition from rural to urbanlife: the case of la Chicana. Women’s Studies, 13(3), 199–222.

Massey, Douglas (1997). Migration, ethnic mobilization and globalization –

causes of migration. In Montserrat Guibernau, & John Rex (Eds.), TheEthnicity reader – nationalism, multiculturalism and migration(pp. 257–269). UK: Polity Press.

Massey, Douglas S., Alarcon, Rafael, Durand, Jorge, & Gonzalez, Humberto(1987). Return to Aztlan: The social process of international migration fromWestern Mexico. Berkeley: University of California Press.

Melo, Sonia (2003). Protagonistas de um Brasil imaginário: faxineiras em Boston.In Ana Cristina B. Martes, & Soraya R. Fleisher (Eds.), Fronteiras Cruzadas:etnicidade, família e redes sociais. São Paulo: Editora Paz e Terra.

Menjivar, Cecília (2000). Fragmented ties: Salvadoran immigrant networks inAmerica. Berkeley, Los Angeles, London: University California Press.

Morokvasic, Mirjana (1984). Birds of passage are also women. InternationalMigration Review, 18(4), 886–907.

Padilla, Beatriz (2007). A imigrante brasileira em Portugal: considerando ogénero na análise. In Jorge Malheiros (Ed.), A Imigração Brasileira emPortugal (pp. 113–134). : Colecção Comunidades.

Parreñas, Rhacel S. (2000, August). Migrant Filipina domestic workers andthe international division of reproductive. Gender & Society, 14(4),560–581.

Pessar, Patricia R. (1999). The role of gender, households, and socialnetworks in the migration process: A review and appraisal. In CharlesHirschman, Philip Kasinitz, & Joshua Dewind (Eds.), The handbook ofinternational migration: The American experience (pp. 51–70). New York:Russell Sage Foundation.

Page 46: Womens Studies International Forum 46 (2014)

44 G. de Oliveira Assis / Women's Studies International Forum 46 (2014) 33–44

Piscitelli, Adriana (2007). Gringos nos trópicos: gênero e nacionalidade no marcodo turismo sexual em Fortaleza. In Mary Garcia Castro (Ed.), Migraçõesinternacionais: contribuições para políticas (pp. 589–613). Brasília: CNPD.

Piscitelli, Adriana (2011). “Papéis”, interesse e afetos, relacionamentosamorosos sexuais e migração. In Arend, Silvia F., Rial, Carmen S., &Pedro, Joana M. (Eds.), Diásporas, Mobilidades e migrações (pp. 103–129).Florianópolis: Editora Mulheres.

Piscitelli, Adriana, Assis, Gláucia O., & Olivar, José Miguel N. (2011). Gênero,sexo, amor e dinheiro: mobilidades transnacionais envolvendo o Brasil.Diásporas, Mobilidades e migrações. Handbook of teratology, v.1, . .Campinas: Unicamp/Pagu.

Pisicitelli, Adriana (2008, jul/dez). Interseccionalidades, categorias dearticulação e experiências de migrantes brasileiras. Sociedade e Cultura,11(2), 263–274.

Pontes, Luciana (2004, julho-dezembro). As Mulheres Brasileiras na MídiaPortuguesa. Cadernos Pagu Núcleo de Estudos de Gênero – Pagu/Unicamp,23. (pp. 229–256), 229–256.

Reis Rossana, R., & Sales, Teresa (1999). Cenas do Brasil Migrante. São Paulo:Boitempo.

Ribeiro Gustavo, L. (1999). 0 que faz o Brasil, Brazil: jogos identitários emSan Francisco. In Rossana R. Reis, & Teresa Sales (Eds.), Cenas do BrasilMigrante (pp. 45–87). São Paulo: Boitempo.

Rollins, Judith (1990). Ideology and servitude. In: Colen, Shellee and SanjekRoger (Org.) At work in homes: household workers in world perspective.series 3 (pp. 77–88). American Ethnological society monograph,Washington: American Anthropology Association.

Sales, Teresa (1991, jan./dez.). Novos Fluxos da população brasileira. RevistaBrasileira de Estudos de População, 8(1/2) (São Paulo).

Sales, Teresa (1992, jan./jul). Imigrantes estrangeiros, imigrantesbrasileiros: uma revisão bibliográfica e algumas questões para

pesquisa. Revista Brasileira de Estudos de População, 9(1), 50–64 (SãoPaulo).

Sales, Teresa (1999a). Brasileiros longe de casa. São Paulo: Cortez.Sales, Teresa (1999b). Identidade étnica entre imigrantes brasileiros na

região de Boston. In Reis, Rossana R., & Sales, Teresa (Eds.), Cenas doBrasil migrante (pp. 17–44). São Paulo: Boitempo.

Sales, Teresa (1995). O trabalhador brasileiro no contexto das migraçõesinternacionais. In L. Patarra Neide (Ed.), (Coord.), Emigração eImigração internacionais no Brasil contemporâneo (pp. 89–101). SãoPaulo: FNUAP.

Sassen, Saskia (2003). Contrageografías da globalización: género ecidadania em los circuitos trasnfornteirizos. Madrid: Traficantes desueños, mapas.

Scudeler, Valéria (1999). Imigrantes valadarenses no mercado de trabalhonos EUA. In R. Rossana Reis, & Sales Teresa (Eds.), Cenas do BrasilMigrante. São Paulo: Boitempo.

Siqueira, Sueli (2006). Migrantes e empreendedorismo na Microrregião deGovernador Valadares – sonhos e frustrações no retorno. 200f. Tese(Doutorado em Ciências Humanas – Sociologia e Política). BeloHorizonte: Faculdade de filosofia e Ciências Humanas, UniversidadeFederal de Minas Gerais.

Soares, Weber (1995). Emigração e (i)mobilidade residencial: momentos deruptura e reprodução continuidade da segregação espacial no espaçourbano. In Reis, Rossana R., & Sales, Teresa (Eds.), Cenas do Brasilmigrante (pp. 175–179). São Paulo: Cenas do Brasil migrante.

Zhou, Min, & Logan, John (1989). Returns on human capital in ethnicenclaves: New York City's chinatown. American Sociological Review, 54,809–820.

Zhou, Min (1992). Chinatown: the socioeconomic potential of the urbanenclave. Philadelfhia: Temple University Press.

Page 47: Womens Studies International Forum 46 (2014)

Women's Studies International Forum 46 (2014) 45–53

Contents lists available at ScienceDirect

Women's Studies International Forum

j ourna l homepage: www.e lsev ie r .com/ locate /ws i f

Domestic work–affective labor: On feminization and thecoloniality of labor

Encarnacion Gutierrez-RodriguezJustus-Liebig University Giessen, Germany

a r t i c l e i n f o

http://dx.doi.org/10.1016/j.wsif.2014.03.0050277-5395/© 2014 Elsevier Ltd. All rights reserved.

s y n o p s i s

Available online 17 April 2014

This paper argues for an understanding of domestic work as affective labor. It engages with theaffective quality of reproductive labor by interrogating the organization of paid and unpaiddomestic work in private households. Thus, while it attends to debates on emotional labor, itsmain focus is on the affective dimension of the social.It does so by focusing on reproductive labor, in particular, domestic work and developing afeminist critique of affective labor through the analysis of the cultural predication of feelingsassociated with and infused in domestic work. In this regard, the cultural predicationprescribing the social meaning attached to domestic work will be explored within theframework of feminization and coloniality. Thus, domestic work will be discussed as affectivelabor surfacing at the juncture of feminization and coloniality. Following this argument, thearticle firstly engages with feminist analyses on reproductive labor, feminization and domesticwork. Secondly, it looks at private households and affective labor. Thirdly, it examines therelationship between paid domestic work and migration regimes from the angle of thecoloniality of labor. Using these insights, the article explores the sensorial corporeality ofracialized affect negotiated in and around domestic work. It concludes by arguing for aconceptualization of domestic work as affective labor.

© 2014 Elsevier Ltd. All rights reserved.

Introduction

This article's focus is on domestic work as affective labor.It engages with the affective quality of reproductive labor byinterrogating the organization of paid and unpaid domesticwork in private households. Thus, while it attends to debateson emotional labor (Carrington, 1999; Hochschild, 1983;Illouz, 2007), its main focus is on the affective dimension ofthe social. As such this article engages with the impact offeelings and emotions on social relationships and spaces(Ahmed, 2004; Brennan, 2004; Sedgwick, 2004). FollowingSpinoza's (1994) observation that affect drives us to act, thearticle explores the twofold character of affect as a texture ofthe social and as socially textured. It does so by focusing onreproductive labor, in particular, domestic work and devel-oping a feminist critique of affective labor through theanalysis of the cultural predication of feelings associated

with and infused in domestic work. It thus contributes to thedebate on affective labor in feminist theory (Corsani, 2007;Federici, 2012; Gutiérrez Rodríguez, 2010; Precarias a laDeriva, 2004; Weeks, 2011).

Engaging with the affective corporeality of domestic work,this article argues for an understanding of feelings andemotions as interlaced in the social semantics of place andtime. In this regard, the cultural predication prescribing thesocial meaning attached to domestic work will be exploredwithin the framework of feminization and coloniality. Thus,domestic work will be discussed as affective labor surfacing atthe juncture of feminization and coloniality. Both processesdescribe social classification systems related to the creation of ahierarchical social order. In order to illustrate this ratherabstract yet material dimension of corporeal affectivity indomestic work, the article uses interview extracts from a studyconducted with colleagues on the interpersonal relationships

Page 48: Womens Studies International Forum 46 (2014)

46 E. Gutierrez-Rodriguez / Women's Studies International Forum 46 (2014) 45–53

between female migrant domestic workers and their femaleemployers in Austria, Germany, Spain and the United Kingdombetween 2002 and 20041 (Caixeta, Gutiérrez Rodríguez, Tate, &Vega Solís, 2004), other observations from research onundocumented Latin American domestic workers employedin private households in Germany and the United Kingdomconducted by the author between 2007 and 2013 are alsoconsidered (Gutiérrez Rodríguez, 2010).

The discussion engages firstly with feminist analyses onreproductive labor, feminization and domestic work (Caixeta etal., 2004; Corsani, 2007; Federici, 2012; Precarias a la Deriva,2004). Secondly, it moves to look at private households andaffective labor. Thirdly, it explores the relationship betweenpaid domestic work and migration regimes from the angle ofthe coloniality of labor (Quijano, 2000, 2005, 2008). Using theseinsights, the article explores the sensorial corporeality ofracialized affect negotiated in and around domestic work. Itconcludes by arguing for a conceptualization of domestic workas affective labor. First, let us begin with the debate onreproductive labor, feminization and domestic work.

Reproductive labor, feminization and domestic work

According to the ILO, majority of domestic workers arewomen (82%), many of whom are migrants or children whose“work is undervalued, underpaid, [and] poorly regulated”(ETUC, 2012: 10). These characteristics resonate with features,which feminist activists and scholars have discussed asassociated with the feminization of labor (Bair, 2010; Bakker,2007; Elson, 1998). Domestic work epitomizes the socialdevaluation of feminized labor (Mies, 1999). This is articulatedeconomically as the productive contribution of domestic workis consistently ignored in official calculations of GDP (cf. Ferber& Nelson, 1993; Folbre, 1994; Hewitson, 1999; Himmelweit,1995; Pérez Orozco, 2004, 2010; Waring, 2004). It is alsoarticulated socially as domestic work continues to be perceivedas unproductive and unskilled labor, devoid of any societalvalue (Gutiérrez Rodríguez, 2010; Weeks, 2011).

Feminist theory has challenged this perception (cf. Barrett,1980; Dalla Costa & James, 1972; Delphy, 1984) and insisted onthe constitutive value of domestic work for social reproduction(Bakker & Gill, 2003; Barker & Feiner, 2010; Bedford & Rai,2010; Benería, 1979; Dalla Costa & James, 1972; Federici, 2004;Kofman, 2012; Molyneux, 1979; Peterson, 2009). This calls intoquestion Marxist views that restrict this labor merely tothe sphere of reproduction by underscoring its productiveforce (cf. Jacobs, 2010; O'Hara, 1998; Redclift, 1985). Morerecently, feminist research has highlighted the emotionalcharacter of domestic work (cf. Boris & Parreñas, 2010;Carrington, 1999; Hochschild, 1983, 2003; Lan, 2006).

Taking these observations on board and considering thetransformation of the organization of labor in post-industrialsocieties, feminist theorists and activists in Spain and Italy haveplaced a renewed focus on the question of reproduction(Benería & Sarasúa, 2011; Corsani, 2007; del Río, 2004;Fantone, 2007, 2011; Federici, 2006; Pérez Orozco, 2004;Precarias a la Deriva, 2004; Ruido, 2008; Sconvegno, 2007;Vega Solís, 2009). In doing so, they consider care work(Spanish: trabajo de cuidados) in particular as a pivotal axisfor organizing precariouswork. For example, theMadrid-basedfeminist group Precarias a la Deriva has drawn attention to the

significance of care work for social reproduction by focusing onpersonal caring activities and re-evaluating the ethical impli-cations of care for society (Precarias a la Deriva, 2004). Thus,Precarias has complicated the Marxist division of productiveand reproductive labor. Introducing care work as a hybridcategory, Precarias defines care work as a hinge betweenreproductive and productive labor. Care work articulates theincreasing interpenetration of these spheres in post-industrialsocieties, a tendency that they coined “the feminization ofprecarity” (Precarias a la Deriva, 2004). In a similar vein, otherfeminist analyses of the impact of the economic crisis in Spainand Italy suggest that we depart from taking the feminizationand the precarization of labor as vantage points from which tounderstand crisis capitalism (Benería & Sarasúa, 2011;Carraquer Oto, 2013; Carrasco Bengoa, 2013; Federici, 2012;Martín Palomo, 2008, 2013). Acknowledging that feminizationdoes not simply refer to the quantitative dimension of thegendered division of work, that is, to the overrepresentation ofwomen within low-income and insecure work sectors, thisdebate has drawn attention to the historical and culturalimplications of feminization as a process of labor devaluation.Thus, feminization connotes the cultural predication of workhistorically delivered by feminized subjects as “inferior.”

While the feminist analysis of crisis capitalism empha-sizes the relevance of reproductive labor through the lens ofcare work, some feminist research warns us not to subsumereproductive labor under the umbrella term “care work.”“Care work” refers to a specific range of activities engagingwith direct or indirect personal care (Folbre, 2006) and toprofessional pathways such as nursing, child care or care ofthe elderly. In contrast, domestic work is not considered aprofession—with the exception of the “housekeeper” inGermany and Austria, which involves the management ofthe household and household workers. Subsuming domesticwork under the term “care” may obfuscate the “dirty work”of physical activities dealing with dirt (Anderson, 2000). Yet,as numerous studies have shown, in light of everydaypractices care workers very often need to deliver domesticwork and domestic workers are requested to do care work(c.f., Anderson, 2000; Hondagneu-Sotelo, 2001; Lan, 2006).As Silvia Federici (2012) notes, despite the interchangeabilitybetween domestic and care work, the assumption thatreproductive labor is care work and thus “affective labor”needs to be critically interrogated.

Federici stresses the historical conditions through whichreproductive labor has been imposed on women and becomea terrain of women's agency and struggle. She notes that thepractices developed in this field that have been passed onover generations represent the creation of common wisdomand collective knowledge acquired through experiences ofoppression and resistance (Federici, 2004). For Federici,subsuming reproductive labor under the label “affectivelabor” fails to acknowledge the persistence of a gendereddivision of work, whereby reproductive labor addresses aspecific quality of labor that is related to certain physicaltasks, personal and emotional skills. Thus, “the fast-foodfemale workers who must flip hamburgers at McDonald'swith a smile or the stewardesses who must sell a sense ofsecurity to the people she attends to” (Federici, 2012: 122)are not synonymous with the care workers who need tocomplete specific physical tasks and deploy emotional

Page 49: Womens Studies International Forum 46 (2014)

47E. Gutierrez-Rodriguez / Women's Studies International Forum 46 (2014) 45–53

faculties in caring for people. Considering “affective labor” as“a component of every form of work rather than a specificform of (re)production” (Federici, 2012: 122) blurs theprevalence of the gendered division of work and itsconstitutive role in supporting the cycle of capital productionand accumulation. For example, caring for a child, an elderlyperson, or a disabled person requires specific physical andemotional skills that are not interchangeable with the skillsexpected of teachers, office clerks, or flight attendants. Whilethe latter are required to deal with emotional demands atwork, they are not immediately involved in the physical careof a person, and their work is not explicitly defined by theparameters of feminization.

While this critique rightly warns of an overgeneralization ofterms that might obscure the intricacies of the gendereddivision of work and the substantial significance of feminizedlabor in supporting capital accumulation, the analysis of theaffective dimension entails more than a consideration of carework. The analysis of affect is not intrinsically related to carework, though care represents one of the axes of analysis.Rather, as previously mentioned, the focus on affect addressesthe affective fabric of our being as it highlights the socialtexture of our affective becoming. Social encounters andrelations of production and reproduction unfold in spaces ofaffective (dis-)encounters such as private households.

The private household and affective labor

Private households are saturated with people's feelings andemotions. While these feelings and emotions are individuallyexperienced as sensations, their affective character goesbeyond personal experience. The relational and spatial charac-ter of affect transcends the sphere of the personal as affectdefines the impact of feelings on bodies, objects and spaces(Ahmed, 2004; Brennan, 2004; Massumi, 2002; Tate, 2009). Assuch, affect addresses the impact of feelings circulating in adispersed manner, which are expressed in fleeting encounters,and have an impact on people's bodies and psyches. Theorientation towards a specific addressee is not always ratio-nally conceived in the expression, impression, and circulationof feelings; and, because they evolve within a social context,they become tangible and intelligible because they bear socialmeaning.

Within the private sphere of the household, domesticworkers are immersed in the immediate intimate relations ofthe household members. While not always explicitly part ofthese relationships, domestic workers unwillingly and un-wittingly become involved in them purely by inhabiting thespace. Thus, they not only cognitively participate in emo-tional work, that is, by attending to the caring needs of thehousehold members, they are also (in)directly addressed bythe emotions and feelings that circulate within the house-hold. Thinking about affect in private households points tohow feelings affect us and how we are affected by theenergies of others. In contrast to emotions that engage withthe cognitive dimension of feelings, affect are sensations orstimuli (Spinoza, 1994). These are driven by life forces suchas desire (cupiditas), joy (laetitia) and sadness (tristitia)(Spinoza, 1994: 160ff.). Affect motivates our thinking andactions as these energies, intensities and sensations compelus to act and transform passion into action.2 As such, affect is

a relational category, an outcome of encounters and energeticcircuits, which permeates our bodies, and results from ourability to feel.

In everyday encounters feelings are transmitted from oneperson to another, from one space to another. This transmis-sion of affect, called affection (Brennan, 2004), may increaseor diminish our energies. It can affect us in positive ornegative ways. Thus, we might feel relieved, enhanced ordepleted as a result of the impact that energy has had on us,leaving us with a positive or negative sensation, intensity orfeeling. For example, in the case of joy and love, one can feelenergized, but “when one carries the affective burden ofanother, either by a straightforward transfer or because theother's anger becomes your depression” (Brennan, 2004: 6),we are left depleted, sad, exhausted, or apathetic. Ouraffective encounters can increase or diminish our energies,they can “animate” or “disanimate” us.

Considering this affective dimension of domestic work inprivate households, the first striking aspect in the narrativesof female domestic workers and their female employers isthe depletion of energy associated with this work. Numerousaccounts speak of the “draining,” “exhausting,” “monoto-nous,” “repetitive,” “apathetic,” “lifeless” feelings attached tothe delivery of tasks like cleaning, making the bed, sweepingthe floor, washing clothes or dishes; but also the routinestructure of preparing meals, grooming or dressing children,as well as just doing the tasks that need to be done, which noone feels like doing and no one notices when they are done.The feelings connected to these tasks are not felt just becausethey are supposed to be boring. Instead, boredom is attachedto tasks because boredom is related to the cultural perceptionof this work as “banal” and lacking any social, professional orfinancial recognition. Within this context, the feelings andemotions that are ingrained in domestic work and felt by thepeople delivering the work are expressed, impressed,exchanged, and circulated in the private households. Affect,therefore, not only unfolds context (Massumi, 2002), but isalso produced in a specific context. Thus, while they areexpressions of immediate bodily reactions and sensations,which are neither rationalized through language nor situatedin a dominant semantic script, they impact people and places,and are situated in a social space, such as a private household.

The affective energies attached to the organization anddynamics of unpaid and paid domestic work in privatehouseholds evolve within the logic of the feminization oflabor. As Annie Phizacklea and Carol Wolkowitz assert,feminization describes the “declining terms and conditions ofemployment, so that a large proportion of the labor force hascome to experience “feminized” (that is, poor and insecure)conditions of work in some cases through deregulation at thenational level.” (Phizacklea & Wolkowitz, 1995: 3) Domesticwork signals this terrain of deregulated work, which is abjectand devalued in society. The subjects providing this work areculturally predicated by signifiers of “inferiority,” producedthrough processes of feminization, and, as we will see below,also by racialization.

In conversations with female employers and domesticworkers alike, the sentiment of “inferiorization” is expressedin their reflection on their positioning as “mothers” and“housewives.” Employing another woman to perform do-mestic work releases them from this positioning and enables

Page 50: Womens Studies International Forum 46 (2014)

48 E. Gutierrez-Rodriguez / Women's Studies International Forum 46 (2014) 45–53

them to experience positive feelings. For example, Antonia, ateacher who lives with her daughter in Vienna, stresses how“happy” they are with the person they have employed to dothe household work. She describes her in the followingwords, “Our fairy, she simply helps us—we both sense it quitestrongly. This just makes us happy. She is like our fairy. I say ittoo. She is a fairy [laughs]—a real miracle. She lives andflourishes! And she works like one too.” (MAIZ, 2004: 69),(Antonia, employer, Vienna).

Antonia uses flowery terms to refer to their domesticworker. Her vital (“she lives”) and productive (“flourish”)potential is noted. Employing a domestic worker enables thehousehold to engage with positive feelings, which affect thehousehold and its members in animating ways, while thedomestic worker takes on the negative affective burdeningrained in this work. For the employer, employing anotherperson to deliver this work means that she is exempt, at leastfor some of the day, from doing this work. Thus, most of theemployers share the feelings that Stephanie summarizessuccinctly, “I would get sick of the sight of these four walls. Imean, I stayed at home for four and a half years. I wasbasically the cleaning woman, the nursemaid. Ultimately, Iam happy, that I have a job outside of the home where I amnot confronted with the household or children—my ownchildren.” (MAIZ, 2004: 63) (Stephanie, employer, Linz).

Stephanie's feelings toward the option of becoming astay-home mother are associated with illness, connoting astate of physical and psychic deterioration. What is interest-ing about Antonia's and Stephanie's accounts is that thefeeling of “happiness” appears in association with anotherperson doing the household work. “Happiness” seems to befelt the moment somebody takes over the depleting aspect ofdomestic work, thus freeing up time to engage withprofessional ambitions and aspirations, or just to relax withthe family, children, partner, or friends. This sense ofrecouping time and personal autonomy might be what theemployers experience when they employ another woman todo the cleaning or caring work. In employing another womanto do the domestic work the disanimating feelings related todomestic work are shifted from the female employer to thedomestic worker. The female employers succeed in escapingthe cultural inscription associated with domestic work asdevalued labor and evading the ascription of their bodies asinferior under the signifier “femininity”. Releasing them-selves from the responsibility of performing household workenables these women to regain a sense of well-being in aterrain that is historically determined and symbolicallyprescribed by correlating femininity with subordination,serfdom, and exploitation. In addition, the affective burdenimpressed upon their bodies is transferred to other women.While employing a domestic worker allows them to partiallyescape the abjection projected onto them, they still remainresponsible for delivering and managing this work in thehousehold (Caixeta et al., 2004; Gutiérrez Rodríguez, 2010).Further, in most cases, the female employer monitors thework performed by the domestic worker. This is also due tothe fact that the male members and/or the children in thehousehold often fail to fulfill their household duties or are notinvolved in household work. As managing household work isdelegated to the adult female members of the household, incases where they refuse to perform these tasks and where

finances allow, outsourcing domestic work becomes anoption. However, this option is subject to economic andpolitical conjunctures such as migration regimes.

Migration regimes and the coloniality of labor

The contemporary organization of domestic work inprivate households in Western Europe is sustained byoutsourcing domestic work (Anderson, 2000; Anthias &Lazaridis, 2000; Lutz, 2008; Triandafyllidou, 2013). Women,who are poor, migrant or minoritized have taken over thiswork. In the case of migrant women, migration policies play asignificant role in determining their access to the labormarket. It is through migration policies that differences areestablished between the national (citizen) and the newcom-er (migrant) populations. Thereby, this process of differen-tiation reactivates a mental matrix that is rooted in colonialracial classification. While not explicitly operating within theracial matrix, migration policies reactivate this logic ofdifferentiation through the classification of the populationinto different categories of citizens, denizens, and aliens.

When a migrant woman, and in particular an undocu-mented migrant woman, is employed to deliver domesticwork in private households, the differential system estab-lished through migration policies becomes tangible in theencounters between domestic workers and their employers.Coupled with discursive regimes and historical perceptions ofthe “Other” to the nation as racially and culturally different,migration policies are experienced cognitively and alsosensed. These feelings and emotions become palpable in theeveryday encounters between domestic workers and em-ployers, as well as in the places they inhabit. When a migrantdocumented or undocumented worker is employed, migra-tion policies enter the field of the private household and withit, as previously mentioned, a whole cultural realm ofimagining the nation's Other is reactivated. These imaginingsare transferred and negotiated in direct and indirect ways insocial spaces, finding a particular expression in the feelingsassociated with objects and places as we see in Carmen'swords next.

“The worst for me, we could say, are the toilets! So, you,you see people who are really spick and span, but you canforget it. Really! So I wear gloves everywhere. You knowrubber gloves?! […] Because I don't know?! It could be,they are people who may be super clean, but to theoutside world! But you, you know the people in thekitchen and in the toilets! So, really! Brushes are availableeverywhere! Thank god we drink only tea now! [Carmensmiles] Brushes, these toilet brushes are availableeverywhere! At least, what can you do? What you cando is make it a little bit cleaner. But it is sprayed all over!Pee all over! The men cannot pee properly at all! (….)”,(Carmen, domestic worker, Hamburg)

With a gesture of disgust, Carmen comments on thesituation in the household where she works. Finding thetoilet dirty leaves her with a sensation of neglect andignorance. As Rosie Cox (2006) notes in regard to therelationship between dirt, cleaning, and status, the way adomestic worker is socially perceived and treated is related

Page 51: Womens Studies International Forum 46 (2014)

49E. Gutierrez-Rodriguez / Women's Studies International Forum 46 (2014) 45–53

to society's cultural conception of “dirt.” Those who workwith sewage, collect rubbish, or clean are poorly paid andmostly exposed to unsafe and insecure working conditions.Doing “dirty work” signals the lowest rank on the socialladder (see also Anderson, 2000). As Cox notes, the “status ofthe worker becomes inseparable from the status of the workand it is impossible to improve dramatically the standing ofeither without challenging deep-seated feelings about dirt”(2006:7). Yet, dealing with “dirt,” as Carmen tells us, not onlysignals a low social status, it also refers to the affectiveeconomies within society and private households.

All the domestic workers told the research teams aboutthe depleting feelings they encountered in the householdswhen it came to the more “banal,” “simple,” and sociallydisregarded tasks, such as cleaning the floors, removingfluids, hairs, and general dirt. These tasks are intrinsicallylinked to our basic needs and are a constant reminder of ourhuman condition. As the domestic worker deals with thephysical and affective traces of our lives, the barriers andboundaries of social differences crumble. This destabilizes thepower asymmetries engraved in the relation between thedomestic worker and the employer. Not only do affectivebonds emerge between the female employer and femaledomestic worker, as well as between the domestic workerand other household members, the domestic worker alsobecomes a silent bystander in moments of absolute intimacyand an addressee of emotions and feelings circulating in thehousehold (Gutiérrez Rodríguez, 2007). Getting to know thetoilets, as some of the domestic workers told us, is getting toknow about the “inner lives” of their employers. Toilets areinstilled with people's energies, which have an impact on thedomestic worker's body and mind.

Leaving the toilet dirty or not cleaning it with the brushtransmits an explicit message of contempt. For WilliamMiller, “contempt” transmits the sensation of not being“noteworthy” (1998: 215). While the users of the toiletmight not intend this, the feeling of contempt is conveyedthrough the non-use of brushes, fluids, and dirt found. Theway the household members use the toilet reveals a lack ofcare in regard to the person cleaning this space. This attitudeuncovers the invisibility attached to domestic work and tothe person delivering this work. Involuntarily, the domesticworker must face these energies, which negatively impacther and cause her to feel revulsion and disgust.

Disgust is a strong feeling and, as Sianne Ngai (2007)suggests, it is “a structured and agonistic emotion carrying astrong and unmistakable signal” (335). For Ngai, disgust is notambivalent about its objects. Within the context of domesticwork, the feeling of disgust expresses a sensation carried by thesocial meaning denoted by this labor. In addition to this, thesocial significance of domestic work is defined through thehistorical legacy of colonialism, slavery, indentured labor,serfdom (Banerjee, 2004; Davis, 1983; Morgan, 2004; Rollins,1985; Romero, 1992), and the contemporary organization of aheteronormative social order (Carrington, 1999; GutiérrezRodríguez, 2013).

While the feelings expressed in and impressed upondomestic work do not always emanate from it, they do unfoldwithin it. The impression of feelings of invisibility andworthlessness are negotiated within this social context,reviving the cultural logic of abjection as these evolve from

a racializing and feminizing script of power. When it comesto employing an undocumented migrant domestic worker,the social context prescribing how domestic work in privatehouseholds is seen is defined through gender relations andthe logic underpinning migration policies. Within the logic ofthe coloniality of power, migration control and managementpolicies enforce processes of subalternization and thedynamics of inferiorization are enforced.

Anibal Quijano (2000, 2005, 2008) identifies the “colonialityof labor” as one of the axes along which the “coloniality ofpower” establishes a societal system of exploitation based onthe correlation of “race” and value. As Quijano states, duringcolonial times labor was racially codified. While the laborextracted from those codified as “white” was consideredproductive and “superior,” the labor power extracted from theindigenous and enslaved populations was seen as “inferior”and, as such, conceived as “free exploitable” labor (Quijano,2000). As Quijano (2000, 2005, 2008) and Enrique Dussel(1995) state, what lies behind this model is the colonizers'perception of the colonized population as pure objects ofexploitation. As Dussel notes, perceived as the “Other” ofEurope, the colonized population was “subsumed, alienatedand incorporated into the dominating totality like a thing orinstrument” (1995: 39). In other words, this population wasreduced to a “thingness” and treated as “raw material.” Thus,Spanish and Portuguese colonialism established a “new modelof global power” on the basis of which the capitalist mode ofproduction would evolve (Quijano, 2008).

This logic of subjugation inherent in the establishment of aracially coded social system still reverberates in the construc-tion of the nation's Other in Western Europe, although it doesnot always explicitly operate in racial terms. Though wrappedin a vocabulary of culture that refers to differences in regard toreligious beliefs, language, norms, and values, the constructionof the “Other” in Western Europe bears the traces of a patternof thinking reminiscent of colonial times. As such, theconstruction of the “Other” as the negation of a European self,imagined aswhite and Christian, remains the neuralgic point in“the structure of power” that “was and even now continues tobe organized on and around the colonial axis” (Quijano, 2008:216). Thus, the coloniality of labor refers to the socialorganization of labor and division of work sustained by thissystem of cultural codification.

This same model is reactivated within the context ofmigration policies. While a colonial system of classification isnot explicit within contemporary national migration policieswithin the EU, the divide between “citizen” and “alien” (migrantand refugee) reverberates with the logic of coloniality. Thisbecomes apparent, for example, in the requirements formigrantsand refugees to be granted entry and settlement, so that theymay establish themselves within the EU. Indeed, migrants fromnon-EU states must comply with the constantly shifting andincreasingly restrictive national requirements. Changes in familyreunification policies (Kofman, 2011; Kraler, 2010; Kraler,Kofman, Kohli, & Schmoll, 2012), visa policies and, in particular,student visas are making entry and settlement in EU memberstates increasingly difficult.

Most of the participants in my research from LatinAmerica arrived in Germany with a tourist visa. Some ofthem were planning to pursue postgraduate studies, butencountered barriers when attempting to enroll in university

Page 52: Womens Studies International Forum 46 (2014)

50 E. Gutierrez-Rodriguez / Women's Studies International Forum 46 (2014) 45–53

programs, as their Latin American degrees were not consid-ered valid or, as they were told, they needed to improve theirlanguage skills. The three-month time frame they receivedwith their tourist visa was insufficient, leaving them with theoption to return to their countries, to continue onward toanother country, or to go underground. The decision toremain in the country is also a decision to live without legalresidency status. Being without legal residency meansavoiding places where they could be detected by the police,which severely restricts their access to the regular labormarket. Obtaining employment in a private household as adomestic worker represents one of the few options formaking a living.

The private household remains an employment sector thatis precluded from standard labor regulations, whichmakes thehousehold shaky ground when it comes to the protection ofworkers' rights (ETUC, 2012). This is shown in the workingconditions of domestic workers, which are characterized byoral contracts, unregulated working hours, as well as unsafeworking conditions. Being without legal residency meansfalling through the cracks of official protection schemes(Düvell, 2005; Maroukis, Iglicka, & Gjmaj, 2011). In thiscontext, it is not only the universal human rights principle ofthe right to a dignified life that is suspended, but also basic civilrights. Indeed, this process of exclusion produces the “undoc-umented migrant” as an “exteriority” to the civil and legalnational norm.

Coming back to the affective dimension of being subjectedto the logic of inferiorization, the position of “exteriority” thatundocumented migrants inhabit is not only conveyed throughthe low social status and devaluation of their labor, but alsothrough the affective circuits to which they are exposed andwith which they engage. Negotiations around domestic workbetween domestic workers and their employers in privatehouseholds occur on both the social and affective levels. Whilethe social dimension speaks about the legal and laborconditions attached to this work, the affective dimensiondraws our attention to how these conditions are felt andsensed. Sensing the social, feeling the cultural script imposedon bodies, spaces and objects refers to the realm of affectiveencounters.

Thus, the transmission of affect between the domesticworkers and their employers relies on affective bondsdeveloped, on the caring tasks performed, and on the spatialityand relationality within which this work unfolds. As Carmentells us, the arrangement of objects can transmit energies thatmight have an impact on the person sorting out the space. Theattribution of inferiority and worthlessness can thus beaffectively transmitted through the arrangement of objectsand through those racialized affects which haunt spaces.

Racialized affects

While not directly addressed to a person, affect pervadesspaces and has an impact on people's feelings, bodies andminds. When a domestic worker faces revulsion, contempt, orbeing despised, these feelings can produce reactions of refusaland revolt. Set in this context, Carmen's feelings towards thetoilet remind us of what Ngai (2007) describes as theracialization of affect. While Ngai develops this approachthrough an analysis of the cultural representation of racialized

bodies, the affect transmitted to Carmen seems to attend to asimilar dynamic. As Ngai notes in regard to racialized affect thecontext of racialization “turns the neutral and even potentiallypositive affect of animatedness ugly pointing to the moreself-evidently problematic feelings” (2007: 32). In the case ofundocumented domestic workers in private households, thiscontext is underscored by the exclusionary boundaries set upby migration policies that subtly underlie the encountersbetween domestic workers and their employers. It is in thisregard that feelings circulating in the space and expressed inthese encounters turn “ugly,” imprinting residues of a racial-izing script and attributing “inferiority” onto the body of thedomestic worker. Feelings circulating within a context ofracialization do not always incite us to act, they can alsoimmobilize us. The feeling of being made invisible, of beingignored, infuses the domestic worker with the feeling of socialinsignificance, it carries the sensation of disanimation. Thisstands in contrast to the impact that domestic work has on thehousehold as an animating force. Thus, while not explicitlyspelled out in any job description, the domestic worker'spresence in the household, and the performance of quotidiantasks that contribute to the household members' well-being,infuses this place with life. Therefore, domestic work has an“animating” quality, although this effect is not commonlyperceived or valued.

While the domestic worker is required to care for thehousehold and, as such, contributes to the creation of positiveaffective energies, the tasks she is supposed to deliver as well asthe dynamics she occasionally encounters, affectively impressupon her a low status and a sense of devaluation. Furthermore,her position as an undocumented migrant places her in ahuman rights void and makes her more vulnerable toexploitation and denigration. When delivered by an undocu-mented migrant woman, domestic work becomes a neuralgicnode in which the “multilayered texture” (Combahee RiverCollective, 1983) of oppression crystallizes.

Workingwith and through the affective texture of domesticwork not only makes us aware of how our ability to act isemotionally driven, but also how emotions and feelingsintersect with as well as carry social meaning. The analysis ofdomestic work illustrates that affect focuses on the transgres-sion of the script of representation and (re-)cognition. As wellas this affective dynamics evolve in a social context, eventhough they emanate from the spontaneity of our existence,instantly transforming the normative texture of intelligibility.Affect dwells in-between the agitations and arousals of ourrelational lives and our attempts to make sense of them. Thus,while emotions and feelings reflect the affective immediacy ofordinary encounters, they are mediated through social mean-ings that also reflect historical becomings and geopoliticalpositionings.

Set within the context of racism and labor exploitation,the analysis of affective encounters in private householdsthat employ undocumented migrant domestic workers shedssome light on how emotions and feelings are not justintentional, but are impressed upon objects and spaces.Thus, the exploration of affect in domestic work tells us aboutthe impression of “ugly” feelings on people's bodies and howthis affects their minds and well-being or, in other words,how these feelings can work to immobilize or “disanimate”them. Yet, as we have seen through Carmen's words, the

Page 53: Womens Studies International Forum 46 (2014)

51E. Gutierrez-Rodriguez / Women's Studies International Forum 46 (2014) 45–53

response to feelings related to attempts at denigration andsubjugation can also be answered with feelings of rebellionand resistance.

Considering these affective dynamics, the analysis ofdomestic work as affective labor not only addresses theemotional and physical qualities of this labor. Rather, it placesthe impression, expression, and circulation of feelings andemotions in the private household as a point of departure inany such analysis. Thus, seeing domestic work as affectivelabor draws our attention to the productive character of careand to the extensive dynamics of societal reproduction. It alsobrings to awareness the emotional economies organizing theimmediacy of employment relations. The feelings andemotions circulating in the private households whichdomestic workers are exposed to and need to deal with, areexpressions of social asymmetries and articulations of globalinequalities.

Conclusion

The employment of undocumented migrant domesticworkers in private households articulates the re-shuffling ofglobal inequalities on a local level (Hondagneu-Sotelo, 2001).Set in this context, the understanding of domestic work asaffective labor asks us to contextualize the expression andimpression of feelings within a societal context. Immersed inthe energies of private households, crossed by the racializingeffects produced through migration policies and the logic ofthe labor market, domestic work is mediated by and derivesfrom affective dynamics.

Focusing on the emotional faculties deployed in caring forothers and the affective fabric of the social, the analysis ofaffective labor addresses the expression, impression andcirculation of feelings and emotions and how these dynamicsimpact bodies, objects, relations and spaces. Thus, theanalysis of affective labor addresses the sensorial corporealityof the social. As such, it speaks of the more legible cognitivedimension of feelings and emotions as well as of thedispersed and unruly dynamics of feelings, emotions, sensa-tions, and intensities.

The analysis of domestic work as affective labor reveals thesociability of affect and the affectivity of the social. This couldlead us to raise the question on the value this discussion couldadd to domestic workers' organizations' claims for standarddomestic workers' rights. Thus, on the level of domesticworkers' rights, the conceptualization of domestic work asaffective labor not only invites us to address demandsregarding the professionalization and regulation of domesticwork,3 but also to go a step further. While the claims fordignified working conditions (ETUC, 2012; FRA, 2011), “porta-ble workers' rights”4 (Piper, 2007) and the social recognition ofdomestic work as work are fundamental claims in the strugglefor domestic workers' rights, putting affect on the agenda ofworking rights strengthens the argument raised that domesticwork is fundamental for the reproduction of society. Looking atthe affective dimension of domesticworkmakes us realize howthis work is intrinsically linked to the pulses of life, therelational, reciprocal, and interconnected character of oursocial being. Labor in general, and domestic work in particular,as such need to be conceived in relation to the feelings,emotions, sensations, and intensities that drive them, which

become embedded in places and experienced in ordinaryencounters. Indeed, the tasks performed in domestic work areintrinsically linked to sustaining personal well-being. There-fore, recognizing this affective dimension in domestic workrequires that we place it at the core of debates on convivialfutures.5 This perspective enables us to set domestic work asaffective labor at the center of the capacity for creating a ‘goodlife’ and enabling a ‘living together’.

Endnotes

1 This study was an EU research project on the interpersonal relation-ships between domestic workers and their employers in private householdsconducted in Austria, Germany, Spain and the UK between 2003 and 2004.Twenty-five in-depth interviews and ten focus groups were held in eachcountry with domestic workers from Eastern Europe, West Africa and LatinAmerica, and with middle-aged, professional White women who were theiremployers.

2 For Spinoza “passion” is affect produced by external causes (Spinoza,1994: 154).

3 For further discussion, see “European Parlament Resolution onRegulating Domestic Help in the Informal Sector”. Retrieved fromhttp://www.europarl.europa.eu/sides/getDoc.do?pubRef=-//EP//TEXT+TA+P5-TA-2000-0542+0+DOC+XML+V0//EN.

4 With this term Nicola Piper discusses the possibility for migrantworkers to claim rights while they are on the move. In particular, she makesthe case for workers that have worked in a country and have been deportedor needed to leave the country of employment to be able to claim theirsalary, redundancy rights and other workers' rights, although they no longerreside in the country (Piper, 2007).

5 As I argue elsewhere in regard to a decolonial ethic of care (GutiérrezRodríguez, 2010), domestic work is driven by the principles of care andsustainability. This affective fabric of domestic work urges us to reconsiderCarol Gilligan's (1982) ethics of care. Gilligan's discussion of the ethic of careaddresses the question of moral judgment departing from contextualizedand interconnected moral claims. In this sense, the sphere of care becomes a“moral imperative” from which to approach questions of reciprocity andresponsibility. As feminist theorists engaging with the ethics of care(Bowden, 1997; McDowell, 2004; Tronto, 1993, 1995) have argued, asociety based on solidarity needs to share the common understanding thatcare and love (Kittay, 1999) are fundamental for communal life (Paredes,2008).

References

Ahmed, Sara (2004). The cultural politics of emotions. Edinburgh: EdinburghUniversity Press.

Anderson, Bridget (2000). Doing the dirty work? The global politics ofdomestic labour. London: Zed Books.

Anthias, Floya, & Lazaridis, Gabriella (2000). Gender and migration inSouthern Europe: Women on the move. New York: Berg.

Bair, Jeniffer (2010). On difference and capital: Gender and the globalizationof production. Signs, 36(1), 203–226.

Bakker, Isabella (2007). Social reproduction and the constitution of agendered political economy. New Political Economy, 12(4), 541–556.

Bakker, Isabella, & Gill, Stephen (2003). Power, production and socialreproduction. Basingstoke, NY: Palgrave Macmillan.

Banerjee, Swapna (2004). Men, women, and domestics: Articulating middle-class identity in colonial Bengal. Delhi: Oxford University Press.

Barker, Drucilla., & Feiner, Susan F. (2010). As the world turns: Globalization,consumption, and the feminization of work. Rethinking Marxism, 22(2),246–252.

Barrett, Michele (1980). Women's oppression today. Problems in Marxistfeminist analysis. London: Verso.

Bedford, Kate., & Rai, Shirin (2010). Feminists theorize international politicaleconomy. Signs, 36(1), 1–18.

Benería, Lourdes (1979). Reproduction, production and the sexual divisionof labour. Cambridge Journal of Economics, 3(3), 203–225.

Benería, Lourdes, & Sarasúa, Carmen (2011). Crímenes económicos contra lahumanidad. El País, 29.03.2011. Retrieved from. http://elpais.com/diario/2011/03/29/opinion/1301349604_850215.html (accessed 30March 2011)

Page 54: Womens Studies International Forum 46 (2014)

52 E. Gutierrez-Rodriguez / Women's Studies International Forum 46 (2014) 45–53

Boris, Eileen, & Parreñas, Rhacel S. (2010). Intimate Labors. Cultures,technologies and politics of care. Stanford: Stanford University Press.

Bowden, Peta (1997). Caring: Gender-sensitive ethics. London: Routledge.Brennan, Teresa (2004). The transmission of affect. Ithaca: Cornell University

Press.Caixeta, Luzenir, Gutiérrez Rodríguez, Encarnacion, Tate, Shirley A., & Vega

Solís, Cristina (2004). Homes, care and borders—Hogares, cuidados yfronteras. Madrid: Traficantes de Sueños.

Carraquer Oto, Pilar (2013). El redescubrimiento del trabajo de cuidados:algunas reflexiones desde la sociología. Cuadernos de RelacionesLaborales, 31(1), 91–113.

Carrasco Bengoa, Cristina (2013). El cuidado como eje vertebrado de unanueva economía. Cuadernos de Relaciones Laborales, 31(1), 39–56.

Carrington, Christopher (1999). No place like home. Relationships and familylife among lesbians and gay men. Chicago: The University of ChicagoPress.

Combahee River Collective (1983, orig. 1977). The Combahee RiverCollective Statement. In B. Smith (Ed.). Home girls. A Black feministanthology, (pp. 272-282). New York: Kitchen Table/Women of ColorPress.

Corsani, Antonella (2007). Beyond the myth of woman: The becoming-transfeminist of (post-)Marxism. SubStance, 36(1), 107–138 (112).

Cox, Rosie (2006). The Servant Problem. Domestic employment in a globaleconomy. London: I.B. Tauris.

Dalla Costa, Maria, & James, Selma (1972). The power of women and thesubversion of the community. London: Butler and Tanner Ltd.

Davis, Angela Y. (1983). Women, race & class. New York: Vintage Book.del Río, Siria (2004). La crisis de los cuidados: precariedad a flor de piel. Rebelión.

Retrieved from. http://www.rebelion.org/hemeroteca/economia/040308sira.htm (accessed 12 June 2013)

Delphy, Christine (1984). Close to home: A materialist analysis of women'soppression. London: Hutchinson (Translated by D. Leonard).

Dussel, Enrique (1995). The invention of the Americas: Eclipse of “the Other”and the myth of modernity. New York: Continuum (Translated by MichaelD. Barber).

Düvell, Franck (2005). Illegal immigration in Europe. Beyond control.Houndsmills: Palgrave/MacMillan.

Elson, Diane (1998). The economic, the political and the domestic:Businesses, states and households in the organisation of production.New Political Economy, 3(2), 189–208.

European Trade Union Conference (ETUC) (2012). 5th Annual ETUC 8 MarchSurvey 2012. (Brussels).

Fantone, Laura (2007). Precarious changes: Gender and generational politicsin contemporary Italy. Feminist Review, 87, 5–20.

Fantone, Laura (2011). Genere e precarieta. Naples: ScriptaWeb.Federici, Silvia (2004). Caliban and the Witch. Women, the body and

primitive accumulation. San Francisco: Autonomedia.Federici, Silvia (2006). Precarious labor: A feminist viewpoint. Retrieved

from. http://inthemiddleofthewhirlwind.wordpress.com/precarious-labor-a-feminist-viewpoint/ (accessed 12 May 2012)

Federici, Silvia (2012). Revolution at point zero: Housework, reproduction, andfeminist struggle. London: PM Press.

Ferber, Marianne, & Nelson, Julie (1993). Beyond economic man. Chicago:University of Chicago Press.

Folbre, Nancy (1994). Who pays for the kids? Gender and the structures ofconstraint. New York: Routledge.

Folbre, Nancy (2006). Measuring care: Gender, empowerment, and the careeconomy. Journal of Human Development, 7(2), 183–199.

FRA (European Union Agency for Fundamental Rights) (2011). Migrants inan irregular situation employed in domestic work. (Vienna, Austria).

Gilligan, Carol (1982). In a different voice: Psychological theory and women'sdevelopment. Mass. Harvard University Press.

Gutiérrez Rodríguez, Encarnacion (2007). The ‘hidden side’ of the neweconomy—On transnational migration, domestic work and unprece-dented intimacy. Frontiers: Journal of Women Studies, 28(3), 60–83.

Gutiérrez Rodríguez, Encarnacion (2010). Migration, domestic work andaffect. New York: Routledge.

Gutiérrez Rodríguez, Encarnacion (2013). The precarity of feminisation: Ondomestic work, heteronormativity and the coloniality of labour.International Journal of Politics, Culture and Society, 12, 1–14.

Hewitson, Gillian (1999). Feminist economics: Interrogating the masculinity ofrational economic man. Cheltenham, UK: Edwar Elgar.

Himmelweit, Susan (1995). The discovery of ‘unpaid work’. The socialconsequences of the expansion of ‘work’. Feminist Economics, 1(2), 1–19.

Hochschild, Arlies R. (1983). The managed heart: Commercialization of humanfeelings. Berkeley: University of California Press.

Hochschild, Arlies R. (2003). The commercialization of intimate life. Notesfrom home and work. Berkeley: University of California Press.

Hondagneu-Sotelo, Patricia (2001). Doméstica: Inmigrant workers cleaning andcaring in the shadows of affluence. Berkeley: University of California Press.

Illouz, Eva (2007). Cold intimacies: The making of emotional capitalism.Cambridge: Polity.

Jacobs, Susie (2010). Gender and agrarian reform. London: Routledge.Kittay, Eva F. (1999). Love's labor. London: Routledge.Kofman, Eleanor (2011). Family reunion legislation in Europe: Is it discrimi-

natory for migrant women? Brussels: European Network of MigrantWomen.

Kofman, Eleanor (2012). Rethinking care through social reproduction:Articulating circuits of migration. Social Politics, 19(1), 142–162.

Kraler, Albert (2010). Civic stratification, gender and family migrationpolicies in Europe. Final Report. Vienna: BMWF/ICMPD (Revised andupdated public version).

Kraler, Albert, Kofman, Eleonore, Kohli, Marin, & Schmoll, Camille (2012).Gender, generations and the family in international migration. Amsterdam:Amsterdam University Press/IMISCOE.

Lan, Pei-Chia (2006). Global cinderellas: Migrant domestic and newly richemployers in Taiwan. Durham, N.C. Duke University Press.

Lutz, Helma (2008).Migration and domestic work. A European perspective ona global theme. Farnham: Ashgate.

MAIZ (Rappold, S., Caixeta, L., Haas, B., Haidinger, B., Rechling, D. & Ripota, P.)(2004). Housework and care work in Austria: Migrant women in privatehouseholds. Unpublished Report.

Maroukis, Thanos, Iglicka, Krystyna, & Gjmaj, Katarzyna (2011). Irregularmigration and informal economy in southern and central-easterneurope: Breaking the vicious cycle? International Migration, 49(5),129–156.

Martín Palomo, Maite (2008). Domesticar el trabajo: una reflexión a partirde los cuidados. Cuadernos de Relaciones Laborales, 26(2), 13–44.

Martín Palomo, Maite (2013). Tres generaciones de mujeres, tresgeneraciones de cuidados. Apuntes sobre una etnografía moral.Cuadernos de Relaciones Laborales, 31(1), 115–138.

Massumi, Brian (2002). Parables for the virtual. Movement, affect, sensation.Durham, N.C. Duke University Press.

McDowell, Linda (2004). Work, workfare, work/life balance and an ethic ofcare. Progress in Human Geography, 28(2), 145–163.

Mies, Maria (1999). Patriarchy and accumulation on a world scale: Women inthe international division of labour. London: Zed Books.

Molyneux, Maxine (1979). Beyond the domestic labour debate. New LeftReview, 115, 3–28.

Morgan, Jennifer L. (2004). Laboring women: Reproduction and gender in newworld slavery. Philadelphia: University of Pennsylvania Press.

Ngai, Sianne (2007). Ugly feelings. Cambridge, Mass. Harvard UniversityPress.

O'Hara, Patricia (1998). Partners in production. Women, farm and family inIreland. New York: Berghahn Books.

Paredes, Julieta (2008). Hilando Fino. Desde el feminismo comunitario ComunidadMujeres Creando Comunidad. La Paz: CEDEC, Asociación Centro de Defensa de laCultura.

Pérez Orozco, Amaia (2004). Estrategias feministas de deconstrucción delobjeto de estudio de la economia. Foro Interno: Anuario de Teoría Política,4, 87–118.

Pérez Orozco, Amaia (2010). Global perspectives on the social organizationof care in times of crisis: assessing the situation. Gender, Migration andDevelopment Working Paper 5. Washington: UN-INSTRAW.

Peterson, Spike V. (2009). Interactive and intersectional analytics ofglobalization. Frontiers: A Journal of Women Studies, 30(1), 31–40.

Phizacklea, Annie, & Wolkowitz, Carol (1995). Homeworking women: Gender,racism and class at work. London: Sage.

Piper, Nicola (2007). New perspectives on gender and migration—Rights,entitlements and livelihoods. London: Routledge.

Precarias a la Deriva (2004). A la deriva: por los circuitos de la precariedadfemenina. Madrid: Traficantes de Sueños.

Quijano, Anibal (2000). Colonialidad del poder y clasificación social. Journalof World-Systems Research, VI(2), 342–386.

Quijano, Anibal (2005). Colonialidad del poder, eurocentrismo y AméricaLatina. In E. Lander, & S. Castro-Gómez (Eds.), La colonialidad del saber.Eurocentrismo y ciencias sociales: Perspectivas Latinoamericanas. (pp.201–246). Buenos Aires: Consejo Latinoamericano de Ciencias SocialesCLACSO.

Quijano, Aniba (2008). Coloniality of power, eurocentrism, and socialclassification. In M. Moraña, E. Dussel, & C. A. Jáuregui (Eds.), Colonialityat Large. Latin America and the postcolonial debate. (pp. 181–224).Durham, N.C: Duke University Press.

Redclift, Nanneke (1985). The contested domain: Gender, accumulationand the labour process. In M. Redclift, & E. Mingione (Eds.), BeyondEmployment. Household, gender, and subsistence. (pp. 93–115). London:Basil Blackwell.

Rollins, Judith (1985). Between women: Domestics and their employers.Philadelphia: Temple University Press.

Romero, Mary (1992). Maid in the U.S.A. New York: Routledge.

Page 55: Womens Studies International Forum 46 (2014)

53E. Gutierrez-Rodriguez / Women's Studies International Forum 46 (2014) 45–53

Ruido, Maria (2008). Just do it! Bodies and images of women in the newdivisionof labor. Retrieved fromhttp://caringlabor.wordpress.com/2011/02/08/maria-ruido-just-do-it-bodies-and-images-of-women-in-the-new-division-of-labor/ (accessed 12 December 2013)

Sconvegno (M. Galetto, C. Lasala, S. Magaraggia, C. Martucci, E. Onari, F. Pozi).(2007). A snapshot of precariousness: voices, perspectivas, dialogues.Feminist Review, 87, 104–112.

Sedgwick, Eve K (2004). Touching feeling. Affect, pedagogy, performativity.Durham, N.C. Duke University Press.

Spinoza, Baruch (1994). The ethics. In E. Curley (Ed.), A Spinoza Reader. Theethics and other works. (pp. 85–126). Princeton: Princeton UniversityPress.

Tate, Shirley (2009). Black beauty: Aesthetization, stylization, politics.Farnham: Ashgate.

Triandafyllidou, Anna (2013). Irregular migrant domestic workers in Europe.Farnham: Ashgate.

Tronto, Joan (1993). Moral boundaries. New York: Routledge.Tronto, Joan (1995). Care as a basis for radical political judgments. Hypatia,

10(2), 141–149.Vega Solís, Cristina (2009). Culturas del cuidado en transición. Barcelona:

Editorial UOC, Niberta.Waring, Marilyn (2004). Counting for nothing. What men value and what

women are worth. Toronto: University of Toronto Press.Weeks, Jacquilyn (2011). Un-/re-productive martenal labor: Marxist femi-

nism and chapter fifteen of Marx's Capital. Rethinking Marxism, 23(1),31–40.

Page 56: Womens Studies International Forum 46 (2014)

Women's Studies International Forum 46 (2014) 54–62

Contents lists available at ScienceDirect

Women's Studies International Forum

j ourna l homepage: www.e lsev ie r .com/ locate /ws i f

Inferiorization and deference: The construction of socialhierarchies in the context of paid domestic labor

Débora Gorbán a, Ania Tizziani b,⁎a Conicet/Idaes- Universidad Nacional de San Martín, Parana 145, 5th Floor, Buenos Aires 1017, Argentinab Conicet/Universidad Nacional de General Sarmiento, Juan María Gutiérrez 1150, Los Polvorines, Provincia de Buenos Aires, 1613, Argentina

a r t i c l e i n f o

⁎ Corresponding author.

0277-5395/$ – see front matter © 2014 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.wsif.2014.01.001

s y n o p s i s

Available online 28 January 2014

In Argentina, domestic work is one of themain occupations for women from low-income sectors.As in other Latin American societies, it is one of the most paradigmatic forms of contact betweenthe different social classes. As such, this labor relationship has been analyzed in numerous studiesas a critical location for the reproduction of social differences and inequality. The interpersonalrelationships between employers and workers mobilize categorization criteria and stereotypedimages that revealwider dynamics regarding the construction of social hierarchies. On the basis ofa qualitative study, the objective of this article is to analyze, in the city of Buenos Aires, theprocesses of constructing social hierarchies that are implied by this particular labor relationship.This analysis seeks to reveal the operations through which employers construct a stereotype ofsocial inferiority for domestic workers through which they legitimize their dominant position inthe labor relationship, and to examine the tensions and ambiguities of this.

