Fields of difficulty and dialogue

Like many ethnographers before me (e.g. Hammerley & Atkinson 1995; Clifford &

Marcus 1986), at the writing-up stage I found it hard to pinpoint where and when my fieldwork actually started or ended. Obviously, I remember and wrote down in my fieldnotes the date and place of my first ‘actual’ fieldwork day. I arrived at the first maternity healthcare clinic, where all the nurses had consented to participate in the research, overcome with feelings of fear and anxiety. I had barely slept the night before, I had my brand new notebook packed in my bag, and my hands were shaking when I set up the cameras in the consulting rooms. Similarly, I clearly remember the last day I spent at each clinic. I recall feeling relieved but at the same time sad as I sat in the coffee rooms with the nurses, drinking farewell coffees and discussing my research and the future.

Yet as I look back on my research process, I realise that many activities, places and times before and after the ‘actual’ fieldwork at the clinics also must be incorporated into it.

These include the first contacts with people in the maternity healthcare organisation, applying for formal research permission and seeking consent from different levels of the actual work organisation, discussions with the nurses about preliminary results and research materials, and almost monthly meetings and hundreds of emails with informants from the local municipal administration. After the actual fieldwork was carried out, we researchers in a research project arranged workshops in each of the participating clinics for the nurses who participated in the study. In the workshops, researchers and nurses together watched and reflected on extracts of video recordings of each individual nurse’s own

appointments with clients. The initial purpose for the workshops was for the researchers to communicate their preliminary analyses to the participating nurses and to offer the nurses a space to reflect on their work practices, based on the activities seen on the videos and our preliminary analyses. All these places and spaces, or sites, are part of my fieldwork, not just in the sense of gaining access to the field and reporting back on the research, but also in the sense that those experiences, encounters and observations contributed to the knowledge produced in my study and reshaped its framework.

Thus my research field includes not only the multiple sites of the everyday, actual and textual, frontline setting of care work and the encounters taking place in it, but every site that has contributed to how I came to know and describe my field. Furthermore, the ways I know and describe my field involve my own presence or interference in it. I agree with writers such as Jessica Mesman (2007) and Donna Henriksen (2002) that in ethnography interfering (intervening in the original) is not really a choice, but an unavoidable condition.

In this sense my study differs from classic descriptions according to which ethnographers go to the field, observe, but do not change anything. Just by being in the field the ethnographer makes a difference, and in my case I am not trying to deny it. I have not aimed to make interventions of any specific kinds during my fieldwork, but I acknowledge and reflect on the fact that neither I nor the participants in my study have full management of knowledge, resources or identities, which all fluctuate in the research process (see also Mesman 2007; Holmes & Marcus 2008).

Overall, I share the view that participation always characterises feminist research in general and ethnography in particular.⁴⁰ This can be theorised as interfering, intervening,

40This brings me to the question of how to conceptualise all the public health nurses, the pregnant women, their partners, their children, the doctors, psychologists, social workers, family care workers, administrative staff, and the lecturers in the various training and counselling classes who participated in my research. Are they informants, the researched, co-researchers or participants, and do these positions vary over the phases of research (design, implementation and evaluation/reporting back)? After many headaches and with the help of earlier literature on the topic, I have decided to call them participants (and informants), but with quite strict meanings. Obviously, as a feminist scholar, I do not want to conceptualise them as the researched, because to do so would entail objectifying connotations and fixed power relations between the researcher and

researched (see also Mesman 2007). I also dislike the terms co-researcher and co-ethnographer, because they blur the differences and conceal the different institutional positionings of the researcher and the participants (cf. Mol 2008a, 10). To render all the participants, including the researchers, ‘the same’ would ultimately

collaboration or feminist action research, to name but a few alternatives (see e.g. Haraway 1991a; 1997; Mesman 2007; Reinharz 1992). In fact, over the last twenty years or so, collaborative approaches to ethnographic practice, termed a reflective turn (Van Maanen 2006, 16), have become widespread. Ethnography and collaboration are viewed here as intertwined practices (e.g. Henriksen 2002; Messman 2007; Holmes & Marcus 2008).

