The Darker Side of Integration Policy

Despite its remarkable receptivity to immigrants entering the country, Sweden has one of the worst track records in Europe in terms of employing immigrants. However, although there are various studies devoted to integration issues, little scholarly attention has been directed toward public employment officers and their role as professional mediators in the integration process. By investigating the language used by public employment officers when they explain how they assist female immigrants in the process of entering the Swedish labor market, this study attempts to add to our knowledge of these officers and their role. It uses ideas from postcolonial theory and critical discourse analysis to show (a) how public employment officers make use of prevailing discourses to transform female immigrants from passive welfare beneficiaries into active and responsible job seekers, (b) how prevailing discourses contribute to the reproduction of culture and gender differences reminiscent of colonial discourses, and (c) how unequal power relations may favor the segregation of mainstream and alternative cultures.


Introduction
This article aims to investigate the language used by public employment officers when they explain how they assist female immigrants in the process of entering the Swedish labor market. It takes its point of departure from current debates about immigration and the challenges that the 21st century faces as a result of increased movement of human capital across national borders. An expected shortage of labor in the European market due to a decreasing population poses a challenge for politicians, who have to negotiate between a growing demand for an influx of non-European labor and a promise to their voters to limit immigration. The difficulty faced by immigrants in entering the Swedish labor market is a well-known problem among politicians, and the effort to come to terms with the issue has long been on the political agenda. Thus, despite policymakers' concern about diversity and a relatively well-developed framework for integrating immigrants into Sweden, immigrants continue to experience discrimination in the labor market. This is evident from studies showing not only that immigrants have higher unemployment rates and lower incomes than native Swedes but also that they are overrepresented in low-skilled and low-status jobs (De los Reyes & Wingborg, 2002;Rydgren, 2004 SCB -Statistiska Centralbyrån, [Statistics Sweden]; SCB, 2009). Figures from the Central Bureau of Statistics show that the income of all immigrant groups is significantly lower than that of native Swedes, and this hits immigrants from Africa and Asia the hardest as these groups are also the most exposed to unemployment. In 2008, the unemployment rate within these groups was 31.0% and 22.5%, respectively (SCB, 2009). It is noteworthy that their level of education is often higher than that of other immigrant groups as well as higher than that of native Swedes. Studies further show that highly educated immigrants often get stuck in low-skilled jobs that do not match their qualifications, and that those who get jobs within their professions earn lower incomes than members of the native-born population with similar training. Immigrants who obtain employment are often found in poorly paid jobs that seldom correspond to their education or earlier experience (Rydgren, 2004). According to De los Reyes and Wingborg's (2002) study, immigrants are particularly overrepresented among manual workers, cleaners, 575631S GOXXX10.1177/2158244015575631SAGE OpenMoulettes research-article2015 1 Lund University, Sweden restaurant assistants, and health care assistants. Furthermore, Bakshi, Hatlevall, and Melchert (2009) call attention to the fact that only 20% of newly arrived immigrants receive professional support from the Public Employment Service Agency (PESA; Arbetsförmedlingen). Their study further shows that the number of male immigrants who are given the opportunity to participate in work-supportive activities is higher than the number of female immigrants who are given such an opportunity. Besides the Swedish studies mentioned above, a number of international studies draw attention to the fact that immigrants are still subjected to both explicit and implicit prejudice in the labor market (e.g., Dietz, 2010;Esses, Dietz, & Bhardwaj, 2006;Hakak, Holzinger, & Zikic, 2010;Hosoda & Stone-Romero, 2010;Jackson, Kendrick, Tony, & Bryant, 2001;Petersen & Dietz, 2008). Esses, Dovidio, Jackson, and Armstrong (2001) argue, for instance, that even if the economies of nations are more interrelated and interdependent today, prejudice toward minority groups within the country as well as toward immigrants from other countries still characterizes the attitude of many individuals.
Although the integration of immigrants is fairly well documented in the literature, the main interest has so far been focused on the existence of discriminatory practices among employers and on unemployed immigrants' feelings about being excluded from the labor market. In spite of the apparently large amount of interest in integration issues, little scholarly attention has been directed toward public employment officers and their role as professional mediators in the integration process. Given the figures referred to above and the fact that Sweden has one of the worst track records in Europe in terms of employing immigrants (Sanandaji, 2012), I contend that public employment officers' role in the integration process deserves scholarly attention.
This article takes an interest in culture and gender discourses and the language used by public employment officers when describing how they help female immigrants to enter the labor market. The focus here on public employment officers is mainly due to the fact that unemployed job seekers have to register at their local employment agency to receive economic support from the government. This means that public employment officers are deeply involved in the employment process. Also, because the study focuses on a marginalized group, in terms of both nationality and sex, I am especially interested in how the power of prevailing discourses on culture and gender is reflected in public employment officers' discursive accounts with respect to this group. Thus, we seek to show (a) how public employment officers make use of prevailing discourses to transform female immigrants from passive welfare beneficiaries into active and responsible job seekers, (b) how prevailing discourses contribute to the reproduction of culture and gender differences reminiscent of colonial discourses, and (c) how unequal power relations may favor the segregation of mainstream and alternative cultures. For this purpose, we are drawing on Phillips, Lawrence, and Hardy's (2004) and Bové's (1995) definition of discourse as a collection of meaningful texts (in this case, spoken words) that are embedded in powerful institutions and practices. To achieve our aim, we will use a conversation technique and analyze the language used by public employment officers when they describe their encounters with female immigrants. Conversation technique refers to the use of open interviews resembling conversations rather than traditional interviews that follow a strict interview guide. We will use ideas from postcolonial theory and critical discourse analysis (CDA), as we see them as a potentially insightful lens for bringing to the surface representations of colonial imprints that may be hidden in the discursive accounts. Furthermore, an interpretative method will be used in the analyses and the presentation of the study.