© 2014 Elsevier Ltd. All rights reserved.

Introduction

InArgentina, domesticwork is, andhas beenhistorically, oneof the main ways in which women participate in the labormarket, particularly women frompopular social sectors (Gogna,1993; Pereyra, 2012). As in other Latin American societies inwhich this type of work is widespread, it is one of the mostparadigmatic forms of contact between the working class andthe middle and upper classes. As such, domestic service hasbeen analyzed in numerous studies as a critical location for thereproduction of social differences and inequality.

In recent decades, domestic work has been the focus ofrenewed attention by social scientists. Although gender in-equalities are the starting point for many studies, the impor-tance of migratory flows in the structure of paid domestic laborthroughout different regions has turned migration studies intoone of the most relevant approaches for debating this issue(Ehrenreich &Hochschild, 2002; Hondagneu-Sotelo, 2007; Lutz,2002, 2008; Parreñas, 2001). These specific migratory flows,

ll rights reserved.

which are generally referred to as the “globalization of carework” (Ehrenreich & Hochschild, 2002), bring women intocontact across borders, creating asymmetrical relationshipsbetween employers from the central receiving countries andmigrant workers (Anderson, 2000; Gutiérrez Rodríguez, 2010;Hondagneu-Sotelo, 2007; Ibos, 2012; Parreñas, 2001). However,other studies reveal that the origins of these hierarchicalrelationships do not lie exclusively in the South–North migra-tion processes that exacerbated issues related to citizenship.Internal migratory dynamics, migratory flows between coun-tries in the South, and class/race distinctions also create theconditions for asymmetrical relationships (Brites, 2001, 2007;Kofes, 2001; Lan, 2003, 2008; Rollins, 1985; Romero, 2002).When considering the particularities of paid domestic labor inLatin American societies, researchers have privileged thisperspective (Chaney & García Castro, 1993). In these societies,this type of work has been the primary employment option forwomen from popular social sectors.

This article is framed by these perspectives, which revolvearound the analysis of domesticwork as one of the crucial spacesfor the construction and reproduction of social hierarchies basedon class position and racial belonging. From this point of view,

Page 57: Womens Studies International Forum 46 (2014)

55D. Gorbán, A. Tizziani / Women's Studies International Forum 46 (2014) 54–62

we maintain that domestic labor not only expresses thedynamics of social inequality, but also contributes to updatingand reproducing these in day-to-day life. We aim to analyze, inthe city of Buenos Aires, theway inwhich domesticworkers and,above all, employers perceive and manage the interactions thattake place within this labor relationship, paying particularattention to the emotional dimension (Lan, 2003, 2008; Rollins,1985; Romero, 2002). The objective of this analysis is to identifythe hierarchy and categorization criteria that come into play inthese interactions, and the tensions, ambiguities, and conflictspresent within them.

Different studies (Rollins, 1985; Romero, 2002) emphasizehow the interpersonal nature of the interactions that areestablished through domestic work plays a central role in theway in which the dynamics of this hierarchy are organized.These studies suggest that the interpersonal rituals that unfoldwithin the relationship between employers and domesticworkers mobilize categorization criteria and stereotyped im-ages that reveal wider dynamics regarding the construction ofsocial hierarchies. In this article, we analyze some of thesedynamics, through which a set of personal and social featuresattributed to domestic employees configure their social inferi-ority in the context of this labor relationship. This constructionjustifies the material exploitation of domestic workers while atthe same time reinforcing employers' class identity. Suchdynamics have significant effects on thewaywork is configuredwithin the sector as a strongly undervalued activity inwhich theprevailing labor and salary conditions are particularly unfavor-able for workers.

To analyze the hierarchy dynamics involved in day-to-dayinteractions between employees and employers, we turnedto certain concepts elaborated by Erving Goffman in his studyon social interactions, The Presentation of Self in Everyday Life,in particular the concept of the “front.” According to Goffman,the personal front is made up of the expressive features orelements that the performer identifies with. The features thatcharacterize the personal front are the main components ofthe way in which those interacting define the social situationthat brings them into contact: they provide informationabout the differences in social status that separate them andthe role that each party plays in the interaction. In the case ofdomestic employees and employers, these personal featuresare central components of the way in which they perceiveand handle the connection that is established through thislabor relationship (Goffman, 2009).

We will develop our argument in five sections. After a briefdiscussion of methodology, our analysis begins with a descrip-tion of this employment sector, which allows us to introduce theways in which domestic employees characterize their experi-ences of work. We will show how workers emphasize that thehighly undervalued nature of their work is one of the mainoccupational problems they face. In the following section, weseek to explore how this undervaluing of domestic employees'work is constructed from the point of view of their employers.Within this process, a series of characteristics (migration origin,poverty, ignorance) comes to define the workers' inferiority, inconnection with the profound inequality of their social andeconomic situations. These features define a front for domesticworkers that provides information about the subordinate rolethat employers attribute to them within the labor relationshipin order to sustain their own position of superiority. In the

following section, we focus on analyzing the tensions andconflicts arising from these attributed roles. In effect, the socialinferiority attributed to domestic workers leads to those hiringthem perceiving them as a threat that must be managed andcontrolled. This threat mainly consists of the possibility thatworkers will not fit in with the subordinate role attributed tothem, thus destabilizing the labor relationship.

Methodology

The reflections in this paper are based on a set ofqualitative data from different sources. During 2009, a seriesof twenty in-depth interviews with domestic workers wascarried out in Buenos Aires. These workers were contactedvia different organizations involved in the sector (unions andassociations), where both interviews and observations ofactivities were carried out. At the same time, over fourmonths, we carried out observations and a series of informalinterviews at two city playgrounds, where we were able tomake contact with workers who take care of children (inaddition to cooking and cleaning).

The ages of the workers in question ranged from 16 to65 years at the time of the interviews. Five of themwere live-inworkers, while the remainder consisted of “live-out” or dayworkers, that is, they resided in their own homes. Three-quarters of the workers interviewed were migrants: four camefrom different Argentinean provinces and eleven from othercountries (mostly Paraguay, but also Peru, Bolivia, andUruguay). Only in two cases were these migration experiencesrecent — the vast majority had been living in the Buenos AiresMetropolitan Area for decades. At the time the interviewswerecarried out, most had become legal residents in Argentina andmany had started their own families in the country. However,aswewill examine later in this paper,migration origin is one ofthe filters through which employers and employees perceivetheir class positions, and it plays a significant role in theprocesses of constructing the social hierarchies implied in thislabor relationship.

The second data source is a series of twelve in-depthinterviews carried out between 2010 and 2011 with peoplewho employ domestic workers. The sample is made up of fourmen and eightwomen between the ages of 35 and 69. Three aresingle, two divorced and the rest married; nine have betweenone and three children. Most contract domestic employees as“live-out” workers, but four have live-in workers. All theemployers interviewed belong to the middle or upper–middleclasses: they are professionals (teachers, lawyers, psychologists,economists), civil servants, havemanagement positions in largecompanies or run their own small businesses.

From our perspective, the study of this labor relationshipcannot ignore the structural inequality that shapes it, acondition that also affects the relationship between researchersand interviewees. As Pierre Bourdieu pointed out, the relation-ship present in an interview is subject to the effects of the socialstructure inwhich that interview is carried out. This relationshipis shot through with asymmetry (Bourdieu, 1993: 609):

It is the investigator who starts the game and sets up itsrules, and is usually the one who, unilaterally and withoutany preliminary negotiations, assigns the interview itsobjectives and uses. (On occasion, these may be poorly

Page 58: Womens Studies International Forum 46 (2014)

56 D. Gorbán, A. Tizziani / Women's Studies International Forum 46 (2014) 54–62

specified— at least for the respondent.) This asymmetry isreinforced by a social asymmetry every time the investi-gator occupies a higher place in the social hierarchy ordifferent types of capital, cultural capital in particular.

To reduce the effects of these asymmetries, Bourdieuproposes the practice of “active and methodical listening” inorder to incorporate the social logic that affects the constructionof data into the analysis. This listening (Bourdieu, 1993: 609),

combines a total availability to the person being questioned,submission to the singularity of a particular life history –

which can lead, by a kind of more or less controlledimitation, to adopting the interviewee's language, views,feelings, and thoughts – with methodical construction,founded on the knowledge of the objective conditionscommon to an entire social category.

Bourdieu's point is to create conditions so that researcherscan “situate themselves at the point in social space fromwhich all the respondents' views over that space emanate.”According to the author, social proximity and familiarity canhelp create the conditions for this type of listening.

In terms of the fieldwork undertaken for this research, thismeant trying to place the social proximity between theresearchers and the employers at the service of the investiga-tion. The fact that we, the researchers, are professional, urban,middle-class women with current or past experience of hiringdomestic workers undoubtedly facilitated access to the em-ployers we interviewed and enriched the information obtained.During the interviews, the interviewees shared information,points of view and experiences that they probably would nothave talked about had they not perceived our proximity to theirown class positions. In contrast, in the case of the interviewswith the workers, the possibility of our being assimilated to thesocial position of the employers led to the need to construct adegree of familiarity with the spaces through which theywere contacted, by spending longer periods of time there andmeeting repeatedly.

The racialization of poverty: towards the construction ofsocial inferiority

In 2009, domestic service accounted for the employmentof almost 14% of all female wage earners in Argentina, whichrepresents over one million workers.1 In this highly femi-nized sector, women make up 98.5% of those employed. Theeducation level of this population is lower than that of otherwage earners. In socioeconomic terms, most women whomake their living from domestic work come from sectorscategorized as poor or destitute. More than 43% of them aremigrants, of which 32.6% come from another province inArgentina, and 11% from other countries, particularly neigh-boring countries (Bolivia, Chile, Paraguay and Uruguay) andPeru. Until 2004, Argentina's migrant population fromneighboring countries was affected by legal frameworks,which restricted and hindered them from obtaining legalresidency status.2 However, in 2004, a new law came intoeffect, enabling migrants from other Mercosur countries(Brazil, Paraguay, Uruguay, and Venezuela), and also Chileand Bolivia to remain in Argentina and obtain residency

permits.3 In this context, an initiative was set in motion in2005 to regularize the legal status of those who had enteredArgentina prior to this date.4 The initiative advocatedthe authorization of temporary and permanent residencypermits and the flexibilization of requisites for residencyapplications. These changes implied that migratory status isno longer a determining factor for the working conditions ofdomestic service.

Domestic employees are among the groups with thelowest individual income levels in Argentina: in 2009, theaverage hourly wage for these workers was around 45%lower than the average salary of all other wage earners. Withregard to modes of employment, according to official data,around 72% of those employed in domestic service work for asole employer. The majority are “live-out” staff. Indeed, theproportion of domestic employees residing with theiremployers has decreased sharply in the last few years andaccounted for only three percent of those employed in thesector in 2009. Domestic work is also an area of work withone of the highest levels of informal employment, despite thefact that there has been a trend towards formalization in thelast decade: the proportion of workers listed with socialsecurity institutions has gone from five percent in 2003 to15% in 2009.

However, domestic labor is characterized by the low levelof social protection offered to those working in the sector andthe limited rights to which they have access, in comparison toother wage earners (see Gogna, 1993; Machado, 2003). Forthe past fifty years, activities connected to domestic workhave been regulated by a special regime (the DomesticService Statute, Decree 326/56), according to which domesticemployees are excluded from regular worker health andsafety law, have no access to unemployment benefits ormaternity leave, and have longer working days, shorter leaveallowances and lower severance pay than other workers. InMarch 2013, a new law regulating the occupation was passed(“Regimen Especial de Contrato de Trabajo para el Personalde Casas Particulares” [Special Contract Regulations forEmployees of Private Houses]). This law seeks to match thelabor conditions of domestic workers with those stipulatedfor general salaried workers.

These features make domestic work one of the occupationswith the least favorable labor and salary conditions on theArgentinean job market. In Argentina, as in most Latin Americancountries, it is also one of the most significant occupations forwomen from poor backgrounds (Valenzuela & Mora, 2009). AsAvila points out in the case of Brazil, domestic employees are “ledinto” their occupation by the limits imposed by class, ethnicityand patriarchy. Domestic employment is the closest option onthe horizon of possibilities for women from poor backgrounds,and presents itself as an opportunity for those who have littleformal education and whomove from the country to the city, orwho live on the outskirts of major cities (Avila, 2008: 67). Theseunfavorable aspects of the work represent a major part of theway in which the domestic workers interviewed during ourfieldwork tell of their experiences:

I had a pretty bad time there. Mymistress treated me badly;she kept telling me I was good for nothing and that I wouldnever get another job. On top of that they were really stingyabout food. I was only given one cooked meal a day, in the

Page 59: Womens Studies International Forum 46 (2014)

57D. Gorbán, A. Tizziani / Women's Studies International Forum 46 (2014) 54–62

evening, and it was always spaghetti with just a tiny bit ofolive oil, because it's very expensive, as she (her employer, awoman) used to tell me […]. I would start my job very earlyin the morning and had a one-hour break in the afternoon; Iwouldfinish around 10 or 11 at night. Sometimesmymasterarrived later, around 12, and Iwould endup going to bed at 1in the morning. On top of that they only used to pay me500 pesos under the table!5

(Lidia, 26, “live-in” domestic employee)

Such accounts that reveal acts of mistreatment anddiscrimination are not the most frequent. However, otherelements that appear in Lidia's account are also present in theexperiences of other women we interviewed, notably theconditions that characterize their occupation: never-endingwork days, low remuneration and no access to the rights andbenefits established by the existing legislation. Nor are theseconditions exclusive to live-in positions, as shown by theway Andrea narrates one of her experiences as a “live-out”domestic employee:

Later I got a “live-out” job… but I just couldn't take it becausethere were five of them. And that woman was so annoying,she was so annoying. Nothing I did was alright with her.Everything… she always found fault with something, and Iwould work like crazy from 8 in the morning to 5 in theevening. Besides, it was a lot of work; I had to cook and cleanand all that. It was hell, I'm telling you, but the paywas good.

(Andrea, 42, “live-out”domestic employee)

As in Lidia's narrative, Andrea highlights the harshness of herworking conditions. Her discourse reveals, in addition, twocentral elements of the way domestic workers talk of theirexperiences: the lack of recognition and the difficultiesunderlying the relationship with employers. Even though theexperiences she referred to as “hell”may not appear frequentlyin such accounts, they significantly influence theway employeesrepresent their occupation.

These unfavorable labor and salary conditions are closelyrelated to the undervaluing of the work domestic employeescarry out:

The work is very undervalued. I see it in my own case, youknow, the pay is bad. OK, sometimes people aren't interestedin learning, in some cases, not always, but it's like peopletake advantage of that… because someone hasn't been toschool, because they're ignorant, because they're… youknow, from another country or province. The thing is thatpeople sometimes take advantage of that. People withmoney take advantage of that: “Ah well, this one's justanother damn negrita de mierda [a commonplace racistexpression that is discussed in greater detail below], we canpay her peanuts and that's that.” And maybe they makeyou… I don't know… stay all day long, you know, and don'teven give you a glass of water. It's totally undervalued.

(Dora, 59, “live-out” domestic worker)

In Dora's narrative, low salaries and never-endingwork daysreflect the fact that, within this labor relationship, the employeris able to “take advantage” of workers. This potential exposureof domestic workers to abusive situations is a reminder ofthe fact that it is the employer who defines, almost unilaterally,

the conditions governing the labor relationship, which isestablished through individual negotiation, with little externalregulation.

The unequal positions of employers and employees in theconfiguration of the labor relationship are anchored, in Dora'sdiscourse, in the profound inequality of their social andeconomic situations, in the distance separating “people withmoney” from the “ignorant” women “from another countryor province” they hire. This description not only reflects theemployers' opinions, but it also points to the way in whichthe least protected social sectors are characterized. Theexpression that Dora uses, negrita de mierda, is commonplacein Argentina and sums up a whole range of discriminatoryopinions that are often used by those from wealthier back-grounds to describe those from poor social sectors. Roughlytranslated, it means “dark-skinned girl from the provinces orurban poor sectors.” As we discuss in the following section, thisexpression invokes a process of racialization of socioeconomicstatus, through which “working poor” and “blackness” becomeone and the same thing, to which the condition of migrant isadded, also in negative terms. This process is characteristic ofthe Argentinian context in which any differentiations bynational origin or ethnic group tend to dissolve into anall-encompassing class-based label (Briones, 2008; Grimson,2006; Margulis & Urresti, 1999).

The expression negrita de mierda thus identifies certainphysical characteristics with an inferior position, not only insocioeconomic terms but also in moral ones. In this way, ifdomestic work is “undervalued,” it is not so much because ofthe intrinsic characteristics of the activity itself, but becauseof the social features associated with those who carry it out.These features contribute to the construction of a stereotypedimage of domestic employees, the lynchpin of which is theirsocial inferiority. 6

Managing intimacy: the undervaluing of domesticworkers and emotional labor

In the employers' accounts, hiring a domestic worker is nosmall matter. Expressions such as “If Perla wasn't here I'd die”(Cami), “When she appeared it was my salvation” (Elena),and “I wouldn't be able to survive without Mariela” (Julia) arecommonplace. In these statements, the presence of thedomestic workers is described using words such as “salva-tion”, whereas their absence is associated with desperationand conflict: “I'd die” or “I wouldn't be able to survive.” Formany of those interviewed, hiring a domestic worker is notperceived as an “option,” but rather as a “necessity.” In thissense, the point at which they decide to hire someone is oftencharacterized as a breaking point or moment of crisis.However, availing themselves of domestic and care servicesis not perceived as contracting a worker in the conventionalsense, but rather as finding “help”:

When I didn't have any help I did everything myself and Ispent the whole time swearing and cursing… what abloody mess they [my children] make, why can't they justdo this, that, the other. But they didn't, and the onemaking sacrifices was always me. But then I got help and Istopped cursing.

(Elena, 50, company manager, divorced, three children)
Page 60: Womens Studies International Forum 46 (2014)

58 D. Gorbán, A. Tizziani / Women's Studies International Forum 46 (2014) 54–62

Though in some cases, the employers share a minimum ofdomestic chores betweenmale and female family members, alarge portion of these were exclusively assigned to thedomestic workers. However, it is paradoxical that the sametasks that employers described as insurmountable whenjustifying their decision to hire a domestic worker becomeless complex and important when transferred to the workerin question.

In effect, these discourses reveal a double operation inwhich tasks that are initially characterized as difficult for oneperson (the male or female employer) to carry out becomenothing more than “help” when they are undertaken by adomestic worker. The testimonies gathered all refer to thework performed by domestic employees in similar ways, inthat they do not actually describe it as being work. As such,they refer to “help,” to “someone being at home” or to“having someone,” all expressions that negate or conceal thelabor relationship, as what the person “at home” is actuallydoing there is working.

All of this reveals a very specific way of characterizingpaid domestic work; or rather, it underlines a particularfeature of existing representations of this activity: theseparation between the function being performed and theperson carrying it out, in that tasks are characterizeddifferently depending on who is performing them. In Elena'stestimony, the use of the word “help” to describe the workcarried out by her employee characterizes this work as beingsomehow incidental. At the same time, it differentiatesbetween domestic tasks within the home, and between thehelper and the person being helped. The distinction isfundamentally based on a differential valuing of domesticwork according to who is performing it. In this sense, it is notso much about the men and women who hire domesticworkers undervaluing the domestic tasks themselves, butrather the fact that they undervalue the person hired to carrythem out.

According to Romero (2002), this differential valuingprocess suggests that the nature of domestic work is notintrinsically degrading or inferior. As Romero points out, thedegrading nature of the activity arises from the interpersonalrelationship between employers and employees; specifically,the practices through which employers structure theiremployees' work in order to include issues that inferiorizethem (control over their food, the spaces they move in, theuse of uniforms, etc.). These practices reveal the way inwhich structures of domination based on class position andracial belonging permeate interpersonal relationships withindomestic labor. In effect, these degrading aspects of domesticwork are connected with the behavior that employers expectof their employees in terms of their social, racial and ethniccharacteristics. Romero (2002: 144) (among other authors)describes this expected behavior as “deferential interaction”,one of the nuclei that define the emotional work domesticemployees must provide. This author notes that in additionto physical labor, the job also implies a significant amount ofemotional labor.

This type of work is related to the way in whichemployees handle their emotions in order to respond totheir employer's psychological needs (for company, to belistened to, etc.). However, this emotional labor is notreciprocal, in that employers are not there to respond to the

psychological needs of their employees. A large part of theemotional work that is involved in this occupation consists ofthe creation of deferential behavior on the part of thedomestic workers in order to reaffirm, through theirinferiority, class and racial differences and the status of theemployer's family. In her terms, “the process that affirms thestatus of white middle-class women employers involvesdeferential interaction that treats non-white working-classdomestics as inferior” (Romero, 2002: 162).

In the previous section, we saw how Dora, whendescribing her experience as a domestic worker, emphasizedthe situation as being one of undervaluing, in which “peoplewith money take advantage.” The possibility of “takingadvantage” is, in her account, directly related to the way inwhich employers represent workers as “negritas de mierda.”The expression suggests that the differential valuation ofdomestic tasks is anchored to this inferiorizing and discrim-inatory representation of workers. It also reveals theimportance of the emotional dimension of the work carriedout by domestic employees in terms of the effort they mustmake to handle the demand for subordination that thisparticular labor relationship implies, and that is experiencedin their everyday interactions with employers.

Being migrants, being poor, being domestic workers

In employers' discourses, the bases on which representa-tions of domestic workers are built are a series of social andeconomic characteristics that are attributed to these women.One of the first such characteristics to be mobilized is that ofemployees' origins. In effect, when those interviewedreconstruct their experiences of hiring someone, the countryor province of origin of the workers frequently represents asignificant piece of information:

I've had a few people working for me. At one point, whenthe kids were very little, there was a girl who was with usfor several years, Emma, who was from Santiago delEstero… She was good, but she had to leave, she had kidsof her own, so she left. Then I went through several peopleand then came a spell with Carmen, who was Chilean, shewas a lovely lady, she stayed with us for a few years. Andthen the last person, who's been with us for more than 11or 12 years, Federica. She's a young girl, she's from EntreRíos. No, I mean Corrientes. She's from Corrientes.

(Ana, 58, professional, married, two children)

More than the domestic workers' names, it is the referenceto their country or province of origin that differentiates workersfrom one another in the employer's perspective and organizesthe employers' narrative of their experiences of hiring domesticworkers. Beyond the particular nature of their origins, thesereferences in employer testimonies underline a social charac-teristic shared by all the workers: the experience of internal orinternational migration, specifically the fact that they comefrom regions marked by critical social and economic situations.The repeated mention of domestic workers' origins performs adouble function in employer discourse. First, it reaffirms acrucial difference between the workers' origins and socialcharacteristics and their own, which are linked to theirbelonging to the urban middle classes. Second, it associates the

Page 61: Womens Studies International Forum 46 (2014)

59D. Gorbán, A. Tizziani / Women's Studies International Forum 46 (2014) 54–62

workers' origins with certain predictable behaviors or ways ofbeing:

What I've seen… is that women from Paraguay… I don'twant to categorize this as something about the national-ity. But what always happens to me is that the girlswho've come to my house from Paraguay do things likethis, they're there three or four months and then from oneday to the next they say: “I have to go becausesomething's happened in Paraguay”. I don't know, I neverknow if they're telling the truth.(Julia, 34, professional, business owner,married, two children)

In the above quote, Julia associates women from Paraguaywith a lack of stability or reliability. In other cases, womenfrom Peru are valued because they supposedly are bettereducated than other domestic workers. Regardless of wheth-er the characteristics in question are positive or negative, thereference to domestic workers' national origins helps con-struct a stereotype and define predictable behavior and waysof being.

These statements regarding workers' national origins areimplicitly permeated by ethnic and racial characterizations.However, regardless of specifically national features, within thecontext of Argentina's migration dynamic, these classificationsoverlap strongly with class distinctions. According to Grimson(2006), since the 1940s, Argentina has been characterized by aprocess of invisibilization of racial and ethnic diversity and theprimacy of a representation based on homogeneity: a “Europeanenclave”with no “black” or “indigenous” populations. Given thiscontext, the specific origins of these populations have becomeinvisible at the same time as they were being socially andpolitically incorporated into the development of import substi-tution industrialization and the rise of Peronism. Migrants fromneighboring countries were not considered as such within thiscontext. Instead, they were absorbed into the mass of cabecitasnegras (literally, “little dark heads”), a pejorative name used tostigmatize the working-class population with some indigenousancestry who moved to Argentina's urban centers, mainly fromthe provinces in the north. In this context,

any differentiation by national origin or ethnic grouptended to dissolve into an all-encompassing class-basedlabel, although this was racially marked by “darkness”.The poor were said to be “black” even though […] theywere not actually black in that they were not of Africanorigin or descent.

(Grimson, 2006: 23)

In a similar vein, the recurring reference to the national orprovincial origins of domestic workers within employerdiscourses seems to function as a powerful indicator of classdifference.

This indicator is reinforced by the representation of theworkers' places of residence. In effect, not only do domesticworkers come fromdifferent places to their employers, but theyalso reside in spaces that are far from their places of work. Thephysical distance between the employers' homes in well-offneighborhoods of the city of Buenos Aires and the places wheredomestic workers live crystallizes, in the employer's discourse,the distance between their class positions. It is a social andgeographic distance that creates difference and a hierarchy

between those involved in this labor relationship. The charac-teristics of the spaces where domestic workers live (shantytowns, slums or precarious housing) also constitute, inemployer discourse, significant references to the position ofdomestic workers in the social structure.

This stereotyped image of domestic workers that em-ployers construct through their discourse, marked as it is byprecarious social and economic situations, is not just anotherreference to the social paths of the women they hire. It is asocial, economic, and symbolic location that is associatedwith a series of features that are intrinsic to domesticworkers. The stereotype is also linked to certain predictableways of being and behaving that have important effects onthe way employers configure their interactions with theiremployees. One of the features that recur most frequently isdomestic workers' low level of formal education:

Because you even get the feeling, when you have a maid,they're generally ignorant, so it's as if I have a kind ofeducational commitment. You know, when you teachsomeone how to behave.

(Norma, 45, employer, two children)

Domestic employees' education levels are not mentionedmerely as part of their social paths, but are instead presentedas an essential feature of theirs. That is, rather than referringto the fact that women who do domestic work for a livinghave been unable to go to school, such comments designate away of being: ignorant. This intrinsic characteristic is onemore in a long list of features associated with differentaspects that define the individual, like their ways of dressingand talking, their tastes and what they consume:

With the maid I had at that point, I could buy six packetsof biscuits one day and the next there'd be none.Something was going on, I said. “No, I ate them”, she said,“I ate them all”. There was a voracity about her, youknow? […] What I mean is that it's a problem becausetheir origins mean that when they see so much food theybecome desperate for certain things.

(Julia, 36, employer, two children)

In the discourses of the employers interviewed, thesecharacteristics gradually outline the social inferiority ascribedto domestic workers, and thus play a central role in thelegitimization of the subordinated position of workers withinthe labor relationship established through domestic service. Inline with Goffman's analysis, these features seem to construct,from the employers' point of view, a “front” for theworkers thatsituates themwithin the interaction. This front ismade up of thefeatures that are identified with the performer: “As part ofpersonal front we may include: insignia of office or rank;clothing; sex, age, and racial characteristics; size and looks;postures; speech patterns; facial expressions; bodily gestures;and the like.” In general terms, front includes both “appearance”and “manner” of the individual: the former tells about theindividual's social status, the latter about the “interaction rolethe performer will expect to play” (Goffman, 2009: 24).

As Goffman points out, it is to be expected that appearanceandmanner confirm one another; that is, that the differences insocial status between performers are expressed to a certaindegree through the differences in the roles played by each in the

Page 62: Womens Studies International Forum 46 (2014)

60 D. Gorbán, A. Tizziani / Women's Studies International Forum 46 (2014) 54–62

interaction. This is the case because performances “tend toincorporate and exemplify the officially credited values of thesociety” (1990: 35), particularly forms of social stratification andclass differences. From this perspective, the stereotyped imagesof domestic workers play a central role in the way in whichemployers construct and legitimize their dominant position inthe configuration of the labor relationship. In this way, theyform part of a discourse, which, by highlighting differences insocial status between employers and employees, seeks to puteach performer in the labor relationship in her place:

Ana really knows her place. For example, Eleo, the one Ihad before, used to say things like, “I love those shoes,oooh, can I try them on?” I mean, she wanted to be likemy friend or something. Ana never says things like that,she knows her place…

(Cecilia, 38, lawyer, married, three children)

“Not knowing their place”: the dislocation of roles in theemployer–employee relationship

This set of social characteristics attributed to domesticworkers allows employers to construct and argue theirdominant position in the labor relationship via a processthat is not exempt from ambiguity and tension. In effect, ifthe social inferiority attributed to domestic workers is thecondition that enables the labor relationship, this brings withit certain threats that must be controlled and managed.

I always talk to my friends about this. I always say thatwhen they leave, they steal things, they take things fromyou, which for me is, how can I explain it… um… therelationship between the family and the maid is veryparadoxical. It's paradoxical, because it's a woman whohas a lot of needs and who sees a lot of things in yourhouse, she uses them, she sees them, because she's in yourhouse. So I really find it hard to see how a woman whocomes to my house every day and opens my children'scupboard, if she has a daughter the same age, how can shenot want to take everything? How does she manage that?

(Julia, 36, employer, two children)

One of the first references to the threat resulting from thesocial inferiority attributed to domestic workers is thepossibility of them stealing something. As previous studieshave analyzed (Brites, 2007; Kofes, 2001), the suspicion oftheft seems to be an integral part of the link betweenemployers and employees. This is frequently justified by theworkers' “humble origins” and the contrast between whatthey do not have and what they see everyday at work. Facedwith this threat, employers develop a series of practices thattend towards control over objects and assets, the definition ofspaces to which workers have access, and the supervision ofworkers as they carry out their jobs. These forms of control,which appear “naturalized” within the accounts, also operatein another sense, in that they reaffirm the stereotype of theworkers' social inferiority, as they are presented as being“self-evident” effects of inequality.

The potential for theft is not the only source of tensionposed by this labor relationship. If theft can be seen as acomponent of the position of social inferiority attributed to

domestic workers, another, more troubling, source of conflictis linked to the possibility that workers might not fit in withthe stereotyped images of them or perform the rolesattributed to them. In effect, domestic workers frequentlydo not “know their place,” as some employers put it. In hertestimony, Norma observes that “I don't like it if they're tooclever. I mean if they're too cocky, over-confident, or if theyseem to want to stand out from the rest.” The characteriza-tion of workers as “cocky” or “over-confident” refers to thosewho do not fit in with the performance of social inferiorityexpected by their employers.

As such, employees are supposed to “know their place,” toplay their part, acting out inferiority and transmitting itthrough deferential behavior towards those who are in asocially and morally superior situation within the relation-ship. Knowing your place in relation to the employer's familyimplies that the worker is capable of upholding a convenientsocial distance, the limits of which are threatened inday-to-day coexistence. This can lead, on the part of workers,to an express performance of the role attributed to them.Goffman cites an example of this sort of performance (1990:38):

The ignorant, shiftless, happy-go-lucky manner whichNegroes in the Southern states sometimes felt obliged toaffect during interactions with whites illustrates how aperformance can play up ideal values which accord to theperformer a lower position than he covertly accepts forhimself.

This performance of inferiority is interesting for the way itdisplaces employees from the position of passive agents,lacking initiative, which has often been used to characterizethem.

From this point of view, to perform the role of the employeeis to sustain a front that does not disturb what is expected ofthose who work in domestic service: fulfilling their obligationsimplies doing so in such a way as to not call the socialsuperiority of the employer into question. However, thetestimonies gathered here reveal moments in which the frontbreaks down and the hierarchical relationship is subverted. “Notknowing your place” sums up the way employers characterizethis new situation, in which the domestic worker stops fulfillingthe role that has been socially assigned to her and, through thisrupture, becomes threatening. The following account clearlyreflects this situation:

At the weekends I would go to my club, and one day Icame home at around three in the morning. I arrived andcouldn't believe my eyes; the whole living room wastotally turned upside down. I mean, the ironing board wasin the middle of the room, the television was facing thewrong way, the table had been moved. It was bizarre. Ihad no idea what was going on. I went to the maid's room,um, Lali I think her name was, I remember her… she haddark hair… She always had her hair tied back and shewore glasses thick as bottles. What a face she had… poorthing. So I went to Lali's room, I used to let her sleep thereat weekends because she had nowhere else to go. Shedidn't answer me, so I thought she must be with someman. But then a woman's voice answers and says, “Laliisn't here”. So I said, “Can you please open the door?” A

Page 63: Womens Studies International Forum 46 (2014)

61D. Gorbán, A. Tizziani / Women's Studies International Forum 46 (2014) 54–62

girl who I'd never seen in my life opened the door. So Isaid, “Where is she?”, “She's gone out dancing…”; “Andwhat are you doing here?”; “Sometimes I sleep here at theweekend.” I… well… it was like I went away at theweekend and my house was… I felt awful, really awful, itwas so invasive, so… so… they had taken over somethingof mine… it was… horrible. I had a window that lookedonto the street, it was three or four in the morning and Istood at the window saying, I'm going to kill Lali. Butalways in the sense of giving her a good telling off, I'mgoing to kill her… out of sheer anger…

As if you were telling off…?…a child. And suddenly I see Lalicoming up Avenida de los Incas, without her glasses. Imean… and her hair was all… you know? When someonelets their hair down and is radiant, beautiful. And it wasme… my shoes, my jeans, my jumper, my jacket, myhandbag. No, no…well… I swear you had to be there to seeit. I couldn't believe it [laughs]. It was… She looked amazing,dressed up as me!

(Norma, 45, employer, two children)

Within Norma's account, Lali, stripped of the centralcomponents of her front that make up the deferentialbehavior expected of domestic workers, reveals her capacityto be like her employer. The image the worker offers to theperson who has hired her, dressed in her clothes, without herglasses, with her hair down, “radiant, beautiful”, is the imageof an attractive woman and not of an inferiorized “child.”That image challenges the superior position in which theemployer has placed herself and reveals the constructednature of the differentiation between the two and thepossibility of discovering, in the other woman, an equal.When Lali is surprised by Norma, the employer's anger is notonly due to the way her employee has used her house, butalso for having transgressed the limits set out for her. Instead,her anger is mainly connected to the fact that Lali isdiscovered in a position of equality with those who hireher, which subverts the labor relationship.

Returning to the categories used in Romero's analysis,emotional work becomes central to creating the deferencethat confirms and upholds the employer's status. As we sawin this example, this status depends not only on theemployee's social class, but also on her racial and ethnicorigins and her physical appearance. The threat emerges withthe possibility that the worker might stop carrying out theemotional work that sustains the hierarchies implied in paiddomestic work. These processes of constructing hierarchiesare the condition for the existence of this labor relationship,and they reproduce and update themselves in the everydayinteractions between the parties involved in this relationship.

By way of closure

Throughout this text, we have gradually uncovered differentoperations through which employers construct the socialinferiority that permeates the way in which they representdomestic workers and manage their relationship with them. Aswe have seen, this process of inferiorization includes, in turn,different practices of control through which employers “handle

the threat” they suppose is implied by the presence of a workerin the intimacy of their homes. In effect, this is a laborrelationship that has been made invisible, in which theconnection between the two parties is defined by asymmetry,which is reinforced in the way the relationship plays outeveryday.

The workers are not recognized as such, but are insteaddescribed as “the maid,” “the girl who helps,” and the personwho “is at home.” These ways of referring to domestic workersembody, in turn, certain features that make up the stereotypeswithinwhich thosewhodo the job for a living seem to, ormust,fit. As we have pointed out, these stereotypes reveal a processof social, economic and symbolic positioning associated withcertainways of being and behavingwhichmake up the sociallyexpected role which those working in domestic service areexpected to fulfill. This role locates employees in an inferiorposition to their employers in the social structure and supposesthat “appearance” and “manner” confirm one another. In otherwords, the very role of domestic employees supposes that theymust act deferentially, “knowing their place” — namely, that of“maid.” From the employer's point of view, “knowing yourplace” becomes a central feature when evaluating those whodo domestic service for a living, as it supposes the upholding ofconvenient social distances.

These dynamics reveal mechanisms of constructing socialhierarchies that are not only present in domestic employ-ment, but also configure it as such. Through the way theyhandle the link with workers, employers put into practice thedynamic and conflict-ridden operations that have beenanalyzed throughout this article in terms of how theydevalue and inferiorize employees. One of the main sourcesof tension in these processes concerns each party's ability tofit in with established roles. As we observed in the finalexample analyzed, when the behavior and image of workersfall outside the parameters expected of their role, this candisturb the connection between the parties, as it implies adislocation of the roles socially assigned to each of them. IfNorma was so surprised by Lali's transformation, it is becausewhat she discovered through the scene was the fragility ofthe symbolic struggle unfolding at the heart of this relation-ship due to the negotiation of class-related positions. In short,what is revealed when the person who is perceived associally inferior does not seem to “know their place” is theprecariousness of the social construction of inequality as anessential feature of inter-class relations.

Endnotes

1 All statistical data presented in this paragraph and the next comesfrom the report Caracterización del servicio doméstico en la Argentina [ACharacterization of Domestic Service in Argentina], created by the Sub-secretariat of Technical Programming and Labor Studies of the Ministry ofLabor, Employment, and Social Security (cited as Contartese, 2010).

2 In the first instance, this took the form of the Agreement on Residencyfor Nationals from Mercosur member states, Bolivia, and Chile, signed inDecember 2002. In turn, in 2003, a new Immigration Law, no. 25.871, waspassed, “which implied a change of direction in policy discourse byincorporating two new features: a human rights perspective and a regionalfocus” (Pacecca & Courtis, 2008: 43). This law “establishes the right tomigration as a human right and incorporates the right to the reunification ofthe family” (Pacecca & Courtis, 2008: 45). It also mentions the state'sresponsibility for ensuring that all foreigners legally residing in Argentinaare treated equally and recognize the unrestricted right of access toeducation and healthcare, regardless of migratory status.

Page 64: Womens Studies International Forum 46 (2014)

62 D. Gorbán, A. Tizziani / Women's Studies International Forum 46 (2014) 54–62

3 This refers to the National Program of Migratory Document Normal-ization for nationals of Mercosur member and associate states, whichincludes immigrants from Bolivia, Brazil, Chile, Colombia, Ecuador, Paraguay,Peru, Uruguay, and the Bolivarian Republic of Venezuela. The program isknown as Patria Grande, which translates roughly as “the greater homeland.”For an analysis of the program, see Gallinati (2008) and Pacecca and Courtis(2008).

4 This working experience corresponds to 2005 and 2006. In the latteryear, the minimum wage earned by live-in domestic workers was 650 pesos.

5 The origins of these stereotypes can be traced back to the socialtransformations of the twentieth century in Argentina, specifically thoseaffecting the conformation of the middle classes. Various studies haveexplored these changes and the transformation of the models of domesticityassociated with them, including Adamovsky (2009), Pite (2011), Pérez(2012) and Cárdenas (1986).

6 There is extensive literature concerning the concept of emotionallabor. Developed by the sociologist Arlie Hochschild, this concept facilitatesthe analysis of certain occupations that require the worker to produce anemotional state in another person through the manipulation and control oftheir emotions (1983). Although Hochschild did not develop this concept inconnection to domestic work, it has been widely used in this field of study.

References

Adamovsky, Ezequiel (2009). Historia de la clase media: Apogeo y decadenciade una ilusión, 1919–2003. Buenos Aires: Planeta.

Anderson, Bridget (2000). Doing the dirty work? The global politics of domesticlabour. London and New York: Zed Books.

Avila, Bethania (2008). Algumas questões teóricas e políticas sobre empregodoméstico. In Bethania Avila, Milena Prado, Tereza Souza, & Vera Soaresy Verônica Ferreira (Eds.), Reflexões feministas sobre informalidade etrabalho doméstico (pp. 63–71). Recife: SOS CORPO.

Bourdieu, Pierre (1993). The weight of the world. Stanford: University ofCalifornia Press.

Briones, Claudia (2008). Formaciones de alteridad: Contextos globales,procesos nacionales y provinciales. In Claudia Briones (Ed.), CartografíasArgentinas. Políticas indigenistas y formaciones provinciales de alteridad(pp. 9–38). Buenos Aires: Antropofagia.

Brites, Jurema (2001). Afeto, desigualdade e rebeldia: Bastidores do serviçodoméstico. Tese de Doutorado, Programa de Pós-Graduação em AntropologiaSocial. Porto Alegre: UFRGS.

Brites, Jurema (2007). Afeto e desigualdade: Gênero, geração e classe entreempregadas domésticas e seus empregadores. Cadernos Pagu, 29, 91–109.

Cárdenas, Isabel (1986). Ramona y el robo: El servicio doméstico en barriosprestigiosos de Buenos Aires (1895–1985). Buenos Aires: Ediciones Búsqueda.

Chaney, Elsa, & García Castro, Mary (Eds.). (1993). Muchacha, cachifa, criada,empleada, empregadinha, sirvienta y…más nada. Caracas: Editorial NuevaSociedad.

Contartese, Daniel (2010). Caracterización del servicio doméstico en laArgentina. Buenos Aires: Subsecretaría de Programación Técnica yEstudios Laborales del Ministerio de Trabajo, Empleo y Seguridad Social.

Ehrenreich, Barbara, & Hochschild, Arlie Russell (2002). Global woman: Nannies,maids, and sex workers in the new economy. New York: Metropolitan Books.

Gallinati, Carla (2008). ¿Inmigrantes o ciudadanos? La construcción del“verdadero” modelo de política migratoria en el marco de la “patriagrande”. Revista Pensares, 5, 479–500.

Goffman, Erving (2009). The presentation of the self in everyday life. London:Penguin.

Gogna, Monica (1993). Empleadas domésticas en Buenos Aires. In Elsa Chaney, &Mary García Castro (Eds.),Muchacha, cachifa, criada, empleada, empregadinha,sirvienta y más nada (pp. 81–98). Caracas: Editorial Nueva Sociedad.

Grimson, Alejandro (2006). Nuevas xenofobias, nuevas políticas étnicas enArgentina. In Alejandro Grimson, & Elizabeth Jelin (Eds.), Migracionesregionales hacia la Argentina: diferencias, desigualdad, derechos (pp. 22–37).Buenos Aires: Prometeo.

Gutiérrez Rodríguez, Encarnación (2010). Migration, domestic work and affect.London: Routledge.

Hondagneu-Sotelo, Pierrette (2007). Domestica: Immigrant workers cleaningand caring in the shadows of affluence. Berkeley: University of CaliforniaPress.

Ibos, Caroline (2012). Qui gardera nos enfants? Les nounous et les mères. Paris:Flammarion.

Kofes, Suely (2001).Mulher Mulheres: Identidade, diferençia e desigualdade narelaçao entre patroas e empregadas. Campinas, SP: Editora da UNICAMP.

Lan, Pei-Chia (2003). They have more money but I speak better English!:Transnational encounters between Filipina domestics and Taiwaneseemployers. Identities: Global Studies in Culture and Power, 10, 132–161.

Lan, Pei-Chia (2008). Migrant women's bodies as boundary markers:Reproductive crisis and sexual control in the ethnic frontiers of Taiwan.Signs: Journal of Women in Culture and Society, 33(4), 833–861.

Lutz, Helma (2002). At your service madame! The globalization of domesticservice. Feminist Review, 70, 89–104.

Lutz, Helma (Ed.). (2008). Migration and domestic work: A Europeanperspective on a global theme. Hampshire: Ashgate.

Machado, José Daniel (2003). Acceso al ámbito de protección del decreto326/56 para trabajadores del servicio doméstico. Revista de DerechoLaboral, 2, 277–317.

Margulis, Mario, & Urresti, Marcelo (1999). La segregación negada: Cultura ydiscriminación social. Buenos Aires: Biblos.

Pacecca, Maria Inés, & Courtis, Corina (2008). Inmigración contemporáneaen Argentina: Dinámicas y políticas. CEPAL Serie Población y DesarrolloNo. 84. Santiago de Chile: CEPAL United Nations.

Parreñas, Rhacel Salazar (2001). Servants of globalization: Women, migration,domestic work. Stanford, CA: Stanford University Press.

Pereyra, Francisca (2012). La regulación de las condiciones laborales de lastrabajadoras domésticas en Argentina. In Valeria Esquivel, Eleonor Faur,& Elizabeth Jelín (Eds.), Las lógicas del cuidado infantil: Entre las familias,el Estado y el mercado (pp. 163–196). Buenos Aires: IDES — UNICEF —UNPFA.

Pérez, Inés (2012). El hogar tecnificado: Familias, género y vida cotidiana(1940–1970). Buenos Aires: Biblos.

Pite, Rebekah (2011). Entertaining inequalities: Doña Petrona, Juanita Bordoy,and domestic work in mid-twentieth century Argentina. Hispanic AmericanHistorical Review, 91(1), 97–138.

Rollins, Judith (1985). Between women: Domestics and their employers.Philadelphia: Temple University Press.

Romero, Mary (2002). Maid in USA. New York and London: Routledge.Valenzuela, María Elena, & Mora, Claudia (Eds.). (2009). Trabajo doméstico:

Un largo camino hacia el trabajo decente. Geneva: International LabourOrganization.

Page 65: Womens Studies International Forum 46 (2014)

Women's Studies International Forum 46 (2014) 63–71

Contents lists available at ScienceDirect

Women's Studies International Forum

j ourna l homepage: www.e lsev ie r .com/ locate /ws i f

Domestic service, affection and inequality: Elements of subalternity

Jurema BritesUniversidade Federal de Santa Maria, Campus Camobi - Av. Roraima 1000, Prédio 74C, Sala 202 - CEP, 97 105 900 Santa Maria, RS, Brazil

a r t i c l e i n f o

http://dx.doi.org/10.1016/j.wsif.2014.03.0090277-5395/© 2014 Elsevier Ltd. All rights reserved.

s y n o p s i s

Available online 29 April 2014

In this paper I intend to present an ethnographic description of the movement of things,people, and affection in the context of domestic service in Brazil. Looking at everydayinteractions, I explore the sociological dimensions (family organization, gender relations, andclass structure) and the symbolic constructions (concepts of motherhood, childcare,reciprocity, care, and affection) as well as the political and infra-political dimensions(domination, subordination, and rebellion) of domestic service in order to better understandthe elements at play in the Brazilian context.

© 2014 Elsevier Ltd. All rights reserved.

Introduction

Domestic work1 provides a good starting point for ananalysis of Brazil. It has existed since the early years ofBrazilian society and was based on the exploitation of slavelabor during the colonial period. Although the country'sconstitution now recognizes domestic work as a profession,and domestic workers enjoy the same rights as otherworkers, the sector is still predominantly composed ofwomen who are poor, mostly Black and have little schooling.My aim in this paper is to unravel the social configurationsthat contribute to maintaining domestic work as a space ofthe reproduction of inequalities.

Employing ethnographic research methods, I describe, onthe one hand, the differences in the daily organization ofworkers' families and employers' families, suggesting that astratified complementarity makes domestic work functionalfor both parties. On the other hand, taking the feelingsexpressed by the people I researched, I would suggest thatthe inequalities reproduced in the course of domestic serviceare sustained largely through affective ambiguities. I draw aconnection between the accusations of theft against domesticworkers and the possibility of extra-wage payments commonin these situations, because both are agreements that occurbehind the scenes. I investigate these elements in terms ofthe transmission of a certain heritage in which things andvalues circulate, serving to maintain the relations between

employers and their employees as spaces of subordination,rebellion, and resistance.

This study is the result of intensive ethnographic experienceover the past ten years in Brazil, including episodic research inthe homes of employers, twomonths of live-in fieldwork in thehome of a domestic worker in Vitória (Espírito Santo), anddaily involvement in the domestic workers' syndicates in Juizde Fora (Minas Gerais) and Porto Alegre (Rio Grande do Sul),in addition to over one hundred semi-structured interviewscomposed of open-ended questions with domestic workers inthese three states.

Following an analytical tradition typical for contemporary(and post-colonial) anthropology, I insist on two points. Thefirst concerns situated knowledge, demanding that theresearcher assumes the limited vantage point implied in hisor her particular position within race/class hierarchies (Brah,1996; Spivak, 1994; Strathern, 2006). Like nearly all otherresearchers in Brazil, and the majority of people who are notconsidered poor, I have extensive personal experience as anemployer of domestic workers, and this fact –which involvesa structural inequality in the relation between researcher andinformant – should not be omitted.

Nonetheless, the second point concerns the insistencethat our partners in debate are sophisticated, multidimen-sional characters who are not so easily “dominated” by thedominant classes. Members of so-called subordinate groupshave any number of ways of, at least symbolically (and,

Page 66: Womens Studies International Forum 46 (2014)

64 J. Brites / Women's Studies International Forum 46 (2014) 63–71

in certain circumstances, politically), reshaping the powerrelations that permeate their existences (Ortner, 1995; Scott,1990; Thompson, 1998). My article aims to highlight thiscomplex creativity among domestic workers in Brazil.

Although residing away frommyownhome, at the house ofdomestic workers, sharing their everyday life, struggling tomaintain non-hierarchical relationships, my social standing asa “boss” was hardly ever forgotten. I was offered many con-cessions on a daily basis, for example, choosing themenu of theday or sitting in the only chair of the house. I was, however,expected to pay for my privileged space. People asked me tobuy all sorts of things: new showers, lice medicine, or anythingthat could be sold to help enhance their income. At any rate,precisely by not hiding that I was a wealthier person, I enjoyedmuch intimacy and friendship. I heard confessions, secrets, andcomplaints. I read and discussedmy final composition with theresearch subjects collectively on a number of occasions. Eventhough they could read, it was difficult for them to understandan academic manuscript. With key informants, I negotiatedhow some issues would appear in the research. It was hard, forinstance, to persuade them to change their real names. Theywanted their stories to be recognized and their names printedin the text. However, we reached a deal that would not putthem in excessive exposure, and to compensate for the absenceof their names, I organized photographic exhibitions that dealtwith their life stories so their reputation would be safeguarded(Brites, 2000).

In this article, my paradoxical position in the field isespecially apparent when I describe theft. I suppose thatbeing able to listen to the domestic workers' stories and thencomparing them with the situations/versions within my ownsocial networkwas advantageous for showing the usefulness ofClifford Geertz's proposition: “Finding our feet, an unnervingbusiness which never more than distantly succeeds, is whatethnographic research consists of as a personal experience”(Geertz, 1973: 21).

Domestic service and “stratified complementarity”

In the Brazilian middle and upper classes, family relationsare routinely permeated by the presence of domestic workerswho normally take care of all the housework as well aschildcare. In the performance of these tasks, just as in thewages and relationships that go along with them, one canobserve the reproduction of a highly stratified system ofgender, class, and race (Azeredo, 1989). The domestic andfamily organizations of employers, as well as the chances ofeconomic improvement and social prestige for them and theirfamilies as awhole are based on a division of tasks that involvesanother woman (generally not a relative). This second womanmay have other notions of family, male/female and mother/child relations, but eventually these notions adjust as theygrow to complement those of the employer. By observing dailyinteractions, practices, and values in these different settings, Ibegan to realize that the different patterns of family organiza-tionweremutually reinforcing, interacting in a sort of stratifiedcomplementarity. By this, I mean the imbrication of socialpatterns deriving from different sorts of lifestyles, familyorganizations, and values that, although based in differentpositions within a social hierarchy, act in a complementaryfashion, reproducing social inequalities.

My analysis is largely inspired by Colen's (1995) studies onCaribbean nannies and domestic workers in the United States.Colen coined the concept of stratified reproduction to describehow “reproductive” tasks have been distributed according tohierarchies of class, race, ethnic group, and gender. The authorobserves that relations of political inequality and economicexploitation within various forms of family organization andgender experiences end up being functional for both parties. InBrazil, the inequalities between employers and domesticworkers are evident. Differences in class, ethnicity, and accessto consumer goods, education, and better jobs are recognizedand broadly discussed within specialized literature (Goldstein,2003; Mori, Bernardino-Costa, & Fleischer, 2011; OrganizaçãoInternacional do Trabalho, 2013).

Furthermore, the hierarchical system underlying domesticservice has been bolstered, in particular, by emotional ambigu-ities in the relation between employers – especially children andwomen – and domestic workers (Goldstein, 2003).

In thenegotiationof extra-wagepayments, in the exchangeofservices completely outside any contract, in the gossip betweenwomen, and in the relations between workers and children, it isimpossible not to recognize the existence of a large amount ofaffection. This fact, however, does not impede a hierarchicalrelationship, with clear demarcations between employers andsubordinates, i.e. between those who can buy domestic servicesand thosewho, by offering their services,manage to access one ofthe less severe alternatives of survival in Brazil.

Domestic workers are usually considered the country'spoorest-paid women. They have little formal education, aremainly migrants, and their cultures and ethnicities arestigmatized by the hegemonic system of values. Accordingto official data from 2008, 61% of the domestic workers inBrazil are Afrodescendants, 38.2% are Caucasian, and 0.4% areIndigenous (Fraga, 2010). There are very few studies on theemploying families (Brites & Picanço, 2013).

According to the IBGE (Instituto Brasileiro de Geografia eEstatística (IBGE), 2009), domestic service in Brazil represents17% of female employment, constituting the second largestprofessional category in the country. This is a massively femaleactivity — about 92.4% of domestic workers are women. Thiscategory also registers very high levels of informality. Only 26%are formally employed, versus 58% of other workers, and 27%earn less than the minimum wage. In April 2013, the BrazilianCongress granted domestic workers the same rights as otherworkers. However, changes in the law, although desirable, donot signify an immediate end to inequalities and prejudices.Domestic servicemay not be considered a preferred occupationin the spectrum of career choices for workers. However, whenother options for entering the labor market are beyond reach,domestic service appears as a ready alternative for work on theinformal labor market.

Nevertheless, one of my ambitions is to show that, despitethe negative aspects of domestic service, infra-politicalrelations2 between employers and workers, though at timesundemocratic, still make this activity interesting for workers.This was one of my most bewildering discoveries duringfieldwork.

The domestic workers I listened to found advantages inthose elements of domestic service that feminist intellectualswho study work and gender tend to denounce as instrumentsof subjugation: extra-wage payment and the possibility of

Page 67: Womens Studies International Forum 46 (2014)

65J. Brites / Women's Studies International Forum 46 (2014) 63–71

negotiating time off, exchanges resulting from personal andclient relations. Trying to understand the afore-mentionedattitude, as anthropologists do, from a native point of view, Ifound meanings that went well beyond the patriarchal andcapitalist domination suggested in much of the literature onthis theme.

Since I myself am an employer, following the anthropo-logical tradition of thinking “the other,” I chose to emphasizethe domestic worker's point of view. This option, togetherwith the use of ethnographic research methods, leads tointeresting considerations on the reproduction of inequalityin Brazilian society, which I have resumed in three key issues:

1. Differences between employers and employees in terms offamily organization and financial income end up generat-ing a stratified complementarity, which justifies thecontinuity of paid domestic work in contemporary society;

2. In the family dynamics typical of working-class groups,3

domestic service often turns out to be more advantageousthan other forms of employment.

3. Despite the experience of subordination, it is possible toobserve domestic workers (in the workplace as much as intheir own families) deploying practices and tactics that allowthem to obtain profit from highly unfavorable situations.

The employer's family

In the middle-class families included in the sample,domestic workers were expected to clean the house and takecare of the children, the elderly, and the family pets. These taskswere to be carried out in a discreet and affectionate manner,allowing the adultmembers of the sample families to engage intheir full-time jobs, usually outside the home.

In general, in these families, the mother, besides having afull-time job, was in charge of health care, hygiene, and homedecoration. She was also largely responsible for maintaining andmanaging the affections of the family's wider network ofsociability.4 The husband took responsibility for the greater partof family expenses, assuring investment in school and socialcareers. He was assigned a few domestic tasks, such as groceryshopping, taking the children to school, or conducting minorrepairs in the house or on the car. Therewere no domestic choresfor children and youngsters, especially if they were boys. Ingeneral, they had their days entirely filled with school andadditional courses in English, Mathematics, Music, Dance, andSports.5

The domestic organization of this group corresponds tothat of the modern nuclear family model, anchored inmarriage, with the creation and promotion of children asthe most important function of the family enterprise. In thesefamilies, the “home” is valued as an ideal locus of expressingintimacy and developing mature and healthy emotionalstructures for the members of the home (Ariès, 1981;Donzelot, 1977; Fonseca, 1995).

According to Duarte and Dias (1986), these families endup having increased visibility in Brazilian society, because –

being made up of intellectual and consumer elites – theypromote their own patterns of organization and values,which reflect the standards desirable within a “normal,”“healthy,” and “democratic” society.

The domestic worker's family — interdependenceand antagonism

The families of domestic workers researched are, unliketheir bosses' families, generally extended families, in whichprimary attachments are established with relatives, notnecessarily within the couple. These are families character-ized by high rates of female-headed households, consensualunions (instead of marriage), and children in circulation(fosterage). In these groups, the responsibility for children isnot restricted to the biological parents, but may be shared byseveral homes.

Men are expected to be the providers in the poorestfamilies, but this role is hardly consistent with the precariousemployment and sporadic income available to this population.Marital unions are easily undone, often lasting no more than afew months or years. There is a high rate of recomposedfamilies, with children being distributed around the kinshipand neighborhood network when the mother remarries.6

Men are relevant as sons and siblings. Paternal ties aremaintained by the presence of female paternal relatives(particularly grandmothers and aunts) in children's lives(Fonseca, 1985). Fathers are rarely present in the home, but,since male social spheres and manhood itself are forged in the“street,” this absence is not considered entirely strange. In likemanner, notions of womanhood include relegating issuesrelated to domestic and familiar territory to the female domain(Duarte & Dias, 1986). It is women, in matrifocal organizations,who represent the multi-generational family group andtransmit the cultural assets of the group.

