My mind as transitional space

Intersubjectivity in the process of analyzing emotionally disturbing data

Marian Tankink


Researchers in the field of violence have to deal with confronting the destructive side of humankind and the pain of their informants. This creates strong feelings of both distance from and proximity to the informants, which can create counter-transference reactions such as over-involvement or feelings of anger or grief. Those feelings do not stop after the interviews but are also present during the phase of analyzing the data. During the analysis of the data I have gathered, I have again experienced a kind of intersubjectivity, this time in my own mental space where the appalling stories of the women with their traumatic war experiences have affected my emotions and self, and vice versa. The central question in this paper is why the process of analysis is more difficult than face-to-face contact in the context of a study of refugees who have experienced violence. On the one hand it appeared that during the process of transcribing, the meaningfulness of the interviews in terms of human values and political action was to some extent lost. On the other hand, the personal emotions of the researcher were colored during the analyzing the data.


Vesna told me during the interview how the Serbian soldiers, in the last war in the former Yugoslavia, kept her in a camp. The stories were so horrible that I could hardly bear to hear them. I saw and felt her pain, and it affected me. But her story remained her story. I tried to understand how the experiences she told me about had affected her and what they meant for her. In this interaction we created an intersubjective (emotional) space in which I could verify my own emotions in our (non)verbal communication.

While transcribing the interview at home some time later, it was much more difficult to listen to the story again. I was alone with the story and I missed that (non-verbal) communication with Vesna in which we had created an emotional atmosphere that made the story bearable for both of us. I transcribed her story and tried to keep it away from my own feelings, as if I was a machine. I tried not to let the appalling story enter into my life.

In the next phase, analyzing the information, I could not maintain this ‘disconnection’. I had to let Vesna’s story enter my own intra-psychic space and the story started to interact with my own self, my own (un)conscious difficulties and pain in life. It was difficult to get a grip on the story, and Vesna’s narrative affected my mood and made me emotionally unstable. It was often impossible for me to analyze the story, because in the moment I wanted to continue I would feel physical disgust and reluctance. I realized that I needed to get some insight into these intrapsychic processes in order to stay mentally and physically in balance, but also because, if the appalling stories of the women I have interviewed have had such an impact on me, how can I know that the story I create in my text will not become too much a story of myself?

Little has been written about the emotional impact of doing research with people who have experienced severe trauma, or doing research in violent or traumatizing situations, on the research process and its outcome. In the mental health care field it is well known that working with people who have experienced serious trauma, especially as a result of human violence, can affect health care workers psychologically. Often the health care workers face counter-transference [note 1]Counter-transference is a concept used in psychotherapy. It signifies the intra-psychic process in the therapist, where his/her personal life is the unconscious source of the therapist’s thoughts about, feelings toward and reactions to the client.
reactions that can result in various forms of stress, a heavy emotional impact and eventually burn-out. Therapists can develop secondary traumatic stress and even post-traumatic stress disorder. Some undergo personality changes such as becoming cynical or unhappy; some develop a tendency to withdraw, or lose their self-confidence (Lansen & Haans 2004: 320).

Researchers in the field of trauma can face similar problems. The process of confronting the destructive side of humankind and the pain of the informants creates strong feelings of both distance from and proximity to the informants, which can create counter-transference reactions such as becoming over-involved, or feelings of anger or grief. I came to realize that those feelings did not stop after the interviews but became even more present during the phase of analyzing my data.

My main question in this paper is why the process of analysis is more difficult than face-to-face contact in the context of a study of refugees who have experienced violence. I will argue that the intersubjectivity developed in the first phase of the research remains, even when I, as a researcher, am not in direct contact with my informants any more. Analyzing data makes the mind of the researcher a transitional, intersubjective space. This paper is a first attempt to find an answer to my main question and also to understand how the researcher’s individual struggle during the process of analysis continues to influence the interpretation of the informant’s story: in other words, how the researcher’s personal emotions shape the analyses and interpretation. I will first pay attention to the process of interviewing people in the context of anthropological research with traumatized refugees; subsequently I will discuss, with the help of the concepts of transitional space and intersubjectivity, some difficulties of the analysis.

