Managing Contact Difficulties: A Child-Centred Approach



2.5 What Are the Major Formulations of Contact Difficulties?

The literature concerning contact difficulties has grown exponentially since 1985. Two formulations for understanding difficult post-divorce contact between children and parents dominate the literature and are described below (Gardner, 1992; Johnston, 1993; Johnston and Kelly, 2001). Stoltz and Ney (2002) extended the Kelly and Johnston formulation to consider multiple contributing factors. More recently, Ney and Blank (in preparation) argue that both formulations (Gardner, and Kelly and Johnston) are embedded in a medical or legal perspective. They conclude that resolution of conflict is more difficult when one perspective is given more weight than other perspectives.

Each formulation (Gardner, and Kelly and Johnston) leads to recommendations for intervention with children and parents. Yet there is a dearth of scientifically sound and valid research in this area of practice. Bruch (2001: 550) reminds us to proceed cautiously, "Lawyers, judges, and mental health professionals who deal with child custody issues should think carefully and respond judiciously when claims based on either theory [Gardner or Kelly and Johnston] are advanced."

Richard Gardner (1992), and Janet Johnston and Joan Kelly (2001) have proposed two different formulations for conceptualizing and addressing contact difficulties. They are described below, along with related contributions by other authors.

2.5.1 Gardner

Gardner (1992: xviii) stresses that parents who alienate a child from the other parent perpetrate "a form of emotional abuse" because the child is denied a loving relationship with one parent. Furthermore, he argues that alienation induces lifelong psychiatric disturbance in the child. His definition is included in Appendix B.

Gardner (1992) also suggests that in some cases the parent may use the child's internal state (e.g. temperament) to contribute to or promote the alienation. Such factors include:

Over the years, Gardner has refined and amended both his original definition and his comments about the gender of alienating parents (2001a, 2001d, 1999a, 1999c, undated b, undated c, undated d). The major shifts that have occurred in his thinking take account of conscious as well as subconscious and unconscious factors within the alienating parent that influence the child's alienation from the other parent.

More recently, Gardner expanded the definition to take account of factors within the child or the situation that may contribute to the presentation. These factors, he suggests, operate independently of the alienating parent's contribution. He explains that:

The disorder refers to a situation in which the parental programming is combined with the child's own scenarios of denigration of the allegedly hated parent. Were we to be dealing here simply with parental indoctrination, I would have probably stuck with brainwashing and/or programming (Gardner, 1992: xvii).

He argues that it is the exaggeration of minor weaknesses and deficiencies of the alienated parent that is indicative of PAS (2002, 2001b, 2001d, 2001e).

Gardner (1992) attributes the following characteristics and reactions to children who experience alienation:

Gardner speculates (1992: 62) that two factors have caused an apparent increase in the incidence of parental alienation: the shift from the concept of tender years doctrine to best interests of the child, and the increasing popularity of shared parenting arrangements. He suggests that these factors result in more litigation and competition between parents and argues that mothers perceive their bond with the child as threatened because courts look to other factors than the mothers' traditional nurturing role when making parenting plans and schedule decisions for children— "… these changes have placed women at a disadvantage in custody disputes". Consequently, some parents adopt parental alienation as a strategy to help ensure victory in litigation. He also comments that "… when a sex-abuse accusation emerges in the context of a PAS—especially after the failure of a series of exclusionary manoeuvres—the accusation is far more likely to be false than true" (1998: xxvii).

In 1992, Gardner argued that the vast majority of "programmers" are mothers and estimated they are responsible for ninety percent of cases of alienation. Since that time, his position about gender has changed and he notes that "… in the last few years I have seen a shift that has brought the ratio now to 50/50" (Gardner, 2000: 442).

