Mediator as a Story Recipient in Narrative Mediation: What Types of Questions Should I Ask?

This study adopted the theoretical framework of narrative mediation to investigate a storied conflict talk between a landlord and her tenant in which the mediator played the role of a story recipient in the co-construction of stories with disputants. The focus of this research is on the function of questions posed by the mediator in the production of turning points which are favourable to the evolution of “better-formed” stories. The results of this study indicate that there are at least two types of questions mediators ask: 1) the questions that can help disputants reflect on their imperfectness; 2) the questions that awaken disputants’ memories of their good stories from the past. It is shown that the de-legitimacy for Self laid a foundation for the production of a good story towards a meaningful outcome for the mediation. The inadequacy of the context formed by first having legitimacy for Other followed by the de-legitimacy for Self led to the failure of destabilizing the problematic story in the mediation. The lack of the dominant party’s legitimacy for Other resulted in the absence of legitimacy from the marginalized side and would likely cause unfavourable consequences to the mediation in the long term.


Introduction
This study of narrative mediation draws inspiration from the Allegory of the Cave in Plato's work: Republic, in which prisoners chained to the wall of a cave were unable to turn their heads. All they could see in their lives were the shadows projected on the wall from the puppets held up by puppeteers passing in front of a fire behind them. These prisoners took the shadows as reality for themselves.
Through the theory of Forms, Plato argued that the "non-material abstract known to us through sensation, possess the highest and most fundamental kind of reality" (Cheng, 2014). Disputants in conflict like prisoners in the cave grasp the matter at the core of the story through contemplating its form presented in the narrative.
And in the narrative approach, we explore the relationship between the story and its narrative which reflects the connection between matter and its form. Regarding the question: "What do we know when we know about a person?", McAdam (1993) proposed that we know a person from his or her own life stories and the stories the person lives by. However, we have no direct access to the stories or the "truth" in people's minds. We can only know a person through his or her narrative or their way of describing stories in their lives. Varying ways of storytelling creates different forms of "truth".
Over the past decades, an increasing body of research has focused on studying mediation through the narrative approach. This approach provides a new angle for mediators to take mediation as a process of storytelling in which the disputants' stories can be told and the mediator leads the narratives along the direction of "better-formed" stories until the arrival at the settlement. The disputing parties are the chief storytellers in mediation. On the other hand, mediators, as story recipients, listen and actively co-author stories with disputing parties, while simultaneously using their metacognitive resources to monitor the whole process of storytelling.
Nonetheless, limited research has been conducted to study the choice of questions in narrative mediation. Based on Brandon's (2011) work, "questioning is what a dispute resolution practitioner does most". Therefore, we can see that it is beneficial to investigate the impact of the questions asked by the mediator as he or she plays the role of a story recipient in narrative mediation.

Mediation Approaches
The present study focuses on a case of interpersonal conflict, that is, the conflict between at least two persons, in the institutional setting of a mediation center. Conflicts occur in all human relationships, societies, and cultures. Conflicts become intractable when people in conflict are unable to put distrust and animosity aside and work together to develop solutions that can satisfy their individual as well as shared needs and interests (Moore, 2014). According to Morasso (2011), "mediation is a reasonable way out of conflict". It provides disputants with opportunities to create openness, trust, and understanding with the facilitation of a neutral third party (Hyde & Bineham, 2000). Any attempt at understanding conflict resolution through mediation cannot be successful without first studying the approach taken by the mediator. There is a diverse range of mediator styles and one of the first attempts to construct a framework encompassing the practices in mediation was through Riskin's grid (1996). Although Riskin's grid had been criticized for its deficiency in the illustration of goals of mediation and the roles mediators play (Kovach & Love, 1998;Riskin, 2003), it nevertheless served as a plank in the platform for mediators to clarify their roles. This grid consisting of four quadrants showing mediators' "orientations", when proposed by Riskin (1996Riskin ( , 2003, was intended to generalize, describe and categorize different approaches to mediation. Researchers have also described the general pathways in negotiation as delineated by the "adversarial approach", "problem-solving approach" and "transformative approach" (Moore, 2014;Wilmot & Hocker, 2011). The transformative approach is favored by mediators for its advantages in exploring deeper identity and relational concerns of two juxtaposed positions to solve the conflict in the long term instead of obtaining a written settlement agreement only (Bush, 1994;Stewart & Maxwell, 2010;Wilmot & Hocker, 2011).
No matter which approach the mediator uses, the result of mediation is the joint product of the choices of disputants with the intervention of a mediator, not the result of isolated tactics employed by either disputants or mediators (Stewart & Maxwell, 2010;Wilmot & Hocker, 2011). Moreover, the complexity of reaching fruition in conflict resolution is compounded by the emergent nature of conflict. As such, the goals of the disputants are not prepackaged and will probably shift in the process of mediation (Stewart & Maxwell, 2010;Street & Cappella, 1985).
To further improve the practices of mediators, more research is therefore needed to explain how disputants and mediators interactively co-construct conversation that shifts the conflict from the extreme of intractable to the other end of tractable.