Some pages of my fieldwork diary, though a bit extensive,may illustrate with more vivacity the descriptions above:

Edilene's house, where I originally thought I was going tobe staying, was too small: only one bedroom, a kitchen,and a porch. I was travelling with my two-year-old sonand, although I was open to the ethnographic meeting, Iassumed sharing the same room with Edilene and herhusband would go beyond the boundaries of my culturalindividual necessity. For different reasons than mine,Edilene approached me with another solution. Withoutgiving up the prestige of hosting me, she offered the bestquarters of her domestic group — a room at her mother-in-law's house, Claudina, who lived next door. In a shorttime, I found the ability of having me under her mother-in-law's roof was symptomatic of the social organizationof that place.Although there were walls that separated the houses, thegroup's everyday life indicated a common family organiza-tion, not only because theywere side by side, but – from thefamily history to themost casual everyday occurrence – alsoshowed an interdependence between residents of differenthouseholds.Themain character of this groupwas Claudina (54 years oldat the time), domestic worker at the same household for23 years. She was born in Bahia and had “worked on aplantation since she was very young.” She had ninechildren, of which only three survived. 30 years ago shemoved to Vitória, along with a sister, and both beganworking as domestic workers. Of an ill-fated relationshipwith a married man, Edinha, the youngest daughter,

Page 68: Womens Studies International Forum 46 (2014)

66 J. Brites / Women's Studies International Forum 46 (2014) 63–71

was born. After a few years of living in Vitória, she metNorberto — a retired night watchman and her currentpartner.Mr. Norberto, along with a nephew of Claudina's, was oneof the first invaders of Jardim Veneza, the neighborhoodwhere they live. After Claudina established herself therewith Norberto, she asked her two married sons to sharethe land. Today three houses have been built, wherethe solutions of a domestic group overlap its nuclearorganization.The mother and her youngest daughter live on one side ofthe land, the eldest son (Tonho) at the other end. Betweenthem there is the house of the middle child (Clodoaldo).With the old couple resides Michele, granddaughter ofClaudina, a seven-year-old girl whose mother, starting anew marriage, decided to leave the daughter with her ex-mother-in-law. Edinha (16 years old), Claudina's youngestchild, who left hermother's house sixmonths agowhen sheannounced she was pregnant, lives at the back of the land,along with her husband.In the middle is Clodoaldo and Beatriz's “barraco” (smallwooden hut) along with their three children (six, five, andtwo years of age). Clodoaldo currently joins his brother on ajob as a bricklayer at Praia Velha. The couple gets up veryearly— especially Beatriz, whoworks as a domestic workerand walks four kilometers each day to her employer'shouse, so she can exchange her transportation vouchers formilk for the kids.On herway towork she leaves her two youngest children atthe only municipal day care center in the neighborhood,which is open only part-time. The oldest child, who is sixyears old, stays at home alone until two o'clock when heand his cousin Michele both go to the same day care center.In a third part of the land lives Tonho (Claudina's firstborn)and his wife Edilene. Tonho was responsible for theconstruction of his mother's beautiful house. As he waswithout a regular job for a long time, the son repaid hismother's efforts by building the house from the ground onlywith the help of old Norberto.Edilene is sometimes disrespected by her relatives and herhusband because she drinks too much, but she is the onewho feeds the whole family group with small leftoversshe brings from her employer's house, sappy fruits andvegetables tradesfolk discard from their stores and pig fator lard she exchanges with the neighbor's son for somemarijuana cigarettes.Among these three houses, the transit of people, services,and objects is incessant, creating tensions as well ascomplicities (Brites, 2000).

Similar patterns of family organization appear in histor-ical accounts of Medieval and Modern Europe (Ariès, 1981),as well as studies concerned with colonial Brazil (Samara,1989). There is evidence of similar family forms in manynon-Western societies as well. Hence, rather than seeingthese arrangements as a momentary breakdown of normalitydue to economic precariousness, one might think of them interms of reproducing traditions in subordinate groups.Nonetheless, it would be naive to think purely in terms ofcultural and unilateral determinations. It is obvious thefamily forms I studied are part of a social system within a

weak state, occurring in a context of material deprivationwhere child labor, and not schools, has long been children'sprincipal mode of access to education (Fonseca, 1999).

The usefulness of the concept of stratified complemen-tarity can be observed here. The co-existence of two familymodels – one for the domestic workers and the other for theiremployers – operates to guarantee the promotion of womenand families among the better-off employers, while main-taining the subordination of the more disadvantaged domes-tic workers.

On one hand, it is not difficult to explain why domesticworkers readily take care of their employers' children withdedication and even affection: they come from social groupswith a cultural tradition of family organization in which thesocialization of children goes well beyond the four walls of thenuclear family, where children are often left in the care of otherfemale relatives or neighbors. On the other hand, with thistype of collaboration, middle-class women easily achieve higheducational and professional accomplishmentswhile guarantee-ing the maintenance of the status quo of their group.

Affection and inequality

In almost every employer's home I investigated, children,when not in school, would spend the majority of their timewith the domestic workers. The parents, away at work,would leave their children with the domestic worker fortwo-thirds of the day. The intensity of contact betweenchildren and their domestic workers created, in varioussituations, a bond that went beyond professional ties. I foundphotographs of employers' children in the domestic workers'personal photo albums, often near the pictures of their ownfamily (a mother in her coffin, a wedding portrait, somesnapshots of brothers and nephews).

When domestic workers are in their own homes, afterworkhours, they can hardly stop talking about the achievements ofthe children they care for: what the little girl said, what thelittle boy did, etc. Their neighbors and family appear to knowallabout these children (birthdays, favorite clothes, etc.). It'salmost as if theywere part of the domestic worker's family. It iscommon – even once the employment contract has ended – forthe domestic workers to continue to follow the lives of thechildren they have taken care of. Occasionally, they will call totalk to them, consult a colleaguewho isworking for someone inthe social network of their former boss, or simply calculatetheir ages, from a distance, remembering their birthdays orkeeping the children's photographs in their albums. In thiscontext, an employment change canmean a huge affective loss.In fact, one reason that employees tend to endure poorly paidjobs is the difficulty of separating from the children they carefor.

There are similar signs of children being attached to “their”domestic workers. One employer, for example, reportedhow her son fell ill when their nanny had to quit her job. Iheard variations on this theme in countless stories, highlight-ing an intensity of contact that is not without importantconsequences.

The following conversation is very revealing of theambiguity of affection and distance present in these relation-ships, which I will further elaborate on in the next section:

Page 69: Womens Studies International Forum 46 (2014)

67J. Brites / Women's Studies International Forum 46 (2014) 63–71

Edilene, a domestic worker, told me, with evident pride,what her boss's daughter a five-year-old girl, said to her:_Lene, you could win the lottery, right? Then you couldcome around just to play with me. You could have lunchand lie in mommy's bed to rest, like she does. (Edilenecloses the story saying) Lie in her bed! What an idea!

Subtle distance

The question I would like to pose is: if there is suchintimacy and affection between children and their domesticworkers, how do these children grow up to be adults whoreproduce such rigid hierarchies? At what point do theirworlds separate from one another?

The employers I investigated do not, as a rule, treat theiremployees rudely. Children apprehend the social distancebetween themselves and the domestic workers throughother channels — subliminal information, their parents'words, and the organization of domestic space.

“The maid's room,” “the maid's bathroom,” and “themaid's quarters”7 are segregated areas where the respectmiddle-class children are taught regarding other people'spossessions disappears. The spaces reserved for domesticworkers in employers' homes do not respect the workers'individuality. They can be full of rubbish, brooms, buckets,and anything else that is useless or should remain out ofsight, so as not to disturb the beauty and order of the home.Another ambiguous dimension is the domestic worker'sbody. This woman can hold a baby in her arms, prepare themeals for the family, clean the house, and wash the clothes,and though it is not frequent, in some cases, she may evenhave to sexually satisfy her male boss.8 But it is out of placefor her to sit on the living room sofa, lie on her employers'bed, dine at the table with them, or use their bathrooms. Inthese cases the domestic worker's body pollutes because itis not performing servile activities and could strain thehierarchical order of the home.

Heritage transmission

A walk through the rooms of the domestic workers' homesgenerally reveals a huge amount of furniture and appliancesthat previously belonged to their employers: old clothes,furniture, mattresses, windows, and toys. Many scholars havenoted this kind of donation, referring to it as a form ofexploitation used by employers to supplement or replace partof the wage paid to the employee (Chaney & Castro, 1993). Onthe other hand, I propose looking at this exchange of goodsand services – an exchange that goes along with most paiddomestic work – as a sort of patrimonial heritage. Thisperspective has the advantage of understanding the processas something that goes beyond the narrow sense of moneyrelations to include the idea of a communication system inwhich, besidesmaterial things, socialmeanings are transmitted(Douglas & Isherwood, 1996). To treat this “transit” of assets as“patrimonial heritage” (Neves, 1982) signifies that somethingbesides material things is being exchanged in this relationship(Mauss, 1967). Objects “do not exist autonomously.” As“material, physical and immediately concrete support ofproduction and reproduction of social life,” they should be

considered “products and vectors of social relations” (Menezesapud Magni, 1994: 11).

Besides wages, employers give their employees objects,but only those that are of no further use to them. They rarelybuy new things to give to their domestic workers. Domesticworkers only get what their employers consider useless: oldbras and panties, clothes that are too small for the children, astained blouse, a used refrigerator, a discarded decoration,leftovers from Sunday lunch, etc. One might wonder justwhat this donation of “second-hand things” represents.

The logic of the social relation between donor and recipientvaries according to the things that are given. One of themessages conveyed through gift-giving is social hierarchy.According to this ideology, socially inferior people receivesecond-hand things: they are “second-class people.” Tradesalso happen within the family, but they depend on the type ofobject passed on and especially on its state of conservation.Consider family heirlooms, for example, whose spirit, accord-ing to the Polynesian concept of hau, is in the object passeddown over generations. It is unimaginable that a domesticworker would “inherit” the wall clock that belonged tograndmother. Furthermore, one does not offer old things,which would otherwise be thrown away, to a “superior.”

The gift given to a domestic worker signifies the donor'sprecedence/excellence over the recipient (Mauss, 1967). Inthe objects offered by employers, there is a definite message:I used it first, I sat on it first, and I ate it first. You can have myleftovers. The place of things reflects social standings. Just asthe domestic worker occupies residual spaces in the em-ployers' homes, the things she receives are also “remainders.”The logical inference here is that, in the relationship betweendonor and recipient, the person who gives away second-handthings has a higher standing in the hierarchy.

The butler did it

During my fieldwork, I came to understand that theftsattributed to domestic workers present an interesting perspec-tive on the tensions that permeate domestic service. In thetransit of things given– aswell as those allegedly stolen – I couldsee a particular mix of affection and antagonism, which reflectsand strengthens the unequal relations of power. I treated thisflow of objects, furniture, clothes, and food, transferred fromonehome to another, as “patrimonial heritage.”

In the same way that, in their oral narratives, employershighlight the generous gifts given to their domestic workers,they often complain about the domestic workers' pettypilfering. In this not entirely consensual patrimonial heritage,one encounters, so to speak, the underside of the giftexchange. Consistently described as a surprise (“Can youimagine? I caught her robbing9 me!”), as something inad-missible, theft, as a rule, triggers indignation from employersand is, many times, what leads to dismissal.

However, no domestic worker ever confessed to stealing.Theft appears in their stories, more often than not, as anunfounded accusation made by the employers. At times theydo admit theft occurrences, but it appears as somethingcommitted by other domestic workers. The thief is alwayssomeone at a considerable distance, so that no suspicionsmight fall on the narrator and her network of friends. Tulia,who has 43 years of experience in housework, and is a

Page 70: Womens Studies International Forum 46 (2014)

68 J. Brites / Women's Studies International Forum 46 (2014) 63–71

central figure in her neighborhood network (even acting as asort of informal employment agency), admitted to only onecase in which she was accused of stealing.

Tulia: _It happened to me once. I was working for Ms.Norma and her ring went missing: My ring is missing, myring is missing! It can't have vanished just like that! Sheshouted this and other such things at me. We alreadyknew they were suspicious of us. I worked with anothergirl, Osmarina, the cook. You see, there was such a ruckusuntil we found that ring! I, who had more access to theclosets, spent a full day searching for it. It was in thepocket of a jacket Ms. Norma had used. She had forgottenthat she put it there and [so she decided] it was “themaid” who robbed it, you know?Jurema: _ And what happened later?Tulia:_ I told her that before she accused us, she shouldfirst look for it very well. Then, if she didn't find it, shecould pressure us on the matter (sic).

The main point here is not to determine whether theworker is telling the truth, but to realize how much domesticworkers recognize this sort of behavior as an ordinary fact, beit a potential accusation or an admission of the possibility ofstealing. These narratives not only indicate a discursiveinstance, but they are also clues for understanding culturalconceptions. Tulia's narrative is a vivid indication of herwisdom regarding the social rules that apply to suchsituations. First, she reveals that the confirmed disappear-ance of something generally triggers accusations against thehousehold's subalterns. Second, the accusation leads to aninvestigation, because her boss – as many do – announces themissing indirectly, thereby giving time for employees to“find” whatever has disappeared. Third, Tulia reports findingthe object, which, according to her, was dropped somewhereby her employers. Thus, she points to her employers'sloppiness or neglect, perhaps due to the little importancethey really give to their belongings (which they know can bereplaced with relative ease). Perhaps the employers rely toomuch on the domestic workers for maintaining order. There-fore, she can also easily exempt herself of any possibility ofsuspicion. Finally, Tulia admits that domestic workers oftenpilfer and, in this case, the employers have the right to demandback what belongs to them.

Even though the lost object is often found or discovered tobe in use by someone in the family, this “detail” is notincorporated into the employers' repertoire as a reference forthe next event. The underlying belief is that domesticworkers obviously steal. Examples of this conviction appearnot only in the fieldwork, but also in everyday Brazilianaffairs. At no time is the possibility of “theft” implied betweenfamily members; it is always attributed to “a stranger.”

Maria Suely Kofes (1991) analyzed the accusations of theftagainst domestic workers as an effect of their indeterminatestatus in their employers' homes. According to the author, thenature of the work performed by domestic workers within thedomestic space is ambiguous, as the home is the locus of familylife par excellence, “organized by kinship relations of affection,trust, fidelity and intimacy” (Kofes, 1991: 236). The accusationof theft against domestic workers in this context, theanthropologist suggests, has the symbolic effect of removing

them from the family relations. It is interesting to think of thedomestic worker as a dangerous element, a “polluter,” in thewords of Mary Douglas (1976). However, placing thesecomplaints in a communicative context of knowledge andvalues, it might be reasonable to suggest that these situationsand accusations of theft represent a chance for dialog betweenemployers and employees. It might be possible to see the fearof theft attributed to the poor as a simple imputation of danger,impurity, or criminal pathology. On the other hand, theassurance that “it was the maid” could be result of a tacitacknowledgment of the extreme inequality that separatesdomestic workers from employers. If, in the eyes of employers,domestic workers are likely to pilfer, is it not for lack of basicnecessities?

This fact leads me to believe that theft committed bydomestic workers is not only expected, but also tacitlyaccepted, because a situation of theft, even when verified,seldom leads to legal or police intervention. Why, despite theaccusations of theft attributed to Tulia, has she continued towork at the same job for 23 years? Why, when faced withcuriously empty shelves, do employers complain indirectly,make insinuations, but rarely decide on a radical charge?

Carrier ants: rebellion, rivalry and fun in theft

My ventures in fieldwork (particularly in workers'homes), as well as my experience as an employer, have ledme to believe that the following anecdote – about the missingclothes of Ms. Dina (Levi-Strauss's wife) – might not be purecolonialist fantasy.

The anthropologist in Saudades do Brasil (Lévi-Strauss,1994) reports on the unusual privileges of being a youngprofessor in Brazil where he could afford the services of adomestic worker. He writes that he unfortunately had to lether go, because she had the lousy habit of “borrowing” Ms.Dina's clothing for Carnival balls.

My impression is that it is not unusual for a domesticworker to take things without permission from her employer'shome. Nevertheless, when this occurs, the objects are generallyof insignificant value, at least in terms of the employers'purchasing power: a can of peas, a bar of soap, uncookedkidney beans, grocery bags, and spare change. Sometimesdomestic workers temporarily borrow something (a dress for aspecial occasion, panties, or bras) that employers watchdisappear and come back, after making insinuations about themissing object. So it would be naive to think that theft occurssimply for reasons of survival. How then are we to understandthis frequently observed activity that brings to mind thetireless work of carrier ants?

In my field notes, I registered the narrative of theft told byan employer (Maria da Penha) to her manicurist (Manoela):

Manoela:_ Did you know that Edilene, ‘the maid,’ wasscheming to rob things with Tulia? Both of them!Everything was all arranged! Maria da Penha – theemployer's daughter – went to the kitchen after lunch.Edilene had already cleaned the kitchen and she waswashing some clothes. Maria da Penha found bananas inthe garbage bag. Of course, since Tulia is the one whotakes the garbage out, she was passing the stolen goodson to the other, got it? Maria da Penha, playing the fool,

Page 71: Womens Studies International Forum 46 (2014)

69J. Brites / Women's Studies International Forum 46 (2014) 63–71

asked Edilene: _Why are you going to throw these goodbananas out? She said the maid blushed and came upwith the lame excuse that she was taking them for Tulia'slunch._Hey, Edilene, you know Mom doesn't like it when youtake things without asking. Why don't you serve a mealand call Tulia to eat here?Jurema: _ And what happened then, did they fire her?Manoela: _ No. But Maria da Penha hates ‘the maid,’ right?She doesn't know how her mother stands such insultingbehavior. She said she wanted to slap [Edilene] in the face.

The revenge of nemesis: the performative expression ofrelations between employers and domestic workers

These narrative examples of accusation suggest that,together with asymmetric authority relations, the accusationsmade by bosses and foreseen by domestic workers point to acommunication process between the parties. Both partiesinvolved in the conflict develop together the grammar involvedin accusations of theft, almost in a performative ritual, whereboth the actors and the audience already know the script of theaffair. A code of speech, gestures, and practices that, althoughclearly manifest, is never explicit, but is rather a silentagreement between the parties about power relations.

In this shared knowledge lies the notion of a hierarchicsociety, in which the possibility of justice is less committed tothe equality of individuals than to the relation of reciprocitybetween those of unequal status. That is, the asymmetry ofrights and obligations is not questioned — it simply exists, asa fact of life. It is because of this inequality that the charge oftheft can be so easily made against subalterns. However, thisinequality is not a simple relation of oppression between therulers and the ruled. There is space for negotiation in whichtheft committed by subalterns is a predictable action,inherent in the relation of social inequality.

When Tulia's boss sets a time limit for the object to befound, when Tulia admits the possibility that some domesticworker might have picked up the ring, and when employerssee pilfering as part of the relationship between employers andemployees, they are all recognizing the logic of gift circulationbetween the involved parties. In these terms, pilfering could beinterpreted as a sort of “revenge of nemesis,” a moral principleexpressed in the gift-exchange scheme proposed by MarcelMauss, denouncing the imbalance between the abundance ofsome and the poverty of others:

“Alms are the result on the one hand of a moral idea aboutgifts and wealth and on the other of an idea about sacrifice.Generosity is necessary because otherwise Nemesis willtake vengeance upon the excessivewealth and happiness ofthe rich by giving to the poor and the gods. It is the old giftmorality raised to the position of a principle of justice; thegods and spirits consent that the portion reserved for themand destroyed in useless sacrifice should go to the poor andthe children” (Mauss, 1967: 15–16).

Duringmy fieldwork, the subject of theft had been extremelyrecurrent in employers' stories (and confirmed by domesticworkers). Nonetheless, “theft” has rarely been analyzed in social

sciences literature. Specialized literature about domestic servicein Latin America briefly mentions the pilfering issue, connectingit to the discussion of the adverse and unfair conditions faced bydomestic workers. In these analyses, “theft” assumes only oneconnotation: accusations by employers against their domesticworkers (Chaney & Castro, 1993; Kofes, 1991).

Zaluar (1985) proposes demonstrating how representa-tions of the world of crime, violence, and power amonglower-income families in Rio de Janeiro are constructed fromtheir own concrete, everyday experiences “instead of anabstract idea of justice or democracy” (1985: 140). Theseexperiences take place within a complex process of interactionbetween workers and outlaws involving gender, age, location,and honor codes that goes beyond amere reaction to dominantmorality. However, when examining the occasional “theft” or“pilfering” (even citing the increased domestic theft “well-known to housewives in the wealthy classes,” Zaluar, 1985,footnote 12), the author refers the discussion to the concept ofclass and to “the revolt” resulting from the progressiveimpoverishment to which the working-class Brazilian popula-tion has been submitted.

If, on one hand, Zaluar (1985) draws attention to a typicaldimension of working class actions, on the other, it lends itselfto a sort of economic reductionism already criticized by E.P.Thompson. The historian (1998) comments on the limitationsof analyses that view the insurrections of eighteenth-centuryEngland asmere “stomach rebellions.”According to the author,the riots should be analyzed as a form of popular action basedon custom. Thompson places these eventswithin the context ofclass relations in eighteenth-century England, and the guidinginfluence of traditional plebeian moral economy. According tohis reading, the riots occurred in reaction to transformationscaused by the incipient capitalist production being introducedwithin all kinds of commercial transactions. Working class“resistance” is not a “given,” it emerges in particular situationsof confrontation with the aristocracy and the new modelof bourgeoisie. By analyzing social relations during a periodof transition, the author perceives new meanings withinworking-class action that were previously treated as the mereresult of the mechanics of misery.

Following this line of thought, theft and accusations of theftare analyzed here as constituent parts of domestic workrelations. Besides wages and gifts offered by employers totheir employees, the continuous acts of theft that are suppos-edly committed by domestic workers illustrate the ambiguitiesof class relations and the ambivalences that this type of serviceprovokes in Brazilian society. The examination of anecdotesabout theft collected during the fieldwork heightens thecomplexity of the notion of class “resistance,” pointing to ahistorical configuration of the actions of subordinate popula-tions that are not fully inclined to submit to the dominantgroups, yet cannot do without them. Theft (as well asextra-wage income given as gifts) opens up a communicationalfield between classes and, in this sense, may be thought of as aperformative expression, reflecting a pedagogical space ofpower relations in the country.

The recycling practices associated with “carrier ant”behavior – the pilfering of objects in employers' homes –

are much like the “direct appropriation of assets,” of“splinters,” “patches,” and “leftovers,” which, according toLinebaugh (1983), constituted the traditional system of

Page 72: Womens Studies International Forum 46 (2014)

70 J. Brites / Women's Studies International Forum 46 (2014) 63–71

domestic production in the eighteenth century. As stated bythe same author, when the monetary system was introducedinto the logic of the production system, workers did noteasily adhere to the work process that intensified their ownexploitation. In order to wipe out popular practices ofsupplementary income, a severe punishment system wasdeveloped that brought thousands of apprentices to thegallows. Scott (1987,1990) – undoubtedly an intellectual heirof Thompson – helps read these events as a struggle for thedemarcation of boundaries between “public scripts” and“occult scripts,” in which subalterns take advantage of thegaps in a system of well-defined behaviors in order to plantheir moves. Their strategies – generally with a specificobjective – for getting some scraps from the abundance of therich rarely take the form of direct confrontation or suicidalrebellion. Instead, they assume a manner of superficialcompliance and false deference, and, with great skill, try tomake their superiors bite their own tongues, holding themto their own implicit promises. Here, it is necessary toremember that, in this tense game, important elements are atplay, such as the fact that domestic workers know muchabout their employer's intimate lives and their possible moralmissteps. Nevertheless, these trumps are not enough toensure a sure victory for subordinates. These are cards that, inorder to be effective, must be played with a great deal ofwisdom, winning minor victories or, at times, just a laugh.

Conclusion

In this article, I have ethnographically described therelations that are enacted between employers and domesticworkers in Brazil in terms of a circulation of values, symbols,and belongings. Far from supposing a symmetrical circulation, Iattempted to demonstrate how, in the flow of things (gifts andminor theft), of people (the workers that leave their neighbor-hoods to work in the homes of their employers, as well as thepractice of the circulation of children), and also of affections(the love these workers have for their employers' children, butalso the vengeance, the suffering, and false accusations) areelements that complement one another as a result of thestratified reproduction of different lifestyles of employers anddomestic workers.

Domestic work has always been present in Brazilian societyin a subaltern manner. In former times, it was performed byslaves, and Black and indigenous women. Nowadays, paiddomestic labor is carried out by a great number of womenbelonging to lower-income segments of society and to ethnicminorities who have few opportunities for formal education.Despite some legal achievements, domestic labor persists as alocus of precarious working rights, a lack of social acknowl-edgment and the stigmatization of the women employed inthis sector.

Based on the ethnographical research carried out in bothemployers' and workers' homes, I described the intimacy offamily life in order to show how differences in lifestyle,economic power, and values are articulated in ways thatreproduce inequalities.

I attempted to demonstrate how the division of repro-ductive tasks sustains the social standing and lifestyle ofthe employers. These practices are surrounded, in turn, bydifferent conceptions of childcare and maternity between

social classes, which end up generating inequalities thatpersist in Brazilian society.

Even though different conceptions and ways of life areoperative for both parts, the relationships are not always welladjusted or harmonious. Tacit consent about inequality ismaintained by redistributive performances as seen in thepractices of gift-giving and extra-salary payments. Behind thedeference demanded by their bosses, the workers developmany forms of resistance, aiming to attain small gains in anunfavorable situation.

Endnotes

1 In this article I refer to domestic work as reproductive activitiesperformed by anyone within the household. The term domestic servicedistinguishes the domestic work that is hired and/or paid for.

2 According to James Scott (1990) infra-politics are everyday forms ofresistance.

3 In Brazil, there is a wide debate on the definition of subordinate groupsas working class, poor, or popular groups for which no parameters werefound in the Anglo-Saxon discussion. See the discussion in Brazil: Fonseca(2005).

4 In this aspect, middle-class family organization in Brazil does not seema lot different from the kin work usually attributed to the North-Americanhousewife, as described by Di Leonardo (1992).

5 See Tânia Salem (1980) for a description of task distribution accordingto gender and generation in middle-class Brazilian families. For a perspectiveon the experiences of the middle-class Brazilian family, see Gilberto Velho(1989), Tânia Salem (1986) and Heilborn (1993).

6 Brazil admits a large diversity of family organizations according toclass, age, race, and gender perspectives (Berquó & Oliveira, 1992; Costa,2002). Legal marriage is equally common in the various sub-groups. A largepart of the Brazilian population, in general those from the poorest classes,marks the beginning of marital life through consensual unions and not legalmarriage. Legal marriages generally happen more often in the middle andupper classes.

7 I used the term “maid” in a few parts of the text to highlight thecultural Brazilian reality whose nomination of domestic workers denotessocial inequalities and a notion of a hierarchical distance.

8 Testimonies of sexual and amorous relationships between domesticworkers and their male employers are not uncommon in Latin Americanliterature, for example, in “El Amor en los Tiempos de Cólera” by GabrielGarcia Márquez. In Brazilian social sciences, one example is “Casa Grande eSenzala” by Gilberto Freyre. Felicie Drouilleau (2011) has recently dedicatedherself to writing about the situation in Bogotá.

9 The Portuguese language distinguishes “robbery” (roubo) from “theft”(furto) by defining the first as the unlawful taking of the property throughthe use of violence. Ordinarily, in common everyday speech, this differenceis not noted. My informants would always refer to “robbery” in talking aboutany sort of theft in their bosses' homes — without relating these incidents tothe use of violence.

References

Ariès, Philippe (1981). História Social da Criança e da Família. Rio de Janeiro:Zahar.

Azeredo, Sandra (1989). Relações entre empregadas e patroas: reflexõessobre o feminismo em países multirraciais. In Albertina Costa, & C.Bruschini (Eds.), Rebeldia e Submissão. Estudo sobre a condição feminina(pp. 195–220). São Paulo: Vértice.

Berquó, Elza, & Oliveira, Marcelle Colares (1992). Casamento em Tempos deCrise. Campinas, 9(2), 138–154.

Brah, Avtar (1996). Cartographies of diaspora: Contesting identities. London/New York: Routledge.

Brites, Jurema (2000). Afeto, Desigualdade e Rebeldia: bastidores do serviçodoméstico. Tese de Doutorado. Porto Alegre: UFRGS/PPGAS.

Brites, Jurema, & Picanço, Felice (2013). O emprego doméstico em números,tensões e contradições: alguns achados de pesquisas. 37° EncontroAnual da ANPOCS, Águas de Lindóia. Online: http://www.anpocs.org/portal/index.php?option=com_docman&task=doc_view&gid=8638&Itemid=429

Page 73: Womens Studies International Forum 46 (2014)

71J. Brites / Women's Studies International Forum 46 (2014) 63–71

Chaney, Elsa, & Castro, Mary Garcia (1993). Muchacha/cachifa/criada/empleada/empregadinha/sirvienta y… más nada: trabajadoras domésticasen América Latina y Caribe. Venezuela: Ed. EPU.

Colen, Shelle (1995). Like a mother to them: Stratified reproduction andWest Indian childcare workers and employers in New York. In F.Ginsburg, & R. Rapp (Eds.), Conceiving the new world order: The globalpolitics at reproduction (pp. 78–102). Berkeley: University CaliforniaPress.

Costa, Carolina (2002). Pirâmide da solidão ou pirâmide dos não casados? Core estado conjugal na terceira idade no Brasil. XIII Encontro da AssociaçãoBrasileira de Estudos Populacionais, Ouro PretoAnais do XIII Encontro daAssociação Brasileira de Estudos Populacionais. (pp. 1–26). Belo Horizonte:ABEP, 1–26.

Di Leonardo, Micaela (1992). The female world of cards and holidays:Women, families, and the work of kinship. In N. Thorne, & M. Yalom(Eds.), Rethinking the family: Some feminist questions (pp. 246–261).Boston: Northeastern University Press.

Donzelot, Jacques (1977). A Polícia das famílias. Rio de Janeiro: Graal.Douglas, Mary (1976). Pureza e perigo. São Paulo: Perspectiva.Douglas, Mary, & Isherwood, Baron (1996). The world of goods — Towards

anthropology of consumption. London and New York: Routledge.Drouilleau, Felicie (2011). Parenté et domesticité à Bogotá, Thèse de doctorat

en Anthropologie Sociale et Ethnologie. Toulouse: École des Hautes Étudesen Sciences Sociales.

Duarte, Luis, & Dias, Fernando (1986). Da vida nervosa nas classestrabalhadoras urbanas. Rio de Janeiro: Jorge Zahar.

Fonseca, Cláudia (1985). Valeur marchande, amour maternel et survie:aspects de la circulation des enfants dans un bidonville brésilienne.Annales-ESC, Paris, 40(5), 991–1022.

Fonseca, Claudia (1995). Amor e família: vacas sagradas de nossa época. InIvete Ribeiro, & Ana Clara Torres Ribeiro (Eds.), Família em processoscontemporâneos: inovações culturais na sociedade brasileira (pp. 69–89).São Paulo: Loyola.

Fonseca, Cláudia (1999). Quando cada caso não é um caso: pesquisaetnográfica e educação. Revista Brasileira de Educação, São Paulo, 10,58–78.

Fonseca, Cláudia (2005). La Clase y su Recusación. Etnográfica. EtnografíasContemporáneas. Buenos Aires, 1(1), 117–138.

Fraga, Alexandre (2010). De empregada a diarista: as novas configurações dotrabalho doméstico remunerado. Rio de Janeiro: UFRJ/IFCS.

Geertz, Clifford (1973). The interpretation of cultures. New York: Basic Books.Goldstein, Donna (2003). The aesthetics of domination: Class, culture, and the

lives of domestic workers. Laughter out of place: Race, class and sexualityin a Rio. Shanytown, Berkeley: University of California Press.

Heilborn, Maria Luiza (1993). Visão de mundo e ethos de camadas médiassuburbanas. Caxambu: Communication in VII ANPOCS.

Instituto Brasileiro de Geografia e Estatística (IBGE) (2009). Pesquisa nacionalpor amostra de domicílios. Rio de Janeiro: IBGE. Set/2011. http://www.ibge.gov.br/home/estatistica/populacao/trabalhoerendimento/pnad2009/pnad_sintese_2009.pdf

Kofes, Maria Sueli (1991). Mulher: mulheres. Diferença e identidade nasarmadilhas da igualdade e desigualdade - interação e relação entre patroase empregadas domésticas. São Paulo: Tese de Doutorado. São Paulo: USP -Faculdade de Filosofia, Letras e Ciências Humanas.

Lévi-Strauss, Claude (1994). Saudades do Brasil. São Paulo: Companhia dasLetras.

Linebaugh, Peter (1983). Crime e Industrialização: a Grã-Bretanha no séculoXVIII (pp. 102–141). In P. S. Pinheiro (Ed.), Crime, Violência e Poder. SãoPaulo: Brasiliense.

Magni, ClaudiaTurra (1994). Nomadismo Urbano: uma etnografia sobremoradores de rua em Porto Alegre. Dissertação. Programa de Pós-Graduação em Antropologia Social - Mestrado - UFRGS.

Mauss, Marcel (1967). The gift forms and functions of exchange in archaicsocieties. New York: W Norton & Company.

Mori, Natalia, Bernardino-Costa, Joaze, & Fleischer, Soraya (2011). Tensões eExperiências: um retrato das trabalhadoras domésticas de Brasília eSalvador.

Neves, Delma (1982). Nesse terreiro galo não canta: estudo de carátermatrifocal de unidades familiares de baixa renda. VI Encontro Anual daANPOCS. Caxambu: GT Família e Sociedade.

Organização Internacional do Trabalho (2013). Domestic workers across theworld: Global and regional statistics and the extent of legal protection.http://www.ilo.org./wcmsp5/groups/public/-dgreports/-dcomm/publ/do

Ortner, Sherry (1995). Resistence and the problem of ethnographic refusal.Comparative Studies in Society and History, 37(1), 173–193.

Salem, Tânia (1980). O velho e o novo: um estudo de papéis e conflitosfamiliares. Petrópolis: Vozes.

Salem, Tânia (1986). Família em Camadas Médias: uma perspectivaantropológica. BIB, 21, 1–80.

Samara, Eni (1989). As mulheres, o poder e a família: São Paulo século XIX. SãoPaulo: Marco Zero/Secretaria de Estado da Cultura de São Paulo.

Scott, James (1990). Domination and the arts of resistance: Hidden transcripts.New Haven: Yale University Press.

Scott, James (1987). Weapons of the near. Everyday forms of peasantresistance. New Haven: Yale University Press.

Spivak, Gayatri (1994). Quem reivindica a alteridade? In HeloisaBuarquedeHollanda (Ed.), Tendências e impasses: o feminismo comocritica da cultura. Rio de Janeiro: Ed. Rocco.

Strathern, Marilyn (2006). O gênero da dádiva: problemas com as mulheres eproblemas com a sociedade na Melanésia. Unicamp: Campinas.

Thompson, Edward Palmer (1998). Costumes em comum: estudos sobre acultura popular tradicional. São Paulo: Companhia das Letras.

Velho, Gilberto (1989). Subjetividade e sociedade: uma experiência de geração.Rio de Janeiro: Zahar.

Zaluar, Alba (1985). A Máquina e a revolta: as organizações populares e osignificado da pobreza. São Paulo: Brasiliense.

Page 74: Womens Studies International Forum 46 (2014)

Women's Studies International Forum 46 (2014) 72–80

Contents lists available at ScienceDirect

Women's Studies International Forum

j ourna l homepage: www.e lsev ie r .com/ locate /ws i f

Intersectionality and female domestic workers' unions in Brazil

Joaze Bernardino-CostaDepartment of Sociology, University of Brasília, Campus Universitário Darcy Ribeiro, ICC Centro, Asa Norte, CEP. 70.910-900 Brasília/DF, Brazil

a r t i c l e i n f o

0277-5395/$ – see front matter © 2014 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.wsif.2014.01.004

s y n o p s i s

Available online 29 January 2014

The paper uses the concept of intersectionality to explore the central role played by the categories ofrace, class and gender in the biographies of female domestic workers in Brazil. While showing howthese categories are implicated in the inequalities and subalternization experienced by theseactors, the paper also reveals how female domestic workers have appropriated them to promotethemselves politically as a professional class. Adopting a historical viewpoint, the second part of thepaper shows the formation of a public agenda for female domestic workers' unions and theirnegotiation with class-based, feminist and black movements in Brazil. It concludes by showing thatunionized domestic workers have developed an original form of feminism that combines aspectstaken from all these movements.

© 2014 Elsevier Ltd. All rights reserved.

Introduction

Domestic labor in Brazil is a symbol of gender, class andrace inequalities, as the majority of the domestic workers inBrazil are black lower-class women. According to a recentcensus, 7.2 million people are professional domestic workers,93% of whom are female; 61.6% of those females are blackand 38.4% are white. The over-representation of black femaledomestic workers can get even more evident: 12% of whitewomenwith a job are domestic workers; the rates increase to21% for black women (IPEA, 2011).

The existence of a domestic labor forcemeans that there arehigh-income families with the means to pay another person'swages. On the other hand, it means a service that compensatesthe lack of basic public service (daycare, for instance). Because ofthat, families with a higher income are able to overcome the lackof some public services by privatizing them, by hiring privateservices. Such economic inequalities are connected to bothgender and race naturalization. Domestic labor is naturally seenas a woman's job and, as such, not worthy of a fair pay, as itsupposedly does not involve special skills. On the other hand,due to Brazil's colonial history, domestic labor is also seen as theblack woman's ‘natural place.’

One of the most evident consequences of those families'private solutions for the lack of public services is the fragilestate regulation concerning female domestic workers and their

ll rights reserved.

employers. Indeed, female domestic workers have recentlyenjoyed legal equalization compared to other jobs. However,there isn't reliable supervision for domestic labor. As a con-sequence, only one-third of the domestic workers in Brazil workunder legal conditions. Therefore, the relationship betweendomestic workers and their employers shows not only noncom-pliance of rights, but also a naturalized code of conduct in whichgender, class and race inequalities make domestic workerssusceptible to disempowerment and a violation of rights.

Inside their employers' home, a fragile state regulationsets the relationship between the female domestic workerand their employers. In the public sphere, on the other hand,their unions actively struggle for better state regulation.Therefore, the female domestic worker unions have tried foryears to create a partnership with the black movement, thefeminist movement and other class unions in order toimprove their rights. In this case, the axes of gender, raceand class power, unlike what happens inside their em-ployers' home, mobilize for democracy and the domesticworkers' empowerment.

This paper aims to explore two dimensions of domesticlabor: the face-to-face relationship between the domesticworker and their employer, which happens inside theemployers' home, and the female domestic workers' politicalmobilization through their unions. Our analysis is based on theconcept of intersectionality. As we will demonstrate, this

Page 75: Womens Studies International Forum 46 (2014)

73J. Bernardino-Costa / Women's Studies International Forum 46 (2014) 72–80

concept is helpful in understanding the interwoven relation-ships of diverse axes of power, themost important of which arerace, class and gender, in the production of both subjugation andpolitical agency. This approach is effective for comprehendingdomestic labor in Brazil, as it takes us beyond discourses thatisolate individual markers of difference by dynamically joiningthem.We focus on how the axes of gender, race and class poweract in that face-to-face relationship which disempowers thefemale domestic worker. On the other hand, we explore howthose same axes of gender, race and class power acted as a toolfor empowerment and democratic mobilization through thedomestic workers' political organizations. Concerning the latter,such axes of power are mobilized from exchanges among thedomestic workers' political organizations and the feminist andthe black movements, as well as their syndicates. As a result ofthis exchange, the female domestic workers have succeeded insome aspects concerning their profession's regulations.

This article is divided in five parts. The first part describesour research methodologies. The following part shows thetheories about the concept of intersectionality. It is good tohighlight that such concept may refer to either disempow-erment or empowerment. The third and fourth parts useempirical data and the concept of intersectionality to explainhow the axes of gender, race and class power interact, on theone hand, to trigger disempowerment in the female domesticworkers' workplace and, on the other hand, to trigger politicalmobilization through their unions. Finally, the last part bringsour final considerations.

Methodology: Listening to domestic workers' voices

Based on themain role of the concept of intersectionality, thisarticle aims to study how gender, race and class dimensionswork in the private sphere, and how it causes disempowermentand inequality; and, as far as the public sphere is concerned, howthey result in empowerment and democratic mobilization.

The data on which this research is based on was collectedin two different periods, all from unionized domestic workers. In2006, as part of a research for my doctorate study, I carriedout semi-structured interviews with twenty-three unionizedworkers from five out of approximately forty existent unions inBrazil. Those unions were based in Campinas (São Paulo state),São Paulo (São Paulo state), Salvador (Bahia state), Recife(Pernambuco state) and Rio de Janeiro (Rio de Janeiro state). Ialso interviewed the leaders of the Domestic Workers NationalFederation anddid some research in the files of each one of thoseunions, as well as in documents and resolutions of the domesticworkers' national congresses.1 During my research, I traced thehistory of the Santos Domestic Workers Association, which wasthe first political association of female domestic workers inBrazil, founded in 1936. In 2011, I interviewed five domesticworkers of the unions of the New Iguaçu district (Rio de Janeirostate) and the Franca district (São Paulo state). Those sevenunions, especially the five from my 2006 research, have beenin charge of the organization of the domestic workers' nationalcongresses. They are considered the core of the domesticworkers' movement not only because of that, but also becausethey give to us a historical glimpse into the domestic workers'political organizations (Bernardino-Costa, 2007, 2011).

The interviews were carried out in each one of the above-mentioned unions and they took 90 min in average. All

interviewees were asked questions about their backgrounds,first job, union affiliation, their union's political struggles,experience exchanges and cooperation with the black andfeminist movements, as well as with other class unions,among others. Interestingly, all of them filled out the consentand confidentiality terms of the University of Brasília. In2011, however, during my presentation as a special guest inthe 10th National Congress of Domestic Worker, in Recife, theinterviewed domestic workers asked me to list their realnames inmy articleswhen I was tomention the history of theirpolitical organization; their anonymity was to be maintainedonly when I was to mention a specific personal history. Thisarticle, thus, follows their request.

From themethodological point of view, one of the guidelinesin this article is the use of what Enrique Dussel (1996) calledethical listening. That means to ethically recognize the existenceof the ones who were faded and silenced by the hegemonicepisteme. By listening to the domestic workers, we tried toundermine the power of the hegemonic speech in the Braziliansociety, especially about black women, who are viewed by thehegemonic episteme as deprived of rationality and thereforeunable to tell their own histories.

Therefore, as we listened to the interviewees, we intendedto deconstruct a Brazilian belief of harmonybetweenwhite andblack people, and between rich and poor. The idea of Braziliansociety as one that forged harmonious relationships, especiallybetween whites and blacks, has been propagated over thecourse of the nation's history. We question this perspective;not only does it obscure diverse forms of violence, but it alsofails to “ethically listen” to both the black and poor populationsof the country. In contrast to such social and racial harmonyspeech, the private sphere of the Brazilian society shows theaxes of gender, race and class power moving towards thedisempowerment of the female domestic worker. On the otherhand, as we listen to the domestic workers telling their storiesabout their political organization, we do not notice a process ofvictimization, but resistance to oppression, exploitation andthe suppression caused by the country's hegemonic speech.

To analyze the relevant data, our arguments were struc-tured around the concept of intersectionality.

A brief discussion of the concept of intersectionality

Especially in the political and academic fields of gender andrace studies, the concept of intersectionality has been employedto stress the interconnections between certain categories,including race, gender, class, generation and sexuality.

Two aspects stand out in the use of intersectionality as aconcept: on the one hand, the homogenizing dimension ofvarious categories such as class, race, sexuality and genderhas been questioned in favor of highlighting intra-class,intra-gender, intra-sex and intra-race differences. On theother hand, rather than utilizing an additive logic in whichthe axes of subordination simply compound one other, asin the idea of double or triple discrimination, the concept ofintersectionality focuses on the interaction between two ormore of these axes of power (Brah, 1996, 2006; Brah & Phoenix,2004; Carneiro, 2003a, 2003b; Collins, 2000; Crenshaw, 2002,2006; McClintock, 1995; Yuval-Davis, 2012).

In the 1990s, Kimberlé Crenshaw, in a dialog with blackfeminists on the supposed universalism of “woman” as a

Page 76: Womens Studies International Forum 46 (2014)

74 J. Bernardino-Costa / Women's Studies International Forum 46 (2014) 72–80

political category, was largely responsible for popularizingthe concept of intersectionality, and she herself became anindispensable reference point in discussions of the topic. In herwork, intersectionality is used to describe the way in whichracism, patriarchal relations, class oppression and other axes ofpower generate discrimination and inequalities. Crenshaw(2002, 2006) emphasizes how the intersectionality of race,class, sexuality and gender are responsible for oppression anddisempowerment. The metaphor of a crossroads allows us tounderstandwhat the authormeans by this concept. The axes ofpower – race, gender, sexuality and class – overlap and crosseach other. Crenshaw (2002: 177) writes: “racialized womenfrequently find themselves in a position where racism orxenophobia, class and gender meet each other. Consequently,they are liable to be hurt by the intense flux on these roads.” Aperson subject to intersectionality, following the author'smetaphor, is like a pedestrian at a crossroads, suffering thedamage caused by collisions from multiple directions. Hence,the concept utilized by Crenshaw shows the disadvantages,vulnerabilities, oppressions and disempowerment suffered bywomen situated at the meeting point of two or more axes ofpower. Crenshaw's examples of sexual and domestic violence,sexual harassment, discrimination and inequalities in terms ofaccess to jobs and education reveal precisely the dimensions ofoppression and disempowerment implied by the concept ofintersectionality (2002, 2006).

Crenshaw notes that the dynamics of the axes of power –

race, class, gender, sexuality, nationality, etc. – are not unilateralin the sense of generating only oppression, since members ofmarginalized groups are able to resist and mobilize politicalwill, individual and collective. The intersectionality approach isshared by Patricia Hill Collins (2000: 13), who proceeds from thenotion of a dialectic between oppression and activism.

Another author of significant contribution to the discus-sion of intersectionality is Avtar Brah (1996, 2006). Whiledrawing attention to the dynamics of oppression, discrimi-nation and exploration in power relations, Brah highlightsthe importance placed on the dimensions of activism andpolitical mobilization in intersectionality. Depending on thehistorical context, the difference informing the notion ofintersectionality often leads to more democratic forms ofpolitical agency (2006:16). To understand Brah's position, itis important to keep in mind that she researched Britain'sblack movements in the 1960s and 1970s and became awareof how the term ‘black’ was employed as a political categoryto mobilize and unite people with African, Caribbean andSouth Asian origins in response to their shared experience ofstigmatization, inferiorization and discrimination in work,education, housing, the legal system and welfare. Initiallytaken as a category of subalternization, the concept of‘blackness’was later used to mobilize colonial subjects whowould never have been previously identified with the term.Through Brah's research, we become aware that racemay havediverse meanings according to context.

In considering class, race, sexuality and gender as axes ofpower, it behooves us to recall Foucault's insights concerningpower. Power is not a property, but a relation. Power relationsalter continually, new conflicts and new points of resistancespring up all the time, leading to the emergence of new subjects(Foucault, 1995). Hence, depending on the context, the notion ofintersectionality can be utilized not only to examine negative

effects like oppression and disempowerment, but also to explorepolitical mobilization.

The concept of intersectionality has the advantage ofallowing us to view two dimensions of power relations: onthe one hand, the production of disempowerment, oppressionand discrimination; on the other, the production of politicalagencies, democratic mobilization and political subjects.

The use of this concept also affords us an analysis based onthe interaction of a multitude of axes of differentiation andpower. This concept emphasizes that different dimensions ofsocial life cannot be analyzed discretely, in search of absolutistexplanations about the processes of power and inequality; onthe contrary, we are reminded of the importance of analysesthat bring together the various existing systems of differenti-ation in specific local contexts.

Intersectionality becomes an important tool for under-standing how discrimination, oppression and domination ofBrazilian domestic workers are produced and stabilized. It alsoemerges as an important concept for our comprehension of thepolitical mobilization of workers in Brazil. Discrimination,oppression and domination occur primarily in the workplace,whereas political mobilization operates primarily in the spacedelineated by the trade unions representing this professionalcategory.

Intersectionality between class, race and gender inthe workplace

Revisiting interviews from earlier research on femaledomestic workers' unions in Brazil, the explanatory potentialof the concept of intersectionality becomes clear. From theoutset, it is important to stress that the explanatory potentialof intersectionality varies according to the context invokedby the domestic workers themselves. When the women talkabout their experiences before joining political associationsor unions, the dimension of intersectionality emphasizingoppression, disempowerment and inequality becomes useful.But when they discuss their experiences after joining theunions, the dimension emphasizing agency, empowermentand political and democratic mobilization comes to the fore.It is also important to stress that we are not talking about purecontexts: that is, a context in which intersectionality impliesdisempowerment only, and another in which it emerges as akey element in political mobilization. Rather, I am arguing that,in any given context, one of these meanings of the concept ofintersectionality prevails over the other without implying thecomplete absence of the latter.

Most female domestic workers begin work as children,recruited from the contingents of poor women with minimaleducation who migrate to towns and cities from rural areas.Their culture, language, dress and race are considered inferiorto those of dominant urban classes. They usually work aloneand sometimes without pay — the employers claiming thatthey are “raising” them, as if they were their own children(Bernardino-Costa, 2011; Chaney & Castro, 1989). Such situa-tions, experienced by domestic workers in Brazil and elsewhere,especially various Latin American countries, reinforce the factthat the predominant kind of intersectionality experienced bythese women is one of disempowerment and oppression. Inother words, race, class and gender are experienced in terms ofracism, class discrimination and patriarchy.

Page 77: Womens Studies International Forum 46 (2014)

75J. Bernardino-Costa / Women's Studies International Forum 46 (2014) 72–80

In their life narratives, domestic workers frequently citemarkers of their race, class and gender identities to explainthe difficulties and hardships of their everyday lives. All thedomestic workers interviewed started to work at an earlyage, when they were children, often after they had beengiven to the employer's family to be raised, with the promisethat they would be provided with a formal education andbetter living conditions. Instead, they faced a life of daily toil,very different from the employer's promises to their mothers.The story of one of the women interviewed, aged 66 at thetime of the interview, clearly illustrates the conditions facedby many domestic workers in Brazil.

Madalene (not her real name), currently affiliated withthe Recife Female Domestic Worker's Union, remembers herfirst job as a domestic worker in the northeastern region ofthe country in during the 1950s. She, like other children, wastaken in by the employing family with the promise that shewould be treated like one of their child, with access to goodhousing and schools. What in fact followed was an experi-ence marked by discrimination, inequality and differentia-tion. Her job was to clean the house, wash clothes by hand,iron clothes with a coal iron, and serve breakfast to heremployers. While the family's children went to school, shehad to stay at home and perform physically exhausting work,without the right to go to school.

Although Madalene participated in the intimate life of thefamily, overhearing their conversations, witnessing their mo-ments of joy and sorrow and generating well-being through herwork, her feeling was that she and the other domestic workersemployed in the home were treated like slaves, their humanityneglected. Her employers seldom talked to her and often talkedabout her in her presence, as if she were not there.

The reference Madalene makes to slave labor is notadventitious. Although it has metaphorical content in thisreference, there is also real evidence that that job retainedconcrete traces of slavery, officially banned in Brazil in 1888.She worked from 6:00 am to 8:00 pm, without remunerationand was exposed to physical and emotional violence, andoccupied spaces set apart within the home. Similar to theslavery period, in which the slaves' quarters were isolatedfrom the overseer's house, Madalene reports that her roomwas separate from the employing family's home.

Another female domestic worker, Carla (not her real name),younger thanMadalene, also describedworking at an early age,this time in Salvador, also in the Northeast. Although theirexperiences took place in different cities and during differentperiods of the twentieth century (the 1950s and 1990s), wecan observe similarities in their narratives, including workingunpaid or for unlawful pay, being overworked and the youngage at which they began working.

Carla told her story as follows:

A woman visited my town and asked mymother to let herraise me. So mymother allowed her to take me away (…).My experience in that job was very long, very harsh andvery sad. I was nine years old and I had to take care of twochildren and clean the whole house. So it was very tough.I didn't have time to study or play. My playtime amountedto taking care of two children younger than myself. Ilooked after them, bathed them, cleaned the house. I evencooked. I ironed clothes (…). My work was very heavy, it

was humiliating. I was beaten. They would burn me. He(the husband) tried to rape me. I started work there whenI was nine years old and left when I was fourteen.

[Carla, twenty-three years old in 2005, affiliated to theBahia Female Domestic Workers' Union]

Both Madalene's and Carla's narratives describe an every-day experience of exhausting work and denial of freedom,including sleeping at theworkplace and receiving nopay.Whatis notable in their life trajectories is that theywere both subjectto their employers' will because they started working indomestic service while they were still children. Madalene'snarrative – like those of other interviewees who cannot beincluded here for space reasons – is also notable for the linksshe herself makes between domestic work and slavery: “Therewas a separate room in the house, in the backyard. It was like theslaves' quarters.” Carla also recalls an argument she hadwith herfemale employer in the 1990s, when the latter remarked thatslavery should never have been abolished in Brazil.

The encounter between two women in the same domesticspace, one the employer and the other themaid, does notmeanthat they will inevitably build a relationship of solidarity justbecause they are women. Class and race differences frequentlyintersect with gender, producing a hierarchical difference inthe positions occupied by women. The fact that such youngchildren became domestic workers indicates how much classforces are a determiner of this fate. Class does not operate inisolation, but rather in combination with race. If it is true thatchild domestic labor is an option for poor girls in the Braziliansociety, it is more so for girls who are poor and black.

Class and race are evident not only at the moment ofemployment; they operate throughout the new labor rela-tionship. There is a clear distinction between class habitusand the racial distinctions that become forms of humiliation,denigration and dehumanization. On the other hand, in thiscontext of power relations, gender is also a dimension of lifethat disempowers these workers, making them vulnerable tosexual violence.

The interviewees' life narratives are very similar. Whetherthey come from Brazil's northeast (Bahia and Recife), or thesoutheast (São Paulo and Rio de Janeiro), we come across thefollowing facts: (a) they begin work as domestic employeeswhile they are still children; (b) the female employer promisesto raise the child, but she actually exploits her as an unpaiddomestic worker; (c) the women are recruited from poorfamilies from rural areas; (d) as children, they are obligedto follow an adult's work routine with no respect for theirphysical capacity and well-being; (e) they frequently recountstories of physical violence, verbal abuse and very often sexualharassment and violence; (f) their freedom is restricted, thewomen are unable to enter and leave their employer's housewhenever they want; (g) they frequently receive no regularpayment; (h) the majority of the women are black (all inter-viewees were black women).

Examining the interviews, we can note the presence ofmultiple axes of disempowerment in their social relation-ships. The variables of class, gender and race work as markersof difference, subordinating one woman to the other. BothMadalene's and Carla's cases show us how the intersectionalityamong gender, race and class subdues them. The simple factthat they are women does not trigger any gender affinity from

Page 78: Womens Studies International Forum 46 (2014)

76 J. Bernardino-Costa / Women's Studies International Forum 46 (2014) 72–80

their boss' wife, since they are both from different classesand races. Similarly, Carla and Madalene come from poor,rural backgrounds, which explain why they became domesticworkers. Finally, their racial origin – black women – adds up tothe chain of vulnerabilities, completing the process of natural-ization that subjected them, as children, to sexual abuse and, as aconsequence, not worthy of wage. Therefore, the concept ofintersectionality shows us the interaction among the variousaxes of power. That enables us to analyze the origins of dis-empowerment, inequalities and domination. Now, the conceptof intersectionality is different from the idea of double and triplediscrimination. In this given case, there is a dynamic process inwhich each one of the axes of power applies their ownenergy, and those energies are not nullified by one another. Allconsidered, Madalene and Carla are subjects of the discrimina-tion, disempowerment and oppression caused by the dynamicenergy of each of the analyzed dimensions, all of which actequally. The explanatory potential of intersectionality is alsosignificant, as it builds an inter-relationship among a macro-context affected by a racist, sexist and class-biased speech, andthe domestic workers' life dynamics.

The intensity of the subjugation is directly proportional tothe fact that these domestic workers are describing a periodof their lives when they were still children. Female domesticworkers gradually become subjects in their workplace asthey acquire experience and reach adulthood. They developvarious strategies to reduce their workload, such as performingtasks slowly, not cleaning the entire house every day, but justsome areas, and negotiatingwith their employers (Brah, 2006).

However, one of the decisive factors in domestic workersbecoming subjects again is ceasing to live and sleep intheir employers' house. This independence enables them tore-establish control over their work time and take days offwork for whatever reasons. Some of the reasons for this changeinclude: (a) recent urban changes in Brazilian cities in whichthe ‘maid's quarters’ have disappeared from the new architec-tural designs for middle-class residences; (b) a public campaignby some domestic workers' unions to combat the idea that theyare the family's daughters and simultaneously encourage thewomen to find their own home to live in.

