A Critique of Private Sessions in Family Mediation

This study represents a critical examination of private sessions (caucuses), seen as a sub-process within the core family mediation process and defined as involving separate and confidential conversations between mediator(s) and each disputant during the main session. In the study, the views of family mediators were explored revealing that considerable support for the use of the tool was juxtaposed with a range of fundamental ethical concerns. Emerging from the study is strong evidence of coercive practice as an inherent component of caucusing thus posing a challenge to the positive benefits.

The employment of a private session, or caucus, within a family mediation session was acknowledged by Moore (1987) as a vehicle for the promotion and satisfactory completion of a settlement. However, its use may not be popular with all practitioners, since according to Taylor (2002, pp. 161-166), the utilization of private sessions arises from users of "problem-solving/negotiative" approaches to mediation practice, whereas practitioners holding to a "systems" perspective often eschew its use.
In the main, discussions on the need for a private session arises as a result of consultation between mediator(s) and disputants, usually during sessions, but also range from being scheduled prior to the mediation session proper, frequently as part of the intake process (Swaab & Brett, 2007) to a type of shuttle between disputants in caucus following a preliminary joint session, when upon settlement a final joint session is held (Calkins, 2006). Despite there being only a small amount of published research on the nuances associated with private sessions, there are available enlightening descriptions and reviews of caucus use. For example, Jacob (1991) discussed the use of private sessions with postdecree disputants where extensive caucus use was seen as beneficial in exposing "blockages" to gain understanding of the nature of the conflict. Significantly, most of the family mediation models reviewed by Bowen (1999) used caucusing at least once per session, ideally equal time being allotted to each disputant. In addition, Parkinson (1997) observed that the decision to include a caucus stage varies according to the mediation model used, as a function of agency practice, directed at the perceived needs of participants, or as a result of mutual consultation between mediators and disputants. Furthermore, mediators might assume that each disputant may need to disclose or discuss their concerns in a private session (Tillett & French, 2006) or for each client to examine their separate positions (Rifkin, Millen, & Cobb, 1991). In addition, Fisher and Brandon (2009) indicated that mediators may identify concerns that might benefit from exploration, or issues that appear to be contributing to repetitive loops, a deadlock, or an impasse. Interestingly, Fisher and Brandon (2009, pp. 184-187 and pp. 29-30) observed that a significant value of caucus is in promoting negotiating equality and positive negotiating strategies and also to allow participants "time-out," while Taylor (2002, pp. 161-166) argued for the employment of caucus in circumstances where disputant safety issues are identified as a significant feature of the relationship. Although Haynes and Charlesworth (1996) saw limited use for private sessions except where specific difficulties presented, a study by Swaab and Brett (2007) concluded that pre-family mediation caucuses contributed to a reduction of interpersonal conflict. Importantly, the authors also asserted that caucusing should be used to establish trust between disputants and practitioners rather than to work through "agreement issues." Bush and Folger (1994, pp. 270-271) acknowledged the value of an environment where parties are able to raise concerns that they feel unable to in joint session, but at the same time, the authors were critical of sessions dedicated to the urging of disputants toward specific courses of action. The authors observed that practitioners should be adept in ensuring that caucusing assists disputants in the understanding of their own position while considering others' perspectives. Reviewing issues likely to benefit from caucusing, Moore (1987) and later Moore (2003, pp. 369-377) identified examples ranging from pressures arising out of preexisting problems in the clients' relationship to those emerging from the session in progress or as a result of the negotiation process. Pope (2001, p. 93) listed several general uses for caucusing such as reflection on the future and the importance of settlement, discussions around alternative options for settlement, ventilation of emotions arising from the dispute, mediator(s) acting as a "sounding board" prior to rejoining joint session, and examination of premature concessions or commitments. Significantly, comediators may also benefit from a private session to "take stock" of the joint session's progress, according to Fisher and Brandon (2009, p. 238). Furthermore, an important and difficult dimension likely to emerge in a session was identified by Saposnek (1998, pp. 126-129) and Taylor (2002, pp. 161-166) relating to situations where children are interviewed separately from each other or their parents.
Views vary as to the most appropriate time to hold a caucus meeting or under what circumstances (Taylor, 2002, pp. 161-166). Moore (2003, pp. 370-375) suggested that practitioners who break proceedings for a private session with one party should time the break appropriately: not too early, particularly where there is still the possibility of a productive settlement; and not too late when all is virtually settled. However, an early caucus is often indicated when it is considered that ventilation of an issue or emotional concern might aid negotiation flow. Despite there being no ideal time for a caucus break, the location, duration, and order in which the disputants are "seen" are generally subject to an agreed standard.
This study attempted to add to the relatively small amount of published research on the efficacy of caucusing by focussing on a critical review of its use in a family mediation context. The views of practitioners were explored on their rationale for using caucus and the potential for ethical concerns or dilemmas likely to arise. Accordingly, the central question addressed was "What are the special characteristics associated with the use of caucus?" An associated question was "Are there ethical implications inherent to caucusing?"