According to approaches that stress the collaborative dimensions of ethnography, ‘doing ethnography’ entails not just giving descriptions of practices of the given field but also changing them either deliberately (as in action research) or unintentionally (just by being in the field). It has even been claimed that the research participants in contemporary ethnographic research settings increasingly expect outside interlocutors, other professionals, researchers or wider public deliberation to be part of their everyday life. The settings where such people operate include contexts that require people to reflect on their own actions and engage in a communicative relationship with the outside world. (Holmes

& Marcus 2008.) In my view, such settings are often to be found within contemporary institutional/organisational contexts, such as NGOs and state services, where communicative practices are increasingly constituted within a discourse of dialogue, participation and empowerment (Phillips 2011): that is, workers are encouraged to work in multiple and equal ‘partnerships’ with participants from various field of expertise and lay people who are constructed as ‘active’ informants or even co-researchers (Homanen 2012, 214–215). As I will show, the Finnish maternity healthcare organisation I analyse in my

mean paralleling scientific interests with the health and social workers’ (often politically minded) interests, thereby neglecting the question of the positions of ‘insider’ and ‘outsider’ and conflating the multiple roles and responsibilities of all the participants (cf. Mesman 2007). All of the people from the fields of maternity healthcare that have taken part in my research have played a double role as participants and informants.

Depending on the person, the emphasis on participation and information-giving has varied, and it may have changed over time. However, all the people who contributed to my research material have in some way or another given me relevant information concerning the social relations and orders that structure their everyday activities. This has been the case both in the discussions I have had inside and outside the clinics and in my recorded and transcribed (expert) interviews with pregnant women and public health nurses. All the informants are also participants to varying degrees. (the model adopted from Anna Rastas [2007].) By calling them participants I want to stress the dimensions of collaboration with people in the field. I have for the most part set the research agenda myself, and have kept my research project from becoming part of theirs (cf. Caswill & Shove 2000). Yet I recognise that the concerns and even politics I encountered in the field redirected my research interests and analysis.

study is a good example of one such institutional site (see e.g. Handbook of Maternity Healthcare 2007).

Of all the terms conceptualizing participation and collaboration I prefer Donna Haraway’s (1991a, 1997, 272-273) term interference, which refers to critical feminist studies (of science) that diffract rather than reflect. What Haraway means by this is that feminist studies, and scientific studies and knowledge in general, should diffract existing patterns of science: the concept contains the idea of interference patterns that can shift existing meanings (1997, 16). Thus, unlike the reflected image of ‘the same’, diffracted patterns of interference bear ‘the effort to make difference in the world’ (1997, 16; 2004a, 69–70).

I understand the model of diffraction as re-visioning reflection and participation through situated knowledges. Indeed, the model draws upon Haraway’s earlier writings on reflexive knowledge and connective politics in Situated knowledges: the science question in feminism and the privilege of partial perspectives (1991/original in 1988) andSimians, cyborgs and women: the reinvention of nature (1991b) (see also Campbell 2004).

Responding to Sandra Harding’s (1986) concept offeminist objectivity, Haraway wrote of situated knowledges as feminist objectivity in relation to both feminist knowledges (of science) and scientific knowledge itself (1991b, 188/1988). The concept refers to the partiality of knowledge, all knowledge. It is a ‘location, partial embodiment and partial perspective’ (Haraway 1991a, 191; cf. Smith’s standpoint). The particularity and specificity of the embodiment of the one who knows is reflected in the concept, as is her position in social orders or ‘networks’ (Haraway 1991a, 190; Campbell 2004, 170).

I interpret interference, diffraction and situated knowledges as mapping positive interventions in a critical, reflective and dialogic manner. Dialogue is the key in Smithian methodology to achieve experience-based knowledge so essential to discovering how individual actions are connected with and coordinated by the actions of others in institutional orders. Experience, as mentioned before, is understood in IE as what people come to know through everyday bodily being and action and emerges for the ethnographer

in dialogue among particular people in particular times and in particular places. (Smith 2005, 123–141.)