For domestic workers, having their own home opens up anew world, much in the same way as participating in unionactivities, since it allows them to build relations based onsolidarity among equals, breaking with the values imposed bytheir employers. As we shall see below, the unions comprise asocial space in which the markers of race, class and gender areseen positively rather than negatively, variables that interactwith each other to empower, building solidarity betweenfemale domestic workers and other social actors that enablespolitical mobilization. Therefore, unlike the face-to-face rela-tionship between domestic workers and their employers, theunion becomes a place for the construction of a collectiveidentity, in which the intersectionality among the axes ofgender, race and class power moves towards a democraticmobilization and the domestic workers' empowerment.

Intersectionality between class, race and gender in thefemale domestic workers' unions

Becoming involved in the political activities of a tradeunion is a watershed in domestic workers' lives. Unions are

social spaces that rupture the isolation experienced withinthe four walls of the employer's house. Moreover, as arguedhere, the unions are also spaces in which the typicallyhierarchical relations between domestic workers and theiremployers are left behind. Today there are approximately133,000 unionized domestic workers, corresponding to just2% of all Brazilian domestic workers (IPEA, 2011).

When we move our analysis to the space of unions, weencounter narratives of democratic mobilization and em-powerment of domestic workers, rather than stories ofoppression. Over the years, since the emergence of the firstpolitical organizations of domestic workers, the recognitionof difference on the part of working class and informedwomen, as homogeneous categories, has meant the politicalempowerment of women. In addition to strengthening thecollective, the union is a space for the restructuring of thesubjectivity of these agents, in which they collectively affirmtheir humanity by transcending the sexist, patriarchal andracist determinations that they face on a day-to-day basis. Inother words, the unions are spaces for the construction offriendship, affection and attention, where each woman isrecognized as a unique and full individual by her companionsand friends.

The political and social movement of domestic workersbegan in 1936 when Laudelina de Campos Melo founded theDomestic Workers Professional Association in Santos (SãoPaulo state). The organization's explicit objective was toacquire the legal status of a trade union, which would enableit to negotiate with the Brazilian government and conse-quently acquire the official recognition and labor rightsalready achieved by other professional classes. Although thedenial of labor rights was the recurrent theme of the firstpolitical organization of domestic workers in the country,it was no coincidence that Laudelina de Campos Melo(1904–1991) was a black activist. She had participated inblack organizations in various cities of the Minas Geraisand São Paulo states in the 1920s and 1930s. In the 1930s,Laudelina had contact with activists from black associa-tions in São Paulo and was herself an activist with theBrazilian Black Front (Frente Negra Brasileira, FNB), themost important black organization in Brazil of the period(Pinto, 1993). Laudelina was also a member of the CommunistParty, thus able to count on other members' assistance inwriting the statutes of her new association.

In the late 1950s, therewas a resurgence of public debate onthe situation of female domestic workers in Brazil, promotedin particular by black organizations. For instance, the BlackExperimental Theatre (Teatro Experimental do Negro, TEN)included among its members the actress and domestic workerArinda Serafim, who encouraged other domestic workers toattend TEN's literacy classes and involved them in politicaltraining projects emphasizing their labor rights (Sermog &Nascimento, 2006). In addition, we should note the publicdiscussion of a law to regulate the profession and guarantee thefirst legal rights for domestic workers (Nascimento, 2003).

Starting in the 1960s, the campaigns run by black organiza-tions in support of domestic workers were given fresh impetusby the Catholic Church through an organization called YoungChristianWorkers (Juventude Operária Católica, JOC) (Hutchison,2010). In 1960, the Young ChristianWorkers promoted the FirstNational Meeting of Young Female DomesticWorkers in the city

Page 79: Womens Studies International Forum 46 (2014)

77J. Bernardino-Costa / Women's Studies International Forum 46 (2014) 72–80

of Rio de Janeiro. The following year, the JOC also organized theFirst Regional Congress of Female Domestic Workers in the cityof Recife, which brought together workers from the states ofCeará, Rio Grande do Norte, Paraíba and Pernambuco (Soares,2002).

It is important to note that as a labor organization, YoungChristian Workers supported workers as a universal andhomogeneous category. Female domestic workers were notincluded in the organization's official campaign agendathough, due to their specific condition, including the lack ofany legal recognition, while the majority of workers in Brazilalready possessed some rights since the 1930s. Despite theirspecificity relative to other professional categories, there weregroups of domestic workers at themeetings of Young ChristianWorkers. Odete Maria Conceição, one of the founders of theProfessional Association of Female Domestic Workers of Rio deJaneiro in the 1960s, reported an incompatibility betweendomestic workers and the working class, when the latter wasthought of as a homogeneous group. As soon as the domesticworkers recognized their differences vis-à-vis other workers,they set out to establish their own associations in orderto take measures concerning issues specific to them: aprofessional category not yet recognized by the State, thereforewithout rights, formed by blackwomen from the poorworkingclass.

While the domestic workers' movement gained groundthrough dialog with the Catholic Church, Laudelina de CamposMelo, living in Campinas, São Paulo, founded the Association ofCampinas in the early 1960s. However, instead of the CatholicChurch playing the main role, in Campinas we note a strongcoalition between the black movement, especially Black Ex-perimental Theatre and the Association.

In the 1960s, the female domestic workers' movementspread across the country as the result of interactions betweenthe Catholic Church (with its emphasis on the working classcause), the black movement and the trade unions. Differentfemale domestic workers' groups and associations allied withtheir partners in distinct ways. During this phase of thedomestic workers movement, the class-based interpretationof the condition of female domestic workers prevailed at thenational level.

This is also the impression we get of the movement onreading the resolutions producedduring theNational Congressesof the 1960s and 1970s. It does not mean that gender and raceissues were absent, but rather that the political mobilization ofdomestic workers was centered on being recognized as part ofthe working class and consequently winning the same rights asother workers. Domestic workers were finally recognized byBrazilian legislation for the first time in 1972, when theyobtained the right to register their employment, take twentydays' vacation per year and receive basic welfare coverage. Thislaw represented the successful outcome of the femaledomestic workers' struggle and the movement's primaryaim of achieving recognition of these working women asmembers of the working class.

It is important to note that although class struggle had beenthe recurring theme of much of the domestic worker move-ment's activism, such as the campaign for domestic employeesto have their own home, racial–colonialist issues were alsopresent. For example, domestic workers frequently comparedthe ‘maid's room’ in the employer's house to slaves' quarters, the

house to the owner's mansion and domestic work to slave labor(Bernardino-Costa, 2007).

The relationship with working class political organizationshas changed over the decades and had regional variations.In some cities, partnerships were established between thedomestic workers' unions and other trade unions, while inother cities there were conflicts and mutual distrust. The sameapplies to the feminist movement. The relationship betweenfemale domestic workers and feminist organizations beganwith great distrust, since freeing oneself from domestic workwas – and still is – seen as the precondition for being amiddle-class woman and able to work outside the home.

Although less intense, this distrust between the domesticworkers' union and the feminist movement persists in someunions. But taking the country as a whole, feminist organiza-tions have evidently become important allies in the domesticworkers' political organization. It should be noted that whilethe hegemonic feminist movement is an important partner indomestic workers' setting forth their political agenda vis-à-visthe Brazilian State, there are marked differences between thedemands of the domestic workers' movement and the feministmovement. Besides the hierarchical differences among femaleemployers and their “employed” domestics, the themes ofliberation of feminists from the wealthiest classes often do notresonate with workers who have other notions of the body,femininity and motherhood.

Positive productive interaction with the feminist movementincreased following the Fifth National Congress of FemaleDomestic Workers, held in the city of Recife in 1985, when thefeminist organization SOS Body (SOS Corpo) helped organizethe Congress. On the national scale, the women's movement –despite lingering wariness – became a true partner of the femaledomestic workers' movement when it supported their causeduring the drawing-up of the 1988 Brazilian Constitution.

The period between the Fifth National Congress of FemaleDomestic Workers in 1985 and the enactment of the 1988Brazilian Constitution was one of intense mobilization amongfemale domestic workers. The women visited Brasilia numer-ous times to pressure members of the national congress togrant rights to those working in the profession. After the 1988Constitution the struggle continued – led now by the legallyrecognized domestic workers' unions – since the movementhad obtained just nine of the thirty-four labor rights guaran-teed to other workers. From the nine rights, the following onesare worth mentioning: minimum wage, thirteenth salary,weekly rest day, paid vacation days, pregnancy license andretirement.

During this period, we can observe the growing nationalinfluence of the Campinas and Salvador unions. The activismof these two unions also led to the strengthening of theposition of racial and feminist issues on a national level. Thisdid not mean that union or class-based approaches vanishedor become devalued, but that new connections were madebetween class, race and gender. For this new configuration tocome about, the historical dialog between these two unionsand the black and feminist movements was crucial.

The testimony of Creuza de Oliveira, a founding memberof the Bahia Female Domestic Workers' Union, affiliated tothe Unified Black Movement (Movimento Negro Unificado,MNU) since 1983, shows how the race-based approach wasinternalized by domestic workers:

Page 80: Womens Studies International Forum 46 (2014)

78 J. Bernardino-Costa / Women's Studies International Forum 46 (2014) 72–80

The partnership between the Unified Black Movementand our professional group was intensified when I tookpart in a meeting in which I asked for support for us fromthe black movement. When I first began to participate, Ihadn't actually been invited, I just turned up. At the startof the 1980s, the language used in the black movementwas highly academic (…). It was difficult for a domesticworker to understand what they were saying. I knew itwould be difficult for me to understand what they weresaying, but I thought to myself: ‘the language they use isdifficult and I understand almost nothing of what theysay, but I know that the issue concerns me because theyare talking about black people. So regardless of whetherthey have PhDs or better economic conditions than me, asa domestic worker, it concerns me because I'm a blackperson too.’ So I thought the space belonged to me, too,and I was determined to stay. I stayed and I've participat-ed in the Unified Black Movement for years now.

[(Creuza de Oliveira, founding member of the BahiaFemale Domestic Workers' Union and President of the

Female Domestic Workers' National Federation)]

Creuza de Oliveira calls our attention to the fact thatdomestic workers do not adhere unthinkingly to thediscourse and political viewpoint of the black movement;rather, its race-based approach is critically interpreted andadjusted to the specific issues faced by female domesticworkers and in particular the life experiences of the blackworking class women.

Over the more than seventy years of the domesticworkers' movement's formation, we can see not only theformation of a resistance movement, but also a movement torestructure the lives of domestic workers.

This resistance movement is in line with the mobilizingdimension of the concept of intersectionality, as highlightedby Avtar Brah (1996). We see in the actions of the 1930s howdomestic workers mobilized a discourse of class and racialsolidarity for the foundation of their first political association.Perceiving differences between themselves and the country'smiddle and urban class, they teamed up with the blackmovement of the era for political action. Similarly, race wasemployed in the 1950s with the objective of achieving theirfirst legal gains. However, it was only in the 1960s that themovement gained a national dimension. Essential to thiswere the discourse and political mobilization of the YoungChristianWorkers. It is noteworthy that the workers were farmore interested in the political aspects of the organization ofthe Catholic Church than in the religious aspects. Someworkers reported to me that their bosses thought that theywere going to the meetings to pray or read the Bible, when infact they were involved in a process of political awareness andactivism. The perception of difference between themselves andother workers is another highlight of that historical moment.Although they lacked an academic education, the domesticworkers' perception that the working class contained raceand gender aspects also becomes evident in that historicalmoment and hence the need for the formation of specificdomestic workers' associations. The interpretations, demandsand political organizations based on gender and race reappearin the national movement of domestic workers with greatintensity from the 1980s up to the present day. If, from the

1930s to the 1960s, there were coalitions with the blackmovement, it was on a regional level, only gaining nationaldimension in the 1980s. Initially coalitions with the feministmovement were productive; although essential for domesticworkers, they were regarded with suspicion for being themovement of their bosses, whose political agenda did notrepresent the issues and demands of a professional categorydifferentiated by race and economic conditions. Domesticworkers repeatedly said in the 1980s, “If emancipation forbosses is getting rid of housework, going freely about, andturning usmore andmore into slaves in their homes,we do notsee liberation.” As for the black movement, we perceive agreater affinity of interests and worldview with the domesticworkers' movement. Although the possibility for solidaritybetween the black movement and the movement of domesticworkers always existed, since most domestic workers areblack, this coalition on a national scale, which negotiated withsociety and the State, only materialized in the 1980s. Decisiveto this was the black movement's vehement criticism of themyth of racial democracy, which purported the absence of racein Brazil (Bernardino-Costa, 2004). Consequently, the demandsof the black movement lacked resonance until the late 1980s.From that time forth, the black movement gained a place inthe national political arena and domestic workers, and uponjoining forces with the former, strengthened their own position.

Both coalitions with the class movement and the feministand black movements were fundamental for the legalequalization act, through a constitutional amendment passedin the beginning of 2013. That act removed a paragraph in theBrazilian Federal Constitution that restricted several socialrights to the domestic workers. Since then, Brazilian domesticworkers have had the right, under official regulation, to afixed workweek, overtime pay, unemployment insurance,guarantee fund for severance pay (FGTS in Portuguese) andnight-shift rates, among others. Nonetheless, this constitu-tional amendment does not apply to domestic day laborers,i.e. those working less than three days per week in the houseof a single employer. This subset comprises one quarter of theentire domestic labor force in the country.

To summarize, what one sees in this political history of theformation of unions by domestic workers is the emergence of ablack working class women's movement, in which we canidentify the intersectionality between class, race and gender. Inother words, the axes of power of class, race and gender,once seen as categories of inferiorization and subordinationbecame and continue to be central to the political and democra-tic mobilization of domestic workers, thus acting as positivevectors for the political agency of these women.

Also noteworthy in the context of the unions is thesubjective experience of the intersectionality of race, genderand class. The unions are spaces not only of political resistance,but also of the restructuring and resignification of lived ex-perience. The unions act as spaces for the recognition of theother as a complex subject with feelings and emotions. Thepolitical education and valuation of black women are achievedthrough activities that include the revisiting and reclamation oftheir historic contributions to the nation and the reinforcementof the value of black women's bodies and esthetics. In sum, apolitical project emerges through the unions' activities andactions that, by weaving together issues of race, class andgender, seeks to overcome the dynamics of hegemonic power,

Page 81: Womens Studies International Forum 46 (2014)

79J. Bernardino-Costa / Women's Studies International Forum 46 (2014) 72–80

which inferiorizes the black woman. It is in this sense that weare able to understand interviewees' discourse of participationand membership in union activities as rebirth: only in thissocial space were they truly valued as full human beings.

Conclusion

The study of female domestic workers in Brazil showshow the concept of intersectionality can be used in two mainways. First as a concept that focuses on the combined effectsof disempowerment and oppression. Second as a conceptthat emphasizes the possibility of political and democraticmobilization.

We have also seen that the meaning of intersectionalityvaries according to the social context under study. In theworkplace, the prevailing notion of intersectionality is one inwhich the differentials of race, class and gender interact,disempowering the social actors involved. In the spaces inwhich female domestic workers become politically orga-nized, on the other hand, these markers of social differencework to strengthen the solidarity among themselves andbetween these women and other social actors (such asactivists from the union, feminist and black movements).Female domestic workers gained strength by breaking awayfrom the socialization and values imposed by their employers,a rupture achieved in a number of ways, including: reachingadulthood, acquiring more life experience and learning howto deal with employers, leaving their employers' house tolive in their own homes, participating in union activities, andso on.

Another aspect worth mentioning is how domesticworkers subjectively experience intersectionality in the twospaces discussed above: in the workplace and the union. Ingeneral the workplace is a space clearly characterized byrelations of power, race, class and gender, where the axes ofdisempowerment, humiliation, bondage and dehumanizationare experienced, materializing in the simplest of day-to-dayattitudes of the employers, such as not saying good morning tothe domestic worker, not talking to her, and talking about heras if she were not present. The contradiction within the socialrelations taking place inside the home is that while the workertakes care of the family and creates comfort, well-being, andcleanliness, she is treated as a piece of equipment or a slave,disregarded as a human person.Within the framework of tradeunions, the crossing of the axes of power of race, class andgender are subjectively experienced as factors of empower-ment and the achievement of autonomy. In their activism, aswell as in the daily activities of the unions, race, class andgender are important dimensions in the workers' lives forthe achievement of value as full human beings. Thus, theraising of the self-esteem of these women, black andbelonging to the poor working class, becomes joy, happinessand rebirth.

We saw their criticism of political action based only onsocial class, as if theworking class were homogeneous, withoutinternal distinctions, and of the hegemonic feministmovementwhen the latter fails to see the women's movement acknowl-edging differences among women, especially of class and race.Finally, although we observe a greater proximity between theblack movement's agenda and the demands and elaborationsof the domesticworkers' movement, the former also becomes a

target of criticism when it becomes a black man's academicmovement. Echoing Sueli Carneiro (2003a, 2003b), wecan say that by combining class, race and gender perspec-tives, the black feminism of the domestic workers' move-ment blackens and feminizes the demands of the unionmovement.

Acknowledgments

I thank Encarnación Gutiérrez-Rodríguez, Jurema Britesand the two anonymous reviewers for their contributions tothis article, and I take full responsibility for the analysisdeveloped here.

Endnote

1 Female domestic workers have organized the following ten nationalcongresses: First National Congress in São Paulo City, 1968; Second NationalCongress in Rio de Janeiro, 1974; Third National Congress in Belo Horizonte,1978; Fourth National Congress in Porto Alegre, 1981; Fifth NationalCongress in Recife, 1985; Sixth National Congress in Campinas, 1989;Seventh National Congress in Rio de Janeiro, 1993; Eighth National Congressin Belo Horizonte, 2001; Ninth National Congress in Salvador, 2006 andTenth National Congress in Recife, 2011.

References

Bernardino-Costa, Joaze (2004). Levando a raça a sério: Ação afirmativa ecorreto reconhecimento. In Joaze Bernardino-Costa, & Daniela Galdino(Eds.), Levando a raça a sério: Ação afirmativa e universidade (pp. 15–38).Rio de Janeiro: DP&A editor.

Bernardino-Costa, Joaze (2007). Sindicatos das trabalhadoras domésticas noBrasil: Teorias da descolonização e saberes subalternos. Unpublishedmanuscript.

Bernardino-Costa, Joaze (2011). Trabalhadoras domesticas no DistritoFederal e suas condições de trabalho. In Natália Mori, Soraya Fleischer,Angela Figuereido, Joaze Bernardino-Costa, & Tânia Cruz (Eds.),Tensões e experiências: Um retrato das trabalhadoras domésticas de Brasíliae Salvador (pp. 133–180). Brasília: Centro Feministas de Estudos eAssessoria.

Brah, Avtar (1996). Cartographies of diaspora: Contesting identities. London/NewYork: Routledge.

Brah, Avtar (2006). Diferença, diversidade e diferenciação. Cadernos Pagu,26, 329–376.

Brah, Avtar, & Phoenix, Ann (2004). Ain't I a woman? Revisitingintersectionality. Journal of International Women's Studies, 5(3), 75–86.

Carneiro, Sueli (2003). Enegrecer o feminismo: A situação da mulhernegra na América Latina a partir de uma perspectiva de gênero. InAshokaEmpreendedoresSociais, & Takano Cidadania (Eds.), Racismoscontemporâneos (pp. 49–58). Rio de Janeiro: Takano Editora.

Carneiro, Sueli (2003). Mulheres em movimento. Estudos Avançados, 17(49),117–131.

Chaney, Elsa M., & Castro, Mary Garcia (1989). Muchachas no more: Householdworkers in Latin America and the Caribbean.Philadelphia: TempleUniversityPress.

Collins, Patricia Hill (2000). Black feminist thought: Knowledge, consciousness,and the politics of empowerment. New York and London: Routledge.

Crenshaw, Kimberlé (2002). Documento para o encontro de especialistas emaspectos da discriminação racial relativos ao gênero. Revista EstudosFeministas, 10(1), 171–188.

Crenshaw, Kimberlé (2006). Mapping the margins: Intersectionality, identitypolitics, and violence against women of color. In Linda Martín Alcoff, &Eduardo Mendieta (Eds.), Identities: Race, class, gender, and nationality(pp. 175–200). Malden/Oxford/Carlton: Blackwell Publishing.

Dussel, Enrique (1996). Filosofía de la liberación. Bogotá: Editorial NuevaAmerica.

Foucault, Michel (1995). O sujeito e o poder. In Hubert Dreyfus, & Paul Rabinow(Eds.), Michel Foucault: Uma trajetória filosófica para além do estruturalismo eda hermenêutica (pp. 231–249). Rio de Janeiro: Forense Universitária.

Hutchison, Elizabeth Quay (2010). Many Zitas: The Young Catholic Workerand household workers in cold war Chile. Labor: Studies in Working-ClassHistory of the Americas, 6(4), 67–94.

Page 82: Womens Studies International Forum 46 (2014)

80 J. Bernardino-Costa / Women's Studies International Forum 46 (2014) 72–80

IPEA (2011). Situação atual das trabalhadoras domésticas no país.Comunicados do IPEA n. 90. Brasília: IPEA. http://www.ipea.gov.br/portal/images/stories/PDFs/nota_tecnica/120830_notatecnicadisoc010.pdf (acessed 13.03.2013)

McClintock, Anne (1995). Imperial leather: Race, gender and sexuality in thecolonial contest. New York/London: Routledge.

Nascimento, Abdias (2003). Quilombo: Vida, problemas e aspirações do negro(1948–1950). São Paulo: Editora 34.

Pinto, Elisabete Aparecida (1993). Etnicidade, gênero e trajetória de vida deDona Laudelina de Campos Melo (1904–1991). Unpublished manuscript.

Sermog, Ele, & Nascimento, Abdias (2006). Abdias Nascimento: O griot e asmuralhas. Rio de Janeiro: Pallas editora.

Soares, Odete de Azevedo (2002). Uma história de desafios: JOC no Brasil —1935/1985. Unpublished manuscript.

Yuval-Davis, Nira (2012). Dialogical epistemology: An intersectional resis-tance to the oppression olympics. Gender & Society, 26, 46–54.

Page 83: Womens Studies International Forum 46 (2014)

Women's Studies International Forum 46 (2014) 81–82

Contents lists available at ScienceDirect

Women's Studies International Forum

j ourna l homepage: www.e lsev ie r .com/ locate /ws i f

Editorial

Challenging gender and violence: Positions and discourses inSwedish and international contexts

Since the mid-1990s the problem of men's violence againstwomen has beenwidely recognised in Sweden (Burman, 2010;Wendt Höjer, 2002). Moreover, feminist and women's humanrights perspectives have often influenced official rhetoric ingender equality politics and legal reforms. The main goal forSwedish gender equality policy is that women and men shallhave equal power to form their lives, as well as to engage inshaping society. Regarding violence the policy emphasises thatmen's violence againstwomen in Sweden should be eradicated,i.e. it includes a vision of zero tolerance regarding gender-basedviolence (Prop. 2005/06:155). Legal reforms, especially in thearea of criminal law, with the specific aim to enhance legalprotection for women against gender related violence andpromote gender equality have been carried through. Compre-hensive policy initiatives followed by generous public fundinghave also been undertaken, particularly in the area of intimatepartner violence and so-called honour related violence, in orderto enhance the knowledge in thepublic sector of Sweden and todevelop best practices. Despite these efforts, as highlighted thisyear in media by the National Domestic Violence Coordinator(Götblad, 2013), no decrease in men's violence against womenhas been observed.

The articles in this special issue emanate from thework in aninterdisciplinary research group at Umeå University, Sweden.The research group, ‘Challenging Violence’, includes researchersfrom a diversity of disciplines, e.g. law, public health, politicalscience, and ethnology. In February 2010, a collaborative projectwith other European researchers resulted in a special issue on‘Public policies on violence against women in a Nordic context’(Niemi & Öhman, 2010). In that issue we concluded thatSweden's policy against gendered violence has been relativelysuccessful, and the country has a low prevalence of violenceagainst women compared to other countries. The relativesuccess might be a result of several interacting societal factorsand of active policymaking. At the same time as the awarenessof different forms of gendered violence has increased inSweden, many still unmet needs have also been identified. So,despite this success, the development and implementation oflegal and health policies to improve women's safety and rightsagainst violence is still needed.

The aim of this special issue has been to continue ourcritical investigation on men's violence against women by

0277-5395/$ – see front matter © 2014 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.wsif.2013.12.004

challenging gender and violence from different perspectivesand disciplinary fields. In this collection of articles the issuesaddressed span gender equality policies and law to subjectand identity constructions. The presented challenges are alsodiverse as regards theoretical and empirical focus. Thisdiversity notwithstanding the articles also has some similar-ities. The point of departure for all these articles is Swedenand the Swedish position on men's violence against women,even though some topics are dealt with in both national andinternational contexts. The main violence themes are inti-mate partner violence and so-called honour related violence.Finally, the articles are based in feminist theory with anunderstanding of men's violence against women as related togender and power structures in society.

The starting point for the work with this issue was aworkshop in 2009 on masculinity and critical studies on menwith Professor Michael Kimmel from State University ofNew York, USA. Several of the contributions also aim atproblematising masculinity and violence. First, MichaelKimmel's essay relates to masculinity as described in theMillennium Trilogy by the Swedish author Stieg Larsson, one oftheworld's best-selling novels of the past five years. The authorbrings to the fore a less well-understood aspect of Larsson'swork, the critique of hegemonic masculinity offered in thenovel as well as a template for a new egalitarian masculinity insomemale characters, e.g. Mikael Blomkvist. Kimmel discussesthose characters against the background of shifting genderarrangements, and suggests that a space is emerging for ‘menwho love women’ as well as for men who hate them.

The following three articles challenge different aspects ofintimate partner violence and so-called honour related vio-lence in a Swedish context. Monica Burman's article focuses ona particular aspect of masculinity and violence, how provoca-tion as a male ‘abuse excuse’ mitigates culpability for theperpetrator in the Swedish criminal legal context of men'sviolence against women in heterosexual relations. She chal-lenges Swedish criminal law's suggestion that the individualmale perpetrator's specific understanding of his violenceshould be the perspective from which to understand andjudge his violence. She argues that violent men can and shouldbe held responsible for their emotional responses to women.While problematising the possibility of promoting change by

Page 84: Womens Studies International Forum 46 (2014)

82 Editorial

legal reform, Burman argues that the criminal legal notion ofculpability should be changed in two respects. First, acommonly expressed notion of emotions as ‘factual’ should bereplaced by an evaluative conception on emotions. Second,when provocation is judged, it should be acknowledged howvalues and reasons intersect with power relations.

In the next article, Kerstin Edin and Bo Nilsson investigateviolent men's discourses on violence. Men inclined to violenceare frequently squeezed into ‘one-size-fits-all’ batterer-intervention programmes with high ambitions for changethat often show little evidence of effectiveness. Some researcheven indicates that any changes in men's violent behaviourmight result from factors not at all linked to the programmes.For this study, ten interviews were carried out with men whohad attended anti-violence programmes within the SwedishProbation Service. The overall aim was to analyse genderedidentity constructions in the narratives of men attending theprogrammes, especially how men articulate the course ofviolent events and how they talk about themselves and theprogrammes. The authors found that men defended them-selves by excuses, explanations, and victim positionings.Furthermore, the men's gendered identity constructionscollided with the programmes' ambitions of changing men'sconceptions and behaviour.

The contribution by Maria Carbin deals with Swedish policyinitiatives against so-called honour related violence. Thesepolicy initiatives are relatively ambitious, and have primarilytargeted young women as victims, one aim being to make itpossible for them to speak up. The aim of the article is tochallenge the ideal that is articulated in the policies, namely theideal of speech as emancipation, and to elaborate the connec-tions between speech, silence, and power. Drawing upondiscourse theory and post-colonial feminism, the author showsthat, despite efforts by policy makers to include and notreproduce stereotypes, the possibility of speaking is formulatedwithin a nationalist discursive terrain. The victims are primarilycalled upon to speak as non-Swedish representatives. Paradox-ically, the inclusion of young women into policy discourse hasled to a particular exclusion and thereby produced new silences.

The following two articles aims at challenging political,health, and legal positions and discourses on gender andviolence in Sweden as well as internationally. Ann Öhman andMaria Emmelin discuss current Swedish international develop-ment policies on gender and violence. The article deals with theproblematic relationship between development policies, globalhealth, Swedish gender equality policies, and violence againstwomen in a global perspective, focusing on intimate partnerviolence and the highly promoted gender mainstreamingpolicy. The authors emphasise the need for integration ofgender and feminist theory into development collaboration. Inthat way interventions can be grounded in current and actualknowledge on, for example, gender relations, power structures,and male hierarchies that constrain and subordinate womenand girls. They also claim that stronger links need to be createdbetween local activist groups in low- and middle-incomecountries and international development agencies. The authorsconclude that it is vital to initiate and formalise a North–Southdialogue between such groups, as well as enhancing South–South dialogue and cooperation.

Eva Nilsson's article challenges the complexities of genderand violence within international refugee law, mainly from

a discourse theoretical perspective. The legal issue is thedefinition of ‘refugee’ in the United Nations Refugee Conven-tion and the ‘nexus’ requirement that the persecution must bebased on specific grounds. Women exposed to male partnerviolence face particular problems due to this requirement.For an abused woman to be defined as a refugee it has to beshown that her ‘membership of a particular social group’ — herbeing a woman — is the ground for the ‘persecution’, that is,the violence her male partner or ex-partner exposes her to.Eva Nilsson shows how this essentialist understanding ofthe subject in the convention is defended with reference toa perceived need to preserve the ‘structure and integrity’(the objective, purpose, and language) of the Convention.The author argues that this also means preserving the powerstructures in society. Even though the argumentation suggeststhat it is time to abolish the ‘nexus’ requirement, the conclusionis rather that we must continue to work with our frame ofthought focusing on the ‘refugee situation’ and the discursiveconstitution of the subject in time and space.

In our view, the articles in this issue give convincing evidenceof the need to continuously challenge how gender, power, andviolence are perceived, constructed, and dealt with in a widerange of contexts, in a country such as Sweden that frequentlyscores well in gender equality rankings. Problems stemmingfrom the Swedish position on gender equality are recurrentthemes in some articles, as well as the shortcomings of law andpolicy in promoting changing gender relations. Some contribu-tions also show how discursive processes and essentialist andpower-imbuedunderstandings of gender, in spite of ambitions tochange unequal gender relations, tend to produce new subordi-nate positions for women or contain othering processes that areoppressive for women exposed to violence. However, thispessimistic viewmust be nuanced. The articles also give evidenceof a state of affairswheremen's violence againstwomen is firmlypresent on the political, legal, and societal agenda, both inSweden and elsewhere. This gives us an opportunity to promotechange by continuing our challenges to gender and violence.

References

Burman, Monica (2010). The ability of criminal law to produce genderequality: Judicial discourses in the Swedish criminal legal system.Violence Against Women, 16, 173–188.

Götblad, Karin (2013). http://www.tv4.se/nyhetsmorgon/klipp/carin-g%C3%B6tblads%C3%A5-ska-v%C3%A5ldet-mot-kvinnor-stoppas-2295883

Special issue: Violence against women and public policies in the Nordic context.Niemi, Johanna, & Öhman, Ann (Eds.). (2010). Violence against women, 16.

Prop. 2005/06:155 Makt att forma samhället och sitt eget liv – nya mål ijämställdhetspolitiken [Governmental Bill: Equal power to shape societyand one's own life — New goals for gender equality politics].

Wendt Höjer, Maria (2002). Rädslans politik Våld och sexualitet i den svenskademokratin [The politics of fear. Violence and sexuality in the Swedishdemocracy]. Malmö: Liber.

Monica Burman Associate Professor, LLD⁎Umeå Forum for Studies on Law and Society,

Umeå University, Sweden⁎Corresponding author at. Umeå Forum for Studies on Law

and Society, Umeå University, 901 87 Umeå, Sweden.

Ann Öhman Professor, PhDUmeå Centre for Gender Studies, Umeå University, Sweden

Page 85: Womens Studies International Forum 46 (2014)

Women's Studies International Forum 46 (2014) 83–87

Contents lists available at ScienceDirect

Women's Studies International Forum

j ourna l homepage: www.e lsev ie r .com/ locate /ws i f

Men who love women: Pro-feminist masculinities in theMillennium trilogy☆

Michael KimmelDepartment of Sociology, Stony Brook University, Stony Brook, NY 11794, USA

a r t i c l e i n f o

☆ This article has not been subjected to an externaldoes not confirm to a scientific format. This has been pas chapter “Men Who Love Women: Pro-feministMillennium Trilogy” in “Men Who Hate Women andTheir Asses: Stieg Larsson's Millennium Trilogy in FeEds: Donna King and Carrie Lee Smith (Vanderbilt Uni

URL: http://www.michaelkimmel.com.

http://dx.doi.org/10.1016/j.wsif.2013.12.0070277-5395 © [2014] Michael Kimmel. Published by El

s y n o p s i s

Available online 14 January 2014

This article examines the different iterations of masculinity represented by different characters inLarsson'sMillennium trilogy. Byexaminingespecially the case ofMikael Blomkvist, andhis relationshipto Lisbeth Salander, the article describes the promises and perils of a profeminist masculinity.

© [2014] Michael Kimmel. Published by Elsevier Limited. All rights reserved.

Introduction

Voracious readers have been devouring Stieg Larsson'sMillennium trilogy the world over. They're fast paced,intricately-detailed, and fun to read. A huge draw is punky,scrawny, tattooed hacker-sleuth Lisbeth Salander. Readers areprobably familiar with the story surrounding Larsson's creationof such an uncompromising take-no-prisoners feminist hero inLisbeth. She's indifferent to fashion, to the feminine mystique;she's technically competent (a computer hacker); she's familiarwith violence and unafraid to use it in a retaliatory fashion. Shemay be the least feminine hero in contemporary fiction, and oneof the most feminist.

Many readers will recognize Salander as an expression ofLarsson's long-held feminist views. He claimed that he waspersonally disgusted by sexual violence. As a teenager, he watchedas several of his friends raped a 15 year-old girl he knew — alsonamedLisbeth. Larssonreportedlynever forgavehimself for standingby and allowing it to happen. He eventually became a lifelongcampaigner against violence against women. (Indeed, I met himonce, in 2002, while working with the Swedish Minister of GenderEquality on a project to reducemen's violence against women.)

From the early 1970s until his death in 2004, Larsson wasunswervingly identified as a feminist. For example, in late

review process as itreviously publishedMasculinities in theWomen Who Kick

minist Perspective”,versity Press, 2012).

sevier Limited. All rights rese

2001 and early 2002 there were two well-publicized murders ofwomen in Sweden; a Kurdish woman named Fadime Sahindaland a Swedish model named Melissa Nordell. Focusing onculture differences, the Swedish media described Sahindahl'smurder as an honor killing and Nordell's as a crime of passion.Larsson would have none of it. He called them “sisters in death”and stressed the underlying systematic and oppressive patriar-chal structure: theywere bothmurdered (bymen) because theywere women:

The forms of oppression differ — but not the cause ofoppression. The forms vary dramatically between Sicilianhonorary murders, burning widows in India, or battering ofgirlfriends and wives on Saturday nights in Sweden. Theculture does not explain the underlying causes as to why thewomen of the world are being murdered, disfigured, circum-cised, beaten and forced into different forms of ritual behaviordecided by men — the causes being that men in patriarchalsocieties oppress women.This is a systematic violence againstwomen – for this is exactly what it is about – and would bedescribed as such, if violence of the same proportion weredirected against trade unionists, Jews or handicapped people.Feminism and anti-racism are two sides of the same coin.

[www.thefirstpost.co.uk/54145,people,news,stieg-larsson-remembered by Eva Gabrielsson]

Please note that thesewordswere notwritten by some crazedradical feminist male-hating harridans of the fevered popularimagination, but by the most popular novelist in the Westernhemisphere at the moment—who also happens to be a man.

rved.

Page 86: Womens Studies International Forum 46 (2014)

84 M. Kimmel / Women's Studies International Forum 46 (2014) 83–87

While it is clear that Larsson's feminism is reflected in thecharacter of Lisbeth Salander, I want to explore a somewhatdifferent side of the story — one no less compelling, but farless visible. In Mikael Blomkvist and some of the other menwho inhabit the world of the Millennium trilogy, Larsson hascreated characters that embody a complex and ambivalentnew Swedish masculinity – a pro-feminist masculinity – thatis markedly different from more traditional notions of both“good” and “bad” men.

Misogyny and toxicmasculinities in theMillennium trilogy

Mikael Blomkvist is no ordinary masculine hero. Hardly aJames Bond type. He is invariably disheveled, often melan-choly, and a bit feckless. But he also emerges as a deeplymoral character, whose sense of ethics leads him to painfuldiscoveries about the society in which he lives. He isunmarried (divorced, with a grown child who figures almostnot at all in his life). He is a devoted friend to Lisbeth (yes,indeed, a friend with benefits in the first volume), and anattentive lover to Erika Berger, his closest friend andeditor-in-chief (who is married to a man who knows aboutand accepts her affair with Mikael, since it enables his sexualrelationships with other men). He is somewhat standoffishand aloof, self-effacing and even somewhat socially graceless.Yet his heart (and his powers of deductive and inductivereasoning) is always in the right place.

What a contrast to the series of foils that Larssonconstructs for him. In each of the three volumes, Blomkvistis set against “men who hate women,” as the literaltranslation of the first book would have it — men who usewomen, exploit them, assault and rape them, and who showcontempt for them. In Dragon Tattoo that includes severalelder members of the Vanger family, representing old-moneywealthy industrialists, the old bourgeoisie. They are thor-oughly corrupt, venal, and in this case, Nazi sympathizers andcollaborators during World War II (as many Swedishindustrialists exploited Sweden's neutrality during the warto profit by selling to both sides).

We also meet Nils Bjurman, the scoundrel-lawyer whobecomes Lisbeth's legal guardian through conniving anddeceit. He abuses his position and rapes Lisbeth twice. Heconsiders these quid pro quo sexual favors, not at all sexualassaults in return for his promise not to cut her off frommoney she earned working for Milton Security.

In Dragon Tattoo, Lisbeth exacts revenge on Bjurman,incapacitating him, and tattooing on his body “I am a sadisticpig, a pervert, and a rapist.” Having secretly videotaped himraping her, she threatens to expose him should he everretaliate against her. Indeed, she symbolically castrates him,promising to release the video if she ever finds him in bedwith any other woman. And when she discovers he isconsidering tattoo removal, she threatens to tattoo herslogan across his forehead.

Effectively neutered in Dragon Tattoo, Bjurman is readilydispatched in Played with Fire. We now meet a second waveof bad men — men like Peter Teleborian, a psychiatrist andhead of the institution in which Lisbeth was confined. Thisconfinement itself was a conspiracy of menwho hate women—

including unscrupulous communist-hating Säpo agents and thesadistic Soviet defector they are trying to protect. Teleborian,

too, abuses his power for sexual access — treating Lisbeth as asex toy and a human experiment. He knows she's not insane,but he derives sexual pleasure from torturing her. He will beexposed as a pedophile in the third volume, as his testimony isdiscredited and he is utterly disgraced.

We also meet Lisbeth's father Zalachenko and her freakishhalf-brother Ronald Niedermann. Their crimes include thesystematic trafficking of Eastern European and Asian womeninto Swedish brothels, where the women are raped, mur-dered, and abused for men's perverse pleasure and profits.Zalachenko abuses Lisbeth's mother as well, beating her,raping her and eventually incapacitating her. Indeed, theprimal crime that sets Lisbeth's trajectory in motion is heradolescent effort to protect her mother and exact revenge onher abusive and violent father: after he beats and rapes hermother, Lisbeth follows him to his car, throws a bottle ofgasoline on his head and calmly lights a match to set him onfire. (He survives, badly scarred, so that he can return tomenace her in the third book.)

Niedermann is particularly interesting, since he suffersfrom a rare malady, congenital analgesia, the inability to feelpain. What more searing commentary about the man whowould equate masculine power with brute force: thattraditional masculinity is possible only if one is incapable offeeling anything at all.

Finally, in Hornet's Nest, the toxic masculinities Larssonexposes expand to include corrupt and calculating bureau-crats, especially those in the Swedish secret police or Säpo.Men like Evert Gullberg and Gunnar Bjork represent a formof hysterical, masculinity. In their frenzy to make sure thattheir secret deal with Zalachenko remains hidden, they areprepared to sacrifice the lives of any number of people.Women, in particular, are expendable, disposable, collateraldamage. In this way, agents of the state are no better thanthe criminal they are trying to protect. Secretive and corrupt,these hysterical villains are exposed as fraudulent men and,significantly, none of these older misogynistic characterssurvive the trilogy.

What then are we to make of these men who hatewomen? First, Larsson suggests that the sins of the fathershave not yet been peacefully put to rest. Larsson means thisboth literally and figuratively. In the literal sense, the viciousdamage that Zalachenko inflicts on Lisbeth's mother playsout as a prime mover in Lisbeth's idiosyncratic development(not that one necessarily needs an abusive violent father inorder to become a feminist hero). Symbolically, Larsson usesthe depraved misogyny of the Nazi wing of the Vanger familyto suggest that Sweden has not yet come to terms with itsshameful past, and that “neutrality” is often a convenientcover for unspeakable crimes.

Secondly, the sins of the fathers are visited upon thesons, in the remarkable entitlement to women's bodiesand sexual access that many of Larsson's middle-agedmen seem to wield. Men like Bjurman and Teleborian arecontemporary heirs to the violent sadistic misogyny ofolder generations. They use their government-sanctionedpositions to gain sexual access to women in a way that wenow recognize as abuse, assault, harassment and rape.Today men face an existential choice about how they willintegrate such misogynistic legacies into contemporaryrealities.

Page 87: Womens Studies International Forum 46 (2014)

85M. Kimmel / Women's Studies International Forum 46 (2014) 83–87

Finally, Larsson insists that we acknowledge that theSwedish welfare state is no refuge from misogyny. Moreconcerned with image over substance, with making sure theveneer of civility, social progress and compassion for all itscitizens is unsullied by a far seedier reality, Larsson portraysthe state – police, federal crime fighting agencies, intelli-gence services, and the psychiatric-prison complex – asconsistently placing the welfare of men over women.Larsson also suggests that the dynamics of entitlementhave shifted in the contemporary era. Earlier eras werecharacterized by unspeakable misogynistic brutality whilethe social surface remained placid. Misogyny was taken forgranted, seemingly untroubled. But the contemporary erahas unsettled that assumed entitlement, rendering itproblematic so that contemporary men do face a choice.Those who choose to continue to hate women, Larsson tellsus, become increasingly hysterical, caught up in a spiral ofdissembling, a frenzied shoring up of traditional arrange-ments. These are the men (e.g., Bjurman and Teleborian)who are the most humiliated and disgraced. They had achoice about whether or not to act as they did. They chose.They chose wrong.

Promises and ambiguities in Larsson'spro-feminist masculinities

Other men in the Millennium trilogy reach that sameethical crossroad and make very different choices. MikaelBlomkvist is not alone in embodying a new masculinity. Buthe is emblematic of both the promises and the perils ofLarsson's pro-feminist masculinity.

There are several of Larsson's men who act ethically andcaringly — either because they are principled and just, orbecause they are emotionally consistent. There's HolgerPalmgren, Lisbeth's aging former guardian, now infirm andin a nursing home, but still as protective and caring as akindly grandfather. (His name seems deliberately similar tothat of Olof Palme, the former Prime Minister of Sweden,who was the avuncular embodiment of the new progressiveSwedish Social Democratic Party until he was assassinated in1986). Palmgren is a model of what the Swedish welfarestate is supposed to be: providing the social safety netwith care and compassion. Listen to how Larsson describeshim:

He had meticulously fulfilled the requirements of theauthorities and submitted a monthly report as well as anannual review. In all other respects he had treated Salanderlike any other normal being, and he had not interfered withher choice of lifestyle or friends. He did not think it was eitherhis business or that of society to decide whether the younglady should have a ring in her nose or a tattoo on her neck…(Hornet's Nest, p. 203)

Or, this:He trusted her enough still to know that whatever she

was up to might be dubious in the eyes of the law, but not acrime against God's laws. Unlike most other people whoknew her, Palmgren was sure that Salander was a genuinelymoral person. (Played with Fire, p.134)

Finally, he chastises himself for not having Lisbeth's orderof incompetence rescinded when he was well. “He loved thisdamned difficult child like the daughter he never had, and he

wanted to have an excuse to maintain the relationship.”(Played with Fire, p. 133)

There's also Dragan Armansky, the director of MiltonSecurity, for whom Salander works as a freelance privateinvestigator. Armansky looks the other way at Lisbeth'srather unorthodox methods of investigation – she hackscomputers – and reaps the benefits of her results. Armanskyexpresses sexual attraction in what could be a compromisingsituation for Lisbeth, but accepts gracefully when she rejectshis sexual advances. And, as the trilogy unfolds, we see hisaffection for her turn more paternal. Through characters suchas Armansky and Palmgren, Larsson reminds readers thatthere are pockets of honor and integrity in the old Sweden.These men represent a model of Swedish manhood thatseems to run parallel to the misogyny of earlier generations.Like the Swedish state image, they are avuncular, protection-ist, and comforting. They are ethical and dutiful, embodyingthe promise of the welfare state to take care of its citizens.But such avuncular paternalism can also be experienced bythose same citizens as patronizing and even condescending.

By contrast, younger men like Dag Svensson represent awholly new version of Swedish masculinity. An intrepidjournalist, Svensson uncovers a trail of sex traffickers, and iseager to expose them. He is an equal partner with hisgirlfriend, Mia Johansson, who is completing her doctoraldissertation on sex trafficking. Equal partners in life andwork, they are both murdered by Niedermann in large partbecause they are a team.

Then there is Paolo Roberto, a tough yet emotionallyresponsive boxer who is enlisted by Blomkvist to help findLisbeth, and who ends up beaten almost to death byNiedermann (Paolo Roberto is an actual person, a formerboxer and a celebrity chef, who plays himself in the Swedishfilm version of Played with Fire). Larsson seems to suggest inthe character of Roberto that the “sensitive new age guy”can also be a boxer — that real men not only supportwomen, but as the poet Walt Whitman says, “containmultitudes”.

Svensson and Roberto embody a third-wave pro-feministmasculinity, one that is accustomed to women's equality inthe workplace, in the bedroom, and even perhaps in theboxing ring. Other of Larsson's men characters displayambivalence between advocating for women and capitulat-ing to misogynistic and patriarchal gender arrangements.For example, Criminal Inspector Jan Bublanski tells InspectorSonja Modig that she is both a “strong woman and a goodcop” (Played with Fire, p. 362). Yet, when he asks if she wantsto file a sexual harassment complaint against a malecolleague who is hostile to her – and Modig declines to doso – Bublanski does nothing to further challenge the sexistworkplace environment that marginalizes her (and supportshim).

Ethical masculinity and its (sexual) discontents

Similarly, Blomkvist's embodiment of a new pro-feministmasculinity is not without its own contradictions andambiguities. On the plus side, Blomkvist actively avoidsaffairs with women who are not his equal, like the nameless“work experience girl” at Millennium he rebuffs when sheshows up uninvited at his house one evening on a pretense,

Page 88: Womens Studies International Forum 46 (2014)

86 M. Kimmel / Women's Studies International Forum 46 (2014) 83–87

eager for a fling (how ethical can a guy be?). And while hedoesn't care to join in a three-way sexual liaison with ErikaBerger and her bisexual husband, he isn't homophobic; he isportrayed as perfectly comfortable relating with gay mensuch as his colleague Christer Malm.

But there are some gray areas in Blomkvist's pro-feministbehaviors as well. His relationship with family seems merelyfunctional and mostly remote. He's a rather indifferentfather to an adolescent daughter he doesn't seem to knowwell or see very often. And his relationships with otherwomen are equally “functional” and hardly enveloped ingauzy sentimentalism or breathless passion. While he is anattentive lover, his relationships with both Lisbeth andBerger are primarily as friend and colleague. He respects,admires, and clearly cares about them, but he doesn't seemto get too sexually excited or emotionally attached. Theirrelationships are casual and pleasant, the sex sensible buthardly passionate. Indeed the only memorably passionatesex in the entire trilogy is between Lisbeth and Miriam Wu.And it's passionate largely for Mimmi.

So, while Larsson makes sure to tell us that Blomkvist is agood lover – the women have orgasms, and everyone seemssatisfied – sexual passion never drives the narrative. None ofthe good guys ever do anything for erotic love. Friendshipand revenge are animating passions, never sexual desire.Indeed, one could argue that there is a certain desexualizingof the new Swedish masculinity embodied by Blomkvist. Isthis necessary? Can pro-feminist masculinities embrace asexual desire that is neither predatory nor violent?

One reading of Larsson suggests the answer might befound in the writings of some radical second wave feminists.For them, normative cultural depictions of men's sexualityreinforced heterosexuality as predatory, violent, and op-pressive to women. Erections signified domination; inter-course a violation or occupation. Under patriarchy, women'sheterosexual desire was collaboration; lesbian sex could beliberating only if it in no way resembled heterosexuality andremained non-penetrative. Men's sexuality could be politi-cally unproblematic only if it was denuded of what men hadcome to understand constituted desire: aggressive, energet-ic, possessive. In such a post-patriarchal feminist universe,gender equality in sex required that men’s sex come toresemble stereotypical “feminine” sex.

Not so among contemporary thirdwave feminists. Onemighteven say that in their pursuit of sexual pleasure as a feminist act,third wave feminists have embraced a more “masculinized”sexuality—more agentic andmore pleasure-seeking as opposedto pain-avoiding. The encounter between Lisbeth, the epigram-matic third wave feminist, and Mikael, the embodiment ofan ambivalent second-wave pro-feminism, encapsulates thisdilemma. The constant misinterpretations and vacillationsbetween being lovers, friends and enemies, capture thecontradictions of this particular historicalmoment, and, I believe,provide part of the pleasure of reading the trilogy. But thenLarsson resolves this tension temporarily — and in a directionthat leaves this reader, at least, unsatisfied.

Initially, Mikael misreads Lisbeth's love for him — a lovethat takes her over 1000 pages of distance and even enmityto get over. In Dragon Tattoo, Lisbeth is on her way to delivera Christmas gift she bought for Mikael, when she sees himwith Berger on the street:

The pain was so immediate and so fierce that Lisbethstopped in mid-stride, incapable of movement. Part of herwanted to rush after them. She wanted to take the metal signand use the sharp edge to cleave Berger's head in two. She didnothing as thoughts swirled through her mind. Analysis ofconsequences. Finally she calmed down (p. 589–590).

She throws the gift into the garbage. Her momentaryjealousy scares her: she is vulnerable, and she has made it herlife's project to be utterly impenetrable. She vows to never letanyone, including Blomkvist, back into her heart.

In the third volume, Blomkvist and Berger decide to endtheir affair. It's not that there is anything “wrong” with it —they both enjoy it, and her husband has accommodatedhimself to it. It’s just that there is no magnetic attractionthat is enough to sustain it (and Berger recognizes thatBlomkvist's found a new love interest in Monica Figuerola).

And in the trilogy's final scene, Blomkvist shows up atLisbeth's apartment. They'd been lovers before, back in thefirst volume of the trilogy, although their affair, too, wassomewhat desultory, a bit of momentary comfort amidst allthe fear and violence.

He brings her bagels and espresso. He announces he is“just company.”

Now, however, Lisbeth does something she had not done,feels something she has not felt, since she was a little girl. Shetrusts Mikael. He “had in fact been a good friend to her overthe past year.” He is not a man who hates women.

The cost, however, is their sexual relationship. “She lookedat him for a moment and realized that she now had no feelingsfor him. At least not those kinds of feelings” (Hornet’s Nest, p.599). She is no longer attracted to Mikael as a lover. She is nolonger vulnerable, doesn't love him like that.

When Lisbeth Salander opens her door and invites him intoher apartment, she speaks for all the abused women who haveshown themselves to triumph over their victimhood, whohavesteeled themselves to prove themselves resilient and indeedheroic. She also allows the possibility that the term “feministman” is not an oxymoron — that men can be friends withwomen. Mikael can actually be a “man who loves women.” Hejust may not be able to have sex with them.

A true crime postscript

Followers of Larsson's work probably also know about thecontroversy surrounding his estate. Larsson died before any ofhis books saw the light of day, and his literary estate is nowworth tens of millions of whatever currency you fancy. Aswith many Swedes, he died without a will, having lived,unmarried, with his partner of more than 25 years (peoplewho live in countries with national health care, freeeducation, and other benefits of citizenship have no economicincentives to marry). As a result, his partner, Eva Gabrielsson,has been frozen out of the estate, while his brother and father,from whom Larsson was estranged for most of his adult life,inherited the entire estate, as the law dictates. Andwhile theyhave offered Gabrielsson a modest amount, they have refusedto honor her legitimate claim as the rightful heir to the estate.Talk about men who hate women! This scandal makes onewish that Mikael Blomqvist and Lisbeth Salander were on thecase. No doubt she'd dig up some email that revealedconspiratorial culpability by his male relatives, and he's

Page 89: Womens Studies International Forum 46 (2014)

87M. Kimmel / Women's Studies International Forum 46 (2014) 83–87

break the story of how, once again, a conspiracy of old-styleSwedish men have manipulated the laws away from justiceand towards their own interests. As with Zalachenko and

Niedermann, the real-life male relatives have proven venaland conspiratorial, and the woman, the true friend, isdispensed with, as if she were unwanted baggage.

Page 90: Womens Studies International Forum 46 (2014)

Women's Studies International Forum 46 (2014) 88–95

Contents lists available at ScienceDirect

Women's Studies International Forum

j ourna l homepage: www.e lsev ie r .com/ locate /ws i f

Blaming violent men—A challenge to the Swedish criminal lawon provocation

Monica BurmanUmeå Forum for Studies on Law and Society, Umeå University, Sweden

a r t i c l e i n f o

0277-5395/$ – see front matter © 2014 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.wsif.2013.12.005

s y n o p s i s

Available online 17 January 2014

Feminists have long criticized how provocations narrative of a woman ‘asking for it’ functionsas a legal ‘abuse excuse’ for violentmen and confirms their rationalizations and justifications forviolence. This article aims to challenge a particular aspect of provocation in Swedish criminallaw—namely, a tendency to individualize and subjectivize culpability in a way that suggeststhat the individual male perpetrator's specific understanding of his violence should be theperspective from which to understand and judge his violence. Criminal legal culpability isapproached as an important aspect in the relationships between gender, power, and violence,and the author argues that the notion of culpability should be changed in two respects.The tendency to regard emotions as ‘factual’ should be replaced by an evaluative view onemotions andmen's responsibility for their emotional responses towomen should be judged byacknowledging how values and reasons intersect with power relations.

© 2014 Elsevier Ltd. All rights reserved.

Introduction

In this article, I want to challenge how provocation andviolent men's culpability for their violence against femalepartners or former partners are dealt with in the Swedishcriminal law. Criminal legal doctrines of provocation longhave been criticized in Anglo-American feminist legal studies(Edwards, 2004; Howe, 2002, 2004; Ramsey, 2010; Tyson,2013). This criticism is directed at the provocations narrativeof a woman ‘asking for it’, which functions as a cultural andlegal ‘abuse excuse’ for violent men and confirms violentmen's rationalizations and justifications of their violenceagainst women.1 Moreover, defence laws, both provocationand self-defence, have been criticized for failing to reflect andrespond to the circumstances in which female domesticviolence victims kill their male abusive partners. Because ofthis criticism and public debate, often fuelled by particular cases,legal reforms regarding provocation have been carried throughin Anglo-American jurisdictions. Provocation as a partial defencethat reduces the crime of murder to manslaughter has beenabolished in three Australian states.2 The provocation defencehas also been abolished in New Zealand (Crimes AmendmentBill 2009). In these jurisdictions, provocation is to be considered

ll rights reserved.

instead by the court as a possible mitigating circumstance whendeciding the sentence.

Debates and reform processes in England and Wales andin the United States have not resulted in repealing the defenceof provocation. In England and Wales, the provocation de-fence has instead been replaced by a new partial defence ofloss of self-control (Coroners and Justice Act 2009, Section55). This defence is applicable if the defendant's loss of self-control is attributable to certain ‘qualifying triggers’. One ex-ample of a ‘qualifying trigger’ is ‘things done or said (or both)which caused the defendant to have a justifiable sense ofbeing seriously wronged’. However, the fact that a thingsaid or done constituted sexual infidelity is to be disregardedin determining whether a loss of self-control had a quali-fying trigger. A few US jurisdictions have also introducedcategorical exclusions of some victim behaviours. For exam-ple, in Maryland, ‘the discovery of one's spouse engaged insexual intercourse with another’ does not constitute a legallyadequate provocation (Ramsey, 2010). In several of theabove-mentioned jurisdictions, gender bias in self-defencelaws for women who kill abusive male partners has beenacknowledged simultaneously and resulted in reforms—for

Page 91: Womens Studies International Forum 46 (2014)

89M. Burman / Women's Studies International Forum 46 (2014) 88–95

example, in Queensland, Australia, where a separate partialdefence tomurder in abusive domestic relationshipswas intro-duced in 2010.3

A slightly different picture appears in the Swedish context.Feminist research, feminist advocacy, and public debate so farhave paid very little attention to the problems of gender biasand gender constructions in defence laws in general andas regards provocation in particular. Although several legis-lative and policy measures in the area of criminal law andmen's violence against women in heterosexual relations havebeen introduced in Sweden in the pursuit of more effectiveand gender-sensitive law and policy, the law and adjudicationon provocation has not been up for serious scrutiny anddiscussion.4 The Swedish criminal legal doctrine on provoca-tion and the ways in which the courts deal with provoca-tion differ to some extent from the Anglo-American ones, aselaborated below. However, similar cultural and genderedproblems can be observed in Swedish criminal law. Apologiz-ing gendered discourses fit well into Swedish criminal law bymeans of the provocation excuse, which blames women forthe violence they are exposed to and mitigates culpability forthe perpetrator (Burman, 2010). Similar observations havebeen made regarding discourses among violent Swedishmen themselves (Gottzén, 2012; Edin & Nilsson,2014–in this issue). As mentioned in the beginning, thisgendered-mitigating effect is a familiar problem in feministresearch. With this article, I want to add a specific dimensionto the analysis of this problem, namely, how the Swedishcriminal legal notion of culpability contributes to uphold theopportunity of mitigating blame in this way and counteractspossible change. More precisely, the aim of this article is tochallenge a tendency in Swedish criminal law to individualizeand subjectivize culpability in a way that suggests that theindividual male perpetrator's specific understanding of hisviolence should be the perspective from which to understandand judge his violence.

