GEOFORUM
PERGAMON Geoforum 30 (1999) 337-350
www.elsevier.com/locate/geoforum
Insider or outsider, both or neither: some dilemmas of interviewing in
a cross-cultural setting
Beverley Mullings
Department of Geography, Syracuse University, 144 Eggers Hall, Syracuse, NY 13244-1090, USA
Abstract
This paper contributes to the growing literature on methods and techniques for conducting qualitative research in economic geography, as well as to recent feminist debates on the impact that relationships of power between researchers and their informants have on the rigor of the findings of qualitative research. Drawing upon my own experiences whilst conducting interviews with managers and workers in information processing companies in Jamaica, I will examine the ways that inter-cultural perceptions, interactions and representations influenced the fieldwork process, and their ultimate effect on my interpretation and writing of the final text. This paper includes that because of the dynamic way in which identities and their attendant power relations are created and transformed during business interviews, uncertainty will necessarily remain a residual in the evaluation and interpretation of information received. It argues that recognizing and naming these uncertainties is an important step towards not only establishing rigor in the research process, but also to displacing the indomitable authority of the author. © 1999 Elsevier Science Ltd. All rights reserved.
Keywords: Economic geography; Methodology; Positionality
At a time when feminist, social and cultural geographers are beginning to examine the way in which we conduct research and the subjectivities that are inscribed in our work, the dilemmas posed by some of the research methodologies currently employed by economic geographers are also being reassessed. The writings of a number of feminist critics (Hartsock, 1987; di Stephano, 1990; Haraway, 1991; Harding, 1991; Code, 1996; Rose, 1997) have been instrumental in shaping the current process of reassessment, because they have questioned claims to objective and value-free research and have sought to explore how relationships of power between researchers and their informants influence how knowledge is interpreted and represented. Haraway (1991), for example, argues that as scholars we embark upon research with 'maps of consciousness' that are influenced by our own gender, class, national and racial attributes. A researcher's knowledge is therefore always partial, because his/her positionality (perspective shaped by his/ her unique mix of race, class, gender, nationality, sexuality and other identifiers), as well as location in time and space will influence how the world is viewed and interpreted. Drawing upon metaphors of spatiality, it can be therefore argued that 'knowledge is never pure but is situated in the complex and sometimes contradictory social locations of producers and audiences'
(Women and Geography Study Group, 1997). A number of feminist scholars have argued that researchers need to incorporate methodologies that recognize the existence of multiple viewpoints and the partiality of their own assessments (McDowell, 1992a; England, 1994; Gilbert, 1994; Archibald and Crnkovich, 1995; Lawson, 1995; Pratt and Hanson, 1995).
In the search for alternative ways of knowing and interpreting the world, many social scientists have begun to re-examine the benefits of qualitative techniques, and in particular those related to ethnography (Eyles, 1988; Katz, 1994; Kobayashi, 1994; Code, 1995; Wolf, 1996; Baxter and Eyles, 1997). The conduct of ethnographic fieldwork is firmly grounded in an empirical methodology that is devoted to the analysis of research data acquired by means of first-hand interactions with members of a local community over a substantial period of time. Most ethnographers utilize qualitative techniques such as participant observation, to provide in-depth understanding of the inner working of a particular social group. While the rise in popularity of ethnographic data-gathering techniques has provided new insights into the way in which subjects view, interpret and respond to their world, it has tended to skew debates regarding the fieldwork process towards particular issues. Thus, many of the current reflections on the purpose,
0016-7185/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved. PII: S0016-7185(99)00025-l
338
B. Mullings I Geoforum 30 (1999) 337-350
dilemmas and successes of particular qualitative methodologies in geography have revolved around research into the behaviors, motivations and actions of individuals as representatives of given communities (England, 1994; Katz, 1994; Kobayashi, 1994), rather than as informants on the operations of more abstract entities such as firms, industries and nations.
Similarly, the debates regarding the positionalities of researchers and their subjects and the consequent power relationships that develop between them, have tended to focus on situations where the researcher is in a more powerful position than the participant (Patai, 1991; Sidaway, 1992; Lai, 1996), a reflection perhaps of the relatively large amount of work conducted with poorer/ subordinated informants. Few of the debates have examined the practical difficulties that surround other qualitative research methods or situations where the researcher is not in a position of relative power. Thus while the current popularity of ethnographic methods has provided geography with new issues to consider regarding the way we conduct research, it has not fully satisfied the particular difficulties that face economic geographers. Given the fact that the informants in research conducted by economic geographers are often CEOs, managers and other members of business elite groups, the relations of power that are encountered are significantly different from those encountered with other non-elite groups.
Drawing upon my own experiences whilst conducting interviews with managers and workers in information processing companies in Jamaica between 1994 and 1995, I will examine the way in which inter-cultural perceptions and interactions influenced both the data collection and interpretation processes. As a researcher, my particular combination of gender, race, class and age characteristics, had significant effects on the type of information I sought to collect and on my interpretation of the information that was shared with me. My commitment to inquiry from a feminist perspective also significantly shaped the groups whose opinions I sought and the questions that I asked. In relation to this focus, I will also share some of the dilemmas that such a commitment brings to business interviews, the preferred research instrument of a growing number of economic and industrial geographers.
1. The qualitative turn in economic/industrial geography
Since the mid 1980s, the standardized questionnaire-based survey has diminished in popularity because it has been incapable of providing an adequate explanation for the structures and processes that influence the strategies and behavior of firms and industries. Almost every economy and industry in almost every part of the world has undergone some degree of restructuring since 1980.
In order to compete in a rapidly globalizing and regionalizing world economy, managers have developed new industrial strategies and organizational structures. Making sense of the constraints and choices open to firms and their consequent behavior, has not been possible using the traditional questionnaire-based survey because such methodological tools while sometimes able to reveal regularities in actions, have not been able to uncover the complexity of the motivations behind those actions. In the search for alternative empirical methodologies, many economic geographers have turned to more qualitative research instruments (Schoenberger, 1987, 1988; Gertler, 1988, 1996; Christopherson, 1989; Storper, 1990; Pratt and Hanson, 1995; Clark, 1998). It is believed that certain qualitative research methodologies provide a better understanding of the policies and practices pursued by firms and industries in the 1990s because they avoid the 'top-down' setting of research priorities, that in the process serve to obscure the questions, and knowledges that are not produced in the academy (Lawson, 1995).
As Schoenberger (1988, p. 182) advocates, the unstructured business interview 'permits direct investigation of strategic decision making' by allowing researchers to interrogate the strategic choices, interrelationships and trade-offs that lie behind quantifiable actions. While the unstructured interview has become the most popular method for understanding corporate strategy, the actual process of conducting such research has been largely unproblematized. Only a few geographers have commented on the fact that as a field method, it too is highly affected by issues of power, positionality and subjectivity (Schoenberger, 1991; McDowell, 1992b; Herod, 1993). These issues have become important because of their ultimate impact on the rigor1 and ethics involved in such research.
The use of qualitative methodologies such as business interviews raises particular challenges for economic geographers. The first challenge relates to the nature of the questions that economic geographers are likely to ask and the audience to whom such research is presented. Given the historical embeddedness of much of the sub-discipline in the practice of orthodox neo-classical economics, many economic geographers have found themselves under considerable pressure to adhere to conventional notions of rigor in the quest for 'scientific objectivity'. Traditionally such notions have ignored the significance of social relations in the spatial expression of economic relations, focusing instead on universal geometric concepts and physical processes in the space
1 In this paper I draw upon the definition of rigor proposed by (Baxter and Eyles, 1997). They define rigorous research as accounts that satisfy of the criteria of credibility, transferability, dependability and confirmability.