These dialogically produced accounts of people’s experiences, then, work as entry-points that organize the further inquiry into the social relations coordinating the work of all the participants involved in institutional activities. In theorizing the social, social relations and the epistemological grounds for studying them, Smith draws on Bakhtin’s theorization of dialogue. For Bakhtin (1981; 1986), language is an ongoing and historically developing complex which is responsive to time and people’s intentional utterances. According to Bakhtin, language is inherently dialogic as meanings are produced through the interplay of multiple voices. In that interplay, a unity is formed but, as a result of the play of difference across voices, that unity is a multivocal one, full of contradictions (Bakhtin 1981; Clark &

Holquist 1984). The rules and regulations of language (grammatical, cultural and other) are integrated into language. However, language is not determined by such regulations and each utterance reproduces and elaborates language (Bakhtin 1981; 1986; Smith 2005;

Homanen 2012, 217).

In Smith’s application, the concept of dialogue is more inclusive than that of Bakhtin.

Whereas Bakhtin’s (1981; 1986) concern is mainly language as dialogue, for Smith dialogue characterizes the social practice in general. Social complexes, such as institutions, are ongoing historical processes in which people’s actions are caught up in, and responsive to, what others are doing and responsive to, and shaped by, what has been going on. By adopting this kind of dialogic ontology of the social, IE aims to treat people as agents and subjects and take account of the particularities of their lives and actions. Thus it strives to avoid constructing abstractions that objectify people. Such abstractions, it is argued, are often in play both in institutional activities and in conventional research practices through representation in texts and face to face interaction. (Smith 2005, 50, 62, 66, 68, 123;

Homanen 2012, 217.)

Smith claims that in institutional contexts, privileged, totalising interpretations and representations have a tendency to suppress diverging perspectives and experiences. Here,

she draws on Bakhtin’s critique of monologue as an attempt to force through a singular voice in the face of plural meaning-making (Smith 2005, 62, 66). Dialogue for her is not just a naturally occurring human encounter but can be realized potentially through methods and efforts to avoid monologue.⁴¹ (Homanen 2012, 217-218.)

Experience-based knowledge that IE seeks to acquire in dialogue is, however, never obtained fully and without difficulties. Some of them originated from problems inherent in IE theorizing of dialogue between participants⁴² and some from the design and implementation of the particular research at hand and from the situatedness of the ethnographer doing the fieldwork. IE recognizes that institutional workers accounts always remain to some extent subpressed by the abstract institutional terms and concepts. The original design of my study and the context of suspicion linked to my position in relation to nursing administration that, I will shortly elaborate on, was not productive in that sense.

They resulted in a situation of power imbalance that could not be erased. (Homanen 2012, 232.)

Further, I myself as the embodied instrument of ethnographic knowledge – that seems to be the minimal definition of ethnography (Ortner 1995) – had an effect on the field and the ethnographic knowledge produced. Among other things, I am a childless and academically educated woman in my 30s, I am a PhD student in women’s studies, and I am situated in numerous ways, both similarly with and differently from my informants, in networks of various power relations in and out of the field (see also Fortier 1998). In fact, what is

41 This is not meant, however, to imply that Smith perceives that dialogue is something that can be achieved through a set of fixed methods. That would be in contradiction with her ontological commitments in which realities, discursive and non-discursive, are only realized in practice. Thus, methods for achieving dialogue are realized in each individual practice at certain times, and they may or may not be transformable to other practices or times.

42 Smith includes non-human participants in her concept of dialogue, especially conceptual entities (texts, discourses and so on). However, dialogue has been criticised for being a humanist concept that is unsuited to ANT-influenced STS research (e.g. Plesner 2012). For instance, Ursula Plesner (2012) has proposed a change in vocabulary, from ‘dialogue’ to ‘negotiation’, to include all the elements – not just people – in the formation of (research) knowledge and to enable a shift from the ‘knower to the known’. I see no need to change vocabulary, here, for two reasons. First, the IE concept of dialogue is already quite inclusive

compared to the theories on dialogue with which Pelsner takes issue (e.g. Black 2008; Anderson & Baxter &

Cissna 2004 [eds]). Second, I have already included ANT-influenced insight in my methodological framework (including the most recently introduced concept of diffraction) to account for ‘odd types of assemblages’ (Plesner’s 2012 term) that may escape critical theory focused on power inequalities and using concepts such as dialogue.

constituted here is an enormous material-semiotic web of relations that fluctuates over time, as does my position in it.

In the rest of this subchapter, I focus on the possibilities and limitations and problems for achieving diffraction and dialogue by concentrating on three possible sources of difficulty in my study: 1) conflicting interests the researcher and the participant nurses had; 2) the different (power) positions between researchers and participant nurses; 3) differences in the knowledges of the participant nurses and the researchers, and the meaningfulness and origins of those knowledges.

