change the world badge

change the world badge


Child Psychology Blogs

Concerned About Unconventional Mental Health Interventions?

Concerned About Unconventional Mental Health Interventions?
Alternative Psychotherapies: Evaluating Unconventional Mental Health Treatments

Tuesday, September 22, 2015

Is Parental Alienation Child Abuse?

For some years now, some psychologists and some divorced parents have been arguing that one parent may alienate a child from the other parent by systematically manipulating and exploiting the child’s beliefs and needs. The child then rejects contact with the noncustodial parent simply because of the alienating behavior of the custodial parent, and not because of any realistic fear or expectation of abusive treatment. This process and outcome have been referred to as “parental alienation”, presumably a term derived from the old legal term “alienation of affections”, referring to the seduction of someone’s adult partner.

Accusations of parental alienation have often been used in the courtroom as part of divorce strategies, but over the years such accusations have been met more frequently by counter-accusations about the manipulative and controlling motives of the parent who feels rejected. Although most of the initial discussions about PA focused on the rights of each parent to have contact with a child, and were quite often embedded in fathers’ rights claims, such a focus was obviously an open invitation to accusations of selfishness, egotism, even (of course) narcissism. Possibly as a result of this, possibly because of ongoing research, within the last several years the strategy shifted to the claim that PA is a form of child abuse, and to attempts to support this view by demonstrating poor adult outcomes for children who experienced PA.

Is PA child abuse? Well, the real answer to that is, it is if we say it is—“we” being the public whose shifting attitudes help to determine how child abuse is defined legally and otherwise.

The historical changes in definitions of child abuse are very obvious. To begin with, the very term emerged only with the work of Henry Kempe and his colleagues in the early 1960s. Before that, an adult who was shown to have harmed a child directly might be prosecuted in the same way as would have occurred if he harmed another adult--  or, he might not, any more than if he had harmed his wife. Some levels of abusive treatment (by today’s standards) were considered the perquisite of a parent, and more especially of a father. (In Samuel Butler’s novel The way of all flesh, set in about 1860, a father beats his four-year-old with a stick for saying “tum” rather than “come”, and is thought to be an excellent parent and excellent clergyman by all who know him.)  In addition, behavior that would now be considered abusive, like putting a child in a dark closet or limiting their food,  would usually have been considered just good old-fashioned parenting.

Today, treatment that is considered abusive from a legal viewpoint is not necessarily the same as treatment defined as abusive for research purposes. In Andrea Sedlak’s 2001 document published by the Department of Health and Human Services, for research purposes, spousal abuse in the child’s presence is regarded as child abuse, for example. Sedlak also described how the standards had relaxed from a requirement that an act actually harm the child before it was considered abuse, to a standard in which endangerment, or potential for harm, was defined as abusive. Under the harm standard, for example, failure to give a child adequate emotional support would count as abusive only if judged as serious, but under the endangerment standard could be counted under less serious conditions.

An important point about changing definitions of child abuse is that developmental outcomes of an action do not need to be demonstrated to be negatively affected in order for the action to be considered abusive. Use of terms like “harm” or “endangerment” implies that we consider an action to be abusive because it is known to cause harm to a child in the short or long run. But this is not necessarily the case, nor is harmfulness necessarily the criterion for identifying an act as abusive.

Obviously, many acts considered abusive are thought of in that way because they cause harm, or have the potential to cause harm, either in the short or in the long run. However, some actions that are argued to be harmful in the long run, like spanking, are not considered abusive unless they cause direct harm in the short run. Other actions, like any interaction with sexual implications, may be seen as abusive whether or not either immediate or later harm has been demonstrated. (For example, Cindy Hamilton’s nude photographs of her children[“Hold Still”] were seen as highly controversial  and possibly abusive.) Susan Clancy’s research on people who as children had experienced non-coerced, non-penetrative sexual contact with adults suggested that harm did not result from these experiences, but such contact is nevertheless considered deeply repugnant by most adults in the U.S. and is identified as abusive.