Narrative
All human communication can be understood as narratives of stories in our lives (Fisher, 1989;Rifkin, Millen, & Cobb, 1991). As people participate in the conjoint telling of stories, relationships, identities, and institutions are established (Cobb, 2006). The communication between people is not just the exchange of information, but the process of understanding others' experiences described in the form of narratives (Gubrium & Holstein, 2008).
The concept "narrative" was adopted by many researchers to study the different ways people use to describe the same story. Through the notion of narratives, the narrative approach also affords a way to understand the dynamics of interaction in mediation (Cobb, 2006;Rifkin et al., 1991;Stewart & Maxwell, 2010;Winslade & Monk, 2000).
In the study of narratives, Riessman (1990) noted that people have no direct access to the experiences in our minds. Furthermore, many researchers have proposed differences that separate stories and the process of storytelling (Schiff, 2012;Gubrium & Holstein, 2008;Schiff, 2012Schiff, , 2017 of the characters coming out from the retelling of the same story arises from people's choices of the materials in their memories and the order they organize these materials. Our ways of telling stories have the power to shape the things we describe, change our viewpoints and construct our relationships with others (Winslade & Monk, 2000). The "narrative is an expressive action used by people to present their life experiences and understanding of life" (Schiff, 2012).
Some researchers further argued that for a full understanding of the narrative, it is insufficient to merely examine the internal organization (Gubrium & Holstein, 2008). Hence, narrative ethnography was proposed to emphasize the external and contextual organization of narratives (Gubrium & Holstein, 2008). The environments of storytelling, where the telling of stories happens, shape the content and the internal structure of stories (Gubrium & Holstein, 2008). The storyteller also shapes the materials from their memories into a verbal performance designed for the current context (Norrick, 2000).

Narrative Conflict
Language is not used for information exchange only but as a means for joint intellectual activity (Fisher, 1997;Mercer, 1995Mercer, , 2000. The consequences of how stories are told and interpreted by the other side (Gubrium & Holstein, 2008), may create a harmonious or tense relationship between interlocutors in storytelling.
The tense relationship may escalate to conflict if interlocutors insist on holding the binary, entrenched and polarized position constructed by their social realities (Stewart & Maxwell, 2010). These realities are the reflection of their understanding of the social world and the cultural value systems they live in (Stewart & Maxwell, 2010).
Using language, stories emerge as linguistic phenomena for the narrational understanding of Self. The study of Self is essential for researchers to understand how interlocutors understand themselves in conflict talk (Gubrium & Holstein, 2008). Some researchers make a distinction between the personal self and the relational self (Gubrium & Holstein, 2008;Mead, 1934).
The study of the relational self in narrative research has flourished since George Herbert Mead's (1934) study about the relationship between mind, self, and society. Compared with the personal self, the relational self is more flexible rather than located (Gubrium & Holstein, 2008). Relational self is constructed and shaped in communicative relation to others (Gubrium & Holstein, 2008). The study of relational self focuses more on how self-understanding develops and transforms through interactions in social contexts (Gubrium & Holstein, 2008). Thus, the current study of self is the study of relational self instead of personal self as the narratives are dynamically co-constructed in mediation. Cobb (2006) pointed out that at the core of the narrative conflict are the constructs of legitimacy for Self and de-legitimacy for Other. Based on previous research, Dorcas Quek Anderson (2018) concluded that it is necessary to give disputants a perception of fairness through "the opportunity to voice their views, feel heard by the mediator and the other party and be accorded respect and a sense of dignity". These are important aspects that contribute to construction of an environment for disputants to shift their boundaries and perspectives through de-legitimacy for Self and legitimacy for Other. Hence, narratives in conflict talk, are used as the means, by which disputing parties privilege their centrality (Cobb, 2006). The Self in an entrenched conflict intertwines with the over-simplified stories that each side holds to construct the Other with bad traits (Cobb, 2006;Dirth, 2015). These stories are closed, characterized with a black-white description of the Other and premature solutions in which disputants are unable to take a step back to share the set of rules the other side used to judge a proof (Littlejohn, 1995;Steward & Maxwell, 2010).
In conflict narratives, disputing parties often use specific terms to enhance the legitimacy for Self and de-legitimacy for Other (Cobb, 2006). Each term holds a problematic story repeated by interlocutors over and over again. Every narrative is the description of the stories that cannot be circulated and elaborated by marginalized groups (Cobb, 2006;Monk, 1996).

