3 summary articles in 4 hrs EMERGENCY

Multi-agency collaboration, multiple levels of meaning: social constructionism and the CMM model as tools to further our understanding

Gill Salmona and Jeff Farisb

This study explores the discourse emerging when professionals from a child and adolescent mental health service meet with professionals from other agencies to discuss cases. The study is timely, given the current political contextual forces pushing agencies to work together which run alongside an expanding literature acknowledging the obstacles to achiev- ing this.

A thematic analysis identified nine themes, defined according to their discourse type, including single agency discourse, case complexity dis- course and multi-agency discourse. In this paper, the usefulness of the coordinated management of meaning model (CMM) is examined as an additional tool which may be used in data analysis to help understand the discourse within multi-agency meetings. The two approaches to data analysis are complementary to each other, with both allowing for different layers of context and complexity to emerge from the data.

Introduction

The majority of documents relating to the development of services for children and families including child and adolescent mental health services (CAMHS) published in recent years highlight the importance of multi-agency collaboration (e.g. NHS HAS, 1995; Department of Health, 1997; House of Commons Health Select Committee, 1997; Audit Commission, 1999). It is the view of Miller and Ahmad (2000) that multi-agency collaboration is an efficient way to deliver high- quality services and to ensure their effectiveness in being responsive to service user needs.

r The Association for Family Therapy 2006. Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA. Journal of Family Therapy (2006) 28: 272–292 0163-4445 (print); 1467-6427 (online)

a Senior Lecturer in Child and Adolescent Mental Health, Welsh Institute for Health and Social Care, University of Glamorgan, Glyntaff Campus, Pontypridd, South Wales CF37 1DL. E-mail: [email protected]

b Senior Lecturer in Psychotherapy, The Family Institute, School of Care Science, University of Glamorgan and Consultant Psychotherapist in Private Practice.

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

The Children’s National Service Framework (NSF) in England also recognizes that multi-agency partnerships are essential to delivering coordinated services especially for children who are disabled, have mental health problems or who are otherwise in special circumstances (Department for Health, Department for Education and Skills, 2004). These ideas are progressed in the government Green Paper Every Child Matters (DfES, 2003), in which it is proposed that, in the long term, key services for children and young people are integrated as part of Children’s Trusts.

A literature describing successful attempts to form multi-agency collaborations in respect of services for children is beginning to emerge, although there is still little written about the more conceptual issues. Salmon (2004) provides a comprehensive overview of these issues in relation to children’s services with particular emphasis on CAMHS.

A number of authors (e.g. Kopser et al., 1994; Henneman et al., 1995; Marino and Kahnoski, 1998; Akhavin et al.,1999) suggest that effective communication is an important factor common to successful collaborations. Another factor which has been identified as pivotal to the success or failure of multi-agency collaborations is the presence or absence of a common language across agencies (Bruner et al., 1992; Ranade, 1998; Miller and Ahmad, 2000). In studying the difficulties encountered by social workers when attempting to support children in need, the Dartington Social Research Unit concluded that the communication between professionals working for different agencies was a major issue. In particular they noted that when professionals talk about children they may use the same words as each other but have a different understanding as to the underlying meaning (DSRU, 1998). In line with this, Bruner et al. (1992) recommend that at an early stage of collaboration, participating agencies explore terms, definitions and operational language, and come to a consensus on their meaning in the light of their shared goals.

While the importance of effective communication is recognized, what actually constitutes effective communication is a moot point. To consider communication across multi-agency contexts as a commodity that can simply be made more effective is to ignore different levels of context let alone assuming that all these contexts operate at the same level of logical type. Agendas at different levels of a network of services may compete for improving effectiveness of communication. For example, fund-holders and commissioners of services have agendas at the level of expenditure in relation to certain outcomes at a broader

Multi-agency collaboration, multiple levels of meaning 273

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

level than the case-specific agendas of clinicians. What counts as ‘effective’ may be as varied as the agendas. Construction of meaning is contingent upon management of multiple levels of context (Pearce and Cronen, 1980; Anderson and Goolishian, 1986; Cecchin and Fruggeri, 1986; Imber Black, 1986; Campbell et al., 1994).

These agendas operate at different levels of abstraction such that in one set of circumstances low waiting lists may result in cessation of funding while in another it may be considered a laudable target. In another example, empty beds in an inpatient CAMHS facility, result- ing from an effective preventive service strategy, may ultimately result in the closure of the facility because not enough beds have been used. This closure would effectively undermine the infrastructure of the preventive service set up to offset the necessity for hospital admissions in the first place.

Multi-agency collaboration and systems theory

When the concept of collaboration is viewed from the perspective of systems theory, the idea that the whole is more than the sum of its parts reinforces the value of collaborative practice. Briggs has pro- posed a model for viewing interdisciplinary collaboration which integrates systems theory. The model differs from others proposed in that it takes account of the complexity of teams functioning within the broad context of a community (Briggs, 1997a, 1997b, 1999). She states: ‘With one professional’s expertise, important information can be obtained. When many professionals representing different disci- plinary perspectives integrate their voices, the observations, under- standing and insights attained are richer and deeper’ (Briggs, 1999, p. 365). Systemically trained professionals attempting to work with children and families will be aware of the importance of seeing their clients as functioning within wider communities, social networks and professional systems. The larger policy and political issues surround- ing the services which a professional can provide and the interlocking and interrelated nature of the systems which provide services cannot be ignored (Briggs, 1999).