Doing research with traumatized refugees

I was aware that doing research among people who have had traumatic experiences [note 2]I am working on my PhD thesis on research among refugee women from Afghanistan, Bosnia-Herzegovina, and South-Sudan who experienced sexual violence as part of the conflict in their countries of origin or during their flight. I am primarily interested in the differences and similarities in meaning production and health-seeking strategies, especially silence as a coping strategy.
could have emotional effects on me as a researcher. During the interviews it was often difficult seeing the women struggling with their unbearable memories and the accompanying pain. While my assumption was that in therapeutic sessions revisiting these painful memories could be part of ‘healing’, in my interviews the only aim was to get insight into the women’s health strategies and into how the women give meaning to what happened to them. The women themselves would not benefit from it; quite the contrary, I thought. Most women experienced intensified nightmares, intrusive memories and stress, for at least three to four days after the interviews. I started to question how ethical my research was. Was it really necessary to cause the women so much pain and sadness by my questions? Was my research so important that it legitimized this suffering?

However, while continuing the interviews, I came to realize that for the women the talking was not only painful or harmful, but that most of them also experienced it as helpful. For a number of women it was the first time that they had talked to another person about their rape experiences, and for others it even had a kind of healing effect. Those women experienced talking to me as a relief in the sense that at least one other person on earth knows their stories. Others wanted to tell their experiences as a testimony, so that the world would know what happened. Some women liked to talk in the knowledge that their stories could be helpful for organizing better support and health care for women in the future. To a certain extent, talking could give meaning to their suffering, and they were prepared to bear the sleepless nights, the nightmares and the stress the interviews caused. I came to realize that what I offered the women was the realization that it was possible to create intersubjectivity on such a painful subject as sexual violence, and intersubjectivity is something they could not create in their own environment. By “intersubjectivity” I refer to the symbolic space between two people in which they create a shared reality, an emotional reality characterized by the conviction/sensation that the ‘I’ understands and is understood by ‘the other’ (Tankink & Vysma 2006: 251). This realization, that the women I interviewed  also gained something from the interview, helped me to reduce my feelings of guilt and my fear that I was asking too much of them.

What remained was my shame and anger about the Dutch asylum policy. Although not accountable for the outcome of democratic elections, as a citizen of this country I felt responsible for the harsh asylum practices. These feelings made me vulnerable to becoming over-involved with the women. I helped some of them to find good lawyers or therapists, I brought others in contact with organizations that help asylum seekers who are rejected, and I collected goods from relatives, friends and colleagues for a woman and her children who found themselves in a very complicated situation. I felt it my duty to help my informants.

This struggle with my (over)involvement made me look forward to the end of the interviews. However, as it turned out, the intersubjectivity that I had developed during this first phase of my research changed in character, and had an even greater effect upon me than before.

Violence and fear as part of intersubjectivity

Intersubjectivity as a symbolic space, in which two people create a shared, emotional reality, is a dynamic concept and has to be seen in all its facets: “physical and metaphysical, conscious and unconscious, passive and active, kind and unkind, serious and ludic, dyadic and collective, symmetrical and asymmetrical, inclusive and exclusive, emphatic and antagonistic” (Jackson 1998: 4). In this sense, violence and fear are part of intersubjectivity and have to be taken into the anthropological analysis. Intersubjectivity is more than shared experiences and empathetic understanding. It also entails compassion (which affirms the identity of the two subjects) and conflict (which confirms the differences between the two subjects). The notion of intersubjectivity helps us, according to Jackson, to get insight into how people experience processes that they (as the experiencer) tend to see as extrapsychic processes and we (as the observer) tend to construct as intrapsychic processes. For instance, most women from South-Sudan consider their rape experiences mainly to be a disturbance in their personal relationships. They feel ashamed, but not guilty as in some kind of “crisis of a personal moral principle”. It is very different among the Afghan women. They are also ashamed, but they also feel guilty, because the honor and respect of their relatives have also been affected. The moment I was able to create an intersubjective space together with my informants, I got more insight into the intrapsychic world of the informant.

