Kandel

California law requires that all unsettled child custody cases must be referred to court sponsored mediation prior to any judicial hearing.^{^[1]} Parents meet in confidential session with a mediator^{^[2]} to develop a parenting plan allocating [page 76] custody and visitation time. If an agreement is reached^{^[3]} it is signed by a judge, becoming a legally binding order.^{^[4]}

The purposes of the mediation proceeding are as follows: (a) To reduce acrimony that may exist between the parties. (b) To develop an agreement assuring the child close and continuing contact with both parents that is in the best interest of the child. (c) To effect a settlement of the issue of visitation rights of all parties that is in the best interests of the child. (Cal. Fam. Code Section 3161)

The court sponsored mediation program and the enabling statutory scheme were introduced in the 1970s,^{^[5]} together with substantive rules compelling gender neutrality in custody decision making and authorizing joint physical custody (with both parents actively participating in day-to-day child care).^{^[6]} California's pioneering scheme has served as the model for similar mediation programs throughout the United States where more than one third of the jurisdictions now have mandatory or discretionary court sponsored child custody mediation. In [page 77] populous, multi-ethnic Los Angeles County alone, mediators hear approximately 11,000 cases per year,^{^[7]} and resolve roughly half of them.

Inspired by the somewhat idealized views about dispute resolution in traditional societies which were current with legal reformers in the 1970s,^{^[8]} mediation is conceived as a flexible process through which parents draw upon their continuing multiplex relationships with each other and their mutual love for their child to rework their psychological and social understanding, address areas of cooperation, and reach an agreement regarding custody that copes efficiently with the plethora of personal scheduling details that shared parenting involves.

Over the past twenty years child custody mediation has generated reams of scholarly discussion. This is roughly divisible into two camps. Proponents of mediation view the process as empowering and transformative (Milner et al. 1995; Bush 1996). In their view mediating parents tell their own stories and forge their own solutions. Ideally, the working out of a custody agreement under the facilitation of a trained mediator also serves as an experiential education in the techniques for cooperative co-parenting. (See Bush 1989; Fuller 1971; Folberg and Taylor 1984; Keltner 1984; Menkel-Meadow 1991; Moore 1986; Saposnek 1983.)^{^[9]} Opponents of mediation view it as a Foucauldian nightmare in which the [page 78] powerful discourses of law and psychology are produced and reproduced in disguised but coercive form. Without attorneys or the procedural safeguards of formal law, these critics maintain, the weaker parent is at increased risk of being manipulated into concessions on custody by the more powerful parent and the even more powerful mediator (Abel 1982: Delgado et al. 1985; Fineman 1988; Grillo 1991; Nader 1984, 1994, 1995; Merry 1982).

However, both proponents and opponents focus on the processual aspects of mediation, albeit emphasizing that certain vague social values and preferred outcomes are sometimes selectively facilitated (Cobb 1992; Dingwall, Greatbatch and Ruggerone 1995; Greatbach and Dingwall 1989).^{^[10]} I suggest we can achieve a [page 79] fuller and more nuanced understanding of mediation by focusing on the substance of mediation discourse, and the emergence in mediation of substantive legal principles. Rather than stigmatizing these substantive principles as 'hidden agendas' we should render explicit and subject to analytical scrutiny the way norms are crystallized into legal principles in the context of multi-directional persuasion.

This article uses representative segments of three child custody mediation sessions, each facilitated by a different mediator, selected from the texts of 40 sessions audiotaped during my fieldwork in the Los Angeles County Superior Court, Conciliation Services Division, between 1991 and 1993. It traces the transformation of a concept originating in developmental psychology - the concept of what is 'developmentally appropriate' for a child - into a principle of the customary law of mediation. Guidelines derived from developmental psychology about what is good for children of different ages are used to gloss and flesh out the vague statutory standard of 'best interests'. The guidelines become objective, highly persuasive, norms. In the multi-directional persuasive context of mediation, the norms operate much like legal presumptions, requiring a parent or parents who want a different custodial arrangement to be very strong and convincing speakers indeed.

'Developmental appropriateness' is a useful, salient concept in mediation discourse because it serves as a coping strategy for dealing with four problems. The first of these is that the legal standard for determining custody, 'the best interests of the child', is supremely, ubiquitously important but hopelessly and admittedly vague. Second, child custody mediators have academic degrees in the mental health professions and professional training in dispute resolution; thus their unique expertise lies in the hybrid space between law and psychology. Third, judicial calendars, time pressures, court bureaucracy and enormous case loads^{^[11]} constrain and intensify goal directedness in the open-ended interpersonal space of mediation.

Fourth, mediators are obliged to negotiate the inherent role conflicts of their job,^{^[12]} which flow directly from the statutory scheme. They must defend the child's best interests, while helping parents to reach a cooperative understanding, empowering them, and formalizing agreements in a time-efficient way. The law directs mediators to encourage parents to initiate their own plans (Cal. Rules of Court Appellate Division I Section 26), to conduct negotiations "in such a way as to equalize power relationships between the parties (Cal. Fam. Code Section 3162(b)), and to ensure the parents' close and continuing contact with the child. But at times mediators must act as advocates for the absent child,^{^[13]} juxtaposing themselves to the parents' demands and making practical suggestions.^{^[14]}

The touchstone of such suggestions is the child's 'best interests', an overarching standard which pervades the statutory scheme.^{^[15]} The statutes provide minimal guidance as to its meaning.^{^[16]} The mediators flesh out the bare bones of the standard in accordance with a praxis born of their situated professional expertise. They have their own way of 'doing child custody' - which both partakes of and differs from the formal written law of statute and precedent and references their distinctive blend of psychological and legal understanding. (This nature of mediation as a hybrid of law and psychology has been remarked on repeatedly by scholars and practitioners alike, e.g. Harrington 1985; Hoefnagles 1985; Merry 1990a, 1990b, 1992; Saposnek 1992; Steinberg 1985; Wallerstein 1986‑1987; Weaver 1986.) Even in statutes and case law, the 'best interests' standard carries a heavy freight of psychological connotations which have increased steadily in the past half century. As between fit parents, the term is usually interpreted to mean the child's total well-being and is informed with knowledge and theory about the child's developmental needs and the dangers to children of interparental conflict. Lawyers and judges, as well as mediators, take this approach, but judges hear [page 82] only the high conflict cases, and so do not work with the detailed time allocation and scheduling issues as closely as the mediators do. It is the mediators who have made notable use of the concept of 'developmental appropriateness' in the sense of an age-appropriate parenting plan.

California child custody mediators comprise a distinct group of dispute resolution professionals.^{^[17]} By statute mediators must possess a master's degree or the equivalent in a behavioral science.^{^[18]} The vast majority are licensed social workers [page 83] (MSWs), or Marriage, Family, and Child Counsellors (MFCCs - holders of a graduate degree which is unique to California). Some hold Ph.D.s in psychology, a few are also lawyers. Academic training in the mental health professions distinguishes the child custody mediators from judges or attorneys, and professionally predisposes them to a sensitivity and concern with interpersonal dynamics and individual emotions. Yet mediators view their job as dispute resolution, not as therapy. They use their mental health understanding to steer parents towards agreement despite their mutual anger and hurt and/or to suggest custody arrangements which are regarded as developmentally appropriate for the child.

Among these mediators, 'developmental appropriateness' is a highly developed concept often used as a 'term of art' by the mediators. Although precisely what is 'developmentally appropriate' is subject to change with changing scientific understanding, during the period of research the mediators had definite ideas about 'developmentally appropriate' parenting plans. For example: very young children should have frequent short visits with either parent. Overnight visitation with non-primary custodial parents should not begin before the age of two; pre-schoolers could appropriately spend weekends and long school holidays with the non-primary custodial parent; and children in pre-adolescence and teenagers should have voice and some choice in deciding when to spend time with each parent and formulating the custody plan.

The mediators themselves do not think of the concept of 'developmental appropriateness' as a legal principle. Rather they think of the 'plan-for-age' ideas described above as psychologically sound guidelines which may be introduced as suggestions when the mediator intervenes because the parents are arguing, having difficulty devising a plan, or proposing something unworkable. Nonetheless, an analysis of mediation texts reveals that it is the use of the guidelines in the context of mediation with its persuasive process, role ambiguities, and shifting alliances that transforms the psychological concept.

Mediators lack authority to render decisions or impose direct sanctions. They do not balance the competing purposes of mediation in the rationalistic way that judges would, purport to weigh evidence and make findings of fact, or strive to attain the child's 'best' interests in the sense of what they consider 'ideal'. Rather, through topic-focused conversation with the parents, they try to shape an agreement that is within the range of what I call the 'tolerable possible' - tolerable because it is consistent with the child's interests and possible because the parents will agree to it. They are consequently intuitively alert to both the parties' [page 84] positions and their range of flexibility, attributes that, in turn, may be functions of the parents' rhetorical skills and prior relationship, fashioned and constrained by the power-laden emotions of the marriage and its aftermath. Further, mediators, who often appear to the parents as knowledgeable and neutral, use their trained powers of persuasion to promote their vision of what is 'developmentally appropriate'. They use such skills of conversational argument as elaborating some statements while ignoring others; using parents' narratives to make unintended points; and, by sharing the floor with parents, co-constructing an understanding of what is needed. Thus good mediators exercise substantial skill in knowing just when and how to intervene (e.g. Donohue 1991; Jones 1988, 1989). It is the nature of the mediators' specific interventions - when and how they introduce 'developmentally appropriate ideas'; how they interpret disputants' narratives and issues in light of the 'developmentally appropriate'; and how they persuade or fail to persuade the parents to agree to them - that engenders the jurisprudence of the 'developmentally appropriate'.

The customary law of mediation has no formal precedential value and leaves no 'paper trail'. Taking place in the courthouse in the law's 'shadow' it is a conversational hiatus in a flood of pre-trial litigation print. As already noted, mediation sessions are confidential. No transcripts are made. Neither mediators nor parents can testify at trial to what has been said at the session. Further, in approximately half of all California jurisdictions, including Los Angeles (the site of this research), mediators are prohibited from making any recommendation to the judge if the parties do not reach agreement.^{^[19]}

Finally, although an agreement reached during mediation is signed by a judge and becomes a court order the agreement is not like a judicial opinion: neither facts, nor reasons, nor rules of law are stated in the agreement. Rather, all parenting plans have a fundamental similarity, which disguises a multitude of real diversity. [page 85] They state simply which parent will have the child on what days and times.^{^[20]} Disguised behind the superficial similarity of the parenting plans, it is possible for a rich jurisprudence of child custody to flourish.