The central feature of Briggs’ (1999) systemic model for viewing interdisciplinary collaboration is that of professionals having a com- mon purpose. This is said to be derived from a sense of shared meaning or belonging, with mutual values, beliefs and goals. Accord- ing to Briggs (1999), shared meaning usually evolves gradually as professionals begin to recognize where their beliefs and biases overlap

274 Gill Salmon and Jeff Faris

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

with others. Individuals who work together in teams may formalize shared meanings or have a common purpose, whereas those who work separately may have to seek out other individuals who share beliefs, values and common practice methods. Professionals coming together across team or agency boundaries or from different depart- ments within a service will not necessarily have developed such a shared worldview. A key feature here is not so much that a group conform to the same set of ideas or beliefs but that they have evolved a baseline shared set of assumptions within which it becomes possible to manage differences of viewpoint including strongly held differences (Speed, 2004). Briggs (1999) comments on the importance of the concept of dialogue in deepening group discussions and allowing for collective thinking and a pool of common meaning to be created. Jargon, she states, often poses a significant barrier to effective communication between interdisciplinary groups of professionals. She suggests that each professional commit to reducing jargon and remember to translate terms and phrases that are unique to a particular discipline.

We consider that, even when terminology is clarified, terms of reference may still remain unclear. Here we wish to draw a distinction between denotative meaning as per a dictionary definition and connotative meaning which contains emotional response, social mean- ing and symbolic association. The importance of language and dialogue is, of course, also central to social constructionist theory which, alongside the concepts of constructivism and postmodernism, has lead to fundamental theoretical shifts in systems thinking over the past twenty years.

Social constructionism

Mary Gergen (2002) defines social constructionism as a metatheore- tical position that describes how our sense of ‘reality’, as we know it, is achieved. Social interaction of all kinds and particularly language is of great interest to social constructionists. The goings-on between people in the course of their everyday lives are seen as practices during which shared versions of knowledge are constructed. The possibility of alternative constructions of the self and others’ ‘events’ in one’s world, through language, is fundamental to this social constructionist view (Burr, 1995). Social constructionist premises include an emphasis on con- text and interpersonal processes in creating joint actions and mutually constructed meanings, strategic interaction, an acknowledgement

Multi-agency collaboration, multiple levels of meaning 275

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

of the importance of power and of the exchange of ideas or feedback (Dallos and Draper, 2000). Consistent with Briggs (1999), Anderson and Goolishan (1988, 1992) see meaning and understanding as socially constructed, and argue that they are not arrived at in the absence of a communicative action or a meaning-generating (dialogi- cal) conversation.

Coordinated management of meaning (CMM)

Developing on from the ideas of Bateson (1972) and social construc- tionist theory, Pearce and Cronen (1980) and Cronen et al. (1982a) proposed the coordinated management of meaning (CMM) model as a way of conceptualizing a hierarchical organization of meanings such that the more abstract contextualizes and defines the less abstract. Having found that people co-construct their own social realities, these authors offer a framework that attends to the general and specific contexts of meaning and behaviour.

The authors of CMM have endeavoured to construct a template for analysing the interaction between frames of reference or contexts (i.e. all those features of a situation that make a difference to participants in deciding between alternative courses of action or making alter- native interpretations). For example, Cronen and Pearce (1985) typically employ the following five embedded levels of context in relation to the analysis of families:

1 Speech acts; 2 Episodes; 3 Relationship; 4 Life-scripting; 5 Family myths.

Adaptations to the hierarchical framework have been suggested: for example, adding a sixth level of sociocultural norms (including the legal system and legislation) and by modifying the model to take account of socially constructed meanings in the life of a therapist as a professional as well as a private person (Lindsey, 1993). It may be useful or indeed highly relevant to include further levels such as economic, community and political contexts as well as ecology.

The number and nature of the levels of context is not fixed and there is an implicit hierarchy of organization of meanings whereby the more abstract levels of context will tend to define lower levels. Drawing upon Wittgenstein’s (1953) idea of a ‘language game’ and

276 Gill Salmon and Jeff Faris

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

Bateson’s (1972) idea of ‘punctuation’ as the ability to discriminate the start and end of an episode or a change from one episode to another, Cronen proposes that the episode is the basic unit of social analysis insofar as human beings organize activities into episodes or contexts with distinctive rules for understanding what is required or accepted action (Cronen, 1995).

An episodic emphasis is also a good correction for the tendency to carelessly generalize from particular kinds of episodes to the exclusion of others. This has particular significance for the study outlined in this paper. The counting of a number of acts or even the quantity of episodes within which these acts occur will not inform us of the import of such acts nor of their significance. It is the contextual relevance that is essential to understanding the reciprocal interaction between meaning and actions.