According to Gardner, PAS can be mild, moderate, or severe. These distinctions are linked to guidelines for contact and type of intervention (1992, 1998a, 1998b, 1999b, 1999c, 2001b, 2001c, undated c). He recommends that courts base their parenting plan decisions on the "… stronger-healthy-psychological-bond presumption" (1992: 263). This presumption reflects a three step process involving preference for the parent with whom the child has the stronger psychological bond, preference for the primary caretaker during the child's early years, and recognition that the psychological bond can change over time and be less important for older children.

Gardner has a clear position about forcing contact. He states that court orders with respect to contact are not required for alienation judged to be in the mild category. If litigation regarding custody ceases, he believes the alienating behaviour is reduced and the children become asymptomatic. He argues that for moderate cases, parents need threats and sanctions if they fail to honour the child's schedule. However, he points out that programming parents are "notoriously uncooperative". Admonishments and requests to "cease and desist", in Gardner's opinion (1992), are ineffective and court orders are required. When cases fall into the severe category "… custodial transfer is the only hope for the children if there is to be any alleviation of their PAS symptomatology" (1992: 64). Contact between the programming parent and child, according to Gardner, should be prohibited for a period of time to allow the child to adjust. Eventually, contact should gradually resume. In an address to members of the Family Law Bar in Toronto in May 2000, Gardner also stated that when children continue to refuse to spend time with a parent, hospitalization or incarceration of the child might be appropriate.

2.5.2 Johnston

Johnston (1993: 111) took issue with Gardner's formulation because it fails to differentiate between alienation and a child's reluctance to spend time with a parent. She stated that:

Reluctance to visit includes a broad range of observable behaviour in which the child, for any reason, verbally or gesturally complains about and resists spending time with the nonresidential parent. The resistance may be manifest only at the time of transition or it may involve intermittent or ongoing complaints about visits. In extreme cases, it can encompass a complete refusal to have contact with the other parent...the child may or may not be hostile or negative to the extreme cases there is often expressed fear and negativity...Parent-child alignment and parent-child alienation are defined as the child's making an overt or covert attitudinal or behavioural preference for one parent and, to varying degrees, denigrating and rejecting the other parent...[it] involves a negative, conflictual, or avoidant relationship between the child and the rejected extreme cases of strong alignment with the residential parent the child usually refuses to spend time with the nonresidential alienated parent.

Johnston believes that resistance to contact is a complex phenomenon that has its "… origins in diverse and multiple psychological, developmental, and family system factors" (1993: 133). Drawing on samples from research she and Linda Campbell conducted with low and middle income high-conflict families (Johnston and Campbell, 1988) and children from upper income high-conflict families (1993), Johnston noted that approximately forty five percent of the children had formed an alignment with one parent. She identified six explanations for a child's alignment and reluctance or refusal to spend time with a non-residential parent. However, she indicates (1993: 132) that the findings are tentative and their interpretation speculative because they are "… based on simple correlations and clinical observations from relatively small samples of high-conflict divorcing families". In more recent work (in press), she states that the first two explanations are developmentally normal responses unrelated to pathology in either parent or the child, the next three are related to pathological family processes, and the sixth explanation is a tertiary or systemic process.

Johnston's first explanation for resistance to contact is a child's basic anxiety about separating from the primary attachment figure, especially when parents are overtly conflictual with one another. Johnston observed that the child's anxiety was heightened when there was ongoing conflict and overt aggression between the parents. She also noted that many of the parents were:

... ambivalent or skeptical of the value of visitation, especially when the child was symptomatic and resistant at transitions; these parents were not well suited to soothing the child and making the child feel safe and competent in handling the changes (1993: 118).

She concluded that these children were not significantly disturbed and their resistance was a developmentally explainable divorce-specific separation anxiety.

Second, the child's limited cognitive capacity to be aware of both parents' opposing viewpoints and feelings may result in an alignment that becomes the resolution of painful loyalty conflicts. She reports a significant proportion of children were not able to separate themselves from the parental conflict and that boys were more likely to be symptomatic than girls were. Third, the intensity and longevity of parental disputes may result in the children developing alignments as a defensive mechanism to cope with the ongoing conflict.

