Because many social workers are interested in and trained in family systems approaches, they may do important work toward family reunification following various kinds of separations of parents and children. The term reunification has for years involved the re-establishment through therapy of parent–child relationships after children have been in foster care because of substantiated abuse or neglect by parents (Miller, Fisher, Fetrow, & Jordan, 2006). In the last decade or two, the term reunification therapy has also come to be used for work with high-conflict divorcing families whose children show reluctance or refusal for visitation with one of the parents (VRR; see Garber, 2007).

Reunification therapies for VRR in high-conflict divorce cases are often court-ordered and involve all the challenges characteristic of treatment of reluctant clients. Generally, these treatments have used counseling methods and talk therapies. However, some authors have conceptualized severe cases of VRR as matters of “parental alienation”, a child’s or adolescent’s intransigent rejection of one parent caused by the intentional or unintentional persuasion of the other, preferred parent (see, e.g., Warshak, 2010). Proponents of the parental alienation (PA) concept recommend special forms of treatment, parental alienation treatments (PATs), that purport to change an alienated child’s views of and behavior toward a rejected parent (see, e.g., Childress, 2015). These treatments, which are generally proprietary and court-ordered, separate children from the preferred parent for some period of time and enforce contact with the rejected parent.

Serious concerns about PATs as they are used today have been raised by the reports of young adults who have experienced these treatments. Journalists’ descriptions of children’s reported PAT experiences have stressed the unpleasantness of the treatment, including not only the time spent in the treatment facility but also the time spent with youth transport service workers (Tabachnick, 2017). A.R., a young Washington State woman, has described to the present author and in a television interview (Nguyen, Witte, Rutanashoodeh, Villareal, & Horn, 2018) her experiences when she (then 17) and her 14-year-old sister were ordered by a family court to be transported to California for treatment by the Family Bridges PAT program (to be discussed later in this paper); the girls wanted to live with their father, whom the mother accused of causing PA. A.R. and her sister were taken in handcuffs from the courtroom by youth transport service workers and taken to an airport, where A.R. attempted to stop the process by telling TSA staff that they were being kidnapped. The transport service workers presented the court order, but eventually were told they could not fly from that airport and instead drove the girls to California. They stayed in a hotel for four days, with two psychologists working for the Family Bridges program, their mother, and their mother’s partner. A.R. experienced the program as both naïve, in that she was told to watch video material of the kind she would expect in an introductory psychology class, and threatening, in that she was told that if she did not cooperate she would be sent to a wilderness camp or a residential treatment center where she would not be able to contact friends or family. She was also told that her father would go to prison if she did not participate in the program as instructed. A.R. eventually returned home and successfully petitioned for emancipation, but the younger sister now lives with her mother and mother’s partner in another state and has not seen her father or A.R. for years. Other young adults interviewed by Nguyen et al. (2018) told similar stories.

PA and PATs were introduced to the social work community a decade ago in an article in Social Work Today (Baker, 2008). Along with psychologists, counselors, and other mental health professionals, clinical social workers have for some years been involved with PATs. They may participate in court-ordered or voluntary attempts at treatment for posited PA, sometimes working with families prior to PAT use, or providing aftercare for some time following the PAT events. Clinical social workers may also provide therapy as ordered for preferred parents who are alleged to have created PA in their children. In a recent California child custody case (Sahar vs. Sahar, Yolo County, CA, 2019), social workers provided supportive therapy to children who refused contact with their father; the father demanded that the children be isolated from their mother and placed in a PAT program.

The present paper will discuss a number of issues related to PA and PATs. Social workers should be aware that the PA concept is considered problematic by many mental health professionals and has been questioned in terms of both plausibility and evidentiary foundation (Mercer, 2019); reasons for this skepticism will be discussed, as both the PA diagnosis and the possible risks and benefits of PATs should be understood before social workers participate in therapies targeting PA allegations. Alternative hypotheses about VRR and appropriate new directions for research will also be considered.

Method

I examined 181 documents dealing with PA following a search of “parental alienation” on Academic Search Complete and on Google Scholar (which yielded 294 documents) and a hand search on bibliographical material which yielded 12 more. Publications from before the year 2000 were removed, as were copies, non-English-language documents, handbooks and popular books of advice. All but one of these were published in professional journals, the exception being a book published by a small local press (Childress, 2015) whose author has been very active in promulgating his view of PA but has never published his work in peer-reviewed form; this author’s elaborate attachment-based theory of PA seemed important to consider in light of the current popularity of the attachment construct. Because not all PATs have been clearly described in published form, some Internet descriptions in the form of blogs or websites were also examined after a further Google search.

Of the 181 published documents considered, 39 contained reports of systematic empirical findings of any sort, not including clinical reports. These documents were used to evaluate the effectiveness of PATs and the plausibility of PA diagnosis and related constructs.

In addition to the published documents, I had available five statements by young adults who as minors had experienced PATs. I also had access to a child custody trial transcript of expert testimony by the PA proponent Childress (2018b).These materials were used in evaluation of the safety of PATs and their potential for harm to children and adolescents.

Diagnostic Approaches and Identification of PA

Description

PA is not in the usual sense a diagnosis, as it does not appear in any formal diagnostic systems. PA is thus not a conventional diagnostic category, although proponents of the idea occasionally present it as if it were. A campaign by advocates of the PA concept (Bernet, 2008) was mounted at the time of the development of the 5th edition of the Diagnostic and statistical manual of mental disorders (DSM-5; American Psychiatric Association, 2013), with the intention of having PA included with its own diagnostic code in that edition, but the campaign was unsuccessful. The upcoming edition of the International Classification of Diseases (ICD-11, to be published in 2023), will include PA as an index term and will link it to the diagnosis caregiver-child relationship problem, but will not include PA as a specific diagnosis. However, proponents of the PA concept have described what they consider criteria that allow identification of children with this problem.