A social constructionist viewpoint on research

The phrase ‘empirical research’ means literally to be ‘guided by experience’ and thus the aim of good empirical research is to reflect the world as it is. Empirical scientists try to keep a dispassionate distance between themselves and their subjects so that they can experience the world directly and not allow biases to cloud their observations. Kenneth Gergen (1999) argues, however, that research is rarely carried out for no reason and the researcher’s biases will enter into the research at every point, from the words selected to frame the problem to the description of the subjects’ actions. We do not construct our interpretations in isolation but against a backdrop of shared understandings, practices, languages and so on (Schwandt, 2000).

Most social constructionist approaches treat dialogue and discourse as the central organizing principle of construction. For social con- structionists, interview data used in research must be interpreted carefully and tentatively, with a willingness to reflect upon the possible contributions of the researcher and the research community to the outcomes (Gergen, 2002). This concept of reflexivity requires re- searchers to become aware of their feelings and biases, their history, values and assumptions, and to scrutinize them closely (King, 1996). According to Gergen and Gergen (1991), the foremost feature of reflexivity is probably an attempt to redress the power inequalities between the researcher and the researched in order to construct meaning and to achieve an expansion of understanding.

Multi-agency collaboration, multiple levels of meaning 277

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

Social constructionism and the CMM model in particular as a potential framework for discourse analysis

CMM is an important theory of human communication for two reasons. First, it deals with an important social process: the way in which individuals achieve, through their interactions with each other, some degree of mutual understanding to reach a mutually desirable end. Second, work involving CMM theory has a notable record of success in research and practical application (Philipsen, 1995). CMM theory has been applied to a variety of different situations: family communication (Harris, 1980), organizational communication (Harris and Cronen, 1979) and intercultural communication (Stanback and Pearce, 1980). Caroline Lindsey (1993) has also adapted the CMM model to explore contexts which are significant to both professionals and families in the reconstruction of ‘the family’. Thus the CMM theory has shown itself to be a useful tool to highlight the complexities of human action and meets with accepted standards of empirical verification (Cronen et al., 1982b).

As a new field of research begins to open up in the arena of multi- agency collaboration, the authors hypothesize that ideas stemming from systems theory and practice and in particular social construc- tionism and the CMM model might be informative in the develop- ment of appropriate research methodologies. In this study, the authors sought to examine the quality of discourse in some episodes of inter-agency multi-professional dialogue using both a thematic analysis as well as the CMM model to see if these two approaches are complementary and additive to each other in furthering the understanding of multi-agency collaboration within a CAMHS con- text.

Methodology

This was a clinical qualitative research study whereby the first author (GS) set out to study ways of working within a CAMH team of which she was a part. The research aims were partly derived from the existing literature on multi-agency collaboration but, in line with qualitative research methodology, also evolved as the data were collected and analysed. The aims were:

1 To explore the discourse that emerges when professionals from a specialist CAMHS meet with professionals from other agencies to discuss cases;

278 Gill Salmon and Jeff Faris

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

2 To do a pilot study using CMM as a framework within social constructionist theory to help understand the discourse that emerges.

Data collection

At the time of the research, a number of professionals within the specialist CAMHS were offering consultation to groups or individuals from other agencies. In addition, alongside regular in-house team meetings, liaison meetings between the team and other agencies involved with children occurred which provided a forum for joint case discussion and planning. Ethics approval and participant consent was sought and granted to audio-tape, transcribe and further analyse the eight meetings listed:

1 Two consecutive consultation meetings with a small group of professionals who work with looked-after children;

2 Two separate consultation meetings about specific cases with social workers;

3 Two consecutive liaison meetings with teachers from a school; 4 Two ‘in-house’ meetings of the specialist CAMHS team: one during

which new referrals to the service were discussed, the other during which currently open cases were discussed.

Data analysis using the thematic analysis framework

Data were analysed initially using a thematic analysis framework. There are differing descriptions of thematic analysis in the literature but all have as an essential principle the identification of increasing levels of abstraction in the data (Leininger, 1985; Aronson, 1994; Gantley, 1999). The eight transcripts were first checked for accuracy and anonymity was maintained by removing identifying data. All data were then coded into categories. Categories were grouped into themes.

To increase the validity of the findings, a collaborative approach was taken whereby a clinical colleague of similar professional back- ground and training as well as a research consultant with a special interest in qualitative techniques looked in detail at one of the transcripts and met to discuss emerging categories with the researcher (GS). The clinical colleague then went on to examine and discuss the seven other transcripts with the researcher so that a consensus could

Multi-agency collaboration, multiple levels of meaning 279

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

be reached on emerging categories and themes (Van Manen, 1990). The data were finally coded with the aid of a qualitative computer package – Atlas-ti. The use of computer-assisted qualitative data analysis allows for the coding of the same data more than once, for example, at a process as well as a content level, and is said to increase validity of data findings and rigour in analytic methods.

Emerging themes

For the purposes of this study, we use Burr’s (1995, p. 8) definition of discourse:

A discourse refers to a set of meanings, metaphors, representations, images, stories, statements and so on that in some way together produce a particular version of events. . . . If we accept the view that a multitude of alternative events is potentially available through language, this means that surrounding any one object, event, person etc, there may be a variety of different discourses, each with a different story to tell about the object in question, a different way of representing it to the world.

