Empowered spaces? Management articulations of gendered spaces in apparel factories in Karachi, Pakistan

Abstract Feminist scholarship has often focused on gendered workspaces within the apparel sector, where it is taken for granted that it is work conventionally attracting neophyte women. Within it, the task of managers is to discipline these young women to become docile and malleable workers. While this may have held to be the case temporally and regionally, South Asia’s experience has exhibited country-specific facets. This article focuses on these gendered workspaces in three factories in Karachi, Pakistan, in which we undertook research. In this context, there was a deliberate change in place facilitated by a United Nations Development Program’s Gender Promotion (GENPROM) initiative – to recruit and retain women workers, even though they acknowledged skilled workers were men. The factory managers we interviewed and spoke with used discursive tropes of gender equality and culturally appropriate women’s-only spaces as ways of justifying their labor recruitment strategy. However, digging deeper through interviews with managers at various levels suggested that their recruitment tactic had similar undertones to that revealed by early feminist research – although articulated via different mechanisms. We argue that this creation of empowerment spaces in particular Pakistani apparel sector factories requires careful tracing because it suggests how management interpellations reconfigure worker subjectivities. We also want to suggest that attentiveness to these practices is important because they may have specific bearings on temporal and spatial realities faced by Pakistan.

¿Espacios empoderados? Articulaciones de gestión de espacios generizados en fábricas de ropa en Karachi, Pakistán RESUMEN La investigación feminista a menudo se centra en los espacios de trabajo generizados dentro del sector de la confección de ropa, donde se asume que es un trabajo que atrae convencionalmente a mujeres novatas. Dentro de esto, la tarea de los gerentes es disciplinar a estas jóvenes mujeres para volverse trabajadoras dóciles y maleables. Aunque tal vez este haya sido el caso temporal y regionalmente, la experiencia de Asia del Sur ha exhibido facetas específicas de cada país. Este artículo se centra en estos espacios de trabajo generizados en tres fábricas en Karachi, Pakistán, donde llevamos a cabo la investigación. En este contexto, hubo un cambio deliberado facilitado por una iniciativa del Programa de Desarrollo de Naciones Unidas para la Promoción de la Equidad de Género (GENPROM, por sus siglas en We use this snippet from an opening group interview with management at a Karachi-based garment factory to capture the contradictions involved in claims made by Pakistani managers, in this context around creating women-friendly working spaces in their apparel factories. Yet, this conversation suggests that inglés) -para reclutar y retener mujeres trabajadoras, aunque reconozcan que los trabajadores capacitados eran hombres. Los gerentes de fábrica que entrevistamos hablaron con tropos discursivos de igualdad de género y espacios sólo para mujeres, culturalmente apropiados, como formas de justificar su estrategia de reclutamiento de trabajadorxs. Sin embargo, profundizando, en las entrevistas con los gerentes en varios niveles sugirieron que sus tácticas de reclutamiento tuvieron matices similares a aquellos revelados por las primeras investigaciones feministas -aunque articuladas a través de diferentes mecanismos. Sostenemos que esta creación de espacios de empoderamiento en fábricas específicas del sector de la confección de ropa en Pakistán requiere un cuidadoso análisis porque sugiere cómo las interpelaciones de los gerentes reconfiguran las subjetividades de lxs trabajadorxs. También queremos sugerir que la atención a estas prácticas es importante porque podría tener impactos específicos sobre realidades temporales y espaciales enfrentadas por Pakistán. as feminist scholars we need to pay attention to the incongruities in such dialogs to appreciate the complex registers invoked as women's employment is created. The management team we interviewed initially emphasized how their group of factories was a precursor in shifting the workforce demographic profile -moving from predominantly working men to working women. Our initial introduction to the factory setting made us cautiously optimistic of the promises and potential it held: at Factory A, 1 we were greeted by three managers, two men and a woman, at the entrance to the factory before we were whisked to a conference room on the upper floor and greeted to a panel of 11 management staff. While they were mostly men, the presence of five young women who made up the team was a welcome sight; we thought that it did not suggest mere tokenism. Yet as our days in the group went by, the gendered hierarchies -although subtle and inflected -which persist became apparent.

培力的空间？巴基斯坦卡拉奇的服饰工厂中的性别化空间
We were on a visit to these factories to assess the ways in which ethical codes were negotiated around public sector procurement practices, as the factories studied supply uniforms worn by public sector workers in Europe, and our expectation was to hear from the management team around this theme. Instead, they started offering us an overview of the company profile, its position as a supplier within a niche market before they went onto discuss a raft of measures they undertook to adhere to be compliant with various technical requirements. Because we were not necessarily hearing that which we thought we were going to be briefed upon, we made a decision to gently nudge them around various dimensions of the codes that we were aware of. It is within this context that one of the managers, although a man, intervened and started talking to us about their involvement in a Gender Promotion (GENPROM) program facilitated by the United Nations Development Program (UNDP), which aimed to recruit women workers into their factories. We sensed that the entire management team was proud of their involvement in this program and packaged these efforts as an attempt to 'empower' women entering the formal labor market. They also linked this to the ethical code spectrum by saying it offers us evidence of how they were not discriminating against women.
Our curiosity was stirred. We sought to hear more about this program, from everyone present -and because we sensed that there was an implied subtext, we recognized the need to explore its practices and effects. Our findings were a byproduct of our larger study. However, because great efforts were undertaken by the management team to make claims about creating a safe working space for women, the need for further scrutiny was evident. We, thus, want to explore what managers mean when claims around creating secure spaces for women workers occur in an industry generally dominated by working men. What do these efforts signify and how does it speak to feminist concerns?
Feminist scholarship has often focused on gendered workspaces within the apparel sector, where it is taken for granted that it is work conventionally drawing in neophyte women (Elias 2005;Ong 1987). Against this backdrop, the task of managers is to discipline mostly young women to become a docile, malleable, and even disposable workforce (Lynch 2007;Wright 2006). While this may have held to be the case temporally (1980s and 1990s) and regionally (East Asia and Mexico, for instance), South Asia's experience has exhibited country-specific variations. In Sri Lanka, women have historically made up the bulk of apparel sector workers (Hewamanne 2008;Lynch 2007), while contemporary findings signal how this is in flux for apparel producers (Goger 2013;Ruwanpura 2014). Within India, whether women or men make up the bulk of the workforce varies according to region (De Neve 2005; Mezzadri 2014a), suggesting that cultural factors and local institutions may engender or inhibit the type and kind of women's participation in the formal or informal spheres of the apparel sector (carswell and De Neve 2013). In the two South Asian countries dominated by Muslim communities -Pakistan and Bangladeshthere is variation, yet again; in Bangladesh it is predominantly women (Kabeer 2004;Siddiqi 2009), while for Pakistan it is predominantly men who dominate employment in the manufacturing sector (Lund-Thomsen 2013). Around 70% of garment factory workers in Pakistan were men prior to the GENPROM initiative (GENPROM Pakistan 2006).
The UNDP's GENPROM initiative in Pakistan began in August 2006 to: facilitate employment of women in the garment industry, primarily through training and skills upgrading; encourage the garment sector to promote women's participation; record and evaluate progress in terms of women's participation in the garment sector; and to share learning between the UNDP and participating companies with regard to skills training initiatives (GENPROM Pakistan 2009). The objectives are strongly linked to the UN's Millennium Development Goals and the overarching objective of poverty reduction through private partnerships (interview with UNDP officers). Since 2006 Technopak Advisors, the Indian arm of Kurt Salmon Associates, has been the consultancy firm involved in the development of the training programs central to the initiative. This has mainly revolved around the training of (predominantly women) machine operators, but also has extended to the areas of quality assurance and compliance. At its inception, the aim was for at least 80% of trainee machine operators in the program to be women. Satellite Training Units (STUs) became the spaces where training was conducted, normally housing some 20 training machines, and these could be located either within large factories (as in the case of the factory group we studied) or as independent regional units to be used by smaller garment factories. A Program coordinator from UNDP establishes the STU 'to transfer the [GENPROM] philosophy to the unit' (interview with front end manager) and to monitor training, while Master Trainers (8-10 per unit) deliver the quality training to machine operators (GENPROM Pakistan 2009). By the end of 2010, a total of 47 STUs had been established, involving 17 Program coordinators, 498 Master Trainers and 9666 trainee operators (GENPROM Pakistan 2011).
Local architectures around labor recruitment and training strategies are worthwhile tracing for multiple reasons. It not only offers a glimpse into how global supply chains intersect with local realities, but as Mezzadri (2014b) remarks it also facilitates acknowledging how labor outcomes are principally about capital-labor relations. This focus, she points out, is vital if we are to appreciate how 'global and local capitalists share common agendas, primarily centered on competitiveness concerns' (Mezzadri 2014b, 340). Feminist scholars in particular have suggested that labor process analysis is further strengthened by examinations concerning itself with how women workers' locations shape their experiences, identity, and subjectivity as workers (Pearson, Anitha, and McDowell 2010). Subjectivity formation, however, is not only centered upon what workers bring into working lives, but also how managers inform and shape workers having particular traits over others, thus feeding into their profiling. How management discourses and practices structure and reproduce myths around women workers, whether as disposable, passive, nimble-fingered, or supplementary earners, for instance, all normalize subject formation (Goger 2013;Werner 2012;Wright 2006). While Goger (2013) rightly emphasizes that this particular subject formation may no longer hold across the global supply chain and is temporal, we also want to suggest that the disruption to the conventional feminist script need not follow a linear pattern. As the entry point to this article suggests, tropes of 'empowerment' deployed tended to be less about women's enfranchisement as workers; instead, it was about giving purchase to assumed benefits that emanate from access to economic resources via work (Rao 2014a). Yet within the multiple threads surfacing from our interviews with managers, the contradictory ways in which work opportunities availed themselves to women also signal latent possibilities arising from paid work. Hence, we do not want to deny that some women may find the space to exert their agency within the factories or their families; still as Rao (2014a) suggests 'experience of work is complex, so is agency: neither can be unraveled without taking into account other elements of women's or men's identities ' (2014a, 80). Reinforcing gendered identities via factory spaces is what our article hopes to disentangle so as to appreciate how practices associated with social and ideological spaces may be quite distinct from the empowerment discourses (see also Rao 2014b).
In the three factories in which we undertook research during December 2012, there was deliberate change facilitated since 2007 through the UNDP's GENPROM initiative; namely, to recruit and retain women workers, even though there was tacit acknowledgment that the skilled workers were men. Significantly, this group of factories was among the first to participate in the program, and the current Production Manager was originally employed by the UNDP as one of the Program coordinators before moving to the group and transferring GENPROM approaches to corporate strategy and the factory floor. All three factories have continued with the training program since the set up of STUs in-house, and have not only learnt from GENPROM, but are actively informing GENPROM evaluations and best practice. 2 The factory managers we interviewed invoked discursive tropes of gender equality and culturally appropriate women's-only spaces, such as separate canteens, as ways of justifying their labor recruitment strategy. However, digging deeper through interviews with managers at various levels suggested that their recruitment tactic had similar undertones to that revealed by early feminist research -although articulated differently, namely the GENPROM program and voluntary corporate codes for labor, which are critically associated with neoliberalization. We want to argue that this creation of 'empowered' spaces in particular Pakistani apparel sector factories requires careful tracing, because it suggests management interpellations are a powerful tool for reconfiguring worker subjectivities, which have a bearing on how labor value is created. It offers an important backdrop for appreciating possible future research on how it may have a bearing on embodied practices of a new cohort of working women.
Having synthesized feminist contributions above, we go on to offer an overview of the fieldwork methodologies used in section two. In section three, we discuss our main findings through a feminist analytical lens. Our purpose is to tease out the ways in which management efforts at creating safe spaces for women workers are also about crafting particular worker subjectivities (Goger 2013; see also Werner 2012). 3 Differently from Goger (2013) however we want to suggest that in the case of our field sites, the transition is not from disposability to empowered -but rather how women's very entrance into the formal labor market is articulated as a way of empowering them so that they can serve their home and nation. Yet, underpinning these laudable efforts is the subscript, which is about using a more malleable labor force to serve the interest of capital (and the nation) much more effectively. What we also find is the ways in which corporate social responsibility commitments of buyer(s) combine with shifting global and national discourses around gendered participation. And it is through a combination of these that worker subjectivities are constructed and articulated. It is this theme that we hope to explore in our article because what we found is the ways in which corporate social responsibility commitments of buyer(s) combine with global and national discourses around gendered participation to construct and articulate particular labor identities (see also Siddiqi 2009). 4 To discuss our findings, our fieldwork in Karachi, Pakistan is outlined next.

Accessing the field via favored connections in Pakistan?
Our overall research purpose in Karachi was to undertake a preliminary assessment of the degree to which ethical codes frame public procurement practices. This interest stems from the morphing research agenda of ours on debates around ethical trade and global production processes merging together with a buyer's sourcing strategies and corporate social responsibility commitments within the sphere of producing for the corporate wear sector (as opposed to fashion wear). So while our primary purpose was far from interrogating feminist research themes, the gendering of workspaces became an increasingly prominent analytical focus.
Our scoping study was conducted during the initial week of December 2012 and it was an intensive, although short, period of fieldwork. We conducted our field research in collaboration with Dimensions, the UK's leading work-wear supplier, as they were interested in our evaluation of the effectiveness of ethical trading codes applied to one of their first-tier manufacturing suppliers of uniforms. While Dimensions paid for our travel expenses, we obtained no further remuneration to conduct our scoping study. In the UK, academics are actively encouraged to engage in knowledge transfer efforts with relevant partners, where their academic research is to have an impact on various constituents of the public. Within this backdrop, we saw the invitation from Dimensions to disseminate our decade-old research on ethical trade and labor practices, including in South Asia, as evidence of the value that academic research can have on industrial relations. Since British academia supports collaborative initiatives, we went through all relevant University procedures -from each respective institution -to travel and do a short spell of field research in Karachi, Pakistan.
The corporate owner of the group of factories in our case study has a total sales turnover of USD 350 million, with seven factories making up its gloves division and three factories in its clothing division. The three clothing production facilities in Karachi produce a wide range of work-wear for European clients, including uniforms worn in the UK's National Health Service. The three clothing factories in Karachi were based within its export Processing Zone (Factory A) as well as outside the zone (Factories B & c). 5 The sizes of the factories varied, with the largest of them employing 1500 workers, the mid-size factory having a workforce of 600 and the smallest of the three engaging about 350 workers. The varied workforce and plant sizes offered us a sense of the diverse built landscape housing apparel production facilities in Karachi as well as the different labor-management relations in each plant. It helped offer texture to an otherwise brief fieldwork visit.
We used factory observations as well as group interviews and individual interviews with management, and almost all of these interviews were conducted in English. In a couple of instances where interviews were conducted in Urdu, a translator aided us. However, we are cognizant that on these few occasions nuances to language use may have been missed, which a native speaker is likely to have picked up. The fact that one of us is a bilingual South Asian aided a sense of camaraderie and familiarity that sometimes managers extended to assuming that the first author was familiar with Urdu. We had to mention that this was not the case. Overall because most interviews were done in English with managers, we feel confident of our ability to interpret and interrogate managerial and transnational frames of reference -which has been an important strand of research in feminist studies evaluating labor-capital relations (Goger 2013;Werner 2012;Wright 2006).
We interviewed 15 managers at Factories A & B, which were clothing production facilities, and observed the workings of the shop floor -with and without local managers supervising our presence. At Factory c, the glove-producing factory, we only observed the large-scale production processes. These interviews with management staff at the factories were supplemented with an interview at the offices of the UNDP in Karachi. While we were partly encouraged to conduct the interview with the UNDP by senior managers at the factory, we found this interview equally revealing of the ways in which global, national, and corporate motives coagulate around the language of empowerment of women; and each institutional actor was not necessarily at odds with each other.

Safe spaces and empowered women workers?
The creation of employment opportunities for women workers in these factories, shaped by the UNDP's GENPROM, was packaged and presented to us as an unusual initiative. We hence wanted to learn about the context within which recruiting women was undertaken. How was a rare occurrence normalized and made acceptable? We were told at a group interview with managers that the following were key mechanisms set in place to undertake these changes: F1R1: 'Because of the work environment, we provide full security, full transport -very near from their home. ' F1R2: 'Full lunch' F1R3: 'We trained them; and skilled workers in skill work. ' F1R4: 'They were unskilled at one time, most of them, we have got them through certain trainings; so may be, they would like to be inside our factories and they feel more happy with us. We train them so as to improve their grade levels' F1R3: 'Actually, the poverty rate in Pakistan is getting higher, so in our society right now, it is preferred that not just one family member and it should be that more than one or two family members should work. So that is why the women right now are coming out of their homes and starting to work, and in this environment that actually we trained them. So it is the environment in our organization in which we are retraining or training the workers. ' Our respondents began by answering queries about how the workspaces are regularized for women workers in a society perceived to be conservative, by speaking about the efforts that they were putting in place to assure women's physical security -on-and off-site. Providing transportation between the workers' homes and the factory premises was impressed upon as a necessity for securing their mobility since there was the risk that the public environment may be hostile or risky for working women. 6 Providing amenities that assured physical security for women workers was supplemented with initial skilled training by the factory. They also mentioned how consequently it was not simply one or two women from particular neighborhoods who were recruited to their factories, but rather multiple family residents and extended kin. This had the effect of familiarizing the work setting to women, mostly starting jobs in the formal sector for the first time; employer efforts at normalizing the workspaces by the recruitment of women through kinship and friendship networks also had the potential to blur the separation between work and neighborhood spaces. While Rao (2014b) captures the unsettling of existing status hierarchies through mobility, it is also likely to be the case that 'naturalizing' new workspaces halts insecurities stemming from mobility and entering formal employment. Managers claimed that it was these facets that attracted women to seek employment at their factory, although it still did not answer our questions around why women opt to go into work in the first place -given that public mobility is frowned upon (charania 2014; Toor 2014). Perhaps because our facial expressions gave away our dissatisfaction, suddenly a young woman manager broke into the conversation and offered us a sense of the milieu that enabled factories to engage in worker recruitment. She talked about the increasing poverty rates that afflict Pakistani society, breaking down the traditional male breadwinner model and compelling other family members to find, supplement, or substitute income. In other words, despite the patriarchal contract, economic circumstances may propel a collapse of its idealized form and force women to become income earners -even among Muslim women (see also Rao 2014aRao , 2014bRuwanpura 2004Ruwanpura , 2006Ruwanpura and Humphries 2004;Siddiqi 2009). The implication of her intervention was that their group of factories was capitalizing on a dramatically changing economic situation, which was making women enter the formal labor market in Karachi.
Scholars working on Pakistan have recurrently noted how women's economic participation is low and their labor force participation rate hovers well below 20% (Rehman and Azam Roomi 2012), despite women counting for 50% of the total population. However, their involvement in the labor market has been slowly growing, shifting upwards from 11.4% in 199411.4% in -199511.4% in to 16% in 200411.4% in -2005 Azam Roomi 2012). Ali (2012) links women's increasing participation in the formal economy to rising poverty in Pakistan post-1990, which he associates with the impact of structural adjustment programs. Their visible presence in the urban spatiality brings with it a plethora of challenges to an emerging set of women workers, navigating the public sphere on their own -often for the first time. Scholars' record how Pakistani women invoke fear in their daily lives, and they proceed to remark how 'the insecurity and worries … comes from threats or violence that they experience in domestic and public spheres' (Ali 2012, 593). Women workers entering garment factories then navigate the stresses that come from material deprivation and economic necessity together with fears around gendered violence they have to encounter -at home and in public settings.
As a first generation of women workers in the formal sector, the factory space then needed to be attentive to a number of potential challenges around recruitment and retention. Similar to rural South India and Bangladesh (carswell 2013; Rao 2014a, 2014b), economic forces -namely deprivation, income inequality and poverty -were powerful factors pushing women into work in Karachi too. During our interviews, this wider context was frequently invoked as to why working women was a boon at multiple scales. The central import of working women and their benefits to the national economy was articulated during our interview with the officers at the UNDP's GENPROM program too. The GENPROM initiative was formulated by the UNDP with the objective of reducing poverty via private partnerships and offering skill development to women. The apparel sector was one of the target sectors, given the conventional dominance of working men. Taking a cue from Western (liberal) feminism, the UNDP officer mentioned 'In Pakistani environment, where male domination is very obvious … we were providing training to women who were dependent on their brother/father for everything. Now they are actually contributing to the family. ' Not only the officers at the UNDP but also our management interviewees mentioned the important contribution women could economically make to the family. This, they contended, offers women the prospect of gaining confidence and a sense of social standing within their communities: UNDP-R1: 'When more women … want to improve their social standing in Karachi, they will work …' UNDP-R2: ' … exactly, but women … they are more confident when working, and you know in this working context … we are expecting to get at least, to get extra benefits. ' Our respondents, including and especially managers, were then keen to emphasize how gaining confidence within neighborhoods and communities was an essential precursor for ensuring that women felt empowered by and through work. Feminist scholars working on South Asia have frequently probed whether economic resources alone have the capacity to shape women's agency (carswell 2013;Rao 2014a;Siddiqi 2009). Moreover, Ali (2012), by specifically focusing on women garment sector workers'as a subset of the working class neighborhood he studied -notes the challenges they face both on-and off-site. On-site women workers concerns ranged from the prevalence of gendered wage discrimination to the lack of separate toilet facilities and flexible work hours. Outside factories, women workers have to negotiate everyday violence both within their households and in traversing between spaces. In the rendition offered to us by the UNDP officers, there was little reflection on whether the domestic stresses and gendered tensions that Ali (2012) outlines evaporate as easily simply through women earning an income via the formal sector. Expressions of doubt or uncertainty around women's empowerment through paid employment hence were non-existent among our respondents. They all took great pains to emphasize the ways in which working women in the apparel sector were likely to be 'empowered' women and work was the means through which women's sense of self-worth was engendered. In contrast to 'disposable' women of Mexico (Wright 2006), or disposable to empowered women in Sri Lankan apparels (Goger 2013), Pakistani women workers in the industry were constructed as empowered from the outset. A senior manager, in fact, used the illustration of Sri Lankan recruitment of women workers in the apparel sector as an ideal worthy of emulation. 7 This particular representation of empowered women had much to do with the ways in which working women contributed to the nation. From the scale of the family and community, crucially both UNDP officers and factory managers also proceeded to refer to the vital contribution that working women make to the national economy -and hence the nation (see also Siddiqi 2009). They talked about how women's involvement as workers contributed to the reduction of poverty levels within their families, which in turn they contended was likely to help reduce overall poverty rates in the country. Hence, women's entrance to the labor force was seen as a vital step in assisting the nation's development trajectory, where women workers' narratives were interlaced with that of the nation's well-being (Bergeron 2004).
In contemporary Pakistan, the need to be attentive to the ways in which these tropes are interwoven is crucial. While Islamist feminists are recurrently reinventing what it means to be modern and yet religious, so as to facilitate low-income women's access to the public sphere (Jamal 2009; see also Rao 2014b), it is not without detrimental consequences for progressive feminist aspirations (Zia 2009). The ways in which these multiple scales are fused together then necessitates unfolding the processes at play. One manager explicitly fused these diverse scales when talking about the benefits women workers bring to their immediate households, as he also casually connected it to the nation's development and welfare: ' … on promoting (women), to train them to relative skills so she can contribute to their families, for economic growth and for quality of life. ' Similar sentiments were already shared with us during the initial group interview with managers; the intricate connection between women workers and the nation's welfare was presented as a foregone conclusion. The challenges that women workers may be facing within factories or families were hardly points of reflection. curiously, however, the discursive trope of managers and workers alike working together as a family or being part of the same family was not invoked at any time during our formal interviews or short phases of observation. Our finding thus appears to contrast with other feminist inquiries showing the multiple ways in which language around 'family' and kin idioms are deployed to produce a malleable workforce (De Neve 2001;Hewamanne 2008;Lynch 2007;Wright 2006).
The justification for getting women to work in the apparel sector then extended from the family and community to the nation in a seamless connection of multiple scales. This interlaced representation likely heeds multiple liberal feminist calls around employment as an important facet of strengthening the position of women (Siddiqi 2009). We, however, want to probe the extent to which this rendering held when workplace politics and social relations were explored. Would individual interviews with managers bear out similar views? While fissures became apparent during the group interview with UNDP personnel, our initial group interview with the managers offered a tightly woven script. They continuously mentioned how the factories belonging to the group offered safe work spaces for women and took all necessary steps to make the space conducive to their safety and public mobility, with their concerns extending from within to the outside. creating empowered women workers by being sensitive to the local spatial context was not simply their aim, but was seen to create a central platform from which women's sense of self-worth could be engendered. As we trace below, workplace hierarchies prevailed over liberal feminist claims and empowerment assertions made by managers. Our next section, therefore, traces these disjunctures to appreciate how previous critical feminist arguments continue to resonate with contemporary times -even as empowered women are constructed. We want to suggest through this analysis that our managerial respondents' articulations are not mere rhetorical maneuvers but actually speak to the particularities of gender and social politics in Karachi, Pakistan.

Localized articulations of gendered workplace politics
We had the opportunity to probe our curiosities about the enthusiasm for hiring, training, and retaining women workers during our interviews with individual managers. Our prompts stemmed partly from the implicit assumptions of the respondents at the UNDP, but also because a casual remark or another during one-on-one conversations disclosed that there may be multiple rationales that needed investigation. These exchanges alluded to the ways in which employing women workers is beneficial for management; we will explore the degree to which these facets fed into changing the labor demographic profile of the factories. By attending to these threads, our purpose is to unpack discourses around empowering women through labor.
Despite the UNDP's enthusiasm for promoting women's entrance into the formal labor market, it transpired that their mid-term review revealed: UNDP R-1: ' … in our society it is very easy to manage, control and train women, because men in general seem to be more hostile, more aggressive; whereas women, they are -I won't say docile -but easily manageable … Now when they have the opportunity of earning something they feel very comfortable actually -they develop this organized component of their behavior that they want to be part of it. They want a premium job and that in itself leads to their disciplined approach towards their work. ' Within an internal review conducted by the UNDP, potential issues surrounding the hiring practices of women revealed that advantages around their disciplined approach toward work came bundled together with their 'manageability' . The suggestion was that women's own enthusiasm and embracing of their laboring lives resulted in attitudinal shifts conducive to workplace politics -the disciplining came from within women and not outside. These practices are not unusual for South Asia, as feminists have already shown that 'specific tropes … tended to be invoked repeatedly in differing locales' , where the 'effect was greater social and self-regulation of workers' lives, inside and outside the shop floor' (Siddiqi 2009, 157). The 'manageability' of women workers was complemented by talk of their allegedly apolitical nature: 'women usually do not have political leanings or leanings to fundamentalism or any violence in them, ' and in the febrile political context in Pakistan this was viewed as a boon. 8 Yet, they also noted how women were more likely to squabble with each other on a weekly basis or result in higher labor turnover. However, the overwhelming perspective was that ' … women are better workers, period. ' The extent to which women workers were viewed as 'better workers' , however, depended on the attributes emphasized by various managers. Women workers lacked tailoring talents according to many managers, but because of their disciplined approach toward laboring life were nonetheless constituted as 'better' workers. Several managers echoed the sentiments of the UNDP colleagues when they recurrently mentioned that by schooling budding women workers they were also creating a favorable labor force: … training is going to benefit … But it is a long route, but a sustainable route; plus you can make your people disciplined as to the way you want them to work. And they are better to organize. When you train someone his or her behavior is totally changed … (said one senior manager) Having an organized workforce, where behavior is altered and the workforce is disciplined seemed to be his underlying concern in validating the extensive recruitment and training program for women. Another said: … because when women are between men it is a very good environment. Otherwise may be some men talk harshly or abusive language is used by men. If there are five women between fifty men, all are very … err … disciplined and cultured.
The ideal of an appropriate workplace culture incorporating gendered spaces in this way was therefore very clearly mobilized by managers to justify the recruitment of women workers. We observed that Factory A operating in a modern, horizontal building included a mix of men and women machinists on the same factory floor, while Factory B tended to keep men and women machinists in separate spaces and sometimes on separate floors within an older building and vertical factory format. However, both men and women are present in both factories. While the effects of either separating or mixing men and women workers were not the focus of the study, the claims made by the managers about the disciplining potential of women's presence are nonetheless indicative of the discourses at work regarding women workers as docile and beneficial for manageable industrial relations.
The management rationales offered echo Siddiqi's (2009) observations for Bangladesh around the greater surveillance and regulation that comes with shop floor work. Women workers in our Karachi case studies were not simply useful for their own self-regulation, but also for how they regulated the loutish behavior of working men. Moreover, however, by instilling discipline and a culture devoid of abusive or harsh talk, the manager also implicitly signaled the likelihood that women were unlikely to retort or challenge supervisory or mid-level supervisory orders.
Yet another manager mentioned how recruiting women was an effective management strategy because 'Women, they are good workers; obedient and cooperative -and those who are supporting … the family will normally have a long association with the company. ' In this instance, it was not simply that women workers were more likely to be obedient and cooperative but that there was a greater likelihood of them remaining with the factory for longer. He opined that if young women were recruited at 16 years of age, then they were more likely to work at the factory for 7-8 years before marriage -which he said in Karachi was usually 23-24 years. Altering the workforce profile at this group of factories then extended beyond the purported need to 'empower' women.
This partiality toward recruiting and training women workers is all the more remarkable because with regards to skill and ability, managers across the board seem to intimate that the skilled workers were men given their tailoring skills or previous work experiences. One of these respondents said 'But errr … as … may be … men can produce some more goods as compared to women with their physical initiative' , suggesting that working men have higher productivity levels. He went on to say, 'It's the general perception, some of the operations or some of the machines, we may perceive that a man is better at this position' , suggesting that there are some operations and machines where having physically able men is appropriate. A front-line manager shared a similar sentiment when he said that women were easier to 'handle' , but as far as output and productivity go men workers surpassed their working women peers. These views partly emanated from the setting in which the formal interview was carried out. When doing the interview, we were disturbed because a floor-level manager brought a faulty piece of garment to the attention of the interviewee, which disrupted the dialog in providential ways. It was evident that the faulty garment piece was not a one-off; and after some discussion between a number of front-line managers and supervisors the necessity to redo the garment was clear. Because the first author witnessed this entire episode, the formal interview got sidetracked -although fortuitously for the research. Hence we had a conversation about how productivity levels are affected because of such faults, with him attributing it to the company policy of overlooking recruiting skilled laboring men and hiring women workers instead. From the perspective of production, production targets and attaining better productivity levels, there seemed little doubt in his mind that men workers were more suitable for work wear production. Yet, women were the more malleable workforce and since this was the company view and policy, production managers had to yield. Then, as if to justify this less than ideal hiring policy from a production perspective, he went on to say that men were more likely to find other jobs and so were more likely to leave work. He said that since working men did not face the same constraints that women confronted, who needed assurances around the safety of the workplace environment, the turnover rates among laboring men tended to be higher. characterizing women as 'better' workers then hinged on representations that brought with it various tensions, some grounded in practicalities around production processes and pressures, others around having an obedient and disciplined workforce, while others still over controlling labor turnover. Each of these pressure points divulged how gender stereotyping also shaped the drive toward recruiting women workers. crucially, our exchange revealed that the discursive terrain around 'empowering' women was constantly connected to their position within the family and not as empowered women workers. This discrepancy is important to tease out because it signals the contentious, incongruous and localized ways in which empowerment discourses are mobilized (see also Rao 2014a; Zia 2009). The penultimate section of our article traces how these tropes were mobilized and how they too were categorized as gendered attributes rather than empowering labor rights for women workers.

'Empowered': Women or women workers?
During our time at these three factories, our interviewees recurrently mobilized notions of 'empowerment' where women were associated with gendered roles as mothers, wives and/or dutiful daughters doing their bit for the family, community and nation. Management did not talk about women as an empowered labor force or even the need for empowered women workers. 9 Indeed their obedience, pliability, and disciplined attitude become the attributes accentuated and appreciated on the part of women workers. There was little reflection on how these women's 'qualities' may stem from social tension and everyday violence that shape domestic social relations, which are hardly empowering for women's identity and subjectivity (see also Ali 2012). This lapse was particularly telling given decades of feminist interventions around workplace politics pointing to the problematic uses and representation of women workers (Goger 2013;Mezzadri 2014a;McDowell, Anitha, and Pearson 2012;Pearson, Anitha, and McDowell 2010;Siddiqi 2009;Wright 2006). Moreover, since most managers connected their efforts of hiring and training women workers to 'empowering' women in their stereotypical roles, our case study illustrates limits to potentially promising initiatives that make claims around gender equitable initiatives. Yet the fact that managers saw their efforts as laudable and genuine deserves reflection too. Since they speak to a particular historical location in Karachi, Pakistan, from which empowerment discourses are shepherded, their exertions suggest the inherent danger in assuming that development buzzwords herald progressive shifts universally (charania 2014; Zia 2009).
There was an overwhelming consensus that women were ever more entering the labor force given escalating inflation and mounting poverty levels, pushing them to secure regular income (see also Ali 2012;Rehman and Azam Roomi 2012). Their economic needs were constantly emphasized: … in fact, the inflation rate has been really bad in the last 2-3 years. So it is very difficult for a single person to earn and feed 4-5 people in the home. So nowadays 2 or 3, or 4 or 5, people work to earn for their homes.
The pecuniary motivations, he said, moved women to seek work and these needs dwarfed any cultural (read: patriarchal) constraints. The economic pressures they were facing meant that there was a steady group of women workers pursuing jobs to supplement family income. Another manager evoked the importance of the patriarchal family and its sustenance through their job creation efforts for women. From his perspective, it was not simply that women workers were able to contribute toward their family, but also by extension the group of factories were helping to sustain the patriarchal family. … We are promoting women to support the family … as far as I understand we have some national issues. We have 53% women of our total population. And we have some societal challenges and in the recent past most women are not part of the labor market, but now there is a need to contribute from their side and as far as social values are concerned if a woman is working … doing work, they also support their head of family. Indirectly if we support women workers, then we are supporting the head of the family, who is normally male.
In their opinion, it is through contributing to family income, domestic needs, and even jehez 10 (dowry) those women workers, old and young, were empowered. The group of factories' role in bolstering the capabilities of their women workforce remained resolutely gendered and relegated to the familial sphere. Many managers also declared how women earning much needed income for the household tip the balance in their favor, facilitating their empowerment. In the words of a middle manager, he conceptualized empowerment as: Empowerment: I understand that in our society in Pakistan women are … we see women as deprived members of society … they are dependent on men … always do what their husbands say. But after the empowerment of women, right, they have the money with their own earnings; so our society … now I feel there is a balance coming … It is the capacity to earn a secure wage packet and contribute to household expenses that was seen to empower women; and all our interviewees repeatedly shared this theme. Rarely, if ever, was there reflection on what it meant for women to be empowered as women workers.
The exceptions were a couple of women managers at varying management levels. One was a woman 'Master Trainer' who articulated her experience in entering the workforce as an operator and then moving to becoming a low-level/front-line management staff member. She said that for the workforce empowerment means 'awareness of why the job is important for women, and … if they (are) empowered how this is good for the workforce' . She then proceeded to make linkages to her personal experiences, noting how working and earning a stable income has enabled her to pursue her part-time graduate studies and persevere as an upwardly mobile worker. While her response was rare, she also serves as a caution to the conventional feminist critique; as she suggests management deployment of gendered tropes does not necessarily mean that it is how women workers may appropriate and inhabit their new identities (see also Rao 2014b). Their earning power may well shape and have a bearing on their social relations that matter not only within the household, but also in the wider social and laboring realms too (see also Kabeer 2002).
Placing emphasis on the need to register difference was also important for Pakistani women workers. A middle-level manager noted how having a workspace of predominantly women workers afforded them a sense of comfort that mixed-working space does not provide. She mentioned how in the factory (that has) more women workers then the women are more relaxed. They … sometimes … do not want to sit beside any men or men workers, … especially when the women are new in such a work place, they are not used to doing a job … when they move from the STU they preferred to sit in between women and not in between men. It is something like that.
In her vocalization, the newness of women workers into the factory sphere required managers to be attentive and manage women worker expectations sensitively. Even here, however, the emphasis was not on empowering women workers, but rather accommodating their expectations given sociocultural norms. Her reactions echo Wright's (2006) observation that 'economic and cultural processes work through each other continually such that cultural entities … constitute the discursive stuff of its materialist core ' (2006, 49). Tellingly, this middle manager did not invoke the discourse of empowerment during our dialogic interview; instead she accentuated difficulties associated with earning a livelihood within a context of spiraling inflation and how the GENPROM program offered the group of factories to offer up-skilling opportunities to women workers entering the formal labor market. Always, however, with necessary deliberation to cultural contours that shape particular aspects of women's everyday lives, outside of workspaces.
Overall, then, the focus of empowering women workers was suggestive of strengthening bargaining positions of women within their households, families, and communities. Because all these facets were seen as improving the overall quality of life, how empowered women contributed to the nation's development and welfare was emphasized. Yet there was no reference made to empowered women workers or steps being taken to empower them within the workplace -the only occasion was when a woman Master Trainer talked about the confidence she has gained within the workplace and how more awareness of rights and empowerment opportunities on the part of women workers is likely to benefit the entire workforce. She was speaking to and through her experiences. These were not points of evocation or reflection for managers, who instead were involved in gender stereotyping as rationales for recruiting and training women workers, and how this became the underlying motivation to change the demographic profile of the workforce.