Baker’s 2008 paper, mentioned earlier, repeated the views of the psychiatrist Richard Gardner (2002), whose work was foundational to thinking about PA. Gardner proposed that child or adolescent VRR should be identified as PA, and thus the result of the preferred parent’s actions, when certain criteria were met. These were “1. A campaign of denigration [by the preferred parent, and the following items on the part of the child– JM]] 2. Weak, absurd, or frivolous rationalizations for the deprecation 3. Lack of ambivalence 4. The ‘independent-thinker’ phenomenon 5. Reflexive support of the alienating parent in the parental conflict 6. Absence of guilt over cruelty to and/or exploitation of the alienated parent 7. The presence of borrowed scenarios 8.Spread of the animosity to the friends and/or extended family of the alienated parent” (Gardner, 2002, p. 97). The meaning of most of these criteria is self-evident; the “independent-thinker phenomenon” refers to the child’s or adolescent’s statement that his or her attitude toward the non-preferred parent is self-determined rather than influenced by others, and “borrowed scenarios” are narratives that the diagnostician believes have been influenced by the preferred parent. (It is notable that the first criterion item, a campaign of denigration, is usually assumed rather than demonstrated.) Like later authors writing about PA, Gardner employed specialized terminology reflecting his assumptions about the causes of VRR, speaking of the non-preferred parent as the alienated or targeted parent and the preferred parent as the alienating parent or alienator. (These terms are still commonly used, but one recent author has referred to the preferred parent as the pathogenic parent (Childress, 2015).)

Gardner (2002) also proposed that children showing severe VRR were displaying emotional disturbances caused by the preferred parent’s behavior and would have long-term emotional difficulties as a result, and that therefore the preferred parent had been behaving abusively. This proposition is of great importance with respect to diagnostic methods, as the conclusion that PA exists becomes an argument that child abuse has occurred, and separation from the preferred (and therefore, it is claimed, abusive) parent should be ordered by a court as part of a protective and therapeutic approach. The burden of proof for such a significant diagnosis is on proponents of the diagnosis (Emery, 2005), especially because of the potential stigma of a mental health diagnosis (Walker & Shapiro, 2010) and the possible consequences of separation and treatment for all family members.

Identification

It is notable that although PA identification should presumably begin with evidence of VRR, this is not always the case in practice. In one case known to the present author, a father has alleged PA even though the child seeks contact with him, claiming that if he allows visitation he will be accused of sexual abuse. In another case, a girl who had for many years alternated weeks at father’s and mother’s homes proposed to spend more time at her mother’s house in order to “have a home”; she did not refuse contact with her father, but he alleged PA. Such allegations become part of family court discussion and potentially lead to PAT orders even though the cases may not involve refusal of contact by the child.

Gardner (1999, 2003) offered a differential diagnosis chart purporting to discriminate between PA and other problematic child attitudes and behaviors. Bernet (2008) has claimed that the evidence exists for a Parental Alienation Disorder (PAD) distinct from other relationship problems, but Bernet, von Boch-Galhau, Baker, and Morrison (2010) also argued that PA could exist in the absence of an alienating parent, implying a departure from Gardner’s view. The difference between Gardner’s criteria and this statement by Bernet et al. is more than a technicality, for, as will be discussed later, the assumption of the alienating parent is at the heart of the demands for custody change related to various Parental Alienation Treatments (PATs). However, as none of the attempts to identify cases of PA have been based on demonstrably reliable and valid methods, this point does not help us identify any diagnostic approach to PA as well-founded.

Childress (2015, 2018b) has argued that some VRR cases may be defined as “attachment-based parental alienation”. Children and adolescents whom Childress diagnoses in this way are described as grandiose, haughty, narcissistic, and suffering from encapsulated delusions about their victimization by the non-preferred parent, as well as being involved in an inverted family hierarchy in which they believe that they may judge their parent.

Warshak (2010) and others have argued that children identified with PA also suffer from deficits of critical thinking such as “black and white” thinking and a lack of ambivalence. However, it is not clear how such cognitive problems might be evaluated, and published material does not indicate that there is attention to such issues when PA is identified.

Diagnosis of the Preferred Parent

Although a few DSM-5 diagnoses include questions about etiology of a problem among their criteria, some PA proponents take the unusual further step of assuming that a child’s behavior and attitudes are keys to diagnosis of mental health problems in the preferred (“alienating”) parent. For example, Childress (2015) states that in the absence of substantiated abuse, VRR indicates the presence of “pathogenic parenting” by the preferred parent. Childress argues that preferred parents in these cases have narcissistic or borderline personality disorders, and that such disorders in adulthood result from earlier disorganized attachment, a claim not supported by current research in this area (Granqvist et al., 2017). The diagnosis of the parent in these cases is based on the child’s attitude and behavior, not on psychological testing or on observations of parent behavior or history. Childress (2018b) further states that VRR without evidence of physical or sexual abuse is prima facie evidence of psychological abuse by the preferred parent and should lead to a period of protective separation from the preferred parent of 9 months or more.

Differences Between Evaluation of PA and of Other Relational Problems

Any assessment of a relational problem is difficult and complex because it must deal with the separate characteristics of at least two people as well as with the interactions between or among them. Some such problems, like Parent Child Relational Problem (PCRP), have been listed in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5; American Psychiatric Association, 2013), but this listing has followed field trials of the evaluation (Wamboldt, Cordaro, & Clarke 2015). PCRP is described in DSM-5 as a “V-code”, indicating that it is not considered a mental disorder. Claims about PA assessment have not been supported by field trials so far, and such claims argue that PA is an indication of a mental disorder in the preferred parent and at least the potential for a mental disorder in the child or adolescent.