We can see, then, that there is an overlap between acts considered abusive and acts shown to be harmful, but the categories are somewhat different. This leads to a look at efforts to show that PA is abusive and that a parent shown to be alienating deserves to be treated as an abuser and separated from the child, for the child’s own good. Amy Baker, in her book Adult Children of Parental Alienation Syndrome, reports that she has interviewed 40 adults whom she contacted because of their statements on the Internet or by word of mouth. Baker looked at interviews with     these people in terms of their experiences with emotional abuse, as defined some years ago by James Garbarino, and including actions like isolating the child, terrorizing the child, and corrupting the child. She reported that these emotionally abusive acts were recalled by the adult interviewees as accompanying or forming part of parental alienation efforts, and that the adults remained troubled by their experiences.

Was this approach a good way to demonstrate that PA is child abuse? As a beginning for research, it was probably the best that could be managed for the time being. However, take a look at some of the problems about drawing any clear conclusion from this work. First, we have the fact that the persons chosen for interviewing were already so concerned about their present situation and past history that they were telling their stories on the Internet. These were not the people who say, “My father always made nasty remarks about my mother and didn’t want me to see her, but actually I got along fine and am doing well now. I think I understand where both of them were coming from, although of course I didn’t want them to divorce when they did.” Baker’s work of necessity focused on people who felt that PA had occurred and that it still influenced their lives in a negative way.

Second, Baker’s work completely conflates PA with emotional abuse. It would not be surprising if the interviewees whose parents were emotionally abusive would be unhappy and troubled as they recall those events--  whether or not there had been PA involved, or even whether or not there had been a divorce (and indeed Baker acknowledges that a child may reject a parent while the marriage remains intact).
To show that PA is harmful in the short or long run would require longitudinal studies in which children who appeared to be in a PA situation were followed into adolescence and adulthood and evaluated with respect to adaptive coping and mood. These individuals would need to be compared with a similar group without the PA experience, but matched on experiences like divorce and emotional abuse. That work is not very likely to happen, for many reasons—but unless it does, we cannot responsibly say that PA in and of itself is harmful, and, therefore, abusive.   
As I said earlier, there are a number of actions toward children that are considered abusive without evidence that they are harmful, just because we as a society find them reprehensible. It’s possible that PA might someday join that category, but at present it appears to me that American and Canadian society in general see PA as bad manners and a bad example to children, but not as harmful to children, and not to be considered abusive because of its effects. Because this is the case--  evidence of harm is not clear, and the consensus does not see PA as child abuse--  I believe it is a mistake for the courts to accept arguments about PA as a reason to alter custody arrangements or to order children into therapies claiming to treat PA. I consider it an especially egregious mistake for courts to refuse a parent contact with a child because that parent will not confess to PA-related actions; this is all too reminiscent of the old Satanic Ritual Abuse accusations, in which a parent could not be with a child if he or she denied SRA, but certainly could not contact the child if he or she “confessed”.


  1. It is exceedingly rare for a parent to judicially be restricted from contact with a child due to accusations of PA. In my state there has not been a single incidence. There have been cases where contact has been restricted due to apparent psychological damage to the child that appeared to be related to certain behaviors of the parents. The parental behaviors were those typically associated with cluster B personality disorders. Would you find these decisions disturbing due to lack of empirical evidence?

    1. None in your state? As it happens, I know of some.

      But it is certainly one thing to restrict contact when observable damage has already occurred, and another to restrict it because of a weakly-supported prediction of long-term damage.

      Also, I might point out that in some cases there is no definite order for restriction of contact, but the accepted parent may be given the choice of giving up contact for a period (e.g., Childress' protective separation) or losing contact altogether and being found in contempt of court into the bargain. If this latter is not happening, how is Childress getting these presumably oppositional accepted parents to allow the child to be separated from them? If it is happening, and parents are forced into making a choice that effectively restricts their contact, that restriction would not get counted as restriction due to court order.

    2. Thanks.

      Can you supply me with references for the cases you mention where parental access was lost due to allegations of PA?


    3. No, I can't. The matter is still in litigation and I was told about it in confidence. I have seen the court documents, though.