A fourth explanation involves the child's inability to extricate his or her feelings and ideas from those of the distressed residential parent. She points out that this group of children tends to reflect whatever the parent needs in an attempt to ensure their own needs are met. They may become vigilant and highly attuned to the alienating parent. Consequently, there is limited opportunity to experience feelings and ideas separate from this parent. Their concern for this parent often makes it difficult for them to leave the residential parent to spend time with the other parent.

The fifth factor is exposure to emotional abuse and physical violence between parents. Johnston (1993: 129) notes that an alignment may be rooted in "… early, barely remembered trauma of domestic violence". In some cases, these children did not feel it was safe to leave the residential parent (mother), their home, or pets. Other children in this group:

... entered into a state of folie à deux with an abused mother, with a narcissistically injured father, or with a paranoid parent, a state in which reality, fears, and fantasies about the excluded parent were inextricably entwined for the aligned parent and child (1993: 129).

The sixth explanation is the child's sense of counter-rejection and retaliation by the rejected parent and his or her social network. Johnston points out that rejected parents become hurt and sometimes outraged by the child's response. Attempts to assert their parental position or pursue the child by letter and telephone are usually unsuccessful. These strategies usually lead to more avoidance on the child's part. The alliance, according to Johnston, is intensified because the child is hypersensitive to and hurt by counter-rejection. From a clinical perspective, she concludes that the child is confused and overwhelmed by guilty feelings and wishes to be rescued from the "intolerable dilemma". The child's behaviour continually tests the depth of concern of the rejected parent.

2.5.3 Kelly and Johnston

Kelly and Johnston (2001: 251) reformulated Johnston's earlier concept of the alienated child. They defined the alienated child as one who "… expresses, freely and persistently, unreasonable negative feelings and beliefs (such as anger, hatred, rejection, and/or fear) toward a parent that are significantly disproportionate to the child's actual experience with that parent". They suggest that a reformulation to an objective and neutral focus will make it easier to distinguish between an alienated child and one who resists visitation.

Their model views the child-parent relationship on a continuum ranging from positive relationships with both parents to a progressively stronger alignment with the residential parent. The other end of their continuum is complete alienation from the non-residential parent. The extent to which the estrangement from a non-residential parent is realistic decreases as one moves along the continuum.

Kelly and Johnston (2001: 254) argue that effective diagnosis and intervention is based on a systems framework that takes into account "… the multiple and interrelated factors influencing the child's response during and after separation". Their model includes other factors that contribute to the alienation, such as parental attitudes and behaviours, in addition to developmental or cognitive factors within the child. Reasonable and unproblematic child responses are explained by factors such as the child's age and stage of development and ongoing conflict between parents. Reasonable but problematic child responses (at the more extreme end of the continuum) are explained by factors such as child abuse, the dynamics of separation, parenting capacity and angry, rigid, and restrictive parenting styles. They point out that even when a child does not meet the criteria to be considered alienated, many of these factors may be present. If this is the case, the potential to be considered alienated in the future must be considered.

According to Kelly and Johnston, the risk factors that predict alienation are:

They also identify the range of behaviours common to both alienating and rejected parents. According to Kelly and Johnston (2001), the alienating parent has extremely negative views of the rejected parent and maintains a deep distrust and fear of his or her former partner. The alienating parent does not believe that the child needs the other parent and may interfere with contact. In more extreme cases this parent will also remove references to the rejected parent. These parents argue that the child should have the right to make decisions about schedule issues. The rejected parent is viewed as a danger to the child and is often seen as never having cared for the child. Kelly and Johnston maintain that empirical research and clinical observation support their conviction that there is significant pathology and anger within the alienating parent.

Rejected parents contribute to the alienation, however their behaviours "… do not by themselves warrant the disproportionately angry response of the child nor the refusal to have contact" (Kelly and Johnston: 258). Their parenting capacity, usually within the normative range, may be compromised by one or more of the following factors: the ongoing high conflict between parents, counter-rejection of the alienated child, a rigid and harsh parenting style or immaturity exaggerated by the ongoing conflict, attribution of new meaning to their behaviour and diminished empathy for the aligned child.