Craig and Alicia,^{^[21]} a young couple, are the divorcing parents of a one and a half year old boy, Ben. The dialogue turns on the idea that overnights are developmentally inappropriate for children under two years old. Craig and Alicia come to the mediation session having agreed between themselves that Alicia will have primary physical custody and Craig will have overnight or weekend visitation with Ben whenever Craig requests it.

Alicia wants Craig to have 'monitored' visitation (visitation only in the presence of a third party), an idea which Craig adamantly resists. Normally visitation is monitored only when a parent poses a danger of violence or abuse to the child. But in this case Alicia seeks monitored visitation because she claims Craig is 'careless' with Ben. Alicia explains that Craig does not properly discipline Ben while she, herself, spanks Ben when he needs it. Alicia's explanation is problematic because California has a stringent standard of child abuse which renders virtually all physical discipline subject to possible criminal sanction.^{^[22]} The [page 86] mediator's view of what is developmentally appropriate, and her standard of what constitutes child abuse, which also diverges from the formal statutory law, influence Craig and Alicia to agree to a custody arrangement significantly different from the one they originally ask for. The mediator begins by insistently explaining that overnights are not developmentally appropriate for children under two.

Mediator: ... generally kids of this age don't go overnight. Usually just developmentally, there's a separation problem with ... from what we call the primary caretaking parent, which it sounds like pretty much has been - not that you've not been involved - but it sounds like maybe she's been the primary one...

Mediator: Well, usually... usually we don't suggest overnights until the child is maybe two or two and a half or so. It's sometimes hard for them to be away... but generally ... generally, a year and a half is a little young to be starting overnights.

Craig: Well, she started leaving the baby over at her mother's house two weeks prior to the separation.... I would also like to have him on Friday nights. To where, he would stay with me the whole day and the whole night.

Mediator: Yeah, well, as I said, generally we don't suggest overnights with a kid this young. That doesn't mean you can't build toward that as he gets closer to two... two and a half...

Mediator: Um, yeah, but it's a little young and sometimes there's a separation problem. You know, the kids become a little clingy-er and sometimes start having sleep problems or something like that. And, again, I'm not saying this has to happen or that he'll be irreparably scarred for life, but's [page 87] generally ... it's a little bit young. Usually, you know, we kind of feel that maybe if they wait a little while ... maybe until they're at least two or two and a half.

The mediator repeats four times that overnights are inappropriate for children under two, supporting her argument with a discussion about separation anxiety. It is evident that the mediator is not talking about Ben's development but about some abstract, objective concept of a developmentally appropriate toddler. She uses the word 'generally' five times and ignores Craig's statements that Ben is nearly two, and has already been spending overnights at his maternal grandmother's house. It seems that, for the mediator, the developmentally appropriate plan is presumptively correct. Craig will have to be very sure of his position to persist against the mediator's repeated insistence that overnights are 'usually' not recommended, and that small children develop 'sleep problems' and become 'clingy-er'. Even if Craig knows that Ben doesn't suffer emotional effects after sleeping at his grandma's, Craig would have to be a very convincing speaker to get the mediator to stop insisting. He fails to do this and when the mediator finds Ben impervious to her speech about developmental psychology, she endeavors to induce him to accept the 'developmentally appropriate' plan by repeatedly proposing that he have one weekend day and one weekday with Ben.

Discussion then shifts to the question of physical discipline as the mediator probes the reason for Alicia's request for monitored visitation. Holding views radically different from those expressed in the California statute, Craig and Alicia both regard spanking children as essential for proper discipline.

Alicia: He shows the child no discipline... and I'm like 'No, he has to be disciplined'. And it's like ... I admit, I do spank my child.

Alicia: You know, and um ... I'll spank him, or I'll hit him on his hand and I'll tell him, 'No, you do not do that!' But I'll tell him and I'll tell him and I'll go, 'Ben, do not do that', and it's like, after I get tired of saying his name I either go ahead and either spank him or slap him on his hand. Or on his arm. And I'll tell him 'Didn't I tell you not to do that?' I go, 'You're not deaf, and you do understand me'. You know, and he'll look at me and like sometimes, the child has, like, my personality whereas it's like ... he'll be like, 'You hit me'. You know? And everything, and you know sometimes he'll keep...

Alicia's disciplinary methods, although neither unusual nor dangerous, skirt the borderline of technical illegality. The mediator could respond to this in several ways. She could be an alarmist, remembering her legal obligation to report child abuse.^{^[23]} But neither parent is alleging abuse, and the experienced mediator has heard many parents' narratives and is, herself a parent who knows both the motives for and values of an occasional spanking. Her response is empathetic.

Mediator: I have four kids who are grown, and I think that almost every parent has probably given their kid a swat on the bottom at one time or another ....

An alternative response might focus on the fact that Alicia has portrayed herself as a young mother who is easily frustrated and angered by her caretaking responsibilities. The mediator might have found this to be a justification (or at least a rationale) for deviating from the 'developmentally appropriate' norm, [page 89] giving Craig the overnight visitation he seeks, and suggesting to Alicia that more free time might increase her patience and improve her parenting skills. This might seem a particularly good idea, in the light of a story Craig has told about how Alicia once threw Ben onto the bed when he woke her up crying in the middle of the night:

Craig: The baby was about 10 weeks old and he was crying at night. Alicia couldn't sleep 'cause she had to work the next day. And I was in the kitchen getting the baby's bottle ready, and when I came into the room, she started saying 'Goddamn it Ben!' and shaking my son. 'Will you go to sleep, Goddamn it? I gotta get up and go to work in the morning!' And she threw him to the bed ok? Well, I had a king size bed and it was low to the ground and she was standing up, so this was a nice two foot drop, and from then after he hit that, he started slipping off the bed.

Instead, the mediator chooses the approach of supporting and accepting Alicia's parenting while simultaneously reconstructing Alicia's narrative to fall within what the mediator regards as acceptable physical discipline. Using the linguistic strategy of a 'collaborative floor', the mediator and Alicia each take turns adding a phrase to a common story, putting a gentler slant on Alicia's methods. (Such co-construction of meaning between mediators and parents occurs frequently in mediation: Cobb 1993, 1994; Cobb and Rifkin 1991; Folger and Bush 1994; Greatbatch and Dingwall 1994; Kandel 1994.) Simultaneously the mediator's semantic suggestions educate Alicia as to what constitutes legal compliance and non-compliance.

Mediator: But you know, it is against the law to strike a child. You need to be very careful about that. [The mediator here both warns her what the legal standard is and suggests that there are technicalities which can he violated.]

Alicia: No, I know. I know. I've never struck him to where I've, you know, intentionally meant to hurt him. You know, I'll hit him on the bottom of his, you know, he has a diaper on, you know, it's like...

Mediator: Yeah, I know what you're saying. I'm just mentioning it because if there ever should be, for example, bruises or anything like that, ... and of course kids that age fall.

Alicia: Yeah, because, I mean I'm sorry to say this, but he's come out clumsy like I am. I mean, I'm serious. He could be walking across the kitchen floor and boom! he'll fall and I'll be, 'Be more careful'.

This collaborative dialogue has multi-faceted ramifications. Alicia appropriates the mediator's buzz words and tell tale signs of child abuse. Readily incorporating them into her description of Craig's childcare methods, she talks graphically about bruises and bumps, the better to argue her case for monitored visitation.

Alicia: I don't know because for a time he was taking care of the baby and it's like, I was like, 'Well, how did he get this and how did he get that?' and he said, 'Oh, he fell. He fell, Oh, he bumped himself'... And it's like, he had bruises on him and I ask, and it's like 'Oh, he fell, Oh, he scraped himself. Oh, he cut himself.' Or, 'oh, he was playing with a fork and he stabbed himself or something.' And I'm like; 'Weren't you watching him?' He said, 'Well, I was watching TV you know, and I got into the program'.

Alicia's compelling narrative is filled with rich detail and reported speech. The mediator hears that Craig is a careless and short tempered father. But despite the descriptions of bruises and bangs, the narrative lacks any allegations of the 'serious physical abuse'^{^[24]} or deliberate physical injury that would persuade the mediator to suggest monitored visitation. Instead of accepting it as helping Alicia's cause, the mediator enlists the narrative as further justification for her view that overnights with Craig would be inappropriate for Ben until he is two years old. The mediator appropriates Alicia's anecdotes, ignoring the 'bruises and bumps' language that the mediator had initially introduced into the conversation. The mediator suggests:

Mediator: Well, if he's that negligent or short tempered, I guess my thought would be instead of two days in a row, maybe do something like say one week Saturday and one week Sunday. I mean, it's the same amount of time but it's spread out because two days in a row, if he's going to get a little careless, he's probably more likely to do it the second day.

The mediator has reframed the tale, using it for a purpose unintended by the speaker. The mediator's interpretation of it as showing that Craig's carelessness warrants a reduction in his parenting time contrasts strikingly with her failure to conclude that Alicia's explosive frustrations warrant a reduction in her parenting time. In short, the mediator interprets the narratives about parenting technique in accordance with her dominant idea of what is developmentally appropriate.

Even this one excerpt illustrates how the customary principles and processes of mediation diverge from both the formal written law and the practices of mediating parents, and how 'developmental appropriateness' emerges as an abstract, objective norm which mediators regard as presumptively correct. To steer the parents towards the developmentally appropriate parenting plan, the mediator has employed didactic explanations, ignored some of the parents' narratives, engaged in the collaborative construction of an acceptable story with Alicia, reframed Alicia's narratives as support for the mediator's position, and repeatedly presented and justified a specific alternative proposal. Where the 'developmentally appropriate' presumption is at work a parent has to be a strong speaker to steer the consensus towards the conclusion that it is inapplicable or unacceptable. Towards the close of the session the mediator guides the parents towards an agreement by proposing several similar, very specific plans, none of which give Craig overnight visitation with Ben, and allowing Craig and Alicia to thrash out verbally their differences over the specific details.

The dialogue in this case turns on the idea that extended summer visitation with the non-primary custodial parent is appropriate for a child of four and a half. The parents, Jane and Steven, recently divorced after two years of separation, have stipulated in open court to a custody arrangement in which Steven will have an alternating division of summer vacation time in two week long blocks which is to start when their daughter Clara is five and a half (two summers after the mediation). Steven has petitioned for a change. He wants the alternating two week blocks to commence the next summer, when Clara is four and a half. This is an area of very 'soft' law, in which a judicial opinion is unpredictable. A court might grant Steven the increased visitation or admonish him to abide by the agreement to [page 92] which he has but recently stipulated. As in Case One, the mediator repeatedly advocates the position which he considers developmentally appropriate: that the extended visitation should start the next summer when Clara is four and a half.