Although the anthropologist’s task is to relate personal, intersubjective experiences to larger social, political and economic societal processes, he or she needs to gain insight into the intrapsychic processes of the informants as well in order to understand the how and the why of those processes. In other words, in order to understand why the women in my research with experiences of sexual violence tended to keep silent about their experiences as a kind of health strategy, I needed to get some insight into their fantasies, fears, pain, their moral principles and internalized notions and their ideas of how others would react. However, in doing so, as an anthropologist, I also needed to have insight into how my own fantasies, fear and pain and internalized notions shape the intersubjective moment and the data. In this respect, it is important to realize that our own intrapsychic processes are as important as those of our informants.

It is not only mental processes that we need to be aware of; our physical feelings can be part of the intersubjective space as well. I often felt physical tension just before a woman in my research revealed that she had experienced sexual violence. For me, those physical tensions are a kind of bodily communication. It was always very clear to me, after I discussed these tensions with the women, that they were not my own feelings of stress but rather belonged to the woman I was speaking with.

Intersubjectivity is in large part shaped by “unconscious, habitual, taken-for-granted dispositions,” and thus not only by conscious worldviews (Jackson 1998: 9). Our awareness fluctuates between a sense of self that is distinctive, ontological and secure and a sense of self that is unstable and can be occupied by the other, a sense of self that can be overwhelmed and engulfed. The same kind of experience can occur with our body. The body can be in perfect harmony with our will, but can also be experienced as alien. Our awareness shifts between those positions.

Intersubjectivity cannot occur without empathy. Therefore, I think, an aspect of the self always corresponds with aspects that are recognized in the other. It is questionable whether or not an intersubjective space can be created between people who do not have this aspect of recognition, although this recognition is not always on a conscious level. For example, for me it is not totally clear why I became deeply emotionally involved with one woman who is an illegal refugee in the Netherlands because she had not been able to tell the Immigration and Naturalization Service (IND) that she was raped, who could have qualified to migrate legally if she had only been able to include these experiences as part of her application, and why I did not feel the same level of emotional involvement with another women who was illegal for the same reason and whose traumatic experiences were beyond imagination. I might have lost the intersubjective feeling of mutual understanding, but in this case it is very difficult to analyze how it is lost. These processes are not always conscious.

Working in the field of violence is not easy, but it is easier to feel empathy with victims of violence than with perpetrators. Empathy with people who are seen as ‘evil’ can be emotionally very confusing. As Pumla Gobodo-Madikizela (2004) asks in her book A human being died last night, can you have compassion for someone whom you see as evil? and if you have compassion, what does that mean? I will not go into this question because it is beyond the scope of this article, but it is one aspect of why working with perpetrators can be even more difficult than working with victims.

The anthropologist as translator

As noted above, I thought that the most difficult tasks in my research would be the interviews and transcriptions, but it turned out that analyzing my data intensively was far more difficult than expected. What are the characteristic features that make the analyzing process so hard? Analyzing your data and conveying it into an anthropological text is, according to Churchill (2005), comparable with the process of a translator who is translating poems or a book. He quotes the translator of a book by Freud who said that “translators are always trying to get inside the minds of authors on whom they are working” (ibid.: 4). To translate a text the translator has to get to know the writer: not only his or her psychological inner spaces but also the context in which he or she lives or lived. Especially if the writer is dead and lived in another timeframe or environment, the mediating influence of the translator and the translation should be taken into account.

Churchill argues that the process which the anthropologist undergoes between the fieldwork and the moment of writing is comparable to translating a foreign text. Both translator and researcher are reshaping their data “in the subjective territory of the mind” that he calls the transitional space. Both translator and anthropologist search for the meaning of what has been said, or why it was said that way. In order to do this, the anthropologist has to enter the psyche of the informants “to capture the subtle and nuanced meanings of their actions and words” (ibid: 6). Ethnography, as Churchill points out, is not an interpretation of the data, but rather something that requires “a process of empathetically entering the psychic space of other human beings and, till the extent possible, translating the actions of those subjects by way of seeing the world from their point of view” (ibid.: 5). By being a participant, observer, interviewer and analyzer the anthropologist tries to get insight into the psychic and physical domain of the informant.

The most important task is to grasp the ‘point’ of the informant’s narrative. The anthropologist does not need to have the same experiences, but does need the empathic capacity to enter into the feelings and way of thinking of the informants and to understand how they are in the world. At the same time the anthropologist uses his or her own perspective and analysis, even if they are different from the perspective or interpretations of the informant. As an example, most women in my research told me that they had never told their husbands that they were raped. They were convinced that their husbands would divorce them or would start to abuse them physically or mentally. I know that there are also many caring and supportive husbands, but for the women the ‘fact’ that their husband will want to divorce them is real. For me as an anthropologist I have to find out how this fear is conditioned, knowing that not all husbands will show this behavior that the women expect. Where does this fear come from? Why do they not believe that their husbands can be caring and supportive? In order to understand this, some insight into the interaction between the subjective reality of the women and the socio-cultural context is a necessity. In this case the women’s opinions seem be influenced by group narratives. Women only hear stories of men who cannot accept the fact that other men have had sex with their wives, and they become insecure and frightened. The anthropologist needs the capacity, while working with the informants and with the data, to constantly move dialectically between context and inner space, between distance and proximity in interaction (c.f. Richters et al. 2007). In the case of ‘heavy’ topics, such as violence, such movement can be demanding.

The mind as intersubjective space

The theories of Churchill and Jackson can be helpful for understanding what happens when I analyze my data and why doing research on violence is difficult. They do not, however, give a full answer to the question of why the analyzing phase was more problematic for me than the previous phases of my study.

Sharing time with a respondent involves communication on verbal, nonverbal and emotional levels, subject to subject. We are creating the intersubjective space together, me and the other. This gives me the possibility to check whether my feelings and ideas are right and it also gives me the possibility to connect with the other’s story. For instance, returning to the example of the physical tension I felt just before a woman revealed her secret story, I could discuss with the woman how she felt just before she told me, while she was telling me and after she had told me. I came to know that she felt very tense and was very anxious, but that at the same time she also felt a deep need to open up. In other words, I was able to analyze and understand that my physical tensions were connected with the woman’s story.

Back at home, in my own familiar surroundings and drinking a cup of coffee, I found that the stories of the women easily fell out of place. I felt reluctant to let the horrible stories into my own safe environment. This was, however, only a minor aspect of my problems in the analyzing period. More important was that during the process of analysis, I had to go again into the psyche of the informant, as I had also tried to do during the interviews, but now this process was entirely in my own mind. This attempt to enter the psychic world of the informants in order to grasp the meaning of what they said and did is difficult, since you have to go into the informant’s pain and fear. For example, a Bosnian woman who was raped several times in a camp had not told her husband anything. She explained to me the difficulties she experienced with her husband who could not understand her behavior. Although she often sensed beforehand that her husband wanted to have sex with her and was able to swallow some tranquilizers, there were other moments when she completely panicked. One night the fear became so intense that she ran outside naked and screaming. She told me that she even tried to avoid normal daily physical contact with her husband, since even a simple touch could bring her into a state of total anxiety. In my attempt, during the analyses, to fully understand how this fear influenced her life and how she coped with it, in order to give a right interpretation, I felt the tension in my stomach. This “empathetically entering the psychic space of other human beings” as Churchill calls it, is not only a mental exercise but also an emotional and even physical experience, and this combined process is strenuous and caused a struggle for me not only with the emotions of the women involved but with my own emotions as well.

The women’s experiences had to be ‘ingested’ by me, the anthropologist, and my mental territory became a transitional space in which ‘I’ created another kind of reality out of the stories of the women. Their lives and stories became part of my own subjective space, which made it difficult to distance myself from their intrusive and painful stories. I was no longer able to verify whether or not my feelings were ‘right’; in the absence of the other, the other had changed.

Only after discussions about my draft paper with other people did it become clear to me that the dialogues created during the interviews were meaningful in terms of human values and political action and reaction. For an informant to tell a personal story and for the researcher to hear that story is a constructive process “grounded in a specific cultural setting, interaction and history”, of both the researcher and the informant (Mattingly &Garro 2000: 21). Text, context and meaning are intertwined and are not always easy to transfer to the transcription. In the transcriptions and during the analyses I had lost that sense of meaningfulness of the stories the women told me that had been so valuable in the intersubjective space and that had made the stories bearable. It was mainly the traumatic stories and pain that remained in my mind and body; I ‘remembered’ the informants more as victims, or at least sufferers, than also as ‘survivors’. This repressed the empowerment and resilience that the women had also shown me. The pain of the appalling stories and suffering became too dominant during the analyses, resulting in experiences of all kind of mixed emotions, such as compassion, pain, sadness, physical disgust, attraction, emotional solidarity with the informant, fear of being infected by the trauma, and strong feelings of distancing, denial, vulnerability, and insecurity. For example, I interviewed a Bosnian woman who was, together with her husband and children, illegal in the Netherlands. She told me that during the war she was in a camp for a couple of months where she was raped on a daily basis by several soldiers. She had never told anyone about her experiences, including the Immigration and Naturalization Service (IND) in the Netherlands. I was shocked by the terrible experiences she told me about and asked her if her lawyer knew this story, because she might have been granted permission to stay in the Netherlands if she had revealed her secret. She answered that she was too afraid her husband would read it in their files, and asked me if I could perhaps inform her lawyer. I promised her to write a letter about her situation, which I gave to her and which she handed over to her lawyer.

The family was living in the Netherlands in terrible circumstances. The older children were forced to stop their education the moment they reached the age of eighteen; the family had no money; they were constantly on the move, often separated in two different places because the apartments and rooms they were able to get from churches and other organizations were often too small for five people. It was very difficult for her to talk and she suffered from serious symptoms of depression and post-traumatic stress disorder, but she insisted on continuing the interviews. She told me that however painful it was that all the memories came back, even now―in a period when she could hardly handle her daily problems―she felt it as a relief that at least one person would listen to her story. But at the same time she also showed a powerful side of herself, telling me how she tried to support her children and husband, how she tried to play with the children and to make jokes together, and how she tried to make their lives better. I was very impressed by her strength.

During the analyses, however, her grief, pain, troubles and despair became dominant in my mind and colored my interpretation. In an attempt to fully understand how such a life must be for all the family members, I became more and more angry and ashamed that this could happen in the Netherlands. Sometimes I could not continue working on this case; I just became sad and angry. Those feelings blocked my ability to see the woman’s strength, her resilience, her persistence in getting a permit and her ability to support her husband and children, all things I had noticed during the interview. In this way my personal feelings that became dominant blocked my ability to see the ‘whole woman’ and to see her agency and moral responsibility. But it also kept me away from my own moral drive and my responsibility to present her with all her strength and weakness in the whole context. I did not know any more how this research could contribute to gaining more understanding and support for these women after the terrible experiences they had endured and could contribute to ensuring better health care for women in similar situations.

As soon as I began to make presentations of my research within the context of the Dutch and European asylum policy, I was able to recover what had been lost in the process of analyzing. This helped me to overcome those feelings which my colleague had called a “crisis of conscience”.


Doing fieldwork means trying to place oneself into the world of the respondents. Empathy, transference and compassion are used to bridge the gap between me and the other, while during the phase of analysis the world of the informants enters into my inner mental space. The process is twofold: on the one hand to bridge the gap, and at the same time to create a clear distinction between me and the other. Understanding the other is not only a cognitive process of intellectual interpretation; it is lived through all the senses and is, as Merleau-Ponty calls it, intercorporality. Re-experiencing the other is not a process of reading but of feeling. As anthropologists we have to work through the specificity of our own self in order to contextualize and transcend the data.

We have to try not to intermix the self with the data without being aware of it, and we also have to stay away from the tendency to split or exclude ourselves, a tendency that, according to Okely results from the “canonical stress on ‘neutral, impersonal, scientific work’” (Okely 1992 in Clough 2006: 261). Okely states that the reflexive knowledge of fieldwork is acquired “not only from examination of the outside categories, but also from more intangible inner experience” (ibid.). Therefore it is important to understand the inner experience of the self, how it interacts and intertwines with the data, what makes it meaningful, and what of the lived-through intersubjective experiences can be lost in the process of transforming the interview into written words. In my case I lost the strength of the women and my own moral responsibility. To recover it and―even better―to keep it requires a constant movement back and forth in ethnographic investigation between experience, consciousness, memories and reflections, not only with the data of the narratives but also with what was created in the intersubjective space during the interviews.

For the anthropological production of knowledge, it is important to realize why a particular research field was chosen, and to understand what for me as anthropologist is important to make it meaningful and bearable. The special engagement with the topic of research is at the same time the strength and the weakness of the research. Without such insight into what it can give and, as in my case, what can be lost in the process after the interviews, the research can become too much a personal burden and can obstruct the creation of a ‘fresh’ analytical discourse or the ability to do right by the data.