Method
The study was part of a larger project using mixed-method research (Hall, 2008), which utilized two sources of qualitative data collection: (a) a large 56-question questionnaire divided into 12 topic areas, 1 being the subject of this article; (a) several in-depth semistructured interviews.
For the questionnaire, the unit of analysis was family mediation practitioners and the sampling, as defined by Denscombe (1998, p. 15) and Hall (2008, pp. 194-196) was purposive. Invitations to participate in the project were posted to mediation agencies known to provide family mediation services (Bowen, 1999). In the invitation, the project was presented as involving a self-completion postal questionnaire (Hall, 2008, pp. 194-196) that would probe practitioners' opinions on a range of topics associated with family mediation practice. The 56-question questionnaire was subdivided into 12 main topic areas (Richards, 2009, p. 101) 1 of which being the subject of this article. There were 129 assenting responses resulting in 129 serial numbered packages being sent to agencies for distribution to willing practitioners. Each of the packages contained a covering letter introducing the researcher and the institution, an outline of the project and its significance, appreciation of respondents' cooperation in the project and assurance of confidentiality, and the questionnaire (with completion instructions) and a return envelope preaddressed to the investigator. Of the 129 packages mailed out, there were 47 completed questionnaires received by the investigator who was blind to the identity of respondents. The response rate was therefore 36%. The 47 respondents' locations were Australia (27), Canada (2), Puerto Rico (1), the United Kingdom (7), and the United States (10). The topic area reported here was titled "Use of Caucus" and consisted of two principal precoded questions.
Seven Australian mediation practitioners were also recruited as volunteer participants to participate in semistructured in-depth interviews (Hall, 2008, pp. 201-204) conducted by the investigator. As with the questionnaire, the sampling was purposive, the participants being known to the investigator through his work in family mediation and chosen for their "first hand experience and knowledge," an attribute which Rubin and Rubin (2005, pp.64-67) regarded as critical to a successful investigation. Inclusionary criteria were formal mediation training and participation in 400 or more mediation sessions post-training. The interviewees provided informed consent based on the cautionary elements expressed by Padgett (1998, pp. 35-36), and agreed to be audio recorded more than 1.5 to 2 hr. When the tapes were transcribed, the content was subjected to a form of thematic analysis (Grbich, 2007, pp. 16-32). A phenomenological style of interview was used, described by Roulston (2010, pp. 122-129) as useful in generating detailed explanation and meaning to interviewee experiences and permitting opportunities to explore special points of significance identified by interviewer and interviewees. Participants were asked to discuss their perception of the family mediation process and reflect on their experiences and understanding of the topic area. Discussion allowed for clarification, exploration, and congruent discussion of meanings and interpretations of emergent issues, which, in the case of this article, were associated with private meetings (caucus).