At their request, the parents in this session never meet face to face. Instead, they have alternating private sessions with the mediator, who negotiates the agreement through a kind of 'shuttle diplomacy'.

Mediator (to Steven): I think it's appropriate for the child to spend time with you more than two weeks.

Mediator (to Jane): [Steven's request] is reasonable ... She's going to be four and a half years this summer; instead of 1994, it is appropriate to ask for the summer to start this summer.... I think it is appropriate for the child and if you get this perspective and you agree then fine...

As in Case One, the mediator uses 'developmental appropriateness' as an abstract, objective principle, rather than as a standard by which to judge what is suitable for the particular child, Clara. The parents, who are both elementary school teachers, share the mediator's respect for age-appropriate parenting. The parents, however, each offer specific evidence about their child, Clara, to support their positions about what is developmentally appropriate for her. Jane explains:

Jane: ... the reason I was sticking to five years old is that, according to my school district, five years old is when you are able to relate an autobiographical incident with clarity and she can tell me if something is disturbing her ...

Steven is willing, arguendo, to accept Jane's interpretation of developmental appropriateness (ability to recount an autobiographical incident), but presents evidence that Clara can already do this because she is ahead of her age group.

Steven: I know for a fact that she is well ahead of her age because we had her given a PPBTD Test (Peabody Picture Book Test) and she was seven months ahead of her age group.

Steven and Jane's concept of 'developmental appropriateness' centers on a child's ability to accurately report narratives. By contrast, 'developmental appropriateness', as mediators usually deploy it, is grounded primarily in children's abilities to adjust to transitions between houses and to different adult caretakers with possibly different parenting styles. Before opening up fuller conversation on Clara's cognitive and verbal capacities, the mediator would have [page 93] to believe that the child's ability to report accurately was somehow linked to her best interests in a way which might justify giving the principle of 'developmental appropriateness' in this case the special gloss, 'able to recount an autobiographical incident'. Thus 'developmental appropriateness', understood as turning upon 'plan-for-age' functions, as it was in Case One, raises a presumption against which Jane must argue to make her case.

To meet her rhetorical burden Jane must explain why Clara's welfare depends upon her ability to relate to Jane what happens when Clara is at Steven's house. This she is unable to do. Jane is a relatively weak speaker, whose stories end in anti-climax and whose sentences trail off unfinished.^{^[25]} Her flashiest tale is that Clara was once accidentally bruised by a falling bass drum when Steven took her to a concert. She says that Clara once came home 'glazed over' and that "Right now she'll come back from a visit and something disturbs her, it takes me days to drag it out of her". But she does not convince the mediator that there is anything suspicious, careless, or amiss about Steven's parenting that would justify Jane's dragging stories out of Clara as a way to keep her eye on things.

Rather, the mediator concludes that the issue of autobiographical reporting would be relevant to an assessment of Jane's needs, not of Clara's. Jane wants Clara to be an information conduit because the residuum of the control Steven exercised during the marriage has left Jane able to communicate with Steven only by 'putting a note in the bag' when Clara goes to visit.

Jane: While we were married he had a lot of mental control over me ... Even now when I see him and I'm dropping my daughter off and I visit, just his look and his tone of voice is enough to send me wanting to dive under the dashboard of the car. So it makes it really hard for me to communicate with him.

Sadly for Jane, whose rhetorical inadequacy is likely in part the result of her subordinated position during the marriage, the mediator persuades Jane to separate her 'issue' from her daughter's interests. The mediator's persuasive tactics reveal [page 94] how 'developmental appropriateness' assumes its jurisprudential identity through contextualization within the broader mediation paradigm which includes teaching parents to separate their emotional entanglements from their children's needs. For example, the mediator tells Jane cautionary tales based on previous cases he has mediated.

Mediator: [The mother] comes to pick up her five and a half year old daughter at [the father's house]. [The father] has dinner with his own family and children at 7:00. So he asks her, to know whether to wait or not, 'Are you going to feed her dinner?' But people aren't like that, that is not what she hears. People have very twisted prisms. She hears, 'Are you going to control me again?' That's what happens unfortunately people who come through these doors always behave in a manner which are most oblivious to the other side. I have seen maybe two or three cases where the people are able to place themselves in the shoes of the other and just address the issue of the children. That is rare unfortunately. Invariably, they use the child - sometimes even maliciously.

Through this tale the mediator advises Jane that her emotions are inappropriate for the situation. This is normal but unfortunate. The tale sets out the task Jane faces: do not misperceive Steven's present concern with the child's interest as an attempt to manipulate you just because Steven has manipulated you in the past. It presents the goal: separating parental emotions from children's needs. In it, Steven is clearly on the 'right' side. Although Jane expresses some legitimate doubts about extended visits with Steven (the child comes home with bruises, or 'glazed over', she doesn't talk about her experiences, declares she won't go again, or 'pitches a fit' before visits), no doubts are legitimated by the tale in which the father only wants to know whether to feed his child dinner.

The second tale builds upon the first, combining the challenge to separate parental emotions from children's interests, with an illustration of how absurd mothers seem who do not abide by what is developmentally appropriate.

Mediator: Earlier today there was a mother here who honestly thought that morally it was not appropriate for her 16 year old to visit dad because he was living with a 'bimbo'. So this is a 16 year old and she wanted the father to send the 'bimbo' away for the weekend the child was there. But I talked to this teenager and he wanted to be with dad and he liked his girlfriend but he didn't want to offend his mother. Sometimes people use those moral issues seemingly in a congruent manner but, I think, [page 95] inappropriately. If you feel it is inappropriate for the child [not to have extended summer vacations in 1993] then you should say that.

Through the tale, the mediator lets Jane know that if she doesn't permit her four and a half year old daughter to have extended summer vacations with her father she will look as ridiculous as the mother who doesn't want her almost grown-up son to spend time with his father's girlfriend. 'Developmental appropriateness', then, in the sense of the right custody plan for the age group, acquires its jurisprudential contours and shadows through interplay with the assumptions and paradigms of child custody mediation. It is not about being mature enough to read and tell stories. It is about power, and control; about knowing when to hold on (as in Case One) to a child is who is too young to visit, even though a noncustodial parent desperately wants to spend the night with him. And about knowing when to let go (as in Case Two) of a child old enough to have an independent relationship with a noncustodial father, and of the entangled emotional issues which make it hard for the custodial parent to release her.

At the same time, 'developmental appropriateness' is an objective standard deemed to fit the cultural norm; a standard that, in Bernard Jackson's term makes 'narrative sense' and is understood to reflect the typical case (Jackson 1995). It is not a rule for weighing evidence, whether Ben's overnights with his grandmother demonstrate his resilience to the stress of separation or Clara's educational test scores prove her precocity. It is a recommendation about the typical child's 'best interest' by the mediator who has the greater authority and expertise in determining the topics and turns of talk. To get a different plan, a parent must be convinced and convincing enough to persuade the mediator or, at least the co-parent, or, as the next case shows, simply insistent and determined enough to say 'No'.

The 'developmentally appropriate' issue in this case is how far voice and choice in the allocation of custodial time are appropriate for pre-adolescents and teenagers. During the mediation session, Marcia and Jerry, divorced parents of four school age children (the oldest of whom is thirteen) rapidly haggle out some details of the visitation and holiday schedule and then disagree strongly on the vacation plan for the following summer. Marcia, the noncustodial parent, wants the children for half the summer while Jerry wants to continue the present plan of alternating weekend visitation. They also disagree as to what the children want

Marcia: Well, see, they're telling him one thing and last night when they were out shopping with me, "Oh, Mom when you get your house, we're going to do this, and I want to do this..."

Jerry: "The kids do not want to go half the summer with her.... As a matter of fact, the oldest said - told me- that he don't want to go with her every other weekend. OK? The next to oldest, James, said the same thing and the other two, they gonna do what the older two are gonna do. ... They-they follow. Ok. My kids do not want to go ... half the summer at all.... As ... as for a summer schedule - no. Every other weekend.

Jerry, who has custody of all four children, belongs to a fundamentalist Protestant church around which the family's life has revolved; church related activities consume all free time except for Monday evenings. In addition to worship on Sundays and Wednesdays, the children attend the church day school, play on school related teams, sing in the church choir, and have other lessons and recreational outings associated with the church.

Marcia was formerly a teacher in the church school. When she fell in love with another teacher and divorced Jerry, Marcia lost her job and was excommunicated from the church. Ostracized from her children's school and extracurricular activities at what was once also her workplace, Marcia feels even more distanced from her children than the typical noncustodial parent. Her request for a block of summer visitation time is an ordinary one. The mediator even opens up the subject saying:

Mediator: Now what about summers? Summers ... Now, here's what .... again there's a lot of things parents do. Some split the summer in half, in different ways. Some have like June and July with one and August and September with the other. Some, some don't do it that way. They have different combinations of weeks here and there...

In an initial effort to bridge the gulf between Marcia and Jerry, the mediator reframes the problem as an issue of the children's needs, a standard technique used by mediators. His words resonate with the dialogue in Cases One and Two as he emphasizes the importance of understanding what is developmentally appropriate for the children. Because these children are between later middle childhood and the teenage years, the developmentally appropriate approach accords more deference to the children's peer group activities and more respect to the children's wishes. Long summer vacations (and even other substantial time) spent with the non-custodial parent often pose conflicts for older children between [page 97] their ties to the parent and their ties to their peer group in the neighborhood of their primary residence. The mediator understands non-custodial parents craving to spend this time privately with their children. Inevitably, the children's own friendship and community ties disadvantage the noncustodial parent. The mediator's reframing softens the blow by posing the matter as a compromise between the needs of parent and children, rather than between the ex-spouses. But it also pinpoints the issue.

Mediator: Part of being a parent is that I guess your parents want your kids in some good activities - is that right? ... And so you're a family and part of being a parent means both parents, seeing if they can coordinate all that... But when you're now in two houses... you're not interacting with the children, right, if you're not having time with them? When they're not with you, then you're not interacting with them. So that makes the time precious because you don't have as much of it as you had before. Right? ... But now, what that means is that you're going to have less time, so there's more of a choice that Mom may be concerned about saying, I'm in a bind here because if the kids do exactly the same things they did when we were together, then there'd be a problem in terms of what I'd like to be interacting. I want to get the kids home, or have dinner with them, and so they may not be able to ... and so, you see how it's a bit more complicated with these activities because now there's you and the kids having time with them and doing your activities with them, might compete with some of the other activities...