Treatment

Parental Alienation Treatments (PATs, as opposed to more general reunification therapies) are based on the idea that children’s VRR with respect to non-preferred parent has resulted from the influence of the preferred parent. Removing the child from the influence of the preferred parent, and increasing the influence of the non-preferred parent, thus become foundational actions in treatment of PA. These steps are unlikely to be voluntary on the part of a child or a preferred parent, but instead depend on court-ordered custody changes and subsequent treatment as proposed to courts by PA advocates. An important aspect of PATs is their shared claim that they are not psychotherapies (see, e.g., Warshak, 2018; Warshak & Otis, 2010; Childress, 2018a). It is argued that these treatments are in fact psychoeducation rather than therapy, although the treatments as described do not meet the definition of psychoeducation as teaching “individuals and families affected by psychiatric disorders about the disorder’s symptoms, course, prognosis, management, and treatment” (Fristad, Ackerman, & Nick, 2018, p. 253). The practical result of this claim is that PATs need not be conducted by licensed mental health professionals, as educational programs may be presented by persons with other types of training.

Mercer (2019) described and evaluated five PATs with some shared features. Proponents of these treatments, described in the next section, made various poorly-substantiated claims about their effectiveness.

Family Bridges

Family Bridges (Warshak 2010, 2018) is probably the best-known of these PATs and has been the subject of a number of publications. This program involves a 4 or 5 day “workshop” attended by the non-preferred parent and by the children, who may be brought to the premises by youth transportation workers following court orders, and who have cellphones and money taken away from them. Children are told they may leave at any time, but given their situation in an unfamiliar place, this is not practically possible. The preferred parent does not attend. Videos on cognitive and perceptual biases are presented, and communication and role-playing practices occur. Goals are improved critical thinking for the child and prevention of VRR for the future. Children and the non-preferred parent are expected to take a vacation together following the program. As is the case for other court-ordered PATs, the cost of the program is to be borne by the preferred parent, and appears to be around $20,000. The preferred parent is also to seek “aftercare” in the form of therapy by a practitioner trained in the Family Bridges approach.

High Road

A second PAT, which unlike Family Bridges has not at the time of this writing been discussed in any peer-reviewed journal, is High Road to Family Reunification (Childress, 2015). High Road, a program conceived and implemented by the life coach Dorcy Pruter, closely resembles Family Bridges. Children usually attend the program because of a court order and may be brought by youth transport service workers. The non-preferred parent attends the program but the preferred parent does not. Again, a “workshop” of several days employs videos, games, role-playing, and so on, as well as informing the children that they will be separated from the preferred parent for months, and making any contact with the preferred parent contingent on their behavior and attitudes being approved by the non-preferred parent.

Overcoming Barriers

A third PAT program, Overcoming Barriers Family Camp (Sullivan, Ward, & Deutsch, 2010) is a treatment that uses psychoeducation and milieu therapy for 5 days and 4 nights, generally at court order. This treatment is unusual among PATs in that all family members, and both preferred and non-preferred parents, are to attend. The non-preferred parent may watch the child, participate in activities with the child, and take part in family meetings. Preferred and non-preferred parents meet in co-parenting groups, where they may be challenged by other parents on inappropriate parenting behaviors or beliefs.

Family Reflections

A camp structure is also used by a fourth PAT, Family Reflections Reunification Program (Reay, 2015). Children initially attend with siblings but without either parent; the program assumes critical thinking problems in the children, and this may be the time in which psychoeducational methods are directed toward improving critical thinking. The non-preferred parent is later introduced and participates with the child in outdoor experiential programs, eventually staying with the child in a large shared room. The preferred parent does not attend the program. Both parents are expected to continue counseling with a Family Reflections-trained practitioner.

Transitioning Families

Finally, the Transitioning Families Reunification Model (Judge et al., 2015) is a PAT that has received little attention. An associated program, Stable Paths (Isicoff, 2015), has been described as using experiential methods like horseback riding, swimming with dolphins, and cooking family meals, as ways to reunify children with non-preferred parents. Court orders preventing contact with the preferred parent are an aspect of this program.

Commonalities of PATs

As described by young adults who have experienced PATs and by PAT proponents, PATs are simple programs with shared features that differ from most psychotherapies. They emphasize the importance of separation of the child from the preferred parent, with severely limited contact even by phone. They create conditions for increased, sometimes almost continual contact with the non-preferred parent. They often stress the idea that the child avoids a parent because of problems with critical thinking and with understanding how the preferred parent has caused the rejecting attitude, and show videos or provide lessons that are intended to improve the child’s thinking skills. PATs generally include the common-sense practice of encouraging pleasurable experiences like holidays, to be shared by the rejecting child and the non-preferred parent with the goal of reducing the child’s avoidance due to anxiety or anger. However, negative experiences like isolation of the child from other familiar people and direct or implied threats of the consequences of non-cooperation are often parts of PAT programs.

Evidence Supporting PA Diagnosis and PATs

Mental health practitioners today are encouraged to seek out well-researched diagnoses and therapies that can be designated as evidence-based treatments (EBTs; alternately, empirically-supported treatments, ESTs). Choosing treatments with the highest possible level of evidentiary support, and making that choice while considering other factors such as client preference and practitioner wisdom, has been called evidence-based practice (EBP). Unfortunately, it is not always very clear whether a treatment is or is not empirically-supported or at what level of evidence (as defined, for example, by the California Evidence Based Clearinghouse for Child Welfare, www.cebc4cw.org) that empirical support exists. Claims made by proponents of a diagnosis or treatment are not necessarily based on the highest levels of evidence. Although randomized controlled trials (RCTs) may be quite difficult to carry out, so that it could be unreasonable to expect that high level of evidence, we should demand that acceptable evidence include a design with at least some form of comparison group rather than a simple before and after measure taken from a single treatment group.