B. Mullings I Geoforum 30 (1999) 337-350
339
economy (Lee, 1994). While geographers who utilize qualitative methods often do so because they seek to understand how social relations of production shape economic geographies, they are often haunted by the discipline's demand for accounts of the 'real' world that conform to the conventional criteria of validity, reliability and objectivity found in quantitative methodologies. The second challenge comes from the groups of informants that economic geographers are likely to seek. Economic geographers who utilize qualitative research methods often conduct interviews in environments where much of the information sought must come from elite informants to whom access is severely restricted. Business elites present greater difficulties of access than most other groups because they are often the exclusive sources of information essential to understanding a firm or industry's operations. As Thomas observes:
Penetrating the social life of a neighborhood can be difficult, but it usually does not take as concentrated a form as it does in a large company. You cannot just walk into an office suite and expect to strike up a conversation or hang out and observe the scene - the courtesies a letter carrier or a drugstore clerk might extend to a stranger in the neighborhood are generally not extended by executive secretaries to intruders who obviously 'don't belong' (Thomas, 1995, p. 82).
Given the strict security and time constraints that most elites operate under, researchers often find themselves with only a brief window of opportunity to convince those from whom they seek information that such an endeavor is worthwhile. In this brief encounter, a researcher's positionality whether perceived or represented often has a crucial impact upon whether he/she is granted an interview. The unstructured interview in the context of economic geography therefore raises new challenges for researchers. To embrace this method requires more than an engagement with techniques that are not statistically quantifiable, it also requires a recognition of the relationships of power played out during the interview process and the effect that researchers have on the final outcomes. This raises specific questions about the strategies used by researchers to make their informants (or their gatekeepers) inclined to participate in the research, and the effect that strategies such as self-representation, have on the faithfulness of the information provided.
Positionality and subjectivity in business interviewing
A number of economic and industrial geographers have begun to address the role that positionality and power play in shaping the findings of the fieldwork process.   Regarding  these  influences, Schoenberger
(1991) argued that while qualitative interviews allowed researchers direct access to the events and decisions that determine actions, there was a danger that such knowledge went through a filtering process that was susceptible to misinterpretation. Recognizing the subjectivities and positionalities underlying many supposedly objective interviews, she argued that researchers sometimes made analyses that reflected their interpretation of an interviewee's own experiences. For example, the cultural meanings attached to particular descriptors; 'a good investment' or 'a bad job' - could result in invalid interpretations. While this insight was pioneering in its recognition of how aspects of positionality such as cultural identity could influence data collection and analysis, Schoenberger did not extend this claim to other attributes of difference such as race, class or gender. By not examining how other signifiers of difference affected the interview process, Schoenberger's analysis limited her examination of the dilemmas encountered in economic geography fieldwork to issues of research design and interview strategy.
In the context of the business interview, the interviewer is often expected to be the 'holder of power', given his/her role as the person asking the questions, but this need not be the case. Herod (1993), for example, argues that not only might the gender of the interviewer and interviewee influence the research process, but gendered assumptions may also influence models of interviewing chosen as well as the subsequent interpretation of the information gathered. Herod argues further, that the asymmetrical distribution of institutional power between men and women has implications for the sort of information and insights that interviews produce. Information provided by male sources, for example, is likely to be influenced by their own gender, class, racial and ethnic position and experiences. He elucidates that male control over corporate information channels can exclude women's experiences and opinions thus providing the interviewer with a partial preliminary set of interpretations. In addition, gender can also be an important factor shaping whether or how an interview takes place. Turner and Martin, for example, state:
... social characteristics of an interviewer and a respondent, such as age, race and sex are significant during their brief encounter; different pairings have different meanings and evoke different cultural norms and stereotypes that influence the opinions and feelings expressed by respondents (Turner and Martin, 1984, p. 271).
In discussions of the power relations that arise during qualitative field research, the focus has been largely static, that is, it is assumed that researchers and those with whom they conduct research occupy fixed positions of power due to their position within specific gender,
340
B. Mullings I Geoforum 30 (1999) 337-350
class or racial hierarchies. Women, for example, are likely to find it difficult to gain access to spaces that are considered 'exclusively male', elites are likely to direct the issues and direction of interviews and poor informants of color run the likely danger of their words being misrepresented or completely unheard. While I am not saying that the debates regarding the dilemmas of authenticity embedded in qualitative methods rely on simplified stereotypes, there has been a tendency to understate the dynamism of the encounter between researcher and researched and the shifting nature of the power relations that ensue.
2. Positionality and power in the collection of data
There is currently much debate regarding how a researcher may gain access to privileged or more balanced viewpoints (Abu-Lughod, 1988; Hill-Collins, 1990; Archibald and Crnkovich, 1995). Writers such as Abu-Lughod (1988) and Hill-Collins (1990) argue that 'insiders', researchers who study a group to whom they belong, have an advantage because they are able to use their knowledge of the group to gain more intimate insights into their opinions. By contrast, 'outsiders' argue that by not belonging to a group under study, they are more likely to be perceived as neutral and therefore be given information that would not be given to an outsider (Fonow and Cook, 1991). 'Outsiders' also argue that they are likely to have a greater degree of objectivity and ability to observe behaviors without distorting their meanings. The binary implied in the 'insider/outsider' debates, however, is less than real because it seeks to freeze positionalities in place, and assumes that being an 'insider' or 'outsider' is a fixed attribute. The 'insider/ outsider' binary in reality is a boundary that is not only highly unstable but also one that ignores the dynamism of positionalities in time and through space. No individual can consistently remain an insider and few ever remain complete outsiders. Endeavors to be either one or the other reflect elements of the dualistic thinking that structures much of Western thought.
As implied in the title of this paper my experiences suggest that to acquire information that faithfully represents the real world, researchers must often seek, what I will refer to as positional spaces, that is, areas where the situated knowledges of both parties in the interview encounter, engender a level of trust and co-operation. These positional spaces, however, are often transitory and cannot be reduced to the familiar boundaries of insider/outsider privilege based on visible attributes such as race, gender, ethnicity or class. In fact, in interviews with business elites it may be better to seek shared spaces that are not informed by identity-based differences, because these are rarely failsafe indicators of an individual's positionality. Making the wrong assump-
tions about the situatedness of an individual's knowledge based on perceived identity differences may end an interviewer's access to crucial informants in a research project.
Traversing the landscape of power relations in interviews, by attempting to create momentary spaces where the interviewer and the interviewees positionalities complement each other, is easier said than done. Particularly where information is sought from individuals who occupy highly disparate positions within an economic or industrial hierarchy, the creation of positional spaces with individuals who are viewed as part of one group might negatively affect the perceptions and willingness of other groups to share information. These dilemmas are particularly heightened for those who seek to understand the functioning of an industry or economic system at a range of conceptual scales that would require an interrogation of perspectives of a range of actors within an economic or industrial hierarchy.
2.1. The politics of self-representation
The process of self-representation is an important component of the search for shared positional space. How individuals represent themselves can make the difference between being granted an interview or not. Particularly among CEOs and company managers, being able to build trust and creating a sense of impartiality is often a crucial requirement for being granted an audience. Impartiality here is not a search for the sort of distance that is viewed as a sign of objectivity, it is instead a desire to create a space during interviews that allows interviewees to share information freely. Particularly because maps of consciousness reflecting gender, race, class or sexuality are imprinted upon all information provided and received, trying to occupy spaces during interviews that minimize these excessive distortions is desirable. But how might such a feat be achieved?
