A last methodological issue to be considered is how it is possible to read ‘feelings’ out of research interviews. This question is especially pertinent when the theoretical perspective on feelings draws on psychoanalytic theory, which is developed for and validated in a clinical setting. An in-depth study of emotional meanings is admittedly hard to accomplish in empirical research outside of clinical practice. This has been a target of much debate in the field of psychosocial studies, where some argue that it is not only a possible but also a vital resource for social researchers to make use of transference/countertransference processes in the reading of the informants’ unconscious processes, while others have been critical to this (for the discussion on this, see, for instance, Walkerdine 1997; Hollway and Jefferson 2000; Bornat 2004; Frosh and Emerson 2005; Roseneil 2006; Frosh and Baraitser 2008; Hollway 2015; Woodward 2015). My position on this is that using psychoanalytic perspectives in social research does not imply that the researcher should relate to their informants as psychoanalysts to their patients. One issue to consider is that even though qualitative interviews as compared to surveys, experiments and tests may be seen as rich materials and ‘thick descriptions’, they are indeed very ‘meagre’, as Chodorow (1999) puts it, when compared to clinical data that stems from often many years of clinical encounters between the patient and the analyst. Another issue is that most social researchers lack proper training in handling and interpreting what goes on in the transference in the interview situation, so the danger of ‘wild analysis’ cannot be excluded. A third issue is that the analyses of interview transcripts are not (and in my opinion should not be) validated in an interpersonal space that includes the informant. What social researchers basically work with are texts, and texts are not persons (and definitely not when analysing interviews that are 25 years old).
However, when psychosocial methods are implied in social research, it is not primarily the psychobiographical specificity of individual stories but the social patterns to which they speak that is the target (Roseneil 2006: 848). In my case, the whole-sample analysis does not work on the level of individual psychobiographies, but on the level of shared psychological patterns. It is not the defended subjects I am aiming at, but rather the patterns of defence across individual cases, no matter how varied the individual trajectories to this pattern were. Furthermore, the concept of the unconscious I work with is not only or even primarily understood with reference to the dynamically repressed, but rather as a part of the self that organises experience and informs creativity and agency on an emotional level (see Chap. 2). For this reason I relate to my interview transcripts not as persons, but as texts. I regard the interviews as a corpus of text in which I try to find the emotional meaning as an integrated part of what is said, directly or indirectly. Just as narrative perspectives open up ‘outwards’ to social change, they also open up ‘inwards’ through conveying the emotional meanings attached to such change (Rudberg and Nielsen 2011).
It is possible to work with a psychoanalytic ontology (Hollway and Jefferson 2000; Roseneil 2006) and to employ a context-sensitive textual analysis and a psychologically sensitive reading that acknowledge that an emotional meaning is always present in what is said, without analysing individual unconscious conflicts. From the way the informants talk and the words they use, it is possible to learn something about their personal images of parents and thus to gain some insight into how also emotional aspects of meanings of gender are part of a person’s biography. I have found the psychoanalytic concepts of identification, disidentification, projection, disavowal, idealisation and ambivalence (see Chap. 2) useful as ways to interpret what the informants say about their relationships with others and their own bodies. They may work as ‘small pockets of feelings’ in the text. What people say about their feelings is sometimes relatively straightforward, while at other times it may be more hidden or ambiguous. The relation between utterance and textual context is an important cue here. If an informant says ‘I had a good relationship with my mother’ and then proceeds to describe her relationship with her father with much more enthusiasm and detail, then this textual context lends significance to the claim. Another informant could utter exactly the same phrase, but the context of the interview could give it a different meaning compared to the first case. In addition, non-verbal cues like voice, tearfulness, pauses, humorous comments, slips of the tongue or simply the intensity of the tone can be important contextual cues, which are preserved in transcripts (Nielsen 2003). Feelings may also reveal themselves from the textual organisation: what are the contradictions or difficult points in the stories, when do they not make sense? What model of the world and of self-other relationships makes itself known in the way in which the informants talk? In looking at interviews as textual structures with conscious and unconscious layers of meaning, I follow the insight of Ricouer that both meaning (content as structure) and significance (content as reference to the world) are vital parts of the interpretation of texts. Life, Paul Ricoeur says, can be seen as ‘an activity and a passion in search of a narrative’ (Ricoeur 1991b: 29). When we construct a narrative of our life, we produce a textual structure that ‘underscore(s) the mixture of acting and suffering which constitutes the very fabric of life. It is this mixture which the narrative attempts to imitate in a creative way’ (Ricoeur 1991b: 28). When we tell a story of our lives, the point is not only to make our lives more intelligible, but also more bearable. We can make ourselves heroes of our own story— however, we cannot actually become the author of our own life. Thus, even though a narrative strives towards homogeneity, it will always be a synthesis of the heterogeneous—a structure of ‘discordant concordance’ (Ricoeur 1991b: 31).
Valerie Walkerdine (1997) and others have argued that there are also processes of transference in reading texts and that the employment of the researcher’s subjectivity in the process of interpretation is indispensable and should not be seen as a source of error. This concurs with Paul Ricouer’s view that the indecisiveness of textual meaning first turns into social communication (the significance of the text) when it meets its reader (Ricoeur 1991a: 63). However, in research it is problematic if the significance that emerges says more about the reader than about the text (see also Bornat 2004). It is therefore important to be able to document the interpretation on the textual level: interpreting the emotional layers of the text is not reading ‘between the lines’, but a broader interpretation of what is ‘in the lines’ than just reading the words in a very literal way. The German psychoanalyst Alfred Lorenzer and his colleagues, working in the field of cultural studies in the tradition of the late Frankfurt
School, have developed a method for ‘deep-hermeneutic cultural interpretation’ of texts (Lorenzer 1986). The point here is not to analyse the person behind the text, but to understand how the unconscious structures in the text reflect the world we live in. Lorenzer does not reduce the manifest meaning of a text to a latent meaning. While the latent meaning of a dream will often require additional information from the dreamer, the latent meaning of a written text has to be in the text, otherwise it makes little sense. The connecting point between the conscious and the unconscious is found in the symbolic accounts themselves. According to Lorenzer, in order to find the unconscious structure in a text, we have to read the text as openly as an analyst should listen to the talk of a patient. We look at the spots where it ‘irritates’ (Prokop 1996)—where something ‘does not fit’ or seems to be missing, where the text becomes contradictory or maybe too coherent, or where the rhetoric is experienced as ambiguous, touching or untrustworthy. It may have to do with content, but more often with form—selection, ordering, rhetorical figures and textual images. My focus of the generational patterns in feelings may actually have more in common with the way in which Lorenzer analyses the conscious and an unconscious structure of meaning in cultural texts than with psychoanalysis of individuals. According to Lorenzer, the purpose of drawing attention to the unconscious structures in symbolic forms is not to understand the specific biography of the narrator, as it would be in therapy, but to understand how unconscious desires and cultural activity are intertwined. While the psychoanalytic interpretation in therapy aims at changing the analysand, the deep hermeneutic cultural interpretation, by drawing attention to hidden ‘life sketches’, aims at changing the analyst (or the readers). Lorenzer calls this the distinction between ‘analysing the production’ and ‘analysing the effects’ of the text (Lorenzer 1986: 68). Through an autobiographical text, we may gain insight into how a narrator can contribute to the change of culture through the impact his or her symbolic behaviour—stories told or practices engaged in—has on others. In my case, looking for the feelings of gender has exactly this aim: to see how feelings of gender have been used as a creative potential of change.