Age and cognitive capacity, the extent to which the child views the divorce as abandonment, and temperament and personality vulnerabilities, moderate a child's response to alienation. Other parent-child relationship factors may also explain the child's response. These can include the child's dependence on the aligned parent, reliance on alienation as a coping strategy or attempts to "rescue" a vulnerable parent. Kelly and Johnston argue that a history of infrequent or no contact with the rejected parent increases child vulnerability. When contact is interrupted, a child has little or no opportunity to evaluate her assessment of the rejected parent. They also suggest that supervised contacts often implemented during abuse investigations reinforce the child's belief that the rejected parent presents a danger. The child's vulnerability is further increased if they are emotionally isolated and have limited access to external resources (i.e. therapists or significant others.) Kelly and Johnston identify the similarity between their observations of the clinical presentation of this group of children and that of Gardner (1992) and Wallerstein (1985).

In summary, Kelly and Johnston recommend that the complexity of these cases require:

... a full assessment to understand the multiple determined factors and influences leading to the children's abrupt rejection of a previously acceptable and meaningful relationship. Each of these influences has there [sic] own particular weight and significance for a particular child in a particular family. No one factor produces the alienated child (Kelly and Johnston, 2001: 264).

They argue (Kelly and Johnston, 2001: 264) that only a "full understanding of this pathological development in the parent-child relationships" leads "to an effective plan and structure for legal, judicial, and therapeutic interventions directed at resolving the profound alienation of the child from the parent."

2.5.4 Stoltz and Ney

Stoltz and Ney (2002: 222) extended the Kelly and Johnston (2001) reformulation of the alienated child. They take issue with how Kelly and Johnston explained children's reasonable and unreasonable responses to alienation and state:

These would be reasonable criteria for assessment if it were not for the failure to continue to take into account other powerful contextual factors. What we are suggesting is that the responses seen as unreasonable are also reasonable, adaptive responses when considered in context and that the failure to see them as such has serious implications.

They propose that excluding the context in which the child's response occurs creates a flawed understanding of the dynamics unfolding in the post-divorce family.

Stoltz and Ney (2002: 36) prefer the term resistance to visitation. Resistance:

...incudes the broad continuum of behaviors of all parties involved (parents, children, lawyers, family, professionals, etc.), ranging from (for example) voiced complaints, to repeated incidences of lateness in dropping the child off, to a child's refusal to go with the noncustodial parent, and so on.

Their objective in reframing the problem as one of resistance is to focus attention on the dynamic, as opposed to the individuals. Stoltz and Ney (2002: 227) argue that this reformulation allows for other contributing factors to be considered, including "… the adversarial influence of the legal system and the possibility that professional intervention (e.g. psychological assessment) can be interpreted through the legal framework and perceived as threatening, thereby adding to resistance".

Stoltz and Ney hypothesize that if divorce were managed in a less adversarial manner, there would be little resistance to visitation. Working from the revised formulation of resistance, they propose a number of practice implications, including the need for:

2.5.5 Related Work

Other authors have based assessment and intervention recommendations on the concept of alienation and have attempted to extend Gardner's formulation.

Lund (1995) provides a therapist's view of mediation. She suggests that alienation is usually reflective of high parental conflict and psychopathology in both parents. She broadens Gardner's framework for understanding these families in order to move away from a blame-based formulation and to consider viable treatment options. Lund, like Johnston (1993), proposes several possible explanations for parent rejection, such as developmentally normal separation problems, deficits in the non-custodial parent's skills, oppositional behaviour on the part of the child, high parent conflict, serious relationship difficulties (excluding abuse) and child abuse. She acknowledges the possible contribution of extended family and support systems. Like Gardner, she believes that the presence of alienation is determined by "… the extent to which a child is consciously or unconsciously being programmed" (Lund, 1995: 311). Based on this formulation, Lund calls for strong direction from the court to manage these cases. Intervention recommendations are predicated upon a team approach that includes parent-child sessions, individual therapy for parents, mediation to resolve conflict and communication between therapists working with the family.

