Research as Therapy Research as Therapy: Measuring effectiveness of Men Against Violence workshops.

Kay O'Connor, Counsellor, Massey University College of Education

Jon Patrick, Professor, Information Systems, Massey University

Abstract: This paper describes our current study in which we are analysing language change as a measure of effectiveness of therapeutic intervention with men who have been perpetrators of domestic violence. We include partners in the study to involve the family system in establishing new ways of relating. We interview workshop participants and their partners before, immediately after and three months after the workshops. Transcripts of interviews with ten workshop participants and their partners (twenty in total) are analysed to assess shifts in (1) content, (2) imagery and (3) linguistic usages that effect distancing or specificity from the subject matter. In planning the method, we have been concerned with the ethical issues implicated in researching therapy, for example informed consent, timing of interviews, access to interview tapes and transcripts, involvement of partners, availability of ongoing counselling for all participants.

The plan

From November 1995, men who have completed the Stage 1 "Stopping Violence" workshop (29 hours) with Manawatu Men Against Violence have been invited to take part in a voluntary Stage 2 "Healing your Violence History" workshop (22 hours) which is facilitated by Jon. This workshop focuses on patterns of violence from families of origin. We set out to measure the effectiveness of this workshop, in ways which would gather more than self report data. We decided to interview partners as well as the participants, partly to increase validity by triangulation (Lather, 1986). The objects of analysis are the interview texts. We track change in language use as well as in content, and since the speech patterns we are interested in are outside consciousness, these aspects of our analysis escape the limitations of self-report. That was our plan. What happened makes a much more interesting story.

Epistemological positions

Research has traditionally positioned the researcher in a neutral observer position, "the view from nowhere". Pare (1995) tells this as a tale of three moves. In positivism, the aim was to discover some objective truth in the observed world as object. This perspective placed the researcher or therapist as expert and the people and their lives as passive objects of study. The second move was the shift to constructivism, where the observed person is the subject with their own view of the world. Meaning is made by and within individuals. Qualitative research and client centred therapy methods were devised to allow access into that subjective experience. The third move is to the focus on the spaces between the subject and object in the intersubjective domain. In this, the social construction approach, meanings are made in community, and the activities of therapy and research are collaboration between people (Gergen,1985; Hoffman, 1990; Weingarten,1991).

Our position(s)

We take various positions within this enterprise and shift around the three positions Pare (1995) describes. We take the first position as counsellors, who lead workshops where we are directive about process, and we are researchers with expert knowledge about how we are going to analyse data and interpret it. In the second position we are fully respectful of each subject/client's unique experience of the world at their own crossroads of historical and cultural experiences and we are respectful of their interpretations of those experiences. The third position enables us room to exercise our philosophies of working and relating, and to invite the participants in the workshops and research to collaborate with us in our exploration of the unfolding stories of their lives.

In moving between these three positions there is a dynamic and spontaneity of transitions that is not necessarily controllable nor needs to be. Debates between us, the researchers, can instantly switch to conversations between us the counsellors, to reflections of us the collaborator-participants. With this admixture of roles and responsibilities we have found that we can only perceive _Research as Therapy_ and work to make one as full and complete as the other, motivated first and foremost by our desire to do the best possible job of counselling. For these reasons we choose to tell you a tale of therapy and research as story making.


We believe that post modern understanding of what constitutes knowledge has disrupted scientific illusions of objectivity and neutrality. The act of observation alters the observed. Rather than attempt to control interview bias, we admit it, accept it and utilise it. We turn the impossibility of neutrality into advantages of partiality. This is good news, since partiality emerges as a strength in research as much as in therapy. We like the narrative metaphor. While we do not claim to be 'narrative therapists', we find narrative theory useful in describing how we work. In the workshops we use time lines and perceptual position exercises which are designed to revisit and reconstrue the past and to clarify positive outcomes and resources for the future. New readings of past, present and future are made available. In like manner Michael White and David Epston explain narratives:

In striving to make sense of life, persons face the task of arranging their experiences of events in sequences across time in such a way as to arrive at a coherent account of themselves and the world around them. This account can be referred to as a story or self-narrative. The success of this storying of experience provides persons with a sense of continuity and meaning in their lives. and this is relied upon for the ordering of daily lives and for the interpretation of further experiences. (White and Epston, 1990; p.10 ).


The narrative metaphor frames both psychology practices of therapy and research as collaborative enterprises. We cannot abdicate all of our expert power as therapists nor as researchers. We need to be aware of privilege and power in these activities. We are two people who have not experienced physical violence in our families; we are psychologists, we are counsellors, we are researchers. All these circumstances distance us from the people we work with. This distance has advantages, in that we recognise that the clients/participants are the experts in their own lives and in their stories in and around violence. We therefore regard them as collaborators with us in the process of therapy and of research.