Narrative Mediation
The purpose of narrative mediation is to ensure "one party's story is not privileged over another" (Stewart & Maxwell, 2010), liberate the oppressed side from dominant stories, and allow the elaboration of the disputants' stories. In narrative mediation, disputants are granted with the equal rights to utilize stories in which their own experiences and motivations are given airtime to defend their positions and correctness of their perspectives (Gergen & Gergen, 2006).
Truth is not the focus and critical criterion in narrative evaluation (Stewart & Maxwell, 2010). The disregard of truth creates an intersubjective space for both sides to write for themselves and is more empowering (Wylie & Pare, 2001). An individual can only genuinely change his or her perspectives when self-reflection happens (Isaacs, 1999). The problematic stories can be turned into better stories when the elaboration of stories is done based on shifted self-awareness (Cobb, 2006). For every term offered by disputants to enhance their legitimacy, it is always possible to find a linguistically opposite term with the same measure of legitimacy (Cobb, 2006). To discern and address the causes of conflict, the task of the mediator, in narrative mediation, is to help disputants find their imperfectness through the terms they provided (Cobb, 2006) and their way of describing experiences.
The "ironic" move is, first and foremost, the most crucial move in narrative mediation (Cobb, 2006;MediatorAcademy, 2016). In the videos provided by MediatorAcademy (2016), Sara Cobb gave a more detailed explanation about the "ironic" move, she proposed in her paper "A narrative perspective on mediation: Toward the materialization of the 'storytelling' metaphor". The "ironic" move enables the construction of Self as less than perfect. This Self in the "ironic" move is a representation of the shifted self-awareness about one's identity and position in conflict talk, which can lay a solid foundation for the further establishment of intersubjectivity and the final settlement of the conflict in mediation.
The critical moment from legitimacy to de-legitimacy for Self or de-legitimacy to the legitimacy for Other is called a Turning Point (Cobb, 2006). A new reciprocal narrative logic can help disputants resolve conflict in the long term (Cobb, 2006). From a conflicted narrative to a "better-formed" storyline, there are three dimensions that mediators need to consider for the realization of narrative transformation: plot, character roles, and themes (Cobb, 2006). Sara Cobb (2006) suggests that there are five stages that a mediator can follow to facilitate the production of "better-formed" stories (See Figure 1). Each stage contains a corresponding turning point to change either of the three dimensions mentioned above to destabilize the tightly woven story held by each disputant. These five turning points together contribute to the creation of a "better-formed" story, transformation of disputants as people and positive evolution of a relationship (Cobb, 2006).

Figure 1. Turning Points in Narrative Mediation
Note. This figure is from "A Developmental approach to Turning Points: Irony as an Ethics for Negotiation Pragmatics," by S. Cobb, 2006, Harvard Negotiation Law Review, 11, 147-197.   Note. This figure is from "A Developmental approach to Turning Points: Irony as an ethics for Negotiation Pragmatics," by S. Cobb, 2006, Harvard Negotiation Law Review, 11, 147-197. Without making some shifts with regards to the characters, an excellent story that is conducive for the resolution of conflicts does not come into existence with only the shift in plot and themes alone (Cobb, 2006). Among these five stages, the first two stages focus on the change of characters in the problematic stories. These two stages are: 1) High in Legitimacy for Self and Low in Legitimacy for Other; 2) Moderate in Legitimacy for Self and Other.
The Turning Point 1 corresponding to the first stage reduces legitimacy for Self and the Turning Point 2 leading to the second stage increases legitimacy for Other. Stage 1 is the prerequisite for the generation of the Turning Point 1. At the second stage, however, the Turning Point 2 generates the stage (Cobb, 2006).
The current study will focus on how the mediator helped the disputants generate these two turning points in the process of the narrative mediation.

Mediator as an Evaluator or/and Facilitator
The value of de-legitimacy for Self and legitimacy for Other in narrative mediation is its self-reflective process in which parties can look back at their problematic stories. Only by self-reflection, can there be a greater chance for the destabilization of the hegemonic control of old narratives. The plot, character roles, and themes interactively emerge through "turns" in mediation to become "better-formed" stories that can make the proposals of both sides more reasonable. Many researchers have studied the role that a mediator plays in the process of mediation (Cobb, 2006;Riskin, 1996;Stewart & Maxwell, 2010;Wilmot & Hocker, 2011). Some researchers proposed that the mediator should be the evaluator in mediation who provides assessments, predictions, or directions directly to the parties.
In contrast, other researchers proposed that the mediator should be more like a facilitator who creates a conducive environment for disputants and leads them to create a resolution suiting their personal as well as mutual interests (Bush & Folger, 2004;Stewart & Maxwell, 2010;Wilmot, 1995). The advocates of the facilitator role of a mediator believe "the more mediators ignore disputants' relational concerns, the more difficulties they will experience in reaching agreement" (Donohue, Darke, & Roberto, 1994).
In evaluative mediation, the role of the mediator is more like a teacher in small group discussions who encourages the parties to develop their understanding and outcomes. As Riskin mentioned (1996), this facilitative approach produces a degree of education or transformation in which the parties experience the process of "empowerment" and through which they strengthen their ability to deal with the problems and choices in life.
The grid proposed by Leonard Riskin (1996) for describing the mediators' approaches to mediation breaks the dichotomy of categorizing the mediator's role as either facilitative or evaluative. He emphasized this dichotomy does not give appropriate recognition to the interactive and dynamic nature of mediation (2003). The roles mediators play ranges from one extreme of a facilitator to the other end of the evaluator in a continuum and depends on participants as well as the current situation of mediation (Riskin, 1996(Riskin, , 2003. However, Riskin (1996) mentioned that the static quality of his framework and emphasized dynamics and flexibility of using his framework in practice.
While a significant amount of research has been devoted to examining the mediator's role in facilitative approach, evaluative approach and Riskin's grid, few studies have been conducted to identify the role that mediators play in narrative mediation. As narrative mediation shares the characteristics of transformative approach (Bush & Folger, 1994;Hansen, 2003), some researchers define the mediator as a facilitator for storytelling (Rifkin et al., 1991).
Narrative mediation also shares some characteristics of a problem-solving approach. The stories in mediation do not develop spontaneously on their own, because mediators have already had different versions of good stories in their hand at the beginning of mediation. Mediators lead the problematic stories that each side holds along the good storylines to the final settlement.