Criteria for Support of PA Treatments

The problems about PA diagnosis described earlier in this paper raise some initial questions about evidence for PA treatments. If a mental health or behavioral issue cannot be reliably and validly identified, it is not possible to determine whether or not a treatment is effective. Studies of PATs, notable Family Bridges, have emphasized the satisfaction of the non-preferred parent as the primary criterion for successful treatment of children who have shown VRR.

Evidence for Effectiveness of PATs

Although proponents of PATs claim evidentiary support for their methods, these claims are not in line with levels of available evidence, and supportive arguments may or may not be relevant. For example, Warshak (2010, 2018) has claimed that Family Bridges is evidence-based. However, the supportive research that has been offered is at a modest level, involving before-and-after studies rather than the randomized controlled studies or the nonrandomized controlled clinical studies usually considered necessary for evidentiary support (see www.cebc4cw.org for an example of analysis of levels of evidence). Warshak and Otis (2010) have also argued that the instructional methods used in Family Bridges are evidence-based, but this is not relevant to questions about the effectiveness or safety of the treatment. Mercer (2019) evaluated the research evidence in published studies of five PATs and concluded that the research evidence is not adequate to support claims of effectiveness for these methods. It is notable that none of the PATs discussed here is listed in any one of three online services that provide information on the evidence basis of psychotherapies for children (National Registry of Evidence-based Practices and Programs [https://www.samhsa.gov/nrepp; recently criticized for its weaknesses]; California Evidence Based Clearinghouse for Child Welfare [www.cebc4cw.org]; Effective Child Therapy [a service of APA Division 53; www.effectivechildtherapy.org]).

Concerns About Risks of PATs

Although treatments for mental health or behavioral problems are intended to benefit clients, over the last decade increasing concerns have been voiced about potential harms associated with some treatments. For example, conversion therapy as an effort to change same-sex orientation has been pointed to as a source of depressive and even suicidal reactions among clients (American Psychological Association, 2000). Some authors have called attention to the need for identification of potentially harmful treatments (PHTs) for both adults and children (Lilienfeld, 2007; Linden, 2013; Mercer, 2017). PATs need to be examined for possible identification as PHTs.

Potential Harms to Children

A special concern about PATs, and the PA concept as a whole, has been suggested by authors who note the connections between domestic violence and other forms of abuse, and later allegations of PA when children avoid the non-preferred parent (Dallam & Silberg, 2016; Meier, 2014). Women living in shelters have been reported to be especially likely to be accused of alienating children from the non-preferred parent (Saunders & Oglesby, 2016). Successful argument of the non-preferred parent that PA has occurred, followed by a court-ordered change in custody from the preferred to the non-preferred parent, thus has the potential for placing custody in the hands of an abusive parent, increasing the risk of harm to a child.

Linden (2013) introduced the concept of an “emotional burden” of treatment, an idea highly relevant to children’s experience of PATs. Linden suggested that unless there is no other possible effective treatment for a mental health problem, treatments that are distressing to clients (for example, make them cry) should be avoided. Descriptions of PATs as experienced by children suggest strongly that at least some find a disturbing level of emotional burden associated with PATs. As Dimidjian and Hollon (2010) have pointed out, although anecdotal material is not acceptable support for the effectiveness of a therapy, anecdotes about harms experienced by clients can be the first line of defense against unwitting continuing use of potentially harmful treatments.

Developmental Issues in PA

People who have come forward to report their experiences with PATs are all young adults, and most of them went through PAT during their middle to late teens. PA is most often alleged when children of 8 or more exhibit VRR. From a developmental point of view we would expect potential harms of PAT to these children to be associated with their ages and stages of development, as well as to the emotional burdens of treatment. Children of 8 and older are sad and depressed when separated from a preferred parent and that parent’s home, but we would not expect them to develop attachment disturbances as a result. Their social relationships are increasingly oriented toward peer relationships, the school, and the community rather than toward parent–child relationships. Changes in living situation are likely to demand a good deal of adjustment if they involve new schools, friends, and communities, and these adjustments can be challenging to children who are already stressed by parental conflict, sometimes resulting in academic difficulties (in the Sahar vs Sahar case discussed earlier, teenage children ordered to live with the non-preferred father had many school absences and did poor work, but on returning to the mother’s household returned to high levels of academic achievement). For older preteens and teenagers, court-ordered custody changes are an impediment to the development of autonomy characteristic of this developmental stage (see Erikson, 1950; Havighurst, 1972).

It is unusual to see the PA concept applied to infants or preschool children, who might indeed feel the impact of attachment disturbances as the result of a court-ordered abrupt change of custody. The present author is aware of two cases of allegations involving young children. In one, a Canadian case, children aged 2 and 4 years were ordered into their father’s custody, with a no contact order against the mother and grandmother; 10 years later, the children have still not seen the mother, who has been ordered under penalty of law not to appeal. In a recent case in the United States known to the present author, children 5 years and 1 year of age, who cried with distress when in supervised visitation with their father, were the subjects of PA allegations against the mother, who was living with them in a shelter. As these young children are unable to report their experiences, any understanding of harm done to them will have to await systematic research.

Anecdotal Evidence of Harm to Children

Anecdotal reports are generally considered inadequate evidence for claims of effective treatment, so these anecdotal reports of harm should also be questioned. It is quite possible that only a few children find PATs distressing, or that unpleasant experiences and emotional burdens are confined to certain methods or certain practitioners. Nevertheless, in the absence of systematic investigation of PATs, their risks and benefits, it seems appropriate to pay attention to the role of anecdotes as the first line of defense (Dimidjian & Hollon, 2010) against potentially harmful treatments. This cautious posture does not assume that all PATs are harmful to all children who experience them, but does remind us that not all interventions are therapeutic and that safety as well as effectiveness needs to be established before a treatment is assumed to be beneficial.