Conflicting interests or complementary projects?

In the beginning of my fieldwork, the nurses talked a lot about politics with me in relation to a new nursing intervention, the family-centered MCH, involving new and additional working methods⁴³ implemented by the municipality. They seemed to view me and the other researchers of the project as representatives of the local administration responsible for the intervention, or at least someone(s) who could influence the decision makers. They voiced their resentment toward the new nursing intervention, pointing out that they felt that all their and their predecessors’ experience-based knowledge and work were being overlooked by the administration. At times, I was also an object of their resentment and was treated with hostility and suspicion. For example, I was told not to use some discussions that took place in the coffee room in my research because the nurses were afraid the administrative staff would make their lives hard, and sometimes I was denied access to the coffee room all together by the door slamming in front of my face. (Homanen 2012, 221-222.)

Regardless of my repeated assurances that neither I nor the other researchers in the project were advocates for the reform or representatives of the administration, I cannot blame the nurses for being suspicious of me. Acquiring their consent to participate in the first place had followed a top-down model where contact was first made with the people in charge of

43 The new methods were aimed at ‘client-centeredness’, addressing ‘psychosocial concerns’ and ‘family-orientation’.

reforms in the administration, and the consent of the regional administrative personnel (regional head nurses and doctors) had been obtained before the nurses themselves were contacted.⁴⁴ Furthermore, the administrative personnel selected the specific clinics that were to be contacted for individual consent. The nurses at the clinics were informed of all this and knew about other contacts we had with administrative personnel in charge of the reform, and I would not be surprised if they had felt obliged to give their consent given that the authorities had given theirs. (Homanen 2012, 222.)

The nurses’ interest in influencing the administration through us seemed to conflict with our research-based interest to gain knowledge of the actual work practices. Although I do believe that I eventually gained the nurses’ trust, the question of conflicting interests still haunted me through out my fieldwork. The nurses’ interest and the agenda to influence administration as well as our research interests were still the same despite the trust gained in the field. As Caswill and Shove (2000) point out, there is a danger that such conflicts of interest lead to the loss of a space for interaction among the research participants.

(Homanen 2012, 222.)

How, then, could the nurses and I work together given our conflicting knowledge interests? At first, I feared that the nurses would merely speak within the abstract administrative discourse, albeit from a critical position. But in practice when administratively trendy abstract terms and care principles were brought up, the nurses sometimes, criticised how, in formulating new initiatives, the administration overlooked nurses’ practice-oriented knowledges and experiences. For example, the nurses told me that care standards and objectives, such as ‘client-centeredness’, addressing ‘psychosocial concerns’ and ‘family-orientation’, that were presented as ‘new’ by the administration had been realized in the nurses’ work practices for a long time prior to the intervention. When I asked the question ‘how’, this sometimes led to accounts of how such standards were met

44 In detail the process of acquiring consent went as follows: First, formal permission from the municipal Committee of Research Permissions for Welfare and Health Institute (21.8.2006) (SOTE:3827/403/2006).

Then consent from administrative staff (regional head nurses and doctors for three different MCH regions in the city) had to be obtained before the clinic staff could be contacted. What happened at this point was that the administrative staff discussed possible clinics for fieldwork with each other and with the clinics’ own staff. In one region we were denied access altogether, and in another we were told to contact only one of the clinics. In the third region we were pointed towards more possible clinics to contact.

in everyday work. It could be argued that my and the nurses’ differing knowledge interests led to complementary projects: my research interviews and workshops arranged in my research project for the participant nurses after the fieldwork was conducted provided the nurses with a space for articulating their critical stance toward the administration and also provided me with experience-based knowledge on nurses’ work practices (concerning the actualities linked to administrative interests). (see also Homanen 2012, 222.)

in everyday work. It could be argued that my and the nurses’ differing knowledge interests led to complementary projects: my research interviews and workshops arranged in my research project for the participant nurses after the fieldwork was conducted provided the nurses with a space for articulating their critical stance toward the administration and also provided me with experience-based knowledge on nurses’ work practices (concerning the actualities linked to administrative interests). (see also Homanen 2012, 222.)