The Dance of Reflexive Monitoring in a Family Co-Mediation Context

This article describes a qualitative study of the co-mediation process from the perspective of mediation practitioners, and what salient features distinguish it from sole mediation. The study examined the nature and character of the interactive patterns of communication performed by all participants as central features of the process. Forty-seven mediators were surveyed by questionnaire in relation to the co-mediation process. A further seven mediators were subjected to in-depth interviews. Their combined responses revealed co-mediators as “agents of change” tasked in a distinctive way with facilitating fair and just outcomes to family disputes. Exploration of the characteristics unique to co-mediation revealed a dynamic reflexive monitoring interplay with interconnected functions comprising a complex “dance” that involved all participants.

Creative Commons CC BY: This article is distributed under the terms of the Creative Commons Attribution 3.0 License (http://www.creativecommons.org/licenses/by/3.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access page (http://www.uk.sagepub.com/aboutus/openaccess.htm).

Article
Mediation, according to Ridley-Duff and Bennett (2011), is driven by a commitment to developing dispute resolution skills directed at future equitable relationships. To this end, Mayer (2012) recognized a "constellation" of approaches and identified some of the activities that mediators perform within the process, the core being to promote constructive communication. This study focused on the interactions presenting where the mediation arrangement consists of two rather than a single mediator practitioner. Regarded as a strategy by Hayes (2012) and as an approach by Walker (2010), the co-mediation configuration was defined by Walker, McCarthy, and Timms (1994) as involving two practitioners working collaboratively on a certain dispute. Characteristics of such teams include mediators exhibiting complementary skills (Roberts, 2008;Zariski, 2005), though often presenting with different backgrounds (Walker et al., 1994). Co-mediation teams therefore, engage with clients in a supportive partnership, and according to Walker et al. (1994) and Mason and Kassam (2011), execute defined tasks with the objective of arriving at a mutually satisfactory formal agreement, or in some cases, a less formal Heads of Agreement (Hanaway, 2012;Whitehead & Buncle, 2013). Furthermore, descriptions of co-mediation teams include those by Cornfeld (1985) and also Keys (2009) who noted that team members range from those having similar professional or disciplinary backgrounds to those selected to work together because they have attributes appropriate to a particular case or dispute. For example, co-mediators from different backgrounds or disciplines engaging in interdisciplinary co-mediation (Gold, 1984;Walker et al., 1994) form an effective partnership that might involve pairing of a lawyer and a "therapist" (Gold, 1982;Irving, 2011;Mosten & Biggs, 1986). Further benefits can be attached to co-mediation teams according to Parkinson (1997), not least of which was access to an extended range of perspectives and outlooks, skills, and professional expertise. The importance attached to an effective co-mediation team was illustrated by Coates and Damas (1997) who suggested that in cases of parent-teen mediation, it was prudent to defer the session if an appropriately matched team is not available. To reduce the impression by clients of "gender bias," co-mediator gender "balancing" was considered essential by Love and Stulberg (1996). Also concerned with "balance," Charkoudian and Wayne (2009) examined the impact of racial matching, in addition to gender matching (Charkoudian & Wayne, 2010), of participants engaged in community mediation. Their research indicated that, although the effect on outcomes was unclear, co-mediation presented more opportunities to facilitate racial and gender balancing of all participants. Whether that would apply to family dispute mediation was not discussed by the authors. Strahl (2007) recognized that the benefits of the co-mediation experience include enhanced opportunities for debriefing after the session. In a similar manner, Zariski (2005) and Roberts (2008) added that a particular advantage is the opportunity to monitor each other's contributions. Both Kranitz (1985, pp. 71-79) and Bowen (1999) noted that while co-mediation had much to recommend it, such as the opportunity for conflict resolution modeling, it was expensive and prone to "communication problems." Irving and Benjamin (2002) also recognized the considerable benefits to be gained from co-mediation but highlighted a number of limitations. These shortcomings identified cost factors plus a range of logistical problems including a susceptibility to negative role modeling by mediators finding it difficult to resolve their own differences, and co-mediators failing to "gel," with the incompatibility likely to adversely