Potential Harms to Other Family Members

Social workers and other mental health professionals who work with children and families often give priority to children’s best interests when considering treatment or custody choices. However, as a family systems perspective emphasizes, the impact of a situation on one family member can affect all members of the group, including new partners and stepchildren of divorced parents.

There has been no systematic investigation of the effects of PAT on siblings or stepsiblings of children sent for PAT, but there is potential for disturbing effects on the other children, who may understand the events to mean that their own wishes and feelings may be considered of little importance. When one child is identified as needing PAT, others in the family may feel they have received permission to scapegoat the “problem child”. Siblings may be separated if one is identified as resistant to PAT, and until both are emancipated, they may have no contact, thus robbing them of a stable sibling relationship that could be helpful and comforting as the adjustment to divorce occurs over time.

The preferred parent, who is accused of causing PA in the child, is exposed to potential harms of emotional, legal, and financial natures. Formal accusations of psychological child abuse (as defined by PA proponents) are threats to identity and reputation, the latter potentially serious when the accused parent is employed in teaching, child care, or mental health services. Legally, the preferred parent risks imprisonment if found in contempt of court for lack of cooperation with court-ordered PAT, including reluctance to attend counseling by a PAT-oriented mental health provider. Court orders for payment of PAT fees by the preferred parent may be financially devastating.

As PAT interventions are not only potentially harmful to children, but have yet to be supported by strong research evidence, we should also consider their negative impact on the non-preferred parent who has chosen to employ a PAT professional. The best case scenario here is that the non-preferred parent simply undergoes opportunity costs, as resources that could have gone to more effective treatment are spent on PAT. However, at least some children who have been through PAT are extremely angry and resentful that the non-preferred parent and the court have forced them to have disturbing experiences, diminishing the chances that child and non-preferred parent will be able to work out a positive later relationship. In cases where domestic violence, poor parenting skills, or mental health problems on the part of the non-preferred parent have caused a child to avoid that parent, failure to acknowledge and treat these issues may confirm the parent in the belief that his or her behavior has been acceptable, making it likely that problem behavior will recur, to the detriment of the non-preferred parent as well as of the child and the preferred parent.

Potential Harms of Failure to Treat

In the interests of fairness, however, we should consider the arguments made by PA advocates who express concern about outcomes for children who display VRR but are allowed to make their own decisions about custody arrangements rather than entering PAT programs. Some PAT advocates argue that children are in fact harmed when their opinions are given consideration (Croezen, Ander, & Jones, 2018). Childress (2018b) stated in his testimony in the Sahar vs Sahar case that for a child to “judge” and criticize a parent is indicative of an “inverted hierarchy” in the family, even when children attribute their positions to specific actions by the non-preferred parent; he described attention to the children’s views as “appeasement”. Warshak and Otis (2010) have emphasized the need for two parents to guide developing children, although PAT research has examined only the child’s relationship with the non-preferred parent rather than looking at relationships with both parents. A posited association between PA and narcissistic personality disorders, said to be present in the preferred parent and developing in the child as a result of PA, has been argued by Childress (2015, 2018b) as a danger when PATs are not used. PA proponents thus consider potential harms of forgoing PATs as at least equivalent to potential harms of the treatments.

Plausibility of PA Concepts

Real understanding of the potential risks and benefits of PA principles and practices must wait for further systematic investigation. However, it is likely that if a concept and associated treatments are implausible, they will be ineffective, and their use may involve risks ranging from minor to serious. Plausibility of concepts of mental health treatments for children involves three factors: (1) the internal logic by which conclusions are drawn and cause and effect relationships are determined, (2) the congruence of theory and practice with established facts about child development and changes in attitude and behavior, and (3) the nature of the treatment itself.

It is by no means implausible that some children avoid and reject one parent as a result of the manipulation and social pressure exerted by the other parent. In this sense, PA no doubt exists. The issue of plausibility arises, however, when a child’s VRR is explainable by mechanisms other than the influence of the preferred parent posited by PA proponents, but PA is claimed as the sole explanation. Plausibility concerns also appear when children’s experience of domestic violence is dismissed as a possible cause of VRR, as was the case in Sahar vs Sahar, mentioned earlier in this paper. PA advocates have implausibly labelled children’s reasons for avoidance of a parent as “frivolous rationalizations” if there is no substantiated sexual or physical abuse, although from the children’s perspective there may be many concerns leading to avoidance, from worries about menstrual hygiene in an unfamiliar home to problems with stepparents or stepsiblings to psychological abuse by the non-preferred parent. These latter situations involve implausible applications of PA concepts. It is not necessary to declare PA concepts impossible in order to question the plausibility of these concepts as often applied.

Internal Logic

It is possible for a preferred parent’s manipulation and persuasion to influence a child to avoid a non-preferred parent; however, it is an error of logic to assume that when a child avoids a non-preferred parent, the preferred parent must have persuaded the child to do so. This fallacious assumption is a common error of critical reasoning called affirming the consequent, in which it is wrongly thought that if B follows A, B must always have been preceded by A, even though B might also (or instead) have been preceded by C, D, etc. Affirming the consequent also causes confusion when it is assumed that when A precedes B, it, and it alone, also causes B. In the case of PA, there might well be correlations between the attitudes of a child and a preferred parent toward the non-preferred parent, but both sets of attitudes might be caused by characteristics of the non-preferred parent or by situational factors, rather than the preferred parent’s attitudes causing the child’s attitudes.

PA proponents (Gardner, 2002) have also used arguments that are logically paradoxical in their discussion of child psychological abuse. Recommendations for custody change from the preferred to the non-preferred parent are sometimes based on the idea that the posited influence of the preferred parent causes mental illness in the child (as revealed by VRR) and therefore is psychological child abuse. When non-preferred parents insult a child (“you’re crazy just like your mother”) or threaten to send a child to residential treatment or boarding school when they get custody, as has occurred in some cases known to the present author, PA proponents see these actions not as psychological abuse but as the proper conduct of a parent for preservation of the family hierarchy (Childress, 2018b).