Looking at the data as a whole it becomes apparent that one way of categorizing much of the discourse is in terms of the intent behind any particular utterance. For example, when a particular utterance was made, was the speaker trying to communicate information about case material to others in the meeting, was she either asking for or offering some clarification on an issue, was she trying to set an agenda for the meeting or intending to offer some input that could be construed as therapeutic?

Nine themes were identified and defined according to their discourse type. The themes and frequency of their appearance across transcripts are shown in Table 1.

While it is outside the remit of this paper to discuss all nine emerging themes in detail, the authors will elaborate briefly on two: discourse around intervention, and discourse around the individual, as these are directly relevant to the extracts of transcripts which were chosen to be examined further using the CMM model.

Discourse around intervention

When groups of professionals discuss case intervention, the discourse tends to reflect the perspective of the agency(ies) offering the service to the child/family.

280 Gill Salmon and Jeff Faris

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

The conversations between professionals about interventions fre- quently referred to the perceived lack of willingness shown by a family to take up offers of help. This can bring out in some professionals a tendency to try even harder to engage in helping activity which may be construed as ‘bending over backwards’, going ‘beyond the call of duty’ in their efforts to try to be helpful or may be construed as being flexible, tenacious and not giving up. A similar response in profes- sionals may occur when a child who has been struggling is now thought to be succeeding in some way. The interaction between participants in conversation and levels of context framing possible meaning will influence how such attempts to try harder may be construed.

Whenever professionals from different agencies or different disci- plines within the same agency gather, the different views emerging about interventions tend to be both about the nature of the interven- tion required for the child/family and which professional or agency is responsible for providing it, and how far the available resources can go.

Discourse around the individual

This theme groups together categories of discourse personal to individual professionals. There is some overlap between the categories appearing in this theme and those appearing in other

TABLE 1 Themes and frequency of appearance across transcripts

Themes emerging in the analysis

Frequency of appearance across transcripts

Single agency discourse 650 Discourse around intervention 453 Case complexity discourse 328 Multi-agency discourse 302 Inter-agency communication 177 Discourse around alternative viewpoints 173 Discourse around the individual 151 Social context discourse 54 Other discourse, e.g. � time 84 � change 65 � success 21 � pressure 11 � uncertainty 11 � crisis 5

Multi-agency collaboration, multiple levels of meaning 281

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

themes. Examples of this are the categories ‘staff roles and remits’ and ‘bending over backwards’. Thus a category may be found in more than one theme depending on the different levels of context used to explain the meaning.

Discourse of a more personal nature to the professional tends to be a feature of the smaller, more informal consultation meetings rather than larger, more businesslike multi-agency meetings. In the former context, professionals seem to be more willing to divulge information which gives clues to the influences upon them at the level of, for example, life and family scripts.

Data analysis using the CMM framework

To illustrate the use of the CMM hierarchical framework in greater detail, two sections of transcript from one of the meetings were selected for further examination. In the initial thematic analysis, both sections of transcript were coded under the category ‘bending over backwards’. This category was used to code discourse relating to a professional or carer appearing to try to do everything they possibly could to help in a situation, and in so doing, in the researchers’ opinion, going over and beyond that which might ordinarily be expected of them. The ‘bending over backwards’ category is a member of two themes, ‘discourse around intervention’ and ‘discourse around the individual’, reflecting the different contextual levels of meaning at which such actions may be understood. Examples from the ‘bending over backwards’ category are selected here, as, in the authors’ opinion, they clearly show how, when subjected to further detailed analysis using the CMM framework, utterances can have very different meanings depending on the context in which they are understood.

The sections of transcript chosen for further analysis are taken from a consultation meeting offered by CAMHS staff to a small group of professionals working with looked-after children. Participation in meetings such as this is thought by the authors to be representative of the type of multi-agency collaboration which occurs within a CAMHS context and is an important aspect of the work of a CAMHS team (Audit Commission, 1999). Note that identifying details have been removed in the following description of the analysis to preserve the anonymity of the professionals taking part in this consultation and the young person being discussed.

The extracts chosen centre on a discussion in which one of the consultees is describing the efforts made by the team and the foster

282 Gill Salmon and Jeff Faris

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

carers to try to help a young person on their pathway to indepen- dence. In line with this, the young person has a goal which requires additional funding if it is to be achieved. The proposal is that the cost will be shared between the foster carers and the responsible local authority, and that the young person should take a part-time job so that they can also contribute. While finances appear to have been agreed from the first two sources, the young person is not following through on the suggestion of the consultee and the foster carers to find a job. The consultee and foster carers thus appear to be ‘bending over backwards’ in their attempts to help the young person achieve the negotiated goal while the young person apparently does nothing to aid the process.

In line with the CMM hierarchical framework, the transcript of the consultation meeting from which these examples are taken was first examined to look for meaning within the context of the meeting. The analysis was then extended to look for possible meanings in the context of the outside world.