I will carry through my challenge by utilizing an analyticalapproach based on feminist legal theory and theories concerningmen’s violence against women as related to gender and powerdeveloped within feminist research and critical masculinitystudies. Criminal legal culpability thus will be approached andanalyzed as an important aspect in understanding the dynamicrelationships between gender, power, and violence. I particularlyaim to challenge howprovocation and culpability discourse tendto construct male rage towards women as an ‘inevitable’ excusefor violent men that is beyond possible change.

I will start by outlining my theoretical and analyticalframework regarding gender, power, violence, and criminallaw. The next section presents how provocation and culpa-bility are conceptualized and dealt with in Swedish criminallaw, with a focus on case law on men's violence againstwomen in intimate relations and criminal legal scholarship onculpability and provocation. Here I will also develop my claimthat the Swedish notion of culpability suggests that theindividual male perpetrator's specific understanding of hisviolence should be the perspective from which to understandand judge his violence. In the third section, I will outline threeproblems that follow from the way provocation and culpa-bility are conceptualized and dealt with and argue for theimportance of including gender and power into the analysisof criminal legal culpability. These problems concern how

violence can be contextualized, how power operates throughculpability in criminal law, and howmasculinity is construct-ed as an ‘abuse excuse’ in a Swedish context of strong genderequality discourse. Finally, in the last section, I will problem-atize the possibility to promote change by legal reform andargue for how the criminal legal notion of culpability shouldbe challenged and changed in the Swedish context.

Analytical framework

Feminist theorizing regarding men's violence againstwomen frequently concerns the specific relations betweengender, power, and violence. The violence is seen often ashaving two interrelated functions: violence is used on anindividual level by men to exert power and control over indi-vidual women, and, on a structural level, it has the effect ofperpetuating systems of domination related, for example, togender, race, and class (McCarry, 2007; Thiara & Gill, 2010).Domestic violence is thus seen as the result of menwishing todominate women through violence and coercive control aswell as of a culture that encourages or condones it (Raphael,2004). In this way, links are created between power systems,individual acts of violence, agency, and social/legal responsesto the acts. ‘Power systems’ are in this context conceptualizedas dynamic and contested forces that produce a context ofopportunity within which people choose to resort or not re-sort to violence. Power systems are seen as ‘structuring forces’defining possible acts and the consequences emanating fromthem and thus affecting how people act, the opportunitiesthat are available to them, and the ways in which their be-haviour and how they are situated are understood and so-cially defined (Burgess-Proctor, 2006; Connell, 2009; Thiara &Gill, 2010).

Men and masculinities have become increasingly morecentral in feminist theorizing about men's violence. The re-lationships between men, masculinity, and violence are welldocumented (McCarry, 2007). It is also argued that there is agreat deal of evidence that men react violently to challengesto their authority, honour, and self-esteem as men (Dobash,Dobash, Wilson, & Daly, 2011). Notions of masculinity cancontribute both to making men's violence possible and to itsbeing excusable and excused (Enander, 2009; Tyson, 2013).

In several contexts, however, neither the problem of vio-lence nor the perpetrator is explicitly gendered (Hearn &McKie, 2010). And if the perpetrator actually is genderedexplicitly as male, it might well be just a new way of locatingthe blame for the violence away from themenwho perpetrateit. For example, if violence is understood as being ‘naturally’associated with men, this ‘naturalness’ can be invoked tojustify such violence (Hearn, 2012). Men might also bedisembodied from masculinity with the result that the focusis directed away from the material reality of men's violentbehaviour and interaction and onto ‘masculinity’ (McCarry,2007). Or causal power may be attributed to ‘masculinity’ or‘hegemonic masculinity’, which thereby becomes the explana-tion and excuse for the behaviour (Connell & Messerschmidt,2005; Hearn, 2012). Another possibility that follows withsome approaches to masculinity—for example, some psycho-social narrative approaches—is according to Tyson that morestories are told that ‘permit the long held cultural habit ofreading male violence as an effect of anxiety and/or latent and

Page 92: Womens Studies International Forum 46 (2014)

90 M. Burman / Women's Studies International Forum 46 (2014) 88–95

somewhat defensive response to feeling feminised. Moreover,according to this story, originary responsibility for these feelingsis said to lie with a woman’ (Tyson, 2013;176).

My legal theoretical point of departure is feminist legaltheory, in which law is regarded as a societal discourse inconstant interaction with other discourses, and the criminallegal subject is regarded as gendered, classed, and so forth(Lacey, 1998; Naffine & Owens, 1997; Niemi-Kiesiläinen,2004; Smart, 1998). This stands in stark contrast to main-stream Swedish criminal legal theory, which emphasizes thatcriminal law must be regarded and constructed as an autono-mous system as free as possible from values, identity, andsubjectivity in order to achieve the equal treatment of indi-vidual offenders and to counteract misuse of state powersover its citizens. Context, structure, and differences are there-fore conceptualized as very problematic to take into consid-eration in criminal legal theory and practice (Lernestedt,2010). Consequently, the criminal legal subject is normallytheorized as detached from aspects such as gender and sexu-ality. In contrast, feminist legal theory emphasizes that it isimpossible to conceive of a criminal law and a criminal legalsubject without interrelations to, for example, social catego-ries, value systems, power, and socially constructed catego-rizations such as gender. These aspects are already present inlaw at the same time as law interacts with them in variousways. In this article, I alsowill regard criminal legal culpabilityas inevitably interrelated with such aspects.

Viewing criminal law as a societal discourse includes a no-tion of criminal law as influencing the production of normativebeliefs and gender norms in the social world (Hamilton, 2010).Criminal law emphasizes that violence and harm to otherpeople is morally wrong and that domestic violence is nota shameful private matter but a societal as well as individualinjustice. It is argued that criminal law has a potential to un-dercut cultural tolerance for violence against women andpromote human rights norms that protect women fromviolence (Coker, 2004; Larsen & Petersen, 2001; Raphael,2004). Law can be seen as even a particularly powerful tool inthe social construction of reality because it is the only discoursethat is backed by the legalized use of force (Hamilton, 2010;Niemi-Kiesiläinen, 2004).

However, feminist research has been concerned seriouslyand for a long time about the ways law constructs genderand the consequences emanating from such constructions.When legal discourse produces or reinforces gender norms andconstructs gendered realities, it is often done in a damagingor constraining manner, which increases vulnerability effects,reproduces social injustices, ignores differences betweenabused women, and impedes effective intervention (Albertin,2009; Coates & Wade, 2004; Hunter, McGlynn, & Rackley,2010). Law also very often fails to comprehend the ways inwhich women are socially harmed by the violence (Hunter,2008). These problems often are said to be particularly true asregards criminal law (Coker, 2004; Douglas, 2012; Hunter,2008). Nevertheless, inspired by Smart and others, my positionis that because violence against women is already in the legaldomain, it must also be addressed there (Howe, 2008; Smart,1989). Further, I intend to follow Smart by regarding law as asite for discursive struggle (Smart, 1998) and by conceptual-izing law ‘as a site of conflict and dispute and not a place ofrefuge or of resolution’ (Smart, 2012: 164).

Provocation in Swedish criminal law

In Swedish criminal law, provocation is a partial excuseapplicable to all criminal offences—that is, it is not restrictedto mortal violence, which often is the case in Anglo-Americanjurisdictions. Further, provocation is both a defence and apossible mitigating circumstance in deciding the sentence. Asa defence, it may lead a court to judge a criminal act as a lessserious crime than an assessment of the other circumstances(for example, bodily harm) would lead to. It might, for ex-ample, mean that an assault that in respect of the physicalharm caused should be judged as aggravated is judgedinstead as a non-aggravated assault and therefore subject to alower scale of punishment. As a mitigating circumstance,provocation may (sometimes in conjunction with a defence)lead a court to mitigate the punishment within the applicablescale of punishment.

In comparison with Anglo-American jurisdictions, Swedishcriminal law is lacking a detailed legal doctrine on provocation.The Swedish Penal Code defines provocation as being at hand if‘the crimewas occasioned by an evident offensive behaviour ofsome other person’.5 According to criminal legal scholarship,provocation is constituted by two aspects: first, a factual aspect,which means it has to be established whether the perpetratorwas provoked, that is, if he or she actually was in an emotionalstate of anger evoked by the behaviour of the victim; second, anormative aspect prescribing that a judgement has to be madeas to whether the perpetrator's reason for being provokedshould be accepted (Jareborg & Zila, 2007; Lernestedt, 2010).However, there is very little guidance in the preparatory worksof the Penal Code (for example, in relevant governmental bills)or in case law as regards what is included in the two aspects,how they are to be judged, and how they relate to each other(Burman, 2011; Lernestedt, 2010).

Further, neither the preparatory works of the Penal Codenor the case law clarifies the justification for the mitigatingeffect of provocation. Swedish criminal legal scholars arguethat the mitigating effect of provocation should be justifiedby the argument that provocation places the perpetrator intothe situation of a moral conflict involving contradictory ethi-cal reasons: a basic demand on self-control conflicts with amorally acceptable reason for reacting and giving vent to anger.The actor's sense of anger is seen as an appropriate emotionalresponse to the provocative behaviour, and a person who,because of such an emotional response, is placed in a moralconflict and does the wrong thing is regarded as deserving lessblame than a person who has all the moral aspects speakingagainst him (Jareborg & Zila, 2007; Ulväng, 2009; Von Hirsch &Jareborg, 1987).

Because of the lack of a detailed criminal legal doctrine onprovocation, the Swedish courts seem to have considerablefreedom in how to judge statements and defences of provo-cation. Apart from an article in 1996 concerning a SupremeCourt case of ‘homosexual advance defence’,6 part of a book(Lernestedt, 2010, which is discussed below) and a bookchapter on emotions and provocation (Burman, 2011), verylittle interest has been shown so far in mainstream Swedishcriminal legal scholarship in studying how provocationis dealt with in the courts or in theorizing and problema-tizing provocation. However, a feminist legal study of caselaw concerning men's violence against women in intimate

Page 93: Womens Studies International Forum 46 (2014)

91M. Burman / Women's Studies International Forum 46 (2014) 88–95

relationships has shown that it is common for the SwedishSupreme Court and the Appeal Courts to accept without anyfurther comments or discussions a man's statement that hewas provoked by ‘his’ woman (Burman, 2007, 2010). Forexample, in the following Appeal Court case, where the mankilled his wife, the court stated,

[Theman's] story further shows that [thewoman], whowasheavily under the influence of drink, after she woke him upsaid things that [theman] took to be very provocative. Afterthat [the man] fetched his shotgun, loaded it and shot [thewoman] at a close range with one shot from behind. Thus[theman] has actedwithout premeditation and in a temperafter feeling he had been seriously offended.7

In another case, where the man was convicted of repeatedassaults and unlawful threats during a period of five years,the man argued that the quarrels and ‘disturbances’ that,according to him, characterized their intimate relationshipwere because of the woman's provocative behaviour towardshim. The Supreme Court concluded,

It should also be noted that it must be assumed that thebehaviour of [the woman] under the influence of drinkon certain occasions might have had an impact on thedevelopment of the rows and disturbances between thespouses.8

The normative aspect of provocation was tested in a fewcases, but in a clear majority of cases where provocationwas an issue, the courts, without clarifying how the issue ofprovocation was judged, either implicitly accepted the man'sstatement that he was provoked or, only on the basis ofthe man's statement of being provoked, explicitly establishedthat the woman had provoked the man (Burman, 2010).Thus, the way Swedish case law deals with provocation in suchcases gives the impression that the individual maleperpetrator's personal and specific understanding of hisviolence as caused by ‘his’ woman's behaviour is theperspective from which to judge his violence.

Lernestedt (2010), a Swedish criminal legal scholar, hassuggested that the normative part of provocation, if notabolished, should be changed. His arguments are in line withhow culpability seems to be individualized and subjectivizedin case law on provocation andmen's violence againstwomenin heterosexual relations. He starts from the commonly usedliberal idea that general preventive ambitions with criminallaw—the wish to influence how people act and think—mustbe limited in order to prevent punishment being used merelyas a means to promote some good for civil society. The mainfunction of criminal legal blame is thus to restrict thepossibilities for punishment. In this way, culpability becomespart of an exculpating and mitigating discourse that tells uswhy we should not punish at all or why punishment shouldbe mitigated, but that fails to present positive criteria for anindividual having to bear the full moral and legal conse-quences of a crime (Lacey, 1988).

Lernestedt uses the concept of ‘true’ individual blamewor-thiness to promote the idea that blamemust be individualizedin a certain way in order to meet the retributive prerequisitefor allowing punishment of individuals as a tool for societal

purposes. The concept of ‘true’ blameworthiness, as argued byLernestedt, must be constructed and adjudicated without anyconsideration of aspects such as general prevention, impactson people's behaviours, and values or educational ambitions.Therefore, in his view, blame should be judged only with thehelp of retrospective measures located very close to theindividual (Lernestedt, 2010). His conclusion is in line withthe negative retributive theory on criminal law that domi-nates Swedish criminal law and legal scholarship.

However, Lernestedt does not elaborate what ‘retrospec-tive measures very close to the individual’ could be. I suspectthat aspects such as the unique emotions, beliefs, or values ofthe individual perpetrator and his reasons at the time of thecriminalized act could count as such measures. This is, asshown above,much in linewith how this issue already is dealtwith often in case law on men's violence against womenin intimate relationships. Further, a notion of individualizedblame in line with Lernestedt constructs the precedence ofthemale perpetrator's point of view on his own violence as aninevitable consequence due to how punishment is consideredhaving to be justified.

Gender, power, and culpability

In this section, I will outline three problems emanating fromhow provocation and culpability are dealt with, as demon-strated in the previous section. In my view, these problemscounteract a feminist understanding of the violence in criminallawand point to the importance of including gender and powerinto the analysis of criminal legal culpability. First, I will arguethat the current construction limits the ways in which theviolence can be contextualized. The second problem regardshow power operates in criminal law through culpability. Fi-nally, I will highlight how provocation and culpability facilitatea construction of masculinity as an ‘abuse excuse’ in a Swedishcontext of strong gender equality discourse.

The presented view on culpability as regards men's vio-lence against women in intimate relationships facilitates adiscourse in which the violence is understood as caused bysomething internal or external to the perpetrator. The contex-tualizing is directed at finding particular events or ‘factual’aspects that might answer the question of why an individualman performed the violent act against a female partner orex-partner. In this context, power and control are seldomseen as such ‘factual’ aspects. In Swedish criminal law, theinterest is instead mostly directed towards aspects such aspsychological deviance or ‘normal’ psychological distur-bances inside the violent man, the latter most often presumedto be ‘caused’ either by the behaviour of the woman or byproblems in the intimate relationship as experienced by theman and interpreted from his perspective (Burman, 2007).

A feminist discourse on violence, on the other hand,emphasizes the need to contextualize men's violence againstwomen in heterosexual relations in a way that acknowledgesthat this violence occurs in a particular context of perceivedentitlement and institutionalized power asymmetry (Dobashet al., 2011). Consequently, violence is not regarded as aresult of something primarily inside or outside the perpetra-tor. Violence is instead viewed as a choice and as a means ofachieving, aspiring for, or maintaining power and control.Violence is comprehended as functional in that it for example

Page 94: Womens Studies International Forum 46 (2014)

92 M. Burman / Women's Studies International Forum 46 (2014) 88–95

may end an undesired argument, prove that the violent mandeserves ‘respect’, or hold women in relationships (Dobash &Dobash, 1998; Enander, 2009). Such a feminist discourse willfind it difficult to influence criminal law, however, because ofthe current notions of provocation and culpability.

Contextualizing violence in criminal law by taking accountof differently situatedmale perpetrators' experiences of angerand of being violated by women is an easy operation. Contex-tualizing violence in a way that acknowledges the situationand experiences of differently situated abused female part-ners or former partners is, on the contrary, difficult. Thisimpacts on, among other things, what constructions of abusedwomen's agency can be included in criminal law. Allowingspace for the view thatmen justifiablymay be violent towardswomen leads the law to view women as ‘true victims’ onlyif they can be regarded as free from guilt (Bumiller, 1998)and being free from guilt is often equivalent to being anon-agent (Mahoney, 1994). Therefore, abused women riskhaving to downplay or even deny their agency in order to beregarded as ‘true victims’ by the criminal legal system. Fur-ther, if women show agency by resisting violent men, forexample, by arguing with the man or leaving the relationship,they risk being considered as providing ‘morally acceptablereasons’ for men to feel violated and to react violently. In thisway, abused women's claims of agency in relation to abusivemen mitigate violent men's culpability. In order to open upfor an understanding of abused women's agency and expo-sure to violence outside the dominant dichotomy of agency—victimization—the agency and culpability of violent men haveto be problematized (Burman, 2010).

According to feminist legal theory, an individual man'sviolence against his individual female partner or ex-partnercannot be reduced to a decontextualized act of an individual.Also, the manner in which culture and power ‘enter’ a conflictand merge with it must be recognized. From such a point ofview, criminal law should acknowledge that men's violenceagainst women in heterosexual relations is very often adysfunction of power, not of intimacy (Coker, 2004; Stark,2004). Instead, when Swedish criminal law enters an indi-vidual case of men's violence against women in heterosexualrelations, it tends to merge with the point of view of the maleperpetrator and with discourses that condone violence. Ac-cordingly, Swedish case law subjects abused women to moraljudgements contained in a masculinist discourse, whichseldom takes account of the position and rationality of abusedwomen. The discourse is masculinist because it justifies andnaturalizes male domination in a set of gender relations inwhich the point of view and power of men are taken forgranted (Brittan, 2001). Provocation and culpability, as dealtwith in the Swedish context, can be conceptualized as a legalconstruct which gives violent men power to repeatedly attri-bute blame for their own violence onto women. From myfeminist perspective, such constructions in law help to per-petuate violence and are counterproductive in preventingmen's violence against women.

This gendered power structure in law allows individualviolent men to defend a masculine position of power and aconstruction of themselves as normalmale subjects by blamingthe women exposed to their violence (Hearn, 1998: Sørensen,2002). In a context where being a man who is violent towardswomen is a negative or stigmatized identity, using provocation

and masculinity as an abuse excuse can be an escape routefrom being constructed as a violent man. In a Swedish context,the strong discourse on gender equalitymust be acknowledgedbecause there is a problematic relationship between Swedishgender equality ideology and men's violent practices againstwomen. According to Swedish gender equality ideology, vio-lence against women is connected to power relations betweenmen and women in society and incompatible with the goaland basic values of gender equality (Burman, 2010; Gottzén,2012). Therefore, as pointed out by Gottzén, a gender-equalman cannot be violent towards women and a violent mancannot be gender equal. And because all ‘ordinary’ Swedishmen are comprehended as gender equal, a violentmanmust besomeone else; he becomes the ‘other non-ordinary Swedishman’ (Gottzén, 2012).

In Gottzén's interviews with Swedish men who had beenviolent towards their female partners or former partners, themen oriented themselves towards the gender equality dis-course when they talked about their violence. Defining them-selves as ‘good’ and ‘ordinary’ Swedish men demanded thatthey in one way or another confessed to the values of genderequality (see also Edin & Nilsson, 2014–in this issue).However, according to Gottzén, men's violence againstwomen is disguised as a consequence of this process inwhich violent men define themselves as gender equal.Therefore, men's violence against women can continue atthe same time as Sweden might go on naming itself as themost gender equal country in the world (Gottzén, 2012).

It is obvious, from my feminist legal perspective, thatthe concept of provocation should not be allowed to facil-itate a construction of masculinity as an ‘abuse excuse’.However, Swedish case law and the idea of ‘true’ blamewor-thiness facilitate the status quo. In this way, uncomfortableand power-imbued questions about violent men's agency, therelation between masculinities and violence, and the distri-bution of blame beyond violent men's own experiences andemotions can be avoided. From a feminist legal perspective,this counteracts not only efforts to reduce men's violenceagainst women in heterosexual relationships, but also con-travenes change in the overall unequal gendered power rela-tions between men and women in society.

Challenging for change

Feminist scholars have long demonstrated the difficultiesin implementing feminist understandings of men's violenceagainst women into criminal legal practice. Legal reformwithout parallel social and cultural reform cannot counteractmen's violence against women in heterosexual relations,neither can a criminal justice system lacking professionalswith the specialized knowledge needed to adequately addressabused women's needs and provide supportive treatment(Albertin, 2009; Bailey, 2010; Bell, Perez, Goodman, & Dutton,2011; Connelly & Cavanagh, 2007; Hunter, 2008; Meyer,2011). Even if law is changed with an explicit feminist am-bition, notions of gender often find new ways of influencinglegal practice. One such example is the reform in Victoria,one of the Australian states mentioned in the introduction,that has abolished provocation as a defence that reducesthe crime of murder to manslaughter. The reform seems tohave resulted in some improvements in how violence against

Page 95: Womens Studies International Forum 46 (2014)

93M. Burman / Women's Studies International Forum 46 (2014) 88–95

women in heterosexual relationships is contextualized. Be-cause provocation instead is relevant as a mitigating circum-stance at the sentencing stage, however, excuses for maleanger and violence againstwomen still can influence case law.Further, there is some evidence that provocation-type argu-ments might still be successful in the guise of other offencessuch as manslaughter by an unlawful and dangerous act(Tyson, 2013).

With this feminist knowledge in mind, my main line ofargument is not related to how the law on provocation shouldbe changed or whether provocation should be abolished. Abol-ishing provocation as a defence and a mitigating circumstanceor changing it in order to respond to the feminist critiquewithout simultaneously changing the notion of culpability willprobably lead provocation-type arguments to find new waysinto the Swedish criminal legal discourse. As mentioned inthe section on the analytical framework, I do not primarily viewlaw as simple tool for change. Instead, I consider the law onprovocation and the notion of culpability as a site for strugglewhere dominant discourses that are oppressive for abusedwomen can be challenged. By arguing that the notion of cul-pability should be changed, I wish to take on a particularfeminist legal method by disrupting the process of genderconstruction in law and introducing different accounts thatmight be less limiting for women (Hunter et al., 2010).

In my view, the notion of culpability (at least in relation toprovocation) should be changed in two respects. First, thetendency to regard emotions as ‘factual’ should be replaced byan evaluative view on emotions. This should be in line withhow current criminal legal doctrine justifies why provocationshould mitigate blame (i.e., contradictory ethical reasons), butit seems as if this notion is forgotten repeatedly, especially inlegal practice. Second, men's responsibility for their emotionalresponses to women must be judged by acknowledging howvalues and reasons intersect with power relations.

An evaluative conception does not regard emotions asmerely physical bodily sensations. Emotions, such as anger,are instead seen from a constructionist point of view as beingconstituted by thought and reason. This is because thoughtsare needed in order to identify and define a specific emotionand separate it from others (Lupton, 1998; Nussbaum, 2004).According to Kahan and Nussbaum,

A person experiences anger when she perceives thatanother has slighted her in a significant way. This per-ception presupposes conventions that specify from whoma person may legitimately demand respect (from hersocial subordinates; from her peers; from certain mem-bers of her family; from all members of the community)and what forms of behaviour count as disrespectful(insulting words; the failure to include the person in someimportant activity; an inappropriate sexual overture).For this reason, anger can be viewed as a mechanism bywhich a person defends her status in the community andthe social norms on which her status depends. (Kahan &Nussbaum, 1996:347–348)

Social norms and conventions are thus necessary compo-nents in dealing with anger and provocation, for individualsas well as for those who are to judge them. Even if the norma-tive part of provocation was abolished, social normswould still

be needed in order to judge whether or not an individualperpetrator's anger was evoked by the behaviour of the victim(Burman, 2011). Social norms, and thus also aspects relatedto, for example, gender, cannot simply be excluded from thenotion and treatment of provocation and should instead beexplicitly addressed.

But this will not be enough to facilitate a more feministapproach to provocation. Violent men should—in line with anevaluative conception of emotion—be regarded as responsi-ble for their emotional responses to women. How a persondeals with emotions and constructs reasons for reacting andacting, violently or not, should be regarded as a matter ofagency and choice. This responsibility should be judged byacknowledging how values and reasons intersect with powerrelations. This argument is not to be confused with a view infavour of ‘collective’ blame or punishment of ‘groups’ or of‘men in general’. It is simply to argue for dealing with theissue of individual responsibility for the way we reason byjudging the values the reasoning is based on and by includingaspects of power on individual and societal levels, instead ofhiding these dimensions within a liberal notion of a value-freecriminal law.

In my view, criminal legal intervention is needed in orderto try to prevent violent men from depriving women oftheir rights to autonomy, integrity, and well-being. However,there is no guarantee that such a notion of culpability thathas been sketched out here will deliver an outcome which ismore favourable from a feminist point of view. It might aswell lead to an even more outspoken confirmation by thecriminal justice system of violent men's rationalizations andjustifications of their violence against women. However,hopefully a changed notion of culpability will provide bettertools for taking the discursive struggle in Swedish criminallaw one step forward in challenging provocation and genderconstructions in law. Maybe it will also facilitate the imple-mentation of gender equality as a value in criminal law in afeminist direction, instead of gender equality being a dis-cursive tool which ‘ordinary’ Swedish violent men, with somehelp from the courts, can use to distance themselves fromtheir abusive behaviour. And one thing is for sure, withoutsuch a challenge to Swedish criminal law, the criminal lawwill do no more than continue to facilitate the use of violenceas a tool for men to achieve, to aspire to or to maintain powerand control over women.

Endnotes

1 Regarding violent men’s discourses on violence and culpability, see,e.g., Hearn, 1998; Mullaney, 2007.

2 Tasmania (Criminal Code Amendment Act 2003), Victoria (CrimesHomicide Act 2005), and Western Australia (Criminal Law Amendment Act2008).

3 The Criminal Code (Abusive Domestic Relationship Defence andAnother Matter) Amendment Act 2010, see Tyson, 2013, 39–42.

4 The most comprehensive reform in Sweden is the Women’s PeaceReform in 1998 in which, among many other legal and policy measures, anew crime designed to better reflect and respond to domestic violence asexperienced bywomenwas enacted. The new crime is named ‘gross violationof a woman’s integrity’ and is partly described and analyzed in Burman, 2010,2012.

5 The Penal Code Chapter 29, Section 3, 1st paragraph, with amendment1 July 2010 (SFS 2010:370).

6 ‘Homosexual advance defence’ is available to heterosexually identifiedmen who have murdered a homosexual male victim and claim provocation

Page 96: Womens Studies International Forum 46 (2014)

94 M. Burman / Women's Studies International Forum 46 (2014) 88–95

because of a sexual advance from the victim. In the article, the author posesthe question when it should and when it should not be justified for a court todismiss a perpetrator’s own opinion regarding the reasons for killing anotherperson (Träskman, 1995–96). Regarding ‘homosexual advance defence’ incommon law jurisdictions, see, e.g., Tyson, 2013.

7 RH 2003:64, p. 312.8 NJA 2005 s. 712, p. 725.

References

Albertin, Pilar (2009). Vulnerability effects in the criminal justice system onwomen who suffer physical abuse in their couple relationships (Spain).Journal of Health Management, 11, 209–228.

Bailey, Kimberly D. (2010). Lost in translation: Domestic violence, ‘thepersonal is political’ and the criminal justice system. Journal of CriminalLaw & Criminology, 100, 1255–1300.

Bell, Margret E., Perez, Sara, Goodman, Lisa A., & Dutton, Mary Ann (2011).Battered women's perceptions of civil and criminal court helpfulness:The role of court outcome and process. Violence Against Women, 17,71–88.

Brittan, Arthur (2001).Masculinities andMasculinism. In StephenM.Whitehead& Frank J. Barrett (Eds.), TheMasculinities Reader (pp. 51–55).Malden: PolityPress.

Bumiller, Kristin (1998). Fallen angels: The representation of violence againstwomen in legal culture. In Kathleen Daly & LisaMaher (Eds.), Criminologyat the Crossroads. Feminist Readings in Crime and Justice (pp. 37–53). NewYork and Oxford: Oxford University Press.

Burgess-Proctor, Amanda (2006). Intersections of race, class, gender, and crime:Future directions for feminist criminology. Feminist Criminology, 1, 27–47.

Burman, Monica (2007). Straffrätt och mäns våld mot kvinnor. Om straffrättensförmåga att producera jämställdhet [Criminal Law and Men's Violenceagainst Women. The Ability of Criminal Law to Produce Gender Equality].Uppsala: IUSTUS.

Burman,Monica (2010). The ability of criminal law to produce gender equality—Judicial discourses in the Swedish criminal legal system. Violence AgainstWomen, 16, 173–188.

Burman, Monica (2011). Provokation—Förnuft eller känsla? [Provocation—Emotion or reason?] In Ulrika Andersson, Christoffer Wong, & HelenÖrnemark Hansen (Eds.), Festskrift till Per-Ole Träskman (pp. 102–113)Volume in honour of Per-Ole Träskman (pp. 102–113). Stockholm: NorstedtsJuridik.

Burman, Monica (2012). Immigrant women facing male partner violence—Gender, race and power in Swedish alien and criminal law. Feminists @Law, 2, 1–26.

Coates, Linda, &Wade, Allan (2004). Telling it like it isn't: Obscuring perpetratorresponsibility for violent crime. Discourse & Society, 15, 499–526.

Coker, Donna (2004). Race, poverty and the crime-centered response todomestic violence: A comment on Linda Mill's insult to injury: Rethinkingour responses to intimate abuse. Violence Against Women, 10, 1331–1353.

Connell, Raewyn (2009). Gender in world perspective. Cambridge: Polity.Connell, R. W., & Messerschmidt, James W. (2005). Hegemonic masculinity:

Rethinking the concept. Gender & Society, 19, 829–857.Connelly, Clare, & Cavanagh, Kate (2007). Domestic abuse, civil protection

orders and the ‘new criminologies’: Is there any value in engaging withthe law? Feminist Legal Studies, 15, 259–287.

Dobash, Rebecca, & Dobash, Russel P. (1998). Violent men and violentcontexts. In Rebecca Dobash & Russel P. Dobash (Eds.), RethinkingViolence against Women (pp. 141–168). Thousand Oaks, California: SAGE.

Dobash, Russel P., Dobash, R. Emerson,Wilson, Margo, & Daly, Martin (2011).The myth of sexual symmetry in marital violence. In Michael Kimmel &Amy Aronson (Eds.), The gendered society reader (pp. 618–630) (4th ed.).Oxford University PressNew York.

Douglas, Heather (2012). Battered women's experiences of the criminaljustice system: Decentring the law. Feminist Legal Studies, 20, 21–134.

Edin, Kerstin, &Nilsson, Bo (2014).Men's violence: Narratives ofmenattendinganti-violence programmes in Sweden.Women Studies International Forum,46, 96–106 (this issue).

Edwards, Susan S.M. (2004). Abolishingprovocation and reframing self-defence—The law commission's options for reform. The Criminal Law Review, 80,181–197.

Enander, Viveka (2009). Jekyll and Hyde or ‘who is this guy?’—Batteredwomen's interpretations of their abusive partners as a mirror ofopposite discourses. Women Studies International Forum, 33, 81–90.

Gottzén, Lucas (2012). Att (inte) bli en kvinnomisshandlare [To (not) becomean abusiveman againstwomen]. In Lucas Gottzén&Rickard Jonsson (Eds.),Andra män. Maskulinitet, normskapande och jämställdhet (pp. 149–165)Other men. Masculinity, creating norms and gender equality (pp. 149–165).Malmö: Gleerups Utbildning AB.

Hamilton, Melissa (2010). Judicial discourses on women's agency in violentrelationships: Cases from California. Women Studies International Forum,33, 570–578.

Hearn, Jeff (1998). The violences of men. London: SAGE.Hearn, Jeff (2012). Male bodies, masculine bodies, men's bodies: The need

for a concept of gex. In Bryan Turner (Ed.), Routledge Handbook of theBody (pp. 307–320). London and New York: Routledge.

Hearn, Jeff, & McKie, Linda (2010). Gendered and social hierarchies inproblem representation and policy processes: ‘Domestic violence’ inFinland and Scotland. Violence Against Women, 16, 136–158.

Howe, Adrian (2002). Provoking polemic—Provoked killings and the ethicalparadox of the postmodern feminist condition. Feminist Legal Studies, 10,39–64.

Howe, Adrian (2004). Provocation in crisis—Law's passion at the crossroads?New directions for feminist strategies. The Australian Feminist LawJournal, 21, 53–75.

Howe, Adrian (2008). Sex, violence and crime. Foucault and the ‘man’ question.New York: Routledge Cavendish.

Hunter, Rosemary (2008).Domestic violence law reform andwomen's experiencesin court. The implementation of feminist reforms in civil proceedings. NewYork: Cambria Press.

Hunter, Rosemary, McGlynn, Clare, & Rackley, Erika (2010). Feministjudgments. From theory to practice. Oxford and Portland: Hart Publishing.

Jareborg, Nils, & Zila, Josef (2007). Straffrättens påföljdslära [The doctrine ofcriminal sentencing]. Stockholm: Norstedts Juridik.

Kahan, Dan M., & Nussbaum, Martha C. (1996). Two conceptions of emotionin criminal law. Columbia Law Review, 96, 269–374.

Lacey, Nicola (1988). State punishment. Political principles and communityvalues. London and New York: Routledge.

Lacey, Nicola (1998). Unspeakable subjects. Feminist essays in legal and socialtheory. Oxford: Hart Publishing.

Larsen, Ann-Claire, & Petersen, Alan (2001). Rethinking responses to ‘domesticviolence’ in Australian indigenous communities. Journal of Social Welfareand Family Law, 23, 121–134.

Lernestedt, Claes (2010).Dit och tillbaka igen. Om individ och struktur i straffrätten[There and back again. On the individual and structural in criminal law].Uppsala: Iustus förlag.

Lupton, Deborah (1998). The emotional self. A sociocultural exploration.London: SAGE.

Mahoney, Martha R. (1994). Victimization or oppression? Women's lives,violence, and agency. In Martha Fineman & Roxanne Mykitiuk (Eds.), Thepublic nature of private violence: The discovery of domestic abuse (pp. 59–92).New York: Routledge.

McCarry, Melanie (2007). Masculinity studies and male violence: Critique orcollusion? Women's Studies International Forum, 30, 404–415.

Meyer, Silke (2011). Seeking help for intimate partner violence: Victims'experiences when approaching the criminal justice system for IPV-relatedsupport and protection in an Australian jurisdiction. Feminist Criminology,6, 268–290.

Mullaney, Jamie L. (2007). Telling it like a man: Masculinities and batteringmen's accounts of their violence. Men and Masculinities, 10, 222–247.

Naffine, Ngaire, & Owens, Rosemary J. (Eds.). (1997). Sexing the subject oflaw. Sydney: The Law Book CO Ltd.

Niemi-Kiesiläinen, Johanna (2004). The reform of sex crime law and thegender-neutral subject. In Eva-Maria Svensson, Anu Pylkkänen, & JohannaNiemi-Kiesiläinen (Eds.), Nordic Equality at a Crossroads. Feminist LegalStudies Coping with Difference (pp. 167–194). Aldershot: Ashgate.

Nussbaum, Martha C. (2004). Hiding from humanity. Disgust, shame, and thelaw. Princeton and Oxford: Princeton University Press.

Ramsey, Carolyn B. (2010). Provoking change: Comparative insights on feministhomicide law reform. Journal of Criminal Law&Criminology, 100(1), 33–108.

Raphael, Jody (2004). Rethinking criminal justice responses to intimatepartner violence. Violence Against Women, 10, 1354–1366.

Smart, Carol (1989). Feminism and the power of law. London: Routledge.Smart, Carol (1998). The woman of legal discourse. In Kathleen Daly & Lisa

Maher (Eds.), Criminology at the crossroads. Feminist readings in crimeand justice (pp. 21–36). New York and Oxford: Oxford University Press.

Smart, Carol (2012). Reflection. Feminist Legal Studies, 20, 161–165.Sørensen, Bo Wagner (2002). Ideen om vold som følelseudtryck [The idea of

violence as expression of emotions].Nordiska ministerrådetNordic Councilof Ministers (pp. 423–430)Kön och våld i NordenGender and violence in theNordic Countries. (pp. 423–430) (TemaNord 2002:545).

Stark, Evan (2004). Insults, injury and injustice: Rethinking state interven-tion in domestic violence cases. Violence Against Women, 10, 1302–1330.

Thiara, Ravi K., & Gill, Aisha K. (2010). Understanding violence against SouthAsian Women. What it means for practice. In Ravi K. Thiara & Aisha K.Gill (Eds.), Violence against women in South Asian Communities. Issues forpolicy and practice (pp. 29–54). London: Jessica Kingsley Publishers.

Träskman, Per-Ole (1995–96). Normalfall eller dödande med förmildrandesärdrag—gränsen mellan mord och dråp [A normal case or a mitigated

Page 97: Womens Studies International Forum 46 (2014)

95M. Burman / Women's Studies International Forum 46 (2014) 88–95

killing—the borderline between murder and manslaughter]. JuridisktidskriftThe Law Review, 434–443.

Tyson, Danielle (2013). Sex, culpability and the defence of provocation. London:Routledge.

Ulväng, Magnus (2009). Rethinking ‘in affect’—disturbed states of mind andcriminal responsibility. In Peter Wahlgren (Ed.), Scandinavian Studies in

Law. Criminal Law, Vol. 54. (pp. 165–188). Stockholm: Stockholm Institutefor Scandinavian Law.

Von Hirsch, Andrew, & Jareborg, Nils (1987). Provocation and Culpability. InFerdinand Schoeman (Ed.), Responsibility, character and the emotions:New essays in moral psychology (pp. 241–255). Cambridge: CambridgeUniversity Press.

Page 98: Womens Studies International Forum 46 (2014)

Women's Studies International Forum 46 (2014) 96–106

Contents lists available at ScienceDirect

Women's Studies International Forum

j ourna l homepage: www.e lsev ie r .com/ locate /ws i f

Men's violence

Narratives of men attending anti-violence programmes in Sweden

Kerstin Edin a,b,c,⁎, Bo Nilsson a,d

a Umeå Centre for Gender Studies, Umeå University, SE-901 87 Umeå, Swedenb Epidemiology and Global Health, Department of Public Health and Clinical Medicine, Umeå University, SE-901 87 Umeå, Swedenc Department of Nursing, Umeå University, SE-901 87 Umeå, Swedend Department of Culture and Media Studies, Umeå University, SE-901 87 Umeå, Sweden

a r t i c l e i n f o

⁎ Corresponding author at: Department of Nursing, U87 Umeå, Sweden.

0277-5395/$ – see front matter © 2014 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.wsif.2013.12.006

s y n o p s i s

Available online 19 January 2014

The efficacy of batterer-intervention programmes for men has frequently been questioned,inviting additional research and development. Men inclined to violence have multifacetedproblems but are frequently squeezed into ‘one-size-fits-all’ programmes with high ambitionsfor change that often show little evidence of effectiveness. Some research even indicates thatany changes in men's violent behaviour might result from factors not at all linked to theprogrammes.For this study, ten interviews were carried out with men who had attended anti-violenceprogrammes within the Swedish Probation Service. The overall aim was to analyse genderedidentity constructions in the narratives of men attending the programmes — how menarticulate the course of violent events and in what way they talk about themselves and theprogrammes.According to our results, men defended themselves by making excuses, explanations andvictim positions. Furthermore, the men's gendered identity constructions collided with theprogrammes' ambitions of changing men's conceptions and behaviour.

© 2014 Elsevier Ltd. All rights reserved.

Introduction

Men's violence against women has, again and again, beendescribed as a major social and public health problem (Walby &Allen, 2004) and also as a human rights crime (Hearn & McKie,2010). The perpetrator is almost always a current or previousintimate partner; this seems to be the case internationally andalso holds for Sweden, where this study was carried out(Renzetti, Edleson, & Bergen, 2001). Preventing and bringingintimate partner violence (IPV) to an end require the under-standing of complex causal connections that are explainedmostly in terms of personal, societal and socio-cultural factorsimpregnated by gender and power structures (Firestone, Harris,

meå University, SE-901

ll rights reserved.

& Vega, 2003; Heise, 1998). Consequently, taking action againstviolence requires a long-term change at many different levelsin society. Until sustainable alterations have been implement-ed, most public and research opinions suggest short-termsolutions such as efficient programmes to stop perpetrators ofIPV (Gelles, 2000).

For decades, the US and several Western countries have run‘Duluth-inspired’ batterer-intervention programmes that arereasonably consistent with cognitive behavioural therapy andapply a confrontational, pro-feminist and ‘one-size-fits-all’ ap-proach (Dia, Simmons, Oliver, & Cooper, 2007). However, theanti-violence programmes might have the wrong agenda,believing in accomplishing change by confronting all violentmen in the same way (Gadd, 2002). When challenging violentmen's gendered identity, there is a risk in ignoring thecomplexity, the emotional processes and the defence mecha-nisms involved. It is that the intervention might not only beineffective, but also that it can sometimes increase tension and

Page 99: Womens Studies International Forum 46 (2014)

97K. Edin, B. Nilsson / Women's Studies International Forum 46 (2014) 96–106

stress, and thus be dangerous for the partners of these men(ibid.).

Despite some positive reports saying that the programmesare making attendees less violent compared to non-attendees(Gondolf, 2004), it is still tricky to measure behavioural changeand prove whether a noticed change is a result of the pro-grammes or caused by other factors (Silvergleid & Mankowski,2006). However, and overall, most reviews of interventionprogrammes for men inclined to violence have reported quitepessimistic efficacy reports, inviting additional research andimprovement (Babcock, Green, & Robie, 2004; Feder, Wilsson,& Austin, 2008; Gelles, 2000; Smedslund, Dalsbø, Steiro,Winsvold, & Clench-Aas, 2007).

Studies of the participants' subjective perceptions of inter-vention programmes and therapy show different resultsregarding to what extent such interventions can facilitate achange of behaviour. According to Dobash, Dobash, Cavanagh,and Lewis (2000), Scott and Wolfe (2000) and Silvergleid andMankowski (2006), the participants reported positive changesdue to the treatment, irrespective of the fear of legal punish-ment. However, according to Shamai and Buchbinder (2010),men can experience treatment as positive and meaningful, butstill use a power scheme in creating relationships. Buttell's(2003) study shows similar results, illustrating that batterer-intervention programmes are not effective in changing the levelof moral development. Alexander and Morris (2008) argue thatintervention programmes are more efficient if the men havealready started to change when they start the treatment.

Intimate partner violence in heterosexual relationshipsincludes two parties, and both of their perspectives areneeded to understand the violence and how women's andmen's agency interact while influenced by hegemonic socio-cultural codes of male dominance (Cavanagh, Dobash,Dobash, & Lewis, 2001). Our research deals with reportedIPV against women and the insider perspective from one side,i.e. the male side.

This study was carried out with men attending anti-violenceprogrammes in Sweden, a country ranked as one of the bestregarding the promotion of gender equality (Hausmann, Tyson,& Zahidi, 2010) and welfare systems in general (Balkmar,Iovanni, & Pringle, 2008). Gender equality ideologies areembedded in, for example, legislation and organisationalstructures (Eriksson & Pringle, 2005). Nevertheless, men'sviolence against women exists, and the Swedish criminaljustice system,which is supposed to be gender neutral, i.e. treatmen and women in the same way, is criticized by researchersfor not being so (Burman, 2010). Consequently, men areneither expected to take full responsibility to rectify theirbehaviour nor are they held entirely responsible for theconsequences of their violent deeds, and this reduces women'sscope and ability to take action regarding their situation (ibid.).For this study, we presume that the men's stories reflect acontinuous negotiation not only with programmes and pro-fessionals' discourses but with more general discourses abouthow aman should be and act (Edin, Lalos, Högberg, &Dahlgren,2008). Moreover, in this dynamic process of negotiation, webelieve that hegemonic constructions of dominant mascu-linities are both contested and (re)produced (Cavanagh et al.,2001; Connell & Messerschmidt, 2005).

The overall aim was to analyse gendered identity con-structions in narratives of men attending anti-violence

programmes: how men articulate the course of violentevents and in what way they talk about themselves and theprogrammes.

Method

Theoretical framework

In this paper, we use men's narratives and view them asrepresentations of the social world (Riessman, 1993). How-ever, this does not mean that there is a linear connectionbetween “reality” and narratives. Instead, narratives areinterpretations of “the real” (Edin, 2006), and depend onseveral factors, i.e. actual discourses, situations, traditionsand ideologies, etc. People use narratives of different kinds inthe creation of individual and collective identities. Narrativescan be seen as “identity claims” (Mishler, 1999), e.g. in theform of contrasting or differentiating oneself from others(Georgakopoulou, 2007). To regard narratives as both repre-sentations of and producers of “reality” is to say that personalnarratives are something more than just personal. They can,for example, reflect experiences that others can identify with,and they contribute to the reformulation of public discourses(Shuman, 2005). Moreover, narratives emerge not only in butalso through the interview dialogue (Presser, 2008, p. 143).Somers (1994) describes narratively mediated processes asconstituting social identities and guiding social actions andinterplays.

In line with the narrative approach, we view genderedidentities as socially, relationally and culturally constructedand as intimately related to discourse. Discourses – systemsof meaning (Laclau & Mouffe, 1985) – are constructed in andthrough the language in use, offering gendered positions(masculine and feminine) that individuals can identify with(Edley, 2001). However, these positions are not fixed; insteadthey are subject to continuous change. Accordingly, identitywill be viewed as a process and as an on-going attempt atidentification (Glynos & Howarth, 2007). Any identity iscontingent, and “identities are always ‘failed identities’whichnever fulfil the telos of subjective identification, thus ren-dering them vulnerable to further dislocation” (Glynos &Howarth, 2007 p. 129). Thus, the men's narratives are viewedas attempts to establish a “fixed” masculine identity, but this“fixation” will always be temporary. An important part offixation processes is disidentification, which is defined as therejection of unwanted positions and attributions (cf. Skeggs,1997).

In order to explain men's violence and in the quest for achange, one ought to understand masculinities (and femi-ninities) as being contingent, inconsistent, changing andcomplex normative discourses reproduced in sexuality and indaily interactions between men and women. However, evenif gender and the meaning of violence are changeableconstructs, they are part of the multifaceted practice andculture where they were formed. For that reason, a change isnot unproblematic because of allied opposition, includingpower, and hence a process of negotiation is needed(Connell, 2005; Edley, 2001; Kimmel, 1987).

The theory of hegemonic masculinity (Connell, 1995) hasreceived a lot of attention in gender studies, and refers to aparticular idealized image of dominant masculinity to which

Page 100: Womens Studies International Forum 46 (2014)

98 K. Edin, B. Nilsson / Women's Studies International Forum 46 (2014) 96–106

images of other masculinities (and femininities) are subor-dinated and marginalized. More recently, the idea of ahegemonic masculinity as a singular characteristic has beendisapproved of, and the importance of analyses in terms ofhegemonic masculinities has been emphasized (Connell &Messerschmidt, 2005; see also Coles, 2009; Wetherell &Edley, 1999). Moreover, criticism has been directed towardsa non-reflexive use of the concept without local, regional,national or global considerations (Connell & Messerschmidt,2005; Lusher & Robins, 2009). Others have been critical ofhegemonic masculinity being treated as a set of properties orfixed character types, and have emphasized the need to viewhegemonic masculinity as dominant as well as a changinginteractive process in which marginalized masculinitiesplay an important role (Connell & Messerschmidt, 2005;Demetriou, 2001; Hirose & Kay, 2010). In this article, notionsof hegemonic masculinities will be used in an analysis of howmen account for their violence and, in the narratives, recon-struct dominant identities and how these identities relate tothe programmes‴ ambition of changing men's conceptionsand behaviour.

Ethics

The study was approved by the local Ethics Committee atthe Faculty of Medicine, Umeå University, Sweden.

Subjects and setting

The starting point for this paper is qualitative researchinterviews with ten men participating in similar anti-violenceprogrammes at three different institutions within the SwedishProbation Service. The agenda for these programmes included aminimum of 20 obligatory group sessions but some men couldbe mandated to attend further sessions. All the participantspresented themselves as if they had already attended amajorityof the sessions (two menmentioned attending 20 sessions andone 13). Eight men had been reported for violence against theirfemale partners (and were court-mandated into the pro-gramme), while two had admitted themselves voluntarily asthey had become aware of aggression and violent behaviour intheir relationships. We assume that they had all used violencein one way or another, but we did not assume anything eitherway in this regard, nor did we search for any evidence incriminal or similar records. The men had a mean age of around40 years (29–60) and all but onewere Swedish born. Themen'slevel of education varied, and they had different employmentpositions, fromas a blue-collarworker to chief executive officer.

In awrite-up from the SwedishNational Board of Health andWelfare (Socialstyrelsen, 2001), the programmes (attendedby the interviewed men) presented themselves as having adrop-out rate of about 30%, a low number when compared tolarge international reviews where the estimated drop-out rateis commonly reported as around 50% (cf. Daly, Power, &Gondolf, 2001). They also described the programmes as in-spired by Manscentrum in Sweden (Eliasson, 2000) and byAlternatives to Violence (ATV) in Norway (Råkil, 2002) and assimilar to ‘Duluth’-inspired programmes (Pence & Paymar,1993). The focus of the programmeswas to stop the violence byworking on the men's conceptions of masculinity and makingthem take unreserved responsibility for their violent behaviour

(Socialstyrelsen, 2001; and see also Edin et al., 2009, 2008; Edin,Högberg, Dahlgren, & Lalos, 2009).

Interviews and analyses

All the interviews were carried out by the first author(KEE): the first three in 2003 and the remaining seven in2007. The interview themes had proceeded from a broadproject focusing on IPV and pregnancy (Edin, 2006). Thequestions were open-ended and dealt with the men'sbackground, experiences of intimate relationships, pregnan-cy and parenthood, their current situation and the pro-gramme. The interviews lasted about two hours and most ofthem turned out to be detailed life stories. Eight interviewswere recorded and transcribed verbatim, while, for ethicaland methodological reasons, handwritten notes were thesource of information for the remaining two. Fictional namesare used for the quotations.

First, the entire narratives were coded into categories andthemes and comparisonsweremade to look for similarities anddivergences in what themen said, but also in what they did notsay. With inspiration from Bamberg (2004), we then analysedthenarratives at three levels,making it possible to study: i) howthey comprise arguments, characters and positions, ii) how themen talk and explain the violence, and finally, iii) how, throughdifferent narratives, arguments and explanations, they (re)construct masculine identities. Thus, an advantage with thismodel is that it makes the different steps of the analyticalprocess visible to the reader. Another advantage is thatnarratives can be studied as a result of individual agency(level 1), as interactively created (level 2) and as discursivelyproduced (level 3). See also the introductory level presenta-tions in each of the three sections below.

Results

1. Changing narratives–changing charactersThe first analytical level deals with construction of intentand central characters/positions in the stories, how theseare (re)created and in what way these changed during theinterviews and in relation to the purpose of what is told.We will, moreover, put forward narrative themes andgenres from the interviews. The gendered identity aspectsof the stories are dealt with further on.Most of the interviewees described how their relationshipwith their partner was good initially. They were happyand did a lot of enjoyable things together, as is shown insome of their comments: “So in fact it has been a superrelationship” (John); “It was very good at the beginning.We had a really good relationship” (Simon); and “Wethought we were the perfect, happy family couple then”(Kevin).Besides being a good partner, a wife/girlfriend can also bepresented as a fantastic woman/person. This was exem-plified when John described how he and his wife had hadreally rough times, but nevertheless they had never raisedtheir voices or quarrelled:

“Yes, I can only say that I think she is a wonderfulperson. I would, yes, up to a month ago, I would havedone anything to make us become a couple again.”

Page 101: Womens Studies International Forum 46 (2014)

99K. Edin, B. Nilsson / Women's Studies International Forum 46 (2014) 96–106

However, the interviewees also described how problemsarose and the relationship got more and more strained. Thestarting point was often a key event or a fateful moment(cf. Chatman, 1978; Giddens, 1991). In studies of narratives,key events are often mentioned as turning points used tostructure life stories. In this case, the narrators evaluate theirlives and,when looking in the rear-viewmirror, focus on themore or less explicit situations that seemmost important forthe violent events to come. The key event was usuallyfollowed by a history of decline, and, according to this genre,the picture of the partner and the relationship changed.In Mikael's description of how he had loved his girlfriendand how much fun they had had together, a key eventbecame apparent when hementioned that after a year or soshe suddenly started to change and became “nasty” and“bossy”. She stopped being nice to him and did not bother tocomfort him when he was sad. Moreover, she did notrespect him, used bad language, laughed at him and madehim “lose face”. According to the interviewee, she decidedeverything, and he was trapped; he could not do anythingwithout annoying her: “She provoked me a lot during thattime, at home, you know, very much.”Thus, in an interview narrative, the same woman can bedepicted as both very nice and very nasty. Which of these ischosen depends on the intent of the story, the actual genreand the chronological position. For example, Allan describedhis wife as a warm and understanding person. However,when the interview homed in on the topic of violence, thepicture changed and she became a nagging woman:

“She has been so pushy; whereas before she was sosoft and easy to talk to, nowadays I only feel accused,accused, accused.”

Other characters, who played a central role in the narratives,were the perpetrator and the victim. The perpetratorwas, forexample, visible when the men told a history of awakening(cf. Flinck & Paavilainen, 2008). Because the men were partof the anti-violence programme, in oneway or another theyrelated to their identity as a perpetrator (cf. Edin et al.,2008). In the interviews, this is reflected through differentself-reflexive presentations:

“And every morning I have to wake up with this, withwhat I have done.” (John)

Thus, many of the men did not deny the violence and someof them did not have any real problems describing theirexperiences and the violent incidents. This is contrary toother findings, which claim that men often rationalize anddeny violence (Flinck & Paavilainen, 2008). However, thisdoes not mean that they accepted the position as offendersby presenting themselves as malicious offenders. Instead,they used different narrative strategies in order to make thecharacter of offender seem less vicious, e.g. by the use ofdifferent explanations. We will elaborate on this under thenext heading.The position as victimwas attributed to the partner and thechildren, but sometimes also to the men themselves. Forexample, they (re)created suffering characters by makingreferences to a problematic childhood, unemployment or a

bad relationship with the partner, and also by talking aboutthe social or economic effects of their own violence. Theydescribed an unfair life situationwhere the consequences oftheir “crime” far surpassed the suffering of the partner:

“I have lost both my driving licence and my job, and Ihave only done nine years of compulsory schooling.…How shall I pick myself up?” (John)

To sumup, at this analytical levelwe can see self-presentationsin which different and situation-dependent characters/positions are presented. These examples also illustrate howpositioning is dependent on the narrated situation and theactual genre. In relation to a “history of awakening”, themen to some extent accept the position as an offender, whilethey as victims can distance themselves from being abusers.

2. Interactive accountsThe second level is directed towards the narrative (interview)situation itself and how the narrator positions himself or ispositioned in interaction with the interviewer. However, aswill be shown, the stories are also directed towards a fictive“other”, because it seems as if the men address someoneother than the researcher with their stories. Moreover,they also address themselves, and the “telling” becomes asituation where the interviewees can convince themselvesthat their version of what happened is the right one. Thus,the interactive character resulted in the use of differentnarrative strategies by which the men were focusing onmaking sense of the violence and their behaviour in relationto the interviewer (cf. Mullaney, 2007) and other potentiallisteners.According to theories of attribution, observers usually seekexplanationswithin the actors (here, men inclined to violence),such as that something is wrong with them, while the actorstend to use external or situational redefinitions to explain andso make excuses for their own behaviour (Augoustinos &Walker, 1995). There are both direct and indirect explanationsfor the violence in the interview stories, and the directexplanations have a more obvious interactive character. Oneindirect explanation is when themen tell touching life storiesas a background to what happened later on. The violence isthen presented as a consequence of special life circumstances.John described how he and his wife had had rough times;they stopped communicating and drifted apart. She lostinterest in him, making him jealous, and it all accumulateduntil it exploded under the influence of alcohol. So all thisfunctions as indirect explanations ofwhat happened and howalcohol was the direct situational trigger.Another man (Mikael) mentioned his cultural background asa direct explanation for his behaviour; he had threatened tokill his girlfriend's parents. He said that, according to hisbackground, it was normal to use this kind of abusivelanguage and to make threats. Living a life under pressure isalso used as a direct explanation; for example, one man(Allan) talked about stress and “being an old lecher” asreasons for his violent behaviour, while Kevin describedchildlessness as driving a wedge between him and his wife.The violence seems, in these cases, to be reactive (cf.Mullaney,2007); from the men's point of view, it is a “reasonable”reaction to an untenable situation. Another example of this is

Page 102: Womens Studies International Forum 46 (2014)

100 K. Edin, B. Nilsson / Women's Studies International Forum 46 (2014) 96–106

when one of the interviewees (Mikael) described how hisgirlfriend forced him to quit soccer training and how thisindirectly “contributed” to the situation of violence.All these explanations also have the character of excuses,and excuses can be a way of denying responsibility becausethey state that under “normal” circumstances there wouldnot have been any violence. Thus, excuses draw a linebetween the violence and the “real” me of the perpetrator(cf. Mullaney, 2007).Some men described experiences of a strange self or beingout of character (cf. Hearn, 1998). They did not recognisethemselves in the situations where violence occurred, i.e.they did not see themselves as being perpetrators andthought they were a different person at the relevant time.This is both an explanation for the use of violence and a kindof negotiation with the position as perpetrator. However,this negotiation also (re)produces the “perpetrator” as beingsomething odd and unusual (see also analysis, level 3), andbeing someone they usually are not; Simon said, “It is notme standing there.”Allan acted similarlywhenhe described the violent situation:“I don't recognize myself in this.” This is one way of con-structing otherness, and by this he can draw a line betweenhis true self and the offender, who becomes somebody else.The responsibility for the violence is transferred to a “fictive”person.One observation can be made here regarding the interac-tional character of identity constructions. The men usedifferent explanations and excuses in their efforts to makesense of the violence in relation to the interviewer andother potential listeners. This also means that they arereconstructing masculine identities through processes ofdis-identification — a theme further developed under theheadings below.