Churchill notes that psychoanalysts are psychoanalyzed as part of their training, as an attempt to make them aware of their own unconscious problems and tendencies so that they can be fully aware of their own subjective response to the analysand and keep this out of his analysis. Anthropologists do not have this kind of intensive training, but I would plead for a kind of supervision, either in groups or individually, that helps the researcher to become aware of his/her own motives, needs and processes. With this insight the anthropologist can better maintain the mental distance in which this constant moving back and forth with his/her data is possible. This is a necessity: in the first place, in order to stay emotionally in balance; in the second place, in order to be able to analyze one’s own position in the field and in the process; and finally, because the anthropological product is a scientific approach and interpretation. Theory has to be developed to explicate data and not used as an attempt to avoid data that are painful and tough to digest.

Anthropological research and writing will always be a craft. We will all create different stories from the same data. Anthropological writing is always a reformulation of reality. This is not a problem as long as the data of the informants is the material that is used. No human being can fully enter the consciousness of another and thereby definitively know the other’s meanings and motivations. To speak along with Geertz: it is the ‘said’ of speaking that we understand and that does not correspond to meaning, as ‘meaning’ is not said (1993: 19). We can never grasp the inner self of another fully enough to know what is in another actor’s mind when he or she verbally indicates an object or an action. This is also not necessary, since the task of anthropology is to identify general patterns. But I hope I have shown that insight into our own processes, emotions and needs will bring anthropological research to a higher level. Intersubjectivity and analyzing one’s own contribution to the intersubjective space is an anthropological tool for obtaining this insight, not only during fieldwork but also when the informants have become part of one’s own mind during the analysis.

In closing, doing research in the field of violence is not only difficult and painful. My own introspective research and all the moments of intersubjectivity with the women in my research also gave me feelings of intense richness. I would never have wanted to miss them.


I would like to thank Annemiek Richters, Margreet Peutz, Rachel Spronk, the members of the Reading group ‘Trauma, memory, violence and narrative’ and the participants of the symposium ‘Intersubjectivity as an analytic tool’ for their fruitful comments.

About the Author

Marian Tankink is a former community psychiatric nurse and a medical anthropologist. She is working on her dissertation (see below) at the Leiden University Medical Center, Department of Public Health &, Primary Care, Section Medical Anthropology. E-mail:


Churchill, Christian J.
2005    Ethnography as translation. Qualitative Sociology 28: 3-24.
Geertz, Clifford
1993    The interpretation of cultures. London: Fontana Press.
Gobodo-Madikizela, Pumla
2004    A human being died that night: A South African story of forgiveness. Claremont: David Philip Publishers.
Jackson, Michael
1998    Minima ethnographica: Intersubjectivity and the anthropological project. Chicago: University of Chicago Press.
Lansen, Johan, & Ton Haans
2004    Clinical supervision for trauma therapists. In: John P. Wilson and Boris Droždek (Eds) Broken spirits: The treatment of traumatized asylum seekers, refugees, war and torture victims. New York: Brunner-Routledge, pp. 317-53.
Mattingly, Cheryl & Linda C. Garro
2000    Narrative as construct and construction. In: Cheryl Mattingly & Linda C. Garro (eds) Narrative and the cultural construction of illness and healing. Berkeley University of California Press, pp. 1-49.
Okely, Judith
1992    Anthropology and autobiography: Participating experience and embodied knowledge. In: Judith Okely & Helen Callaway (eds) Anthropology and autobiography. ASA Monographs 29. London: Routledge, pp. 1-28.
Richters, Annemiek, Marian Tankink, Hishamah Bel Khodja, Janus Oomen & Marianne Cense
2007    Face to face with refugee women's distress in the Netherlands: Distance and proximity in the aftermath of sexual violence. In: Els van Dongen & Ruth Kutalek (eds) Facing distress. Distance and proximity in times of illness. Vienna: Lit-Verlag, pp. 63-84.
Tankink, Marian, & Marianne Vysma
2006    The intersubjective as analytic tool in medical anthropology. Medische Antropologie 18 (1): 249-65.