Mediator as a Story Recipient in Narrative Mediation
Although mediators take on other roles which are specific in the study of narrative mediation, such as managers of the storytelling process or an active participant in the co-construction of the narrative (Rifkin, 1991), limited research has been conducted to examine the mediator as a story recipient in narrative mediation. In the following part, we detailed our identification of the mediator as a story recipient in narrative mediation. Narratives require at least two members of the human community to participate in the process of storytelling (Nair, 2004). Norrick (2000) criticized the narrative theory proposed by Labov and Wletzky (1967) for their ignorance of conversational emergence of oral narratives. The conversational narrative was proposed to study storytelling by focusing on the verbal performance of both the storyteller and recipient in storytelling (Norrick, 2000).
Storytelling is not a fixed phenomenon but a dynamic and interactive process in which both teller and recipient co-construct the story together. However, "listeners as a group have generally been neglected in language research" (Gardner, 2001). Many studies have been conducted to research the roles that listeners play as story recipients in interactive talk (Gardner, 2001).
Many researchers emphasized the recipients' participation in the co-construction and their co-authoring of stories (Goodwin, 1986;Monzoni, 2005;Sacks, 1992). The idea of the co-authored narrative was proposed to study the co-presence of the listener in storytelling (Nair, 2004). Narrative mediation is a joint intellectual activity in which people (mediators and parties) use language to reflect on their actions and create new experiences together.
In mediation, there are at least one mediator and two parties' participants. The disputing parties, as main tellers in the mediation, bring their problematic stories to mediation to enhance their entrenched positions.
Mediators, as story recipients, do not have a simple job due to the following reasons.
Firstly, mediators do not only listen and respond to these stories but also plan, control and evaluate the whole process of both parties' narratives. The story teller "introduces the story to attract recipient's attention, gain control of the floor and ensure recipient's understanding" (Norrick, 2000). During the storytelling, recipients may interrupt and comment on a story and try to change the direction of story development by providing an appropriate response at an appropriate moment (Kim, 2016).
Secondly, mediators engage themselves in mediation with their past experiences or stories; and even actively participate in storytelling and co-author the story with the previous teller (Norrick, 2000). They lead each party through turns to move from their original focus to the less problematic stories created by the new meaningful narratives. While some turns lead to the crucial evolution of positive stories, some turns devolve negatively into stories that obstinately refuse to break the old reflexive reponses in conflict.
The role of mediators as story recipients is not well documented. Although there has been considerable work on the discursive moves adopted by disputants and the function of moves in conflict talk (Stewart & Maxwell, 2010), the research on the linguistic and conversational tactics, used by mediators to influence the development of the problematic stories held by disputing parties remains limited. This study is an attempt to investigate the situated communicative practices and mechanisms by which mediators as story recipients create new narrative logic with disputants.

Data
The video (see Acknowledgements) adopted in this research is about a landlord-tenant mediation in which one landlord (Linda), one tenant (Dennis) and mediator (Terry) participated. The background of the video is a tenancy dispute which arises from Dennis' accrued rent of 2.5 months. Dennis is a tenant as well as an old friend of Linda for four to five years. He is jobless and depends on unemployment benefits to live, and it is hard for him to find new accommodation and relocate during winter. Linda requires Dennis to pay for late fees, court fees, mediation fees, double damage fees on the top of rent. Both parties face a dilemma. Dennis faces the risk of having an eviction on record, and Linda may suffer impairment loss on her rental income without an agreement reached in mediation.

Steps of Analysis
Drawing from previous studies (Moore, 2014;Lovenheim & Guerin, 2004), it is noted that the mediation process can be described in six parts which allow disputing parties to communicate and interact with each other to resolve their differences with the assistance of the mediator.
The six parts of the mediation process are: Referencing the above six components, the mediation discourse of this study has also been divided into six parts.
Many studies have been conducted to investigate the individual strategies that mediators employ in the six parts of the mediation process. However, few studies have been conducted to analyze the discourse of mediators. As such, this gap framed the first research question in this study: What discursive mechanisms were adopted by the mediator to generate turning points?
A preview of the transcript indicates, from Part A to Part D, most of the mediator's turns are questioning turns. A table was made to present the proportion of questioning turns adopted by the mediator as a discursive move, to make our observation more quantitatively accurate.  In Table 1, the questioning turns adopted by the mediator accounts for 55.17% of the mediator's total number of turns. The percentage of questioning turns was more than 50% from Part A to Part D.
The percentage of questioning turns in Part A's opening statement was 80%, which was the highest among these four parts.
At Part D (private caucuses), the questioning turns the mediator adopted with Linda in caucuses (78.57%) was higher than that with Dennis (53.66%).
The percentage of questioning turns in Part E was 38.46% and 28.57% in Part F, which were both lower than 50%.
The percentage of questioning turns at Part F was 28.57% and this was the lowest among the six parts.
All the turning points (Turning Point 1 and Turning Point 2) appearing in the six parts were identified respectively.