3. Discursive effects — masculinity and changing subjectpositionsAt this third level, we look at the effects of the interviewstories in terms of discourses and discursive positions: athow the narrator, directly or indirectly, positions himselfin relation to “cultural discourses and normative (social)positions, either by embracing them or displaying neu-trality, or by distancing, critiquing, subverting, and resistingthem” (Bamberg, 2004). How can the relation be describedbetween, on the one hand, subject positions in singleinterviews and, on the other hand, more general discourses?How can the interview stories be understood as (re)productions of identity and gender?

“The doing of men's talk is just one more aspect of thesocial construction of men and masculinities […]. Men'stalk about violence is not an effect of the past but it is menthemselves doing masculinities in the present” (Hearn,1998, p. 213).

Identification — re-establishing masculinity

An important event in the life stories was when the mendescribed the violence and admitted being violent. Crossingthis border and resorting to violence is degrading for a man,and the interviewees used different (narrative) strategies to

re-establish a lost status (cf. Mullaney, 2007) and an acceptablemasculine identity. One was showing an ability to take control.This is similar to Flinck & Paavilainen's findings (2008) re-garding men's reasons for using violence in the first place,which are an example of “pursuing a sense of control”. One ofthe interviewees described his technique for handling andtaking control over the violent side of him. When he felt upset,he left the house and went for a walk, which was a strategy hehad used on several occasions (Allan).

One seemingly paradoxical aspect of the (re)constructionof identity was that one man described himself as beinghen-pecked. This was evident when Mikael told a story of arelationship in decline and how his partner changed andbecame more and more cocky. She did not respect himanymore; she laughed at him, calling him a mummy's boy.Earlier, we described this as an indirect explanation for theuse of violence. If we now analyse the story at level 3,another observation can be made. Mikael presents a wholescenario in which he becomes less of a man. In the end, hecould do nothing. She had all the power and she usedhis earlier violent behaviour to blackmail Mikael. Eventheir child became, according to him, a “weapon” in theirrelationship:

“Then I came to her just because she wanted it, but shenever did anything to make me happy during that period.She never made me happy. She only made me damnedangry all the time, or damned angry … she made me feelboth subordinate and unimportant and all of that, youknow.”

How can this “anti-masculine” self-description be under-stood in terms of discursive positions? How is it possible for theinterviewee to present himself as being hen-pecked? Again, thevictim position is of importance. A victim position is usuallyassociated with passivity, fear and introverted behaviour,“traits” that are often regarded as feminine. However, in thecontext of violence and with the interview as a situation of“judgement”, the victim position becomes something “posi-tive”. It can be used as a rhetorical mode to explain unaccept-able behaviour — a victim has the right to “defend” himself. Tohit a woman is taboo, and because he crossed this boundary heneeds some kind of explanation or excuse. By presentinghimself as a victim too, he can save face.

What, then, is the effect in terms of gender? One effect isthe (re)construction of a traditional feminine position. Thiswas evident when Mikael described himself as the victim of awoman who represented something non-feminine. He wasnot a victim of some really powerful enemy, but of a womanwho had abandoned a “true” femininity. Mikael also (re)produced a morally superior and stoical masculinity when hedescribed how his girlfriend used violence:

“I have had a cup thrown at my back and things, a coffeecup at my back, and a remote control at my head, youknow, and I have been hit by her without retaliating.”

Thus, he both repositions the identities of victim versusoffender and (re)constructs a stoical masculinity – he wasnot the first to use violence – and he thereby indirectlylegitimates his own behaviour.

Page 103: Womens Studies International Forum 46 (2014)

101K. Edin, B. Nilsson / Women's Studies International Forum 46 (2014) 96–106

Mikael also made statements about gender and ethnicitywhen he mentioned his non-Swedish girlfriend. He pointedout that it is easier to live with Swedish girls because they aremore into gender equality. The effect of this is two-fold. Onthe one hand, he (re)produces an old-fashioned masculinitywhere the man is a real man and hence neither subordinatednor belittled as a “mummy's boy”. On the other hand, heestablishes a modernmasculinity in line with gender equalityideologies, positioning himself as a representative of thismasculinity. Given that Sweden is a country where genderequality ideologies have had a major impact on public debate(Hausmann et al., 2010), through his narrative he can usegender equality ideas to both explain and compensate for hisdeeds.

(Dis)identification — negotiations with masculinities

As mentioned above, a self-controlled masculinity is re-produced in the interview narratives. However, the identifi-cation process is often complex and can seldom be describedin terms of a straight reiteration. Masculinities are oftenblurred by different demands and discourses. At a local level,situational aspects interfere with both re-constructions ofand negotiations with hegemonic masculinities. This wasevident when Peter described a situation where he scoldedhis girlfriend:

“Yes, I told her off. I did not use any four-letter words oranything, but I used very strong language. I told her offaccordingly … Yes, and I was mad, bloody mad then; Iprobably looked really dangerous then. You see, I havethat disadvantage because I'm a big guy and then you lookthreatening as soon as you … like if you suddenly get upfrom a table and kind of turn towards someone and startranting. Then you become threatening. That's how it is.”

In this context it can be a disadvantage to be big and strongand to use rough language. In other contexts, e.g. working-classmasculine cultures, these characteristics are usually powerfulmasculine “traits” (e.g. Willis, 1977). Thus, an establishedmasculine discourse does not necessarily carry a singlemessage. Instead, it can be described as an element with multi-ple possible meanings and it is created depending on certaincircumstances (Hall, 1996).

In the example mentioned above, strong and physicalmasculinity is viewed as something negative. This is because itis articulated by elements or signs that are seen as “dangerous”,“threatening” and “ranting”. Thismasculinity becomes a burdenand a sign of a negative masculine position. However, theinterviewee Peter did not fully identify himself with thismasculine position. He described it in terms of “you”: “Thenyou become threatening.” This generalisation is a kind ofdisidentification (Skeggs, 1997) and a description of how henormally is not. The disidentification was enforced when Peterrelated his behaviour to his profession by which he knows howwrong it is to act threateningly, because it only triggers otherpeople.

Thus, characteristics of the narratives are the processesof identification and the disidentification. These processes aresometimes straightforward, e.g. when an interviewee de-scribes himself as a special “type” of person. However, it is

more common that the processes are complex and char-acterised by constant considerations and negotiations. Theinterviewees admit that they are perpetrators, but at thesame time describe themselves as something other than atypical perpetrator; as Simon said, “It is not me standingthere.” Then the processes of identification and disidenti-fication intertwine.

(Dis)identification — negotiations with theanti-violence programme

How are the treatment sessions visible as narratives in theinterviews and how are these part of the interviewees' identityproduction? An important aspect of this identity construction isthemen's differentways of showing their disidentificationwiththe programmes' aims and intentions. On an empirical level,this disidentification can be identified in the interviewees'critiques and rejections of the programme.

Some of the interviewees disagreed with having toparticipate in the programme at all, because they thought itwas unjust that more violent and worse offenders than themescaped both punishment and treatment on the programme:

“And then you think, there must be things that are muchworse and how the others are even being violent rightnow.” (Allan)

Others were openly critical towards the programme:

“I mean, there are … just because one of us has battered,that does not mean that everyone in the group has donethe same thing, irrespective of whether you have beenconvicted or not. Obviously, if one person ignores a redlight, not everyone needs to ignore a red light … It's kindof… I don't get the point sometimes. And I have told themall the time that I have not hit her. I have not done thatand that and that.” (Adam)

Adam's critique is rooted in his construction of a victimposition. He not only presents himself as a victim of wrongaccusations, but he also disidentifies with a position as anabuser. This disidentification has similarities with Wood's(2004) concept “dissociations”, which refers to how abusivemen attempt to disconnect themselves from the violence andthe identity of “real” abusers.

This was also the case when the interviewees were criticalof the collective nature of the programme. They thought thatit was built on generalisations and that all of the participantswere treated in the same way, as if they had all been jealousand were perpetrators who had hit women. Both Adam andKevin emphasized that the accusations of violence werewrong, and Kevin said that he had never been jealous andalso thought that he was in the wrong place:

“I did not recognize anything of what they were talkingabout.”“That programme was a complete waste of timefor me, I think.”

The interviewees also viewed the programme as beingpart of their punishment (cf. Miller, Gregory, & Iovanni,2005). They had to attend the sessions and they were longingfor the time when they were over. In the interview with

Page 104: Womens Studies International Forum 46 (2014)

102 K. Edin, B. Nilsson / Women's Studies International Forum 46 (2014) 96–106

Simon, this was obvious. When he was describing theschedule for the treatment sessions, he said:

“We have four weeks off in July. But I have to come hereon my holiday anyway.”

According to Simon, he had to attend once a week evenduring his holiday, but then he was “free” in July. If he failedto appear at the sessions he would be sentenced to prisonfor six months. Thus, some of the interviewees were criticalof the programme and they could not identify themselveswith the position (of a punished abuser) offered by theprogramme.

However, in other parts of the interviews the mendescribed their participation in the programme as apositive experience, and as an opportunity to develop andchange:

“You do learn things … I'm sure you do learn somethingfrom this. With those two sitting over there, I mean. Whatis he … a psychotherapist? And the other one is some-thing. And well, they don't really say much now but wetalk. Then they write some things on the blackboard andwe talk about it in the group then, you see. So of courseyou do get some different ideas compared with drawingyour own conclusions. And then you might draw adifferent one, because the majority agrees and then youmight say that this is better than what I would do myselfand stuff like that.” (Mikael)

In this case, the collective character of the treatment is,contrary to Adam's descriptions (see above), regarded assomething relatively good. To have the opportunity to shareexperiences and views with others is seen as one way oflearning about new perspectives and avoiding wrong con-clusions. However, an ambivalent attitude towards theprogramme was more common. This is observable in theinterview with Allan. He thought the programme was good,but:

“first you feel extremely vulnerable and … Yes, you reallydo feel that you are what you are, a big villain […] This isterrible, to sit in front of six, seven, eight other people andthen tell them that you … yes, but even on the second orthird time, it felt so natural. So I think going twenty timesis a necessity in order to make you feel that you havecleansed yourself a bit.”

Thus, this interviewee is positive towards the programmeand he sees it also as an opportunity to cleanse himself.However, he also seems to have problems admitting hiscrime, and he emphasizes how terrible it is to talk about his“deeds” and experiences in front of others.

One concluding example is taken from the interview withSimon. When he described the treatment, he was a little bitcritical and said he did not like role-plays, because they areboring and tough, and they create anxiety. He did not want tomess up and make a fool of himself. However, he also saidthat he thought about the programme every day after asession:

“And you think you have got the tools to get by withoutusing violence from now on?” (KEE)

“Yes, these are better than none at all. Then you have gotsome things to think about too. So then when you comehome, after you have been here, you also reflect a bit. So you,you've been here once a week but you think about it a littleevery day, to some extent.” (Simon)

This example is in some ways typical of the interviewsand the interviewees' identity production. Stories about thetreatment programmes are not stories about dramaticchanges in behaviour or revelations, but rather they areexamples of how the interviewees navigate between differ-ent discourses and demands.

According to a general therapeutic discourse, the partic-ipants in the programme have to take responsibility, admit tobeing real villains and accept the position as an offenderbefore they can become better people (cf. Schrock & Padavic,2007). This is a common way of thinking in therapy ingeneral, and it reflects the former discussion of the confes-sion mode. A change of behaviour is dependent on theperson's understanding and interpretation of the event assomething not acceptable (cf. Ludwig, 1985). Thus, theidentity as a perpetrator has to be established and theindividual has to accept this before any change can occur andhe can be reformed and allowed to return to society. Theinterviewees seem to be aware of this, which also is apossible explanation for their occasionally positive descrip-tions of the programme in a positive manner — they “talkedthe talk” (cf. Schrock & Padavic, 2007).

However, they often explicitly disagreed with the pro-gramme's content, and these tendencies towards disiden-tification are in some ways self-evident: no one would like tobe defined as an abuser. However, the disidentification canalso be understood by making references to gender. A manwho abuses women (and children) is not a “real” man inrelation to general gender discourses; he does not haveenough self-control and he does not live up to the demandsof a breadwinner (Gerzon, 1982) — a “real”man is a providerwho takes care of and is responsible for the family. Suchdemands are (indirectly) reproduced in the interviewnarratives, for example when the men deny or try to explainthe deviant behaviour that they have been accused of. As wasillustrated earlier, they use different narratives, excuses andideas about a strange self and being out of character in theirnarratives. In this way they seem to minimize or neutralizethe severity of the behaviour, disidentify with the position asan abuser and (try to) re-establish a respectable masculineidentity. “Normally”, they represent a chivalric manhood,care about women and never abuse them (cf. Wood, 2004).Thus, by disidentifications and allusions to a strange self andbeing out of character, they can confirm their morality andalso avoid the need to develop explicit plans for reformingtheir lives (cf. Presser, 2008).

Discussion

Violence and (changing) gendered identity constructions

This paper is about the narratives of men attending threesimilar anti-violence programmes within the ProbationService in Sweden. We aim to understand their gendered

Page 105: Womens Studies International Forum 46 (2014)

103K. Edin, B. Nilsson / Women's Studies International Forum 46 (2014) 96–106

identity constructions — how men articulate the course ofviolent events and in what way they talk about themselvesand the programmes.

From the findings, we argue that men construct ways ofdefending themselves as excuses and explanations and usedthem to adopt victim positions. In the following section, wewill discuss how the men's identity constructions mightcollide with the programme's ambition of changing men'sbehaviour. We also want to challenge the programmes withregards to what extent they could more actively involve themen in constructing new identities while condemning theirold ones. Before developing our arguments, we will presentsome research on the issue of change.

Thus far, there has been no clear-cut consensus in re-search on what pronounced main factors of change can beidentified, if any. When scrutinizing evaluations of batterer-treatment programmes in general, it is important to remem-ber that about half of the men drop out and that they aresignificantly different from the completers (Daly et al., 2001).Moreover, some research indicates that the key to changingviolent behaviour in men attending programmes might comefrom factors not even included in the agenda, such as men'scontact with the criminal justice system where there areconsequences connected to the family, which shakes them upenough for them to realize they have a serious problem(Silvergleid & Mankowski, 2006). When men attend anti-violence programmes, many processes have already startedand continue in parallel to each other. Hence it is difficult todistinguish cause and effect (Silvergleid &Mankowski, 2006).

Moreover, changing behaviour is not a straightforwardprocess and it varies between situations and people (Lehmann& Simmons, 2007). When following a process of change in maleviolent behaviour, it seems quite obvious what the varioussteps and concepts look like, and key issues can be identified,such as increased responsibility, better understanding andcommunication skills, and improved independence (Scott &Wolfe, 2000). The reasons behind such successful changesmight be a reduction of factors strongly correlated with violentbehaviour such as anger, mental health problems and sub-stance abuse (Augusta-Scott, 2009). Most theorists agree thatchanges take place by degrees, moving from denial towardsmotivation, dedication and a permanent change (ibid.). Theimplication of this is that a change is not at all possible for aperson in denial, i.e. moving from denial means approachingnon-violence (Scott & Wolfe, 2000). Whatever the changelooks like, the critical issues are still about what makes menchange and why some do and some don't. Where is the startbutton for change?

The analysis in this paper has shown that identityconstruction processes to some extent “compete” with theprogramme's ambition of change. From earlier research (Edinet al., 2008, 2009) and from how the programmes presentthemselves (Socialstyrelsen, 2001), it has been found that theintention of the group sessions is to openly communicate,show feelings and unreservedly take the blame. These waysof behaving contradict several characteristics of the mascu-line identities usually reconstructed in the narratives. How-ever, even with this disparity, we do not know whatinfluence the programmes have actually had on the menwho attended and were later interviewed. The men mighthave started a process of change even if they had not reached

an expected and desired level from the programmes' pointof view. Nevertheless, we believe it is important to understandthis identity (re)construction as one way to consider thecomplexities in the interventions and, through this under-standing, hope to increase the possibilities of change (cf. Edley,2001).

One seemingly paradoxical masculine identity position isthe victim. The interviewees described themselves as victimsof their own childhood, unemployment, the programme, a“strange self” and sometimes also of their wife/partner. Thisposition made it possible for the men to “save face” despitethe use of violence against a woman, which traditionally isseen as a non-masculine act. In other words, due to the victimposition, the interviewees could present themselves asrelatively normal men and the violence as an exception.Against that background, the need for change does not seemto be particularly acute.

Thus, the victim position is, together with other excuses,explanations, and redefinitions, a strategy to secure apotentially threatened masculine identity. These strategiesare dependent on the narrated situation and context, whichcan be related to how the reconstruction of hegemonicmasculinities depends on local contexts and situationalaspects (cf. Connell & Messerschmidt, 2005). One concreteexample is that the men in one situation could reconstruct a“modern” masculinity with references to Swedish genderequality discourses, while, within a different context, theyreproduced a traditional masculinity as breadwinners takingresponsibility for their families. Different and sometimescontradictory constructions such as these might be one ex-planation for the “fragmented” interview self-presentationsand for the tendencies to deny being violent. However, theyalso illustrate processes of domination, because the men could,by reproducing gender equality discourses as well as aspects ofthe breadwinner, sustain a viable and superior position. Theyseemed in both cases to represent a “real”man. Such processesof domination can also represent hindrances for change. Mentalked about the programme sessions in a context wherecommunication is encouraged and there are opportunities totalk about the violent events. In that way they seemed to becorresponding to something good, at least from a therapeuticdiscourse point of view. However, the men's search for acredible identity, in a context where they seemed to have feltmore or less accused, probably does not represent a goodstarting point for change.

What then, can we learn from these interviews regardinganti-violence strategies?

The programmes' intentions to change men's valuesmight appear to work when one listens to what the men inthis study said about sessions that provided opportunities totalk about the violent events. However, when one scratchesbeneath the surface, as in this study, something else may berevealed. The group-session agenda might have failed tocreate a context where it was possible to talk about difficultthings such as setbacks, ambivalence (Dallos, 2006) andstruggles in their daily life interactions (Edin et al., 2008,2009). Exercising responsibility in order to live up to theprogrammes' expectations might cause the men to feelshame, something they have learned to stay away from,but which makes men employ an array of excuses,rationalisations and denials in their efforts to secure a

Page 106: Womens Studies International Forum 46 (2014)

104 K. Edin, B. Nilsson / Women's Studies International Forum 46 (2014) 96–106

potentially threatened masculine identity (Lehmann &Simmons, 2007).

Instead, if men could get more support to realize thatidentities are never fixed but are open to change, this mayopen up possibilities and encourage a self-chosen responsi-bility to decide to build up a different masculine behaviour(Edley, 2001; Lehmann & Simmons, 2007). In programmes,such an alternative could take the form of a discussion withmen to make them understand their violent manner andthat outbreaks of violence are not isolated events but areinterrelated to other parts of daily life (Hearn, 1998). Fromthat point on, men might recognize how they can adjust theirbehaviour and take on different identities and behaviours indifferent situations. This is in line with research showing howindividuals need to actively plan and take control in order tobridge the big gap between changed beliefs and actuallychanging behaviour in real life (Scholz, Schüz, Ziegelmann,Lippke, & Schwarzer, 2008; Sniehotta, 2009). Violent menoften keep their violent behaviour separate from everydayissues instead of seeing the close interaction between them(Hearn, 1998).

Previous research has indicated that programmes canunintentionally encourage men to take only pseudo-responsibility for their actions (Schrock & Padavic, 2007).Studies of different anti-violence programmes in Sweden,including this studywhich looked at three similar programmeswithin the Swedish Probation Service, have revealed rathergender-stereotyped and unreflected professional discoursesand how the agendas in point had also left outmany importantinter-related everyday issues, such as sexuality, children andparenthood (Edin et al., 2008, 2009). Moreover, if gender isseen as an interactive factor, men's violence ought to beunderstood by using an integrated approach, and this meansincluding both men's violence towards other men and men'sviolence towards women (Schofield, Connell, Walker, Wood, &Butland, 2000). Bringing in discussions about such mattersmight be key tomakingmen plan to actively take steps to bringabout a change, not only when it comes to violence but in theclose interactions with men, women and children in daily life,where masculine behaviour is constructed (Edley, 2001;Hearn, 1998).

Besides the use of gender theories to understand thedifficulties in making men change violent behaviour, thereare other more general theories that might be helpful. Withinbehavioural science, theory of reasoned action (TRA) hasbeen predominant in efforts to understand and supportbehavioural change. Lately, this theory has been questionedbecause there seems to be no linear process from changingbeliefs to changed behaviour. The best intentions of changeproduce no transformation without an implementationintention approach, i.e. the person takes control in actingand planning how the change will take place (Scholz et al.,2008; Sniehotta, 2009).

Certain alternative and seemingly reliable interventionmethods challenge programmes similar to the ones in thisstudy and correspond fairly well with the ideas describedabove. Some programmes are currently being developedfrom evidence-based methods, where several of thesemethods converge to a large extent and are in line withso-called strength-based approaches (Babcock et al., 2004).One example of such a method is ‘Motivational interviewing’,

which is supporting men's readiness to end their violentbehaviour by bringing forth motivation from within theclient and seeing ambiguity as an expected step toovercome in the non-linear process of change (Dia et al.,2007). Another is ‘Narrative therapy’, which endeavours tounderstand and externalize ideas about male identities,giving violent men a choice to take a new stance towardsdominant and hegemonic masculinities by understandingthe subsequent consequences on themselves, their part-ners and children and with that possibly taking responsi-bility (Augusta-Scott, 2009). We are not certain that theseintervention methods would have been suitable for themen in our study, but we do think it is important toinvestigate new ways of thinking and novel types oftreatment which may help bring about the possibility ofchange for men who have engaged in violence againsttheir partners.

To sum up, in this paper we have scrutinized genderconstructions by analysing the narratives told bymen attendinganti-violence programmes within the Probation Service inSweden. One important finding was that the men's identityconstructions seemed to collide with the programmes' ambi-tions of changingmen's behaviour. The reason for this might bethat the programmes are based on the men's “unconditionalsurrender”, which supports men's various ways of defendingthemselves by using excuses and explanations and in that wayconstructing victim positions. Thus, in changing violent men'sbehaviour it seems just as important to actively involve men inthe construction of (new) identities as it does to condemn theold ones. How this should take place is a question for futureresearch; however, men's engagement with treatment has,significantly, been described as an important factor in decreas-ing the reluctance to change (Scott & King, 2007; Scott &Wolfe,2000).

References

Alexander, Pamela C., & Morris, E. (2008). Stages of change in batterersand their response to treatment. Violence and Victims, 23(4),476–492.

Augoustinos, Martha, & Walker, Iain (1995). Social cognition, an integratedintroduction. London: Sage.

Augusta-Scott, Tod (2009). Narrative therapy: Addressing masculinity inconversations with men who perpetrate violence. In Peter Lehmann, &Catherine Simmons (Eds.), Strengths-based batterer intervention: A newparadigm in ending family violence. New York: Springer PublishingCompany.

Babcock, Julia, Green, Charles, & Robie, Chet (2004). Does batterers' treatmentwork? A meta-analytic review of domestic violence treatment. ClinicalPsychology Review, 23(8), 1023–1053.

Balkmar, Dag, Iovanni, LeeAnn, & Pringle, Keith (2008). A reconsiderationof two “welfare paradises”: Research and policy responses to men'sviolence in Denmark and Sweden. Men & Masculinities, 12(2),155–174.

Bamberg, Michael (2004). Form and function of ‘slut bashing’ in maleidentity constructions in 15-year-olds. Human Development, 47,331–353.

Burman, Monica (2010). The ability of criminal law to produce genderequality: Judicial discourses in the Swedish criminal legal system.Violence Against Women, 16(2), 173–188.

Buttell, Frederick P. (2003). Exploring the relevance of moral developmentas a treatment issue in batterer intervention. Social Work Research,27(4), 232–241.

Cavanagh, Kate, Dobash, Emerson, Dobash, Russell, & Lewis, Ruth (2001).‘Remedial work’: Men's strategic responses to their violence againstintimate female partners. Sociology, 35(3), 695–714.

Chatman, Seymour (1978). Story and discourse. Ithaca: Cornell UniversityPress.

Page 107: Womens Studies International Forum 46 (2014)

105K. Edin, B. Nilsson / Women's Studies International Forum 46 (2014) 96–106

Coles, Tony (2009). Negotiating the field of masculinity: The production andreproduction of multiple dominant masculinities. Men and Masculinities,12(1), 30–44.

Connell, R. W. (1995). Masculinities. Los Angeles: University of CaliforniaPress.

Connell, R. W. (2005). Masculinities (second ed.). Cambridge: PolityPress.

Connell, R. W., & Messerschmidt, James (2005). Hegemonic masculinity:Rethinking the concept. Gender and Society, 19(6), 829–859.

Dallos, Rudi (2006). Attachment narrative therapy. Integrating narrative,systemic and attachment therapies. Maidenhead: Open UniversityPress.

Daly, Jennifer, Power, Tomas, & Gondolf, Edward (2001). Predictors ofbatterer program attendance. Journal of Interpersonal Violence, 16(10),971–991.

Demetriou, Demetrakis (2001). Connell's concept of hegemonic masculinity:A critique. Theory and Society, 30(3), 337–361.

Dia, David, Simmons, Catherine, Oliver, Mark, & Cooper, Lyle (2007).Motivational interviewing for perpetrators of intimate partner violence.In Peter Lehmann, & Catherine Simmons (Eds.), Strengths-based battererintervention: A new paradigm in ending family violence. New York:Springer.

Dobash, R. Emerson, Dobash, Russell P., Cavanagh, Kate, & Lewis, Ruth(2000). Changing violent men. Thousand Oaks, CA: Sage.

Edin, Kerstin (2006). Perspectives of intimate partner violence, focusing on theperiod of pregnancy. (Medical Dissertation). New series no. 1041. UmeåUniversity.

Edin, Kerstin, Högberg, Ulf, Dahlgren, Lars, & Lalos, Ann (2009). Thepregnancy put the screws on: Discourses of professionals workingwith men inclined to violence. Men and Masculinities, 11(3),307–324.

Edin, Kerstin, Lalos, Ann, Högberg, Ulf, & Dahlgren, Lars (2008). Violentmen: Ordinary and deviant. Journal of Interpersonal Violence, 23(2),225–244.

Edley, Nigel (2001). Analysing masculinity; interpretative repertoires,ideological dilemmas and subject positions. In Margaret Wetherell,Stephanie Taylor, & Simeon J. Yates (Eds.), Discourse as data, a guide foranalysis. London: Sage.

Eliasson, Per E. (2000). Men, women and violence: Understanding andchanging violent behavior: A report from Manscentrum in Stockholm.Bjärnum: Carlsson bokförlag.

Eriksson, Maria, & Pringle, Keith (2005). Introduction: Nordic issues anddilemmas. In Maria Eriksson, Marianne Hester, Suvi Keskinen, & KeithPringle (Eds.), Tackling men's violence in families: Nordic issues anddilemmas. Bristol: Policy Press.

Feder, Lynette, Wilsson, David B., & Austin, Sabrina (2008). Court-mandatedinterventions for individuals convicted of domestic violence. CampbellSystematic Reviews, 12.

Firestone, Juanita, Harris, Richard, & Vega, William (2003). Theimpact of gender ideology, male expectancies, and acculturationof wife abuse. International Journal of Law and Psychiatry, 26(5), 549–564.

Flinck, Aune, & Paavilainen, Eija (2008). Violent behaviour of men in theirintimate relationships, as they experience it. American Journal of Men'sHealth, 2(3), 244–253.

Gadd, David (2002). Masculinities and violence against female partners.Social & Legal Studies, 11(1), 61–80.

Gelles, Richard (2000). Public policy for violence against women: 30 years ofsuccesses and remaining challenges. American Journal of PreventiveMedicine, 19(4), 298–301.

Georgakopoulou, Alexandra (2007). Small stories, interaction and identities.Amsterdam: John Benjamins.

Gerzon, Mark (1982). A choice of heroes. The changing faces of Americanmanhood. Boston: Houghton Miflin Company.

Giddens, Anthony (1991). Modernity and self-identity. Self and society in thelate modern age. Cambridge: Polity Press.

Glynos, Janos, & Howarth, David (2007). Logics of critical explanation in socialand political theory. New York: Routledge.

Gondolf, Edward (2004). Evaluating batterer counseling programmes: Adifficult task showing some effects and implications. Aggression andViolent Behaviour, 9(6), 695–713.

Hall, Stuart (1996). Introduction. Who needs identity? In Stuart Hall, &Paul du Gay (Eds.), Questions of cultural identity. London: SagePublications.

Hausmann, Richardo, Tyson, Laura, & Zahidi, Saadia (2010). The globalgender gap report 2010. Geneva, Switzerland: World Economic Forum.World.

Hearn, Jeff (1998). The violences of men. London: Sage.Hearn, Jeff, & McKie, Linda (2010). Gendered and social hierarchies

in problem representation and policy processes: “Domestic

violence” in Finland and Scotland. Violence Against Women,16(2), 136–158.

Heise, Lori (1998). Violence against women: An integrated, ecologicalframework. Violence Against Women, 4(3), 262–290.

Hirose, Akihiko, & Kay, Pih (2010). Men who strike and men who submit:Hegemonic and marginalized masculinities in mixed martial arts. Menand Masculinities, 13(2), 190–209.

Kimmel, Michael (1987). Rethinking “masculinity”: New directions inresearch. In Michael Kimmel (Ed.), Changing men, new directions inresearch on men and masculinity. London: Sage.

Laclau, Ernesto, & Mouffe, Chantal (1985). Hegemony and socialiststrategy: Toward a radical democratic politics. London & New York:Verso.

Lehmann, Peter, & Simmons, Catherine (Eds.). (2007). Strengths-basedbatterer intervention: A new paradigm in ending family violence. NewYork: Springer.

Ludwig, Arnold (1985). Cognitive processes associated with “spontane-ous” recovery from alcoholism. Journal of Studies on Alcohol, 46(1),53–58.

Lusher, Dean, & Robins, Garry (2009). Hegemonic and other mas-culinities in local social contexts. Men and Masculinities, 11(4), 387–423.

Miller, Susan L., Gregory, Carol, & Iovanni, LeeAnn (2005). One size fits all? Agender-neutral approach to a gender-specific problem: Contrastingbatterer treatment programs for male and female offenders. CriminalJustice Policy Review, 16(3), 336–359.

Mishler, Elliot (1999). Storylines: Craftartist's narratives of identity. Cam-bridge: Harvard University Press.

Mullaney, Jamie (2007). Telling it like a man. Masculinities and batteringmen's accounts of their violence. Men and Masculinities, 10(2),222–247.

Pence, Ellen, & Paymar, Michael (1993). Education groups for menwho batter, the Duluth model. New York: Springer PublishingCompany.

Presser, Lois (2008). Been a heavy life. Stories of violent men. Urbana:University of Illinois Press.

Råkil, Marius (Ed.). (2002). Men's violence against women (Onlyavailable in Norwegian: Menns Vold mot kvinner). Oslo:Universitetsforlaget.

Renzetti, Claire, Edleson, Jeffrey, & Bergen, Raquel (Eds.). (2001). Sourcebookon violence against women. London: Sage.

Riessman, Catherine K. (1993). Narrative analysis. Newbury Park, CA:Sage.

Schofield, Toni, Connell, R. W., Walker, Linley, Wood, Julian, & Butland,Dianne (2000). Understanding men's health and illness: A gender-relations approach to policy, research, and practice. Journal of AmericanCollege Health, 48(6), 247–256.

Scholz, Urte, Schüz, Benjamin, Ziegelmann, Jochen, Lippke, Sonia, &Schwarzer, Ralf (2008). Beyond behavioural intentions: Planningmediates between intentions and physical activity. British Journal ofHealth Psychology, 13(3), 479–494.

Schrock, Douglas P., & Padavic, Irene (2007). Negotiating hegemonicmasculinity in a batterer intervention program. Gender and Society,21(5), 625–649.

Scott, Katreena L., & King, Colin B. (2007). Resistance, reluctance, andreadiness in perpetrators of abuse against women and children. Trauma,Violence & Abuse, 8(4), 401–417.

Scott, Katreena L., & Wolfe, David (2000). Change among batterers:Examining men's success stories. Journal of Interpersonal Violence,15(8), 827–842.

Shamai, Michal, & Buchbinder, Eli (2010). Control of the self: Partner-violentmen's experiences of therapy. Journal of Interpersonal Violence, 25(7),1338–1362.

Shuman, Amy (2005). Other people's stories: Entitlement claims and thecritique of empathy. Urbana: University of Illinois Press.

Silvergleid, Courtenay, & Mankowski, Eric (2006). How batterer in-tervention programs work: Participant and facilitator accounts ofprocesses of change. Journal of Interpersonal Violence, 21(1), 139–159.

Skeggs, Beverly (1997). Formations of class and gender. Becoming respectable.London: Sage.

Smedslund, Geir, Dalsbø, Therese, Steiro, Asbjørn, Winsvold, Aina, &Clench-Aas, Jocelyne (2007). Cognitive behavioural therapy for menwho physically abuse their female partner. Cochrane Database ofSystematic Reviews, 18(3), CD006048.

Sniehotta, Falko (2009). Towards a theory of intentional behaviour change:Plans, planning, and self-regulation. British Journal of Health Psychology,14(2), 261–273.

Socialstyrelsen (2001). Dare to see – be able to act, Utväg Skaraborg – workingtogether for a safe haven for women (Only available in Swedish: Våga

Page 108: Womens Studies International Forum 46 (2014)

106 K. Edin, B. Nilsson / Women's Studies International Forum 46 (2014) 96–106

se-kunna handla, Utväg Skaraborg-samverkan för kvinnofrid). Stockholm:Copie Center.

Somers, Margaret (1994). The narrative constitution of identity: Arelational and network approach. Theory and Society, 23(5), 605–649.

Walby, Sylvia, & Allen, Jonathan (2004). Domestic violence, sexual assault, andstalking: Findings from the British Crime Survey. London: Home Office(Home Office Research Reports 276).

Wetherell, Margaret, & Edley, Nigel (1999). Negotiating hegemonic mascu-linity: Imaginary positions and psycho-discursive practices. Feminismand Psychology, 9(3), 335–356.

Willis, Paul (1977). Learning to labour. How working class kids get workingclass jobs. Farnborough: Saxon House.

Wood, Julia T. (2004). Monsters and victims: Male felons' accounts ofintimate partner violence. Journal of Social and Personal Relationships,21(5), 555–576.

Page 109: Womens Studies International Forum 46 (2014)

Women's Studies International Forum 46 (2014) 107–114

Contents lists available at ScienceDirect

Women's Studies International Forum

j ourna l homepage: www.e lsev ie r .com/ locate /ws i f

The requirement to speak: Victim stories in Swedish policiesagainst honour-related violence

Maria CarbinUmeå Centre for Gender Studies, Umeå University, SE-901 87 Umeå, Sweden

a r t i c l e i n f o

0277-5395/$ – see front matter © 2013 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.wsif.2013.12.002

s y n o p s i s

Available online 22 December 2013

Over the last decade, political initiatives against so-called honour-related violence have beenundertaken in several Western countries, as well as in the UN. Swedish policy initiatives arerelatively ambitious, and have primarily targeted young women as victims, one aim being tomake it possible for them to speak up. In this article the overarching concern is to explore howvictim stories are used in Swedish policy initiatives. Drawing upon discourse theory andpost-colonial feminism, the aim is to challenge the ideal of speech as emancipation and toelaborate the connections between speech, silence and power. The article shows that, despiteefforts by policy-makers to include these young women, and not to reproduce stereotypes, thepossibility of speaking is formulated within a certain nationalist discursive terrain. The victimsare primarily called upon to speak as non-Swedish representatives. Paradoxically, theinclusion of young women into policy discourse has led to a particular exclusion and therebyproduced new silences.

© 2013 Elsevier Ltd. All rights reserved.

Introduction

By the end of the 20th century, single cases of violenceagainst young ethnicised/racialised women were making thenational news headlines all acrossWestern Europe. The violencewas labelled honour related and articulated as exotic, new anddifficult to understand. Following the publicity about ‘honourkillings’, political initiatives have been undertaken in severalcountries aswell as in the UN to combat honour-related violenceand forced marriages. Denmark, for example (and to a lesserextent the UK and Norway) has launched initiatives againstforced marriage as part of their immigration policies and couldbe accused of using them as part of an agenda of restrictionsupon family reunification (Bredal, 2005; Dustin & Phillips, 2008),whereas in Sweden the problem has mainly been constructedas a matter of honour-related violence to be tackled withinimmigrant integration and gender-equality policies. Interesting-ly, it seems as though the issue of ‘honour-related’ violence hasreceived more attention in the Nordic countries than in manyother Western countries, at least judging from the mediacoverage (Keskinen, 2009: 268) and, in terms of policy materialand the amount ofmoney spent, the Swedish state has dealtwiththe issue by far the most extensively. One reason for the Nordic

ll rights reserved.

countries in general and Sweden in particular to devote suchefforts to combating honour-related violence may be traced towhat several researchers describe as the image of Sweden andthe Nordic region in general as the gender-equal society(Keskinen, 2009; Towns, 2002; Tuori, 2007).

The debates and political initiatives against honour-relatedviolence were, however, initiated during the post-9/11 time ofthe so-called War on Terror waged by the Bush and Blairadministrations. Thus, these measures were often launched inthe name of nationalist assumptions about gender equality andintertwined with an agenda of policing immigrant, frequentlyMuslim, communities (Abu-Lughod, 2002; Korteweg &Yurdakul, 2009; Meeto & Mirza, 2007; Phillips & Saharso, 2008;Prins & Saharso, 2008; Wilson, 2007). The war on terror, alongwith the growing surveillance of migrant communities, repre-sents a retreat from liberal multiculturalism and what in Dutchpolicies has been described as a “turn to neo-realism” (Korteweg& Yurdakul, 2009). Thewar on terror has furthermore awakeneda familiar stereotype—the dangerous, fundamentalist brownman who is understood to be too rigid in his views on women(Bhattacharyya, 2008). As a consequence, any identification ofthe specificity and particularity of violence against ethnicised/racialised women in the West is in danger of reproducing an

Page 110: Womens Studies International Forum 46 (2014)

108 M. Carbin / Women's Studies International Forum 46 (2014) 107–114

understanding of Islam and people from the Middle Eastas problematic and particularly hostile towards women (Sen,2005: 42).

Thus, policy-makers have been faced with a dilemma: totalk about ‘honour killings’ as such runs the risk of fuellingracist understandings. On the other hand, formulations of theproblem in universalist terms as part of men's global violenceagainst women risk missing the particularities of thesecrimes. Sweden provides an interesting case in analysingpolicy efforts against honour-related violence not onlybecause the Swedish state has been eager not to constructthe problem as emanating from religious views and as amatter of Islam,1 but also because of the impact of a feministagenda in relation to Swedish policies on violence againstwomen in general (Carbin, 2010; Wendt Höjer, 2002).

Despite major policy efforts and intense political debateon the topic, few academic studies of these policies have beenundertaken so far.2 This article thus examines how theproblem of violence against young women from ethnicised/racialised minorities has been represented within policydiscourses against honour-related violence. The task set uphere is twofold. Firstly, I wish to explore subjectificationprocesses. I intend to illustrate the self-identities available tovictims of violence. This includes an analysis of the potentialeffects in terms of possibilities and restrictions for subjects tospeak politically. Secondly, the article aims to elaboratetheoretically on the connections between silence, violence,speech and power: How are some forms of speech madepossible, whereas other utterances are made impossiblewithin political discourses? What are the terms and condi-tions for speaking? (How) is speech politically enabling?(How) can a feminist post-colonial understanding of theproblem be formulated? Thus, in this article, I intend both toprovide a critical investigation of the complex relationshipbetween speech, silence and power in Swedish policies onhonour-related violence, and to discuss alternative ways ofrepresenting the problem. Before I embark on the analysis, Iwill describe the methods and materials used in the article,and will give a brief overview of feminist postcolonial theory,and how it has been used in the analysis.

Materials and methods

The research material for this article consists primarily ofhandbooks and surveys on honour-related violence fromSwedish County Administrative Boards produced between2003 and 2010. In 2003, nationwide initiatives againsthonour-related violence were launched, which continuedfor the next several years: all Swedish County AdministrativeBoards were given the task of mapping out the extent of theproblem. Some of the boards in the larger counties, alongwith some municipalities in larger cities, not only producedreports on the ‘spread’ of the problem, but also publishedhandbooks on how civil servants could deal with victims. Inthis article I have focused on those documents that includetestimonies by, and stories about, the young women who areportrayed as victims, since the aim is to analyse thepossibilities for victims to speak. The article is thus basedon the analysis of seven out of around 40 of the documentsfrom the County Administrative Boards (see reference list).Added to this, a couple of documents from a women's NGO

(Terrafem—a shelter for immigrant women) have beenincluded (see reference list) in order to explore the extentto which alternative stories can be found outside govern-mental bodies.

In this article, a discourse analytical reading of the policydocuments is used in order to explore the subjectificationprocesses and underlying assumptions within the policies(Winther Jorgensen & Phillips, 1999). Discourse analysis isnot a method in the strictest sense, but rather a way ofapproaching and thinking about a problem. In this case Ihave analysed the representations of the problem bydrawing upon Michel Foucault's work in order to analysewho is able to tell the truth about honour-related violence,the ways in which they are able to talk about the problem,and what the consequences are in terms of power (Foucault,1983). I have searched for utterances or testimonies fromthe victims in the policy material about explanations of theviolence and descriptions of the situation of the victims ofthe violence.

Post-colonial feminist theory

Inmy analysis of the positioning of victims, I am inspired by apostcolonial feminist paradigm. The term postcolonial is hereprimarily used as an analytical concept (and not as a historicalera) in which processes of subjectification are related to colonialstructures. Postcolonial theory provides insights into on-goingand intertwined constructions of race/ethnicity and gender, aswell as exclusions in terms of national belonging and capitalism.Gayatri Spivak (1993), Lila Abu-Lughod (2002) and Sara Ahmed(2002) have, for example, analysed how the establishmentof the white, western, individual, feminist subject requires abinary opposition who becomes the Other woman. This Otherwoman as a victim of patriarchal oppression contributes to theconstitution of the western woman as emancipated, free andautonomous and helpful. Thus, postcolonial feminist theoriesoffer a necessary critique fromwithin feminism, pointing out theimportance of locating feminist struggleswithin colonial settingsand posing questions regarding the naming of ‘others’ as well asthe simultaneous self-representation of ‘us’. The act of naming,the very construction of the policy problem, the naming ofnation, gender and race/ethnicity is thus at the heart of myanalysis. However, these theoretical insights are not only used inorder to analyse the inclusion of the immigrant girl in Swedishpolicy discourse but they are also drawn upon in order to discussthe possibilities of an alternative feminist politics in relation tothe problem of violence against racialised and/or ethnicisedwomen.

Immigrant girls: from margin to centre

At the beginning of the 1990s, few, if any, political efforts inSweden were devoted to combating violence against youngwomen from ethnicised/racialised minorities. At that point intime the problem of violence, threats and control of youngwomen from family members was not taken seriously (de losReyes, 2003; Schlytter, 2004). Nevertheless, only ten years later,initiatives against this type of violence were at the centre ofSwedishpolitical debates (Carbin, 2008). Two cases, themurderof Pela in 1999 and the murder of Fadime in 2002,3 led tointense and polarised media debates in Sweden (Ekström,

Page 111: Womens Studies International Forum 46 (2014)

109M. Carbin / Women's Studies International Forum 46 (2014) 107–114

2009; Hellgren & Hobson, 2008; Reimers, 2005). While manypoliticians, especially from the right, claimed that a naïvemulticultural understanding of migration was part of theproblem, others said that the lack of effort was due to racist orethnocentric understandings of the young women and theirfamilies. Nevertheless, at the beginning of the 2000s, ideals ofmulticulturalism as a positive asset were in retreat (Mulinari etal., 2009). The data analysed here shows that talk of ‘honour’now proliferated in what one could call a discursive ex-plosion. The Swedish language was thereby ‘enriched’ by newwords such as ‘honour-related problems’ (hedersrelateradeproblem), ‘honour-limitations’ (hedersbegränsningar), ‘honour-conflicts’ (hederskonflikter), and ‘honour-related family life’(hedersrelaterat familjeliv). As a consequence, a new politicalsubject position was constituted and introduced into Swedishpolitics: ‘Vulnerable girls’ (utsatta flickor) or ‘Girls with otherethnic and cultural backgrounds’ (Flickor med olika etnisk ochkulturell bakgrund).

“Let the girls speak”

One core theme in the policy discourses analysed hereconcerns the failure of Swedish society to protect youngwomenfrom their so-called patriarchal families. It was stated bypractically every political party, as well as by the government,that Swedish society had ‘betrayed’ youngwomen by not takingthem seriously. This was the starting point for young women toclaim certain rights. There was no prior enabling convention orauthorisation for ‘immigrant girls’ to speak and to tell theSwedish authorities about their situation. In this sense, theefforts of the government on behalf of ‘vulnerable girls’ and ‘girlswith other ethnic and cultural backgrounds’ were to enablethem to speak politically and to make their voices be heard. Itwas also stated in policy documents initially that “we have tolisten to the girls,” since they were regarded as voiceless,silenced and oppressed (Carbin, 2010). The demand for youngwomen to speak emanated partly from an emancipatoryagenda. Indeed, the Swedish authorities wanted to solve theproblem and improve the situation of victims of violence. Forexample, threats and controlling behaviour by family membersagainst young women were more likely to be taken seriously.The need for specific shelters for young women was investigat-ed and victims of violence were not automatically sent back totheir family or other relatives (who could be part of theproblem). Thus, the problem was now recognised as being acollective form of violence within families. The Swedish modelof men's violence against women had up until then focused onindividual men's violence against former or current partners.This specific violence against young women from ethnicminorities did not fit this model, however. As a consequence, itwas not perceived as violence but as ‘generational conflicts’, forexample (de los Reyes, 2003). When the problem wasrecognised as ‘violence’, young women (and men) seekinghelp from the Swedish authorities were more likely to getsupport. To sum up, young women were now able to speak asvictims and thus to receive the support granted to victims ofcrime.

While young womenwere now enabled to speak politicallyand to claim their rights, the opportunities to speak wereformed within certain discursive structures. As Prins andSaharso (2008) put it, being in the political spotlight has

proved to be both a blessing and a curse for immigrant women.In this case, one consequence of the political initiatives againsthonour-related violence was a call upon young women to speakand to articulate their experiences. Instead of being voiceless,‘girls’ were now required to speak, and not only victims ofviolence, but all young women with parents from the MiddleEast in general, and girls with a Kurdish background inparticular.

At the heart of the wish to emancipate oppressed groupsby letting them speak lies a paradox. At the same moment asa subject is called upon, or an event is given a name, thepossibilities of speaking are significantly reduced. This meansthat, at the same time as a specific insight is won, severalother possible insights are lost, as Nikita Dhawan (2007: 62)describes it. The dilemma in this particular case is obvious.With the construction of the problem as ‘honour related’,there already follows a reduction of possibilities. Before 2003,politicians struggled over the question of how the problem ofviolence against young women from ethnic minorities was tobe defined and named. In 2003, this uncertainty andmultiplicity of possible definitions was reduced to ‘honour’.Thus, the opportunity for the subjects to speak was organisedand restricted in advance since the problem was defined as‘honour-related’ violence.

The political goal to “let the girls speak” and “let us listento the girls” is moreover connected with the idea thatspeaking means liberation, and that letting the girls speakup forms an emancipatory policy as such. Therefore, in thefollowing I intend to interrogate the function of thesetestimonies within policy-making: How are stories ofviolence, pain and suffering used in public policy?

Authentic stories

Stories of violence, honour and forced marriages have beencirculating for some time now in Western European publicspaces—not least by way of bestselling novels or drama-documentaries such as the novel “Mariée de force” by Marie-Thérèse Cuny, which describes the life of French-Moroccan Leilawho is married against her will, or the best-selling book “Desertflower” (which also was made into a film) by Waris Dirie andCathleen Miller, which tells the authentic story of a girl inSomalia and deals with genital mutilation. Consequently, ‘we’have become familiar with reports of the oppression of womenin some country far away—preferably in North Africa or theMiddle East. The stories are often marketed as individual,authentic narratives and an imaginary Western reader isassumed to lack knowledge and insight into the strangeness ofthese unknown societies and cultures (Karlsson, 2006). The lackof knowledge on the part of Swedish society is also a centraltheme in the policy field of honour-related violence. Utterancessuch as “this type of violence is new to Sweden” are common.For example, one central theme in policy documents concernsthe idea that “Swedish society lacks fundamental knowledge”and that “Sweden has been unaware” of the problem ofhonour-related violence. The sense of not knowing, of notbeing able to explain or control the situation causes problemsfor Swedish policy-makers and fuels the search for a knowableand easily representable victim, with a coherent narrative. Thus,as a consequence of this lack of knowledge, a need for authentic

Page 112: Womens Studies International Forum 46 (2014)

110 M. Carbin / Women's Studies International Forum 46 (2014) 107–114

stories is articulated in Swedish public policy, as demonstratedin the quotes below:

The handbook offers information about the honour prob-lematic. It entails a short description of honour thinking, itscultural and geographical spread, its connection to familyand clan, but not to a certain religion. This part of thehandbook also includes an authentic description of thesituation for a young woman.

During the last few years, we have gained more knowledgeabout honour-related violence. Above all, through the girls'own stories.

[(Länsstyrelsen Södermanlands län, 2005: 3)

(Socialstyrelsen och Länsstyrelserna, 2005: 10 Nationellt

konsultativt stöd).]

These extracts show that the stories of young women arecentral in the establishment of the new policy field. Hence,one way of acquiring knowledge is through the testimonialsof young women from ethnic minorities. Indeed, their storiesdo provide knowledge about some problems that mightotherwise have been neglected in policy and practical socialwork. The problem is now recognised as violence and effortsare being made to combat it.

There is a tendency to view the victims' stories as theabsolute truth, as if they show reality as it really is. The storiesof young women are thus treated in policy documents aspure, non-ideological, neutral reflections of reality. It is thesuffering of the child that touches the Swedish nation, not thesuffering of the parents, fathers, mothers or women. Thechild or ‘girl’ represents innocence and is seen as beingwithout strategic motives. This means that young women/girls are seen as ideal victims (Žižek, 1999: 299). Despite thefact that almost all policy documents state that both youngwomen and young men can be victims of honour-relatedviolence, the ideal victim is a young woman. Young men arenot articulated as victims in the same way that young womenare. That is, when it comes to the construction of the policyproblem, the young woman as victim has become an identity.Young women can more easily be regarded as ideal victims,as passive and above all as definitely not responsible for theviolence.

The problem as I see it is not so much that young womenare asked to speak and tell their stories. The question israther, as Foucault has put it: “who is able to tell the truth,about what and with what consequences in terms of power”(Foucault, 1983). Interestingly, in this case young womenfrom ethnic minorities have become ‘truth-tellers’ of Swedishpublic policy on honour-related violence, rather than socialworkers, gender researchers or researchers with a compe-tence regarding racism, migration and ethnicity.

Consequently, young women are perceived as both idealvictims and optimal truth-tellers. Not everyone can be seenas an optimal truth-teller. Instead, Foucault summarises therequirements of a truth-teller as a person who speaks out ofduty, who speaks freely and who risks something by tellingher story. In particular the last criterion, to speak againstone's personal interest or to say something that goes againstthe grain, is very important in order for a subject to pass as anoptimal truth-teller. Fadime, for example, who spoke in the

Swedish parliament before her death, has been constructedin political debates as such a truth-teller. Her words havebeen repeated and circulated in parliamentary debates,anthologies and elsewhere.

The question to which I turn next, is what the truth-telleris expected to talk about. We have already concluded that thegirls are required to speak, now let us turn to the content ofthe speech in order explore the relationship between speechand power.

The non-Swedish self

Jamila was 23 years old the first time I met her. It was latesummer 2002 at a café that could be reached through twoentrances. In Swedish society a lot was circulating aroundthe upcoming elections. The election campaigns centredupon schools and public care issues. In front of me sat awoman who made me realise that nothing in modernSweden concerned her. She lived in a medieval worldthat I had only read about in history books. In the middleof Eskilstuna, Jamila sobbed and trembled. It was diffi-cult to understand anything but the words, help me, helpme.

[(Länsstyrelsen Södermanlands län, 2005: 4, s. 13)]

In the search for knowledge about honour-relatedviolence, a recurrent theme can be found in the narratives.Young women are portrayed as markers of the border ofSwedishness (see also Akpinar, 2003). Interestingly, Fadimeand other victims of this violence are seen as truth-tellers inrelation to something wider than just their own individualsituation. They are thus called upon to explain the violenceand to tell Swedish society about ‘their culture’ or theirgeographical background. Individual stories thus becomeparadigmatic examples of ‘how it is’. Added to this, theindividual story should preferably confirm ideas about‘Kurdish/Arabian/Middle Eastern values’ as being the funda-mental problem. Researcher in social work Barzoo Eliassi hasnoted a similar phenomenon in his interview study on howyoung men and women with Kurdish backgrounds negotiatetheir identities within the dominant Swedish society. Youngwomen with a Kurdish background are constantly askedquestions about their families, such as whether their fatherswill allow them to go to the cinema or on a date. As Eliassinotes, when young women say that they are not oppressed,the reception is usually that they as individuals are unique.They are perceived as the exceptions, whereas the fatherwho kills his daughter is constructed as the rule and proof ofthe woman-hostile Kurdish culture (Eliassi, 2010: 188). Thus,individual stories of families in harmony or families withordinary conflicts between parents and children are notregarded as ‘the truth’ about Kurdish culture. The story abouthonour-related violence has to be about values from theMiddle East in order for the speaker to pass as a truth-teller.

The agent of these stories is Swedish society or the Swedish‘listener’ rather than the girl. Here we can see how therepresentation of the Other involves a double representationthat has as much (or even more) to do with the image ofthe ‘Swedish’ listener or viewer. To represent and reflect thevoice of the victim lends the representative an aura of goodnessand pride. A feeling of being the voice of the victim also gives

Page 113: Womens Studies International Forum 46 (2014)

111M. Carbin / Women's Studies International Forum 46 (2014) 107–114

legitimacy to the policy efforts. The embodiment of the reader isthe figure of a Swede who should feel better about stories ofsuccess and that the girls have been saved from their families.The process of saving someone, as Lila Abu Lughod describes it,not only involves the rescue of someone from, but equallyimportant is the fact that it means to save the person tosomething (Abu-Lughod, 2002: 789). In this endeavour,presumptions are being made about the superiority of thesociety to which victims of violence are saved. Thus, the projectof rescuing youngwomenbuilds on ideas about Swedish cultureand society as better, more gender equal and a safe haven foryoung women. Hence, there is a danger in the use of thesetestimonials. As Sara Ahmed has shown, stories of pain involvepower (Ahmed, 2004: 22). Ahmed warns against thefetishisation or commodification of suffering in the politicalarena (Ahmed, 2004: 32). Here, individual pain is appropriatedin the service of creating a national self-understanding ofSweden as a gender-equal and essentially good society.

While testimonies of non-Swedish, sometimes cultural,oppression and pain are common, the young women are also(and sometimes in the same policy document) represented asautonomous and free individuals, as will be discussed in thefollowing.

The Individual(ised) Self

Diana thinks that the trials she has gone through haveincreased her self-esteem. She has been forced to rely onher own ability to solve upcoming problems herself.Today Diana sees herself as emancipated from her familyand from cultural expectations, and this has made herstrong and independent.

[(Länsstyrelsen Södermanlans län, 2004: 4, p.23)]

In this narrative, we understand that the young woman hasleft her oppressive family and chosen a new life. What isapparent in the stories is howyoungwomen are called upon andasked to speak from an individual viewpoint, not as culturalbeings, not ‘as women’, but as individuals. The ideal self in thesehandbooks and surveys is a person who is capable of leaving herhome and family and starting a new life detached from her oldone. These stories are about individuals who, in one way oranother, have been ‘saved’ from their immigrant family and haveovercome the pain involved in their lives. The narrative ishopeful. Despite the pain Diana has suffered, the intended readeris assured that she has changed and that this progress hasoccurred due to the rescue efforts of Swedish society. TheSwedish reader will certainly feel better when reading thesestories of rescue, hope and progress. The (private) pain ofthe girls is evoked in policies on honour-related violence, assomething that demands a public (collective) response (Ahmed,2004: 20). Yet, the point of identificationwith the girls is not thepain in itself, but rather the choice to leave their family and toenter into Swedish society.

In another of the surveys from a National County Board, ayoung woman's words about her situation in her family arerepeated:

It is wrong that I constantly have to adapt to others and Ifind it difficult that I cannot find my self.