Again, as in Cases One and Two, the mediator verbally paints a more generic picture than the parents. He visualizes, and invites the parents to visualize, their situation as a prototypical conflict between noncustodial parents and almost teenage children. He emphasizes that wanting to be with friends is "what happens when you get to be a teenager" and "that's going to be happening more and more".^{^[26]} To the non-custodial parent, he says:

Mediator: You know, what I hear from you is that summer would be one time when you could have more, uh, you know, consecutive time with the kids. Uh ... and, uh, have a, uh, a more consistent experience with them is when they're not going to school. And so, that's what you're expressing here.

Despite the marked disagreement between Marcia and Jerry, the mediator might well have worked hard at mediating this issue, as many mediators do, given its importance to the family. The mediator might have spent considerable time, proposing various options, letting the parties talk it through, perhaps arriving at a tentative or compromise arrangement. In this case, however, when the parents remain firm after a few minutes, the mediator halts the discussion.

Mediator: Ok, well let me ... Ok, well, now as you were talking there, I was thinking to, uh, putting the summer off and that meeting with them - either waiting until you get your ... You know, either come back here in May, for example, I could talk to the kids too...

In postponing the discussion the mediator relies on the aspect of 'developmental appropriateness' which holds that as children reach pre-adolescence or the teenage years they should have more of a say in where they spend the summer. Mediators often want to talk to children, (especially where the parents hold contrary views of the children's wishes) to get a sense of their views and needs. Yet postponing the discussion from late October (the time of the mediation session) to the following May, rather than scheduling it a few weeks hence (for example, in mid-December when the children are spending their winter vacation at Marcia's house) is very unusual.

By emphasizing the generically 'developmentally appropriate', the mediator frames issues in conformity with objective norms that exclude giving heed to Marcia's special pain. He treats the issue here as the ordinary conflict between the peer group interests of the children and the parent's interest in seeing the children. He proffers his suggestion that the children participate as ordinary developmentally appropriate advice. Although this tension between parental time and peer group activities is a problem for every non-primary custodial parent, Marcia's excommunication from the church which is the focal point of the children's lives is especially poignant. She needs parenting time, she insists, to forestall her total isolation and alienation from her family.

Marcia: I mean when ... like you say ... when we all lived together, it was fine, because I worked in the church. I coached the sports with my kids - football, too. Everything was fine and dandy. But what he's trying to do, is still base their whole life around that church and shut me out. And I'll tell you where the church stands on it. They've excommunicated me from the [page 99] church. What that means is that it's a sin for my kids to see or talk to me and its a sin for him to see or talk to me.

Marcia: So I know it's hard, and it's a big adjustment for them to come and see me. But I think even if they don't want to, they need to at times because the church has such a big impact on them. And he'll tell you it doesn't. But it does. And I think I need to get my kids away and let them see that I'm still mom, and I'm still me. Even though I don't go to the church, you know? My love for them hasn't changed at all.

Marcia's plea for an alternative and particularized interpretation of best interests as a requirement that the children spend time bonding with her in order to overcome the negative effects of her excommunication does not overcome the mediator's commitment to the 'developmentally appropriate' voice and choice approach. As the mediator framed the case, the issue was the children's wishes to be in their activities versus the mother's desire to be with them.

He did not consciously frame this as a clash of lifestyles between the parents - or as a case in which, in order to protect the relationship between mother and children, because of the mother's excommunication from the church, there might be a reason why the children had to be directed or seriously encouraged to spend time with their mother even if they did not want to. Such an approach would also have been consistent with the children's best interests. The mediator could have drawn strength for pressing Marcia's case from the several statutory provisions which hold that "frequent and continuing contact" with both parents is a right and need of the children, and from his own statutory obligation to safeguard "the rights of the child to frequent and continuing contact with both parents".^{^[27]}

If the mediator had reframed the issue in this way, Marcia's and the children's needs would have seemed to be the same, while Jerry would have been the parent who was encouraged to disentangle his own feelings and do what was best for the children (like Craig in Case One and Jane in Case Two). The mediator, however, did not use these explicit provisions to help Marcia carve out an exception to the 'developmentally appropriate' plan. Rather, he used the concept of'developmental appropriateness', which lacks any statutory basis whatever, to discontinue the discussion.

The mediator's strategy here, of avoiding the subject and ending the session, is extreme. Other circumstances could have justified curtailing the length of the session. The parents themselves had expressed a need to be finished quickly, and another couple were waiting rather anxiously in the anteroom. But the mediator could simply have continued, having everyone wait, or have scheduled a second session a few days or a week later. Putting off the topic for a half a year was quite unusual. In effect the mediator simply refuses to address the issue - going off on other topics, or putting off discussion. Mediators will often do that where a particular topic is sensitive or polarized, while agreement is negotiated on other issues. Some parents are much more insistent about repeatedly bringing up a subject and trying to take hold of the conversational flow than Marcia is. But where the mediator simply says, as he did here, that a particular subject will be addressed six or seven months hence, the parent has little choice but to litigate or protest. The summer time is critically important to Marcia; she is ready to litigate.

Using their honed skills of persuasion and their subtle authority to control the micro processes of mediation, such as turn-taking, issue-reframing, and choosing topics of talk, mediators exert their influence to get parents to agree to developmentally appropriate plans. Presenting, and pushing for, what is generically and 'scientifically' developmentally appropriate is a benign, even helpful way to reconcile the mediator's conflicting roles of child advocate and parents' aid. The child's interests, when encoded as the 'developmentally appropriate', are recharacterized from positions adverse to those of the parents into a neutral framework, even a model parenting plan, which the parents can use to bridge their differences and construct their future co-parental relationship in a 'scientific', professionally recommended way.

The 'developmentally appropriate' also enables the mediator to 'represent' the often absent child, and to ensure the reasonableness of the average, ordinary agreement.

At the same time, the principle of the 'developmentally appropriate', laden with its objectified and typified vision of how reasonable children behave and how reasonable parents should behave, has the ironic consequence of standardizing parental plans and stereotyping outcomes in a forum designed to maximally individualize them.

The predictable danger of such stereotyping is that the parents with the most non-mainstream ideas about parenting, and accordingly perhaps the most unusual lifestyles or cultures, are forced to sustain the heaviest burden of rhetorical persuasion. But the dialogues suggest that this is a one-sided view of what happens in fact. The fluid discourse of mediation, in which supportive policy can be expressed as empathy and shared experience (every mother spanks her child) and negative policy as interrupted conversation (let's put this off 'til May) interacts with standardizing concepts like the 'developmentally appropriate' to shield and encourage the unlawful or unusual (like spanking and totalistic religious communities), allowing also for the protection of diversity behind the curtain of confidentiality and standardized written agreements.

That the concept of'developmental appropriateness' transforms and is transformed by mediation is evident in the three cases just described. But if we are to call this transformation 'becoming law' and intend more than semantic cleverness, 'developmental appropriateness', as it functions in mediation, must be shown to have at least some of the characteristics which are associated with laws in other contexts. While I find good reason to shy away from the eternal essentialist debates about what 'law' is,^{^[28]} I assume that most readers will agree that an objective norm backed by authoritative sanctions which shapes formal disputes but can still produce indeterminate results is 'a law'. While I have no interest in fitting the data into a taxonomic scheme, it is against the backdrop of that definition that this section discusses how 'developmental appropriateness' operates in mediation. The cases just discussed suggest five general conclusions.

1. 'Developmental appropriateness', understood as a set of custody and visitation agreements or parenting plans appropriate to children of specific developmental ages, constitutes a 'customary law' of mediators which is distinct from both the customs of disputants and the formal written law. The principle of'developmental appropriateness' is not simply an articulation of shared child raising customs or cultural norms which mediators and disputants have in common. To the contrary, each of the three cases attests to a clash of norms and/or a dispute about what principles should control in the context of shared norms. In the first case, both parents believe that their one and a half year old son is mature enough for [page 102] overnight visitation; it seems, he regularly stays overnight at his maternal grandmother's home. The couple's view of the relationships among the three-generation extended family, which includes Ben's spending significant time at his maternal grandmother's, and the distribution of childcare responsibilities between an ex-wife and an ex-husband, reflect the working class Latino^{^[29]} community in which Craig and Alicia live. They differ noticeably from the mediator's ideas. Yet the mediator invokes multiple means of persuasion to induce the parents to accept a 'developmentally appropriate' plan significantly different from that which they both request when the session begins.

In the second case Jane and the mediator generally agree that the child's age should be a significant factor in determining the parenting plan. But the issue of what is 'developmentally appropriate' is hotly contested, with Jane and the mediator referring to different external standards. Jane cites a norm accepted by the school district where she teaches first grade: at the age of five a child can accurately recount an autobiographical incident. Steve concedes that this standard may be the correct measure, but asserts that it is factually inapplicable to Clara whose verbal precocity, on the basis of educational tests, enables her to talk like a five year old at the age of four. The mediator avoids the question of Clara's verbal development, treating it as Jane's psychological 'baggage' which must be shunted aside, and urges the parties to follow the 'customary law' that overnight visitations are appropriate at the age of four and a half.

The clash of standards between Clara and the mediator is especially poignant because, as middle-class professionals in fields that incorporate developmental psychology, they have been socialized into much of the same theory. They perhaps have read the same books, and they share the same general worldview about the proper raising of children, their developmental course, and the importance of putting a child's needs ahead of one's own hang-ups. Jane actually tells the mediator that she doesn't want to negotiate with Steve, but that she has [page 103] come to mediation because she 'wants to do what is best' for Clara. And, ironically, it is precisely by manipulating this common participation in broad, vague norms, that the mediator is able to persuade Jane, through his exemplary tales, to grant Steven increased summer visitation as the developmentally appropriate thing to do.

Similarly, in the third case, Marcia, a former teacher in a religious elementary school, probably shares many norms with the mediator, who also comes from a religious background, and was once a member of the clergy. They implicitly share norms about the appropriateness of pre-adolescent and adolescent children making their own choices, and the moral value and power of intense involvement with organized religious institutions. Yet again, it is these shared norms which form the basis for the difference between Marcia and the mediator. The mediator relies upon the 'developmentally appropriate' as though no religious issues were involved, thus indirectly shielding the religious influences from examination and attack. But Marcia argues that the religious influences compel an exception to the 'developmentally appropriate' in the interests of maternal love.^{^[30]}

Thus Jane clashed with the mediator about the standard by which 'developmental appropriateness' should be measured. Marcia clashed with her mediator over her argument for a principled exception to the 'developmentally appropriate' children's voice and choice. In each case the difference occurred between a mother and a mediator with similar education and perspectives. The clashes reveal the legalistics of mediation to be more than contests of power in the context of shared norms.