Findings
At this point, we have answered the first research question. Table 1 and Table 2 shows that questioning was the most frequently used discursive move by the mediator. The occurrences of Turning Point 1 and Turning Point 2, in this case study, were all triggered by the mediator's questions. The findings are substantive clues for us to explore further about the function of the mediator's questions in helping disputing parties realize the shift of self-awareness and connotating legitimacy for Other. The current study examines not only the questions used to generate turning points, but also those questions having other functions. Finally, the study also investigates how these questions are employed by the mediator to co-construct new narrative with disputants to reach resolutions that are satisfactory to all disputants.
Questioning skills are a prerequisite in the selection, training, and development of teachers, interviewers, therapists and related fields (Hus & Aberšek, 2011;Dillon, 2004;Padesky, 1993;Moston & Engelberg, 1993). In mediation, there has been considerable research on the importance of the mediator's questioning skill in achieving success in mediation (Bennett & Hermann, 1996;Lehman & Page, 2007).
Unspoken needs are ignored by the mediator who does not elicit the narrative necessary to form the contextual background upon which a resolution arises (Bennett & Hermann, 1996).
Neutrality underpins the practice of the mediation and is an essential quality of a mediator as well as a stronghold against bias in the process of intervention (Domenici & Littlejohn, 2001;Riskin, 1996).
However, many mediation practitioners are confronted with difficulties in practicing neutrality in the facilitation of making mutually accepted solutions (Rifkin et al., 1991). The problems in practicing neutrality are embedded in the inevitable conflict and paradoxical relationship between impartiality and equidistance, two aspects of neutrality (Rifkin et al., 1991).
On top of these issues, poor questioning techniques prevent parties from reaching a mutually satisfying agreement. In contrast, a good question can maintain the mediator's neutrality and still encourage parties to express their feelings and the facts in their minds (Lehman & Page, 2007).
However, limited studies were conducted to examine the function of question-asking in maintaining neutrality, which is hard for mediators to achieve in practice. Besides, previous studies on the mediator's questioning skill is static, that is, these studies were conducted to study questioning skill as an individual tactic. Few studies have been conducted to study how the mediator employs question-asking dynamically to co-construct talk in mediation. Nearly no research has been conducted to study how mediators, as story recipients, use questioning skills to transform intrinsically intractable conflict talk via the recipe of a "better-formed" story to reach resolutions that are satisfactory to all parties.
Many researchers have noted the importance of question-asking in narrative mediation (Cobb, 2006;MediatorAcademy, 2016;Nago & Page, 2005;Rifkin et al., 1991). However, no systematic research has been conducted to study the application of questioning skills in narrative mediation. Therefore, the second research question in this study is: How does the mediator uses question-asking techniques in generating the turning points in Sara Cobb's (2006) systematic approach to narrative mediation? Narrative mediation is greatly influenced by how the mediation session is organized and the interventions made by the mediator (Rifkin et al., 1991). "Strategic question asking is the key to implementing narrative mediation" (Nago & Page, 2005). In answering the "how-to" in generating turning points, Cobb (2006) mentioned that one of the essential methods is the "formulation of questions that functions to connote others positively". The answer to "What types of questions to ask?" (Rifkin et al., 1991) is to understand what question influences the unfolding of the story in narrative mediation.
The current study analyzes how questions posed by the mediator generate turning points. The function of the questions asked by the mediator as listed in Table 2 is further explored in the formation of "better-formed" stories. Thus, we attempt to answer the third research question which is: How do the turning points help the mediator co-construct the narrative with disputants to reach resolutions based on "better-formed" storyline?

Discussions
This study adopted Sara Cobb's (2006) theoretical framework on how to apply the five stages in achieving success in narrative mediation. There is one critical turning point, corresponding to each stage of the mediation process, which is essential for the mediator to create a cooperative climate, destabilize dominant stories and overcome relational problems. Sara Cobb emphasized the first two stages and their corresponding two turning points for their importance in facilitating disputants to change their entrenched positions. A long-lasting and mutually beneficial agreement is not attainable without de-legitimacy for Self and legitimacy for Other. In this article, we will focus on the first two stages and study how the mediator's questions trigger their corresponding turning points.