Despite the knowledge that researchers can never be fully located on one side or the other of the insider/ outsider boundary, in business interviews it is still often necessary for researchers to represent themselves as being so located, in order to gain access to information. Many economic geographers, for example, commonly advise researchers to actively display a sound knowledge of the topic under discussion in order to win the respect and confidence of business elites during interviews. Such performances have the potential to create a space where the researchers and their subjects can view each other as intellectual equals. This is, I would argue, one way of representing oneself as a temporary insider. While temporary 'insider' status may be granted to academic researchers who display a knowledge of the discourse of business and industry, there is no guarantee that other visible signifiers of difference such as race, gender or age
B. Mullings I Geoforum 30 (1999) 337-350
341
would not nullify such gains. While there certainly has been much published research on the negative impact of some visible aspects of identity, such as race, there remains no unitary way to predict the impact of these identifiers on a research project. Phoenix (1994) elaborates on this point in her assessment of two studies, one exploring the lives of mothers under twenty years of age and the other examining the social identities of young people. Phoenix argues that while it may be comfortable for a feminist researcher to interview women, that ease does not come from shared gender but may be partly the result of shared class and/or shared color. She is adamant that finding a shared positional space should not be viewed as simply a process of 'racial and gender matching' and advocates that the dynamism of individual identities should be taken into account. Similar findings are reported by Pratt and Hanson (1995) who argue that:
Positions are not static; this is a point that needs to be underlined carefully in the contemporary context, in which "marking" by sexual orientation, class, race etc. is sometimes used not only to open up new conceptual spaces but also to discipline and silence others (Pratt and Hanson, 1995, p. 25).
Lutz (1993) in her examination of the experiences of Turkish women immigrants who worked as social workers with Turkish communities in the Netherlands and Germany also found that assumptions of shared positionalities based upon ethnicity were problematic. In her study she found that immigrant social workers acting as community intermediaries were prone to being marginalized by both others conceived to be insiders, and those considered as outsiders. Among the communities that they worked with, these intermediaries were prone to gendered impositions based on cultural traditions among their non-immigrant, native peers, whilst simultaneously challenging institutional racism within their workplace and wider society. To negotiate the boundaries in this terrain of power, many women regulated the types of information that they would share, representing themselves in ways that appeared acceptable to both communities.
Drawing upon my own recent field experience, the complexities of my own attributes - a black woman of British/Jamaican heritage, from a North American University - made it impossible to be viewed as either an insider or outsider. In fact, as I shall demonstrate, my attempts to interview foreign and local business owners and mangers during research on the information processing industry in Jamaica in 1994-95, involved a constant shifting of the multiple axes upon which my identity rested. In my encounter with local and expatriate managers of information processing firms I found that I was most able to access valuable insights into the
motivations behind corporate strategies of these elites when I emphasized aspects of my identity that I believed would be considered non-threatening, regardless of whether I was considered an insider or outsider. These qualities almost invariably focussed upon a shared knowledge of the industry or a lack of knowledge of the socio-politics of Jamaican society. These positional spaces provided me with a forum for developing an empathetic understanding of the situatedness of those from whom I sought answers. In this regard, I attempted to be both an insider and outsider. I aim to demonstrate that the creation of such a relationship is fraught with ethical and political dilemmas that run the risk of alienating researchers from some of the individuals from whom they seek information.
3. Researching management and workers in Jamaica's information processing industry
The export of information processing services such as data-entry and telemarketing, is a relatively recent industrial activity in Jamaica (government policy to develop the sector only started in 1986). It involves at the most basic level the collection, transmission, storage, processing and display of information using communications technology such as computers and telecommunications equipment. Services that are exported range from the keying in of information for magazine subscriptions or coupons, to the development and conversion of software for a wide range of applications. Much of the industry, however, tends to be concentrated in lower value-added data-entry services where women constitute the majority of the workforce. The Jamaica information processing industry is dual in nature, divided between government assisted, foreign-owned back-offices located in a free zone enclave in the tourist resort area of Montego Bay and local sub-contracting companies scattered largely throughout Kingston, the capital city. Business elites in the Jamaican information processing industry therefore are of two main types: expatriate elites largely from the United States, and local elites, who in terms of firm size are a subordinate group. While both groups are dominated by men (women constitute only a third of the owners of information processing firms), the managers of local companies are largely black and those of foreign-owned firms largely white. While both groups of businesspersons can be considered elite, neither can be considered key players within the global international business community. Nearly all of the firms located in Jamaica operate as the back-offices or sub-contractors to larger multinational corporations. These individuals hold power, but their power is dispersed and expressed largely through their ability to create employment in the communities where they are located.
342
B. Mullings I Geoforum 30 (1999) 337-350
The information processing industry whilst projecting itself as a high-tech industry set to take Jamaica into the 21st century is little more than a service industry largely dependent on sweated female labor. Elsewhere (Mullings, 1998), I describe the industry as flexibly Taylorist because of its combination of elements of labor flexibility with rigid controls over the work process itself. Like the assembly operations in the manufacturing sectors of many developing countries, this service export represents Jamaica's latest attempt to capture a greater share of diminishing international capital. In the wake of declining flows of foreign capital to the region, and increasing levels of competition between the islands, the state has turned a blind eye to the regulation of the capital-labor relationship. While under existing labor laws and statutes workers have the right to choose union representation and are entitled to overtime, holiday and sick leave, few attempts have been made to ensure that these new export-oriented industries adhere to these provisions. As a result, workers in the information industry are not currently unionized, and past attempts to do so have resulted in workers being laid off. In one case, when workers went on strike in protest over pay and conditions their employers, a US-based company, closed their operations in Jamaica within a couple of months. As an economic geographer my focus in this research was to understand how the process of structural adjustment, with its emphasis on the creation of diversified, export-oriented sectors shaped the development and performance of the information industry.
In the first three months of both 1994 and 1995 I conducted interviews with managers and workers in various information processing firms to examine the impact that policies to create this service export had on the structure of the industry and the organization of work. In order to assess the successes and failures of this new private-sector-led, export-oriented industry, it was imperative that I interview both owner/managers and workers. Like Sayer and Morgan (1985) I believed that it was important to interview 'both sides' of the industry in order to understand the structural positions of capital and labor in the restructuring process. Interviews only with managers would have provided an interpretation of the organization of work but would have ignored issues of great importance to workers. By contrast, interviews only with workers might have yielded an overly narrow picture of the industry and the complexity of the wider economic pressures shaping the organization process. In addition because of my commitment to feminist episte-mology it was important for me to connect the global and national institutions and practices that influenced the competitiveness of the industry to changes occurring in the everyday experience of work in the industry and at home. My objective in this endeavor was to uncover the way in which patriarchal structures combined with capitalist structures to produce the realities of everyday
life for the workers in the information processing industry. This therefore necessitated interviews with not only local and expatriate industry elites, but also the women who worked for them.
Due to the fact that workers in the industry were neither unionized nor represented by any collective association, all contact with workers had to be sought through their place of work. I could have sought to contact women who worked in the industry independent of their employers, but I would have risked alienating managers, many of whom were already reluctant for any questions to be asked about their operations. At the end of each business interview, I asked managers to inform their workers of my presence and desire to discuss their opinions regarding the future growth and direction of the industry. This often seemed to be a 'natural' question to ask because often interviews would end with managers lamenting about the attitudes of their workers. The question posed was always non-threatening and non-controversial. Often I would express my inability to understand why workers were not motivated and had no loyalty to the job. I would then proceed to ask whether I could leave a poster inviting workers to participate, or perhaps speak to two or three members of staff about their views of ways that service delivery in the industry could be improved.