At the same time 'developmentally appropriateness' is no mere application of formal law. That law enunciates only the vague 'best interests' standard, but little clarified by case law. Rather, mediator praxis glosses the standard with rules or guidelines derived from developmental psychology. Further, emphasis on 'developmental appropriateness' may sometimes take precedence over statutory concerns like the prohibition of corporal punishment (Case One) or the need for frequent and continuing contact (Case Three). It constitutes a kind of 'customary law', in the sense of a shared, unwritten, habitual way of doing and deciding things which is not reflected in statutes and written decisions. It constitutes a set of mutually understood techniques, assertions, and assumptions for the expeditious [page 104] production of relatively routine results. It is an amalgam of ingrained perspectives, habitual expertise, and privileging of positioning which approximates Pierre Bourdieu's concepts of 'habitus' and 'practice'. (Bourdieu 1977)

Such differences in praxis, generating a virtual 'customary law' of law-related professionals that differs from both the formal law and the customs of disputants, have been found in other contexts by ethnographers in the United States. (Conley and O'Barr 1988, 1990a, 1990b^{^[31]}; Maynard 1984, 1990^{^[32]}; Sarat and Felstiner 1986, 1988, 1990, 1995^{^[33]}; Yngvesson 1993^{^[34]}). Where such dispute resolvers are full or part time specialists, their 'customary law' will reflect their specialized knowledge, training, and role. It will incorporate privileged concepts and perspectives that differentiate it from the 'customs' of litigants, even those of the same society, culture, and class. Those studies and this suggest that we may profitably approach the praxis of specialist advocates, decision-makers, and dispute resolvers as a kind of 'customary law' which exists in tension with both formal law and the customs of disputants in ways susceptible to jurisprudential analysis.

2. 'Developmental appropriateness' is an abstract legal principle, not a scientific law nor a standard for factual determination. 'Developmental appropriateness', carrying the notion that a particular parenting plan is appropriate for a child of a particular developmental age, is presented by the mediators in the cases above as a [page 105] rule, even stated in propositional form. But what kind of rule is it? A rule of science or a rule of law?

Mediators typically credit 'developmental appropriateness' with objective validity. If not exactly a scientific law, it is a guideline derived from the scientific studies of behavioral and social scientists, reflecting a real world wisdom superordinate to diversities of interests and ethnicities. The mediators often (as all three do in the cases discussed above) present it to parents as an objectively worthwhile goal, a scientifically grounded standard for which parents should strive; that is, as a law of science rather than a law of 'law'. While some parents, like Marcia (Case Three) resist, many respond to mediators' discourse on developmental appropriateness as though they are being educated rather than ruled. The seemingly objective quality of 'developmental appropriateness' makes it palatable, facilitating agreement and compliance in a forum with neither sanctioning nor decision-making authority.

However, 'developmental appropriateness' is really more like a 'customary law' (an unwritten rule specifying which plans are considered 'correct') than a law of science. Although the principle of 'developmental appropriateness' derives from studies which may be considered scientific, it is transformed into law by abstraction and change of context. The norms of 'developmental appropriateness' are based on descriptive statistical studies of what is appropriate for children. These descriptive studies are then converted by abstraction into a set of culturally expected and accepted norms about child development and linked appropriate parenting. They become recommendations or rules of thumb which are then presented as sound advice by mediators with training in child development. Finally they are firmed and further abstracted when they are used as guidelines by those who work in a legal arena for drafting legally binding parent plans. Thus, what is initially descriptive becomes prescriptive, while simultaneously becoming more abstract, formulaic, and standardized. The constitutive power of the law exerts a circular irony. The prescriptive gradually becomes descriptive as 'developmentally appropriate' parenting plans influence, and even enforce, normative standards. They also influence the customs of law-abiding divorcees who conform to the terms of their custody agreements. As more and more developmentally appropriate parenting plans are 'so ordered' into law by judges, the 'developmentally appropriate' way of doing things becomes, increasingly, the statistically normative practice. In a society like the United States, where the divorce rates have stabilized in the fortieth percentile, and lawyers negotiate divorce cases against the probability of a particular litigation result, it is hardly far-fetched to argue that the 'customary law' of mediators will influence the customs of law abiding divorcees. By consequence, a parent has to be an increasingly powerful speaker and make out an increasingly powerful case to argue that the 'developmentally appropriate' standard should not apply for reasons [page 106] of either cultural difference, personal preference, or the distinctiveness of a particular child.

3. 'Developmental appropriateness' operates like a legal presumption. In law, a presumption is an inference or assumption which must be made from certain facts. Most legal presumptions are of the variety that affect the burden of proof.^{^[35]} They are legally 'true' unless the party opposing them can produce proof to the contrary. Such presumptions embody policy decisions. For example, in California law, property acquired during a marriage is presumed to be community property (belonging in undivided halves to the husband and the wife), a rule promoting the policy that all family assets be divided equally at divorce.

In court sponsored child custody mediation, the concept of 'developmental appropriateness' operates similarly. The presumption is that a particular 'developmentally appropriate' plan based on the fact of the child's age is to be assumed or inferred to be good. The underlying policy is that such a plan is probably reasonably good for a majority of children. Although, perhaps, they do so more by praxis than deliberate policy, mediators will typically presume that the 'developmentally appropriate' parenting plan should be the basis for agreement or the starting point for negotiations. Thus the burden of persuasion is on the parent who wants something different. In contradistinction to adjudication, persuasion is multi-directional (each parent tries to convince the other parent and the mediator, and the mediator works to persuade both parents towards a plan that is both adequate and acceptable), and the parents speak on their own behalf. The parent [page 107] who would steer the session towards an agreement that differs from the 'developmentally appropriate' bears a heavy rhetorical burden of storytelling, argument, persuasion, insistence, and alternative suggestions.

While many parents do overcome the presumption, the three cases discussed here illustrate the challenges involved in seeking an alternative construction of the issues of the case. In Case One, Alicia's efforts to use the buzz words of abusiveness (bruises and bangs) to obtain monitored visitation do not succeed. Instead, they convince the mediator that Craig is careless (not abusive) and feed directly into the 'customary law' presumption that overnight visits are too long.

In Case Two, Jane endeavors to convince the mediator that 'developmental appropriateness' should be applied to mean that the child's ability to recount an autobiographical incident should be the threshold for extended summer visitation. Her rhetorical skills are unequal to the task. Although Clara once came home 'glazed over' and sometimes complains about visiting Steven, Jane cannot describe any specific secret problems with his parenting for which Clara acted as conduit. Her worst allegation is that Clara was once accidentally hit by a bass drum while at a rock concert with Steven. Steve makes her look ridiculous by describing how she once mistook spaghetti sauce on Clara's forehead for the bloody signs of abuse. The mediator convinces Steve and Jane to accept overnight summer visitation. In Case Three, Marcia fails to overcome the developmentally appropriate presumption that older children and adolescents should have voice and choice regarding the parenting plan. Her plea that "I think even if they don't want to [see me], they need to at times because the church has such a big impact on them" gets no support from the mediator, even though it is supportable by the formal statutory rules of custody.

4. 'Developmental appropriateness' operates within the broader paradigm of mediation as a 'rule of persuasion' rather than a 'rule of decision'. The presumption of developmental appropriateness operates within the broader paradigm of mediation, in which emotions and feelings are incorporated. In trying to change parents' perceptions, in a context that recognizes the mutual validity of feelings and facts, mediators will use persuasive techniques to highlight a line of demarcation between children's needs and parents who need their children. This the mediators did in Case Two by storytelling and in Case Three by reframing the issues. Similarly parents draw upon the broader discourse and values of mediation to make their points. Parents may stress a co-parent's negligence or carelessness to bring their positions under the 'best interests' rubric, as Alicia and Jane do in Cases One and Two. They may talk about the children's emotional needs and feelings, as Jane and Marcia do in Cases Two and Three.

Decisions can be reached in mediation only through multi-directional persuasion, agreement, and consensus. Mediators, therefore, must call upon a range of persuasive devices. This article has illustrated several of those which work at the discourse and linguistic level. These include: avoiding responding to some statements by parents; constructing 'collaborative floors' in which the mediator and a parent co-construct a position; reinterpreting and reframing statements made by parents; telling exemplary myths or cautionary tales as models or anti-models for parental behavior; framing issues in ways which exclude certain concerns raised by a parent; and (very occasionally) ending a session without having discussed an issue a parent has raised.

Undoubtedly in some respects the mediator has the upper hand. She wears the court's aura of authority, She possesses personal professional expertise, and so generally has superior skills of persuasion and accumulated knowledge of how parents respond in mediation sessions. She bears the real authority to control turns and topics of talk. When all or part of the mediation is conducted in separate sessions, only the mediator knows everything that has been said.^{^[36]} Mediators also have authority to intensify judicial intrusiveness by recommending a custody evaluation if parents do not reach agreement.

Parents, who have less control of the session's agenda, must employ even greater rhetorical skills on their own behalf. However, failure by a parent to convince a mediator of the inapplicability of the 'developmentally appropriate' parenting plan does not result in the imposition of the plan if the parent finds the mediator's 'developmentally appropriate' suggestions unacceptable. Parental response may range from acceptance to adaptation, to alteration, but the bottom line is that a parent may refuse to sign the agreement. For example, in Case Three, Marcia remains firm about summer vacation time: "I'll go before the judge and fight for half the deal". Paradoxically, then, even though the parents may have an uphill [page 109] job in convincing the mediator not to urge them towards the 'developmentally appropriate', the mediator can only get the parents to accept the developmentally appropriate by persuasion. Sessions like Case One, where the parents agree to a plan quite different from what either wanted at the outset are relatively rare. And there are also cases, although not many, where parents agree to a plan that is far from anything the mediator would have suggested. The majority of cases reaching agreement are like Case Two, compromises between the parents' respective positions. The compromises are more favorable to the parent who has been the stronger speaker and/or whose position the mediator has most strongly supported, and incorporate some specific suggestions (always developmentally appropriate) which the mediator has made.

5. 'Developmental appropriateness' is hardened in strength and specificity through routinization and indirect sanctions. The need for expeditious and adequate agreements within the range of the 'tolerable possible' exerts a standardizing effect on the application of 'developmental appropriateness'. The approximately 11,000 child custody mediation cases handled by the eighteen full time and twelve part time mediators in Los Angeles County involve parents of all social classes, born on all continents save Antarctica. Most of those which reach agreement are resolved in a few hours' time. The cost conscious eye of the state offices of court administration, and the sense of professional success involved in bringing a couple around to a satisfactory agreement, create an ambience which favors going with what usually works and treating the routine as presumptively appropriate Even as professional peer group and state judicial bureaucracy exert indirect pressures and persuasive but perceptible sanctions on the mediators, the mediators gently nudge and sanction the parents through the verbal strategies discussed above.