Results of the Questionnaire
This topic area ("Using Caucus") involved two precoded questions, grouped according to response category.
In the first question, practitioners were asked whether they used private sessions (caucus) "routinely," "sometimes," "never," or "rarely." All 47 participants answered this question, the responses being 23 (49%) "routinely," 21 (45%) "sometimes," with 3 (6%) stating they either "never" or "rarely" used caucus. The prominent feature of these responses was that 44 of the 47 respondents (94%) used caucus "sometimes" or "routinely." In the second question (consisting of two parts), the intention was to survey the respondents' view of the value of caucusing.
Part 1: Do you consider that the use of the caucus is valuable? Of the 44 (of the 47) who responded, 18 (41%) and 24 (55%) stated that using caucus was "always" valuable or "sometimes" valuable, respectively, with 1 respondent stating "mostly" and another stating "rarely." Part 2: The mediators were asked whether they considered it valuable, to whom was it valuable? Of the 44 who answered the question, 39 (89%) considered caucus was valuable to the mediation process, 36 (82%) considered it of value to clients, and 30 (68%) of value to the mediator(s). The survey exposed limitations on the use of the questionnaire tool in this case. For example, problems were associated with the rigid parameters set by the questions with some respondents stating that they opted for the nearest to their preferred answer.

Results of Interviews
Following transcription and thematic analysis, key interviewee assertions and opinion arising from discussions and selected dialogue were coded for like or linked responses and grouped into thematic clusters (Roulston, 2010, pp. 122-129).

Cluster 1-Aims, Benefits, and Values of Caucus and Caucusing
Despite some reservations, there was consensus among interviewees on the benefits caucus could offer disputants, such as to allow them to express their views without fear of interference or intimidation from the other disputant, enable a client to ventilate emotions in a safe environment, or "rehearse" for a return to joint session. In addition, there was agreement among interviewees that a major aim of caucusing was to gather information with a view to facilitating the management of difference or tension between disputants in joint session. Moreover, some interviewees observed that a topic discussed in private session often assumes a different meaning or connotation in joint session. An amalgam of suggestions as to the aims of caucus use included (a) to explore "real" differences between disputants, and how they are to be managed; (b) how disputants felt about each other; (c) how they regarded discussion of their concerns in joint session; and (d) what has been the reaction to those same concerns prior to attending mediation. Some mediators added that they utilize caucus to discover what compromises parties are prepared to make in agreement setting. In responding to questions about the value of caucusing, three interviewees asserted that they might be unable to facilitate moves toward resolution of issues if they could not equate what transpired in the joint session with discussion in private sessions. For example, one person stated that if you hear similar stories and they appear consistent in both cases, then you have a good idea that whatever is resolved, is going to be valid. If you have a person presenting differently than in joint session, then you have to pursue a different approach.
Responding to questions from the interviewer on the role of the practitioner, without exception all interviewees acknowledged the central role of "controlling the process."

Cluster 2-When to Introduce a Caucus Session and for What Length of Time?
Interviewees stated repeatedly that they offered parties equal time in caucus sessions, regardless of the discussion substance. Two people described how and when private sessions were used in their model, with one person noting that a "small" caucus lasting about 10 min was usually introduced 15 to 20 min into the session, whereas the other interviewee stated that if the decision was made to use a caucus break, she would do so about 30 min into session. Referring to the length of private sessions, an interviewee said that typically, about 15 min per person was offered once, sometime during a session. On the other hand, another person reported that in her agency, caucus was made available whenever needed. She said that if there was seen to be a power imbalance, for example, it might be appropriate to explore individual views on what disputants had tried previously. A mediator stated that he asks participants whether they would like a "break" or "to caucus." He would suggest a caucus where there was an impasse in the proceedings or any suggestion of abusive behavior in the relationship that might be "notifiable under law," for example, child abuse. Another interviewee agreed and added that the model she used usually had a private session, sometimes two, during the assessment stage: "For instance, if we decided not to go ahead with mediation, further down the track we would be inclined to have another private session to check how each person is going to handle this." Also in her model, an aim might be to find out what the disputants really wanted: "If we think counseling is a better option, be it separate or couple counseling, one person might say, 'maybe we'll be better off seeking counseling.' Under these circumstances, you would need to hold a private session with the other party." Answering the question of when and where he would use caucus, an interviewed mediator responded thus: I have used caucus in the beginning, in the middle, and at the end. The reasons were: in the beginning you use it in order to establish a reason for "shock remediation"; in the middle you might use it if there is expression of extreme emotion, or a "disposition" that needs to be clarified and perhaps confronted.
He explained that you can't really have people crying and the other party exploiting the situation, or also becoming upset. If you discover that the person is experiencing difficulties in accepting the system that you are using and maybe questioning it or the validity of the process, you have to suggest a caucus session. It often means that one has to start from the beginning and ask what's bothering them and why they feel the way they do "please tell us because we need to understand it." A caucus towards the end might be utilized in order to test the validity of what has been resolved or to discover how disputants feel about the session's content.
Another interviewee elaborated on the occasions for the use of caucus, noting that most often mediator(s) agree it is time for a session, although there is often conflict between practitioners as to the appropriateness of temporarily stopping the joint session at a particular spot, and holding separate sessions with the clients. When asked how she would manage conflict, she stated that if she wanted to delay the caucus, or needed to ask another question of the client(s), she would press ahead and delay caucus until she considered a more appropriate time! There was consensus among interviewees regarding additional situations likely to precipitate caucusing, including (a) a history of inadequate communication where one party habitually acquiesces to the other, or (b) a superficial level of communication where continuance of the relationship requires agreement at all costs. In private session, the other partner being absent might upset the arrangement thus allowing for concerns or fears to be verbalized.