Congruence with Developmental Processes

One version of PA (Childress, 2015, 2018a, b) has attempted to connect VRR with damage to the child’s attachment to the non-preferred parent, said to be caused by actions of the preferred parent. However, except in rare cases, PA diagnoses and PATs are directed at children older than age 8, with children in their early to middle teens the most likely candidates. Older teenagers and preteens normally retain affectionate connections with their parents, but their attachment attitudes and behavior are very different from those of the toddlers whose behavior was the impetus for attachment theory. By the teens, children are likely to display a wish for proximity to parents only when seriously distressed or threatened, and are able to explore the world freely without the support of an attachment figure. They have developed internal working models of social relations (Bowlby, 1982) and use these to establish effective relationships with peers and others outside the family; peer relationships begin to take on great importance, especially when there are romantic aspects to the relationship. This developmental change creates a reduction in the wish or need to stay close to parents. Parent–child relationships are normally altered through transactional processes (Sameroff, 2000) during these years, as parents respond to children’s changing needs and behavior by altering their own approach to the maturing child, and the child’s further developmental change is in turn shaped in part by parent attitudes.

Expected developmental trajectories for teenagers would involve a great increase over concerns with autonomy, rather than attachment (Erikson, 1950), as well as with peer relationships. The PA concepts that emphasize parental authority as an essential factor in development are thus at odds with expectable development, at least among WEIRD (Western, Educated, Industrialized, Rich, Democratic; Baron, 2014) populations. PAT pressure for children to comply with, empathize with, and be affectionate toward the non-preferred parent is likely to be met with active or passive resistance because of the powerful developmental forces working toward “launching” from the family of origin. That such experiences could have benign effects on the children most likely to be treated is implausible; in fact we might compare the possible results to the effect of trying to prevent a toddler from walking when maturation decrees that she should.

Psychoeducation Versus Psychotherapy

Although PATs have been presented as involving psychoeducation rather than psychotherapy, this reframing is not plausible. PATs are intended to bring about dramatic changes in behavior and attitudes in children who have strongly and intransigently resisted contact with a non-preferred parent, and who are angry with that parent, frightened of the parent, or both. Making serious changes in clients’ emotions, attitudes, and behavior is the task of psychotherapy, whereas psychoeducation is aimed at providing information that will allow clients and their families better to navigate physical and mental health interventions. However, psychoeducation may be provided as part of a “package” that includes supportive psychotherapy (Flatt & King, 2010).

The PATs described earlier in this paper may be as brief as 4 days of intensive work, or may be a matter of a couple of weeks of a camp experience. They are followed by months or years of aftercare by practitioners trained in a PAT, but whose methods are not described in the literature. Months or years of continued separation from the preferred parent, siblings, or other family members may also be part of the treatment package. It is implausible that these intensive and extensive treatments can be considered primarily psychoeducational.

Conclusion

As has been stated earlier in this paper, this article is not intended to argue that PA, as described by its opponents, cannot and does not exist. A problem may be vanishingly rare but very real; the prevalence of PA has not been discussed even by the strongest supporters of the concept. The issue is not about the possible existence of PA, but about its assessment and treatment as proposed by Baker (2008) and later authors.

It appears that PATs have the potential to harm children and their families, and that no adequate evidence has been reported to support claims that these treatments are beneficial. In addition, some of the concepts that are foundational to PA diagnosis and PATs are implausible. These points run counter to the claims of PA proponents that children who show VRR will be harmed if not given a PAT. Because of this risk–benefit situation, social workers involved with high-conflict families should be very cautious about accepting PA claims made by family members or by mental health professionals. They should remain aware that although it is possible that a child’s avoidance of a non-preferred parent is caused by actions of the preferred parent, such a connection cannot be demonstrated simply by observing VRR. It is important to keep in mind that children’s reports of abuse may be accurate even if unsubstantiated by investigation, and that psychological abuse or poor parenting skills on the part of the non-preferred parent may have a powerful effect on a child’s attitude. Court-ordered custody changes in response to VRR may thus put children in the hands of abusers (Dallam & Silberg, 2016; In the Court of Appeals of the State of Washington, 2017 [Develle v Poppleton]) with potential impacts ranging from academic failure to suicide.

Implications for Practitioners

Social workers and other mental health and family support professionals need to be aware of the difficulties surrounding claims of PA and recommendations for PATs. As PA claims and concepts enter public discussion, references to “parental alienation” are used more and more freely. Practitioners need to be aware that a divorced parent who uses the term PA may well not be referring either to any type of assessment or even to a child’s actual rejection of contact, but may simply want to attribute any unwanted child behavior or attitude to actions of the former spouse. It is important to inquire closely into the child’s actual level of rejection and to ascertain whether the parent’s concern is really with child obedience, affection, or even gratitude rather than avoidance of contact.

Practitioners also need to remain aware that in family life, as elsewhere, multiple effects are often the consequences of multiple causes. That a preferred parent’s actions are the sole cause of a child’s attitude and behavior is conceivable, but would be atypical when considered in the context of all developmental processes. Although a preferred parent may manipulate, exploit, even bribe or threaten a child into rejecting the other parent, a non-preferred parent may affect the child’s attitude through poor parenting or general social skills, may be emotionally, physically, or sexually abusive, and may handle anger inappropriately. Children may also prefer one home to another for situational reasons including proximity to school, friends, sports and activities. When parents have new partners, those people can have profound effects on parent–child relationships, especially if the new partner has personal concerns about PA or wishes to interfere in the co-parenting relationship. Without exploring all these avenues, practitioners risk missing important points that could help them work fruitfully with a family.