    4. Just now, the Tsimhoni case. It's playing out in Michigan! google Tsimhoni Judge Gorcyca, Lansat and you will find plenty of information, and there is a dropbox with court documents linked to Facebook page "justice for tsimhoni"

  2. Do you think there might be merit in studying what effect this sort of situation has on the rejected parent?

    Also, do you think there would be any merit in studying how this situation affects "left behind" siblings (including step and half). For example, my mother lost her older brother (her only sibling) when her father took him and ran out on the family. It was not "PA", that I know of (visitation did take place at some point after the initial event), but it had an affect on my mom that remains with her to this day.

    The evidence may not reveal "abuse", but it might reveal a good reason to have some sort of professional intervention to normalize things. (Not talking about removal here.)

    If it actually is a case of bad manners and a bad example, then finding "favored" parents and their families to study shouldn't be too hard. Even if you can't get the "rejecting" child to participate in the study.

    Although, the more extreme cases would probably not want to take part in the study, and so the data might get skewed toward "not as bad as people think".

    The idea of "PA" has been around for 30 years now, are there any credible researchers or credible institutions that have studied this already, or started researching it at all? "PA" seems as though it is definitely something that is going to take a long time to think about and to study to make sure that none of the children are inadvertently harmed by good intentions.

    1. Well, yeah! I think that would be fascinating and extremely valuable as a way to look at PA-ish events on a continuum rather than as a category of their own. Something else that might be looked at is the extent to which one parent's derogation of the other leads a child to think badly of the derogating parent, not necessarily because the child thinks the statements are not true, but because he or she does not see this as acceptable parent behavior-- potentially leading the child to abandon the idea that either parent could be helpful.

      As for credible researchers, I had hopes for Richard Warshak at one time, but my impression is that he may have dropped this work (?). In defense of the non-involvement in this topic, I have to point out that understanding either shared or non-shared environmental effects on development is enormously complicated, and as you point out, getting people to participate is no easy thing.

      I do think that studying the rejected parent might be a good start, both in terms of what you mention (the effect of the situation on that person), and in terms of what I've suggested-- the characteristics of the person that contributed to this kind of enmeshment. Perhaps rejected parents are simply of randomly distributed kinds and luck alone made them marry someone who manipulates child attitudes-- but if anyone is going to claim that family systems are at work, I think they have to consider that rejected parents may have a range of characteristics in common-- and I'm referring to personality traits and interaction styles, not to abusive treatment of other people.

      You speak of the need for professional intervention even if there is no abuse, and I certainly see what you are getting at. But first there has to be a form of intervention that is shown to be effective; second, somebody is going to have to pay for it, and that may be very challenging to people trying to get back on their financial feet after the divorce; third, there continues to be a stigma attached to psychological interventions; fourth, if people are court-ordered into treatment, that becomes a problem in itself. Sorry to be so gloomy about it, but these are the realities.

    2. This is interesting to me because I had a nasty divorce and my ex-husband did a lot of nasty talking in order to turn my son against me. this mostly consisted of presenting everything I did in the worst possible light. It worked for a while, and we went to a therapist, who first brought up that this might be PA syndrome (well, actually I saw a book about it on his desk and said, "What's that?" and he told me about PA). He was a good therapist and focused on positives and what to do from here forward, but the kid's negative attitude got worse and worse, peaking right around age 14 when he joined ROTC. It was so bad my dad had decided to have a talk with my son, but then suddenly, poof, I was no longer the devil. I've never known why. Was it that I tried to keep up positive communication by email? Was it the strong ties my son had to my family of origin? Did his ROTC commander (whom he had huge respect for) tell him what was up? Did he just get mad at his dad? Don't know, and really, don't care. Just glad to have my son back. It put me into a very deep depression (I have major depressive disorder, recurrent, severe, already) and I tend to view myself as a "bad mother," which I didn't even realize until I had another baby, and my second husband said, "I don't know where you get this idea that you suck at motherhood. I think you're a fantastic mother." Which did a lot for me.

      Anyway, just a personal perspective. I don't know if PA is an actual psychological syndrome, but I know manipulation happens. I was very young (20) and was absolutely blindsided by the whole thing.

      Love your blog.

    3. Doesn't everything peak at about age 14?

      Of course manipulation happens, and divorce does not bring out the best in anyone, including the children. But I think your story shows the complexity of events that are simplified as PA. Your youth when your first child was born? Your depression? Your ex-husband's behavior? Your son's need to feel autonomous? Your remarriage, whenever it occurred? Any or all of those could have affected your son's behavior, both negative and positive.