Cluster 3-Empowering Mechanisms Associated With Caucus
One interviewee was circumspect in describing her use of caucus; nevertheless, she emphasized the importance of the caucus tool as being one of empowerment for participants who are then in a strong position in caucus to review aspects of their status. Yet another person also alluded to empowerment as a recurrent theme, stressing the usefulness of private sessions when an impasse has occurred: It is a very powerful strategy if we are "stuck," or if something happens which is inflammatory, or we've nowhere else to go but separate the disputants . . . we're in a better position to say I don't understand what's happening or what's going on, or it seems we're stuck . . . what's your view . . . So, it gives us a chance to check things out with people without doing so in joint session where they may not tell you what was going on without losing face. (emphasis added).
By adding the "loss of face" aspect, she underlined the way caucusing enhanced disputant empowerment. Similarly, she reassured the disputant of her impartiality: "That's where engagement (with clients [emphasis added]) comes in . . . being able to proceed in a way that the person isn't going to feel you're not on their side or that you're advocating for the other person." She stressed the value of caucus to elicit significant information: "Since you don't know what's happening with people until you ask them, it becomes a problem when they are reluctant to reveal much in joint session." Another interviewee's comment was similar when referring to the case of a couple who appeared to have settled the dispute and had merely come to mediation to have it written out! However, he and his co-mediator decided to continue with the mediation protocol anyway. It transpired that there were underlying issues that had not been addressed. It was in caucus that these were identified. His views were consistent with those of the other six interviewees with regard to the empowering effects of caucusing. Emphasizing its role in reassuring clients, he asserted that if used carefully, caucus had considerable ability to empower and afford the practitioner the opportunity to demonstrate their neutrality and impartiality. In addition, another interviewee added that by asking whether disputants would like to take a particular path or even a different approach was facilitating some sense of empowerment for disputants.

Cluster 4-Disadvantages in the Use of Caucus
Although mediators interviewed were, in varying degrees, supportive of the use of caucus, some saw cause for concern with its use. The concerns ranged from some negative aspects in specific circumstances through to significant concerns, prompting references to it being tantamount to unethical practice. One interviewee thought there were few negative aspects to using caucus and because it was enabling an ultimate agreement then, "the end might be said to justify the means." Conversely another interviewee was somewhat uncomfortable with the use of private sessions, having been initially trained in a model that used them only if considered necessary. She was critical of its use from the perspective of its potential for eroding the trust between mediators and disputants: I think that if you're encouraging parties to speak openly and honestly with each other, by separating them and then offering disputants opportunity to talk privately with me because there may be things you don't want to say in front of the other person, you're actually creating an element of distrust. It can be seen by the other party as quite destructive, and they might think "well, here am I being open and honest and I've been asked by the mediator to be so, while at the same time the mediator's saying, you don't have to be open and honest really, because you can tell me whatever you like, and it will stay confidential." So there's something in the whole process of private sessions that might break down that element of trust that you're hoping to build up.
Two interviewees acknowledged the positive aspects of caucusing but were also forthright on the negative perspectives, one person suggesting that the whole idea of caucusing was unethical and "smacked" of unprofessional practice, a view echoed by the other interviewee with particular concerns, who considered its use as a violation of the spirit of mediation where "everyone worked in concert aiming for a resolution to the dispute." Another interviewed participant also had reservations on the use of caucus, stating that it was "underhand to some extent and possibly contrary to basic tenets of mediation." One interviewee added that the whole idea was that everyone works toward a resolution "in the same room . . . and in an 'open manner.'" Significantly, another interviewee saw disadvantages that particularly related to the feelings of the party not in the private session, thus creating the possibility of erosion of trust: "I think it can contribute to suspicions being raised such as, 'what are they going to say when I'm out there.'" She also voiced concern about unilateral decisions to enter "caucus" (i.e., by one mediator): or held at the wrong time. She said that an example might be if one mediator suddenly say's "oh look, I think we'll have a private session now" and hasn't checked with the co-mediator who might feel the session is just getting somewhere. Sometimes we don't say we are going to have a private session, we promptly call a break! When pressed by the interviewer as to why, having such negative views, interviewees continued to use private sessions, typical responses ranged from it being part of the agency model through to stating that in most (not all) situations, the positive aspects outweighed the negative ones!