[(Länsstyrelsen Värmland, 2004: 4, p. 20)]

The policy documents produced by the National CountyBoards are thus either talking in terms of cultural/geographicaldifferences or in a more universalist frame in which the girlscan speak as individuals, who have left their belonging behind.The young victims are thereby allowed to speak, and arerepresented, in a dichotomous relationship: either as cultural/non-Swedish victims or as active individuals who have lefttheir family and thereby can pass as Swedish.

Alternative stories?

The question, however, is whether there are alternativeways of representing and possibilities of speaking outside ofthese governmental, state discourses. In order to find moremarginal stories, I have analysed material from a shelter forimmigrant women. In one of their pamphlets the words of ayoung woman are represented in the following way:

If people just could see us ‘immigrant girls’ as Swedes, Ithink that things would be much easier for us. We wouldnot be treated differently, and maybe then we would knowexactly what we are here in Sweden. ‘Immigrant girls’whofight for what they believe in are often misunderstood. Weare heard as ‘the voices of the oppressed’. No! We are notoppressed in our homes, we are oppressed by the Swedishpeople and their opinions.

[(Terrafem, 2004b, p. 74)]

The discursive violence of racism and stereotyping is herebrought into the public domain and defined as a political issuethat needs to be included when talking about honour-relatedviolence. Swedish society is here seen as the problem, ratherthan the solution. This is clearly an alternative story that aimsto point out other types of problem than only the immigrantfamily. Another of the pamphlets from Terrafem is called: “Letthe girls talk: A book from the project ‘Refuse to be called avictim’”. In the brochure, eight young women are representedwith photos and names, and in the introduction we can readthat:

In your hand you hold a pamphlet by girls who aretalking. They talk about themselves, about what theythink, how they feel, what they experience and what's ontheir minds. They are all different, as you will soondiscover, and they all have a foreign background. They aredefinitely not victims.

[(Terrafem, 2004a, p. 10)]

The central position of the girls themselves in thediscourse on honour-related violence is thus confirmed byall parties, even though the frames for speaking vary slightly.Nevertheless, the possibilities to resist and to tell alternativestories are limited by the fact that these young women arealready called upon to speak as cultural representatives. Theyare presented with the option of either confirming that theyare cultural representatives, or resisting and questioning theproblem as part of their culture, as in the quote below:

Many think that we ‘immigrant chicks’ are oppressed andthat we are not allowed to do what we want. That is yetanother thing that makes me frustrated. (…) I try toignore these things and only showwhat I as a person have

Page 114: Womens Studies International Forum 46 (2014)

112 M. Carbin / Women's Studies International Forum 46 (2014) 107–114

to offer. I am not a victim because of my origin, and Irefuse to be a victim.

[(Terrafem, 2004b, pp. 56–57)]

This young woman wants to underline that she isdefinitely not oppressed, not a victim. Instead she is of theopinion that the major problem she faces is the stereotype ofher as an oppressed immigrant girl. As a consequence, heralternative story tends to be framed in an individual, liberallanguage, closely related to stories of the individualised self.The possibility of speaking ‘as a victim’ is, as Beverly Skeggshas shown, problematic for subjects who are already seen asnot respected and lacking cultural value (Skeggs, 1997).Hence, the possibilities in talking about the problem in termsof vulnerability seem limited. There is instead a need to claimone's respectability as an individual, not as a victim. Thismeans that alternative stories, here represented by thepamphlets from the women's shelter, also have a tendencyto run into the pitfall of talking in terms of either/or. It seemsdifficult for these young women to move beyond thedichotomous thinking of either oppressed or free.

Silence as resistance?

I have found that the testimonies of victims of honour-related violence have two effects:

Firstly Their utterances serve as a double representation.That is, while the girls are allowed to speak, theyare representing themselves and at the same timeSwedish society is represented as good and genderequal. Locating the problem within the immigrantfamily serves as a way of hiding problems relatedto Swedish society, such as ethnic discriminationand racism. Thus, the victim stories presented bythe county boards have the effect of silencinginstitutional racism and the function or dysfunc-tion of Swedish welfare institutions and the failureto consider racism.

Secondly Violence as a problem —and more precisely howcontrol, threats and violence operate – is silenced.The stories represented here are more about themotives and causes of violence, than the violenceand the situation of young women. Thus, as aconsequence, these stories are used in policydiscourse to confirm what Swedish society alreadyknows—namely, that this is a culturally specificform of violence that occurs in immigrant commu-nities from the Middle East.

Speech can indeed form the grounds for emancipatorypolitics and the importance of previously silenced groupsmaking their voices heard in liberal democracies should notbe underestimated. However, speech is today seen as almostidentical with emancipation and automatically perceived as apositive value: speaking up, speaking for oneself, speaking upfor the other, giving voice. To be required to speak also meansto be silenced in a particular way. This is perhaps most visiblein the alternative stories produced by women's NGOs, whichbreak with the cultural script but nevertheless remaintrapped within the logic of either oppressed or free. Thus,speaking is already silence. In order to understand young

women victims, and to develop policy tools, it seems asthough Swedish policy-makers have to reduce the victims toknowable subjects, to stereotypes that can be easily grasped.As if they can gain access to difference, the unknown and theparticularities of the Other through her self-expression. As ifwe can simply hear what she has to say. Yet, as Sara Ahmedpoints out, the Other represents otherness precisely becauseshe is not understandable or easy to represent (Ahmed, 2002:561). Here lies a paradox: Swedish society desires to knowand understand the situation of these young women, yet themoment they are represented within policy discourses, theyare reduced to stereotypes, and as such remain impossible tounderstand.

I would thus like to draw attention to the problematic sideof speech and the underlying assumption that speaking upleads to liberation in this particular case. Speech can beformed by regulatory, hegemonic discourses whereby mak-ing your voice heard partly mirrors the hegemonic ideals. AsWendy Brown puts it:

Silence, as Foucault affirms it, is then identical neitherwith secrecy nor with not speaking. It instead signifies aparticular relation to regulatory discourses, as well as apossible niche for a practice of freedom within thosediscourses.

[(Brown, 2005: 87)]

If there is a requirement for some subjects to speak out, totalk about their situation, then Foucault insists that we ask thequestion ofwhy people have to speak in the first place.What arethe techniques of power involved in “letting the girls speak”?

Both Swedish public policy and the media demand thatyoung women and men with backgrounds from the MiddleEast talk about “How it is to live in such oppressive families”.The very terms and conditions placed upon speaking aretherefore narrow and restricting. In this sense, a niche for thepractice of alternative speech must be created outside ofthese spheres. In relation to these interpellations, silencemight form a strategy that disrupts the discourse. Especiallysince the discourse is so reliant upon the ‘truth-telling’ victimor the representative of the group. To answer the call isalready to accept the interpretation of oneself as culturallydeviant and consequently to legitimise the order of thediscourse. Few scholars have explored the value of silence.The political goal in our mediated culture seems to be tomake oneself visible, but what about trying to escape, beingabsent, being silent: could this be a form of resistance?

Researcher in social work Barzoo Eliassi shows in aninterview study how a pupil with Kurdish background chosenot to talk at all when the teacher kept on bringing newspaperarticles on honour killings to the classroom (Eliassi, 2010: 195).In this case, silence becomes an obstruction to both therequirement to speak and the nationalist assumptions under-lying the discourse of honour killing. It can be read as a refusalto participate for strategic reasons. However, silence can onlyprovide a solution under certain specific circumstances. Whilestories from young women cannot be seen as “the ultimatetruth about their culture” they cannot be dismissed altogether.As Sara Ahmed (2004) writes, on the one hand, pain andsuffering cannot form the ground for feminism, but on theother hand suffering cannot be excluded from a feminist

Page 115: Womens Studies International Forum 46 (2014)

113M. Carbin / Women's Studies International Forum 46 (2014) 107–114

agenda. Thus, stories of violence, threats and control cannotprovide the foundation for feminist politics or knowledgeclaims, but a feminist politics of solidarity demands somethingmore; it is not enough only to listen to these narratives.

Towards a feminist politics without secure grounding

Themajor dilemma for feminists in relation to the problemofso-called honour-related violence is that highlighting domes-tic violence issues in specific minority cultural/religious/ethniccommunities runs the risk of stereotyping these communitiesas particularly violent and backward. It separates out this formof domestic violence as a special cultural phenomenon re-quiring special cultural knowledge. On the other hand, notrecognising the particularities of the violence and the specificsituation of these victims runs the risk of generalising anddrawing upon ethnic majority, middle-class women's prob-lems as the universal problems. How can a political alterna-tive be formulated that does not fall into the trap of eitherculturalising or universalising the problem? The dilemma ofhow to form a politics that does not fall into the trap of eitherreducing young women into tokens, or ignoring the violence,remains.

Swedish policy discourses, as I have shown, are primarilyformed as either culturalist or universalist speech (see alsoKeskinen, 2011). While, in the UK context, a policy dilemmahas occurred that is caught between multicultural toleranceand cultural stereotypes (Meeto & Mirza, 2007), in theSwedish policy context the multicultural discourse has beenmarginalised (Carbin, 2010). Some researchers claim that ahuman rights perspective is what is needed in order to solvethe dilemma and to develop a more neutral perspective(Meeto & Mirza, 2007). Others argue that intersectionalitymight serve as a solution to the problem (Dahrvishpour,2006; de los Reyes, 2003). I think intersectionality (in theversion developed by black feminism) is a necessaryapproach in the sense that it includes different powerdimensions, and as such it is important politically. Yet, I amnot convinced that building a feminist political approach onintersectionality is enough in order to form an ethicalapproach to the problem of how to encounter particularothers, and other others (see Ahmed, 2002). An intersection-al approach would include race/ethnicity and class as sites ofpower and analyse the interconnections between them andthe interconnections with gender, which is necessary. Iwould nevertheless argue that intersectionality alone cannotprovide analytical grounds for understanding the negotia-tions going on in the debate on honour killings. Theintersectional approach risks reducing young women toobjects of knowledge, as transparent and within ‘our’ grasp.In this case, a postcolonial feminist reading must also takeinto account the way in which processes of naming, andthereby constructing Others and the Self, in themselvesgenerate possibilities and restrictions on speech. Only byacknowledging power and privilege within feminist move-ments, and by understanding that there are different agendasand different struggles, can a common ground for feministpolitical strategies be identified and the beginning of aprocess of solidarity be formed. However, this is also notenough to challenge and really make a difference in feminist

political struggles. Added to this, we must also question thevery ideal of reaching the ‘knowable’ other.

According to Judith Butler (2003: 10), we are allconstituted politically by the social vulnerability of ourbodies, attached to, and exposed to others. Vulnerabilityshould not be regarded as an intrinsic quality of an existingsubject (for example a victim of honour-related violence),but constitutive of the human condition. This suggests a newmode of ethics that holds at its centre the vulnerability andinterdependency of being (Shildrick, 2000). An insight thatwe are all vulnerable might also open up the possibility ofspeaking for victims of violence, without being forced tospeak either as a victim or as a free individual. This mightopen up a possibility to talk about vulnerability andspecificity without being reduced to a cultural stereotype. Ifvulnerability is seen as the human condition, it might not benecessary to take a stance against it, and to ‘refuse to be avictim’. In line with Butler and Shildrick, I would also arguethat it is necessary for those in more privileged positions whowish to be allies in struggles for social justice to unlearnknowingness and to become comfortable with their ownepistemic vulnerability. This means that those in privilegedpositions need to develop a politics without secure grounding—without easy solutions such as human rights or intersectionality.The idea that one has to know the other is part of a strivingtowards control. The question is whether one should not insteadtry to unlearn knowingness and to become comfortable withepistemic vulnerability. The very idea that we cannot know andshould not seek to know and control might form a starting pointfor transforming ethical relationships and creating a platform formeeting the particular other.

Endnotes

1 See Carbin (2010) for a discussion of the fact that Islam and Muslimminorities are never mentioned in the policy documents. Instead, theproblem is articulated as having to do with ‘values’ from a certaingeographical region of the world. Today, however, the liberal-conservativegovernment is investigating the issue of forced and arranged marriage, andwhether these inquiries will lead to a tightening of family reunificationpolicies is still unclear.

2 Apart from Carbin (2010), there are few policy studies available.Sabine Gruber (2011) explores school initiatives and policies.

3 I name these victims by their first names since this is how they werediscursively named in the media and in policy documents, and also because Ido not want their family names to be constantly connected with thetraumas.

References

Abu-Lughod, Lila (2002). Do Muslim women really need saving? Anthropo-logical reflections on cultural relativism and its Others. AmericanAnthropologist, 104(3), 783–790.

Ahmed, Sara (2002). This other and other others. Economy and Society,31(4), 558–572.

Ahmed, Sara (2004). The cultural politics of emotions. Edinburgh UniversityPress.

Akpinar, Aylin (2003). The honour shame complex revisited: Violenceagainst women in the migration context. Women's Studies InternationalForum, 26(5), 425–442.

Bhattacharyya, Gargi (2008). Dangerous brown men: Exploiting sex, violenceand feminism in the war on terror. New York/London: Zed Books.

Bredal, Anja (2005). Tackling forced marriages in the Nordic countries:Between women's rights and immigration control. In Sara Hossain, &Lynn Welchmann (Eds.), ‘Honour’—Crimes, Paradigms and ViolenceAgainst Women (pp. 332–353). London: Zed Books Ltd.

Page 116: Womens Studies International Forum 46 (2014)

114 M. Carbin / Women's Studies International Forum 46 (2014) 107–114

Brown, Wendy (2005). Edgework: Critical essays on Knowledge and Politics.New Jersey: Princeton University Press.

Butler, Judith (2003). Violence, mourning, politics. Studies in Gender andSexuality, 4(1), 9–37.

Carbin, Maria (2008). Honour related violence—The invention of a newSwedish policy problem. In Eva Magnusson, Malin Rönnblom, & HarrietSilius (Eds.), Critical studies of Nordic gender equalities (pp. 24–47).Göteborg: Makadam.

Carbin, Maria (2010). Mellan tystnad och tal. Flickor och hedersvåld i svenskoffentlig politik [Between silence and speaking: Girls and honour-relatedviolence in Swedish official policy]. : Stockholms Universitet.

Dahrvishpour, Mehrdad (2006). Invandrarflickor som fyrdubbelt förtryckta?En intersektionell analys av generationskonflikter bland invandrare iSverige [Immigrant girls as four-times oppressed? An intersectionalanalysis of generational conflicts among immigrants in Sweden]. In UlfHedetoft (Ed.), Bortom stereotyperna? Invandrare och integration iDanmark och Sverige. Stockholm: Makadam (pp. 89-111).

de los Reyes, Paulina (2003). Patriarkala enklaver eller ingenmansland? Våld,hot och kontroll av unga kvinnor i Sverige [Patriarchal enclaves or no-man'sland? Violence, threat and control of young women in Sweden]. NorrköpingIntegrationsverkets skriftserie IV.

Dhawan, Nikita (2007). Impossible speech: On the politics of silence andviolence. Sankt Augustin: Academia Verlag.

Ekström, Simon (2009). Hedersmorden och orden [The honour-killings and thewords]. Göteborg/Stockholm: Makadam.

Eliassi, Barzoo (2010). A stranger in my homeland: The politics of belongingamong young people with Kurdish backgrounds in Sweden. Sundsvall:Mittuniversitetet.

Foucault, Michel (1983). Discourse and truth: The problematization ofparrhesia. 6 lectures given by Michel Foucault at the University of Californiaat Berkeley, Oct–Nov. 1983 (http://www.lib.berkeley.edu/MRC/foucault/parrhesia.html)

Gruber, Sabine (2011). In the name of action against ‘honour-related’violence: National nations, gender and boundaries in the Swedishschool's ambitions to combat violence and oppression. Nordic Journal ofMigration Research, 1(3), 126–136.

Hellgren, Zenia, & Hobson, Barbara (2008). Cultural dialogue in thegood society: The case of honour killings in Sweden. Ethnicities, 8(3),385–404.

Karlsson, Lena (2006). Nyfiken på andra kvinnor i andra kulturer [Curiousabout women in other cultures]. Feministiska interventioner. Stockholm:Atlas.

Keskinen, Suvi (2009). Honour related violence' and Nordic nation building.In Suvi Keskinen, Salla Tuori, Sari Irni, & Diana Mulinari (Eds.),Complying with colonialism: Gender, race and ethnicity in the Nordicregion. Farnham: Ashgate.

Keskinen, Suvi (2011). Troublesomedifferences—Dealingwith gendered violence,ethnicity, and ‘race’ in the Finnishwelfare state. Journal of Scandinavian Studiesin Criminology and Crime Prevention, 12(2), 153–172.

Korteweg, Anna, & Yurdakul, Gökçe (2009). Islam, gender, and immigrantintegration: Boundary drawing in discourses on honour killing in theNetherlands and Germany. Ethnic and Racial Studies, 32(2), 218–238.

Meeto, Veena, & Mirza, Safia (2007). There is nothing honourable abouthonour killings: Gender, violence and the limits of multiculturalism.Women's Studies International Forum, 30(3), 187–200.

Diana, Mulinari, Suvi, Keskinen, Sari, Irni, & Salla, Tuori (2009). Introduction:Postcolonialism and the Nordic models of welfare and gender. In SuviKeskinen, Salla Tuori, Sari Irni, & Diana Mulinari (Eds.), Complying withcolonialism: Gender, race and ethnicity in the Nordic region. Farnham:Ashgate.

Dustin, Moira, & Phillips, Anne (2008). Whose agenda is it? Abuses ofwomen and abuses of ‘culture’ in Britain. Ethnicities, 8(3), 405–424.

Phillips, Anne, & Saharso, Sawiti (2008). The rights of women and the crisisof multiculturalism. Ethnicities, 8(3), 291–301.

Prins, Baukje, & Saharso, Sawitri (2008). In the spotlight: A blessing and a cursefor immigrant women in the Netherlands. Ethnicities, 8(3), 365–384.

Reimers, Eva (2005). En av vår tids martyrer—Fadime Sahindal sommediehändelse [A martyr of our time—Fadime Sahlindal as a mediatedevent]. In P. de los Reyes, & L. Martinsson (Eds.), Olikhetens paradigm(pp. 141–159). Lund: Studentlitteratur.

Schlytter, Astrid (2004). Rätten att själv få välja. Arrangerade äktenskap, könoch socialt arbete [The right to choose. Arranged marriages, gender andsocial work]. Lund: Studentlitteratur.

Sen, Purna (2005). ‘Crimes of honour’, value and meaning. In LynnWelchman, & Sara Hossain (Eds.), Honour: Crimes, paradigms andviolence against women. New York: Zed Books.

Shildrick, Margaret (2000). Becoming vulnerable: Contagious encountersand the ethics of risk. Journal of Medical Humanities, 21(4), 215–227.

Skeggs, Beverly (1997). Formations of class and gender: Becoming respectable.London: Sage.

Spivak, Gayatri Chakravorty (1993). Can the subaltern speak? In LauraChrisman, & Patrick Williams (Eds.), Colonial discourse and post-colonialtheory (pp. 66–111). New York: Columbia University Press.

Towns, Anne (2002). Paradoxes of (in)equality: Something is rotten in thegender equal state of Sweden. Cooperation and Conflict, 37(2), 157–179.

Tuori, Salla (2007). Cooking nation: Gender equality and multiculturalism asnation-building discourses. European Journal of Women's Studies, 14,21–35.

Wendt Höjer, Maria (2002). Rädslans politik [The politics of fear]. Malmö:Liber.

Wilson, Amrit (2007). The forced marriage debate and the British state. Race& Class, 49(1), 25–38.

Winther Jorgensen, Marianne, & Phillips, Louise (1999). Diskursanalys somteori och metod. Discourse analysis as theory and method. Lund:Studentlitteratur.

Žižek, Slavoj (1999). Njutandets förvandlingar. Stockholm: Natur och Kultur.

Empirical material

Länsstyrelsen Södermanlands län (2005). Förtryck och våld i hederns namn[Oppression and violence in the name of honour].

Länsstyrelsen Södermanlans län (2004). Frigörelse på egna villkor.Kartläggning av flickor och unga kvinnor som riskerar att utsättas förhedersrelaterat våld [Emancipation on your own terms: A survey of girlsand young women at risk of being victims of honour-related violence].

Länsstyrelsen Värmlands län (2004). Låt oss lyssna. Kartläggning och analysav behovet av insatser för flickor och unga kvinnor som riskerar attutsättas för så kallat hedersrelaterat våld av nära anhörig [Let us listen: Asurvey and analysis of the need of initiatives for girls and young womenwho are at risk of becoming victims of so-called honour-related violenceby a close relative].

Socialstyrelsen och Länsstyrelserna (2005). Nationellt konsultativt stöd[National consultative support].

Terrafem (2004a). Låt flickorna tala! En bok från projektet vägra kallas offer[Let the girls talk! A book from the project ‘Refuse to be called a victim’].Malmö: Terrafem.

Terrafem (2004b). Invandrad, invaderad. Nu höjer vi våra röster. Om att varatjej och minoritet [Immigrant and invaded. Now we raise our voices. To be agirl and a minority]. Malmö: Terrafem.

Page 117: Womens Studies International Forum 46 (2014)

Women's Studies International Forum 46 (2014) 115–122

Contents lists available at ScienceDirect

Women's Studies International Forum

j ourna l homepage: www.e lsev ie r .com/ locate /ws i f

Development policies, intimate partner violence, Swedishgender equality and global health

Ann Öhman a,b,⁎, Maria Emmelin c

a Department of Public Health and Clinical Medicine, Epidemiology and Global Health, Swedenb Umeå Centre for Gender Studies, Umeå University, Swedenc Department of Clinical Sciences, Social Medicine and Global Health, Lund University, Sweden

a r t i c l e i n f o

⁎ Corresponding author at: Umeå Centre for Gender StuSE-901 87 Umeå, Sweden.

0277-5395/$ – see front matter © 2013 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.wsif.2013.12.001

s y n o p s i s

Available online 14 January 2014

This paper discusses current Swedish international development policies on gender andviolence. It deals with the relationship between development policies, global health,promotion of gender equality, and violence against women in a global perspective. The focusis on intimate partner violence and the highly promoted gender mainstreaming policy.Theoretically, our point of departure lies within a feminist notion of gender relations, powerstructures, and male hierarchies that constrain and subordinate women and girls and whichexpose them to gendered violence. We claim that stronger links need to be created betweenlocal activist groups in low and middle income countries and the international developmentagencies. It is important to initiate and formalize a North–South dialogue between suchgroups, as well as enhancing South–South dialogue and cooperation.

© 2013 Elsevier Ltd. All rights reserved.

Introduction

This paper deals with the relationship between interna-tional development policies, especially the latest Swedishdevelopment policy, global health, promotion of genderequality, and violence against women in a global perspective.We want to take a closer look at the strategies, the history,and the goals for development policy and their links topreventing violence against women, promoting genderequality, and global health. We argue that results fromgender research on violence against women and feministnotions of gender inequalities need to be taken into accountin development policies regarding gendered violence. Wealso argue that stronger links need to be created betweenlocal activist and/or feminist groups in low and middleincome countries and the development agencies. It isimportant to initiate and formalize a North–South dialoguebetween such groups, as well as enhancing South–South

dies, Umeå University,

ll rights reserved.

dialogue and cooperation. To the best of our knowledge,there is lack of theoretical discussion as to gendered violence,global health, and gender mainstreaming (GM) in Swedishdevelopment policies.

In 2010, the Swedish government launched a new policyfor gender equality in Sweden's international developmentcooperation for the years 2010–2015 (Sida, GovernmentOffices of Sweden, 2010). The policy strongly emphasizesgender mainstreaming as a means for reaching the goals forgender equality. The overall objective is “gender equality,greater influence for women, and greater respect for women'srights in developing countries”, and the four areas put in focusare “women's political participation and influence, women'seconomic empowerment and working conditions, sexual andreproductive health and rights and women's security, includingcombating all forms of gender-based violence and humantrafficking” (Sida, Government Offices of Sweden, 2010). Thepolicy also acknowledges the role of poverty reduction as ameans to enhance gender equality (and the other wayaround). Women and girls are the most important targetgroups for interventions, but the policy underlines that thesedepend on political will and an involvement of both men andwomen. The human rights perspective is prominent and the

Page 118: Womens Studies International Forum 46 (2014)

116 A. Öhman, M. Emmelin / Women's Studies International Forum 46 (2014) 115–122

policy is in line with the 1979 UN Convention on theElimination of All Forms of Discrimination against Women(UN, 1979). The policy also touches upon the MillenniumDevelopment Goals (MDGs) as a basis for changing genderrelations (United Nations Resolution A/RES/55/2, 2000). Itclarifies that unequal power relations are key aspects forunderstanding gender inequality. In sum, Sweden's develop-ment cooperation policy can be viewed as a progressive policyin line with international frameworks for development.

Although the Swedish development policy does notexplicitly deal with gender equality as a key issue forimproved health, we, as global and public health researchers,will frame the discussion in this paper about gender equality,development policy, and gendered violence from a healthperspective. There is a lack of debate on these topics, sincemost deliberations about gender equality and developmentpolicies focus around issues of economy, democracy, humanrights, and political power. We claim that gender equality is aproxy for improved health of any population, as it empowerswomen and girls, which in turn improves the economicstatus of families and society. We thus start this discussion bydescribing men's violence against women, focusing onintimate partner violence (IPV), as a major global healthproblem. Thereafter, we present a description of the maincharacteristics of development policies on gender in generalduring the last three decades, followed by a broad overviewof the current Swedish development policy with its gendermainstreaming approach and focus on the rights of women(Sida, Government Offices of Sweden, 2010). Finally, wediscuss and problematize the opportunities and challenges ofintegrating research results on IPV into development coop-eration for increased gender equality. We acknowledge theimportance of scrutinizing different forms of gender-basedviolence (GBV) such as female genital mutilation, prostitu-tion, trafficking, sexual torture, and rape as a weapon inwarfare from a gender policy perspective. However, in thispaper we restrict our focus to men's violence against womenwithin a relation, through what in many settings is labeledintimate partner violence.

Thus, the aims of the paper are threefold: (a) to describemen's violence against women as a global health problem;(b) to highlight different approaches (including the Swedish)in development policies for gender equality; and (c) todiscuss and problematize the need for integrating researchresults on IPV and feminist theory into developmentcooperation for increased gender equality.

Intimate partner violence and global health

Intimate partner violence is a major global healthproblem. It is a human rights concern embedded in theimbalance of power between men and women (Campbell,2002; Jewkes, 2002a, 2002b). Intimate partner violence isone form of gender-based violence that involves current orformer partners in heterosexual as well as homosexualrelationships. This paper focuses specifically on the violenceperpetrated by men against women since this is the mostcommon form of IPV and has well documented negativeeffects on women's health (Krantz & Garcia-Moreno, 2005).

In 2005, WHO presented the results from a multi-countrystudy on what they, at that time, labeled domestic violence

(WHO, 2005). The study was performed in 11 countries andwas the first study to use a standardized methodology thatallowed for comparisons among different settings. The cross-sectional surveys, performed both in urban and rural settingswere preceded by formative (qualitative) research to assurethat the questionnaires were adjusted to the socio-culturalcontext of each of the settings.Much effort wasmade to ensurethat the specific ethical concerns involved in studies of partnerviolence were taken into account (WHO, 2001). In the WHOstudy, physical violence included actions such as being beaten,hit, kicked, choked, burnt, or threatened with a weapon by acurrent or former partner/husband, while sexual violence wasdefined as being physically forced or threatened to have sex orto do something sexually degrading. The study confirmed theseriousness of men's violence against women but also showeda large variation in both lifetime and 12-month prevalence ofphysical and/or sexual violence. The lifetime prevalence ofphysical and/or sexual violence reported, varied between 15%and 71% in ever-partnered women aged 15–49 years. Thecorresponding figures for past-year experience ranged from 4%in Japan to 33% in Tanzania, 49% in Bangladesh, and 54% inEthiopia (Garcia-Moreno, Jansen, Ellsberg, Heise, &Watts, 2006;WHO, 2005). These figures were in line with estimates fromother countries of the South, not involved in the WHO study,such as Haiti, Nigeria, and Uganda with prevalences rangingfrom 11% to 52% (Gage, 2005; Koenig et al., 2003; Okenwa,Lawoko, & Jansson, 2009). Figures from the Swedish setting,where also less severe types of violence were included,indicated that 12% of women have been subjected to partnerviolence during the past year (Lundgren, Heimer,Westerstrand,& Kalliokoski, 2002). More recent estimates specify a past12 month exposure of 8% to physical violence and 3.2% to sexualviolence (Lövestad & Krantz, 2012).

Intimate partner violence is associated with injuries as wellas several other severe health-related consequences (Campbell,2002). Depression, anxiety, and post-traumatic stress disordersare well-documented health effects as well as reproductivehealth problems (Campbell, 2002; Deyessa et al., 2009; Ellsberget al., 2008). Feelings of shame, guilt, and poor self-esteemhave also been reported to accompany violence experiences(Valladares, 2005). In addition theWHO study indicated thatexposure to violence increased women's vulnerability toalcohol and drug abuse, suicidality, maternal mortality, andHIV infection (WHO, 2005).

The causes of IPV have been described as multifaceted andelaborated in an ecological model developed in the late 1990s(Heise, 1998) and further detailed in recent years (Ellsberg &Heise, 2005, Heise, 2011). The model illustrates how factorsat the individual, relationship, community and macro-sociallevel interact to influence the risk of violence within intimaterelationships (Heise, 2011). At the macro-social level genderorder, cultural and economic factors are central in influencingnorms, sanctions, attitudes and behavior at the other levels.According to Heise this means that certain individual riskfactors may be enough to “cause” abuse in some socio-culturalor community settings but not in others. In her review Heise(2011) assesses the evidence of links between risk factors IPV,and the effectiveness of interventions to reduce such violence.She concludes that there is strong scientific support for the linkbetween country level of IPV and norms that accept violencefor conflict solving and encourage views on masculinity that

Page 119: Womens Studies International Forum 46 (2014)

117A. Öhman, M. Emmelin / Women's Studies International Forum 46 (2014) 115–122

involve control and aggression. Several studies have shownthat IPV is more common in societies with pronouncedunequal gender power relations, subordinating women andgirls (Jewkes, 2002a, 2002b; Lawoko et al., 2007; Okenwa etal., 2009, 2010; Rani et al., 2004). These gender attitudes andnorms are often internalized also among women themselves,which was evident in the WHOmulti-country study as well asin studies from Nigeria and Sub-Saharan Africa (WHO, 2005;Oyediran & Isiugo-Abanihe, 2005; Uthman, Moradi, & Lawoko,2009, Uthman, Lawoko, & Moradi, 2010). In these studieswomen, to a varying degree believed that a man had the rightto beat his wife under circumstances where she has notcompleted household work adequately, has refused sex, hasdisobeyed her husband, or has been unfaithful. Interventionstargeting the gender order and the accompanying norms arefew but promising. Heise (2011) gives examples fromcountries such as South Africa, India and Ethiopia were gendertransformative approaches have targeted both men andwomen to challenge existing gender norms. The resultsindicate a positive impact but the studies often have method-ological limitations making it hard to show an actual reductionof violence. Health care and judicial systems are also influencedby gendered norms and attitudes on IPV in that women mayreceive limited support and avoid disclosing violence experi-ence. These norms can also decrease health care personnel'sand police' preparedness to ask and act about violence (Edin,2006; Jewkes, 2002b; Laisser, 2011).

Garcia-Moreno and colleagues (WHO, 2005) have pointedout that the elimination of violence against women is notincluded in the specific targets or indicators of the MDGs(http://www.un.org/millennium/declaration/ares552e.htm).In their report, they examine each of the goals for its relevanceand the strategic opportunities it gives for prevention andreduction of violence. They see clear links to all the MDGs andsuggest that the goal to eradicate extreme poverty and hungermay influence violence against women by protecting thepoorest and most vulnerable women. Another example isthat targeting universal primary education will encourageschooling for girls and, if successful, contribute to reductionof violence. The authors also indicate that the goals relating toreducing child mortality, maternal health, and HIV/AIDS, willdepend on taking violence against women into account sinceexposure to violence often hampers prevention and care. Also,the goals to ensure environmental sustainability and to developglobal partnerships for development could have includedgender-empowering interventions advocating against violenceagainst women. However, the authors' strongest claim is, insupport of theUNTask Force on Education, that themillenniumgoal to promote gender equality and empower women should beexpanded to have a clear target to reduce lifetimeprevalence ofviolence against women by 50% by the year 2015 (Grown,Gupta, & Kes, 2005).

This means that intimate partner violence is seen both asa gender equality problem and a global health issue thathinders the progress in achieving the Millennium Develop-ment Goals. It is a public responsibility that has to beacknowledged and responded to. Of course it also poses achallenge for country development policies to advocateactions and interventions to reduce intimate partner violenceand to increase gender equality also outside a country's ownborders.

Development policy, gender mainstreaming and intimatepartner violence

In international debates, there has been ongoing discus-sion about how to integrate gender issues into the efforts ofimproving the life and health of people in less advantagedsocieties. To better discuss and problematize the opportuni-ties and challenges of integrating research results on intimatepartner violence into development cooperation, we will herebriefly review the history of recent international develop-ment policy and gender. This history can be summarised as atransition from the invisibility of gender and women, via anemphasis on women's issues, towards, finally, a gendermainstreaming policy.

The ideas and conceptions of development policy workwere primarily, in the 1960s and 1970s, initiated from theGlobal North with less engagement from local people in theGlobal South, the so-called developing countries. We do notwant to disregard the hard work of feminist activists andadvocates in the Global South and their struggle to changewomen's lives and to combat gender based violence, forinstance in the UN, Gita Sen and Naila Kabeer have beenessential in advocating women's voices from the GlobalSouth. We agree however with feminist, social scienceresearchers that in the larger global development policydiscussions women from the Global South were oftenmarginalized (Connell, 2007; Connelly et al., 2000; Schech,1998). There was a belief that a transfer of goods, know-howand values from the Global North to poor countries wouldfacilitate their development. However, many times, the basicneeds and rights of people, especially women and girls, wereneglected in the governmental development programmes;and men's violence against women was seldom on theagenda in development interventions initiated from theGlobal North. Women were often invisible in all stages oftraditional development policy (Kilby & Olivieri, 2008).Much of this discussion centred on the “Western helper”and the “ThirdWorld beneficiary.” For a long time, it was thisquestion of “we” and “the rest” that distanced people in theGlobal South from those working in development policyprogrammes initiated from the Global North. The conse-quent obvious risk of ethnocentrism marginalized disadvan-taged groups and those who were the focus of developmentaid.

Due to heavy criticism of these policies and the waythey were implemented in the 1970s and 1980s, there wasa shift of focus towards women, which was supposed toplace women's issues at the centre of development policyand aid (Mohanty, 1988; Walby, 2005). Accordingly,development policies started highlighting women's spe-cifically disadvantaged positions in poor settings (Schech,1998). Women's lobby groups and activist organizations inthe Global North as well as politicians urged for a morecomprehensive inclusion of women since prior developmentprogrammes had failed to reach women's needs. The drive toinclude women resulted in a strategy usually labeled Womenin Development (WID) (Connelly et al., 2000; Schech, 1998).The WID strategy resulted in a variety of implementationsthrough special organisations for women's issues such as“women's bureaus,” “women's desks,” and “women's offices.”These were often established in close collaboration with

Page 120: Womens Studies International Forum 46 (2014)

118 A. Öhman, M. Emmelin / Women's Studies International Forum 46 (2014) 115–122

international development policy bodies. It was a reaction tomeet the criticism of the previous exclusion and invisibility ofwomen; thus, it was constructed to enhance women's accessand inclusion in development processes.

Unfortunately, the WID strategy to a large extent failed tointegratewomen, and instead forced gender issues into a specialorganisation focusing onwomenonly, setting aside other bodiesof importance for real change in gender relations (Connelly etal., 2000; Schech, 1998). It never really addressed the basic rootsof gender and gender inequality such asmale-dominated powerstructures and gender regimes in social institutions. The WIDstrategy was permeated with an essentialist notion of thedifferences between women and men and argued for atraditional development in terms of modernization theory andeconomic growth. There was an underlying assumption thatthis would reduce gender inequalities, therefore decrease therole of gender as an analytical category and focus more onindividual achievement (Connelly et al., 2000). We argue that,since gendered power structures were not targeted this mighteven have contributed to an increase in gender segregation.During this period, men's violence against women and IPVwereneither prioritized on the development policy agenda of theGlobal North nor in health research.

The WID strategy was thus heavily criticised and putunder scrutiny by feminist activist groups as well as bygender researchers. Accordingly, during the 1990s, new goalswere established and new strategies formulated for thedevelopment policies that emphasized mainstreaming ofgender issues into all forms of work for development, labeledGender and Development (GAD) (Connelly et al., 2000).Unlike WID, the GAD approach accounts for gendered powerstructures and how these play a key role for both men andwomen, i.e. not specifically for women. The GAD perspectiveacknowledges the social construction of gender relations andhow they affect women's and men's lives. It also touchesupon intersectional perspectives and the importance of therelationship among, for instance, gender, class, and ethnicity.The GAD approach proposes strategies that address practicalneeds and disruption of gendered power structures. The GADtherefore argues for a gender mainstreaming approach and inthe UN World Conference on Women at Beijing in 1995, all189 national governments adopted a strategy to focus ongender equality as a means for development (Rathgeber,2005).

The critique of development policies and the transforma-tion from WID to GAD was initiated in several internationalwomen's and gender conferences such as in Mexico 1975,Copenhagen 1980 and Nairobi 1985. The mainstreamingstrategy was also adopted by the WHO and its framework forgender equality in health. WHO defines gender mainstreaming(GM) as follows:

If health care systems are to respond adequately toproblems caused by gender inequality, it is not enoughsimply to “add in” a gender component late in a givenproject's development. Research, interventions, healthsystem reforms, health education, health outreach, andhealth policies and programs must consider gender fromthe beginning. Gender is thus not something that can beconsigned to “watchdogs” in a single office, since no singleoffice could possibly involve itself in all phases of each of

an organization's activities. All health professionals musthave knowledge and awareness of the ways gender affectshealth, so that they may address gender issues whereverappropriate and thus make their work more effective. Theprocess of creating this knowledge and awareness of – andresponsibility for – gender among all health professionalsis called “gender mainstreaming” (http://www.who.int/gender/mainstreaming/en/index.html)(WHO, 2007).

The World Health Organization strongly emphasizes thatIPV and violence against women are severe threats to publicand global health. WHO suggests a number of preventivestrategies for ending the violence that need to build onevidence and empirical scientific results. WHO also suggestsefforts for increased collaboration with international agen-cies and organizations to reduce IPV (http://www.who.int/mediacentre/factsheets/fs239/en/)(WHO, 2012).

Swedish gender equality policies, development policy andintimate partner violence

Sweden adheres in many ways to the internationalumbrella of strategies and implementation of gendermainstreaming strategies agreed upon in the UN andWHO. As mentioned in the Introduction to this paper, theSwedish government has launched a new policy for genderequality in Sweden's International Development Coopera-tion Agency (Sida) for the years 2010–2015 (GovernmentOffices of Sweden, 2010). The overall objectives of the Sidaplan are to contribute to a reduction of gender-basedviolence and to promote the rights and the economic andpolitical empowerment of those subject to gender-basedviolence – mainly women and girls – in Sida's partnercountries and in humanitarian assistance. The policyemphasizes gender mainstreaming as the means to reachgender equality aiming at (i) actively applying and inte-grating the gender perspective, (ii) targeting specific groupsor issues, and (iii) conducting a gender-aware dialogue withpartners. Women's agency and rights are highlighted as keytargets for intervention for women's increased politicalparticipation, economic empowerment, and working con-ditions. It also highlights sexual and reproductive healthand rights and aims to tackle all forms of gender-basedviolence. In addition, it indicates specific intermediaryoutcomes relating to prevention, legal measures, andservice and care for victims and survivors. In Sida's genderpolicy document of 2005, two main areas of strategicsupport are (i) strengthening rights, and (ii) powerstructures and relations. The 2007 a Budget Bill identifiedas the main thematic priorities: (i) the economic empower-ment of women, (ii) sexual and reproductive health and rights,(iii) the political participation of women, and (iv) women'ssecurity. Sida management is also obliged to give Sida staff“resources, incentives and competence development in thefield of gender equality and gender mainstreaming” and toinclude gender equality in the monitoring and evaluation of itsown actions.

In order to contextualise Swedish international develop-ment policies, it is important to remember that Sweden,together with the other Nordic countries, ranks high ininternational comparisons regarding gender equality (World

Page 121: Womens Studies International Forum 46 (2014)

119A. Öhman, M. Emmelin / Women's Studies International Forum 46 (2014) 115–122

Economic Forum, 2011). Swedish gender equality policies areoften referred to as being both progressive and effective. Theyare often used as a normative standard for the rest of theworld.The gender equality discourse is strong in Swedish society, andpoliticians agree that gender equality is a key issue fordevelopment and wealth in any given society. There is alsopolitical consensus that this gender equality is important toprotect against reactionary forces that would push womenback to household duties and reproductivework (Sörlin, 2011).The main goal for Swedish gender equality policy is thatwomen and men have equal power to form their lives, as wellas to engage in shaping the society. This implies that theyshould have the same opportunities, rights, and obligations inall spheres of life. The policy is formulated so that theoverarching goals are equal distribution of power and influencebetweenmen andwomen, economic equality betweenwomenand men, and equal distribution of unpaid care and householdwork. Last but not least, the policy emphasizes that men'sviolence against women in Sweden should be eradicated, i.e. itincludes a vision of zero tolerance regarding gender-basedviolence (Prop. 2006/06:155).

Therefore, and in line with the main discourse of nationalgender equality in Swedish public policy, gender is stronglyintegrated into the aims and the activities carried out inSwedish development programs in the world (Sida, 2010:1;Silfver, 2010). Linking the issue to development policyimplies hope that the Swedish notion of gender equalitymight be transferred to other social contexts. Sida hasadopted a definition of gender-based violence that includeswhat is being perpetrated by whom, for what purpose, andwith what effect. The types of violence specifically mentionedin the plan are female genital mutilation, violence in thename of honour, violence against lesbians and homosexuals,sexual violence in conflict, and domestic violence. The fourentry points clearly relate to the Millennium DevelopmentGoals and state that gender-based violence is a violation ofhuman rights; working for its elimination is crucial forpoverty reduction, reversing the spread of HIV, and theprotection and promotion of sexual and reproductive healthand rights (Sida, 2007).

The need to integrate research results on IPV intodevelopment cooperation for increased gender equality

Issues of gendered power structures, patriarchal societies,male dominance over women and girls that restrict theiragency and participation, and violence against women, havestarted to be integrated into development policy, but frompolicy to practice, there still remains much to be done. Ourconcern is how gender mainstreaming policies in Swedishdevelopment programs can make use of research results, forinstance from health research. It is therefore vital to pose thequestion of how research on the magnitude and conse-quences of violence can contribute to gender equality anddevelopment policies.

Researchers on gendered violence and health oftensuggest, on the basis of their results, interventions such aschanges in legislation to combat normative systems thatsuppress and violate women's agency in order to decreaseIPV. In addition, they often suggest changes in health systemsso that IPV may be integrated into everyday clinical routines

as well as training of personnel to detect IPV. However, thesesuggestions fall short if they are not acknowledged andintegrated into general health policies and legislation. Thehealth researcher's role is usually to suggest changes inpolicies or evaluate interventions, but seldom are theyactively involved in policy making, as their main focus isacademic work.

Several feminist scholars have pointed to the danger of“exporting” gender policies from theGlobal North to the GlobalSouth and the obvious risk of a continuation of colonialrelations and dependencies (Baell, 1998; Connelly et al.,2000; Kilby & Olivieri, 2008; Moser, 2005; Schech, 1998). Therisk of de-contextualization is obvious if gender analysisframeworks for analysis are adopted with little culturalsensitivity or consideration of differences in public discourses,legal systems, and gendered norms, as often presented inresearch on gender and violence. Further, there is an obviousrisk of essentializing “the woman” to a subject of similarity,regardless of social context and conditions, and withouttaking into consideration the widespread differences insocio-economic living condition and power structures thataffect gender relations. Criticism has been raised, infeminist and post-structural research, of the essentialistnotion of the female subject in development projects (Baell,1998; Kilby & Olivieri, 2008; Schech, 1998). Schech outlinesthis critique by using Nussbaum's dichotomy and the choicebetween a universal measure of quality of life for all menand women or the use of a variety of different normsselected from different cultures (Nussbaum & Clover,1995). Development policy has thus been infiltrated by aparadigm that is largely influenced by a notion of Westernmodernity and a Western approach to development. Thishas been the cornerstone for both the Women in Develop-ment and the Gender and Development policies, and thereis reason to suspect that the approaches have not reallybenefited poor, marginalized, or women exposed to violence(Connelly et al., 2000).

Post-developmentalists, post-colonialists, and post-structuralfeminists have therefore paved the way for a renewed viewon gender and development. As Schech (1998) outlinedalready in 1998, Australia's development policies ought tomove away from a hierarchal model of Western modernityto a more cooperative approach that emphasizes partner-ship. Baell (1998) defines this process of changing ap-proach as moving from aid-dependent relations towardsdevelopment partnerships (Baell, 1998). Further, Kilby andOlivieri raise the question whether women's rights can beadvanced within a neo-liberal framework (Kilby & Olivieri,2008). The point of departure for all this critique is GADand the rights approach. They argue that the focus onwomen-only projects actually marginalized women frombroader (and male) power structures and that this policy didnot empower women. The view of “women of the Global South”in development policies from the Global North has beencriticized by feminists from the Global South for being racist,ethnocentric, and essentializingwomen as a homogenous groupof victimized, poor women without agency (see for instanceMohanty, 1988).

Whereas WID, on one hand, separated women fromdevelopment work, the GAD strategy, on the other hand, had atendency to be too instrumental and technical—an approach

Page 122: Womens Studies International Forum 46 (2014)

120 A. Öhman, M. Emmelin / Women's Studies International Forum 46 (2014) 115–122

whichmight not benefit women and girls. In addition, therewasno guarantee that gender mainstreaming meant includingwomen; instead, mainstreaming gender often meant “male-streaming,” i.e. men in power defined the core issues so thatwomen's prioritised needs and rights were obscured or left out(Moser, 2005; Mukhopadhyay, 2004).

In a recent self-evaluation of gender mainstreamingpolicies, conducted by Swedish development agency Sida,the evaluators stress that it is still important to createincentives for gender mainstreaming, not only internallyinside Sida, but also externally among program partners(Sida, 2010:1). They state that gender analyses are oftenconducted, especially at the beginning of program activity.There is, however, a concern that Sida has been poorer inmaintaining the same incentives later on in the workprocess of program interventions. It is concluded thatgender is largely absent in program monitoring andevaluation. There is thus a risk that gender issues areconcealed and that this hinders the internal learningprocess among Sida staff for further advancement of thedevelopment policy. However, Sida does exercise andemphasize a strategy of dialogue, which would certainlybenefit women if they are approached as collaboratingagents in their own right.

There is also conceptual confusion as to the use of theconcept of gender in development debates (Warren, 2007).This is the case also in current Swedish policies on IPV anddevelopment programs. It is of concern that Sida still uses theold concept of “gender role” in its recent policy for develop-ment collaboration (Sida, 2010). We would have appreciated amore up-to-date and theoretically driven use of conceptualframeworks presented in gender research so that the termi-nology in development policies could be based on contempo-rary gender/feminist theory. Drawing on old sex role theory, aspresented in a rather structural-functionalist notion, to de-scribe and analyse gender inequalities of today is not onlymisleading, but will also obscure analysis (Connell, 2007;Connell & Messerschmidt, 2005).

The need for cultural sensitivity in development policies isobvious, but often this is neglected. Usually the issue ofviolence against women is dealt with separately, which inturn increases the risk of being rendered invisible and notpart of general interventions to improve women's lives andhealth. Also, health research on violence against women is atrisk of falling into this pitfall of neglect, even though genderresearchers on IPV often claim cultural sensitivity as a keyaspect of the outcomes of research. One example wherecultural sensitivity is in focus is the WHO multi-countrystudy of 2005, which emphasizes the importance of adaptingquestionnaires to the specific cultural understanding ofgender and violence (WHO, 2005). The issue is about givingwomen a voice and about always having in mind whose voiceis doing the talking.

It is also important to include local feminist groups andactivists from the Global South in the dialogue. In manycountries of the Global South, which are now targets fordevelopment collaboration, there is an existing body offeminist activists, which have long traditions of activism andengagement in policy making. These NGOs form an importantcounterpart to development agencies of the Global North thatcan contextualise and sensitise the problemswith IPV and give

women a voice in ways to empower women and increase theiragency. This partnership approach would certainly facilitateand probably effectuate development cooperation so that thosein need of interventions on IPV, i.e. women and girls, can have asay.

The gender equality discourse in Sweden is rather top-down, emanating from governmental legislation and nationalpolicy; heading towards local policy and governance inmunicipalities, companies, and even family life. In line withEveline and Bacchi (2005), we argue that there is an obviousdanger that adopting a gender mainstreaming approach willnot enhance gender equality as it may turn into merely aquestion of administration and management. Eveline andBacchi state that the very understanding of gender in aspecific context will guide the whole work at an early stage ofthe process. It will certainly have a bearing on how equalitiesare understood and articulated, which in turn will affectpolicies and strategies for change. Further, they argue thatpower asymmetries are usually not in focus in gendermainstreaming, and they claim that it needs to go beyondtraditional approaches of difference between men andwomen. Instead they suggest so called “gendering-awarenessmainstreaming” in which they advocate for a shift from thenoun and category “gender” to a verb and a process “gendering.”Sweetman (2012), in her editorial on gender, developmentand mainstreaming policies, is also critical of the waymainstreaming policies have been used from the mid-1990sand onwards. She still regards gender mainstreaming as aninternational priority as well as a local reality and sheemphasizes that without local connections to activists andfeminist groups, the policy may hamper real change and becounter-productive to local struggles and initiatives. We agreewith Sweetman and believe that combating intimate partnerviolence requires complex and comprehensive strategies thatinclude bothmainstreaming policies that take into account notonly governmental efforts but also grass-root movements'initiatives in a local context. In this regard, we look forward tostrengthened relations between Swedish development policiesand such local initiatives.

Conclusions

We want to emphasize the need for an integration ofgender health research on IPV and violence against women aswell as an integration of gender and feminist theory intodevelopment collaboration so that proper interventions canbe grounded in current and actual knowledge. Further,we stress that Swedish development policies on gendermainstreaming in general have a good intention but, to beeffective, they need to be more solidly grounded in bothactivist movements in the Global South as well as in currentgender theory developed in gender studies. There is a needfor a sustainable dialogue among development agencies,health researchers, and activist movements in order todevelop more democratic and participatory frameworks forreducing intimate partner violence and violence againstwomen.

The question of who is in focus in these policies isrelevant. It is also a question about who is in power to define“the problem” of IPV and gendered violence. The risk oftaking “the male” as norm is obvious in male-dominated

Page 123: Womens Studies International Forum 46 (2014)

121A. Öhman, M. Emmelin / Women's Studies International Forum 46 (2014) 115–122

societies, often patriarchal in their structure, where womenhave little or no power to speak in public and where menusually talk about women as being in “their place.”

References

Baell, Jo (1998). Trickle-down or rising tide? Lessons on mainstreaminggender policy from Colombia and South Africa. Social Policy andAdministration, 32(5), 513–534.

Campbell, Jacquelyn C. (2002). Health consequences of intimate partnerviolence. The Lancet, 359(9314), 1331–1336.

Connell, Raewyn (2007). Southern Theory—The Global Dynamics of Knowledgein Social Sciences. Cambridge: Polity Press.

Connell, Raewyn, & Messerschmidt, James (2005). Hegemonic masculinity:rethinking the concept. Gender and Society, 19(6), 829–859.

Connelly, M. Patricia, Murray Li, Tania, MacDonald, Martha, Parpart, Jane L.,Parpart, Jane L., Connelly, M. Patricia, et al. (2000). Feminism anddevelopment: theoretical perspectives. Theoretical Perspectives onGender and Development (pp. 51–159). Ottawa: International Develop-ment Research Centre.

Deyessa, Negussie, Berhane, Yemane, Alem, Atalay, Ellsberg, Mary, Emmelin,Maria, Hogberg, Ulf, & Kullgren, Gunnar (2009). Intimate partner violenceand depression among women in rural Ethiopia: a cross-sectional study.Clinical Practice and Epidemiology in Mental Health, 5(8).

Edin, Kerstin (2006). Perspectives on Intimate Partner Violence, Focusing onthe Period of Pregnancy. Academic dissertation. Umeå: Umeå University(http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-838)

Ellsberg, Mary, & Heise, Lori (2005). Researching violence against women: apractical guide for researchers and activists. Washington DC, UnitedStates: World Health Organization, PATH.

Ellsberg, Mary, Jansen, Henrica AFM, Heise, Lori, Watts, Charlotte H., &García-Moreno, Claudia on behalf of the WHO Multi-country Study onWomen’s Health and Domestic Violence against Women Study Team.(2008). Intimate partner violence and women’s physical and mentalhealth in the WHO multi-country study on women’s health anddomestic violence an observational study. The Lancet, 371, 1165–1172.

Eveline, Joan, & Bacchi, Carol (2005). What are we mainstreaming when wemainstream gender? International Feminist Journal of Politics, 7(4),496–512.

Gage, Anastasia J. (2005). Women's experience of intimate partner violencein Haiti. Social Science & Medicine, 61(2), 343–364.

Garcia-Moreno, Claudia, Jansen, Henrica A. F. M., Ellsberg, Mary, Heise, LoriL., & Watts, Charlotte H. (2006). Prevalence of intimate partner violence:findings from the WHO multi-country study on women's health anddomestic violence. The Lancet, 368(9543), 1260–1269.

Government Offices of Sweden (2010). On equal footing—policy for genderequality and the rights and role of women in Sweden's internationaldevelopment cooperation 2010–2015. Article No UD 10.062.

Grown, Caren, Gupta, Geeta Rao, & Kes, Aslihan (2005). Taking action: achievinggender equality and empowering women. UN Millennium Project, TaskForce on Education and Gender Equality. London: Earthscan.

Heise, Lori L. (2011). What works to prevent partner violence: an evidenceoverview. STRIVE Research Consortium, London School of Hygiene andTropical Medicine. London: LSHTM.

Heise, Lori L. (1998). Violence against women: an integrated, ecologicalframework. Violence against Women, 4(3), 262–290.

Jewkes, Rachel (2002a). Preventing sexual violence: a rights-based approach.The Lancet, 360(9339), 1092–1093.

Jewkes, Rachel (2002b). Preventing domestic violence: most womenwelcome inquiries, but doctors and nurses rarely ask about it. BritishMedical Journal, 324, 253–254.

Kilby, Patrick, & Olivieri, Kate (2008). Gender and Australian aid policy: canwomen's rights be advanced within a neo-liberal framework. AustralianJournal of International Affairs, 63(3), 319–331.

Koenig, Michael A., Lutalo, Tom, Zhao, Feng, Nalugoda, Fred, Wabwire-Mangen,Fred, Kiwanuka, Noah, et al. (2003). Domestic violence in rural Uganda:evidence from a community-based study. Bulletin of the World HealthOrganization, 81(1), 53–60.

Krantz, Gunilla, & Garcia-Moreno, Claudia (2005). Violence against women.Journal of Epidemiology and Community Health, 59(10), 818–821.

Laisser, Rose (2011). Health Care Workers' Attitudes and Perceptions on IntimatePartner Violence and their Views in Care and Prevention: A Cross-sectionalSurvey from Tanzania. Academic dissertation. Umeå: Umeå University(http://www.diva-portal.org/smash/record.jsf?parentRecord=diva2:456835&pid=diva2:457041)

Lawoko, Stephen, Dalal, Koustuv, Jiayou, Luo, & Jansson, Bjarne (2007).Social Inequalities in Intimate Partner Violence: A Study of Women inKenya. Violence and Victims, 22(6), 273–284.

Lundgren, Eva, Heimer, Gun, Westerstrand, Jenny, & Kalliokoski, Anne-Marie(2002). Captured Queen: Men's Violence against Women in “Equal” Sweden—A Prevalence Study. Uppsala: Uppsala University (http://www.nada.kth.se/~stefan/CapturedQueen.pdf)

Lövestad, Solveig, & Krantz, Gunilla (2012). Men's and women's exposureand perpetration of partner violence: an epidemiological study fromSweden. BMC Public Health, 12, 945.

Mohanty, Chandra Talpade (1988). Under Western eyes. Feminist scholarshipand colonial discourses. Feminist Review, 30, 61–88.

Moser, Caroline (2005). Has gender mainstreaming failed?—a comment oninternational development agency experiences in the South. InternationalFeminist Journal of Politics, 7(4), 576–590.

Mukhopadhyay, Maitrayee (2004). Mainstreaming gender or “streaming”gender away: feminists marooned in the development business. IDSBulletin, 35(4), 95–103.

Nussbaum, Martha, & Clover, Jonathan (Eds.). (1995). Women, Culture andDevelopment. Oxford: Oxford University Press.

Okenwa, Leah E., Lawoko, Stephen, & Jansson, Bjarne (2009). Exposure tointimate partner violence amongst women of reproductive age in Lagos,Nigeria: prevalence and predictors. Journal of Family Violence, 24(7),517–530.

Oyediran, Kolawole Azeez, & Isiugo-Abanihe, Uche C. (2005). Perceptions ofNigerian Women on Domestic Violence: Evidence from 2003 NigeriaDemographic and Health Survey. African Journal of Reproductive Health /La Revue Africaine de la Santé Reproductive, 9(2), 38–53.