Further, failure to reach an agreement at mediation is fraught with more formal legal sanctions, although they are indirect. Parents of minor children cannot be divorced until custody of the children has been decided. If parents do not reach agreement at mediation they have only two directions in which to go. They can go back into negotiation, where they must reach an agreement consistent with the law of 'best interests' if the judge is to be induced to sign it. Or they can move on in the judicial process, either to an intrusive custody evaluation in which a court-employed social worker inspects their homes and interviews their children, or to a trial which is likely to entail increased bitterness, time, costs and surrender of their situational control to the judge as a third party decision-maker.

Child custody mediators are ever ready to remind the recalcitrant of these alternatives. Legal sanction is never far away. Marcia's refusal to cave in to the mediator's suggestion is a classic example of the resistance paradox in the legal arena which has recently been so well and ubiquitously documented (e.g. by: Abu-Lughod 1990; Coutin 1993; Ewick and Silbey 1995; Hirsch 1994; Hirsch and [page 110] Lazarus-Black 1994; Merry 1990; White 1990; Yngvesson 1993, 1994). By insisting that her own views be recognized, she undergoes the sanction of entrapment in an escalatingly legalistic system. Although her view may ultimately prevail, it will prevail only by judicial decision. To pursue the possibility of getting her own way she must relinquish autonomy and control.

Court sponsored mediation is expanding in many areas of civil law, raising concerns of fairness, justice, and success. Close discourse analysis shows that mediation cannot be regarded simply as an open field for the expression of raw power or personal empowerment. Rather it is a forum in which norms, which originate in science and may end up in law courts, are uniquely transformed into law through dialogue. If the previous section has shown why 'developmental appropriateness' can properly be called a law of child custody mediation, the discussion has also rendered hard and schematic the soft contours of mediation dialogue in which many stories are told and points pursued simultaneously and in which law is made and practiced without the participants even being aware of it.

Law, in the sense of an objective rule for the resolution of disputes, precipitates from the dialogue. The intersection of praxis (skilled expertise driven by the need for efficiency and standardization), sanction (however remote and whether grounded in the community or the state), and disputes requiring resolution at least on an interim basis (as do custody issues do divorce) generates 'law' as an emergent property.^{^[37]} The content may come from custom or (as in the case [page 111] of 'developmental appropriateness') from science, but praxis, dispute, and sanction transform it into 'law'. And the 'law' itself is transformed by the forum.

In mediation the praxis component has distinctive characteristics: an enabling statutory scheme which imposes specific role conflicts and ambiguities on the mediators; a hybrid discourse in which the professional knowledge of the mental health professions and developmental psychology play a large part; and pressure from the administrative apparatus of the state judicial system to settle many cases quickly with parents who come from a genuinely global diversity of ethnic and economic backgrounds.

Further, in mediation 'law' is dialogued in ways specific to that process, with its multi-valent persuasional fact and feeling mix, so different from the adversarial fact and proof nature of litigation. It is this form of dialogue that gives 'developmental appropriateness' its special character as a presumptive rule of persuasion and its special persuasive effect through its resonance with the holding-and-letting-go transition in parent-child relationships. In mediation power and consensus are shared, in shifting and indeterminate ways, among the disputants and the third party 'neutral'. Much of the sanction is indirect and slowly escalating. The concept of 'developmental appropriateness' might, perhaps, also be a source of law in custody trials. There too it would become law, for there too is praxis, sanction, and dispute. But in the forum of litigation, where facts are hardened by being 'found', the roles of argument and decision are parcelled out to separate players and sanction is immediate. It would be a different 'law'.

A perception that has become common wisdom in legal anthropology is that, as the forums for disputes change, so do the disputants, the issues, the norms, and the alliances (Comaroff and Roberts 1981; Mather and Yngvesson 1980; Yngvesson and Mather 1983; Yngvesson 1993). But, although issues as to the nature of law and the comparison of different laws gave rise to seminal work in the history of legal anthropology. (See e.g. Barton 1919; Bohannan 1965, 1967; Gluckman 1955, 1965; Hoebel 1954; Llewellyn and Hoebel 1941; Malinowski 1951; Pospisil 1958, 1971; Schapera 1937), they have for some years been relegated to the background. Law is just beginning to come back into theoretical focus. (See Biolsi 1995; French 1995; Hunt 1992, 1994; Hirsch and Lazarus-Black 1994; Snyder 1981.) A view of law as an emergent property, to be sought wherever praxis, dispute, and sanction co-exist, reveals the mutually constitutive relationship of law and forum. Popularly mediation is viewed as ideally a content-free, egalitarian process; and its 'legalization' as a kind of procedural pollution resulting from its incorporation by the judiciary. But the development of the 'customary law' of child custody mediation discussed above is but minimally influenced by the law's shadow. For the most part it is the sui generis result of the [page 112] processing of many similar cases expeditiously and efficiently according to similar concepts of what is right and good.

When visualized as 'process' mediation and adjudication are seen as poles apart. But the inherent 'legalization' of mediation suggests the possibility of a more incisive and finely honed comparative jurisprudence, which will emphasize substance and elaborate how a single principle - such as a child's best interests - may transform itself as the reigning paradigm shifts from persuasion to decision. This conception points the way to a culturally contextualized and comparative approach to legal paradigms which will move beyond the law/process and formal/informal dichotomies, not merely in the study of court sponsored mediation as a form of 'alternative dispute resolution', but in legal anthropology and comparative law generally.

1982 "The contradictions of informal justice." Pp. 267-320 in Richard L. Abel (ed.), The Politics of Informal Justice, Vol. 1: The American Experience. New York: Academic Press.

1990 "The Romance of Resistance: Tracing Transformations of Power through Bedouin Women." American Ethnologist 17: 41-55.

1995 "Bringing the law back in: legal rights and the regulation of Indian-White relations on Rosebud Reservation." Current Anthropology 36: 543-559.

1967 "Preface." Pp. xv-xix in Paul Bohannan, Justice and Judgment Among the Tiv (repr. 1989), Prospect Heights, Illinois: Waveland Press, Inc.

1989 "Efficiency and protection, or empowerment and recognition? The mediator's role and ethical standards in mediation." Florida Law Review 41: 253-286.

1996 "Models of mediation", panel presentation at the American Association of Law Schools Annual Meeting, San Antonio, Texas, January 7, 1996.

1992 "The domestication of violence in mediation: the social construction of disciplinary power in law." Paper presented at the Law and Society Association Annual Meeting, Philadelphia.

1993 "Empowerment and mediation: a narrative perspective." Negotiation Journal 9: 245.

1994 "A narrative perspective on mediation: toward the materialization of the 'storytelling' metaphor." P. 48 in Joseph P. Folger and Tricia S. Jones (eds.), New Directions in Mediation: Communication Research and Perspectives.

1991 "Practice and paradox: deconstructing neutrality in mediation." Law and Social Inquiry 16: 35-62.

1991 Rules and Processes: The Cultural Logic of Dispute in an African Context. Chicago: University of Chicago Press.

1988 Fundamentals of jurisprudence: an ethnography of judicial decision-making in informal courts." North Carolina Law Review 66: 467-507.

1990 Rules versus Relationships: The Ethnography of Legal Discourse. Chicago: University of Chicago Press.

1990 "Rules versus relationships in small claims disputes" Pp. 178-196 in Allen D. Grimshaw (ed.), Conflict Talk: Sociolinguistic Investigations of Arguments in Conversations. Cambridge: Cambridge University Press.

1978 Structured Mediation in Divorce Settlement: A Handbook for Marital Mediators.

1993 The Culture of Protest: Religious Activism and the U.S. Sanctuary Movement. Denver: Westview Press.

1985 "Fairness and formality: minimizing the risk of prejudice in alternative dispute resolution." Wisconsin Law Review 1359-1404.

1995 "Gender as an interactional resource in divorce mediation sessions." Paper presented at the Law and Society Association Annual Meeting, Toronto, Canada.

1991 Communication, Marital Dispute, and Divorce Mediation. Mahwah, New Jersey: L. Erlbaum Associates.

1995 "Subversive stories and hegemonic tales: toward a sociology of narrative." Law and Society Review 29: 197-226.

1988 "Dominant discourse, professional language, and legal change in child custody decisionmaking." Harvard Law Review 101: 727-774.

1994 "Ideology, orientations to conflict, and mediation discourse." P. 3 in Joseph P. Folger and Tricia S. Jones (eds.), New Directions in Mediation: Communication Research and Perspectives.

1995 The Golden Yoke: The Legal Cosmology of Buddhist Tibet. Ithaca: Cornell University Press.

1971 "Mediation - its form and functions." Southern California Law Review 44: 305-339.

1963 "The Kpelle moot, a therapeutic model for the informal settlement of disputes." Africa 33: 1-11; reprinted as "Two forms of dispute settlement among the Kpelle of West Africa." Pp. 368-378 in Donald Black and Maureen Mileski (eds.), The Social Organization of Law, New York: Seminar Press, 1973.

1955 The Judicial Process Among the Barotse of Northern Rhodesia. Manchester: Manchester University Press.

1985 Dispute Resolution: negotiation, mediation and other processes. Boston: Little, Brown & Co.

1989 "Selective facilitation: some preliminary observations on a strategy used by divorce mediators." Law and Society Review 23: 613-641.

1994 "The interactive construction of interventions by divorce mediators." P. 89 in Joseph P. Folger and Tricia S. Jones (eds.), New Directions in Mediation: Communication Research and Perspectives.

1991 "The mediation alternative: process dangers for women." Yale Law Journal 100: 1545-1610.

1979 Disputes and Negotiations: a cross-cultural perspective. New York: Academic Press.

1985 "Socio-legal concepts in mediation ideology." The Legal Studies Forum 9: 33.

1994 "Kahdi courts as complex sites of resistance: the state, Islam and gender in post colonial Kenya." In Mindie Lazarus-Black and Susan F. Hirsch (eds.), Contested States: Law, Hegemony and Resistance. New York and London: Routledge.

1994 "Introduction. Performance and paradox: exploring law's role in hegemony and resistance." P. 1 in Mindie Lazarus-Black and Susan F. Hirsch (eds.), Contested States: Law, Hegemony and Resistance. New York and London: Routledge.

1954 The Law of Primitive Man: A Study in Comparative Legal Dynamics. Cambridge: Harvard University Press.

1985 "Mediation: a method of law and psychology in the Netherlands." P. 185 in Craig A. Everett (ed.) Divorce Mediation: Perspectives on the Field.

1992 "Foucault's expulsion of law: toward a retrieval." Law and Social Inquiry 171: 1-138.