Cluster 5-Confidentiality During Caucus Sessions
All interviewees asserted that discretion and confidentiality are central to the spirit and ethos of mediation practice. This was an important aspect of the interviews because (a) confidentiality of proceedings is a criterion for successful, ethical family mediation; (b) most codes of ethics of professional bodies emphasize confidentiality. While all interviewees acknowledged the central importance of neutrality and impartiality to the caucus milieu, questions about client empowerment and preservation of confidentiality during and after sessions garnered some specific observations. For example, two people interviewed noted that there was no control on out-of-session disclosure of confidential material from either joint or separate sessions, confidentiality being limited by the desire of all participants to preserve it. Another interviewee observed that after a private session and upon resumption of joint session, it might be beneficial for disputants to disclose those key aspects of the caucus, which may be of assistance in the formation of settlement. In contrast, a particular interviewee emphasized that in her model, breaking of private session confidences was not tolerated, with the exception of statutory notification of child abuse. Another person also cited statutory obligations, affirming her strong belief in the value of confidentiality thus explaining her reluctance to use the caucus tool, holding that its use should be discretionary. She said that those who favor private sessions will say that there may be many people (disputants [emphasis added]) in mediation who feel quite intimidated by the other party who, despite being encouraged to be as open and honest as possible, don't feel they can be so because of: a troubled "history," feeling too nervous, not being at their "best," or the other party being constantly dominant.
She continued by observing that "there may be a number of reasons why in a private session, one party opens up in a way that they haven't been able to in a joint session." She said that no matter how skilled a mediator is, if there is a behavioral aspect present, recent "history" between the parties, or simply a concern for the future, some parties feel that they cannot "disclose" in a joint session with their former partner present, a private session may then be beneficial.
However, she said that she left the option open ended, by contrast to those who applied subtle pressure to include caucus: "There are some mediators who will say that we can have a private session if you wish; or one party might request one."