Meaning within the context of the meeting: Persons present at the meeting and negotiation around case consultation

This was the second consultation meeting between staff from this agency and the local CAMHS. Of the two consultees present, one had also attended the previous consultation, while for the other, this was their first attendance at the CAMH clinic. At the start of the meeting there was some discussion about the purpose of the meeting and how the time available could and should be used. Both consultees agreed to continue discussing a family initially mentioned in the previous consultation meeting. They agreed that on this occasion the focus of the discussion should be on a different sibling. Both consultees were familiar with the case although one was more involved than the other. Prior to one of the consultees touching on the aspects of the story described above, a considerable amount of time had already been devoted during the meeting to detailing the complex background to the case.

Influences on the consultee’s understanding of meaning

So how does it come about that the actions of the adults and the young person are understood in this way? Why is it, for example, that the

Multi-agency collaboration, multiple levels of meaning 283

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

adults in the story are understood to be ‘bending over backwards’ rather than their actions being interpreted, for example, as being flexible and accommodating? The CMM hierarchical model of con- text suggests that although an action may be given meaning at various levels, some levels of a context are higher and more important than others (i.e. an individual’s social reality is organized such that, for example, the individual’s life scripts are more powerful interpretive resources than the content-interpreted stimuli) (Pearce and Cronen, 1980). Thus, in the example shown, it is possible that the consultee is influenced by a personal and/or professional life script emphasizing the importance of putting oneself out and/or perhaps at a higher level of context, this is also the ethos of the agency. Support for the latter hypothesis comes from a quote found later on in the consultation meeting in which the consultee describes the people employed by the agency as being characterized by their willingness to ‘go the extra mile’.

In a similar way, the actions of the young person could be given meaning at various levels of context. The consultee is infuriated by what is perceived to be a lack of effort put in by the young person to achieve their goal, compared to the efforts it is felt that others are putting in on their behalf. The consultee seeks to understand the young person’s behaviour, but how this behaviour is interpreted as well as how the consultee views their relationship with the young person will be influenced by their own personal life script and experience of their families of origin and creation. In addition, as a professional, the consultee will be influenced by the practice of their team and the ethos of the employing agency and ultimately the constraints of their society and its legislation (Lindsey, 1993). Similar influences will be impacting upon others present at the meeting including the CAMHS consultants.

There are many ways in which the young person’s behaviour could be viewed. Perhaps the described behaviour could be seen as fairly ‘normal’ for adolescents who are often viewed as needing pushing to do anything. Or perhaps this is a young person who is taking everyone for a ride and one who should be grateful for the help they are getting, or perhaps one who has been damaged by past experiences and is now very insecure. Within the context of the meeting, the consultee later connects with their own personal experi- ence of parenting and chooses to understand the young person’s behaviour in the same light. In talking about their own family, the consultee may be responding to a regulative rule (guiding her actions)

284 Gill Salmon and Jeff Faris

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

that at a small consultation meeting between professional colleagues with the CAMHS, anecdotes from one’s personal or family life experience may be shared to express a view about the case. This may have been facilitated by the presence of another colleague who was also familiar with the case, as well as by the fact that the meeting was small and fairly informal.

Social reality, coordination of meaning, constitutive rules and regulative rules

Whatever meaning the consultee gives this young person’s behaviour, it will undoubtedly trigger a series of responses. The young person may perceive that the consultee and/or foster carer are talking to them as if they were a child when they would prefer to be related to as an adult. In response to this, perhaps the young person does not do what they are told when they are told to, as they would prefer to do it in their own time. As the young person is never left alone for long enough before a further episode of nagging occurs, they never get around to doing anything in their own time. All communications from the consultee or foster carer to the young person might be interpreted by the latter as the speech act ‘nagging’ and result in the young person understanding the relationship between them as ‘I am the child, they are the adults’.

This can be further explained by the CMM concept of social reality which refers to an individual’s beliefs about meanings and action in the sphere of interpersonal conduct. One of the ways in which these beliefs can be expressed is by way of constitutive rules. These are used to help understand how meaning at one level of abstraction deter- mines meaning at another level. In the example, there may be a constitutive rule which determines that all attempts by the consultee to get the young person to pull their weight are understood by the latter as nagging. Alternatively, the young person may respond to these communications by feeling guilty that others are putting themselves out, or perhaps they might think that the underlying agenda behind the offers of help towards independent living is that the foster carers ultimately want to get rid of them.

These two frames of reference (i.e. the frame by which the consultee understands the young person’s behaviour and the frame by which the young person understands the actions of the consultee) interact to give us what CMM terms the ‘coordination of meaning’. According to Philipsen (1995), coordination is not necessarily the

Multi-agency collaboration, multiple levels of meaning 285

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

coming together of perfectly shared meanings. Rather it is the coming together of actions into a sequence or pattern, regardless of whether the producers of those actions agree as to their meanings.

The meaning of any utterance will be influenced by previous utterances and any utterance will influence future utterances. The authors hypothesized that further insight into the meaning of the behaviour described by the consultee would emerge by examining the transcript of the meeting as a whole. It transpired that earlier on in the consultation, there had been a discussion about the difficulty the consultee had in arranging regular contact for the young person with their family of origin. Later on in the consultation meeting, comments are made about the young person questioning the nature of the placement (i.e. whether it was a ‘foster placement’ or ‘supported lodgings’). The meaning of a ‘foster placement’ to the young person seemed to be somewhere where everything should be done for them and where they should not have to lift a finger to help. Any suggestion that the young person might, for example, take a part in household chores, as the other children in the foster family were expected to, led to the former questioning whether they were indeed in ‘supported lodgings’. The young person’s resistance to sharing the chores with others in the family might reflect a bigger resistance to seeing themselves or being seen by others as part of the foster family, when what the young person really wants is to be back home with their own family. Perhaps the young person deliberately chooses to stay quite separate. The constitutive rule here might be that requests by the foster carers to the young person to do their share of the chores have the meaning of ‘asking the young person to be part of the family’. The regulative rule which then helps guide the young person’s action (e.g. when asked by foster carers to do chores (i.e. to be a member of the family)) would be to ignore the request and do nothing.

Meaning within the context of the outside world

There may also be contextual factors located outside of the confines of the meeting that are significant. Knowledge of these factors is likely to be helpful in determining the meaning of the discourse. A researcher or clinician may have knowledge of some of these factors, but can only hypothesize about the rest. Background factors which may have been influential in this example include, at the level of the agency, that the professional with whom the first author (GS) had initially negotiated

286 Gill Salmon and Jeff Faris

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

the sessions had subsequently left the agency and at the time of the consultation had not been replaced. Thus the original remit for the sessions that had been agreed before the sessions had started needed to be renegotiated with the staff who actually attended.

At a higher contextual level, government policy and legislation is referred to at regular intervals throughout both of the consultation meetings with the agency. Individual professional and agency practice seems to be greatly influenced by the constraints imposed at this level. The issue which seems to present the greatest frustration and which is perceived to be a huge obstacle for professionals working with looked- after children is the issue of needing someone with parental respon- sibility to make decisions on behalf of the child, and such a person is often difficult to contact. In addition, even the simplest of arrange- ments are perceived to involve going through many layers of ‘red tape’. One might hypothesize that in an attempt to avoid or reduce ‘red tape’, certain courses of action or options in relation to children and young people may seem preferable to others.

Limitations of the study

One of the limitations of this study was the authors’ wish to examine the usefulness of the CMM model as a framework for data analysis. Although the CMM model has been applied to a number of situations (Harris and Cronen, 1979; Harris, 1980; Stanback and Pearce, 1980; Lindsey, 1993; Philipsen, 1995), a detailed examination of the litera- ture reveals that in all cases, the researchers had set out to examine the model from the start and designed the research interviews accordingly. In contrast, in this study the researcher conducted an in-vivo study of CAMHS clinical practice. As such, participants were not interviewed using questions formulated with the CMM model in mind. Second, this was only a pilot study and as such the researcher did not attempt to analyse all the transcripts using the CMM model, as this was outside of the scope of the research project.

Another limitation on the study to be considered was the impact that the role of the clinician-researcher would undoubtedly have had upon the research process. Even though the clinician-researcher was essentially researching aspects of her day-to-day clinical practice, the very fact that she was aware of the research aims and background to the study would have had an impact on her practice at the time of the study. For example, while participating in consultation meetings, a lot of attention was paid to asking for clarification of terminology or

Multi-agency collaboration, multiple levels of meaning 287

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

information about case material. It is unclear whether the clinician- researcher would usually have responded in this way. In addition, clinician-researcher participation in the meetings may have affected the consent process, particularly in relation to members of the team in which she worked.

Discussion

The contribution of social constructionism and the CMM model to the development of qualitative research methodology

In this study, the authors illustrate that the application of social constructionist theory, and the CMM model in particular, provides an alternative useful framework for researchers and clinicians at- tempting to understand discourse as it allows for multiple layers of complexity to be explored. Lindsey’s (1993) modified version of the CMM model is particularly useful as it can unpick discourse at a personal and professional level.

While the qualitative technique of thematic analysis and the application of the CMM model would appear to be completely different analytic tools, they both allow for different layers of context and complexity to emerge from the data. Thus, although approaching the research from different directions, the techniques seem comple- mentary and additive to each other. The use of two complementary analytic techniques could also be viewed as a systemic approach to research, as it reinforces the value of viewing data from more than one perspective, with the aim of creating a resulting whole that is greater than the sum of its parts. It is the view of the authors that systemic theory offers a potential method for researching which could sig- nificantly add to theory building as well as the development of qualitative research approaches.

The potential contribution of this study to the literature on multi-agency collaboration

It has been suggested that two of the most critical factors in the success or failure of multi-agency collaborations are the presence or absence of effective communication and a common language used across the agencies (Bruner et al., 1992; Ranade, 1998; Miller and Ahmad, 2000). In the authors’ view, consensus on definitions or terminology alone is unlikely to suffice in achieving collaboration. In multi- professional contexts with divergent foundational histories, values,

288 Gill Salmon and Jeff Faris

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

ethos, standards of practice, and salaries, one might wonder whose definitions or terms would prevail? For all parties agreeing to equal participation in establishing newly constructed definitions would necessitate at least some relinquishment of cherished truths and vested interests. At most, such realignment of meaning would require a major revolution in thinking and practice. The process itself of identifying categories relies upon specific frames of reference or categories that underpin the premises leading to the particular conclusions of the discipline or profession in question.

While the authors support the need for a common understanding of language to be negotiated between professionals from different agencies, they argue that this study emphasizes the impact that the particular context might have on the understanding which is arrived at in any one situation. The authors advise participating agencies in multi-agency meetings to discuss their own aims for the meeting and the individual agency agendas they hope to address by attending at an early stage. Facilitators of such meetings need to be mindful of issues such as power which may impede the ability of participants to challenge others when active negotiation over meaning is required (Mehan, 1983), and of the regulative rules which may be in operation at any one time.

The findings of this study support those of other authors (Hallett and Birchall, 1992; Okamoto, 2001) in emphasizing the importance of case-level or task-level collaboration in the process of multi-agency working, as mutuality of interest seems to facilitate information exchange. This resonates with the CMM concept of coordination which is the coming together of actions regardless of whether the producers of those actions agree as to their meanings (Philipsen, 1995).

Contextual forces may frame the interaction in multi-agency dis- course. It is, however, a possibility that increasing facility for meta- communication and reflective practice among practitioners from varied professional backgrounds and different agency or service contexts may become the norm through the implicative force of such discourse. In the authors’ optimistic view this could open up possibilities for all the benefits of multi-agency and multi-professional communication advocated in the Department of Health and other government papers cited at the beginning of this article. On the one hand, however, there is also unlimited potential for chaos and disorganization and the eventual rigidity or institutionalization of novel ideas, almost as though there is a natural law that determines

Multi-agency collaboration, multiple levels of meaning 289

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

these processes. To ignore this would be to miss the point about CMM, and of course Bateson’s caveat that as human beings we can never fully and finally eliminate paradox or confusion from our commu- nications (Bateson, 1972).

References

Akhavin, P., Amaral, D., Murphy, M. and Uehlinger, K. (1999) Collaborative practice: a nursing perspective of the psychiatric interdisciplinary treatment team. Holistic Nursing Practice, 13: 1–11.

Anderson, H. and Goolishian, H. (1986) Systems consultation with agencies dealing with domestic violence. In L. Wynne, S. McDaniel and T. Weber (eds) Systems Consultation. New York: Guilford Press.

Anderson, H. and Goolishan, H. (1988) Human systems as linguistic systems: preliminary and evolving ideas about the implications for clinical theory. Family Process, 27: 371–393.

Anderson, H. and Goolishan, H. (1992) The client as the expert: a not knowing approach to therapy. In S. McNamee and K. Gergen (eds) Therapy as a Social Construction. London: Sage.

Aronson, J. (1994) A pragmatic view of thematic analysis. The Qualitative Report, 2. Audit Commission (1999) Children in Mind – A Report on Child and Adolescent Mental

Health Services. London: Audit Commission. Bateson, G. (1972) Steps to an Ecology of the Mind. New York: Ballantine. Briggs, M. H. (1997a) Building Early Intervention Teams: Working Together for

Children and Families. Gaithersburg, MD: Aspen Publishers. Briggs, M. H. (1997b) A systems model for early intervention teams. Infants and

Young Children, 9: 69–77. Briggs, M. H. (1999) Systems for collaboration: integrating multiple perspectives.

Child and Adolescent Psychiatric Clinics of North America, 8: 365–377. Bruner, C., Kunesh, L. G. and Knuth, R. A. (1992) What does research say about

interagency collaboration? NRCEL (101/10/02 internet reference). Burr, V. (1995) An Introduction to Social Constructionism. London: Routledge. Campbell, D., Coldicott, T. and Kinsella, K. (1994) Systemic Work with Organizations:

A New Model for Managers and Change Agents. London: Karnac. Cecchin, G. and Fruggeri, L. (1986) Consultation with mental health system

teams in Italy. In L. Wynne, S. McDaniel and T. Weber (eds) Systems Consulta- tion. New York: Guilford Press.

Cronen, V. E. (1995) Coordinated management of meaning: the consequentiality of communication and the recapturing of experience. In S. Sigman (ed.) The Consequentiality of Communication. Hillsdale, NJ: Lawrence Erlbaum.

Cronen, V. E. and Pearce, W. B. (1985) Towards an explanation of how the Milan method works: an invitation to a systemic epistemology and the evolution of family systems. In D. Campbell and R. Draper (eds) Applications of Systemic Family Therapy: Milan Approach. New York: Grune & Stratton.

Cronen, V. E., Johnson, K. M. and Lannamann, J. W. (1982a) Paradoxes, double binds, and reflexive loops: an alternative theoretical perspective. Family Process, 21: 91–112.

290 Gill Salmon and Jeff Faris

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

Cronen, V. E., Pearce, W. B. and Harris, L. M. (1982b) The coordinated management of meaning: A theory of communication. In F. E. X. Dance (ed.) Human Communication Theory. New York: Harper & Row.

Dallos, R. and Draper, R. (2000) An Introduction to Family Therapy: Systemic Theory and Practice. Milton Keynes: Open University Press.

Dartington Social Research Unit (DSRU) (1998) Towards a Common Language, http://[email protected].

Department of Education and Skills (DfES) (2003) Every Child Matters. Notting- ham: DfES Publications.

Department of Health (DoH) (1997) NHS Planning and Priorities Guidance 1997/ 98. London: HMSO.

Department for Health, Department for Education and Skills (DoH, DfES) (2004) National Service Framework for Children, Young People and Maternity Services, Executive Summary. London: DoH Publications.

Gantley, M. (1999) A beginner’s guide to developing a thematic analsyis. In Y. Carter, S. Shaw and C. Thomas (eds) An Introduction to Qualitative Methods for Health Professionals. London: The Royal College of General Practitioners.

Gergen, K. J. (1999) An Invitation to Social Construction. London: Sage. Gergen, K. J. and Gergen, M. M. (1991) Toward reflexive methodologies. In

F. Steier (ed.) Research and Reflexivity. London: Sage. Gergen, M. M. (2002) Social constructionist theory. Paper presented to Emerg-

ing Approaches to Inquiry 9, Conference, Hawkeswood College, Stroud, September.

Hallett, C. and Birchall, E. (1992) Coordination and Child Protection: A Review of the Literature. Edinburgh: HMSO.

Hammersley, M. (1989) The Dilemma of Qualitative Method: Herbert Blumer and the Chicago Tradition. London: Routledge.

Harris, L. M. (1980) The maintenance of social reality: a family case study. Family Process, 19: 19–23.

Harris, L. M. and Cronen, V. E. (1979) A rules-based model for the analysis and evaluation of organizational communication. Communication Quarterly, 27: 12– 28.

Henneman, E., Lee, J. and Cohen, J. (1995) Collaboration: a concept analysis. Journal of Advanced Nursing, 21: 103–255.

House of Commons Health Select Committee (1997) (HC26-1 1997) Report on Child and Adolescent Mental Health Services (CAMHS). London: HMSO.

Imber Black, E. (1986) The systemic consultant and human-service-provider systems. In L. Wynne, S. McDaniel and T. Weber (eds) Systems Consultation. New York: Guilford Press.

King, E. (1996) The use of self in qualitative research. In J. T. E. Richardson (ed.) Handbook of Qualitative Research for Psychology and the Social Sciences. Leicester: The British Psychological Society.

Kopser, K., Horn, P. and Carpenter, A. (1994) Successful collaboration within an integrative practice model. Clinical Nurse Specialist, 8: 330–333.

Leininger, M. M. (1985) Qualitative Research Methods in Nursing. London: W.B. Saunders.

Lindsey, C. (1993) Family systems reconstructed in the mind of the systemic therapist. Journal of Systemic Consultation and Management, 4: 299–310.

Multi-agency collaboration, multiple levels of meaning 291

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice

Marino, T. and Kahnoski, B. (1998) Communication action for case managers: techniques to manage conflict. Nursing Case Management, 3: 36–48.

Mehan, H. (1983) The role of language and the language of role in institutional decision making. Language and Society, 12: 187–211.

Miller, C. and Ahmad, Y. (2000) Collaboration and partnership: an effective response to complexity and fragmentation or solution built on sand? Interna- tional Journal of Sociology and Social Policy, 20: 1–39.

NHS Health Advisory Service (1995) Thematic Review. Together We Stand. The Commissioning, Role and Management of Child and Adolescent Mental Health Services. London: HMSO.

Okamoto, S. K. (2001) Interagency collaboration with high risk gang youth. Child and Adolescent Social Work Journal, 18: 5–19.

Pearce, W. B. and Cronen, V. E. (1980) Communication, Action and Meaning: The Creation of Social Realities. New York: Prager.

Philipsen, G. (1995) The coordinated management of meaning theory of Pearce, Cronen, and associates. In D. P. Cushman and B. Kovacic (eds) Watershed Research Traditions in Human Communication Theory. Albany, NY: State Univer- sity of New York Press.

Ranade, W. (1998) Making Sense of Multi-agency Working. Sustainable Cities Research Institute, http://www.sustainable-cities.org.uk/Database_files/ranade- multiA.doc.

Salmon, G. (2004) Multi-agency collaboration: the challenges for CAMHS. Child and Adolescent Mental Health, 4: 156–161.

Schwandt, T. A. (2000) Three epistemological stances for qualitative inquiry: interpretivism, hermeneutics, and social constructionism. In N. K. Denzin and Y. S. Lincoln (eds) Handbook of Qualitative Research (2nd edn). London: Sage.

Speed, B. (2004) All aboard in the NHS: collaborating with colleagues who use different approaches. Journal of Family Therapy, 26: 260–279.

Stanback, M. H. and Pearce, W. B. (1980) Dissimulation: an analysis of a form of intercultural communication. Paper presented to the International Commu- nication Association Annual Convention.

Van Manen, M. (1990) Researching Lived Experience: Human Science for an Action Sensitive Pedagogy. Ontario: State University of New York Press.

Watzlawick, P., Beavin, J. and Jackson, D. (1967) Pragmatics of Human Commu- nication. New York: W. W. Norton.

Wittgenstein, L. (1953) Philosophical Investigations. New York: Macmillan.

292 Gill Salmon and Jeff Faris

r 2006 The Authors. Journal compilation r 2006 The Association for Family Therapy and Systemic Practice