1994 Explorations in Law and Society: Toward a Constitutive Theory of Law. New York: Routledge.

1995 "Law and dispute resolution in the Hebrew Bible: Hercules or Solomon?" Paper Presented at the Critical Legal Studies Conference: Contested Communities: Critical Legal Perspectives, September 8-10, Edinburgh, Scotland.

1988 "Phase structure in agreement and no-agreement mediation." Communications Research 15: 470.

1989 "Lag sequential analyses of mediator-spouse and husband-wife interaction in successful and unsuccessful divorce mediation." P. 93 in M. Afzaur Rahim (ed.), Managing Conflict: An Interdisciplinary Approach. Westport Connecticut: Greenwood Publishing Group.

1992 "Time and structure of mediation sessions in the Los Angeles County Conciliation Court." Family and Conciliation Courts Review 30: 474-489.

1994 "Power plays: a sociolinguistic study of inequality in child custody mediation and a hearsay analog solution." Arizona Law Review 36: 879-972.

1941 The Cheyenne Way: Conflict and Case Law in Primitive Jurisprudence. Norman: University of Oklahoma Press.

1995 "Introduction." P. 1 in Bruce Mannheim and Dennis Tedlock (eds.), The Dialogic Emergence of Culture. Urbana and Chicago: University of Illinois Press.

1980 "Language, audience, and the transformation of disputes." Law and Society Review 15: 775-821.

1984 Inside Plea Bargaining: The Language of Negotiation. New York: Plenum Press.

1990 "Narratives and narrative structure in plea bargaining." Pp. 65-95 in Judith N. Levi and Anne Graffam Walker (eds.), Language in the Judicial Process. New York and London: Plenum Press.

1991 "Pursuing settlement in an adversary culture: a tale of innovation co-opted or 'the law of ADR'." Florida State University Law Review 19:1-46.

1982 "The social organization of mediation in nonindustrial societies: implications for informal community justice in America." Pp. 17-45 in Richard Abel (ed.), The Politics of Informal Justice Vol. 2: Comparative Studies. New York: Academic Press.

1990a Getting Justice and Getting Even: Legal Consciousness Among Working Class Americans. Chicago: University of Chicago Press.

1990b "The discourse of mediation and the power of naming." Yale Journal of Law and the Humanities 2: 1-36.

1992 "Culture, power, and the discourse of law." New York Law School Law Review 37: 209-225.

1995 "Roundtable: Mediation, Popular Justice and Transformation." Panel presentation at the Law and Society Association Annual Meeting, Toronto, Canada.

1969 "Styles of court procedure: to make the balance." Pp. 69-91 in Laura Nader (ed.), Law in Culture and Society. Chicago: Aldine.

1984 "The recurrent dialectic between legality and its alternatives: the limits of binary thinking." University of Pennsylvania Law Review 132: 621.

1994 "Controlling processes in the practice of law: hierarchy and pacification in the movement to re-form dispute ideology." Ohio State Journal on Dispute Resolution 9.

1995 "Civilization and its negotiations." Pp. 39-63 in Pat Caplan (ed.), Understanding Disputes: The Politics of Argument. Oxford and Providence: Berg Publishers.

1991 "Profile: child custody mediation services in California superior courts." Family and Conciliation Courts Review 30: 229.

1979 Order and Dispute: An Introduction to Legal Anthropology. Harmondsworth: Penguin Books.

1983 "Strategies in child custody mediation: a family systems approach." Mediation Quarterly 29.

1992 "Clarifying perspectives on mandatory mediation." Family and Conciliation Courts Review 30: 490.

1986 "Law and strategy in the divorce lawyer's office." Law and Society Review 20: 93-134.

1988 "Law and social relations: vocabularies of motive in lawyer/client interaction." Law and Society Review 22: 737-769.

1990 "Legal realism in lawyer-client communications." Pp. 133-151 in Judith N. Levi and Anne Graffam Walker (eds.), Language and the Judicial Process.

1995 Divorce Lawyers and their Clients: Power and Meaning in the Legal Process. New York: Oxford University Press.

1981 "Colonialism and legal form: the creation of customary law in Senegal." Journal of Legal Pluralism and Unofficial Law 9: 49-90.

1985 "Through an interdisciplinary mirror: attorney-therapist similarities in divorce mediation." P. 9 in Craig A. Everett (ed.), Divorce Mediation: Perspectives on the Field.

1986-87 "Psychodynamic perspectives on family mediation." Mediation Quarterly 14/15: 75.

1990 "Subordination, rhetorical survival skills and Sunday shoes: zeroing in on the hearing of Mrs. G." Buffalo Law Review 38: 1-58.

1993 Virtuous Citizens, Disruptive Subjects: Order and Complaint in a New England Court. New York and London: Routledge.

1983 "Courts, Moots, and the Disputing Process." Pp. 51-83 in Keith O'Boyn and Lynn Mather (eds.), Empirical Theories About Courts. New York and London: Longman.

[1] Cal. Fam. Code Section 3170 states:

If it appears on the face of a petition, application, or other pleading to obtain or modify a temporary or permanent custody or visitation order that custody, visitation, or both are contested, the court shall set the contested issues for mediation.

[2] Cal. Fam. Code Section 3711 states: "Mediation proceedings pursuant to this chapter shall be held in private and shall be confidential". The effects of the rule of confidentiality are considered below. The reference to privacy is amplified by Cal. Fam. Code Section 3182, which states that the "mediator has authority to exclude counsel from participation in the mediation proceedings". As a matter of practice, the presence of attorneys at a mediation proceeding is extremely rare. All mediators, and most attorneys, prefer the parents to mediate the custody issues without their lawyers present. Mediators will only allow the attorneys to be present where the parents want them to be and will never allow an attorney to be present on behalf of one party unless all the parties are represented.

[3] If agreement is not reached, the case may be referred for a custody evaluation by trained personnel from a different court department who make homes visits and talk to everyone concerned. Alternatively, it may go directly to the judge or be scheduled for trial while allowing some time before the hearing for the parents or their attorneys to try again to negotiate an agreement.

[4] At the conclusion of the mediation session, parents are given copies of the agreement. Either or both parents have ten days to revoke a signed agreement, by written notice to the mediator. This gives them time to consult with their attorneys.

[5] In 1976 court sponsored mediations became a mandatory pre‑requisite to the litigation of custody in Los Angeles County and several other California counties (McIsaac 1983). In 1981 the rule was extended by statute to all California counties.

[6] In the mid 1970s United States Supreme court cases finding gender discrimination to be unconstitutional sounded the alarm to numerous United States jurisdictions which had presumptively awarded children of `tender years' to mothers. Within a few years, virtually every American jurisdiction had overhauled its family laws to render them gender neutral. The issue of joint or shared custody was, and remains today, much more controversial. California was the first state to give it statutory recognition.

[7] In 1991 Los Angeles County handled a caseload of 11,128 mediations

(Ricci 1991). David Kuroda, Division Chief, Mediation and Conciliation, Los Angeles Superior Court, Conciliation Services Division, estimates that Los Angeles County handled approximately 11,000 mediations in 1993.

[8] Mediation was first advocated and inspired by anthropologists and other social scientists whose nonwestern fieldwork impressed upon them the virtues of resolving disputes by working through their human dimensions in a culturally sensitive context, and the way in which persistent multiplex relationships which would endure after the dispute formed an incentive to peaceful settlement (e.g. Gibbs 1963; Gulliver 1979; Nader 1969; Roberts 1979). The idea was seized upon especially by attorneys and others working with divorcing families as an antidote to the proverbial escalating bitterness of divorce litigation and its detrimental effects on children. They argued that contemporary families, like small scale traditional communities, had multiplex interrelationships which must, inevitably, continue even after the divorce was over (Fuller 1971; Coogler 1978; Moore 1986; Folberg and Taylor 1984; Goldberg et al. 1985; Keltner 1984).

[9] See e.g:

[A]lternative forms of dispute resolution ... could lead to outcomes that were efficient in the Pareto‑optimal sense of making both parties better off without worsening the position of the other. In addition, the processes themselves would be better because they would provide a greater opportunity for party participation and recognition of the party goals. (Menkel-Meadow 1991)

[S]elf‑ determination and empowerment are furthered through outcomes that are designed by and for the parties, rather than outcomes designed by and (at least in part) for outsiders. Mediated outcomes empower parties by responding to them as unique individuals with particular problems, rather than as depersonalized representatives of general problems faced by cases of actors or by society as a whole.... Here, then, are the special powers of mediation: It can encourage personal empowerment and self‑determination as alternatives to institutional dependency, and it can evoke recognition of common humanity in the face of bitter conflict. Both powers involve resorting to the individual a sense of his own value and that of his fellow man in the face of an increasingly alienating and isolating social context. These are valuable powers indeed. Further, they are unique to mediation. These are functions mediation can perform that other processes cannot. (Bush 1989)

See also the writing of practitioners of mediation: Coogler 1978; Moore 1986.

[10] Sara Cobb (1992) analyzes texts of mediation sessions dealing with domestic violence to show how the narrative of violence is `domesticated' and disappears from conversations later in the mediation session. Cobb (1992) and Greatbatch and Dingwall (1989) pioneered the application of conversational analysis methods to family mediation texts. They coined the term `selective facilitation' to describe the way mediators employ the underlying values of mediation to persuade the participants towards certain resolutions rather than [page 79] others.

[11] According to David Kuroda (above note 7), as of October 1994 the approximately 11,000 cases a year were being handled by 18 full time mediators, and twelve others who worked part time `as needed'.

[12] Multiple role ambiguities are encoded in the job objectives of California child custody mediators. These include: achieving full parental cooperative understanding v. formulating a parenting plan within a reasonable time period; enabling parents to negotiate their own agreements v. representing the child's best interests; allowing the plan to evolve naturally from interparental dynamics v. empowering a less empowered parent; and permitting parents to talk freely and confidentially v. complying with an obligation to report abuse or give a parent an individual session upon request (Kandel 1992). In jurisdictions where mediators

make recommendations to the judge they also face the ambiguity of being a confidante v. being an expert witness.

[13] In a minor, but not insignificant percentage of cases children (beginning with those aged six or seven and with the proportion increasing with age) do participate in mediation sessions, either because parents bring them or because mediators ask to see them. To avoid putting children in the troubling position of having to "side with" or "choose between" parents, however, much of the discussion between children and mediators takes place without the parents being present and children are rarely asked directly to make choices about their custody. Thus, even when a child does participate in the mediation session, it is appropriate to view the mediator as representing the `absent child'.