High in Legitimacy for Self. Low in Legitimacy for the Other
Conflict arises from the differences between how we describe our stories and and how others interpret our stories. The differences signify a divorce between "the desired path and actual or potential obstacles" (Stewart & Maxwell, 2010). Closed systems and structures often feature in the problematic stories in conflict talk. The plots, characters, and themes are tightly woven and are not easy to be reorganized to create a new circular narrative. According to Cobb (1994), the high legitimacy for Self and low legitimacy for Other form the root of the problematic story. As can be seen in The purpose of the first stage in the mediation is to create a positive climate and a comfortable place to encourage each side to work cooperatively towards a settlement (Moore, 2014). An open question was adopted by the mediator to encourage the flow of information and encourage parties to expand their stories and engage themselves in mediation (Lehman & Page, 2007).
Linda responded to Terry's two successive questions ("Did you…?", "why is that?") with two negative replies ("I don't know...", "I don't think"). Instead of giving the expected answer "yes" or "no", Linda responded to the mediator's closed question "Did you have any questions about orientation" with a negative statement to show her skepticism even as she attended this mediation.
The mediator applied two same open questions ("why is that?") to explore the story behind Linda's show of reluctance. This open question is an invitation for Linda to tell the mediator the reason behind her low expectation for the outcome of the mediation. The second question allows Linda to communicate her subjective perception of her situation. She expressed her disappointment in her failure in talking with Dennis ("I tried talking to him before").
In the initial story, Linda constructed herself as a landlord who had tried asking for rent to be paid. The statement "I tried talking to him before", was used to portray Dennis as the tenant who delayed his rent even after the landlord's repeated attempts to negotiate with him. In the opening statement, Linda said to Dennis, "You owe money and a lot of it and you know that and I've been waiting a long time for my money". The coordinating conjunction "and" was used four times within one sentence by Linda to attribute negative intent to Dennis. Example 3 is the excerpt of Dennis' opening statement. From Dennis' description, the picture of the relationship between Linda and Dennis became clearer. Dennis applied two negative statements "I can't move" and "I don't have a job" to express his attitude towards eviction and legitimize his delay in rent.
Dennis represented himself as a tenant whose traits contrasted to those of the antagonist in Linda's story.
His delay in paying off his rent is due to the current circumstances of his unemployment, and he had tried to find a job instead of staying at home.
At the stage of the opening statement, disputants can describe the dispute respectively (Moore, 2014).
While one person is speaking, the other person is not allowed to interrupt (Moore, 2014). However, in Dennis's opening statement, Linda interrupted Dennis' talk without getting permission from the mediator.
Linda's aggressive action arises from the differences between the image of the tenant in her description and Dennis' description. Despite Linda's interruption, Dennis was given a chance in his opening statement to "defend against the first storyteller" (Stewart & Maxwell, 2010) and "the theory of responsibility" (Stewart & Maxwell, 2010) constructed by Linda's narrative.
In this landlord-tenant relationship, the landlord was the party with the dominant story as the tenant was the marginalized party who delayed rent. Linda's story, to some extent, is privileged by the social moral system to be more "true" than the story of Dennis. Besides, Linda, as the first speaker, "is granted a powerful advantage in the ability to frame the dispute" (Cobb, 1994).
Inadvertently, the "truth" as forged by Linda's narrative and the social morality which availed her position as dominant power may have an impact on the practices of the mediator. Yet, in the narrative approach, mediators are encouraged to balance the power between both sides and make the untold story to be told instead of emphasizing "truth". The lack of turning points at this stage is a result of Linda's emphasis on high legitimacy for Self and de-legitimacy for Other. question "what you would like happen today?" was used by the mediator to invite Linda to express her purpose in coming to this mediation. As can be seen from Linda's response "I like my money!", Linda's sole interest in this mediation is to take her money back from Dennis. This assertive statement constructs Linda as the aggressive landlord who is determined to get rent back. The adverse effect of this opening question is that it may increase the disputants' frequency in repeating their problematic stories which will further drive the wedge in the disputants' soured relationship.
The goal of "what does each person want?" is called a topic goal in conflict by Wilmot and Hocker (2011). While the topic goal is easy for disputants to talk about and for mediators to discover, the "relational goals remain tacit and unspoken" (Wilmot & Hocker, 2011). Our identities are not so much personal but defined by relational stories. In other words, how we connect with each other in stories is realized by the roles of who we are and the themes of what we want in our narratives.
A useful technique for the mediator to create "better-formed" stories is to ask questions to get a fuller and more substantial understanding of the plot sequence. The task for the mediator in the narrative approach is to facilitate disputants to see the other side of themselves and prepare them to be more tolerant of the other disputant. The exploration of the alternative narrative can facilitate parties to give legitimacy to the other side and create chances for the reciprocal talk (MediatorAcademy, 2016). When people in conflict are aware their imperfectness, they begin to tolerant other's imperfectness (MediatorAcademy, 2016).