The many practice issues associated with PA concerns are multiplied enormously when we take into account the legal issues surrounding PA. A practitioner who is too readily accepting of PA claims made by a parent, another practitioner, or a lawyer, risks exposing a family to draconian treatment measures sometimes ordered by family courts. These measures include abrupt and complete separation from the child’s home and the parent with whom he or she is most comfortable, as well as the loss of family resources that should instead be expended in the child’s best interests.

Research Issues

On the basis of the available evidence, it is reasonable to reject indiscriminate use of the PA diagnosis and question any use of PATs. However, rejection of one approach to VRR does not solve the problem of how to help high-conflict families function better in cases when children avoid one parent. Trite though it may be to advise further research, it is clear that we do not presently have much insight into children’s avoidance in cases where neglect or abuse do not seem to have occurred. Social work researchers and other mental health professionals may help solve the VRR/PA puzzle by examining both general and specific factors other than actions of a preferred parent that can contribute to children’s reluctance or refusal.

General Research Questions

Culture

There are few, if any, investigations of the role of cultural differences in development or handling of VRR. Warshak (2015) expressed concern over allowing children to call a parent by his or her first name and saw this as a factor in PA, but this practice can result from cultural as well as familial or individual differences. Is VRR primarily a problem of WEIRD (Western, Educated, Industrialized, Rich, and Democratic) populations, whose children may have grown up believing that they have rights and can make choices? Or does it involve a clash between the “children’s rights” view, held by children and possibly preferred parents, and authoritarian attitudes on the part of a parent whose religious or ethnic identity stresses parental control over children until at least early adulthood? Generally speaking, such a clash would involve an emphasis on fathers’ rights and authority, although in some cases fathers are the target of allegations that they have acted to alienate their children from the mother.

Individual differences

Social work researchers might also pursue increased understanding of characteristics of non-preferred parents, of preferred parents, and of children who avoid a non-preferred parent. Non-preferred parents were described by Fidler and Bala (2010), as lacking “appropriate degrees of empathy… counter-rejecting, punitive and angry with their child, much like a knee-jerk reaction to being treated very poorly and disrespectfully by their child. The rejected parent may be easily offended and ironically react like their alienated child, with aggressive and disrespectful behavior…. Rejected parents may act in self-centered and immature ways, with little or no insight into how their own behavior is contributing and affecting the child” (2010, p. 20). Walters and Friedlander (2010) considered “intractable” rejected parents as rigid and noted that the rejected parent’s behavior, while not technically reaching the level of abuse, may be truly unpleasant, further polarizing, and unreasonable, which together make it very difficult for the child to manage”(p. 432). Johnston (2003) also suggested that the non-preferred parent’s actions could be the cause of a child’s rejection. In spite of these comments about characteristics of some non-preferred parents, however, attempts to create diagnostic measures for PA have not yet incorporated such information.

Little work has been done on the characteristics of preferred parents. Are these adults more empathic than the non-preferred parent, more cooperative or tolerant? Or more vulnerable and in need of care given by a parentalized child? If they do not have these characteristics on objective measures, do the children attribute the characteristics to their parents? What are the gender differences between preferred and non-preferred parents? Richard Gardner was accused of supporting fathers in their allegations that mothers caused children to avoid their fathers, but he responded that there were cases where mothers were the rejected parents, and the present author has encountered such cases. Still, however, we have no clear information about the frequency with which mothers and fathers are preferred or non-preferred. In the Sahar vs Sahar case, expert witness testimony by Childress (2018b), a PA proponent, stated that the preferred parent had a borderline or narcissistic personality disorder, which Childress attributed to her having earlier suffered from disorganized attachment; these statements were made without evidence from her history or present status.

Characteristics of children showing VRR have not been thoroughly described, even in terms of age and gender frequency. The age range for children brought into PATs is generally about 8 years to 17 years, with the upper limit established as 18-year-olds are legally adults and can make their own decisions about contact with parents. In a very small number of cases, PA has been alleged when preschool children have tried to avoid contact with a parent. Other characteristics of children showing VRR or given PATs have not been described on an empirical basis, although PA proponents have claimed that cognitive skills in children showing VRR are inadequate. PA proponents generally state that relationships were good between the child and the non-preferred parent in the past, but evidence for this claim has not been presented. Neither has there been discussion by PA proponents of children’s general or specific anxiety about issues other than contact with the non-preferred parent or management of anger about other issues. It would also be of interest to know how sibling groups function in these situations, for example whether the attitude of an older sibling is communicated to younger siblings. This kind of information is much needed for explanation of VRR cases, and particularly for differentiating cases where a preferred parent might indeed have an undue influence on the child’s attitude toward the non-preferred parent.

Separation and divorce may be preceded or followed by the development of new romantic partnerships by either or both of the parents. Warshak (2000) noted remarriage as a possible trigger for avoidance of the non-preferred parent. However, little attention has been paid to the characteristics of new partners or stepparents, or of stepsiblings, as they may cause a child to avoid the non-preferred parent. These new family members, who of course were not chosen by the child, may act to support good relationships between a child and both biological (or adoptive) parents, or may offer serious challenges to those relationships and may be the first to allege PA due to the actions of one of the parents. In one case known to the present author, a stepmother who had herself been the subject of VRR by her own children in another country was apparently instrumental in bringing a father to allege that his children’s mother was causing their VRR.

Consideration of the roles in VRR of child relationships with persons other than the parents raises unanswered questions about family constellation and the influence of siblings on each other. Although we might expect the attitudes of older, adolescent siblings to have the greatest effect on younger children, it is also possible that a wish to protect an apparently vulnerable younger sibling could exacerbate a teenager’s rejection of the non-preferred parent. Once again, empirical work is needed to clarify these matters.