      Thanks for the story and the kind words!

  3. This is a shocking apologia for child abuse.

    I've followed this debate between Jean Mercer and Craig Childress from the start, and I almost jumped in against Mercer's initial pillory aimed at Childress because it seemed so materially unfounded and remarkably focused on Childress' purported personal shortcomings. I read it five times trying to figure out what Mercer's actual objections are. Therein lies the framework for understanding this dispute which is of expansive import to family law and clinical practice with fractured families.

    It is well documented that adults with personality disorders are more likely to harm people, including their own children. Why Jean Mercer would attempt to have us ignore the implication that such harm to families is real and needs to be understood; well, that is an interesting question about Mercer, but also about the child abuse intervention industry itself....

    1. My point was that people's views of what constitutes child abuse have changed historically. A hundred years ago, nobody would have thought twice about an adult sharing a bed with a child; today, this is seen as having strong sexual implications. It's possible that over time the PA-abuse claim will come to be seen as correct by most people, but this has not yet happened. Neither has there been effective demonstration that long-term problems emerge from PA in and of itself, and separate from the usual list of forms of emotional abuse.

      So why claim that PA is child abuse? The only reason I see is that the law prohibits child abuse and takes a dim view of abusive parents, so if PA can be brought under the abuse rubric simply by assertion, a powerful tool is put into the hands of lawyers acting for rejected parents. Whatever the value of this outcome might be for solution of a social problem, it is deceptive to say that the abuse claim is based either on evidence of long-term harm to the child or on consensus about what constitutes abuse.

      An argument in terms of the rejected parent's own rights would be more straightforward, but of course this does not have the emotional power of referencing the child's interests.

      By the way, my original statement had to do with my doubts about Childress' theory, and the lack of evidence for the treatment he proposes.

    2. Hi Dr. Mercer,

      I agree with you that an assertion of PA is an assertion of child abuse and the child abuse can be assert for a variety of reasons, some nefarious. That being said, do you really think that it is appropriate for a court to ignore rather than address an assertion of child abuse? Are you suggesting that observed indications of psychological abuse should be treated differently than observed indications of physical or sexual abuse?


    3. I think indications of psychological abuse should be treated with much more caution than indications of physical abuse. What's regarded as psychological abuse may be rather different in different cultural groups or even parts of the country. Certainly there can be ill effects of psychological abuse, but these are much more complex and hard to demonstrate than effects of physical abuse.

      How many people have laughed at Bill Cosby's old riff on "I brought you into this world, I can take you out of it"-- but is such a statement not psychologically abusive by current middle-class standards? People would not laugh at a description of a serious beating given to a child, but they did laugh at this possibly abusive statement--- does this not suggest that they are far from sure that it is abusive to threaten a child in this way?

      My point is that although psychological abuse surely exists, our attitudes toward it, and our definitions, tend to be biased by our own experiences and values, and by the fact that its effects are hard to measure. That's why I say we need to be cautious about our assumptions and consider context carefully. One person's psychological abuse may be another's firm discipline or even a necessary technique for saving a child's soul. Only the context can tell us whether the parent's action warrants removing the child.

      As for sexual abuse, coercive penetration aside, I want to point out, as I mentioned the other day, that views of what is sexually abusive have changed to include many actions that no one would have blinked at years ago. Cultural differences about sexually-tinged behavior displayed to children are also huge. Do you remember a case several years ago where a Middle Eastern man living in New England was seen by a baby-sitter to kiss his baby boy's penis? Baby-sitter reported to child protective services, etc., etc., but the final outcome was the conclusion that this behavior was perfectly normal and even admirable within the father's cultural background, and indicated his pleasure and pride in his son. I don't mention this to suggest that sexual abuse of children is no problem, but just to say that like psychological abuse, consideration of sexual abuse must consider the context-- whereas physical injuries are injuries no matter what cultural or other context they occur in.

    4. Jean, you said, "whereas physical injuries are injuries no matter what cultural or other context they occur in."
      This is not entirely true. There are cultural practices, such as coining, that leave injuries on the child, and adults who remember it happening as a child will attest to the pain and discomfort of the practice, but it's not considered child abuse (at least in VA).