During the caucus with Linda, de-legitimacy for Self came before the legitimacy for Other. As can be seen from Example 6, a question ("Okay, so how long has he lived there?") was used by the mediator to seek good stories from the past as an indicator of Linda's future performance in the mediation. This question was useful for the mediator to understand the landlord-tenant relationship between Dennis and Linda in the past. Linda's response ("he was a good tenant") changed the "reciprocal de-legitimacy" (MediatorAcademy, 2016) between her and Dennis.
However, the mediator's attempt to destabilize Linda's problematic and dominant story failed. As can be seen from Linda's following description ("I'm trying to call him, and I'm just sick of it", "I want to get that money", "I am a business not a charity"), she reverted to the problematic stories to emphasize her legitimacy for Self. Although there is another turning point ("I would consider that we were friends at one point"), the extended new story is not stable as Linda cursed Dennis ("Because he deserves it"). Thus, we see that putting the legitimacy for Other before the de-legitimacy for Self produces a context for conflict talk which does not lead to fruition in resolution.
The real turning point in Linda's attitude during the caucuses came with the mediator's question ("Okay, and how much do you think would be fair?"). The mediator presented Linda with a probing question on her idea of a fair amount to be paid by Dennis. Linda's sigh followed by a drawn-out response of "well…" seems evident of her relenting and indicates that she realized her stance towards Dennis may be too aggressive. The probing question by the mediator to rethink about fairness is critical for the mediator to help Linda reflect on herself and awaken Linda's sympathy towards Dennis. This question enables Linda to realize her imperfectness and reduces her high legitimacy for Self. The ongoing talk after this question between mediator and Linda turned to a new proposal which seems fairer to Dennis compared to the prior one.
The private caucuses of Linda show the de-legitimacy for Self is a prerequisite for the reframing of each side's story, to make settlements come out of these stories. This shifted Self is a representation of the shifted self-awareness about one's identity and position in the conflict talk. It lays groundwork for the further establishment of intersubjectivity and the final settlement of the conflict in mediation. The realization of the imperfect Self is called the "ironic" move by Cobb (2006).
The "ironic" move (low legitimacy for Self) enables the construction of Self as the less perfect and is essential to fashion durable agreements out of mediations. Linda's response, in the joint negotiation, to the mediator's question ("what do you think is fair considering that he doesn't have a job?") demonstrates this "ironic" move. Until the last stage (joint negotiation), Linda retained her shifted attitude towards Dennis, gave more legitimacy to Dennis and considered him as someone contrary to the one in the initial problematic story.
Without this "ironic" move, a lasting success cannot ultimately be reached, as can be seen from Linda's attitude which swung between high legitimacy for Self and low legitimacy for Other. Therefore, the question that the mediator should ask first in the caucuses is the one that can facilitate the disputants to reflect on themselves. In this paper, we follow Cobb's (2006) theoretical framework and call this kind of question as an "ironic" question. The reversed order of first connotating legitimacy for Other, followed Example 10 and Example 11 are the excerpts from the mediator's caucuses with Dennis. Example 12 is the excerpt from the joint negotiation between Dennis and Linda. It is observed from Example 10, the hypothetical question ("If you were in her shoes, would that be an acceptable agreement to have a tenant that doesn't pay you any rent?") was applied by the mediator to set up a scenario for Dennis to put himself in Linda's shoes to reflect on himself as a tenant who delays rent. This question helped Dennis give himself more de-legitimacy, which is essential for Dennis to come up with his new solution (selling his mom's jewelry) to the conflict. This led to the problematic story taking on a productive direction as Dennis was willing to take a step back.
As can be seen from Example 9 ("Well...I guess...") and Example 11 ("I don't like selling it at all"), it should be noted that selling his passed mother's jewelry was not an easy decision for Dennis. However, the sentimental value of the jewelry and Dennis' struggle of selling her mother's jewelry seems ignored by the mediator and Linda. No attempt to express sympathy or reciprocality for the sentimental value of the jewelry by the mediator or Linda is demonstrated in Example 11 (Mediator: "well... but she said you hadn't paid January rent") and Example 12 (Linda: "why only $1750?"). The effort that Dennis made in paying his rent was taken for granted by mediator and Linda. Dennis, as the oppressed one in the problematic story, remains as the marginalized one who had the desire to be liberated. Thus, the legitimacy that Linda gave to Dennis is moderate.
Although Dennis' low legitimacy for Self was realized in the caucuses, no Turning Point 2 (Dennis' legitimacy for Linda) could be found in this case. The absence of Turning Point 2 here might be due to the mediator and Linda's lukewarm attitude towards Dennis' effort to construct himself as a good character in the newly created story.
Compared with the talk between Linda and the mediator, we found no questions were posed by the mediator to help Dennis remember his good relationship with Linda in the past. In this new story, there is still an untold story that Dennis held on to about the selling of his late mother's jewelry. The lack of Dennis' legitimacy for Linda may cause unexpected consequences when and if Dennis does not follow through on the agreement at the end of the mediation afterwards.