affect the process. Boulle (1996) speculated upon the many positive advantages of co-mediation and noted the "doubling of resource strengths" (pp. 113-117), the potential for dynamic stability in the interplay between participants, and potential economic benefits from practitioner division of labor. This might benefit situations where there are perceived to be asymmetric power relations that Gerami (2009) regarded as actually inherent to a mediation setting. In fact, Martin and Douglas (2007) opined that it is incumbent upon practitioners to be alert to power imbalances and deal with them effectively. Both Love and Stulberg (1996) and Rosengard (2004) were in favor of co-mediation, considering it a compelling training and mentoring framework, whereas Zariski (2005) reinforced their views by endorsing the model as a progressive training instrument. Conversely, Tillett and French (2006) declared that co-mediation could only be effective if practitioners worked co-operatively in an equal partnership, suggesting limited utility as a training instrument. Parkinson (1997) also identified the likelihood of competition or discord between mediators, and suggested that the presence of two mediators rather than one might prove intimidating for disputants. Roberts (2008) confirmed this view, adding the possibility of one mediator dominating a session. However, Roberts applauded its potential for modeling negotiation and conflict resolution skills noting that co-mediation had a tendency to minimize the possible effects of mediator egotism (Roberts, 2007). Discussing ways of circumventing some of these difficulties, Irving and Benjamin (2002) pointed out that practitioners might ideally exercise choice between sole or co-mediation modes according to particular client needs and the mediator's preferences. However, Bickerdike (2003), as part of a service evaluation, surveyed mediation clients with a view to determining situations in which each of the configurations might be applied optimally. A preference was expressed by 128 male clients and to a slightly lesser extent by 130 female clients, for sole rather than co-mediation.
The distinctiveness of the current work is that the specific interactive elements of the process were explored qualitatively from the perspective of mediators. The research questions of interest to the investigator were essentially twofold.

Research Question 1a:
What are the characteristic interactive mechanisms that are central to the co-mediation process?
Research Question 1b: What salient features distinguish it from sole mediation? Research Question 2: How do mediators perceive the family co-mediation process?
Utilizing data gathered from semi-structured interviews coupled with survey responses, practitioners' views on the "pros and cons" of co-mediation led to some understanding of the nature of collaborative mechanisms within comediation sessions. From the information gathered, a model was proposed that emphasized the complexity and intricacy of the sub-processes present within a family co-mediation context. In view of the assertion by Kelly (2004) that the characteristics of mediation participants and their behaviors are demanding of significant attention and the observation by Sinclair and Stuart (2007) on the paucity of research into process dynamics, the current study afforded greater understanding of the role assigned to co-mediators.

Method
The research investigation was a "component design" under the umbrella of a mixed method design, described by Hall (2008) as a triangulation method involving the use of different approaches intended to increase the validity of the findings. The approach chosen utilized two sources of qualitative data: (a) questionnaires and (b) in-depth semi-structured interviews. In recruiting questionnaire participants, letters of invitation were mailed to mediation agencies. Because sampling was purposive, the emphasis of the unit of analysis was family mediation practitioners. In the invitation letter, the project was presented as involving a postal questionnaire that would probe practitioners' opinions on a range of topics associated with family mediation practice and process. There were 129 affirmative responses from the agencies resulting in 129 serially numbered packages being sent to the agencies for distribution to willing practitioners. Each of the packages contained a cover letter introducing the researcher and the sponsoring institution, a description of the project and its potential significance, appreciation of respondents' co-operation in the project and reassurance of confidentiality, the questionnaire (with completion instructions), and a return envelope addressed to the investigator. One hundred twentynine packages were sent out with 47 totally completed questionnaires received by the investigator who was blind to the identity of respondents. The response rate was therefore 36%. In the analysis, answers to each question were coded for similar or analogous responses and grouped into clusters.