    5. I just said that they are injuries-- not that they are always judged in the same way by every culture. In addition to coining, circumcision, ear piercing, cicatrization, orthodontia, nose jobs-- all are painful and have long-term effects, but are not universal considered abuse.

  4. ...
    Not all fractured families are such as a the result of serious psychological disturbance of one or both parents. But where there is a personality disorder present it is necessary for family court judges, custody evaluators, and family law attorneys to be aware that it must be dealt with in determining the best interests of the affected children.
    This isn't to say that personality disordered parents must be cut out of their child's life.

    Mercer writes (in a crude critique of Amy Baker's study of adults who were used when children to alienate a loving, involved parent):

    "To show that PA is harmful in the short or long run would require longitudinal studies in which children who appeared to be in a PA situation were followed into adolescence and adulthood and evaluated with respect to adaptive coping and mood."

    This is an unconscionable sleight-of-hand clearly intended to distract from the real issue under debate: psychologically disturbed people tend to reproduce in their children features of their own disturbance. This is not a question of research design, although well designed research is always needed as paradigms of inquiry evolve and new hypotheses emerge.

    Moreover, Mercer hoists herself by her own petard. She suggests that parental alienation would have to be compared with effects of emotional abuse via control group assignment, and that this is not likely to be undertaken "for many reasons—but unless it does, we cannot responsibly say that PA in and of itself is harmful, and, therefore, abusive."

    Pay close attention here to what she is saying - and doing! Mercer has set up an unfalsifiable construct which makes it impossible that parental alienation exist. By arbitrarily separating emotional abuse from parental alienation, she uses the former to protect the later from scrutiny (both research and clinical). No syndrome of abuse "exists in and of itself."

    This is exactly the myopia Childress warns about (pathogen, to use his metaphor). Can you imagine a clinician telling a troubled child, 'I'm going to help you but I will have to ignore the causes of your abuse'?

    Following Mercer's twisted logic, a responsible professional would indeed ignore the big picture diagnostic in order to serve up to the family court judge a tidy issue to be judicially resolved. If a lucrative referral to counseling can be made, all the better - ignoring of course the reality that personality disorders are not generally treatable by counseling.

    Finally, Mercer's rationalization of parental alienation as "bad manners", as seen by "society" is a wholesale surrender of the duty to protect children from abuse. Many states, including Oregon - where I live - legally obligate parents to encourage the child's relationship with the other parent.

    Is Mercer so attached to the bias that personality disorders can't really harm kids; is she so attached to that conviction that she cannot even recognize prevailing family court conventions established to protect children from divorce conflict?

    Don't take this question literally. Mercer does not operate in good faith on these important questions. To engage honestly with a prejudice compelled ideologue is to subject yourself to constant frustration and ultimately, especially if you are a child dependent on the care and nurturence of such a person, debilitating emotional distortions.

    Emotional abuse exists. We as parents, clinicians, and family court professionals must respect our obligations to understand and confront all forms of child abuse, not just those that flatter our personal agendas.

    1. One second, Kevin-- you are conflating accusations of, or even evidence for, PA , with personality disorders and emotional abuse, and those in turn with high conflict divorce. You are also claiming without adequate evidence that PA in itself has effects similar to those of all forms of emotional abuse.

      Although all metaphors need to be handled with caution, that pathogen metaphor is a particularly dangerous one, with its attractive medical/scientific quality. Teach people to call a parental behavior a pathogen, and you go a long way toward convincing them that the behavior must have a seriously negative effect. (Why not call the interventions antibiotics, too? Perhaps that won't work because clients know what antibiotics are.)

      As for my good faith, I don't have a financial or career dog in this fight, but it looks as though many PA supporters do.

    2. Assuming Kevin's definition of PA is the same as Childresses he is not conflating PA with personality disorders and abuse. Childress's model requires damage to the child and cluster B traits on the part of the parent to call the observed phenomena PA

    3. Yes, I see what you mean, but not everyone defines PA that way. It would be a good idea if Childress created a new name for the phenomenon he's concerned with-- BTW, does he not say somewhere that what he's working on is NOT PA?