Conclusions
In our article, we have traced how discourses around 'empowerment' are deployed and their resultant practices as apparel sector managers are at early stages of recruiting and training women to become industrial workers. In the context of Pakistan where men have been at the forefront of labor in the manufacturing sector, opening up jobs for women of all ages seems a laudable aspiration since public mobility is frowned upon. However, when these discursive tropes are interrogated further what becomes apparent is that the primary preoccupation of trying to empower women is within the realm of their stereotypical gendered roles as mothers, wives, and potential brides rather than as women workers. Women workers' entitlement to labor rights does not even get lip-service, since they are treated first and foremost as women with ascribed gender roles linked to familial or communal relations rather than as workers. When this scripting is woven together with recurrent invocation of the need for cultural sensitivity by management of UNDP and factories alike, what emerges is legitimation of material practices that reify gendered stereotypes beneficial for pliable industrial relations. While manifestations of 'cultural sensitivity' is likely to vary and be shaped by contingent forces specific to local spaces, its recurrent invocation by development agencies and factory managers needs critical interrogation if a feminist agenda is to be abided by. Empowerment discourses mobilized alongside 'cultural sensitivity' by management and international development organizations alike are hence more about serving the interests of industrial capital rather than a new class of industrial workers. Women workers were not only depicted as easier to manage, but their presence was also viewed as a powerful disciplinary force restraining men's unruly or disruptive behavior. The advantage of having women workers within the shop floor was then at least twofold, if not manifold. Where women's entrance into the formal labor market is not matched by an acknowledgment of the need to strengthen substantive labor rights and access to a living wage, discourses around empowerment are at best contestable. In Karachi's urban sociality, Ali (2012) already notes that women workers' emergence has thwarted meaningful assertions of their agency and empowerment; this implies that invocations of 'cultural sensitivity' signal partiality toward to the existing patriarchal and capital-centric status quo (see also Werner 2012).
Our research in Pakistan shows that the UNDP's GENPROM initiative, in which the factory group we studied participated, has played a role in up-skilling women working in the garment sector against the backdrop of a national apparel industry dominated by male employees. The training programs targeting women workers as part of this initiative appear at first glance to be suggestive of empowerment. This initiative, shaped by the neoliberal strategies of the UNDP to support poverty reduction through partnerships with industry, has facilitated a significant increase in both the employment and the up-skilling of women garment factory workers. This change has developed alongside the proliferation and application of voluntary corporate codes of conduct to ensure minimum labor standards in the workplace, viewed also as a part of neoliberalization involving governance of global supply chains through private regulation. However, management discourses deploying gendered tropes in the factories we visited suggested that improvements in terms of growing numbers of women garment sector workers and their corresponding up-skilling need to be tempered by discourses mobilizing notions of women workers akin to the findings of early feminist labor research regarding the disciplining of malleable women workers. However, we argue that the changing gendered workspaces of Pakistan are neither a replica of that early discourse of managing docile women workers, nor an articulation of gendered workplace empowerment via new global development initiatives and private regulation. Rather, we argue that the ways in which the subjectivities of women workers are produced in the Karachi-based factories result from an intertwining of globalizing development discourses and initiatives around empowerment, national-institutional and cultural contexts of gendered employment expectations and localized articulations of gendered workplace politics on the factory floor. The findings of our study suggest that rather than privileging any one of these scales of analysis, changing gendered workspaces should be understood through an approach recognizing their articulation through trans-scalar spaces of discourse and practice. Notes 1. We anonymize the factories researched in order to protect their identities in accordance with interviewees' wishes.
For the same reasons, we do not provide information on the specific job titles of the interviewees quoted in the article. 2. This pioneering role of the factory group in GENPROM and the group's subsequent contribution to best practice and learning within the initiative was proudly reported to us by the Production Manager, who was previously a UNDP GENPROM Program coordinator, and a senior woman manager. 3. As Goger (2013) shows, it also has ramifications for how labor value is created. We, however, are unable to pursue this line of inquiry in this article, given that we did not interview workers. 4. Another possibility would have been to interrogate protective legislative around women's work, which we have not done due to not interviewing women workers (see: Lahiri-Dutt and Robinson 2008). 5. In line with agreed ethical protocol, factory names are not provided. 6. We witnessed all women workers covering in hijab at the end of their workday, before they stepped outside the working space and into the public domain. Pakistani feminists have noted that the hijab also serves as a refuge against male harassment and offers them a protected way through which to negotiate the public sphere (Zia 2009, 41). {Note: we are not suggesting that women's harassment in the public sphere is an everyday violence applicable only to Muslim women. Bowman (1993) traces the ways in which street harassment of women is common in the USA as much as it is in India and elsewhere.} 7. The construction of Sri Lankan women workers as 'empowered' is recent (Goger, 2013). Sri Lankan apparel workers were conventionally stigmatized and considered disposable from the inception of the garment industry from the late 1970s (Hewamanne 2008;Lynch 2007), and the first author's research in Sri Lanka suggests that some aspects to this stigma still linger. 8. Feminist scholars studying Pakistan may dispute this stereotype, given the spread of Islamist forces and its contradictory effects on the subjectivity of women (Jamal 2009;Zia, 2009). 9. Even in Goger (2013) analysis of managers constructing women workers from being disposable to empowered, she misses out pursuing this feminist interpretation. 10. A manager explained to us that while in the Islamic world, the usual norm is bride price (i.e. here the potential husband pays into the wife's family to make claim over her), in Pakistan -aside from the North-Western Frontier area -the practice was dowry, also known as jehez. He explained that since Pakistan was initially part of India and most Pakistanis initially were Hindus, it was these customary practices rather than Islamic religious practices that shaped marriage negotiations and relationships.
has worked at the University of Southampton, Hobart & William Smith colleges, University of Munich and the International Labor Office (Geneva). Her research explores themes around feminism, ethnicity, labor practices, and ethical trade in South Asia, which has been funded by the AAUW, ESRc, and British Academy, among others. Her research has been published in peer-reviewed journals, edited volumes and as a research monograph.