Seven mediation practitioners, separate from the surveyed subjects, were engaged as volunteers to participate in informal in-depth interviews, considered by Denscombe (2010) as semi-structured, and which were conducted by the investigator. As it was considered important to discuss with participants their understanding of the process and its mechanisms and explore the meaning of their role, the utilization of this method of data collection was similar to the reasons defined by Tovey, Chatwin, and Broom (2007) where a range of conceptualizations and meanings was elicited. The 7 interviewees in the current study were from a pool of 19 experienced practitioners known professionally to the investigator through shared involvement in family mediation. A phenomenological style of interviewing was used, considered by Roulston (2010) as valuable in generating detailed explanation and meaning to interviewee experiences and understandings, thus allowing opportunity to explore points of significance as identified by participants. The interview guide consisted of expanded and additional thematic elements analogous to those in the questionnaire. Participants were asked how they perceived the family mediation process and to reflect on their experiences and understanding of the co-mediator role. Inclusionary criteria were formal mediation training, participation in 400 or more mediation sessions post-training, and extensive familiarity with sole and co-mediation. The interviews were recorded over a period of 1 to 1½ hr, the interviewees having provided "informed consent" based on the cautionary elements expressed by Padgett (1998, pp. 35-36) later reinforced by Denscombe (2010, pp. 172-173). Transcription followed the method described by Richards (2009), and pseudonyms were used to preserve confidentiality. The transcripts were subjected to thematic analysis (Grbich, 2007), coded for comparable responses and patterns (Saldana, 2009) and then categorized into clusters (Saldana, 2009). In the following two sections, patterns of responses to questionnaires and interview questions are presented.

The Questionnaires
The participants' responses to questions in the survey were coded as indicated in italics below, and grouped into clusters of recurrent patterns as described by Saldana (2009), then numbered 1 through 14 for Question A and 1 through 10 for Question B. Coding followed the method as defined by Flick (2007), whereby each of the clusters is developed from the data. The number of responses per cluster is indicated in parentheses.
A. "What do you think are the advantages of a co-mediation model?" Respondents believed that co-mediation provided advantages (1 through 14 below; number of responses in brackets).
1. A gender balanced mediation team contributing to, and aiding client-practitioner communication, enhanced appearance of impartiality, peer support, and matching of mediator skills with disputant gender. Most respondents believed this configuration exhibited to clients a lack of gender bias and reflected client gender differences, so that each party was likely to feel represented. The possibility that men and women would have different perspectives on issues arising in a session was regarded as predictable. One respondent observed that female clients often preferred to talk with female mediators; another noted that a disputant "might gain comfort" from there being a mediator of the same gender in a case where violence was seen by one or both parties as a feature in the relationship. One male respondent, however, held the view that it was mediator quality rather than gender that was important. (35) 2. Enhanced skill sharing conducive to development of opportunities for consultation, "resolution of differences and creative brainstorming"; and, opportunities for mediator reflection and effective debriefing. (26) 3. A wider skill and knowledge base from mediators with varied academic training and/or expertise, affording "different approaches and perspectives," generating a range of alternative intervention strategies. (21) 4. Significantly enhanced ability to observe and monitor all participants and compare notes; for observation of client demeanor, coalitions and collusion between participants, "unspoken signals," and "body language." (21) 5. A mutually supportive, flexible, and "less wearing environment" for all participants and opportunities for mediator "load sharing" during sessions, such as note taking. (19) 6. An enabling, monitoring, and maintenance function of co-mediator neutrality/impartiality and practitioner objectivity, thereby "aiding quality control." (17) 7. Modeling of good negotiation and communication skills, co-operative decision making and effective teamwork. A number of respondents emphasized the importance of "effective teamwork" or "compatibility of colleagues" with significant numbers of respondents being emphatic in their support of choosing to work with colleagues who exhibited "well developed" or "excellent" . . . "communication skills." (17) 8. Provision for one practitioner to monitor errors, to identify and pick up on cues, or "follow up topic areas when colleague is perceived to be having difficulties"; a balance of differing strengths and weaknesses of each mediator; better interaction and mutual support, particularly when working with complicated, difficult, or demanding cases; substantially reduced possibility of coalition or collusion between a mediator and client and between mediators. (14) 9. Attempting "power balancing" of disputants (13). In this context, the concept of "power balancing" was as defined by Mayer (2012, p. 69). 10. The enabling of clients to (a) move from "fixed positions" and (b) to understand significant others' needs and concerns, leading to preparedness for mutual resolution of disputes and concerns. (9) 11. Better regulation of the process and the session as a whole, with less chance of misinterpretation or misunderstanding should "a personality clash" arise between mediator and client(s). (6) 12. Allowing mediators to be "matched" to parties, if necessary; "matching" can be by age, gender, culture, or ethnicity. (6) 13. Lessening the possibility of "triangulation" and bias, and protection against practitioner/client boundary violation including, as some respondents noted, "mediators offering advice." (6) 14. Enhanced safety and security, where mediators can be mutually protective and also "of the clients," particularly where domestic violence is identified by participant(s). (3) B. "What do you think are the disadvantages of a co-mediation model?" Respondent's persuasive views (1 through 10 below; number of responses in brackets) were as follows: (a) That co-mediation was vulnerable to: 1. A lack of mediator cohesion and poor teamwork resulting from, and in, poor communication or conflict between mediators, or both. (11) 2. Different mediator styles likely to hinder resolution progress; different practitioner personalities impeding progress, which can be intimidating for clients, and likely to produce a negative outcome; agencies not "pairing mediators appropriately" or "sensitively." (10) 3. Competitive professionals exhibiting a conflict of values or objectives, styles, and approaches, leading to tensions, which may prejudice outcomes. In such situations, colleagues might hold different views as to what constitutes a preferable outcome or "what was important." A perceived bias on the part of one mediator potentially affects the overall performance of both practitioners. (8) 4. Differing directions being adopted by each mediator resulting in client confusion; in addition, mediators might be, or at least feel, incompatible and unwilling to follow process, or "have adverse views of each other." (8) 5. Issues arising from difficulties in matching mediators for gender or race. (5) 6. Disputants feeling "threatened" where two mediators are of the same gender.
(3) (b) That the co-mediation configuration: 7. Demands a substantial, and in some cases, "unrealistic" level of co-operation, empathy, and understanding between mediators. (6) 8. Is financially expensive, and therefore probably less cost effective in operation, disjointed, "wordy," and time-consuming. (8) 9. Is administratively cumbersome in terms of coordination logistics extending into the application of techniques such as caucusing. (7) 10. Has the potential for displaying a disparity of skills associated with incompatible colleagues; even those with similar experience may see one dominating the other. In situations where mediators have appreciably different levels of experience, the more experienced mediator may feel unsupported by the less experienced colleague. (5) The Interviews All seven interviewees, named as Nicola, Paul, Sarah, Peter, Tina, Joan, and Margot, worked both as co-mediators and sole-mediators. Like statements and patterns (Saldana, 2009) were gathered from the transcripts and categorized into clusters that are reproduced below and numbered 1 through 6.
All the interviewees held the view that mediators should be compatible and well matched, thus enabling effective monitoring of participant interactions, assessing progress of the session, and generating a more productive intervention. Joan commented, . . . I've done excellent co-mediation. In the "early days" we mediated with the same partner most of the time. We knew each other's strengths and weaknesses and were able to "pick each other up" on something the colleague was saying or was missed such as a client's crucial or core messages. Joan emphasized that productive co-mediation required "knowing" your colleague. She added that co-mediators "should not be talking over the top of each other or dominating a session, but should balance each other out." Joan hastened to add that there would be times within a session when appropriately, one of the two mediators did most of the talking.
2. Co-mediation as a framework for pursuit of fairness and justice.
The interviewees regarded co-mediation as structured to maximize fairness. "Fairness" was frequently raised as a topic, as was "justice," though less often. There was agreement that the financial cost of two mediators has to be justified, so to be of value the mediation process must be fair to all parties to a dispute, particularly since they would be the final judges of what was or was not fair. Most interviewees accepted that judging levels of fairness and justice within the mediation process was complex, as clients do not control the mediation process and might hold differing or unspoken views of what is fair and what is just. Interviewees were unanimous that mediators have a primary responsibility to ensure that disputants are satisfied that every effort has been made to achieve a just and fair outcome. In fact, Margot, Nicola, Joan, and Tina suggested that the presence of a comediator demonstrated that the process is directed toward a fair and just agreement. Interviewees also emphasized that transparency in communication as modeled by practitioners was a central tenet of the process, facilitating fair and just outcomes, augmented by intensive reflexive monitoring. Tina and Margot regarded that this particular aspect of comediation gave it ". . . the edge over sole mediation . . ." 3. Co-mediation as a facilitator of client empowerment.
The interviewees associated questions about "communication skills" with "the role of the practitioner." This role included monitoring the relative "power differentials" of all participants including colleagues, and then repairing perceived imbalances. Interviewees held a range of views on issues surrounding "power imbalance." For example, Nicola regarded co-mediation as calculated to facilitate empowerment of disputing parties to enable them to "assume responsibility of their lives . . ." Alternatively, Sarah indicated a preference for "direct intervention" wherever mediators agreed that marked "imbalance" persisted between disputants. Margot and others conceptualized the "power status" present during sessions as "fluid," "ebbing and flowing" even between mediators. Exploring the concept of "power balance" and empowerment, the interviewees revealed little of how they determined power levels, but did acknowledge the presence of fluctuating power imbalance. They all disliked the term power balance because it was incongruent with, as one person put it, ". . . core ethics of neutrality and impartiality." Most of them saw the two factors of client participation and informed decision making, empowering. They mentioned, too, that power relationships established prior to mediation were often destabilized during sessions whether by co-mediator action or as resulting from the passage of the process. However, some interviewees volunteered that they saw themselves as having a further goal: that of enabling empowerment of both parties. Conversely, they shied away from the idea of reducing the power of one disputant so as to "equalize." Statements such as ". . . we try to empower both" and ". . . ensure each party is able to negotiate equally," illustrated that, for them, the empowerment process was linked to client autonomy.
4. Practitioner support inherent to co-mediation.
Interviewees viewed the individual perceptions of each participant as a major contribution to the dynamic environment of sessions, particularly the reflexive stance assumed by practitioners as they constantly monitor the objectivity of their colleague. In this regard, Sarah saw opportunities within co-mediation to exercise critical and reflective practice, affording special advantages over sole mediation by reducing the likelihood of collusion or coalitions between participants. Both Peter and Margot saw the reflexive role as offering a wider benefit, particularly in monitoring the total dialog of the session. Paul's mediation experience was mainly (though not totally) within a co-mediation framework. He said that working with somebody else means that if something goes wrong or one of the parties does not really "take to you," then you can continue with the session. One has support . . . to the extent of a colleague assisting or assuming control of the session. Paul added, "I always feel that the balance of having a man and a woman facing a man and a woman is a much more effective way of establishing . . . the atmosphere necessary to achieve best results." He emphasized the importance of the manner in which co-mediation duties are shared, describing it as, "a shared endeavor coupled with shared responsibility . . ." Three interviewees thought it critical to the integrity of the process that each mediator presents a neutral and impartial stance; this, despite the improbability (or impossibility according to Peter) of practitioners being privately or "internally" neutral. The three noted that presentation of an impartial environment demands a high degree of reflexivity by co-mediators, achieved through constant mutual monitoring of communication. All seven considered it vital to constantly monitor the direction in which the session was progressing to focus on mutually satisfactory outcomes, with clients making autonomous decisions, excluding any "leading" or coercion.
5. Co-mediation as a vehicle for improving communication between disputants.
All interviewees expressed the view that a major focus of co-mediation was to actively facilitate improvements in disputants' communication between themselves, and potentially with significant others. This was reinforced by Margot, who saw a role for herself in educating parties and facilitating improvements to their communication abilities so as to progress toward resolution. She said that her educative function was to "guide" the dominated party toward enhancement of communication skills, thereby attempting to reduce professed intimidation. Tina asserted that while providing a safe environment in which to facilitate improvements in communication skills, co-mediation was more likely than sole mediation to assist both parties address complex questions and re-establish or introduce conflict-solving as desirable. Tina said that this was achieved through ". . . stimulation of cooperative decision-making and resolution of differences . . . to move from fixed positions . . . leading to preparedness to mutually resolve issues . . ." Tina's belief was that differences could be solved through enhancement of interpersonal communication skills via modeling of problem-solving techniques.
Modeling and constant monitoring of, for example, problem-solving skills by practitioners was enthusiastically espoused by all the interviewees as intrinsic to co-mediation practice. Paul was atypical in playing down the importance of the pursuit of agreement and being more interested in facilitating improvements in future communication: . . . for me, the outcome of mediation is not necessarily the agreement. I would not feel there is success, if at the end of the session the people do not indicate in some way or other that they're able now to continue dialog in a way that is likely to be more productive and positive than . . . in the past.
Paul, Nicola, and Tina attached importance to mediators ensuring client core concerns were identified and explored and desired goals linked with outcomes so disputants felt their individual needs are being addressed. All three interviewees credited the reflexive character of co-mediation with facilitating improved communication between disputing parties.
6. Co-mediation as a stimulant to autonomous decision making.
Peter observed that many disputants enter mediation with a perceived loss of personal power, or reduced control over their lives. He considered it a virtue of co-mediation that a person may learn to assume some control and make decisions, where previously this capacity had been "lost." He attributed this to the manner in which co-mediation is practiced, such as the "transparent" pursuit of autonomous and informed client decision making. Like Peter, Nicola felt strongly about the objective of stimulating, "fair and reasonable decision making" to regain control or better manage their concerns. The concerted effort from co-mediators engaged in encouraging clients to "better" improve their decision-making ability was a primary focus for Tina and Sarah. Both of them used the term better in this context. When confronted by the interviewer, both submitted that "better" connoted a value judgment, but stressed that as far as they were concerned, its use was justified, mainly on the grounds that it reflected their commitment to the pursuit of client autonomy.

Analysis of Data, Discussion, and Conclusion
The project began with two questions: First, what distinctive interactive mechanisms are central to co-mediation, and what salient features distinguish it from sole mediation; second, how do co-mediators perceive the family co-mediation process? Drawing on the questionnaire and interview data, the answers to these questions yielded the following integrated findings, recorded and discussed below.
The first finding was that responses from all 54 participants (47 questionnaire respondents plus 7 interviewees) exhibited considerable commonality. For example, participants recognized the benefits of a sole mediator model but regarded two mediators as essential for success (although one respondent had some reservations).
From discussion around the characteristics peculiar to comediation configurations, a second and parallel finding arose. The discussion identified co-mediation as a framework in which autonomous, informed decision making by disputing parties is enabled by a dynamic "reflexive monitoring process." Reflexive monitoring, or more accurately "reflexive monitoring of action," was defined by Giddens (1984, pp. 3-30), and elucidated by Kasperson (2000, pp. 37-40) as entailing recursive, continuous self-aware surveillance and scrutiny of conduct, movements, and interactions in the flow of activities embedded in a particular setting; in this case, a co-mediated session. The informative feedback so derived is subjected to self-evaluation including interintrapersonal feedback leading to informed decision making that determines subsequent action. Activities within sessions are dynamic and affected by labile elements such as perceived "power imbalances." Throughout, the mediators' accounts of mediation process were descriptions of what amounted to "reflexive monitoring," defined in the context of the constant monitoring of performance taking place at a number of levels simultaneously for the duration of the session. The constant interactions and interchange activity taking place within sessions as described above, are represented in this study as a form of dance with related feedback loops. The notion of a dance being executed by all the mediation participants fits the context of a dynamic co-mediation session. The use of the expressions dance and dancing both as figures of speech and in a metaphorical sense was discussed by Lawler (1951) who observed the manner in which the concept of dance and dancing is frequently used by people referring to dancing eyes, dancing in the sky, or where a person's heart leaps for joy to describe a state of intense activity such as in tennis playing. Lawler identified the place of dance and dancing across cultures and emphasized the active role of rhythm and movement as inherent components of most societies. A number of authors have noted the dance metaphor being used to good effect, for example, in a situation of motivational interviewing in social work (Hohman, 2012) where the dance indicates participant empathy, connection and synchronization, and a state of working together. Hillock (2013), when canvasing the views of social workers on their understanding of oppressive practice, noted the expression being used by a respondent as dancing with service users. Similarly, Helleberg and Hauge (2014), reporting a study concerned with the emotional demands made of nursing staff caring for people with dementia in institutional settings, characterized the relationship as a dance between patient and caregiver. Leading from these observations, it can be concluded that the dance and dancing metaphor signifies an intensely active environment and is appropriate to a co-mediation setting. The dance incorporates a simultaneous monitoring and intermeshing mechanism of mediator-client and mediator-mediator interaction that is reflexively focused on outcomes peculiar to a given mediation session. Highlighted was the obligation for mediators to model and facilitate improvement of client communication skills, with the possibility of developing problem-solving approaches in future conflict and the instatement and maintenance (as far as possible) of client autonomy. In this regard, they saw compatibility of mediators as vital. Ideally, the use of a co-mediation model ensures that all participants experience and ultimately achieve a welldeveloped sense of agency or autonomy in decision making, in part, by recognizing the need for autonomy of significant others. A central feature of family mediation is, in fact, a normative assumption that disputing parties desire to act autonomously and are competent individuals, with the capacity to reach mutually inspired agreement or solutions. Unfortunately, this is not necessarily the case, particularly because disputes are by their nature, unpredictable and emotionally charged. In this regard, the view of all the study participants was that co-mediators have an "enabling" function, assisting clients to participate in a transparent "safe" mediation session.
In co-mediation, both practitioners are actively engaged as invitees within the clients' family (informal) system and communicative milieu where their role is principally one of facilitation, with enablement of empowerment a central function. Because family mediation requires disputants to make autonomous decisions, thereby enhancing the client's power of agency, co-mediators are tasked to ensure that decisions made in the course of the process respect individual autonomy while acknowledging that each client belongs to an idiosyncratic family system where entire patterns of family interaction may be enmeshed.
The respondents highlighted some of the risks inherent in the co-mediation mode of practice likely to present a threat to effective teamwork, including "personality factors" with the potential for clashes between mediators. Such difficulties might undermine successful modeling of problem-solving skills and threaten the integrity of a session. In addition, interviewees stipulated associated conditions that should be met in the implementation of co-mediation. The conditions were, first, that co-mediators were compatible with comparable skills; second, that coalitions were not formed between mediators or between a mediator and client(s); third, that a mediator was not to appear to favor or collude with one client over another, nor collude with each other and to be culturally competent.
The findings overall have implications that cover a range of concerns. Foremost among these were the importance of well-developed communication and negotiation skills; "mediator compatibility" with comparable levels of training, expertise, and experience; the centrality of reflexive monitoring to the successful conduct of sessions; the roles of implicit modeling of problem solving, co-operation, and communication skills; the interrelated concepts of fairness and justice in achieving outcomes; and the vexed question of the restrictions imposed by cost factors. The link between justice and fairness, described by Rawls and further elucidated by Audard (2007), was introduced by the investigator into all the interviews.
Markedly, effective teamwork was the attribute of "good" mediation practice that mattered most with participants, which implied harmonizing mediator qualities with clear communication and willingness to explicate when required. Their unequivocal emphasis on the importance of supporting and complementing each other with sensitivity, allowing adequate time to review or "debrief," and engaging in the process respectfully and ethically, cannot be overstated. The participants held a uniform view that an important advantage of co-mediation was the opportunity to choose "appropriate" mediators in terms of "gender balance" and/or the possession of special insight or knowledge that might be of assistance in helping disputants address features unique to their dispute. Appropriate pairing of co-mediators was seen to facilitate better communication between all participants, giving opportunities for the mediators to provide tacit modeling of cooperative behaviors, eschewing didactic instruction.

Conclusion and Implications for Practice
This study set out to identify the special advantages associated with co-mediation practice and emerged to identify a scenario where, in addition to fulfilling their primary obligation as facilitators, co-mediators are part of a complex dynamic system, which is itself a component of the session management progression. This management function embraces the sharing of mediators' skills including modeling of co-operative problem-solving process, activities, and strategies. In this way, the mediation team acts as agents of client-behavior change incorporating elements of improved decision-making competences, which in this sense refers to each party in the dispute gaining more control over their problem solving, identified by Dudley (2014) as central to the empowerment process. The changes they stimulate or facilitate in the course of mediation are likely to positively influence future communication between disputants. There are implications for future family co-mediation practice associated with this as it presents a situation where practitioners, through the mosaic of dynamic reflexive monitoring of their own and their colleagues' interpersonal communication, are able to evaluate, provide feedback, and act on emergent information. In addition, it permits the monitoring of client autonomy and "power status."

Limitations of Current Study and Direction of Future Research
The study had a singular limitation: the low response rate of the questionnaire. This article concerns one topic area, but the total questionnaire had a relatively large number of questions (56) across all topic areas, which might have deterred many respondents from replying. An on-line type of survey might have yielded a better response rate and provided additional capacity to more accurately target subsequent in-depth interviews. With a view to extending the existing knowledge base concerned with the mechanisms inherent to the family mediation system and sub-systems, and acknowledging the appeal by Emery, Sharra, and Grover (2005) for more divorce mediation research, further investigations are planned. The project reported upon here provided insight into those mechanisms and, in doing so, revealed considerable complexity involving all participants in the dynamic intermeshing interactions of the family mediation process. Building upon the research conclusions reported here, it is envisaged that a range of in-depth and integrated studies would explore sociological and broad philosophical aspects of the process along with detailed examination of mediator professional practices during each step of the process.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.