While the decision to interview both managers and workers offered a more comprehensive picture of the impact of the process of restructuring, methodologically this approach held its own share of difficulties. Most managers were very reluctant to provide me with access to their workers because they feared, as one manager expressed, 'that they would have problems getting them to work afterwards'. Particularly in the Montego Bay Free Zone, which housed most of the foreign-owned operations, gaining access to workers via this method proved extremely difficult. In addition, many of the women themselves expressed a reluctance to be interviewed. A reluctance, that in subsequent interviews was attributed to fear of reprisals, and a belief that I would share my information with the government or even worse, management. In most of my encounters with workers, I would first be introduced by either owners or managers, many of whom would 'interpret' for their staff my purpose for being there. In one case, for example, workers were simply told that they should 'go and answer the questions that the lady outside wanted to know'.
Given the tense nature of the relationship between workers and managers in the industry, it was not surprising that attempts to interview workers were frequently met with distrust, with few wanting to participate. In some cases managers would provide me with a quiet and private space to interview workers. While the offer of a quiet place to conduct interviews was gratefully received, it simultaneously served to increase the level of doubt in the minds of workers that I
B. Mullings I Geoforum 30 (1999) 337-350
343
approached, that my motive was to simply learn from them, their experience of work in the industry. The contrast between the cramped airless open-plan spaces in which most of the workers were and the often air-conditioned office I was allowed to use served to reinforce my perceived 'camaraderie with management' and my status as an 'outsider'. After a few attempts to interview workers at the workplace, I decided that I would need to embark on a new strategy, one where I used the workplace as a forum for administering a questionnaire to elicit basic social and economic data. This strategy also provided me with an opportunity to arrange with workers to meet at a later date for a more in-depth interview. In the later interviews, I was surprised at the extent to which workers would 'open up' and share their impressions of the industry and its effect on their everyday lives.
It is not surprising that my desire to understand the perspectives of both groups was problematic, especially given the particularly antagonistic relationships that have traditionally existed between workers and managers in Jamaica (Manley, 1975; Stone, 1986, 1991). Being committed to this particular strategy for knowledge construction, however, represented more than simply a methodological conundrum, it highlighted the contradictions that the pursuit of a traditional economic geography agenda can pose to a feminist-informed one. My commitment to feminist-informed focus, made it necessary to extend my focus, beyond business elites and their perceptions of the macro-economic events affecting the information processing industry, to the effect of restructuring on the relations of power within the workplaces and homes of the women who comprised 99 percent of the industry's workforce. The dilemmas that the resulting methodology posed to the way that I represented myself and my ultimate interpretation of the research made it clear that economic geographers who seek to do more than 'add women and stir' must also examine how the questions that are asked, interpreted and presented may conflict with the dominant modes of representation within either feminist or economic geography.
4. Positionality, subjectivity and ethics in the field
Striking a balance between both groups without being automatically viewed as 'trying to undermine either the managers or the workers' was fraught with difficulties. Being introduced by managers made workers suspicious of my reasons for being there, while seeking to conduct interviews with workers away from work premises made managers suspect my motives for wanting to interview their workers in the first place. Striking a balance between both groups became an issue of finding spaces where I could see the world from the viewpoint of both groups of respondents, and oppor-
tunities where I continue to represent myself as an impartial seeker of information.
With managers, I found that questions demonstrating my awareness of the issues affecting their firms' business strategies provided me with a level of credibility and respect. I asked questions that sought to establish how a range of actors (the state, parent-offices, competitions, subcontractors and workers) influenced the service provided by firms and their plans for future service provision. The unstructured interviews, therefore, served two main purposes. First, the interviews provided me with information on the factors affecting the export of information processing services in Jamaica, and second, they provided an opportunity to develop a level of trust with managers/owners that would allow me in due time to request access to their workers.
While some managers consented to my request to interview their employers, this did not indicate that they fully trusted me. While much effort was made to conduct all interviews away from the watchful eyes of managers and supervisors, it was impossible to fully eliminate their presence. On one occasion while I was collecting the completed questionnaires from the data entry operators at a particular firm and making arrangements to conduct a more probing interview, a supervisor whisked the complete questionnaires out of my hand in order to 'ensure that the they had been fully completed'. Needless to say, after the incident it was almost impossible to secure additional volunteers for further interviews at that firm. As argued by Hsiung (1996), who interviewed factory owners in Chinese garment assembly factories and their women workers, developing reciprocal relationships with elites runs the danger of losing a degree of control and autonomy over the fieldwork process. In Hsiung's case her desire to be a participant observer in the garment industry was exploited as managers increasingly viewed her as a valuable source of free labor.
4.1. Representation and the landscape ofpower in the field
Shaffir (1991) observed that mildly deceptive practices of representation are as inherent in field research as they are in daily life. This I found to be the case whilst reflecting upon my interviews with Jamaica's business elites. Once access had been negotiated, the process of procuring information became an intricate process of identifying spaces and times when it was desirable for me to be an 'insider', and situations when it was more desirable to be an 'outsider' to the social group under inquiry. The decision to emphasize my status as an outsider was often employed when I wanted to know more about the politics of Jamaica's industrial policy or manager's opinions regarding the labor process. I recognized that these were areas that, given the volatility and importance of Jamaican politics to everyday life, were particularly sensitive (Stone, 1986; Bakan, 1990).
344
B. Mullings I Geoforum 30 (1999) 337-350
While the extent to which being perceived as an outsider might have limited the amount of information shared with me cannot be known, anecdotal evidence suggests that my concern to be so identified was valid. For example, at the beginning of many of the interviews, inquiries were often made about my relationship to a high ranking government official, because we shared the same surname. In the context of an island with a small population like Jamaica, people who share a surname are often closely related, so the inquires were probably prompted by a need to establish the level of caution to be exercised when making statements especially about the government. By constructing myself as largely British, with only a vague knowledge of Jamaica and its political culture, I felt that I could represent myself in the least threatening way.
Ironically, the decisions that I made when representing and positioning my identity, were shaped by my partial knowledge as an insider and my desire to be an empathetic listener. As a partial insider I was aware that certain signifiers of my identity (my Jamaican ancestry and past affiliation with the local university) were likely to heighten levels of suspicion and distrust among both local and foreign managers. With this knowledge, I was able to draw upon my affiliation with a North American university and my British nationality to create a persona that was less likely to be perceived as a threat.
Whether such deceptions are ethical or not has been the subject of some debate (Katz, 1994; Keith, 1992; Abu-Lufghod, 1988). Keith (1992), examining issues of research ethics in ethnographic methodologies, argues that this type of research often strays considerably from the reflexive ideals that many geographers espouse. In relation to his participant observation work with the police, Keith argues that many of his self-representational practices were arguably deceitful, undemocratic and perhaps indefensible. Selectivity in the disclosure of aspects of his identity in ethnographic research was a form of unethical betrayal of the trust and confidence that was extended by those who provided him with information. While these practices were unethical, they conformed to the protocols of academic reportage, whose unspoken rules value and prioritize representative accuracy whilst ignoring the politico-ethical considerations that shape the power relations implicit in knowledge production. Particular academic textual and methodological strategies often become an abuse of the power held by the researcher because scholarly texts 'always give the academic author the final world'.
While there is no doubt that researchers who are selective in the disclosure of aspects of identity undemo-cratically manipulate the context in which information is provided, I would argue that there is no guarantee that such selected disclosures are the main determinants behind that type of information than an informant decides to share. My attempt to achieve 'outsider' neutrality did
not give me full access to all the information that I sought. While I sought to represent myself in ways that would minimize any threat that visible aspects of my identity might create, I had little control over the ways that managers interpreted and reacted to visible aspects of my identity such as gender and race, even though I did try to minimize any negative impacts that these biological attributes might have had, however, by seeking initial access through letter or telephone appointments. Just as there are a multiplicity of readings that emanate from individual tests, so too are the meanings attached to a researcher's body, regardless of the way that he/she may choose to represent it.