Darnall (1998) seems to accept Gardner's conceptualization of alienation, but distinguishes between the concept and the syndrome. His definition is included in Appendix B. In summary, he suggests that alienation is any constellation of behaviours that disturb the relationship between a child and one parent. One of his contributions to this area of practice is his categorization of alienating parents as "naïve", "active" or "obsessed".

2.6 Critique of Dominant Formulations

The debate about parental alienation syndrome (PAS) in the literature has tended to focus on four issues:

2.6.1 Terminology

Since the early 1990s, debate has occurred as to whether the concept of alienation and PAS exist. There is considerable scepticism. Many clinicians, and certainly the majority of key informants, view the terminology as moralizing and victimizing (Etemad, 1997). Johnston (2001: 2) believes this terminology is an attempt to medicalize the symptoms. In her opinion "... the term PAS does not add any information that would enlighten the court, the clinician, or their clients". Freckleton and Selby (2002: I-3420) echo this perspective. They suggest that Gardner's terminology draws on the legacy of rape, battered women and child sexual abuse. They state:

The misleading pseudo-scientific patina of objectivity and reliability provided by the word "syndrome" appears not to be justified…it may well be that the disadvantages of focus upon the scientific standing of parental alienation syndrome may be such as to make its reception unhelpful or even counterproductive (2002: I-3420).

As part of this debate, numerous authors suggest that contact problems exist, but that the terminology is not useful. For example, Sturge and Glaser (2000) suggest the term "implacable hostility" would be more helpful. They argue that using the label of PAS assumes a cause that leads to a proscribed intervention. However, in their opinion, the complexity of situations and differences between families require a variety of interventions. Use of terminology such as PAS is generally viewed as pathologizing and creating a focus on power and control issues. It does not promote positive outcomes for children or increase the likelihood of successful intervention.

Some of those who support the concept suggest that the difficulty is the use of the word "syndrome" (Hayward, 1999). Others, such as Turkat (1997), differentiate within the formulation of PAS. For example, he argues there may be acute interference that does not involve a systematic or devious plan. There may be direct, as well as indirect, PAS.

Some key informants suggested that the terminology of PAS was appealing to parents and practitioners seeking easy explanations for complex situations, but that the labels do not adequately capture the nuances of the situation. At the present time, these labels simply elevate alienating behaviours to a quasi-clinical status and draw attention away from what is really happening to the child. Several informants noted that Gardner has stated there might be other factors that explain these situations. They stressed the need for objective clinical judgements to ensure decision-making takes into account a variety of factors (child's age and stage of development, the co-parenting relationship, the non-parental relationship between the adults, parenting capacity and mental health issues).

The terminology issue is part of a much larger debate about language within the context of family change. Mahony's (2001: 5) challenge in this regard is timely. He writes "… the language in our legislation relating to guardianship, custody and access and the way in which terms are defined, need to be brought into line with current ways of thinking about children's rights and interests with corresponding parental responsibilities".

2.6.2 Conceptualization

In addition to concerns about the utility of terminology such as PAS, there is significant criticism of Gardner's formulation. Bruch (2001) argues that Gardner overstates the prevalence of PAS, fails to recognize predictable post-divorce behaviour and confounds a child's developmentally related reactions with psychosis. Bruch (2001: 550) states that "PAS as developed and purveyed by Richard Gardner has neither a logical nor scientific basis. It is rejected by responsible social scientists and lacks solid grounding in psychological theory or research". Others suggest his research methods and techniques lack reliability and validity, and have not been subjected to peer review. They also note that his books are self-published (Freckelton and Selby, 2002; Zirogiannis, 2001). Supporters of Gardner's formulation and terminology acknowledge its value, but recognize the weaknesses of the conceptualization (Etemad, 1997), including its weak presentation of data, moralizing tone, creation of a "good parent/bad parent" scenario and inadequate empirical support.