The article begins by introducing some theoretical discussions that we find relevant to our study. Following an explanation of the research design, findings from the study will be presented and discussed.

Looking at Culture and Gender Through the Postcolonial Lens
The conceptualization of culture and gender differences finds its roots in the colonial discourse and its dichotomized and hierarchical ordering of the Western and the non-Western worlds (Fougère & Moulettes, 2012b;Jack & Westwood, 2006, 2009. As postcolonial scholars (Mignolo, 2011;Said, 1978) remind us, the colonial system favored colonizers over colonized, males over females, and European patriarchy over other forms of gender construction. Succeeding in the colonial conquest and its endeavor to achieve modernity and capitalism required a strategy imposing a social order that made the Europeans masters of the universe. The concept of coloniality and questions of colonialism and racism were highlighted by the Latin Americans Mariategui in the 1930s and O'Gorman in the 1950s (Mignolo, 2011), but it was Said (1978) who made Westerners aware of the way these ideas have influenced our perception of the world. Said's groundbreaking book, Orientalism, informs us that colonialism was one of the most profound and significant experiences that shaped Europeans' perception not only of Orientals but also, and perhaps mainly, of themselves-as well as Orientals' perceptions of themselves (Fougère & Moulettes, 2012a). As argued by Said, the imperialist powers needed to create the Orient as an "Other," to define themselves as the center. To achieve success in this endeavor required a colonial strategy that systematically led colonized people to understand themselves as the inferior and backward Other. From a postcolonial or post-Occidental (a concept coined by Mignolo) perspective, it is not the fear of difference per se but rather the fear of the superiority that the others may possess and of the threat this may pose to our image of cultural identity that causes ethnocentrism and the exclusion of those we perceive as different. Positioning culture as both resemblance and menace, Bhabha (1994) argues that "the colonial mimicry is the desire for a reformed, recognizable Other, as a subject of a difference that is almost the same, but not quite" (p. 86). From this standpoint, therefore, the notion of a homogeneous culture needs to be continuously repeated to be kept alive. The construction of social labels is thus not just a way to make existence easier to comprehend but a way to fix people-a fixation that, in spite of an expressed desire among the colonialists to mold the others into a common pattern and the latter's effort to become integrated, will keep them apart. As Fanon (1967) suggests, this is a culture that is "fixed in its colonial status . . . both present and mummified" (p. 44), and as such, it determines minorities' place in society.
The strategy that was applied also favored men's position in society by attributing to them characteristics associated with power, while undermining women's position by presenting them as disregarded adjuncts to males (A. Prasad, 2006;Said, 1978). The legacy of colonialism and its representation of men as rational, active, assertive, modern, and normal, while women are characterized as emotional, passive, irresolute, backward, superstitious, and abnormal, is still present today (e.g., Hofstede, 2001). It is, as Lugones (2000) points out, "what lies in the intersection of gender/class/race as central constructs of the capitalist world system of power" (p. 746). The continuous repetition of culture and gender differences has imposed a way of thinking about the world and the way we should live our lives. Like national culture, gender has traditionally been perceived as essentialistic and constituted as a homogeneous group identity (Collison & Hearn, 1996;Mohanty, 2003;Moulettes, 2007;A. Prasad, 2003). Hence, just as the inhabitants of a country are assumed to be bound together by the notion of shared culture, men, and likewise women, are assumed to be bound together by a sociological notion of "sameness." As a result, we see both culture and gender dichotomies operating normatively within an organization (Lugones, 2000) and thus contributing to the reproduction of gendered and cultural inequality outside the organization.
One reason why the stress on culture and gender has become so important in contemporary society may be people's enhanced vigilance in the effort to keep the discourse of cultural difference and the classification of people by race, ethnicity, geographic origin, gender, and the like alive in times of increased migration and external pressure and uncertainty. Although professionals may be convinced that they treat everybody the same, it is plausible that their subjective notions and pressure from the environment to keep the image of cultural and gendered homogeneity alive may affect their mind-set and actions. To judge from Meyer and Rowan's (1977) findings, organizations are likely to mirror the myths and values of their institutional environment, and irrespective of whether they wish to conform to these myths and values or not, people consequently contribute to the reproduction of the dominant discourse. According to P. Prasad (1997), organizations are places where colonial imprints are put into practice in the sense that they contribute to the shaping of organizational values and actions. They are also, as scholars such as Gherardi (1994) and Pullen and Rhodes (2012) inform us, powerful places for the (re)construction of gender. However, because the continuous repetition has transformed them into taken-for-granted features, it is a process that generally takes place unnoticed. Like our stereotyped images of culture, gender too passes, as Pullen and Knights (2007) point out, "as unnoticed, denied or disavowed, partly because it is 'done' routinely, repeatedly and unknowingly and with a degree of automaticity that conceals its precariousness and performativity" (p. 505). Keeping this in mind, I ask to what extent these notions may influence public employment officers and how much they are contributing to power asymmetries between them and those they are supposed to assist in finding jobs. We further ask what attitudes they present when describing their encounters with female immigrants and to what extent colonial imprints are discursively reproduced in their accounts. One way to investigate this is to combine postcolonial ideas with a discourse analytic technique and scrutinize the language used by public employment officers when they describe how they assist female immigrants to find jobs.

Discourse and Performativity
Any presentation of discourse should perhaps start by paying homage to de Saussure (1967), who introduced the idea that the relationship between structure (langue) and practice (parole) is arbitrary; to Derrida (1967), who claimed that it is as a result of logocentrism that texts can be seen to be structured around binary oppositions; to Lacan (1966), who claimed that the idea of a true and whole self is a myth; and to Foucault (1963Foucault ( , 1977, who argued that the social world and the relations of power that characterize it are determined by discursive formations. Besides having been important sources of inspiration to postcolonial scholars (e.g., Bhabha, 1994;Said, 1978;Spivak, 1988aSpivak, , 1988bYoung, 2000Young, , 2001, they have also influenced scholars such as Fairclough (1989), Grant, Keenoy, and Oswick (1998), Laclau and Mouffe (1985), Phillips et al. (2004), Potter andWetherell (1987), Van Dijk (1997), and Wodak (1989Wodak ( , 1997, who in turn have contributed to the expansion of discourse across disciplines and to the development of CDA. Broadly understood as talk and texts embedded in powerful institutions and practices, CDA takes an interest in the exploration and delineation of power relations in society. In practice, this approach to discourse generally combines discourse analytic techniques with a critical perspective to interrogate social phenomena (Ainsworth & Hardy, 2004).
There is, thus, a long tradition of critical research in the social sciences and humanities, but the view of what "critical" means and how it is justified in different traditions varies. Referred to as "ideology critique," one of the versions of critical discourse analysis targets the presumption, dissimulation, or manipulation of knowledge that shapes people's mind-set (Stenner & Marshall, 1995;Van Dijk, 1997). Scholars within this tradition argue that power relations in society are accompanied by a hegemonic language that systematically masks the world and claim that the aim of the critique is, therefore, to subvert power by showing the reality behind the ideology and what it really stands for (Winther Jørgensen & Phillips, 1999). For example, although people may say one thing, the continuous repetition of taken-forgranted assumptions often makes them act in a completely different way. Hence, the researcher's role is to unveil the ideology so that people can see what is hidden behind it and get the chance to change the prevailing discourses. By and large, this view corresponds to Foucault's (1963Foucault's ( , 1977 understanding of power as both productive and oppressive. Contrary to Van Dijk's (1993) emphasis on power as an exclusively oppressive force, Foucault explains power both as a discursively constructed subject position that people occupy and perform and as the object of various discourses that attempt to control and normalize people. Drawing on Bentham's design of a model prison, Foucault (1997) argues that all modern institutional forms mimic the patterns of a panopticon's way of organized surveillance and claims that the establishment of rules and judgments around the idea of norms "has become one of the great instruments of power" (p. 184). Rather than forcing people to follow the rules, according to Foucault, applying impersonal surveillance that produces subjects who act as if they are being constantly watched and thus discipline their own behavior so as to achieve normality has become a more successful means of control in modern society.
In this article, we lean on the above approach to CDA, as I use CDA to challenge mainstream research and its essentialistic, monolithic, and static assumption regarding both culture and gender. We are particularly interested in how power asymmetries contribute to the reproduction of takenfor-granted assumptions reminiscent of colonial culture and gender discourses. Said (1978) explains that language itself is a highly organized and encoded system, which uses many devices to express, indicate, and exchange messages and information. Questioning whether there can be a true representation of anything, he concludes that "we must be prepared to accept the fact that a representation is eo ipso implicated, intertwined, embedded, interwoven with a great many other things besides the 'truth,' which is itself a representation" (Said, 1978, p. 272).
Rather than viewing culture and gender identities as essentialistic, monolithic, and static, we agree with the idea that they are talked into being and performed according to prevailing discourses. As Perryman (2006) reminds us, the term performativity was coined by Lyotard (1984), who, in "The Postmodern Condition: A Report on Knowledge," argued that postmodern society is obsessed with efficiency and effectiveness, and that this efficiency is increasingly "measured according to an input/output ratio" (p. 88). Since then, the idea of performativity has been used by a number of scholars who take an interest in the connection between structures, institutions, and practice (Cameron, 1995;González, 2004;Lee, 2015;Perryman, 2006). For example, in her theory of performativity, Butler (1990) argues that gender identities do not simply exist; they are talked into being over time and are performed according to the discourses that existed prior to one's performance of them. The idea that performativity is about performing the normal within a particular discourse is, as pointed out by Cameron (1995), equally valid for any fixed and substantive social label. The continuous striving for normality is also the reason why cultural and gender differences always need to be repeated to be kept alive (Bhabha, 1994).
Drawing on this insight, I argue that employment officers are socialized into prevailing discourses and influenced by mechanisms that impose efficiency (Lyotard, 1984) and normality (Foucault, 1977). It is plausible to assume that they will perform in accordance with prevailing discourses and, in their effort to obtain efficiency, will attempt to transform unemployed people into subjects who secure current discourses, while excluding those deviating from normality. And as this often takes place routinely and unknowingly (Pullen & Knights, 2007), it is likely to favor power asymmetries and the reproduction of prejudice about "Others." This may occur when values and behavior patterns are imposed on a person and/or when a person is reminded that he or she does not belong. A classic example of the latter is provided when a person with a foreign-sounding name, who looks different or speaks with an accent, is asked where he or she comes from and, hence, is presumed to be a foreigner. Other examples occur when it is presumed that he or she will not behave according to the prevailing norms or when it is presumed that he or she lacks sufficient skills to perform efficiently. The latter may occur due to differences in working methods or when inadequate language skills are taken for granted to hinder immigrants' work performance.
In this article, I will scrutinize the impact of power asymmetries in the relationship between employment officers and unemployed female immigrants. I will pay particular attention to (a) how public employment officers make use of prevailing discourses to transform female immigrants from passive welfare beneficiaries into active and responsible job seekers, (b) how prevailing discourses contribute to the reproduction of culture and gender differences reminiscent of colonial discourses, and (c) how unequal power relations may favor the segregation of mainstream and alternative cultures. These issues are interrelated, in the sense that they all contribute to power asymmetries related to the neocolonial discourse.

Methodological Considerations and Research Technique
Using the transcripts of our interviews, I sought to go beneath the surface meaning in the participants' accounts and uncover the meanings hidden within them (A. Prasad & Mir, 2002). I used a convenience sampling as I was interested in the richness and nuances of the participants' accounts rather than figures, objective facts, and general patterns (Broadbridge, 2009). Given the limited number of participants, my study does not allow me to make any generalizations in the traditional sense. Nevertheless, as Denzin and Lincoln (2000) argue, neither an individual nor a case is just an individual or a case. The study of the particular is the study of the general, and the longer, more detailed, and more intensive the transcripts, the fewer the participants who are needed to saturate the empirical material (Rudestam & Newton, 2007).
The sample included 12 in-depth interviews with public employment officers in five PESA branches in the southern part of Sweden. The participants consisted of eight women and four men with different educational and career backgrounds (e.g., they included a former teacher, a lawyer, a former journalist, and several sociologists). They had worked for between 1 and 25 years at the PESA. As it was important to the research that the participants should feel comfortable, I let them choose the location for the interview. In most cases, the choice fell on their office, while a coffee shop in town and the home environment were chosen by some of the participants. The interviews lasted between 1 hr and 3 hr and were transcribed word for word.
As the research was exploratory in nature, I wanted the interviews to be respondent-driven rather than interviewdriven. The interviews were performed, therefore, as conversations instead of conventional interviews based on structural questionnaires, which meant that I had prepared only some broad questions that covered the main purpose of our research. It further meant that I had some general themes to guide the interviews, but I kept them as open as possible to allow respondents to answer as freely as they wished. The interview technique I chose provided an opportunity for the interviewees to reflect on the areas they found to be most pertinent to the professional/client relationship. This meant that in some instances, questions were initiated by the researcher, while in other instances, they were asked as natural follow-ups to the preceding conversation. The chosen approach allowed for flexibility without a loss of focus on the most salient points and enabled the collection of multilayered information without the superimposition of a rigid structure by the researchers (Broadbridge, 2009). All methods have their advantages and disadvantages, and a research method based on a seemingly loose protocol is no exception. Its major disadvantage is that there is no guarantee that it will provide a cohesive pattern, as the interviewees are allowed to speak freely and tell their own stories. However, this can be seen as one of the advantages, as it may provide information that would not be discovered in traditional interviews where structured or semistructured questionnaires are used. Put bluntly, the stories that people provide may unveil attitudes that would otherwise remain hidden.
The interview accounts analyzed in this study have been selected from a larger study of public and private labor market intermediaries' work in assisting immigrants to enter the Swedish labor market. One reason for focusing on female immigrants was that one of the questions asked in the larger study of the integration of immigrants was related to Bakshi et al.'s (2009) earlier study, which showed that female immigrants receive less assistance from the public agency than do men. Another reason was that the interviewees in the larger study often referred to female immigrants when they explained how they helped immigrants to find jobs.
A selection of material for an analysis is a sine qua non for any study, because for obvious reasons, it is neither useful nor even possible to include everything that is transcribed from interviews. As I wanted to draw attention to what is often silenced and suppressed due to the scholarly effort to obtain neutrality, I decided to interpret the accounts through a postcolonial lens. It was thus important to select accounts that contained material relevant to the aim of the article and to include statements about female immigrants. As a consequence of this process, the article does not provide a full picture of employment officers' perception of unemployed immigrants. This means that the positive attitudes toward immigration expressed by two of the interviewees, a man in his 50s and a woman in her 30s, are not included in the analysis. Both of these interviewees defended immigrants' rights and emphasized the economic, social, and cultural benefits of immigration. They also expressed concerns about some of their colleagues' as well as employers' negative attitudes toward immigrants.
The interviews were transcribed in Swedish and later organized into themes such as background and earlier experiences, administration, receiving job seekers, and assisting immigrants. The last-mentioned theme was divided into subthemes such as vocational guidance, vocational training, work practice, attitudes toward immigrants, and statements about female immigrants. In a second phase, a selection of interview accounts that were considered especially interesting for this article was translated into English. Due to the translation from Swedish to English, the excerpts do not always render the exact words or expressions used. However, an effort was made to keep the translation as close as possible to the original texts.

The Swedish Public Employment Service Agency
The first step toward the creation of a public employment agency was taken in 1940 with the establishment of the Labor Market Commission. The aim was to help people earn their living during World War II. Over time, the commission's activities continually developed to reflect changes in the labor market and society. A major change came in the early 1990s, when public sector reforms became a priority and market solutions were introduced within the PESA. Due to the influence of the neoliberal ideas that began to flourish in Europe in the 1970s, new public management was implemented by the right-wing government of Sweden in the early 1990s as a way to come to terms with a bureaucracy that was considered by politicians to have grown out of proportion (Green-Pedersen, 2002).
A significant change in employment officers' work routines came with technological developments and the introduction of a standardized coding system aiming at facilitating labor market control. This meant that the officers' conditions of work changed, in the sense that they now had to spend more time at the computer producing statistics for the authorities. To make the system more efficient, new routines, such as individual activation plans and work applications reports, have recently been introduced. The former are executed by the employment officer at the first meeting with a client, while the latter, which aims at showing the unemployed persons' active search for a job, is produced on a regular basis by the unemployed persons themselves.
Another significant change came in 1993, when the employment agency's monopoly expired and private labor market agencies were established Svensk för fattnings samling (SFS), 1993 [Swedish Code of Statutes]. However, although competition from private labor agencies is growing steadily, job seekers are still obliged to visit the PESA on a regular basis to obtain their unemployment allowance. Today, the public employment service consists of 320 local branches of the PESA and approximately 350 private employment service agencies, of which 165 are operating as coaching consultancies. The consultancies are financially supported by the government but receive their clients from the PESA.

Transforming Clients From Passive Welfare Beneficiaries Into Active and Responsible Job Seekers
One of the reasons behind the implementation of marketbased reforms was the desire to provide cheaper and more efficient services (Green-Pedersen, 2002). Besides the fact that officers have more clients to take care of, their administrative workload has also increased. Maria, a young employment officer, who had been at the agency for some years, explained, I have one hour to prepare this [action plan] before I meet them [the clients]. "Welcome to Sweden. I have 20 questions that I need to ask you," dah, dah, dah . . . and that's it. "I have written an action plan for you; here you are. This is our agreement. Thank you and goodbye." Judging from Maria's account, it appears that she is not satisfied with the current situation, as she now has to spend more time at the computer, entering statistical data and preparing action plans, than in assisting her clients to find jobs. Reflecting on how her work has changed, Maria explained, There used to be those who literally took the immigrants by the hand and helped them getting to the right place. "If you are unsure how to get to L., I can come with you." One might think that it would be easy for them to get there by themselves, maybe-if they have come all the way from Beirut . . . . Even if it was not part of their job description, some of her colleagues used to help immigrants who, they considered, needed special attention, to find their way to "the right place," which could be anything from a local authority office to an educational institution or a potential workplace. This kind of help was now considered too time-consuming and costly and had therefore been abandoned. The abandonment of this kind of assistance may partly be explained by the government's effort to get rid of the image of Sweden as a bureaucratic nanny state (Green-Pedersen, 2002). However, Maria's statement also demonstrates that the officers' freedom to act has been cut down and that they now are more careful to perform according to the rules. An enhanced anxiety among employment officers about being monitored was also demonstrated in the interview with Marta, who, although she was promised anonymity, was careful about her answers and, in response to several of the questions, claimed she did not have the necessary statistics to answer them. Besides demonstrating the existence of power asymmetries between employment officers and their supervisors, the officers' apparently enhanced caution may be seen as an indicator of panoptic performativity (Foucault, 1977).

Supportive Activities for Those Who Deserve Them
The idea of the active society is that exclusion from the labor market can be avoided through social and cultural investment (Caswell, Marston, & Elm Larsen, 2010). In Sweden, this idea manifests itself in a push for cultural and social investment to counteract the exclusion of marginalized groups from the labor market. At the time of the study, some of the staff members at one of the employment offices I visited were working on the guidelines for a new integration policy for newly arrived immigrants, which was going to be implemented later that year. Martin, an employment agency manager, and one of the people responsible for this task, speaks as a representative for the professional core when he explains the politicians' intention in implementing the reform: Within this new reform that we are preparing . . . there's a very important economic "incentive," that is, you can no longer take advantage of the Social Services Act and say that we have this family that needs money. Every adult has to apply for financial support and participate in activities to get it. This means that [for] the family or the woman . . . or whoever takes the ultimate decision in the family that someone shouldn't work but stay home and take care of the household and the children, it will hit them financially.
What Martin describes is a reform that encourages a policy where welfare benefits are associated with economic sanctions. Hence, the reform is intended to encourage all adult family members to register at the employment office, but it is particularly intended to put pressure on female immigrants to leave their lives as housewives and register as job seekers. The idea that female immigrants often stay at home to take care of their families is a view that most employment officers share; they also believe it to be the main reason why female immigrants receive less support from the PESA than male immigrants do. According to employment officers' statements, female immigrants' absence from working life for an extended period of time is also believed to make their later entry into the labor market more difficult. Philip, an employment officer specializing in coaching, was one of those who made the connection between child care and the difficulty of finding employment on later entry to the labor market: I believe that [the difficulty] is because female immigrants stay home and take care of their children. They stay away from working life 10 years before they register. They take care of their families. That's not common among Swedish women. Female immigrants who have stayed home and taken care of their children are very far from the labor market.
When Cesar was asked whether women are equally supported (i.e., to the same extent as men), he replied, Yes, all clients are equally supported. We support them as much as we can. We even offer a coach who can assist them in finding a position as a trainee. Judging by Cesar's statement, one possible explanation for the higher unemployment rate among female immigrants seems to be that they cannot be trusted to give their time to their jobs while they invest so much time in their children. This seems to be what Cesar is suggesting when he explains that these women will leave their job when they get a "call from the day care center" telling them that "their child has got a cold." Apparently, a job-seeking mother should not prioritize her children. It would be intriguing to ask what a Swedish parent (man or woman) would do in this situation. Besides the fact that Cesar's conclusion is simplistic and biased, it reproduces an image of female immigrants as a homogeneous group of child producers who cannot be trusted in employment as long as they give priority to their children. In this respect, Cesar seems to believe that all women coming from outside the Western world can be categorized as one homogeneous group of unreliable female immigrants. Cesar's statement that "as a trainee you have to show that you deserve the job" indicates that the help offered is not guided by a genuine concern for clients' needs but rather conditioned by a counter performance. In this particular case, the counter performance consists of showing that one has proven to be worthy of a job. Unemployed female immigrants, thus, seem to correspond to what Caswell et al. (2010) refer to as "at risk" clients, over whom officials, as professionals, have the power to generate consequences (Jenkins, 2000) if clients are not willing to adjust to the prevailing conditions and act as responsible job seekers. Besides demonstrating the existence of power asymmetries between agency officers and the unemployed, Cesar's statement indicates that the officers, as professionals, feel they are being put under enhanced pressure to normalize female immigrants, so as to make them seem more employable.

Correction of Cultural Flaws to Maintain the Notion of Cultural Superiority
The interview accounts further reveal how the officials guide their clients to make them more attractive in entering the labor market. An important feature of this work is to convince the immigrants of the importance of adjusting to Swedish culture. The view that cultural differences are flaws that need to be corrected is illustrated in the following account provided by Karin: Well, take the teaching profession, for example, where you have different school systems in different countries and where you have different views on teaching methods and we have our view. I had this problem with a woman from the Balkans once. There were quite a few people from the Balkans who . . . it was more about teaching methods and things like that. All the children are forced to write with their right hand. Such a teaching method does not work in the Swedish school. . . . But then I sit here as a wicked person when I tell this woman who has this education and who has worked for many years at home, "You have to reconsider. You have to take a supplementary course." . . . If you are a craftsman you can work as a trainee in a company and a competent supervisor can make an evaluation and say, "Yes, he knows this and this, but he seems to use other methods than the ones we are used to and needs to learn how we work." Karin highlights the differences between countries with regard to teaching methods and ways of working. In doing so, she implies that other educational systems do not match the Swedish standard and that immigrants therefore need to take supplementary courses. She is unwavering in her belief that the supervisor is "competent," and she uses the example of different teaching methods to exclude a client from a profession. She implicitly suggests that all members of a specific group need further education to live up to Swedish standards, rather than acknowledging that the Swedish system may benefit from incoming knowledge and practices. By doing so, she reproduces a notion of logocentrism (Derrida, 1967), which by its emphasis on ethnocentrism favors the image of a Westernized center. The example given above indicates that help has become an instrument of training (Gronemeyer, 2010), in which the participants' cultural adjustment is an important component. For the sake of economic development, it is apparently not the needs of the individuals that are prioritized, but society's need for cultural homogeneity. It may seem a bit strange, though, that something that seems so trivial should require so much further education. However, the search for differences even in the most trivial things may be seen as a strategy to keep the image of cultural differences alive (Bhabha, 1994). Because there were other interviewees who provided similar examples (e.g., that painters in Sweden learn to paint vertically, while painters in some other countries learn to paint horizontally), one may wonder whether the emphasis on further education is just used as a pretext or whether it is grounded on the presumption that the educational system in other countries never lives up to the Swedish standard.

"A Good Performance" Needs Appropriate Clothing?
Implicit in the deficit discourse, one can discern the importance of making immigrants break with their past to become integrated into Swedish society. The call for adjustment reflects a fear of difference or a fear of the possible superiority that the others might possess and what it might do to cultural identity (e.g., Appadurai, 2006;Bhabha, 1994;Said, 1978). What is frightening today seems to be the influence of non-Western cultures, and not least Islam and women dressed in veils. For example, while referring to a Muslim woman who had refused to remove her headscarf when offered a position as a trainee in a restaurant, Lisa, a career adviser, shared her own opinion of the political debate about the burka: . . . to get a job I do think that you actually may need to take off your headscarf. For example, there was this discussion about the burka in preschool and in this case I do think that you should take it off when you are at work . . . so that you are able to do a good job. Then, in your private life, you can wear it if you like.
Lisa's description conveys a value judgment about Muslim women's style of dress, ultimately explaining that she feels the implementation of a restriction is necessary for a client to get a job. It is unclear, however, what kind of headscarf or other covering she is referring to in this statement and whether she makes a distinction between different kinds of coverings or not. Nevertheless, she considers them inappropriate clothing during working hours. She does not clarify what she means by "doing a good job," and in what way a headscarf may be in conflict with a professional role and how it may hinder work performance. Yet her observation implies that wearing any piece of unconventional clothing is unsuitable and might have consequences for the employer. What is interesting about statements such as these is the frequency with which the "headscarf issue" surfaces whenever integration is discussed, but not in terms of the actual way in which it can cause a problem in a Swedish workplace. It is also surprising that Lisa refers to the wearing of a burka in preschools, considering that the number of women wearing the burka is strictly limited in Sweden and thus the use of this covering in schools does not seem to constitute a real problem. Whatever Lisa's general intention may be, the statement reflects the impact of the social disadvantage that the headscarf is causing to the women who wear it and the anti-Muslim sentiments that have become a frequent feature in the Western world including Sweden. The statement also indicates that there is a link between the fear of difference (Appadurai, 2006;Bhabha, 1994) and what it may mean for our supposed cultural identity. The discursive practice of marginalizing those considered different demonstrates power inequality and the supremacy given to one's own culture.