Not knowing

We come from a position of not knowing, and listen for what we do not know. We are curious. (Anderson & Goolishian, 1992; Hoffman 1991). What is the meaning of the story for these people is not our expertise but theirs. We do not seek to uncover any objective truth, but to co-create meaning. As we listen, encourage, reflect, paraphrase, summarise, we influence the story and co-author whether that is our intention or not. We yielded to the inevitability of influence. We cannot turn off or prevent the possibility that narrating one's story in the interview can contribute to the change process we have all decided will be the plot we are unfolding, the desired outcomes of change. In the interviews and in the workshop, the stories are re-told in words. In participants' lives, the stories are performed as texts in action, and it is in performance that the stories are transformative (Bruner, 1986).


A narrative postmodern approach challenges the notion of a unitary consistent self. Rather than 'self', a person is constituted by fragmented constantly constructed subjectivities (Weingarten, 1991); subjectivity is process in the space between people, not a thing inside the individual as container (Kvale, 1992). How we act out constructions of multiple subjectivities varies in different contexts and in similar contexts at different times. Multiple subjectivities are constituted in dynamic ongoing processes in talk and interaction with others (Davies 1989; 1993). Traditional methods of 'treating violence' operate on dominant stories of an internal world, an essential self, and people who struggle with violence usually see violence as part of their essential identity, a raging monster inside their skins. We prefer postmodern narrative views of subjectivities because they enable possibilities for reinterpretation and liberation from a constrained essential and unchangeable self.


We take up the narrative metaphor "externalising", in this case externalising violence along with other unwanted ways of living. This is not to reduce responsibility, but rather to open up a distance which enables change, a different way of relating to past stories in participants' family of origin and subsequent families. When dominant narratives of controlling anger and controlling violence are internalised, choices are rigidly limited. When violence is internalised, it has to be controlled. (White and Epston, 1990). The irony of this metaphor of control is that people who choose to violate tend to overemphasise control already. The language that we use to tell stories can limit us.


Language constitutes what is spoken, it does not represent any objective reality (Gergen, 1985). Language does not mirror nature, but creates the 'natures' we know (Anderson & Goolishian, 1988). What we call 'reality' is constituted in our speaking it into being. Speaking is not neutral or passive, and rather brings forth what is then regarded as having substance. Narrative gives the illusion of logic, reasoning, cause and effect (Gergen, 1985,) and is not representational but performative. Language is a social activity, done between/among people, and is about the space between people even when it purports to be about 'inner realities'. The descriptions of life that are available are not reflections or representations of what is, but rather shape and constrain and enable what is. So we take up the saying that the person is not the problem, but the violence is the problem. Violence is thereby reified, externalised, reduced to an option which has lost the illusion of power over the person.

Dominant narratives of violence

The dominant narratives of domestic violence in New Zealand at present cast men as perpetrators and women as victims. We prefer discontinuous, particular, local stories that allow multiple possibilities for all the people involved. Such stories enable lives that escape the subjugating narratives of divisive gender relations which constrain subject positions for both men and women.

The workshop

The Healing Your History of Violence workshop is based on a range of beliefs and values including the three following:

1. We are positively opposed to increasing degree of control. We have found that people who have used violence tend to already emphasise control as a way of life. Increasing control we see to be unhelpful to them.

2. We believe that before we can create a future where to violate is not a preferred option, the story of the past needs to be rewritten. Present patterns follow past patterns. If the past story of family of origin is re-told with new patterns, possibilities are opened for a new resourceful story.

3. We aim to open opportunities for the people we work with to constitute new subjectivities, a different 'identity' which is positive. Participants tend to frame this goal at first in terms of their relatedness - father, wife, worker. Workshop activities make room for rewriting the 'self', constructing a preferred 'identity' which is 'I' as a person - e.g. relaxed, competent, happy etc., which is more 'useful', that is which enables more choices.

The research

As part of the workshop, Jon interviews participants before, after and four months later. The partners of these men are invited to participate in the study. If these women agree, Kay interviews them before, after and four months later. The women are interviewed after the men in each case. Jon does not see or hear the women's material until the second interviews are completed. As the men's therapist, his aim is to understand the men, their own experiences without the view through the partners' filters. Participants in the final pool are ten men and their partners who agreed to the interviews at each stage. Before the interviews, participants are given information sheets and signed consent forms. Interviews are audiotaped and copies of tapes and interview transcripts are returned to participants before the following interview. Interviews are semi structured, eliciting material about the man's relationship with himself, with his partner and with the children in the family, in the past and in the present and also his expectations and hopes of the workshop outcomes and change in the future.

Interview bias:

Since we cannot have a neutral, objective interview, we choose to make the inevitable bias work for us. We reframe, we celebrate 'news of difference' (Bateson, 1972) we seek 'unique outcomes' (White & Epston, 1990) and we are biased in favour of enabling more choices. The men and women we speak to are the privileged authors of their own stories, and we co-author only little bits. We want these bits to open up possibilities and to enable transformative stories - in performance. We come back and interview again to revitalise that story line and emphasise what can make life worth living rather than how it is routinely lived (Rosaldo, 1986). The participants are able to involve us in their story telling and story making in the time that we are allied with them. The semi structured interviews that result are in accord with our research intentions, since we want texts of 'natural language' to analyse. We interview three times. Before the workshops, after the workshops (about six weeks later), and then four months on. In this way we are involved in the ongoing story and intend to put a twist in the tale.

Before Interview (Men):

The first interview aims to establish rapport for the workshop. Jon as researcher is relating to man as therapist. The therapist role takes precedence in consideration of the client/participant's wellbeing. In this interview Jon works with each workshop participant to create positive outcomes as goals for the workshop. This is not so much to establish a before-after comparison but rather a therapeutic future pacing. This initial conversation is inevitably intruding into the process of the intervention and any before after comparison we do make cannot be separated from the process of the interview.

After interview (Men).

We return the transcripts of the before interviews and copies of the audiotapes, and interview after the participant has read/heard the before interview. This brings to the forefront of the story the hopes and expectations of the workshop, and brings the focus onto the changes that have taken place. The story line of change then precedes the disappointment story of no-change - and both story lines are discernible. We are deliberately partial and biased in favour of the change story and ensure it is retold. Jon encourages workshop participants to re-tell their story allowing the opportunity for the new version set up in the first interview and established in the workshop. In this way the plot thickens (Geertz, 1978). Reality is created in and through language, and interpersonal relating constituted in how we speak it. So in interviews we enable that speaking to constitute a positive, choice filled reality. We use the workshop information to stimulate those positive stories as well as to work through those issues surfaced in the workshop. Evolving meanings emerge from interactions among people (Hoffman, 1990).

Before and After - Women

We wanted to interview partners of the men in the workshop for the following reasons. As researchers, we wanted to check the men's stories of change by including the partners' points of view. We were also curious as to the changes the women might write into their lives to parallel the changes the men talked about, both in what they say and in how they say it. As counsellors we wanted to follow family therapy principles by involving both partners in the workshop goals and evaluation. And so Jon asked the men to ask their partners if they would be willing to participate in the research. When this was agreed, Kay contacted the women and checked their agreement.

I certainly didn't expect any possibility of 'neutrality'. Before the study I had concerns that I would align myself alongside the women to a degree that might conflict with workshop. We clarified that my contact would be only with the women, and that they are my priority in our interconnectedness. Ethically, they become in effect my clients. As I expected I did validate the women's reported experiences by being involved in their stories. However my bias became quickly more complex than that. Because of my involvement in planning the workshop and in talking through the group processes, I also align myself to an optimistic view of the family as a group with hope.

My second interview is deliberately informed by the man's story as told in the first interview tape and Jon's stories about the workshop. I am free to do this - I am not the workshop therapist. As I read and reread the texts of the men's stories, I align myself with their desire for change. I have lost the sight of these two as separate individuals - his story, her story about her and her story about him are no longer separate individual stories and the interweave of these stories brings together the individual threads into a whole. The whole pattern comes to the forefront of my visual field. Her story is one thread of the pattern in the fabric. Other threads include the plot and the characters as they emerge in the content of their stories, and also our detailed familiarity of how each of these two use language to constitute themselves and their worlds. By now any notions of neutrality have long gone, and before-after measures become a co-written story of change.

Follow-up interview.

The purpose of the third interview is to explore the durability of the changes reported in the workshop and in the second interviews. In narrative terms, are the re-told stories transformative in performance. The follow up interviews are informed by thorough familiarity with first second interviews of both partners and the workshop. In the meantime, we have both analysed the transcripts of the mens' and womens' before and after interviews for content and language use. We have discussed our readings of possible story lines in these texts. This third conversation with the men and their partners is explicitly affirming of change, and where appropriate we emphasise the unique outcomes to further 'thicken the plot'.

The format we use has strengths. Including partners in workshop expectations and following up change has been worthwhile. However, only the men have the workshop experience. Our approach to domestic violence emphasises the formative experiences of family of origin, not just for the men, but for the women. The imbalance of the intervention bothers us. So we set up a workshop for couples also titled "Healing your history of violence" and advertised in the community. We have so far run one of these.

Couples workshop

We used the same format for this workshop as Jon uses in the workshops with MMAV. There are two major advantages. First, both partners were present to hear and at times collaborate in telling each other's stories, and in experiencing the workshop. Secondly, both partners in each case identified themselves as involved in stories of violence in their families of origin and as bringing those past patterns into their present family. There were also disadvantages. These participants were not as self disclosing or mutually supportive as in the men only workshop. Whether the caution was a result of the presence of partners or the presence of both sexes we do not know. However working with a mixed group as co-leaders was a positive experience, and we intend to develop other formats which build on the advantages and remove the disadvantages of this trial. For these interviews, neither of us accessed the other's interview materials until after the second interview, when we reached the analysis stage.


Our research focus is on language use. The transcripts of interviews are the objects of study. We are exploring two aspects of the texts, first the use of imagery (visual, auditory, kinesthetic, olfactory, and gustatory), and second specific speech patterns (deletion, generalisation and distortion). We are particularly interested in changes in language use that occur over the time we are working with these people. We believe that how people use language constitutes their ongoing experiences, and that therefore language use is informative about the people who speak.


We have become increasingly confident in some of our hunches about violence in intimate relating.

First, to focus on behaviour change without attending to habits and patterns from family of origin is wasteful of therapy time. This perception we share from our counselling work is borne out in these workshops. Interpersonal relating where people violate is intergenerational and occurs among siblings as well as between parents and children and between partners.

Secondly, encouraging control of violence (and also of anger) is not helpful to enable liberated futures. On the contrary, control has become over emphasised in the lives of those who use violent ways of relating. We will continue to disrupt that view of relating in order to create space for relaxed, spontaneous being.

Thirdly, the simple perpetrator victim narrative is not helpful. The story is much richer and more intricately choreographed than that. Violence is also involved in relating to the environment, animals and objects. We do not believe it is useful for anyone involved to focus exclusively on violence of men towards women. Our social discourses of gender relations disable men as well as women; power inequity disables everyone.

Finally we conclude (at this stage - work in progress!) to work with men alone in respondent programmes is limited compared to involving both. Our current plan is to co-lead parallel workshops for partners, men and women separately, and then to bring them together. The interview process is so valuable we will continue it as therapeutic process regardless of research. Detailed familiarity with speech patterns and content is enriching to the therapeutic process. The dual relationship of therapist and researcher has enabled us as practitioners to review our practice and to be more aware of the therapeutic process. We have become careful to negotiate and contract with the people we work with in counselling and in research. And we have become involved in the story making process in ways that have delighted and energised us.

This paper is a synopsis of the story so far. The plot of our enterprise is still unfolding, and we are caught up in subplots and themes. In time the story will unfold, and we will make coherent narratives of our work to tell you about in the future.


Anderson, H. & Goolishian, H. (1988). Human systems as linguistic systems: Preliminary and evolving ideas about the implications for clinical theory. Family Process, 27, 371-393.

Anderson, H. & Goolishian, H. (1992). The client is the expert: A not-knowing approach to therapy. In S McNamee & K. Gergen (Eds.), Therapy as social construction. Newbury Park, CA: Sage.

Davies, B. (1989). Frogs and snails and feminist tales : preschool children and gender. Sydney : Allen & Unwin.

Davies, B. (1993). Shards of glass: children reading and writing beyond gendered identities. St Leonards, N.S.W. : Allen & Unwin.

Epston, D. !993). Internalising discourses versus externalising discourses. In S.Gilligan & R. Price (Eds.). Therapeutic conversations (pp. 161-177). New York: Norton.

Freedman J. & Combs, G. (1996). Narrative therapy: The social construction of preferred realities. New York: W.W. Norton & Company.

Geertz, C. (1978). The interpretation of cultures. New York: Basic Books.

Gergen, K. (1985). The social constructionist movement in modern psychology. American Psychologist, 40, 266-275.

Hoffman, L. (1990). Constructing realities: An art of lenses. Family Process, 29, 1-12.

Kvale, S. (Ed.). (1992). Psychology and postmodernism. London: Sage.

Lather, P. (1986). Issues of validity in openly ideological research: Between a rock and a hard place. Interchange, 17, 63-84.

Pare, D. A. (1995). Of families and other cultures: The shifting paradigm of family therapy. Family Process, 34, 1-19.

Rosaldo, R. (1986). Ilongot hunting as story and experience. In V.Turner & E. Bruner (Eds.), The anthropology of experience. Chicago: University of Chicago Press.

Sampson, E. E. (1985). The decentralization of identity: Towards a revised concept of personal and social order. American Psychologist, 40, 1203-1211.

Weingarten, K. (1991). The discourses of intimacy: Adding a social constructionist and feminist view. Family process, 30, 285-305.

White, M. & Epston, D. (1990). Narrative means to therapeutic ends. New York: Norton.

This paper was presented at the Psychological Society Conference in Christchurch in August, 1996