Other Relevant Factors

The circumstances of separation or divorce may also play strong roles in the development of VRR. A history of domestic violence, even if it has not been directed toward the children, can provide a powerful motive for children to avoid a non-preferred parent; children do not themselves have to have been physically abused for this to be true. Alcohol and drugs can also be factors, whether they are used by parents, by children, or both. Financial security or insecurity, distant or near locations of parents’ homes, children’s school circumstances and ease of contact with their friends may also have impacts. The effects of any of these variables can change with the child’s age and are likely to have the greatest effects in the teenage years.

Some Specific Research Suggestions

Longitudinal and Developmental Studies

Johnston (2003) proposed that longitudinal studies were needed for understanding of the role of VRR in children’s development. Such studies are the foundation of our understanding of child development. Using repeated measures of a group of individuals over the course of development, longitudinal studies control for individual differences and allow the charting of developmental pathways. As we have established information about developmental trajectories for factors like cognitive responses to ambiguity, relationships with peers and parents, and views of the self and others. Longitudinal studies of these factors in children who have shown VRR (which have not been done) could then be compared to similar studies of individuals with more normative experiences and behaviors, showing to what extent there are long-term effects of childhood or adolescent VRR on development.

Developmental studies could also show whether VRR and related attitudes and behaviors are distributed over the child and adolescent period or whether they peak during a narrower age period and are causally connected with the developmental tasks of that period. A comparison of boys and girls showing VRR at different ages would be of interest because of girls’ “fast track” to puberty, putting them developmentally about two years ahead of boys in early adolescence. A sex difference of this type, if demonstrated, could also give insights into developmental issues that affect attitudes toward the preferred and non-preferred parents.

Other Forms of Refusal

When we examine psychological and behavioral issues that are not well-understood, one useful strategy can be to find some analogous events that are easier to study. Given that there are so many factors that can help to explain a child’s avoidance of a parent, it seems to be a good idea to find some parallel situation that may give insights into this kind of avoidance. Are there other things children avoid with intense emotion and resistance? Yes, and one of them is not uncommon: school refusal. About 35% of children sometimes refuse to go to school without a “rational” explanation– some do this infrequently, while others may manage to avoid school most days for long periods of time. Children who refuse school not infrequently fight against going to school each morning, plead stomachaches, even vomit, scream and have tantrums. The great majority of their parents do not like this, want the child to go to school, fear the consequences for their own work responsibilities, and are disturbed about the social and educational outcomes for the child. The parents don’t know what to do to get the child to go to school, and it is noticeable that they are rarely if ever accused of alienating the child from school.

Pina, Zeir, Gonzales, and Ortiz (2017) discussed children’s motivations for refusing school. They referenced children’s experience of anxiety and depression in response to bullying; aversive social or evaluative situations; a desire for more adult attention; and enjoyment of events found outside school. In each case a possible parallel with refusal to visit a parent may be drawn (by the present author; Pina et al. did not discuss this issue). In addition to this list of motivations for avoiding school (with their possible parallels in avoidance of a parent), Pina et al. noted characteristics of children who refuse school, such as poor social skills, social isolation, high levels of family conflict, and a poor sense of self-efficacy in stressful situations. The fact that some parallel characteristics may be seen in children displaying VRR further suggests that the analogy between the two sorts of refusal may provide a useful direction for research.

Parenting Style and Treatment Style

A second potentially useful analogy between child development research and the investigation of PA and PATs involves the idea of parenting styles and the possibility that these may be generalized to examination of treatment styles. The best-known approach to parenting styles remains that of Baumrind (1971), who used the categories of authoritarian, authoritative, and laissez-faire styles, categories that could also be applied to treatment methods used for children and adolescents. Baumrind reported the best developmental outcomes for children who experienced authoritative parenting with strong parental guidance but without harsh rules or demands for obedience. Descriptions of PATs by their practitioners and by young people who have reported their experiences suggest that PATs could be categorized as employing an authoritarian style instead of the authoritative style considered most desirable by Baumrind. It is possible that the demands of non-preferred parents alleging PA should also be placed in the authoritarian category, but this would of course require empirical verification.

The concept of intrusive parenting through psychological control (Barber, 1996; Bradford & Barber, 2005; Barber et al., 2012) provides another potentially fruitful analogy for examination of VRR cases and of PATs used to treat them. Intrusive parenting involves attempts to gain psychological control of adolescents’ thoughts, feelings, and emotional expression, through methods that include love withdrawal, guilt induction, and manipulative tactics. In VRR cases, either or both of the parents may have been intrusive in these ways; empirical work would be necessary to establish this. In a study of 6th to 8th-grade children of non-divorced parents, Weymouth and Buehler (2016) found that increased parental intrusiveness was associated with increased parent–child hostility. These authors concluded that “parental control… may undermine developmentally normative increases in adolescent differentiation from parents… Higher parental intrusion may constitute boundary violations in the parent-adolescent relationship for adolescents… and adolescents likely attempt to restore balance by establishing less permeable emotional boundaries between themselves and their parents” (p. 716). This conclusion suggests that parents’ demands for more affection from resistant children are likely to create greater resistance.

Descriptions of PATs by practitioners and by young people who have experienced the treatments suggest a high level of psychological intrusiveness and pressure to change thinking and feelings about the non-preferred parent, as such changes are the goal of the treatment. Manipulative tactics were described in one case known to the present author, in which an adolescent was threatened with wilderness therapy or some isolating form of residential treatment if she did not cooperate. A deposition in another case revealed that a PAT practitioner had accused children of cruelty to their non-preferred parent and attempted to induce guilt about the children’s position on contact with that parent. By analogy with the work of Weymouth and Buehler, these treatment approaches may also increase adolescents’ hostility, although the children may make pragmatic choices to behave in ways demanded by powerful parents and practitioners, especially when the real power of family court orders is behind the demands. Empirical work is needed to confirm or disconfirm this speculation about the role of intrusive parenting and psychological control in VRR and PATs.