Exercise of Power in the Name of Civilization
Muslim culture is clearly regarded with suspicion, but this extends to any unfamiliar culture, behavior, and custom. An example of suspicion of unfamiliar behavior is provided by Eva, who admits to being provoked when a woman arrives at a scheduled meeting accompanied by her husband: I have worked with many women . . . women who come from other cultures, where their husbands come along and insist in joining the discussion because they have this feudal culture. In those cases . . . I don't think it's OK. We have to follow human rights. We have to follow the democratic principles and so forth. . . . But it's difficult because these women have to straighten their backs. I can say, "No, I don't accept" [that they bring their husbands along to the meeting] and book a new appointment.
The example demonstrates how Eva makes use of her power as a professional to correct female immigrants' behavior to make them adjust to what is presumably the expected behavior in Swedish culture. It further demonstrates how the reproduction of colonial imprints is used in practice. Judging from Eva's statement, the type of behavior she describes is simply not acceptable in a modern and democratic society. According to her, couples such as the one she describes come from a "feudal culture" that does not respect "human rights" and "democratic principles." So now when they live in Sweden, these women have to learn how to "straighten their backs," by which Eva presumably means that they have to become more independent. She assumes these women are oppressed, that is, that their husbands prevent them from interacting with society on their own terms, while instead they may be bringing their husbands along for mental support or because they feel insecure about their Swedish language skills. Eva's use of words such as "feudal," "human rights," and "democracy" disclose her view of her clients' cultures as backward and undemocratic and, hence, an assumption that all women in these countries are oppressed by their husbands. If in fact this is her assumption, it could be argued that these women are regarded as supplemental not only by their husbands but also by the employment officers.
Besides expressing a stereotypical and depreciatory view of both men and women from "other cultures," Eva's statement seems to be grounded in a fear that stems from views similar to those Lisa expressed about Muslim women who wear a headscarf. In essence, this line of thought corresponds to Jackson et al.'s (2001) suggestion that threat is the closest culturally universal link to prejudice and racism and the way in which people behave toward outgroups. The statements quoted above illustrate how power is exercised by the use of words derived from the language of the dominant cultural discourse. Rather than taking a serious interest in diversity, the statements point at the importance of protecting the notion of cultural homogeneity. Besides supporting Meyerson's (2003) argument that people take advantage of the language of the dominant culture, the statements show that people also act in accordance with dominant ideas within the institutional environment. This way of positioning dominance over female immigrants is clearly reminiscent of colonialism and its idea that failure to achieve a particular level of civilization must be corrected for development to take place.

Conclusion
What we have been concerned with in this article is how public employment officers discursively describe how they assist female immigrants in finding jobs. I have argued that the transfer to a neoliberal society has affected public employment officers' working conditions and that they have been equipped with tools aiming at facilitating a transformation of clients from passive welfare beneficiaries into active job seekers. The aims set out in this article were to show (a) how public employment officers make use of prevailing discourses to transform female immigrants from passive welfare beneficiaries into active and responsible job seekers, (b) how prevailing discourses contribute to the reproduction of culture and gender differences reminiscent of colonial discourses, and (c) how unequal power relations may favor the segregation of mainstream and alternative cultures.
The findings of the study indicate that employment officers are working under increased pressure owing to a growing number of clients and changes in working routines. Society's enhanced striving for efficiency probably puts pressure on employment officers, as they, in their role as civil servants, have to follow political directives as well as meet employers' requirements for a supply of appropriately competent workers. Furthermore, it is likely that the high unemployment rate among immigrants and the fact that employment officers have now been made responsible for making the immigrants employable have put even more pressure on them.
The findings further demonstrate that some employment officers blame immigrants' difficulty in finding jobs on their lack of education and cultural adjustment. This is manifested in the officers' discursive accounts and their performance as cultural role models, for example, when advising Muslim women to take off their headscarves in work-related situations and when implicitly telling a female immigrant to become independent by adjourning a meeting because the female is accompanied by her husband. Taking into account that one of the ideas behind the implementation of new public management was to make public organizations more efficient by measuring employees' performances, it may seem natural to the officers to put the blame on immigrants' lack of cultural adjustment, because although the placement of immigrants is beyond the officers' control, they may fear that the difficulty in integrating immigrants into the labor market will have a negative effect on the officers' performance measures. This may be one explanation of the fact that female immigrants are being presented as liabilities, who need to be corrected to fit a supposedly necessary cultural mold. This may explain why the higher unemployment rate among female immigrants is blamed on a lack of education and cultural adjustment.
However, employment officers' discursive accounts seem paradoxical, as, on one hand, they lay stress on individuals' need to take responsibility for finding a job, and, on the other, they perceive of themselves as responsible for presenting employable job seekers to the labor market. The policy of blame implicit in employment officers' emphasis on a lack of education and cultural adjustment seems in this respect to support Eskelinen, Olesen, and Caswell's (2010) suggestion that it is the employability of the clients rather than their situation that is the focus of employment officers' attention. It also seems to support Esses et al.'s (2006) finding that current activation policies seek to mold unemployed people to fit them into changed conditions on the labor market. This in turn indicates that employment officers act under the influence of the dominant discourses and favor the "work should be rewarded" strategy set out by governments to reward the working population and thereby tone down the "nanny state" image (Green-Pedersen, 2002). Considering, however, that the number of unemployed in Sweden at present amounts to approximately 380,000 people (8% of the able-bodied population) while the available vacancies are estimated at approximately 70,000 (SCB, 2014), this might be perceived as a rather demoralizing strategy by the unemployed, who, despite their desires and efforts, remain outside the labor market.