[14] Thus the virtue and necessity of continuing contact with both parents is fundamental and reiterated throughout the statutory scheme (Cal. Fam. Code Sections 3020, 3040(a)(1), 3161(b), 3162(b)(1)), but this right is double‑edged, belonging to the child as well as the parent. Cal. Fam. Code Section 3162(b)(1), for example, setting forth uniform standards of practice for mediators, requires them to safeguard "the rights of the child to frequent and continuing contact with both parents" (emphasis added). When one mediating parent complains that the other has not paid any attention to the child, and therefore doesn't deserve much parenting time, the mediators respond that it is the child's right to have a relationship with even a non‑deserving parent, provided that parent will not do harm to the child.

[page 81]

Cal. Fam. Code Section 31180(a) states that the mediator is "entitled to interview the child where the mediator considers the interview appropriate or necessary". However, very young children are not interviewed, and children in the middle years are generally not interviewed unless their parents bring them to the mediation session. Further, as a matter of policy, to avoid placing children in a psychologically difficult circumstance or giving them the perception that they must make a public choice between their parents, mediators rarely speak to the children while the parents are present. Consequently, during sessions with the parents it is the mediator who is obliged to take a position on behalf of the child which sometimes differs from the position the parents advocate.

[15] The standard appears repeatedly in the California statutory scheme. Cal. Fam. Code Section 3180(b) (West 1994) states in its pertinent part that "the mediator has the duty to assess the needs and interests of the child involved". Cal Fam. Code Section 3162(b)(1) (West 1994) states that mediation shall be conducted in accordance with standards which include a "provision" for "the best interest of the child...."

[16] Virtually the only `interest' codified in the statute is the interest in having frequent, continuing contact with both parents. See Cal. Fam. Code Section 3161 (West 1994); Cal. Fam. Code Section 3020 (West 1994), providing that "it is the public policy of this state to assure minor children frequent and continuing contact with both parents ... and to encourage the parents to share the rights and responsibilities of child rearing ..."; Cal. Fam. Code Section 3162(b)(West 1994), requiring that mediation standards of practice include "provisions for ... the safeguarding of the rights of the child to frequent and continuing contact with both parents".

[17] Mediators' distinctive and hybrid professional training and knowledge includes specific skill in the facilitation of mediation sessions, knowledge of child development, knowledge of the substantive law of child custody and other aspects of family law, and situationally specific knowledge about which judges are likely to make what sorts of custody rulings in which types of cases. Part of this knowledge is local: the information shared in the courthouse corridor about specific judges and child custody evaluators or the ideas and suggestions exchanged at staff meetings. Part of it is global. Many child custody mediators also identify with and participate actively in the broader professional community of alternative dispute resolution practitioners, learning from and contributing to such professional organizations as the American Association of Family and Conciliation Courts, SPIDR (Society for Professionals in Dispute Resolution), and professional publications such as Mediation Quarterly, and Family and Conciliation Courts Review.

[18] Pursuant to Cal. Fam. Code Section 1815, mediators must hold "a master's degree in psychology, social work, marriage, family and child counselling, or other behavioral science substantially related to marriage and family interpersonal relationships" and have "at least two years' of experience in counselling or psychotherapy, or both, preferably in a setting related to the areas of responsibility of the family conciliation court and with the ethnic population to be served." Additionally mediators must possess knowledge of the California court system and family law procedures, other available community resources,

adult psychopathology and the psychology of families.... child development, child abuse, clinical issues relating to children, the effects of divorce on children, the effects of domestic violence on children, and child custody research sufficient to ... assess the mental health needs of children.

However, "the family conciliation court may substitute additional experience for a portion of the education, or additional education for a portion of the experience." In Los Angeles County positions for child custody mediators are additionally [page 83] advertised as requiring five years of related experience.

[19] Cal. Fam. Code Section 3183(a) provides merely: "The mediator may, consistent with local court rules, submit a recommendation to the court as to the custody of or visitation with the child". Where local court rules do not permit mediators to make recommendations to judges, mediators can, however, always recommend that a child custody evaluator (usually a social worker who works as part of the court staff) be appointed by the judge to make a home visit and issue a report with a recommendation regarding custody. They can also recommend a psychiatric evaluation (which is less frequent). In the California jurisdictions where local rules permit recommendations to judges about what they think the proper custodial arrangement should be, they are then in the anomalous situation of having to recommend a result while being prohibited, by the rule of confidentiality, from stating the facts on which the recommendation is based.

[20] Sometimes they also specify which parent will be responsible for transportation and what notice has to be given in case of a change in the arrangements. They usually also include certain other `boilerplate' clauses such as a statement that neither parent may speak ill of the other.

[21] To protect the confidentiality of the parties, all names are pseudonyms and the names of the mediators are not used.

[22] Cal. Penal Code Section 273a(b) states:

Any person who, under circumstances or conditions other than those likely to produce great bodily harm or death, wilfully causes or permits any child to suffer, or inflicts thereon unjustifiable physical pain or mental suffering, or having the care or custody of any child, wilfully causes or permits the person or health of that child to be injured, or wilfully causes or permits that child to be placed in such a situation that its person or health may be endangered, is guilty of a misdemeanor.

[page 86]

Quite obviously, the words "unjustifiable physical or mental suffering" are `waffle words' which allow for a multitude of differing interpretations depending upon one's philosophy and cultural tradition of child raising.

[23] Mediators usually inform parents at the beginning of a session that child abuse statements are not covered by the confidentiality requirements. Yet mediators interpret this statute commonsensically, veering parents' talk away from disclosures of mild slaps or spankings, while warning them that such behavior is technically illegal. The mediator's introductory remarks to Craig and Alicia are fairly typical:

Mediator: The only kinds of things that are not confidential would be like ... any indication that the child is in danger. Danger does not mean that he was left with a sitter and he stayed up late and didn't eat his vegetables. Danger is serious physical abuse, sexual abuse or something of that sort.

The seemingly redundant qualification in the phrase "serious physical abuse" conveys the mediator's ambivalence towards the statutory standard.

[24] See previous footnote.

[25] For example, Jane says that once Clara came home `glazed over' and she took her to the doctor. The doctor said it could be emotional trauma or "it could be she's coming down with something, I don't know." When the mediator asks Jane what it ended up being, Jane merely shrugs. Steven is a far stronger speaker. He is able to caricature Jane's cautiousness when he tells a story about how Jane mistook a spot of spaghetti on Clara's head for blood, and thought she had been abused (Kandel 1994: 922). For a fuller discussion of the rhetorical strategies used in this case see Kandel (1994).

[26] There was some dialogue in the mediation session to the effect that the children had sometimes wanted to go to parties or other events during Marcia's visitation time - which she was reluctant to allow them, although there were also statements to the effect that they were just as happy not to attend other activities.

[27] Above note 16. Furthermore, Cal. Fam. Code Section 3040(a)(1) states, in the pertinent part: "In making an order granting custody to either parent, the court shall consider, among other factors, which parent is more likely to allow the child frequent and continuing contact with the noncustodial parent...".

[28] I would settle for a dictionary definition if I could. But Black's Law Dictionary, Special Deluxe Fifth Edition (1979) sitting on my bookshelf has nearly two full columns of definitions and still doesn't get anywhere near the

power/law/custom debate.

[29] In both the formal and informal taxonomy of ethnicity used in California `Latino' is a broad generic term. It is usually juxtaposed to African American, Asian American, Native American and Anglo (which includes all Americans of European ancestry from countries other than Spain). Indicia for characterization as `Latino' include having Spanish as a first language, having parents, grandparents or great-grandparents whose first language was Spanish, having immigrated from a Latin American country, having ancestors who immigrated from a Latin American country, having a Spanish surname, and self-classification as `Latino'. The category includes people from a multitude of national origins and an even greater multitude of ethnicities, and includes a substantial portion of the population of Greater Los Angeles.

[30] There is statutory justification for Marcia's position that the children's best interests demand a deviation from the usual application of `developmental appropriateness'. That is to be found in the repeated statutory references to the importance of "frequent and continuing contact" with both parents: above, notes 16, 27. This, of course, is never mentioned in the mediation session.

[31] From which it emerges that, in small claims courts, where formal rules of decision can theoretically give way before the customs of claimants, the majority of judges favor rule-oriented litigants to those whose customary law is grounded in ideas about multiplex social relationships; and that this is the case even when the rule-oriented litigants do not argue for the same rules as those which more formal American courts would apply.

[32] Showing that prosecutors and defense counsel develop quite ritualized ways of plea bargaining depending upon what a case is worth in terms of evidence, severity of the crime, and other matters.

[33] In which analyses of office conversations between divorce lawyers and their clients reveal surprisingly little discussion of formal law. Rather, divorce lawyers emphasize a Realpolitik of customary law, stressing their judicial connections and judges' personalities, while clients demand a holistic divorce including both psychosocial and legal components.

[34] Showing how clerks, as gatekeepers in Massachusetts lower courts, decide what is a `garbage case' and what is a criminal complaint, while using their knowledge of community standards to informally mediate some of the `garbage'.

[35] According to Cal. Evidence Code Section 605,

A presumption affecting the burden of proof is a presumption established to implement some public policy other than to facilitate the determination of the particular action in which the presumption is applied, such as the policy in favor of establishment of a parent and child relationship ...

According to Cal. Evidence Code Section 604,

The effect of a presumption affecting the burden of producing evidence is to require the trier of fact to assume the existence of the presumed fact unless and until evidence is introduced which would support a finding of its nonexistence, in which case the trier of fact shall determine the existence or nonexistence of the presumed fact from the evidence and without regard to the presumption.

[36] When the mediation is conducted in whole or part through separate sessions, each parent learns what the other has said only through the mediator's report. Such reports are never complete nor verbatim, not only because of the cognitive and linguistic impossibility of such reportage and the perceptual idiosyncrasies which inevitably influence how summaries are constructed, but also because of deliberate strategies employed by the mediators. In particular, mediators promise each parent individual confidentiality and do not repeat what parents do not want them to (although such statements inevitably influence how the mediator pursues agreement). Also, when shuttling between separate sessions, mediators act in a negotiational role and will emphasize, downplay, evaluate or interpret the other parent's position and their firmness or flexibility in order to facilitate agreement.

[37] Bruce Mannheim and Dennis Tedlock explain:

The principle of emergency has been most clearly enunciated by biologists, who try to understand how life forms have emergent organizational properties that cannot be predicted from their constituent parts. In the words of Ernst Mayr, `When two entities are combined at a new level of integration, not all the properties of the new entity are necessarily a logical or predictable consequence of the components.' This definition stresses two conditions: first, that the new level of organization have its own

principles of patterning that cannot be reduced to its component principles; and second, that the new level of organization include some degree of contingency. (Mannheim and Tedlock 1995: 9, citations omitted).