Okenwa, Leah, & Lawoko, Stephen (2010). Social Indicators and PhysicalAbuse of Women by Intimate Partners: A Study of Women in Zambia.Violence and Victims, 25(2), 278–288.

Rani, Manju, Bonu, Sekhar, & Diop-Sidibé, Nafissatou (2004). An EmpiricalInvestigation of Attitudes towards Wife-Beating among Men andWomen in Seven Sub-Saharan African Countries. African Journal ofReproductive Health, 8(3), 116–136.

Rathgeber, Eva (2005). Gender and development as a fugitive concept.Canadian Journal of Development Studies, 26, 579–591 (sup1).

Schech, Susanne (1998). Between tradition and post-coloniality: the locationof gender in Australian development policy. Australian Geographer, 29(3),389–404.

Sida (2007). Action Plan for Sida's Work Against Gender-Based Violence 2008–2009.Stockholm: Sida, Department for Democracy and Social Development.

Sida (2010). On Equal Footing Policy for Gender Equality and the Rights andRole of Women in Sweden's International Development Cooperation2010–2015. Stockholm: Government Offices of Sweden.

Silfver, Ann-Louise (2010). Emancipation or neo-colonialism? Global gendermainstreaming policies, Swedish gender equality and local negotiationsabout putting gender into education reforms in the Lao People'sDemocratic Republic. Compare, 40(4), 479–495.

Sörlin, Ann (2011). Health and the Elusive Gender Equality: Can the Impact ofGender Equality on Health be Measured? Academic dissertation. Umeå:Umeå University (http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-46298)

Sweetman, Caroline (2012). Beyond gender mainstreaming—introduction.Gender and Development, 20(3), 389–403.

United Nations (1979). Convention on the Elimination of All Forms ofDiscrimination Against Women. Division for the Advancement ofWomen, Department of Economic and Social Affairs (http://www.un.org/womenwatch/daw/cedaw/)

United Nations Resolution A/RES/55/2 (2000). The United Nations MillenniumDeclaration, New York, United Nations, 8 September. http://www.un.org/millennium/declaration/ares552e.htm

Uthman, Olalekan Abdulrahman, Moradi, Tahereh, & Lawoko, Stephen(2009). The independent contribution of individual-, neighbourhood-,and country-level socioeconomic position on attitudes towards intimatepartner violence against women in sub-Saharan Africa: A multilevelmodel of direct and moderating effects. Social Science & Medicine, 68,1801–1809.

Uthman, Olalekan A., Lawoko, Stephen, & Moradi, Tahereh (2010). Sexdisparities in attitudes towards intimate partner violence againstwomen in sub-Saharan Africa: a socio-ecological analysis. BMC PublicHealth, 10, 223.

Valladares, Elliette (2005). Partner violence during pregnancy, psychosocialfactors and child outcomes in Nicaragua. Umeå, Sweden: PhD thesis,Umeå University.

Walby, Sylvia (2005). Introduction: comparative gender mainstreamingin a global era. International Feminist Journal of Politics, 7(4),453–470.

Warren, H. (2007). Using gender-analysis frameworks: theoretical andpractical reflections. Gender and Development, 15(2), 187–198.

WHO (2001). Putting women first: ethical and safety. Recommendations forResearch on Domestic. Violence Against Women. Geneva: World HealthOrganization.

Page 124: Womens Studies International Forum 46 (2014)

122 A. Öhman, M. Emmelin / Women's Studies International Forum 46 (2014) 115–122

WHO (2005). WHO Multi-country Study on Women's Health and IntimatePartner Violence Against Women: Initial Results on Prevalence, HealthOutcomes and Women's Responses. Geneva: World Health Organization.

WHO (2007). Strategy for Integrating Gender Analysis and Actions into theWork of WHO. http://www.who.int/gender/mainstreaming/en/index.html

WHO (2012). Violence against women. Intimate partner and sexual violenceagainst women. Fact sheet No. 239 (http://www.who.int/mediacentre/factsheets/fs239/en/)

World Economic Forum (2011). The Global Gender Gap Report 2011: Rankingsand Scores. http://www3.weforum.org/docs/GGGR11/GGGR11_Rankings-Scores.pdf (Geneva, Switzerland)

Page 125: Womens Studies International Forum 46 (2014)

Women's Studies International Forum 46 (2014) 123–131

Contents lists available at ScienceDirect

Women's Studies International Forum

j ourna l homepage: www.e lsev ie r .com/ locate /ws i f

The ‘refugee’ and the ‘nexus’ requirement

The relation between subject and persecution in theUnited Nations Refugee Convention

Eva NilssonUmeå University, 901 87 Umeå, Sweden

a r t i c l e i n f o

0277-5395/$ – see front matter © 2013 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.wsif.2013.12.008

s y n o p s i s

Available online 24 January 2014

The challenge that my paper deals with is the complexities of gender and violence withininternational refugee law, taking women exposed to male partner violence as a starting point.The focus is the definition of ‘refugee’ in the United Nations Refugee Convention and therequirement that the persecution must be based on specific grounds, the ‘nexus’ requirement.My analysis shows that the Convention is grounded in an essentialist understanding of thesubject and that the preservation of its structure and integrity also means preserving thepower structures in society. The argumentation suggests that it is time to abolish the ‘nexus’requirement and the limitation of the grounds, but my conclusion is rather that we mustcontinue to work with our frame of thought focusing on the ‘refugee situation’ and thediscursive constitution of the subject in time and space.

© 2013 Elsevier Ltd. All rights reserved.

Introduction

My paper deals with the challenges posed by the complex-ities of gender and violence within international refugee law,taking women exposed to male partner violence as a startingpoint. The focus of my challenge is the definition of ‘refugee’ inthe Convention relating to the Status of Refugees (1951) andthe requirement that the persecution must be based on specificgrounds, the ‘nexus’ requirement. The definition covers onlypersecution for reasons of race, religion, nationality,membershipof a particular social group or political opinion. In many states,such as Sweden, women are recognized as constituting a‘particular social group’, making this a ground of particularinterest in my analysis. In the case of male partner violence it isalso this ground that is often invoked. In this instance it must beshown that the violence stems from her membership of a‘particular social group’. In this way, the persecution (violence)and the ground for persecution (being a woman and thus partof a ‘particular social group’) are separated as beingnon-constitutive of each other. My paper challenges thisseparation between the subject and the persecution. It ischallenged in the light of studies revealing that women – and

ll rights reserved.

these women in particular – have difficulties showing that thepersecution that they fear is grounded in their ‘membership of aparticular social group’. It is also challenged in the light oftheories that the subject is constituted in discourses throughprocesses involving actions such as persecution.

This means that the analysis has both an empirical andtheoretical starting point. The theoretical starting point willbe described more thoroughly, since it is primarily that whichwill contribute to new knowledge in this area of law, whilethe empirical starting point – that the definition of refugeehas been interpreted historically within a framework of maleexperiences – will only be used as a ‘backdrop’ in the analysis,since it is well-established internationally (see UNHCR, 2002a).Hence, this is an argument that has been made, well andextensively, for the past twenty years (see e.g. Bailliet, 2012;Bhabha, 2004; Crawley, 2001; Edwards, 2003; Freedman, 2010;Heyman, 2005; Kneebone, 2005; Macklin, 1995; Mascini & vanBochove, 2009; Randall, 2002; Spijkerboer, 1994; Tuitt, 1996).In Sweden too, studies suggest that female asylum seekers arediscriminated against in the asylum determination process(Cheikh Ali, Querton & Soulard, 2012, p. 30. See e.g. Bexelius,2008; Feijen & Frennmark, 2011; Segenstedt & Stern, 2011;

Page 126: Womens Studies International Forum 46 (2014)

124 E. Nilsson / Women's Studies International Forum 46 (2014) 123–131

Zamacona Aguirre, 2008). I myself have made a contribution tothis argument in a recently published article (Nilsson, 2012).The point of departure for that articlewas the changes thatweremade in the new Swedish Aliens Act (SFS 2005:716) of 2006concerning the definition of ‘refugee’, implying inter alia thatpersecution on account of gender should be covered by thedefinition, and empirical studies showing that these changeshave had little impact in practical applications. The conclusion inmy study is, however, that there is a gap not only between thelaw and practical applications, but also between the interna-tional legal framework and the changes made in nationallaw. Instead, it seems that gender equality is not favored in thepolicy area of asylum,while it is a commonly observed feature inother policy areas. Accordingly, under the threat of masses ofasylum-seekers crossing our borders, there are very limitedpossibilities for taking gender into account, as globalization andthe feminization of asylum-seekers increases this threat. It isparticularly the case if ‘we’ revise our self-image as a goodand equality producing state (‘non-refugee producing state’)and acknowledge that it is not only ‘the other’ women (andchildren) that are being oppressed. Holding on to internationalrefugee law may from this perspective seem impossible,applying it in a ‘generous spirit’ even less possible.

The study, however, also raised questions about the basicpremises in the international legal framework upon whichthe changeswere based. As alreadymentioned, this is the focusof the current article. In other words: this article takes thetheories and studies referred to as starting points for an analysisof problems concerning the international legal regime, but it isthe international legal regime that is the focus of my analysis andthat is the subject of my challenge.

I will start with a description of my theoretical startingpoints. I will then turn to the international legal framework.In this part of the paper I will provide a background to anddescription of the 1951 Convention and its definition of arefugee as amended by the Protocol relating to the Status ofRefugees (1967), giving a more detailed description of the‘membership of a particular social group’ ground. I will then payparticular attention to statements made by the United NationsHigh Commissioner for Refugees, UNHCR, concerning theinterpretation of the concept. The Commissioner's statementsare not legally binding, but they provide legal interpretativeguidance for interpretation of the Convention and Protocol.1 Inmany states, including Sweden, they also form the basis forrecognizing women as constituting a ‘particular social group’whichmakes them important in the analysis. Thereafter followsthe analysis and discussion, where I will problematize theConvention's basic design, the ‘nexus’ requirement and theseparation that it implies, in the light of the theories and studiesreferred to. Finally, I will present my conclusion.

Theoretical starting point

The analysis starts from a critical perspective based on afeminist legal theory. A feminist perspective on law representsan expansion of the field of women's legal research, fromresearch into legal issues with direct relevance to women toresearch from a more complex, systems-critical perspectiveinto the premises and assumptions upon which law is based(Gunnarsson & Svensson, 2009). It involves both a distrust oflawand a desire to change the power system,which is linked to

sex/gender and signifies and shapes the law (Nilsson, 2007a).However, an intersectional perspective is also used. This is acritical perspective that expresses an understanding thatgender does not include one social position but many andfocuses on the interaction between various power structures(Crenshaw, 1989, 1991).2 In contrast to power analysis basedon the asymmetric position of different groups in a hierarchicalorder, however, an intersectional perspective seeks to exam-ine the context in which the definition of groups and theseparation into different (and unequal) categories becomesmeaningful and a basis for the exercise of power. It is, therefore,essential to examine the temporal and spatial design of theexercise of power but, as well as challenging the boundaries oftime and space, an intersectional perspective also offers anopportunity to transcend the analytical divisions created bythe categories of class, sexuality, gender and race/ethnicity.To examine how e.g. class is intersected by both gender andrace stereotypologies, or how the constitutive role of sexualityfunctions in the creation of a gender order, means questioningthe ‘logic of distinguishing’3 which sustains the notion ofhomogeneous and hierarchically divided power relations. Thisis a form of analysis which requires that attention be paid tothe discursive structures and material conditions that makethese categories meaningful and indispensable conditions forthe exercise of power. From this perspective, it is not only therelationship between different categories which is of interest;the question is rather how these categories are created andgiven meaning in specific contexts (de Los Reyes, 2007; de LosReyes & Gröndahl, 2007).

This is a contextualized approach to law and legalknowledge which builds on social constructionist thinking. Asocial constructionist theory of law focuses onhow law relates tothe social, in that it not only regulates something that exists priorto law, but is also in itself part of the social construction of thatsame reality (Burman, 2007). A common approach among socialconstructionist researchers is discourse analysis where theoryandmethod are intertwined, in a theoretical andmethodologicaltotality, focusing on language. However, within discourse theorythere is not one single approach but a series of interdisciplinaryandmultidisciplinary approaches (Winther Jørgensen & Phillips,2000). Some of these include detailed linguistic analysis; somedo not (Fairclough, 2003). A number of key premises, however,constitute the basis for most discourse analysts, however,building on social constructionist theory: First, discourse analysisassumes a critical approach to taken-for- granted knowledge. Itmeans that our knowledge and our worldviews are not seen asmirror images of the reality “out there”, but as a product of ourway of categorizing the world. Secondly, it is assumed that theway we understand the world and the categories and conceptsthat we use are historically and culturally specific. Discursiveaction is a form of social action as it helps to construct the socialworld (including knowledge, identities and social relationships)and thereby preserves certain social patterns. Thirdly, it assumesa relationship between knowledge and social processes. Knowl-edge is thus seen as something which is produced in socialinteraction,where one builds common truths and fights forwhatis true against what is false. This means that all forms of socialinteraction, but especially linguistic ones, are of great interest.Fourth, it assumes a relationship between knowledge and socialaction. In a particular worldview, some forms of actions are‘natural’ and others unthinkable. Different social worldviews

Page 127: Womens Studies International Forum 46 (2014)

125E. Nilsson / Women's Studies International Forum 46 (2014) 123–131

lead to different social actions, and the social construction ofknowledge and truth therefore has concrete social consequences(Burr, 1995; Winther Jørgensen & Phillips, 2000).

In addition to these key premises there are also differences inthe approaches that I have beenmostly inspired by, Faircloughs'critical discourse analysis and Laclau & Mouffes' discoursetheory. Hence, Fairclough, in contrast to Laclau & Mouffe,distinguishes between discursive practices and other socialpractices. This means that some social phenomena functionaccording to other logics and must be studied using other toolsthan discourse analysis. In Laclau &Mouffes' discourse theory nodistinction is made between the discursive and non-discursivesocial practices, i.e. all social practices are seen as discursive. Thatdoes not mean that everything is reduced to language in thediscourse theory, but that discourses are material. Some haveperceived discourse theory, in suchway that when everything isseen as discourse, reality ceases to exist. This perception is builton a misunderstanding. As in critical discourse theory, both asocial and a physical reality exist. In discourse theory, however,the physical reality is completely superimposed, i.e. all socialphenomena are considered to be organized according to thesame principles as the language. As the signs in the languageare relationally defined and only gain their meaning by beingdistinguished from other signs, social actions only gain theirmeaning in relation to other actions. Accordingly, all socialpractices can be seen as articulations, because they reproduceor alter the ordinary meanings assigned to them. In discoursetheory, the analysis aims to (according to a concept taken fromJacques Derrida) deconstruct the structures that make up our‘natural’ environment, i.e. the analysis serves to prove that thegiven organization of theworld is the result of political processeswith social consequences. The struggle to form meaning is,therefore, an important focus in concrete analyses (WintherJørgensen&Phillips, 2000. See also Laclau&Mouffe, 1985, 1990).This means that critical discourse analysis focuses on the use ofdiscourses as a means of exercising power (Niemi-Kiesiläinen,Honkatukia & Ruuskanen, 2007).

Furthermore, discourse theory provides a more developedtheory about the constitution of entities and identities,especially regarding group formation and collective identitywhich makes it of particular interest in my analysis.According to discourse theory, there is no objective relation-ship that determines into what groups a social space isdivided. Instead the formation of groups is always inclusionsin a non-determinable terrain and, like any other discourse, itserves only to rule out alternative interpretations. Throughthe discursive formation of groups we exclude ‘the other’,the one in relation to which we identify ourselves, and alsoignore the differences that exist within a group and hence allthe other ways in which we could also have formed groups.One important element in group formation processes is‘representation’. Since groups are not predetermined in asociety but exist only after being first expressed in words,someone has to talk about or to a group or on their behalf.The situation is not that a group is first created and thenrepresented, but that the group and the representation areconstituted in a single move, and once a group is representedit involves the entire community image, because the group isconstituted in relation to other groups. Group formation isthus part of the struggle concerning how the whole myth ofsociety is to be given its content and, conversely, different

social images imply a direction about how people should bedivided into groups. Group formation is therefore political(Winther Jørgensen & Phillips, 2000. See e.g. Mouffe, 2005).

In my opinion, law, as a social practice, represents adiscursive practice that helps to create and reproduce unequalpower relations between social groups. In addition, legaldiscourses have a privileged status among social discourses inthat they have state power behind them, providing legaldiscourses with an exceptional power to shape social relations(Niemi-Kiesiläinen et al., 2007. See also Burr, 2003; Smart,1989). For instance, construing ‘violence in the private sphere’as emotional, as individual actions taken for personal reasons,also means construing it as a ‘social problem’, placing theresponsibility for the protection on the home country (Nilsson,2012). The consequence of this construction in law is thatmany of thesewomenwill bewithout protection, since in theirhome country they are often dependent on their violent menfor their support and ‘protection’ (Jastram & Newland, 2003).Many of these women may also be threatened by their ownfamily if they attempt to leave their husband or partner, as itmay be perceived as ‘dishonoring’ the family (UNHCR, 2008).The law's constructions of violence and how it is understood inrelation to the subject is therefore of fundamental importancefor these women. As pointed out by Kyambi “the refugee is, infact, constructed in law by the process by which he [!] isrecognized as a ‘refugee’ for the purpose of the law.” (Kyambi,2004, p. 26.)

The international legal framework

Refugee status, on the universal level, is governed by the1951 Convention and the 1967 Protocol relating to the Status ofRefugees. These two international legal instruments have beenadopted within the framework of the United Nations and arelegally binding on those states which have ratified or accededto them. The 1951 Convention was adopted after the SecondWorld War, as the refugee problem had not been solved andthere was a need for a general definition of who was to beconsidered a refugee, to replace ad hoc agreements adopted inrelation to specific refugee situations which had been thesituation earlier (UNHCR, Handbook, para. 7). The definition isgiven in Article 1A. The first paragraph of the article deals with‘statutory refugees’, i.e. persons considered being refugeesaccording to international agreements adoptedbefore the 1951Convention. The second paragraph gives a general definition,according to which the term ‘refugee’ shall apply to any personwho:

“As a result of events occurring before 1 January 1951 andowing to well-founded fear of being persecuted forreasons of race, religion, nationality, membership of aparticular social group or political opinion, is outside thecountry of his nationality and is unable or, owing to suchfear, is unwilling to avail himself of the protection of thatcountry; or who, not having a nationality and beingoutside the country of his former habitual residence as aresult of such events, is unable or, owing to such fear, isunwilling to return to it. […]”

Hence, the application of the Convention is limited toevents occurring before 1 January 1951. The Contracting

Page 128: Womens Studies International Forum 46 (2014)

126 E. Nilsson / Women's Studies International Forum 46 (2014) 123–131

States may also, under Article B choose to restrict theapplication of the Convention to events occurring in Europe.By acceding to the 1967 Protocol, states undertake to applythe substantive provisions of the 1951 Convention torefugees as defined in the Convention without the 1951dateline and without geographic limitations. The definitioncovers only those persons who are outside the country oftheir nationality or, if the person is stateless, outside theirusual country of residence. Accordingly, internally displacedpeople are not covered. Furthermore, the paragraph onlycovers persons who have a “well-founded fear of beingpersecuted”. The persecution feared must also be based onspecific grounds, in that the definition covers only persecu-tion for reasons of race, religion, nationality, membership of aparticular social group or political opinion. Finally, it is arequirement that the person is unable, or unwilling, to availherself/himself of the protection of that country. Apart fromthese provisions of inclusion, the Convention also containsprovisions of cessation (Article 1C) and exclusion (Articles1D, 1E and 1F) from refugee status.

As described in the Introduction, it is the ‘nexus’requirement (the limitation of grounds) that is the focus ofthis paper. Among the five grounds enumerated in thedefinition ‘membership of particular social group’ is ofparticular interest. It is considered to be the least clear andis not defined by the Convention itself. According to theUnited Nations High Commissioner for Refugees, consistentwith the objective, purpose and language of the Convention,this category cannot be interpreted in such a way as to renderthe other Convention grounds superfluous or to function as a“catch all” that applies to anyone fearing persecution. Thus,to preserve the structure and integrity of the Convention'sdefinition of a refugee, a social group cannot be definedexclusively by the fact that it is a target for persecution.Nonetheless, according to the guidelines, persecutory actiontoward a group may be a relevant factor in determining thevisibility of a group within a particular society or may evencause the creation of a particular social group in society. It isalso pointed out that there is no “closed list” of groups thatmight constitute a ‘particular social group’. Instead the termshould be read “in an evolutionary manner, open to thediverse and changing nature of groups in various societiesand evolving international human rights norms”. Accordingto the guidelines a ‘particular social group’ is a group ofpersons who share a common characteristic other than theirrisk of being persecuted, or who are perceived as a group bysociety. The characteristic will often be one which is “innate,unchangeable, or which is otherwise fundamental to identity,conscience or the exercise of one's human rights”. Consequent-ly, the definition includes characteristics which are historicaland therefore cannot be changed, and those which, though it ispossible to change them, ought not to be required to bechanged because they are so closely linked to the identity of theperson or are an expression of fundamental human rights, aswell as groups that are based on a characteristic determined tobe neither unalterable nor fundamental but that, nonetheless,are perceived as a recognizable group in that society. Hence,the definition includes two approaches: the first, referred to asthe “protected characteristics” approach, or the “immutability”approach, and the second, referred to as the “social perception”approach. Both approaches are based in state practice in

common law jurisdiction. According to the guidelines, in civillaw jurisdiction most decision-makers placemore emphasis onwhether or not a risk of persecution exists than on the standardfor defining a particular social group (UNHCR, Handbook, paras.5–8). The requirement that the persecution must be based in aconvention ground is, however, usually still maintained, e.g. inSweden (see Wikrén & Sandesjö, 2010, pp. 167 ff.).

According to the UNHCR guidelines on gender-relatedpersecution the size of the group has sometimes beenused as a basis for the refusal to recognize ‘women’generally as a particular social group (UNHCR, Guidelineson international protection: Gender-related Persecution,para. 31). However, in these guidelines, as well as in theguidelines on ‘membership of a particular social group’, itis emphasized that the size of the purported social groupis not a relevant criterion (UNHCR, 2002a,b). Theguidelines on gender-related persecution also state thatsex can properly fall within the ambit of the category ofsocial group, with women being a clear example of asocial subset defined by innate and immutable charac-teristics, and frequently treated differently from men.Their characteristics also identify them as a group in asociety, which subjects them in some countries to differenttreatment and standards. A distinction is, however, madebetween the terms “sex” and “gender”, according to whichgender refers to the “relation between women and men basedon socially or culturally constructed and defined identities,status, roles and responsibilities that are assigned to one sex oranother”, while sex “is a biological determination” (paras. 3and 30). The importance of this distinction was later empha-sized in the UNHCR Handbook for the Protection of Womenand Girls.

Furthermore, according to the guidelines on ‘membershipof a particular social group’, cases asserting refugee statusbased on this ground frequently involve claimants who facerisks of harm at the hands of non-state actors, with womenwho risk being abused by their husbands or partners beingone example. In this connection it is noted that situationsmay occur in which a claimant is unable to show that theharm inflicted or threatened by the non-state actor is relatedto one of the five grounds. The situation of domestic abuse isgiven as one example, where a wife may not always be ableto establish that her husband is abusing her based on hermembership of a social group, political opinion or anotherConvention ground. In those cases she may be able toestablish a valid claim for refugee status if the state isunwilling to extend protection based on one of the fivegrounds, i.e. the harm visited upon her by her husband isbased on the State's unwillingness to protect her for reasonsof a Convention ground. Hence, the causal link between thepersecution and the ground may be satisfied where there is areal risk of persecution at the hands of a non-state actor forreasons which are related to one of the Convention grounds,whether or not the failure of the state to protect the claimantis Convention related, or the risk of being persecuted at thehands of a non-state actor is unrelated to a Conventionground, but the inability or unwillingness of the state to offerprotection is for a Convention reason (UNHCR, Guidelines onmembership of a particular social group, paras. 20, 22 and23 and UNHCR, Guidelines on gender-related persecution,para. 21).

Page 129: Womens Studies International Forum 46 (2014)

127E. Nilsson / Women's Studies International Forum 46 (2014) 123–131

The ‘structure and integrity’ of the Convention

As described, ‘membership of a particular social group’ isregarded as the ground that has the least clarity. It seems,however, that the position taken by the UNHCR concerning theinterpretation of this criterion is also somewhat ambivalent.Although it is stated in the guidelines that “persecutory actiontoward a group may be a relevant factor in determining thevisibility of a group in a particular society or even cause thecreation of a particular social group”, it is pointed out at thesame time that, in order “to preserve the structure andintegrity” (the objective, purpose and language) of theConvention, a group “cannot be defined exclusively by the factthat it is targeted for persecution”. According to the guidelines,there is no “closed list”, of exactly which groups may becharacterized as ‘a particular social group’. Instead, the term“should be read in an evolutionary manner, open to the diverseand changing nature of groups in various societies and evolvinginternational human rights norms.” Thus, for this reason theUNHCR position is also ambivalent, as it prescribes both a‘historic’ and an ‘evolutionary’ interpretation of the concept. Inmy opinion, these disparate positions reveal theoretical incon-sistencies which can be explained by the fact that theConvention's basic design was laid down during a time whenthe mainstream western perception of the subject as anautonomous and sovereign unit prevailed, while the guidelinesare based on state practice of a more recent date, reflecting ‘thesocial turn’ and changes in this ‘historic’ perception of thesubject. The theories reflected in state practice are however onlypartially implemented in the UNHCR guidelines. Even if ‘thesocial perception approach’ and a couple of other statementsmade by the UNHCR reflect these theories, it is obvious thatthey are “alien” ‘elements’ in a discursive ‘structure’ marked by‘essentialism’ (cf. also the discussion by Verdirame, 2000, pp.588 ff., particularly on p. 594, about the definition of the fourgroups in the Convention on the Prevention and Punishment ofthe Crime of Genocide, 1948.). Perhaps, this is particularlyobvious when it comes to the category of ‘women’. In this case itis explicitly stated in the guidelines that the defining character-istics are “innate and immutable”which reveals that the UNHCRposition is also somewhat ambivalent. This is, however, aperception that can be challenged in light of the theoriesdescribed (but also, today more than earlier, in light of changed“realities”). It can also be challenged in the light of itsconsequences for women. In the following part I will elaboratethese challenges more thoroughly, starting with the empiricalbasis for my analysis (the ‘backdrop’).

Preserving the ‘structure and integrity’ — but also thepower structures in society

To begin with, the requirement that the persecution mustbe based on specific grounds can be challenged in the lightof studies disclosing that women have particular difficultiesshowing that the persecution that they fear is grounded in their‘membership of a particular social group’. As described, thisempirical basis has also been an important starting point forthe UNHCR. The situation for women is, however, starting tochange, and the guidelinesmentioned above on gender-relatedpersecution are one part of a process that is raising awarenessand acceptance of how gender-related forms of persecution

can fall within the refugee definition, according to the UNHCR.Women and children exposed to various forms of gender-related persecution are also now being recognized as refugeesin many countries. It is only very recently, however, that agrowing number of countries have also recognizedwomen andgirls who are victims of domestic violence as refugees (UNHCR,Handbook for the Protection of Women and Girls). There aresubstantial differences in the protection offered, however,among those states which recognize victims of domestic abuseas refugees on the grounds of social group. According to oneline of reasoning recognition is restricted to those cases wherethe state fails to protect the victims of domestic assault becausethey are women. Another line of reasoning simply finds thatwomenwho are subjected to domestic violence are persecutedbecause of their social group (women), since their gender is asubstantial factor in their persecution (Kelley, 2002, pp.559–568, on p. 565, with further references).

Sweden is one of the countries that recognize gender-related persecution. After a revision in 2006, the definition of‘refugee’ explicitly covers persons who have a well-foundedfear of persecution on account of their gender,4 i.e. personswho, according to the previous legislation, were grantedprotection in accordance with a special safeguard provisionas ‘persons otherwise in need of protection’. It was alsoestablished in the preparatory work that gender mayconstitute ‘membership of a particular social group’ andthat persecution in the private sphere may be a basis forobtaining refugee status (prop. 2005/06:6). These changeswere made in order to implement the UNHCR guidelines but,as already mentioned above, the studies carried out so farsuggest that they have had little impact on practicalapplications. In my opinion it is not only the practicalapplications that are problematic; a closer reading of thepreparatory work reveals that the gender perspectiveimplemented in the legislation is also a ‘gender biased’perspective. In the investigation, on which the GovernmentBill is based, there is a discussion about ‘wife-battering’ whereit is pointed out that theremay bemany reasonswhy awomanis exposed to serious battering, for instance because she hasbeenunfaithful, or is accused of being unfaithful, the primary ordriving force normally being jealousy. According to theinvestigation it would be far-fetched to say that the action ofthe man in these cases is a result of her ‘membership of aparticular social group’ (SOU, 2004:31, p. 126). In my earlierarticle (Nilsson, 2012) I argue that it is not so far-fetched,however, to say that the expressions that the jealousy takes oris ‘allowed’ to take (actions available in the situated emotionalrepertoire), are due to her ‘membership of a particular(subordinate) social group’. For, if one assumes (which theinvestigation does) that it ismainlywomenwhoare exposed tosuch persecution, why then is it men who do the hitting? Dowomen not get jealous? If not, is that related to gender? And, ifthey do get jealous, why do they not hit ‘because of’ it?What ismore, if we acknowledge feelings – or the expressions theytake or are allowed to take (violence) – to be gendered, howcan we even differentiate between gender (ground) andviolence (persecution)? There is no analysis of this in theinvestigation. Yet, the structural abuse of which this persecu-tion is an expression is construed as emotional, as individualactions taken for personal reasons, just because it is manifestedin a family context.5 Thus, women who risk abuse from their

Page 130: Womens Studies International Forum 46 (2014)

128 E. Nilsson / Women's Studies International Forum 46 (2014) 123–131

husbands or partners may be excluded from refugee statusbecause the persecution that they risk is not defined as beingbased on ‘membership of a particular social group’ (she is notbeaten because she is a woman). Should the abuse neverthelessbe defined as being ‘on account of gender’, there is also therequirement that the state's inability to provide protection is ‘onaccount of gender’ and not simply due to lack of resources orinefficiency.6 Hence, Sweden embraces the more restrictive lineof reasoning described earlier, among those states that recognizevictims of domestic abuse as refugees on the grounds of socialgroup. The consequence of this restriction is that women fromcountries that lack resources or countries with an inefficientstate administration are discriminated against in comparison tomale victims of persecution, as this is not required innon-gender-related cases when persecution is exercised byprivate actors.

Paradoxically, it was precisely constructions such as thesethat formed the basis of the criticism that was the starting pointfor the UNHCR guidelines on gender-related persecution. Asdescribed above, in the area of refugee law, there has beencriticism both before and since the adoption of the guidelines. Inmy opinion, this is not only because of differences among statesin the protection offered, i.e. a lack of implementation of theseguidelines but the Convention's basic design and some of thepositions taken by theUNHCR are also problematic, in relation towomen exposed to male partner violence. In the guidelines it ispresumed e. g. that it is difficult to demonstrate that the abuse insuch cases is related to a Convention ground, which reproducesthe ‘particularity’ of this abuse— but it also puts the statement inthe Swedish preparatory work about “jealousy” in a differentlight. Hence, to be determined ‘gendered’ or ‘structural’, ‘generalpatterns’ of abuse must be shown, but in the refugee area oflaw it seems that domestic abuse is not acknowledged as a‘universal’7 pattern of domination and control. Accordingly,when the persecution is familiar to our own ‘gendered practices’or when they are otherwise considered too ‘general’ the linkbetween the persecution and the ground is denied. If the‘otheringmechanisms’ are strong enough, however, the violencemay qualify as ‘structural’. So, in order to be considered‘structural’, the violence must not be ‘familiar’ or ‘general’, but‘other’ (Nilsson, 2012. See also discussion by Bailliet, 2012, pp. 44and 46; Freedman, 2010, p. 191;Mascini & van Bochove, 2009, p.130, as well as discussion about the ‘universality’ of this kind ofviolence by Randall, 2002, p. 17, and Charlesworth & Chinkin,2000, pp. 12 f. Cf. also discussion by Barsky, 1994, pp. 1–7 ff.,particularly on p. 5, about constructing an ‘appropriate refugee’or the ‘suitable other’). In other words, only abuse that differsfrom our own (‘naturalized’) structures is seen as ‘structural’.As I see it, ‘gender’ is always present, cutting through everyrelation (Sweden is not exempt from this, despite its commonlyobserved high level of gender equality.), i.e. every action isgender-related in some way, together with other discursivepower relations such as class and ethnicity (Nilsson, 2007a). Itmay be for this reason that it has proved to be so difficult to seeour own power structures, or to accept that there is such a thingas ‘persecution on account of gender’ (cf. Noll, 2000, pp. 9 f.;prop. 1996/97:25, p. 290).

The ‘logic of separation’ (between the subject and the action)in this case presumes a ‘logic of distinguishing’ (see above) thatsustains the notion of homogeneous and hierarchically dividedpower relations, but also preserves the power structures in

society. From this starting point, the distinction that isreproduced in the guidelines between the terms “sex” and“gender” are also problematic, as is the statement that thedefining characteristics of the category ‘women’ are “innate andimmutable”. In my opinion, it is open to question whether it ispossible to distinguish between sex and gender (cf. Folkelius &Noll, 2000, pp. 15 f.; Noll, 2000, p. 9, in polemic with Daley &Kelley, 2000, pp. 148–174) and whether the defining charac-teristics of the category ‘women’ are “innate and immutable”.Furthermore, in my view, persecution, along with otherdiscursive practices or actions, is an essential ‘element’ in theconstitution of subjects, both at a group level and an individuallevel. Interesting, in connection with this, is the criticism ofButler, referred to in the in the above-mentioned Swedishinvestigation. She asks how feminist criticism should respond toscientific discourses which proclaim “facts” about what “sex” isand wonders if sex is not as culturally constructed as gender(Butler, 1990a, p. 10). Butler's views are perceived, in theinvestigation, as a denial of the existence of a physical reality,and questions whether such a theory can be taken seriously(SOU, 2004:31, p. 166). Butler, however, makes it clear that hercriticisms do not mean that she ignores the physical realities orthe importance of physical bodies. She also stresses the need tospeak to and for ‘women’ (Butler, 1992, pp. 4, 8 and 13–17).According to Butler “[t]he foundationalist reasoning of identitypolitics tends to assume that an identity must first be in place inorder for political interests to be elaborated and, subsequently,political action to be taken.” Butler's argumentation is that“there need not be a ‘doer behind the deed’, but that the ‘doer’ isvariably constructed in and through the deed” and that it is “thediscursively variable construction of each in and through theother” that is of interest in her analysis (Butler, 1990a, p. 181.See also discussion by Butler, 1990a about “identity as an effect”on p. 187 and Butler, 1990b, p. 339, and the discussion aboutdiscursive and non-discursive social practices, above).

The understanding of sex as dual and dichotomous(where the defining characteristics of the respective categoryare “innate and immutable”), is, however, a perspective that fitslike a hand in a glove with the liberal (gender) project (see alsoNilsson, 2007a, p. 159; Widerberg, 2000, p. 129). Some peoplewonder how it will be possible to come together politically, ifwe can no longer define what a ‘woman’ is, biologically orculturally. Others believe that thedefinition is preciselywhatwemust let go of if we are to be truly political (Lindén & Milles,2002, p. 18). As described, questioning essential identities doesnot mean that we cannot continue to use the terms ‘workingclass’, ‘men’, ‘women’, ‘black’ or other ‘significants’ to referto collective entities. The possibility of political struggle onthe basis of these ‘significants’ also remains (Mouffe, 1992,pp. 370 ff. See also Mouffe, 2005, particularly on pp. 5 f.).

Conclusion

To sum up, my analysis shows that the Convention isgrounded in an essentialist understanding of the subject andthat the preservation of its ‘structure and integrity’ alsomeans preserving the power structures in society. It has beenestablished by the UNHCR that the Convention's basic design(its ‘structure and integrity’) requires persecution to be onspecific grounds, where the persecution can contribute tobut not in itself constitute, the subject. The Convention's

Page 131: Womens Studies International Forum 46 (2014)

129E. Nilsson / Women's Studies International Forum 46 (2014) 123–131

‘structure and integrity’ did not, however, prevent the 1967Protocol or previous limitations of space and time from beingremoved (events occurring before 1 January 1951 and inEurope). Furthermore, there are no limitations of grounds inthe protection following the International Covenant on Civiland Political Rights (1966) (art. 7) or in the Convention againstTorture (1984) (art. 3). These instruments are, however,limited in other ways. Universal protection against persecutionis thus lacking. Furthermore, there is a lack of internationalprotection for persons fleeing for reasons other than ‘persecu-tion’. This applies e.g. to the so-called climate (or environmen-tal) refugees, an issue that has received considerable attentionin recent years.8 This is a group of refugees that reallychallenges the Convention definition of ‘refugee’, and the‘nexus’ requirement in particular, but also, as with women,‘our’ self-image as a ‘non-refugee producing state’, since thematerial conditions (the social and physical reality) for theserefugees, aswell as for other refugees from the poorworld, is infact, to a great extent, a consequence of ‘our’ actions (see alsoBarsky, 1994, pp. 6 and 246, with further references). There aretherefore, as pointed out by de Los Reyes & Mulinari, reasonsfor problematizing time and the role of history when it comesto the creation of meaning and legitimacy in an unfair worldorder. A discourse which displaces the issue onto other timesand places has namely been conducive to hiding the way livingconditions for ‘the others’ in the poor world has been aprecondition for the welfare of ‘us’ in the rich world (de losReyes, & Mulinari, 2005, in reprint 2010, p. 76.).9

My argumentation suggests that it is time to abolish the‘nexus’ requirement and the limitation of the grounds and focusinstead on the protection available in the home country,regardless of a ‘relevant motive’. That was my conclusion tobeginwith, but I also realized that itmightmake the Conventionless useful as a legal instrument, presuming that law requiresdelimiting — in one way or another. So, my conclusion wouldrather be that we must continue to work with our frame ofthought. As argued, questioning essential identities does notmean that we cannot continue to use the terms. What we needto do is to ‘re-negotiate’ them, focusing on the ‘refugee situation’and the discursive constitution of the subject in time and space,i.e. how categories are created and given meaning in specificcontexts. Perhaps it is ironic that I stress the importance of timeand space, when the previous Protocol was added to make theConvention universally applicable, regardless of time and space.My aim, however, is the same: to increase its ‘universality’. Thisis perhaps also ironic (cf. note 7, above). I believe that this is thechallenge: exploiting the given, dichotomies and categories,acknowledging pluralism.10

Acknowledgements

In addition to the referees of this paper, I wish to thankparticipants in the research theme Violence at Umeå Centrefor Gender Studies (UCGS), Umeå University, for commentson drafts of this paper.

Endnotes

1 See e.g. http://www.unhcr.se/en/resources/legal-documents/guidelines-and-positions.html. In Sweden it is established in the preparatory work (prop.2004/05:170, p. 94) and in case law (MIG, 2006:1) that the UNHCR, 1979,

together with the conclusions that the UNHCR supports, provides guidance forapplication.

2 For more recent international literature on ‘intersectionality’, see e.g.Grabham, Krishnadas & Herman (2009), Lutz, Herrera Vivar & Supik (2011),McCall (2005), and Yuval-Davis (2006).

3 Compare the term ‘principle of keeping apart’ [isärhållandets princip],introduced by the historian Yvonne Hirdman in her theory about the gendersystem. The ‘principle of keeping apart’means that the relationship betweenmen and women is dichotomized. In symbiosis with this principle is theprinciple of the male primacy norm. This means that qualities which areperceived as masculine are valued more highly than features and tasks thatfall within the female sphere, leading to a hierarchization (Hirdman, 2001).Compare also with the term ‘logic of detachment’ [avskiljandets logik],introduced by the legal scholar Eva-Maria Svensson, referring to theseparation between law and society that traditional legal scholarship buildsupon (Svensson, 1997).

4 In the Act, the term ‘sex’ is used, referring to biological as well as socialsex. See prop. 2005/06:6, p. 34. In the English translation of the Swedish Actthe term ‘sex’ is translated as ‘gender’. See Chapter 4, Section 1 of the AliensAct, SFS 2005:716, available in English at http://www.sweden.gov.se/sb/d/5805/a/66122.

5 Similar results can be found in recent research from other countries.See e.g. Freedman (2010). See also Kneebone (2005, p. 24).

6 The UN Special Rapporteur on Violence against Women hascriticized Sweden in this respect and has stated that this interpretationof the law, which would introduce a double persecution requirement,diverges from the UNHCR gender guidelines (see HRC, 2007, 6 February).

7 I use the word ‘universal’ to capture ‘discursive structures’ and‘hegemony’, i.e. I lay no claim to ‘universalism’. Instead I believe that time andplace are central categories in analysing and ‘denaturalizing’ socially andhistorically created inequalities, such asmen's violence againstwomen (see alsode Los Reyes & Gröndahl, 2007, particularly on pp. 14; de los Reyes, 2007 on pp.101 ff., aswell as discussion byMouffe, 2005, particularly on pp. 83 ff. and 120 ff.and Butler, 1992, p. 8). For a discussion of challenges that an intersectionalanalysis grounded in a structural framework provide for understanding the roleof culture in domestic violence, see Sokoloff & Dupont (2005), particularly on p.58.

8 This was i.e. an issue for the Human Rights Days that were held inStockholm in November 2011. Program available at http://www.mrdagarna.nu/index.php/en/programme/fullstaendig-programlista.

9 Migrants also contribute to our welfare in terms of ‘brain-draining’or ‘social dumping’. Migrant women e.g. are often recruited into‘women-specific’ jobs that are unprotected and low paid. See adiscussion by Jean d´Cuhna at http://www.saynotoviolence.org/around-world/news/five-questions-jean-d-cunha, and de los Reyes & Mulinari,2005, in re-print 2010, on pp. 85 and 102 ff. See also Calleman (2011).

10 See also Nilsson (2005, 2007b), as well as Laclau (1996, p. 59),according to which “the impossibility of a universal ground does noteliminate its need: it just transforms the ground into an empty place whichcan be partially filled in a variety of ways (the strategies of this filling is whatpolitics is about).” See also discussion by Laclau (ib.) about ‘the death of thesubject’, ‘universalism’ and ‘particularism’, particularly on pp. 29 f.

References

National Official Documents

Prop (1996/97:25). Svensk migrationspolitik i globalt perspektiv [Governmentsproposal, Swedish migration policy in global perspective].

Prop (2004/05:170). Ny instans-och processordning i utlännings- ochmedborgarskapsärenden [Governments proposal, new order of instanceand process in aliens and citizenship matters].

Prop (2005/06:6). Flyktingskap och förföljelse på grund av kön eller sexuellläggning [Governments proposal, refugee status and persecution on accountof one's gender or sexual orientation].

SOU (2004:31). Flyktingskap och könsrelaterad förföljelse [Governmentcommission, refugee status and persecution on account of one's gender].

United Nations Official Documents

Convention on the prevention and punishment of the crime of genocide, New York.(1948, 9 December). (78 UNTS 277).

Page 132: Womens Studies International Forum 46 (2014)

130 E. Nilsson / Women's Studies International Forum 46 (2014) 123–131

Convention relating to the Status of Refugees, Geneva. (1951, 28 July). (189 UNTS137).

International Covenant on Civil and Political Rights, New York. (1966, 16December). (999 UNTS 171).

Protocol relating to the Status of Refugees, New York. (1967, 31 January). (606UNTS 267).

Convention against Torture and Other Cruel, Inhuman or Degrading Treatmentor Punishment, New York. (1984, 10 December). (1465 UNTS 85).

HRC (2007, 6 February). Implementation of General Assembly Resolution 60/251of March 2006 Entitled “Human Rights Council”. Report of the SpecialRapporteur on violence against women, its causes and consequences,Yakin Ertürk, Addendum, Mission to Sweden, A/HRC/4/34/Add.3.

UNHCR (1979). Handbook on Procedures and Criteria for DeterminingRefugee Status under the 1951 Convention and the 1967 Protocol relatingto the Status of Refugees, Geneva (HCR/IP/4/Eng/REV.1 (re-edited1992)).

UNHCR (2002a, 7 May). Guidelines on international protection: Gender-relatedpersecution within the context of Article 1A(2) of the 1951 Convention and/orits 1967 Protocol relating to the Status of Refugees (HCR/GIP/02/02).

UNHCR (2002b, 7 May). Guidelines on International Protection:”Membershipof a particular social group” within the Context of Article 1A(2) of the 1951Convention and/or its 1967 Protocol relating to the Status of Refugees(HCR/GIP/02/01).

UNHCR (2008). Handbook for the Protection of Women and Girls, Geneva.

Literature

Bailliet, Cecilia M. (2012). Persecution in the Home. Applying the duediligence standard to harmful traditional practices within human rightsand refugee law. Nordic Journal of Human Rights, 30(1), 36–62.

Barsky, Robert F. (1994). Constructing a productive other. Discourse theory andthe convention refugee hearing. Amsterdam/Philadephia John BenjaminsPublishing Company.

Bexelius, Maria (2008). Asylrätt, kön och politik. En handbok för jämställdhetoch kvinnors rättigheter [Asylum law, gender and politics. A handbookfor equality and women's rights]. Rådgivningsbyrån för asylsökande ochflyktingar.

Bhabha, Jacqueline (2004). Demography and rights: women, children andaccess to asylum. International Journal of Refugee Law, 16(2), 227–243.

Burman, Monica (2007). Straffrätt och mäns våld mot kvinnor: Omstraffrättens förmåga att producera jämställdhet [Criminal law and men'sviolence against women: The ability of criminal law to produce genderequality]. Uppsala Iustus Förlag.

Burr, Vivien (1995). An introduction to social constructionism. London Routledge.Burr, Vivien (2003). Social constructionism. London Routledge.Butler, Judith (1990). Gender trouble: feminism and the subversion of identity.

New York Routledge.Butler, Judith (1990, in re-print 1999). Gender trouble. In Linda J. Nicholson

(Ed.), Feminism/Postmodernism (pp. 324–340). New York Routledge.Butler, Judith (1992). Contingent foundations: Feminism and the question of

“postmodernism”. In Judith Butler, & Joan. W. Scott (Eds.), Feministstheorize the political (pp. 3–21). New York Routledge.

Calleman, Catharina (2011). Domestic Services in a “Land of Equality” — Thecase of Sweden. Canadian Journal of Women and the Law, 23(1), 121–139.

Charlesworth, Hilary, & Chinkin, Christine (2000). The boundaries ofinternational law: A feminist analysis. Manchester Manchester UniversityPress. Melland Schill Studies in International Law.

Cheikh Ali, Hana, Querton, Christel, & Soulard, Elodie (2012). Gender-relatedasylum claims in Europe: A comparative analysis of law, policies andpractice focusing on women in nine EU member states, France, Belgium,Hungary, Italy, Malta, Romania, Spain, Sweden and the United Kingdom.

Crawley, Heaven (2001). Refugees and gender: Law and process. Jordans Bristol.Crenshaw, Kimberlé (1989). Demarginalizing the intersection of race and

sex: A Black feminist critique of antidiscrimination doctrine, feministtheory and antiracist politics. University of Chicago Legal Forum.

Crenshaw, Kimberlé (1991). Mapping the margins: Intersectionality,identity politics, and violence against women of color. Stanford LawReview, 43(6), 1241–1299.

Daley, Krista, & Kelley, Ninette (2000). Particular social group: A humanrights based approach in Canadian jurisprudence. International Journal ofRefugee Law, 12(2), 148–174.

de Los Reyes, Paulina (2007). Rosengård är min Shatila!: Hegemoniskaberättelser och gränsöverskridande erfarenheter [Rosengård ismy Shatila!:Hegemonic narratives and transnational experiences]. In Paulina de LosReyes, Satu Gröndahl, Pia Laskar, LenaMartinsson, & Tiina Rosenberg (Eds.),Framtidens feminismer: Intersektionella interventioner i den feministiskadebatten (pp. 101–115)Future feminisms: Intersectional interventions in thefeminist debate (pp. 101–115). Hägersten Tankekraft Förlag.

de Los Reyes, Paulina, & Gröndahl, Satu (2007). Introduktion: Framtidensfeminismer: Vägar bortom törnrosalandet [Introduction: Future feminisms:Roads beyond the land of the Sleeping Beauty]. In Paulina de Los Reyes,Satu Gröndahl, Pia Laskar, Lena Martinsson, & Tiina Rosenberg (Eds.),Framtidens feminismer: Intersektionella interventioner i den feministiskadebatten (pp. 11–22)Future feminisms: Intersectional interventions in thefeminist debate (pp. 11–22). Hägersten Tankekraft Förlag.

de los Reyes, Paulina, & Mulinari, Diana (2005, in re-print 2010).Intersektionalitet [Intersectionality]. Malmö Liber.

Edwards, Alice (2003). Age and gender dimensions in International RefugeeLaw. In Erika Feller, Volker Türk, & Frances Nicholson (Eds.), Refugeeprotection in International Law, UNHCR's global consultations on interna-tional protection (pp. 46–80). Cambridge University Press.

Fairclough, Norman (2003, in re-print 2005). Analysing discourse. Textualanalysis for social research. New York Routledge.

Feijen, Liv, & Frennmark, Emelia (2011). Kvalitet i svensk asylprövning. En studieav Migrationsverkets utredning av och beslut om internationellt skydd[Quality in the asylum determination process. A Study of the MigrationBoard's investigations and decisions on international protection]. UNHCR.

Folkelius, Kristina, & Noll, Gregor (2000). Affirmative Exclusion: Kön, genusoch flyktingskydd i den svenska utlänningslagen [Affirmative exclusion:Sex, gender and refugee protection in the Swedish Aliens Act].Feministiskt perspektiv, 4(1), 14–18.

Freedman, Jane (2010). Protecting women asylum seekers and refugees: Frominternational norms to national protection? International migration, 48(1),175–198.

Grabham, Emily, Cooper, Davina, Krishnadas, Jane, & Herman, Didi (2009).Intersectionality and beyond: Law, power and the politics of location. NewYork Routledge-Cavendish.

Gunnarsson, Åsa, & Svensson, Eva-Maria (2009). Genusrättsvetenskap[Jurisprudence on gender and law]. Studentlitteratur Lund.

Heyman, Michael G. (2005). Domestic violence and asylum: Toward aworking model of affirmative state obligations. International Journal ofRefugee Law, 17(4), 729–748.

Hirdman, Yvonne (2001). Genus: Om det stabilas föränderliga former [Gender:On the changing forms of the stable]. Liber Malmö.

Jastram, Kate, & Newland, Kathleen (2003). Family unity and refugeeprotection. In Erika Feller, Volker Türk, & Frances Nicholson (Eds.),Refugee protection in international law. UNHCR's Global Consultations onInternational Protection (pp. 555–603). Cambridge University Press.

Kelley, Ninette (2002). The Convention refugee definition and gender-basedpersecution: A decade's progress. International Journal of Refugee Law,13(4), 559–568.

Kneebone, Susan (2005). Women within the refugee construct: Exclusionaryinclusion in policy and practice: The Australian example. InternationalJournal of Refugee Law, 17(1), 7–42.

Kyambi, Sarah (2004). National identity and refugee law. In Peter Fitzpatrick,& Patricia Tuitt (Eds.), Critical Beings Law. Nation and the global subject(pp. 19–36). Aldershot Ashgate Publishing Ltd.

Laclau, Ernesto (1996). Emancipation(s). London Verso.Laclau, Ernesto, & Chantal, Mouffe (1985, in re-print 2001). Hegemony and

socialist strategy: Towards a radical democratic politics. London Verso.Laclau, Ernesto, & Mouffe, Chantal (1990). Post-Marxism without apologies.

In Ernesto Laclau (Ed.), New reflections on the revolution of our time(pp. 97–132). London Verso.

Lindén, Claudia, & Milles, Ulrika (2002). Feministisk bruksanvisning: Essäer[Feminist manual: Essays]. Stockholm Norstedts Förlag.

Lutz, Helma, Herrera Vivar, Maria Teresa, & Supik, Linda (2011). Framingintersectionality: Debates on a Multi-faceted concept in gender studies.Surrey Ashgate.

Macklin, Audrey (1995). Refugee women and the imperative of categories.Human Rights Quarterly, 17(2), 213–277.

Mascini, Peter, & van Bochove, Marjolein (2009). Gender stereotyping inthe Dutch asylum procedure: ‘Independent’ men versus ‘dependent’women. International Migration Review, 43(1), 112–133.

McCall, Leslie (2005). The complexity of intersectionality. Journal of Womenin Culture and Society, 30(3), 1771–1800.

Mouffe, Chantal (1992). Feminism, citizenship and radical democraticpolitics. In Judith Butler, & Joan W. Scott (Eds.), Feminists theorize thepolitical (pp. 369–384). New York Routledge.

Mouffe, Chantal (2005). On the political. New York Routledge.Niemi-Kiesiläinen, Johanna, Honkatukia, Päivi, & Ruuskanen, Minna

(2007). Legal texts as discourses. In Margaret Davies, ÅsaGunnarsson, & Eva-Maria Svensson (Eds.), Exploiting the limits of law.A Swedish feminist challenge to pessimism (pp. 69–87). AldershotAshgate Publishing Ltd.

Nilsson, Eva (2005). A child perspective in the Swedish asylum process: Rhetoricand practice. In Hans E. Andersson, Henry Ascher, Ulla Björnberg, MaritaEastmond, & Lotta Mellander (Eds.), The asylum-seeking child in Europe(pp. 73–87). Göteborg Centre for European Research at Göteborg University.

Page 133: Womens Studies International Forum 46 (2014)

131E. Nilsson / Women's Studies International Forum 46 (2014) 123–131

Nilsson, Eva (2007). Barn i rättens gränsland. Om barnperspektiv vid prövning omuppehållstillstånd [Children in the borderland of law. On child perspectives inthe determination of granting residence permits]. Iustus Förlag Uppsala.

Nilsson, Eva (2007). Children crossing borders. On Child perspectives in theSwedish Aliens Act and the limits of law. In Margaret Davies, ÅsaGunnarsson, & Eva-Maria Svensson (Eds.), Exploiting the limits of law. ASwedish feminist challenge to pessimism (pp. 105–126). Aldershot AshgatePublishing Ltd.

Nilsson, Eva (2012). Persecution on account of one's gender: Refugee statusor status quo? Feminists@law, 2(1), 1–23.

Noll, Gregor (2000). Rätt utan gränser? Flyktingjuridik, genusteori och nyaavgränsningsproblem [Law without limits? Refugee law, gender theory andnew limitation problems]. Aktuellt om genusforskning, 2. (pp. 7–10).Centrum för genusvetenskap: Lunds universitet, 7–10.

Randall, Melanie (2002). Refugee law and state accountability for violenceagainst women: A comparative analysis of legal approaches to recognizingasylum claims based on gender persecution.HarvardWomen's Law Journal,25(281), 1–29.

Segenstedt, Alexandra, & Stern, Rebecca (2011). Vad krävs för att få skydd?[What is required to get protection?] Raoul Wallenberginstitutet förmänskliga rättigheter och humanitär rätt & Svenska Röda Korset.

Smart, Carol (1989). Feminism and the power of law. London Routledge.Sokoloff, Natalie J., & Dupont, Ida (2005). Domestic violence at the intersections

of race, class and gender. Violence against Women, 11(1), 38–64.Spijkerboer, Thomas (1994). Women and refugee status. Beyond the public/private

distinction. The Hague Emancipation Council.Svensson, Eva-Maria (1997).Genus och rätt: En problematisering av föreställningen

om rätten [Gender and law: A problematisation of the notion of law]. UppsalaIustus Förlag.

Tuitt, Patricia (1996). False images. The Law's construction of the refugee.London Pluto Press.

Yuval-Davis, Nira (2006). Intersectionality and feminist politics. EuropeanJournal of Women's Studies, 13(3), 193–209.

Verdirame, Guglielmo (2000). The genocide definition in the jurisprudenceof the Ad Hoc Tribunals. International and Comparative Law Quarterly,49(3), 578–598.

Widerberg, Karin (2000). Könskontraktets kropp: Kropp, medborgarskapoch kön [The gender contract of the body: Body, citizenship and gender].In Carsten Henrichsen, Anette Storgaard, & Jens Vedsted-Hansen (Eds.),Lovens Liv (pp. 125–138)The life of the law (pp. 125–138). KøbenhavnJurist- og Økonomförbundets Forlag.

Wikrén, Gerhard, & Sandesjö, Håkan (2010). Utlänningslagen med kommentarer[The Aliens Act with comments]. Solna Norstedts Juridik.

Winther Jørgensen, Marianne, & Phillips, Louise (2000). Diskursanalys som teorioch metod [Discourse analysis as theory and method]. Lund Studentlitteratur.

Zamacona Aguirre, Maite (2008). Riktlinjer för utredning och bedömningav kvinnors skyddsbehov – Ett fungerande verktyg? [Guidelines for theinvestigation and assessment of women's need for protection—A workingtool?] Svenska Röda Korset.

National Case Law

MIG 2006:1.

Internet-based references

http://www.mrdagarna.nu/index.php/en/programme/fullstaendig-programlista(accessed 7 November 2011)

http://www.sweden.gov.se/sb/d/5805/a/66122 (accessed 7 November 2011)http://www.saynotoviolence.org/around-world/news/five-questions-jean-d-

cunha (accessed 7 November 2011)http://www.unhcr.se/en/resources/legal-documents/guidelines-and-positions.

html (accessed 20 December 2011)