Conclusion and Limitation
The ability of a mediator to build durable agreements remains an enduring measure of success in mediation. As disputants become willing to accept de-legitimacy for Self and legitimacy for Other, they move from reflexively keeping their problematic stories to reflectively constructing a narrative along good storylines. For the mediation to produce durable agreements, the mediator must first be seen as a neutral for both disputing sides to perceive a sense of even-handed treatment. It is notable that perhaps only through this perception of even-handed treatment can each side be willing to look reflectively at their own imperfectness to establish de-legitimacy for Self, followed by legitimacy for the Other.
Although the mediator is a story recipient of the disputants' narratives of the "facts" in their heads, it does not mean the mediator is a passive listener in the storytelling process. "The mediator plays an important role in asking the right questions, in raising central issues, and in overall participation" (Gerami, 2009).
So, what is the appropriate approach to mediation that allows for both parties in a conflict talk to take a step back to listen to others, to first connotate a de-legitimacy for Self and subsequently, a legitimacy for Other? To achieve a positive outcome in mediation, the mediator must adopt an appropriate approach that allows for the productive narratives to be co-constructed. A review of the literature available shows that very diverse approaches exist across the globe in the practice of mediation as noted by Stipanowich (2017) and the situation may have been similar at the time of Riskin when he decided to create the Riskin's grid. In the narrative approach, mediators use questions to guide disputants to a meaningful conclusion of the session.
Through narrative mediation, the "ironic" questions asked by the mediator as story recipient play a role in helping the mediator balance the power between parties so that neutrality can be maintained to some extent. The reciprocal narrative emerges as the mediators ask appropriate questions, encouraging parties to discern their underlying concerns and changing their form of talk from "disputational talk" to "exploratory talk" (Mercer, 2000). The outcomes of mediation are shaped through the responses of disputants. The mediator helps disputants develop their responses in turn by turn unfolding of stories by eliciting new narratives along good storylines. The "better-formed" stories in intersubjective space can be co-constructed when there is co-existence of de-legitimacy for Self and legitimacy for Other. The legitimacy for Other can only take root when the ground is prepared with the de-legitimacy for Self. In other words, the disputants must be willing to take the first step in uncovering imperfections in their own perspectives. Before disputants can propose a positive narrative, mediators should afford each party a sense of being treated even-handedly regardless of their "bad images" in the problematic stories. Hence, in narrative mediation, each stage of the mediation process is anchored in a focus on "better-formed" stories by developing constructive responses with less regard for the truth.
Each disputant in mediation has two characters in the conflict narratives. One of these characters is the legitimized role constructed by himself or herself, and the other role is the de-legitimized character built by the other disputant. Conflict comes from the contradiction of the two polarized characters existing for a person. The problematic stories are often characterized by temporality and the lack of complexity, which result in the limited options that disputants could choose. People in conflict are unable to listen to others and attune themselves to others because of the social construction of legitimacy for Self and de-legitimacy for Other. Mediation is a way of pulling people out of the problematic stories (MediatorAcademy, 2016) and helping them to see the world differently.
Narrative mediation facilitates disputants to take a step back to look at the frame created by their stories and come up with new solutions to the conflict (MediatorAcademy, 2016). The new solutions reflect the new value systems that are created by both sides with shifted self-awareness (MediatorAcademy, 2016).
Even though different mediators have their way of interacting with the disputing parties, the mediator is a character in the story who asks questions actively, and who is engaged in the evolution of the narrative (MediatorAcademy, 2016). The narrative approach to mediation is beneficial for the creation of solutions instead of the static one given by evaluative mediator. The mediated agreements represent the work of disputing and mediators, and the solutions are more likely to be integrative, creative and sustainable.
The result of the study shows there are two types of questions the mediator could ask. The first type of question is the one that can help disputants reflect on their imperfect side. The second type is the one that can awaken the disputants' memories of the old stories about their good relationship. The result of this case study also indicates that the order of first seeking legitimacy for Other and subsequently de-legitimacy for Self did lead to the failure of destabilizing the problematic story. These findings enrich Sara Cobb's (2006) research of turning points in narrative mediation in that a mediation may be ruined if there is no correct order in connotating de-legitimacy for Self and legitimacy for Other. In addition, the result of this study indicates that the lack of the legitimacy given to Other, led to the absence of legitimacy given by the disputant at the marginalized side with a lower power position. The exchange of de-legitimacy connotations between the disputants may cause severe problems for the achievement of an agreement that continues to be honored after the mediation. The study of mediation from the conversational perspective can provide theoretical support for the training and selection of a mediator. Specifically, the findings from this case study serves to inform practitioners what oracy skills are essential in guiding disputants' responses to shape meaningful outcomes at the end of the mediation. This article clarifies some of the questioning skills mediators can apply when adopting the narrative approach and enriches the knowledge base of ESP teachers to help learners improve their professional competence to be a mediator in settings where English is the medium of communication. However, the findings of this case study cannot be generalized to other cases, and more research is needed to investigate the function of question-asking in the generation of the first two turning points mentioned in this study. More cases are needed to support the study of how mediators' question-asking triggers the production of the other three turning points in Sara Cobb's (2006) framework. Lastly, the function of other discursive mechanisms and strategies can also be examined to make the picture of narrative mediation more complete.