    4. "Ne noceas, si juvare non potes" is the main precept of the doctor. Too many charlatans, but very few defenders for orphans. If the courts have begun to justify the murderers on the pretext that all the children in the orphanage are sick and have a tendency to self-harm and suicide ..... it's time to begin to protect these children.

      "Can you imagine a clinician telling a troubled child, 'I'm going to help you but I will have to ignore the causes of your abuse'?" (Kevin Hornbuckle)
      Can you imagine a child is after the "treatment" in a psychiatric hospital or in the grave ... but the doctor and the adopter do not receive adequate punishment because the alleged fault orphanage and biological parents. This way you can justify anything.
      If there is no adequate way to help a child .... then do not надо break into the child's head using picks and shovels. Better inaction than risky experiments on the child's mind.
      "Ne noceas, si juvare non potes" !!!

  5. His general practice is to put it in quotes and he has some language where he discusses his reasoning for doing so. My understanding is that he considers the observed phenomena similar to the observed phenomena that Gardner and his followers described as PA or PAS, but that their perception of the phenomena was poor resulting rightfully in limited acceptance. His argument is that there is a phenomena there and Gardner saw it, but it is nothing like what Gardner and his followers described.

    I believe most people in the field see the term "Parental Alienation" as tainted and using the term almost inevitably triggers difficulties for discussion. It is tempting to try to avoid these difficulties but probably inappropriate if you view psychology as a science where progress is made by iterative improvement. An analogy would be with the disease malaria. It was once thought to be caused by "bad air" thus the name, but the name remained even though the perception of the disease changed when the role of protozoans was understood. It was not a new phenomena, just a new perception of an old phenomena, so the named remained.



    1. I don't really understand this argument. If what C.C. is seeing is not what Gardner described, why call it by the same name? And if what he sees is not what Gardner described, how can he know it's the same phenomenon?

      Well, but, be that as it may, do you think these accepted-parent behaviors would be seen as pathological if the child did not then reject the other parent? Are we looking at anything other than Barber's "intrusive parenting", but in a situation where the rejected parent is likely to complain about the accepted parent's behavior?

      I would really like to know about the prevalence of PA-or-whatever-it-is-as-described-by-anybody. I don't seem able to find an estimate anywhere, and the numbers of cases as described by either C.C. or Warshak are quite small. Can you help?

  6. Gardner's descriptions went well beyond what he directly observed. He generalized and conceptualized way more than he reported observations in an objective manner. Nevertheless it is possible to tell that his focus was on this phenomina where the relationship between a parent and child was terminated in a way that had previously been rare. I think the subject matter for Childress is the same.

    With the disagreements over definitions and methods to identify there are no reliable estimates, but proponents usually give percentages in the low single digits, opponents say zero percent. Childress's definition is narrower than earlier definition so probably around 1% or less of children. Even proponents believe it is claimed much more often than it occurs.

    It is a very contentious issue and because of that it is a bit of a no-mans-land occupied only by zealots either for or against. You see a lot of truth-by-assertion and ad homen attacks and not much else. It is unfortunate that so little good work is done on the subject.


    1. Well said, and thank you for these constructive remarks.

      Unfortunately, all these difficulties would seem to need to be ironed out before the question of effectiveness of treatment can even be raised.

      As I've been saying, I'd really like to know the characteristics of the rejected parents. Without at all wishing to ascribe abusive behavior to all of them, I have to wonder what they have in common that would encourage marriage to a potentially manipulative person, or would guide their later behavior in ways that would make rejection more likely. The same questions could be asked about the children-- are there personality characteristics that would make them more ready to reject someone, whether or not they were encouraged to do so? I would assume that any family systems approach would include these questions, so I don't quite see how Childress claims he is systems-oriented.

      I recently read the suggestion that assigning guardians ad litem to these contested children would help us at least tell rational from irrational concerns-- not that I would suggest that irrational concerns should necessarily be ignored.

      BTW, do you mean one percent of all children,or one percent of children in high conflict divorce cases?

    2. Well, in the tsimhoni case, the Guardian at Litem mainly wanted his peace and quiet (as well as money for his services), but did not help the situation become more "rational"...