The findings further demonstrate how prejudices about female immigrants are being reproduced in employment officers' discursive accounts. It demonstrates how female immigrants are being discursively linked together as a homogeneous group of non-Western, perpetually pregnant housewives with little education. The findings further demonstrate that categorizing immigrants as "the others" and ascribing stereotypical qualities to them supports Meyer and Rowan's (1977) argument that people seek conformity and rely on myth and institutional values. The depreciatory views of female immigrants that some of the employment officers express indicate that these officials, besides being constrained in a professional role, are constrained by taken-for-granted assumptions about immigrants and their background. The findings demonstrate how stereotyped and prejudiced images reminiscent of colonial discourse are put into discursive practice through routines and tacit repetitions, thereby reinforcing the continuance of inequality and marginalization (P. Prasad, 1997;Pullen & Knights, 2007). They also reveal how people unreflectively lean on dominant discourses of differences characterized by ethnocentrism and cultural singularities (e.g., Hofstede, 2001), and how prejudices about female immigrants are normalized through the continuous reproduction of cultural hierarchy and patriarchy (Lugones, 2000;Said, 1978). Judging by the examples provided by the employment officers, it is evident that the immigrants find themselves in an inferior position that they have no or little power to change. Their future in the labor market is in that sense very much in the hands of the employment officers.
Furthermore, the continuous reproduction of dominant discourses means that competing discourses are excluded and that the concept of diversity is narrowed down to superficial, essentialistic, dichotomized, and hierarchically ordered categories of culture and gender. What is ignored in the dominant discourses is that culture, independent of time and space, may travel across borders and transform itself into hybrid forms; that people have different histories and experience culture differently; and that countries are therefore characterized by cultural fragmentation (Bhabha, 1994;DiMaggio, 1997) rather than by fixed and homogeneous entities. The desire to make unemployed immigrants adjust to the norms points at the fact that the integration of immigrants follows the process of hegemony, privileging self while neglecting values and behavior systems that deviate from the norm. In response to the question on the effects of unequal power relations, it could thus be argued that unequal power relations between employment officers and immigrants favor the segregation of mainstream and alternative cultures.
One conclusion that can be drawn from the findings of the study is that integration is hard to achieve as long as immigrants are objectified and treated as "the others," and no sincere effort is made to recognize them as individuals or to admit our own bias and the fact that we may be blinded by ethnocentrism and Western epistemology. From this perspective, discrimination awareness training among employment officers may be one way to confront the privileging of self at the expense of others and their value systems. Bearing in mind that research addressing discrimination against immigrants in the labor market to date has been fairly limited in Sweden, more research on this issue is needed. Because categorizations are always accompanied by human judgments, there is a risk that those considered too difficult to help will be unfairly classified in the coding system used at the employment agency's offices. This leads me to the proposition that more research should be carried out in connection with the relationship between the classification system and discrimination against immigrants.
However, as the blame placed on unemployed immigrants' lack of cultural adjustments may be a symptom of a dysfunctional system, this is probably not enough. What we propose is research focusing on structural changes and the transformation of work and its consequences for immigrants' employability. Studies of immigrants and their experiences could help address potential problems embedded in the current system. For example, instead of evaluating employment officers' performances based on measurements connected to job placement rates, evaluations could be carried out through a survey of immigrants enrolled at the employment agency. Also, considering that immigrants' voices are seldom heard, audit studies focusing on their experiences of the employment agency and their interactions with employment officers could contribute to a further deepening of insights into employment officers' role in the integration process. What we suggest is studies investigating immigrants' opinions of the employment agency and the help they receive from the employment officers. What difficulties do they come across in their encounter with the PESA? Would their chances of finding a job be improved if more time was spent on assisting them to find a job rather than on administration? This leads to the question of whether there are alternative and more efficient ways to organize the employment agency's offices and employment officers' work tasks. Another area of investigation should focus on the ways in which immigrants resist the employment agency. How do unemployed immigrants express their exclusion from the labor market? How do they resist employment officers' advice? Also, in consideration of technological advances such as computerization, robotization, and outsourcing, it could be worthwhile to investigate the transformation of work and its future consequences for those who are the most affected by unemployment, for example, immigrants and women. Will further technological advances make it easier or harder for immigrants to enter the labor market?
Finally, with new public management having been in practice for three decades, it has become more and more obvious that it has not been the success that politicians had hoped for. With regard to labor market policies and the integration of immigrants, the question is whether it is not time to reevaluate the efficiency of the market-based reforms. A recent survey (Holmberg & Sommerstein, 2013) of Swedes' confidence in public institutions showed, for instance, that the PESA ranked as the least successful performer. One of the factors contributing to this crisis of confidence may be the fact that the implementation of market-based reforms has left the employment officers with too much power to steer and control the unemployed and too little power for the latter to decide on their own lives. Perhaps this is an indicator that the time has come to take Denhardt and Denhardt's (2006) suggestion into consideration and implement a more serviceminded model based on negotiations, dialogue, and responsibility that is shared between the PESA and the unemployed job seekers.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) received no financial support for the research and/or authorship of this article.