Several types of concerns about Gardner's formulation are identified in the literature. These focus on its:

Other concerns relate to:

(Bruch, 2001; Etemad, 1994; Faller, 1998; Freckleton and Selby, 2002; Johnston, 2001; Peralta-Vaughn, 2001; Rybicki, 2001; Smith and Coukos, 1997; Waldron and Joanis, 1996; Warshak, 2000a; Warshak, 2000b).

In general, Johnston (1993) and Kelly and Johnston's (2001) formulations have been more favourably received. Warshak (2000a) questions how changing the terminology to that of the alienated child helps to ensure more accurate identification of alienation in divorcing families. The Johnston and Kelly terminology, he asserts, is ambiguous, although he acknowledges that they are working to clarify the terminology in their model. He argues that Johnston and Kelly confuse estrangement and alienation. Warshak proposes that the Johnston and Kelly formulation is more similar to Gardner's than it is different with respect to parent behaviours and child responses. However, according to Warshak, Kelly and Johnston recognize the contribution of children, and to a much lesser extent, that of the parent who is alienated.

Gould (1998: 172) comments that whether one " … can definitively establish a "syndrome" or not is less important than the task of helping divorced families heal" and the task of establishing "… a fact pattern of systematic negative influence by one parent upon a child that interferes significantly with that child's ability to form a healthy bond with the other parent". Johnston (2001: 2) concurs, stating that children and parents "…would be better served by a more specific description of the child's behavior in the context of his family".

2.6.3 Evidentiary Criteria

As of this writing, PAS is not included in either the American Psychiatric Association's (2000) diagnostic manual (DSM-IV-TR), except perhaps under the general condition of "Parent-Child Relational Problem." There is no consensus in the mental health literature with regard to the terminology of PAS, either as to whether the concept meets the evidentiary requirements to be considered a syndrome or as to the most effective ways of intervening with the subgroup of post-divorce families who experience contact difficulties.

Warshak (2002) argues that there is an important need to clarify terminology to reduce the incidence of misdiagnosis, and he points out that it is accepted practice for social scientists to use clinical records and observation as a first step in drawing attention to new phenomena. Definitions develop over time based on repeated observations drawn from multiple data sources. Ultimately, the conceptualization that emerges from clinical reports is investigated with larger samples, standardized measures and control groups. The research is normally subjected to rigorous peer review in order for conclusions to be accepted by mental health practitioners (Warshak, 2000a).

Laing (1999) pointed out that a number of people who provided evidence to the Special Joint Parliamentary Committee on Child Custody and Access in Canada referred to clinical syndromes (PAS being one) that have not been subjected to the generally accepted research criteria to which Warshak refers. Cartwright (1993) counters this type of criticism by suggesting that as the phenomenon becomes better understood, it will be redefined. Gardner has, in fact, revised his model several times since 1985. Nevertheless, there is an absence of research with respect to Gardner's concept of PAS.

Recently, authors have addressed the difficulties inherent in utilizing syndrome evidence and clinical and scientific innovations in litigation (Birks, 1998; Freckleton and Selby, 2002; Williams, 2001). Williams, referring to work by Myers, notes that in part these difficulties reflect a lack of direction in the legal literature concerning how to define and use novel psychological evidence.

Williams (2001: 278) points out that the Supreme Courts in Canada and the United States recently addressed a critical question: What are the principles that trial judges should use to determine the admissibility of expert evidence? He argues that the concept of alienation:

... faces considerable difficulty if examined critically with respect to principles of admissibility. Courts have been less than vigilant in exercising their "gatekeeper" role. The admissibility of Parental Alienation Syndrome and/or Parental Alienation should not be benignly taken for granted.