Discussion and Conclusion
Responses to the questionnaire demonstrate that most respondents used caucus and that they regarded it as valuable to the mediators, the clients, and also to the mediation process. Results from the interviews indicated that despite some strong reservations, there was mainly agreement that adroit use of private sessions enables mediator(s) to perform supportive functions such as exploration of disputants' concerns; in this, the results were sympathetic to existing literature, such as discussed by Pope (2001). Private sessions were seen to create a "safe" environment which was defined as a discussion forum without interference from their codisputant, possible "loss of face," or an erosion of the negotiation phase as is possible in joint session. Interviewees indicated that any concerns disputants may hold about the confidential nature of caucus are to a large degree contingent on adequate explanation to them by mediators, about its use and the reasons why it may be used. Concern surrounding preservation of confidentially was identified by some mediators as in need of cautious treatment. A private session may be at the behest of one or all of the participants in a session. Some mediators who participated in this study used caucusing to consider strategies to influence the course of the session, to learn what compromises parties were prepared to make in the process of agreement setting, or to better understand an individual disputant's perspective. It emerged from the study that private sessions enable examination and exploration of concerns surrounding, for example, future care of children which may be seen by a disputant to be more appropriately aired privately before revealing options in joint session. The absence of a dominating partner allows the party in caucus to be heard by a mediator without pressure from the other party to precipitate a decision or agreement under duress. Important additional benefits to be derived from caucus include improvement in the negotiation process, often by the relief of tension between clients and the prevention of premature concessions, and using it as a "sounding board" to check out what the parties are willing to agree to in the journey to a fair and just outcome. Modeling of skills such as active listening by practitioners is also likely to be noted by disputants. Although there are occasions where disputants may call for a private conference break, the mediator(s) in "controlling" the process are the most likely of the participants to actually do so. It is important to recognize the point behind caucusing in this context. The evidence from this study and the literature suggests that caucusing is focused to a large extent on the pursuit of justice for disputants, with justice being seen as a central tenet of mediation. Importantly, caucus can be exploited by a party to reassess their position, even though sometimes as a prelude to undermining session progress. More positively, they may use it to evaluate the way the session is proceeding for their co-disputant, or to consider ways of establishing a more conciliatory approach to issues. A constructive aspect of caucus is that its skilful use by practitioners aids the facilitation of empowerment. This is accomplished by allowing each party to clarify their goals and desired outcomes, reinforce their position, and explore options and feasibility of implementation through reality testing, without exposure to possible intimidation from the other party.
In addressing the associated research question, a range of questions emerged from the study relating to the use of caucusing. Not least of these is in fact connected with empowerment of disputants. If the aim of caucus sessions is to provide a safe and informal avenue for ventilation of concerns, or disclosure of sensitive material, and because the disputants belong to an informal system, how does it affect the other party? This might explain the reluctance of mediators from a "systems" perspective embracing the tool (Taylor, 2002, pp. 161-166). Is there a possibility that empowerment of one party may be at the expense of the other? How can mediators be certain that by seeing both parties separately in a nonthreatening (safe) environment where the discussion is confidential, the parties will be empowered as an automatic result of caucusing? In addition, some ethical concerns surfaced from the study. For example, there may be a form of coercion involved if caucusing is either used routinely or even suggested to clients, however, gently. Furthermore, such coercion may be instrumental in the erosion of trust engendered during or prior to the engagement phase of the process. This might compound the coercive effects from subtle pressure applied to a disputing party such as suggesting that it might be in their best interests to negotiate a settlement rather than pursue a particular path. It is worthwhile pondering on the possibility that coercive processes might be initiated involuntarily prior to a session being planned, because pressure is on both parties to attend mediation to avoid the risk of "losing out." Coercive practice is thus likely to continue throughout the whole session with caucusing playing a significant part, particularly where mediators initiate one or more breaks during a session. Indeed, it is equally plausible to argue that mediators interrupting sessions to caucus is redolent of disarray or dissent when viewed from the disputant position, or even a level of disrespect for the spirit of the process. In the interviews, confidentiality featured strongly difficult since it relies on all participants preserving it. Conversely, it may be argued that caucusing allows for the modeling of dynamic resolution procedures. This then begs the question "Is mediation even applicable in cases where private sessions are considered necessary?" Perhaps counseling is more appropriate. A significant aspect of private sessions is that since family mediation is regarded as an open and honest negotiation process where issues of conflict are mutually aired and ideally, resolved, private sessions may well be seen as the antithesis of conciliation or reconciliation.
Addressing the basic questions set for this study, the conclusions are encapsulated by stating that caucus includes features that are crucial to the ethos of family mediation with impressive support from respondents and interviewees. In the pursuit of justice for clients, there is a constant focus on the clients' potential autonomy. Caucus helps create a situation in which autonomy is recognized, facilitating disputants' arrival at a firmly and fairly negotiated agreement with clients being encouraged in independent decision making. Caucusing at best allows for free autonomous expression in a safe environment and as a mediator of equity in social justice terms. At worst, it is an opportunity for the erosion of trust and is the antithesis of empowerment and a perpetrator of unethical practice, with its extended use appearing to create a series of ethical dilemmas. Family mediation is concerned with the pursuance of informed, joint decision making; accordingly, it behooves practitioners to be aware of the potential for negative unethical effects associated with private sessions.

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.