4.2. Partially inhabiting the spaces of local elites
Among the managers of locally owned information processing companies, my presumed lack of knowledge of Jamaican politics allowed them to open up and provide me with very candid opinions about the industry. Most were quite critical, for example, of the government's policy of providing the mainly foreign firms located in the free zone enclaves with generous tax and location incentives, as well as access to cheaper telecommunication services but not the locally owned ones. Many felt that the reluctance of both the state and the local private sector to provide assistance to the industry was an important factor behind the local industry's failure to thrive. Some managers went further to explain that unless the traditional owners of capital in Jamaica found the industry to be a lucrative endeavor, from which they too could profit, it was unlikely that the local segment of the industry would receive financial support in the future. One manager went as far as to state that the industry was unlikely to receive substantial support without the direct participation of the local telecommunications company, chaired by a member of one of the five richest families in Jamaica.
Being made privy to the fairly strong opinions of the local managers of information processing companies made me feel that I was not perceived as a threat to them. Upon reflection, I concluded that perhaps I was being provided with such 'insider' information because of the relative subordination of this perceived 'elite' group to the rule of even more powerful local elites, because I appeared to embody some of the attributes of the local industry's managers. Though unspoken, as a doctoral candidate at the time, I, like many of the local managers was part of a generation whose occupational mobility had been achieved through access to education at all levels. In the context of the Jamaican industrial sector the information processing industry was unusual because, as firm owners pointed out in interviews, it attracted a new breed of Jamaican entrepreneur, the black middle-class professional, who normally would not have become an owner of the means of production.
B. Mullings I Geoforum 30 (1999) 337-350
345
The shifting of the historically entrenched position held by Jamaica's white, Chinese and other ethnic minorities, in manufacturing as well as banking and finance (Reid, 1977; Stone, 1980; Gordon, 1991), was largely achieved through the educational achievements and occupational mobility of the black population. My positionality as a black doctoral candidate, appeared to invoke some degree of support among local managers, though this was neither a representation that I sought to promote nor one whose value I was even aware of.
4.3. Race and shared positionality
My shared positionality as a person of African descent, also allowed me to partially and temporarily inhabit the space of an insider, among black owner/ managers even though this was not an aspect of my identity that I sought to promote. This was particularly evident in interview with local managers, many of whom shared concerns about the difficulties that racial and ethnic stereotypes posed for their efforts to market the quality of their service. For example, a black, male manager of a local information processing firm, in explaining why he had merged with an American company in order to market his databases stated that:
I am conscious enough to know that the industry is so specialized and depends on so much contacts and impressions ... credibility in the marketplace, that it is just too much for me to get into ... plus in the advertising business I haven't seen one black face so ... I don't believe that I could develop that kind of company in America and penetrate that industry (Interviewee A, 1994).
The extent to which my position as a person of color made non-black managers and owners perceive me to be an outsider cannot be known. While I experienced considerably more difficulty procuring interviews in the free zone, where most of the foreign-owned firms were located, I am not certain that this could be attributed simply to the fact that I was a black women seeking access to firms headed by predominantly white foreign managers. Given the high level of security maintained in the free zone areas and the general distrust with which all outsiders are regarded (Mullings, 1996), it is just as likely that my presence as a researcher was as disruptive as my presence as a person of color.
While issues of race did not enter my discussions with white managers and owners of information processing companies, a number of them indirectly alluded to the euphemistic 'cultural biases' held by their clients. A number of managers, for example, stated that it was difficult to convince clients to outsource documents to the island because many felt that it was a backward and disorganized place where they were liable to lose their
data. As B a white, American male, who managed a data-processing company in the free zone shared:
Some people do think that they are packing up sending all their documents to a Caribbean Island, or offshore from the United States, its scary for them. They worry that they are going to have a riot down here,... that they are never going to get everything back (Interviewee B, 1994).
My vague and partial relationship to Jamaica, and my affiliation with a North American University, allowed me to create a neutral space whereby managers could freely discuss their opinions. Occupation of this space was beneficial because on issues regarding the relative ease or difficulty of investing in Jamaica, I was able to explore the sometimes conflicting opinions and experiences of both foreign and local investors. In interviews with local managers, for example, managers candidly shared their opinions on their relationship with their foreign-based competitors. Interviewee C, a white Jamaican male, for example, stated that the government's policy of offering fiscal and locational incentives to the largely foreign investors in the free zone areas was foolhardy and shortsighted because ultimately:
The Americans have no loyalty, the Americans they'll come and they'll go! (Interviewee C, 1994).
Dissatisfaction was also voiced by the managers of foreign-owned firms, who felt that doing business in Jamaica was fraught with bureaucratic and cultural hostilities that made for a particularly difficult business environment. For example, as interviewee B shared:
The Jamaicans, their culture ... is totally different than ours. They don't understand customer service ... they really don't as far as I am concerned and they need to work on it ... it's an attitude across the population ... and the government promotes it. I'm not made to feel welcome here (Interviewee B, 1994).
While interviewee D, the black male manager of a US based firm stated:
... the first time when I got back to America after being here for a while I got back and thought, 'Thank God! American Soil!!' (Interviewee D, 1995).
Interviewees, based on assumptions of our mutual racial and national positionalities, shared information research (Phoenix, 1994; Stanfield II and Rutledge, 1993; Essed, 1991), they might not have otherwise done. Their perceptions of me as a temporary insider, however, did not mean that the positional space that we
346
B. Mulangs I Geoforum 30 (1999) 337-350
shared was stable. On issues regarding the difficulties of marketing information processing services owner/managers appeared willing to view me as a temporary insider, sharing their perceptions of the effects of nationality, culture and 'race' on sales and investments. On issues, however, where managers were likely to be viewed as the holders of power e.g., management worker relationships, or during the process of negotiating access, I was not welcomed or made to feel like an insider. At these times I was treated with much suspicion and made aware of their power to withhold access to information. In keeping with the observations of Thomas (1995), I was keenly aware that regardless of the extent to which my questions appeared innocuous or even beneficial to the organization of an elite group, I would not be extended the time or the courtesies that would have been extended to other fellow elites.
4.4. An outsider to the spaces of local elites
While my status as a doctoral candidate appeared to provide some degree of 'insidership' with local investors, this positionality developed only after I had gained access to managers. Among both groups my position as a doctoral candidate and an outsider to the information processing industry was initially disempowering as I was constantly symbolically reminded of my status as a seeker of information, wholly dependent upon the managers/owners and their gatekeepers for information and subsequent access to workers.
Thus in nearly every scheduled interview, appointments would be either cancelled at the last minute or, if kept, attended at least half an hour later than the scheduled time, even when the dates and times had been agreed upon weeks in advance. In fact, interviews were often more successful when they were arranged at very short notice (on the same day) and confirmed fifteen minutes before the scheduled time. Even when a meeting was in progress, the relative control that I was able to extend over the interview process was limited by numerous telephone calls, all of which would be attended to during the course of the interview. In the process of negotiating access, the fact that I was neither part of the information processing business community nor of a comparable elite group,2 rendered me an outsider, regardless of my gender, race or nationality. The difficulty that I experienced in having promised interviews honored was noticeably greater among the managers of locally owned firms. Patiently waiting for interviews to be rescheduled appeared to be a significant signifier of respect, because once initial contact was made it was rel-
2 While membership of a foreign university appeared to increase the likelihood of being granted an interview, my status as a doctoral student held little currency.
atively easy to make repeat calls to request further interviews or additional information. Among foreign-owned firms, however, this was not the case, once an interview was completed it was almost impossible to get companies to furnish any further information.
4.5. Gender, class and the dilemmas of representation
While I made choices about the aspects of my identity that I wished to disclose during interviews, my representation was not always in my control. The elites that I interviewed made choices about the level of information that they were prepared to give based on their own perceptions of me. While I had greater control over the attributes that a given national representation conferred, perception and stereotypes regarding my gender, class and race were elusive and uncontrollable. This was most apparent during interviews with women managers. In most of these interviews I naively expected that questions about the wage and labor process would be answered differently than they would be by men managers. Given the disproportionate number of women in the industry and the fact that we were women discussing issues that primarily affected women, I had expected more empathetic accounts that perhaps would reflect our shared positionalities as women. This was an ironic expectation on my part given the fact that I myself had actively sought positional spaces where the meanings attached to the visible aspects of my identity were neutralized. Thus I anticipated accounts that would be cognizant of the difficulties of juggling the demands of the home with that of the industry. From the responses and perceptions that I recorded, it was clear that class was a much more important axis of difference than gender and that women managers viewed me as an outsider to the world of their workers and more closely aligned to their own class positions. So as interviewee E, a black female manager of local data processing firm stated when asked about the ability of workers to earn a living wage:
I don't know how they manage, because when I look at the hairstyles and I know how much they cost, and when at transportation costs ... I can't relate to it but they manage. I worry about paying the light, the mortgage, the phone ... but I probably live in a different environment and therefore have to project a certain image ... they will go out and buy dancehall things and that is why you see them wearing it on the road because they 'shack out'3,
3 Like the English expression 'the Full Monty', to 'shack out' means to dress extravagantly. This term, however, carries with it pejorative, class-biased connotations critical of the style of dress popular in Jamaica's dancehall subculture.
B. Mullings I Geoforum 30 (1999) 337-350
347
and their lives revolve around that (Interviewee E, 1994).
While I was unsuccessful in creating a gender-based positional space, I nevertheless found women managers to be far more willing than men to provide me with access to their workers. Perhaps we did in fact, share common concerns, even if they could not be articulated lest they threatened the differences in class and authority that were expected between managers and workers.
4.6. Representation: an ethical dilemma
My attempts to represent myself in ways that rendered me either a partial 'insider' or 'outsider', had its share of ethical dilemmas. Whilst I did not falsify any of the attributes that constituted my positionality, my search for shared spaces where I could develop some degree of trust made it difficult to assess whether I should have been explicit about all the dimensions of my positionality. Should I have 'come clean' about my opinions regarding the distribution of power between managers and workers in interviews with the business elites that I spoke to? Should I have expressed my doubts about the capital-labor relationship with its racial undertones with managers in the free zone areas? Should I have explained to the data-entry operators that part of my research aimed to examine the extent to which the competitiveness of the industry was affected by the quality of their work. My choice of questions in all of these regards was shaped by my situated knowledge of the world borne out of my own experiences as a middle-class, black, woman. Given the setting, and my choice of methodology it was not surprising that I found myself caught in the middle of the quiet conflict between workers and managers, viewed by neither group as a trustworthy insider. During one field interview, for example, I was confronted with the conflict that my relationship with management posed for my relationship with workers. I had just completed a second interview with the female manager of a locally owned data-entry shop and had requested access to her employees in order to ask them to participate in my research. The manager said that she would introduce me to 'the ladies' so that I could make further arrangements to meet with those who were willing to participate. I stood beside the manger as she introduced me to her employees, and as she turned to leave she began to stumble. I caught her arm to steady her, and as I glanced up at the group of quietly smirking workers realized that my actions had firmly placed me in the camp of 'management' instead of the camp of 'workers'. I felt rather annoyed at this callous lack of regard for what could have been a significant injury. As I tried to ignore the jokes and giggles that continued long after the manager left the room, I realized that I was in an ethical dilemma. How could I
demonstrate in the few minutes that I was given, that I had no particular affinity towards management and that in fact, I wanted to understand their views of the viability of the industry given its particular organization of work, and its impact upon their daily lives. Seeking to tread the narrow path of neither insider nor outsider in this instant failed, and I found it difficult to convince but a handful of women to meet me at another time, away from the work environment.
Seeking spaces that are neutral, that render one neither an insider nor outsider in the context of unstructured interviews with both bosses and workers is a very difficult exercise, which is destined to place the interviewer in the camp of either one or the other at sometime during the research process. I have concluded that had this industry been more open to the public gaze, perhaps I might have had greater success in gaining equal access to both groups, but its closed nature and the divisions between workers and managers made success in interviewing managers the recipe for lack of success in gaining the trust of workers. So great was the difficulty of developing relationships of trust between both groups that I resorted to clandestine meetings with workers outside office buildings in order to set up interviews. This tactic raised high ethical dilemmas for me. How could I seek relationships of trust with managers without informing them that I intended to also interview their employees? What repercussions might my actions have on workers if it were to be revealed that they had met with me?
5. Positionality and situated knowledge in interpretation of qualitative data
While perceived aspects of a researcher's identity may be disempowering in certain interview encounters, power is almost uniformly invested in the researcher when he/she interprets and writes up the research. Certainly in my own study, while I might have felt at times like a disempowered outsider to the worlds of the workers and business elites that I interviewed, I did retain significant authority over the interpretation of the information I was given, the quotations used to support my claims and ultimately the final text presented to the public. I not only had full authority over the theoretical framework that I used to analyze and make sense of the changing relationship between labor and international capital in Jamaica, I also maintained control over the way in which the data that I collected were incorporated into the theoretical and empirical analysis. In both endeavors the situatedness of my knowledge, and the knowledge that I derived from an extensive reading of the work of others, shaped my interpretation and presentation of the research. Establishing certainty in the claims that I made, required the use of a number of
348
B. Mullings I Geoforum 30 (1999) 337-350
techniques ranging from the triangulation of the information from multiple sources, methods and theories to the unveiling of the constructs that guided the research and the limits imposed by the sample used. While making clear the underlying rationale for my claims instituted a certain rigor to my findings, there remained much uncertainty regarding the influence that my own subjectivities and strategies to find shared positional spaces imposed on my findings.
A number of geographers have begun to explore ways that the process by which information is interpreted can be made more transparent in qualitative methodologies (Rose, 1997; Gilbert, 1994; Katz, 1994; Kobayashi, 1994; McDowell, 1992a; Pile, 1991). They argue that researchers can situate their knowledge through reflexive consideration of the relations of power that operate during fieldwork process. Reflexivity then becomes an important process for making clear the landscape of power within which much research is conducted. In an extensive examination of the role of reflexivity in minimizing the distorting effect of power relations in qualitative research methodologies, Rose (1997) argues that feminist geographers tend to advocate two main types of reflexive practices. These practices are based upon what Rose refers to a 'transparent reflexivity'. This particular form of reflexivity 'depends on certain notions of agency (as conscious) and power (as context), and assumes that both are knowable' (Rose, 1997, p. 311). The first of these practices calls on researchers to be inwardly reflexive, making clear through self-conscious and critical introspection, their positionality via-a-vis the research. Using this tactic, it is argued (Katz, 1994; Radcliffe, 1994; Moss, 1995) that researchers can take active steps towards redistributing power by employing specific research strategies such as the shared writing of texts (Mbilinyi, 1989) or the explicit disclosure of the circumstances surrounding the collection of data and their analyses (Dyck, 1993).
The second tactic requires researchers to outwardly make visible their position within the written research, by revealing the circumstances that surrounded data collection and analysis, that is, the context within which fieldwork is conducted. Such a tactic requires researchers to seek ways of linking smaller scale methodologies to larger-scale political concerns thereby revealing how power is distributed between macro and micro scales (Nast, 1994; Staeheli and Lawson, 1994). While it is advocated that these tactics make visible the landscape of power in which research is conducted, Rose argues that as a discourse 'it produces feminist geographers who claim to know how power works, but who are also themselves powerful, able to see and know both themselves and the world in which they work' (Rose, 1997, p. 311). This reflexive practice, she argues is bound to fail because of the impossibility of sustaining a unified and coherent gaze at a landscape of power. No re-
searcher is able to occupy the same positionality as those who are the subjects of his/her research, yet the unavoidable distance that this relationship creates is paradoxically the one that researchers seek to overcome through transparent reflexivity. Rose calls for geographers to develop other forms of reflexivity, ones that recognize the extent to which the interview process is often one where both researcher and those who are the subjects of research create versions of themselves that are re-interpreted and re-presented in different ways. In such a situation researchers should able to point out the uncertainties and gaps in interpretation that necessarily accompany the interview process. In my own research, there were curious silences that reflected my inability to provide definitive truth statements. For example, I could not be sure that the poor performance of locally managed information processing companies was due to their failure to receive the type of concessions given to foreign investors by the state, or whether they would have told a different story or shared other types of information with me had I been foreign white researcher. Neither was I sure that the accounts of home and family life wouldn't have been richer had I been a locally-based researcher, perhaps representing a Non-Governmental Organisation (NGO) rather than the academy. I certainly felt that had my focus not been that of understanding the factors shaping the organization of the industry, and my audience not been the academy my choices regarding the methodology chosen, the groups approached and the way that I presented my findings to my peers would have been different. But without these self-imposed constraints my research would not have satisfied my desire to uncover the complex struggles between labor and capital, local and foreign capital, men and women, daughters and parents that shape the economic landscape in which Jamaica's information processing industry is located.
6. Conclusion
While my research outcomes were influenced by the extent to which I sought to find spaces that evoked the least threat or suspicion from the elites whose opinions I sought, the process of finding shared positional spaces was not as simple as I had imagined. Not only were some of my personal attributes notably, gender and race, beyond my ability to direct, but also, the meanings that these attributes conveyed changed with each person that I interviewed. For example, my positionality as a 'feminized', 'racialized' subject, did not always exclude me from particular types of information. Similarly, as I failed to adequately appreciate when I began my research in the field, the positionalities of the 'elites' that I sought information from were also as dynamic and cross-cutting as my own. This made it equally difficult to
B. Mullings I Geoforum 30 (1999) 337-350
349
determine with accuracy spaces where I would gain access to information ordinarily unavailable to outsiders.
Qualitative methodologies such as unstructured interviews do present a challenge to economic geographers. Researchers often are faced with limited periods of time to gain access to individuals who may or may not provide information that is crucial to understanding firm and business behavior. These individuals often also have the ability to influence whether other 'non-elites' will participate or assist in the research. In the process of exchange between a researcher and an elite subject, there is a convergence of positionalities. It is at this moment, when perceptions of a researcher's position-ality begin to evoke stereotypes that influence the opinions and feelings expressed by respondents, that a researcher should seek to create spaces that foster trust and co-operation. Is it always possible to strike the right balance and if we did, would we know, and how would it affect the validity4 of our claims? Here, I endorse the issues that others have already identified as challenges to geographers who utilize qualitative methodologies (Schoenberger, 1992; McDowell, 1992b; Rose, 1997). While traditional quantifiable questionnaire-based instruments provided us with the 'surety of objectivity', such claims using unstructured interviews must now be reconsidered. Gaining access to a range of respondents and being able to corroborate their views and identify inconsistencies can go some way towards validating findings based on business interviews, but recognizing that the information we as researchers receive will always be partial makes our claims more circumspect and our stance more reflexive. This is a consideration that is particularly important for researchers whose identities rest upon axes that are not only different, but in many circumstances may be dis-empowering. Identifying aspects of difference which may stultify dialog and seeking spaces where some level of trust can be established, to me, is the only way that researchers can gather information that is reliable. Recognizing and naming those uncertain moments when positional spaces may not have been shared, or when dialog may not have been honest are not only important steps towards producing valid accounts of economic landscapes based on qualitative methodologies, but that also important strategies for displacing the indomitable authority of the author.
4 This I would define as the extent to which an account represents the social phenomena that it refers to. While conventionally, the term 'validity' has carried connotations of the search for an objective and indisputable truth, my use of this term seeks to convey that my search was for interpretations that appeared justified given the coherence and consistency of the theory and evidence proposed given the context in which both elements (theory and evidence) were produced (see Baxter and Eyles, 1997 for a more through examination of this topic).
Acknowledgements
I would like to thank Liisa Cormode, Alexandra Hughs, Mark Monmonier, Jennnifer Robinson and the anonymous reviewers of this piece for helping me to improve on an earlier draft. I would also like to acknowledge the valuable comments received from participants in the Industrial Geography Speciality Group session on New Methodologies for Elite Research, at the 1996 Association of American Geographers Meeting in Charlotte, North Carolina, when this paper was first presented.
References
Lughod, L., 1988. Fieldwork of a dutiful daughter. In: Altorki, S.,
Fawzi El-Solh, C. (Eds.), Arab Women in the Field: Studying Your
Own Society, 139-161. Syracuse University Press, Syracuse. Archibald, L., Crnkovich, M., 1995. Intimate outsiders feminist
research in a cross-cultural environment. In: Burt, S., Code, L.
(Eds.), Changing Methods Feminist Transforming Practice, 105-
126. Broadview Press, Ontario. Bakan, A., 1990. Ideology and Class Conflict in Jamaica: The Politics
of Rebellion. McGill-Queen's University Press, Montreal. Baxter, J., Eyles, J., 1997. Establishing qualitative research in social
geography: Evaluating 'rigour' in interview analysis. Transactions
of the British Institute of Geographers 22. Christopherson, S., 1989. Flexibility in the U.S. service economy and
the emerging spatial division of labour. Transactions of the
Institute of British Geographers 14, 131-143. Clark, G., 1998. Stylized facts and close dialogue: Methodology in
economic geography. Annals of the Association of American
Geographers 88, 73-87. Code, L., 1995. How do we know? Questions of method in feminist
practice. In: Burt, S., Code, L. (Eds.), Changing Methods: Feminist
Transforming Practice, 13^14. Broadview Press, Ontario. Code, L. (Ed.), 1996. Taking Subjectivity into Account. Routledge,
New York.
di Stephano, C, 1990. Dilemmas of difference: feminism modernity and postmodernism. In: Nicholson, L. (Ed.), Feminism/Postmodernism, 63-82. Routledge, London.
Dyck, I., 1993. Ethnography: A feminist research method. Canadian Geographer 46, 80-89.
England, K., 1994. Getting personal: reflexivity, positionality, and feminist research. Professional Geographer 46, 80-89.
Essed, P., 1991. Understanding Everyday Racism: An interdisciplinary Theory. Sage, London.
Eyles, J., 1988. Interpreting the geographical world qualitative approaches in geographical research. In: Eyles, J., Smith, D., (Eds.), Qualitative Methods in Human Geography. Basil Black-well, Oxford, pp. 1-16.
Fonow, M., Cook, J. (Eds.), 1991. Beyond Methodology Feminist Scholarship as Lived Research. Indiana University Press, Bloom-ington.
Gertler, M., 1988. The limits to flexibility comments on the post-
Fordist vision of production and its geography. Transaction of the
Institutions of British Geographers 13, 419-432. Gertler, M., 1996. Worlds apart: The changing market geography
of the German machinery industry. Small Business Economics 8,
87-106.
Gilbert, M., 1994. The politics of location: Doing feminist research at
"home". Professional Geographer 46, 90-96. Gordon, D., (1991). Race, class and social mobility in Jamaica, In:
Lewis, R., Bryan, P. (Eds.), Garvey: His Work and his Impact.
Africa World Press, New Jersey, pp. 265-182.
350
B. Mullings I Geoforum 30 (1999) 337-350
Haraway, D., 1991. Simians Cyborgs and Women: The Reinvention of
Nature. Routledge, New York. Harding, S., (1991).Whose Science? Whose Knowledge? Cornell
University Press, New York. Hartsock, N., 1987. The feminist standpoint: Developing the ground
for a specifically feminist historical materialism. In: Harding, S.
(Ed.), Feminism and Methodology, 157-180. University of Indiana
Press, Bloomington. Herod, A., 1993. Gender issues in the use of interviewing as a research
method. Professional Geographer 45, 305-317. Hill-Collins, P., 1990. Learning from the outsider within the sociological significance of black feminist thought. In: Fonow, M., Cook, J.
(Eds.), Beyond Methodology Feminist Scholarship as Lived
Research, 35-59. Indiana University Press, Bloomington. Hsiung, P., 1996. Between bosses and workers the dilemma of a keen
observer and a vocal feminist. In: Wolf, D. (Ed.), Feminist
Dilemmas in Fieldwork, 122-137. Westview Press, Colorado. Katz, C, 1994. Playing the field: Questions of fieldwork in geography.
The Professional Geographer 46, 67-72. Keith, M., 1992. Angry writing representing the unethical world of the
ethnographer. Environment and Planning D: Society and Space 10,
511-568.
Kobayashi, A., 1994. Coloring the field gender race and the politics of fieldwork. The Professional Geographer 46, 73-79.
Lai, J., 1996. Situating locations: The politics of self identity and "Other" in living and writing the text. In: Wolf, D. (Ed.), Feminist Dilemmas in Fieldwork, 185-214. Westview Press, Colorado.
Lawson, V., 1995. The politics of difference: Examining the quantitative/qualitative dualism in post-structuralist feminist research. Professional Geographer 47, 449-457.
Lee, R., 1994. Economic Geography. In: Johnson, R.J., Gregory, D., Smith, D. (Eds.), The Dictionary of Human Geography, 149. Blackwells, Oxford.
Lutz, H., 1993. In between or bridging cultural gaps migrant women from Turkey as mediators. New Community 19, 485-492.
Manley, M., 1975. A Voice at the Workplace: Reflections on Colonialism and the Jamaican Worker. André Deutsch, London.
Mbilinyi, M., 1989. I'd have been a man politics and the labor process in producing personal narratives. In: Personal Narratives Group, (Ed.), Interpreting Women's Lives: Feminist Theory and Personal Narratives. Indiana University Press, Bloomington, pp. 204-227.
McDowell, L., 1992a. Doing gender feminism feminists and research
methods in Human Geography. Transactions of the Institute of
British Geographers 17, 399^16. McDowell, L., 1992b. Valid games? A response to Erica Schoenberger.
Professional Geographers 44, 212-215. Moss, P., 1995. Embeddedness in practice, numbers in context: The
politics of knowing and doing. Professional Geographer 47, 442-
449.
Mullings, B., 1996. Industrial Development in an Era of Structural Adjustment: The Growth of Export Informatic Services in Jamaica. Unpublished Ph.D. Thesis, McGill University.
Mullings, B., 1998. Jamaica's information processing services neo-liberalist niche or structural limitation. In: Klak, T. (Ed.), Globalization and Neoliberalism: The Caribbean Context, 135-154. Rowman & Littlefield, Boulder.
Nast, H.J., 1994. Opening remarks on women in the field. Professional Geographer 46, 54-66.
Patai, D., 1991. U.S. academics and third world women: Is ethical research possible? In: Gluck, S.B., Patai, D. (Eds.), Women's Words: The Feminist Practice of Oral History. Routledge, New York.
Phoenix, A., 1994. Practising feminist research the intersection of
gender and race in the research process. In: Maynard, M., Purvis, J.
(Eds.), Researching Women's Lives from a Feminist Perspective,
49-71. Taylor & Francis, London. Pile, S., 1991. Practising interpretive geography. Transactions of the
Institute of British Geographers 16, 458^69. Pratt, G., Hanson, S., 1995. Gender Work and Space. Routledge,
London.
Radcliffe, S., 1994. Representing post-colonial women: Authority
difference and feminisms. Area 26, 25-32. Reid, S., 1977. An introductory approach to the concentration of
power in the Jamaican corporate economy and notes on its origin.
In: Stone, C, Brown, A. (Eds.), Essays on Power and Change in
Jamaica. Jamaica Publishing House, Kingston. Rose, G., 1997. Situated knowledges: Positionality reflexivities and
other tactics. Progress in Human Geography 21, 305-320. Sayer, A., Morgan, K., 1985. A modern industry in a declining region:
Links between method theory and policy. In: Massey, D., Meegan,
R. (Eds.), Politics and Method: Contrasting Studies in Industrial
Geography, 147-168. Methuen, London. Schoenberger, E., 1987. Technological and organizational change in
automobile production spatial implications. Regional Studies 21,
199-214.
Schoenberger, E., 1988. From Fordism to flexible accumulation:
Technology competitive strategies and international location.
Environment and Planning D 6, 245-262. Schoenberger, E., 1991. The corporate interview as a research method
in economic geography. Professional Geographer 43, 1980-1989. Schoenberger, E., 1992. Self-criticism and self-awareness in research: A
reply to Linda McDowell. Professional Geographer 44, 215-218. Shaffir, W.B., 1991. Managing a convincing self-representation: Some
personal reflections on entering the field. In: Shaffir, W., Stebbins,
R. (Eds.), Experiencing Fieldwork, 72-82. Sage, Newbury Park,
California.
Sidaway, J., 1992. In other words on the politics of research by "First World" geographers in the "Third World". Area 24, 403-408.
Staeheli, L.A., Lawson, V., 1994. A discussion of 'women in the field': The politics of feminist field work. Professional Geographer 46, 96-102.
Stanfield II, John, H., Rutledge, D. (Eds.), 1993. Race and Ethnicity in
Research Methods. Sage, London. Stone, C, 1980. Democracy and Clientelism in Jamaica. Transaction
Books, New Jersey. Stone, C, 1986. Work Attitudes Survey: A Report to the Jamaican
government. Earle Publishers, St. Ann, Jamaica. Stone, C, 1991. Race and economic power in Jamaica. In: Lewis, R.,
Bryan, P. (Eds.), Garvey: His Work and his Impact, 243-264.
Africa World Press, New Jersey. Storper, M., 1990. Industrialization and the regional question in the
third world lessons of post imperialism; prospects of post-Fordism.
International Journal of Urban and Regional Research 14, 423-
444.
Thomas, R., 1995. Interviewing important people in big companies. In: Hertz, H., Imber, J. (Eds.), Studying Elites Using Qualitative Methods, 3-17. Sage, Thousands-Oaks.
Turner, C.F., Martin, E., 1984. Surveying subjective Phenomena. Russell Sage Foundation, New York.
Wolf, D.L.,1996. Situating Feminist dilemmas in fieldwork. In: Wolf, D.L. (Ed.), Feminist Dilemmas in Fieldwork. Westview Press, Boulder, pp. 1-55.
Women and Geography Study Group, 1997. In: Feminist Geographies: Explorations in Diversity and Difference. Addison Wesley Longman, Harlow, Essex.