Warshak (2000a) suggests these terms tend to be applied indiscriminately and regardless of the reasons for difficulties. PAS, he believes, is the "effect" of parent behaviour (Warshak, 2002).

Hayward (1999) comments that alienation allegations can be an effective legal device for fathers. Most authorities express concern about how the concept of alienation is used in litigation due to the limited research in this area (Johnston, in press) and its inherent methodological constraints (Zirogiannis, 2001). Johnston (2001) considers PAS a "fashionable legal strategy." Others think it is introduced to deflect attention from dangerous behaviour such as domestic violence (Bruch, 2001; Smith and Coukos, 1997). On the other hand, Warshak (2000a) contends that labels such as PAS direct our attention to ensuring that strategies for alleviating the problem will be considered. Johnston (2001) suggests that the label PAS creates gender politics and pits fathers against mothers. She argues that the media exploits this tension.

Bruch (2001: 537) conducted an electronic search for reported cases in the United States between 1985 and February 2001 that utilized the term "parental alienation," and reported that:

... numerous mental health professionals in addition to Gardner who [sic] have testified that PAS was present, although far fewer were willing to recommend that custody be transferred and contact with the primary custodian be terminated...The degree to which PAS has been invoked by expert witnesses, attorneys, or judges in these cases and the almost total absence of inquiries into its scientific validity is profoundly disturbing.

Mahony (2001) addresses Bruch's concern when he reminds us of the importance of focussing parenting plan discussions on children.

2.6.4 Remedies

Gardner's formulation is reflected in the remedies he recommends for mild, moderate or severe alienation (refer to Section 2.5.1). Darnall (1998), working with Gardner's formulation, proposes other remedies, including compensating time when there is interference in the child's schedule, contempt findings and incarceration for the alienating parent. He supports Gardner's proposal that in the most severe cases, decision-making authority should be shifted to the other parent. Mason (1999) expresses concern about how the child would interpret such action. Concern is also raised about Gardner's preference for a single therapist as opposed to a team approach (Etemad, 1997).

Many researchers and clinicians reject Gardner's remedies and label them radical, punitive and limiting the potential for a meaningful relationship (Bruch, 2001; Freckleton and Selby, 2002; Johnston, 2001; Murray, 1999). Wall et al. (2002: 90) note that testimony received from the Children and Family Court Advisory Service during a United Kingdom consultation process suggests that when intervention fails to bring about change in these situations "attempting to facilitate some form of indirect contact is more appropriate than resorting to fines or imprisonment." Johnston (2001: 15) argues that these types of coercive court interventions:

... like fines, imprisonment, change of custody, and enforced visitation-have a dim prognosis for transforming family relationships. In fact they can serve to entrench the family disputes and embitter children and youth into long-standing resistance and contempt for the legal system and its associated professionals.

Wall et al. (2002: 97) took a strong position on this issue: "… fines and committal are not only crude methods of enforcement; they are wholly inadequate as a means of addressing the problem".

Warshak (2000a) notes that both the Gardner and the Kelly and Johnston models are based on clinical experience, find support in the literature, but lack empirical research. The primary difference, he suggests, is in the proposed interventions. Bruch (2001: 543) agrees that their work represents improved science, but she believes they "… go beyond their data as they craft recommendations for extended, coercive, highly intrusive judicial interventions". She raises several objections. First, she queries the implicit assumption that all serious interpersonal difficulties can be remedied by a mental health intervention. The Kelly and Johnston proposal, Bruch argues, treats post-divorce parenting more intrusively and in her opinion leads to the growth of a "divorce industry" for professionals.

Bruch's concern reflects Wallerstein et al.'s (2000) suggestion that because alignments are transitory, what they term "over zealous" interventions are inappropriate. Bruch asserts that Kelly and Johnston have not fully considered all the assumptions involved in the roles taken by courts and mental health professionals when there are parenting plan disputes or contact difficulties. She raises other important questions about the Kelly and Johnston formulation and intervention recommendations, including whether: