The use of children as weapons is a central feature of intimate partner violence (IPV). One dynamic of IPV referred to as coercively controlling abuse (CCA; also referred to as intimate terrorism; Johnson, 2008), may be particularly likely to involve using children as weapons given that this type of abuse is characterized by psychological intimidation and control. Parents that have a history of CCA often threaten to harm or block access to their children if their victims were to leave (Hines & Douglas, 2010; Hardesty & Ganong, 2006). The abusive dynamics between the parents do not simply stop when their relationship ends (Eckstein, 2011; Senkans et al., 2021), and shared children provide continuous opportunities to maintain the abusive behaviours.

Harman and Kruk (2022) have drawn direct parallels between CCA and parental alienating behaviors (PABs), which are a pattern of behaviors executed over time by a parental figure (the alienating parent) to make a child come to believe their other parent never loved them, abandoned them, is unsafe or unfit, resulting in the alienation of a child (Baker & Darnall, 2006; Harman et al., 2018, 2021a, b, 2022). PABs have been associated with negative outcomes that are unique for alienated children (e.g., lack of ambivalence towards parental figures, Bernet et al., 2020) compared to children experiencing other forms of family conflict (e.g., Harman et al., 2019a), as well as long-term outcomes similar to those experienced by children experiencing other forms of abuse (Harman et al., 2018). PABs often include the denigration of the alienated parent, interference of contact between the child and the alienated parent, parentifying behaviours, and loyalty inducing behaviours (Baker & Ben Ami, 2011; Harman et al., 2018). Many of these behaviours involve intimidation and control toward both the alienated parent and the child. Indeed, common PABs include threatening the alienated parent with losing access to the child if they do not comply with the alienating parent’s demands, spreading false rumors about the alienated parent to the child and influential figures in the child’s life, threatening the child if the child shows love or affection to the alienated parent, and monitoring and controlling the child’s communications with the alienated parent (Harman et al., 2018).

Research findings that PABs are abusive towards the child (Kruk, 2018; Verrocchio et al., 2016; Verhaar et al., 2022) and the alienated parent (Balmer et al., 2018; Harman et al., 2018; Lee-Maturana et al., 2020; see Harman et al., 2022 for a comprehensive review) have led scholars to characterize these behaviors as a form of family violence (Djikstra, 2019; Harman et al., 2018; Lee-Maturana et al., 2020). Recent research has also determined that abusive and alienating parents have similar characteristics. For example, alienating and abusive parents both tend to have personality traits that are features of Cluster B personality disorders (e.g., narcissistic, borderline, & sociopathic traits; Cameranesi, 2016; Cunha Gomide et al., 2016; Gordon et al., 2008; Harman & Biringen, 2016; Hines, 2008; Spidel et al., 2013). These traits include a lack of empathy, difficulties in regulating emotions, and a tendency to see the world in a dichotomous way.

Alienated children exposed to PABs report lower self-esteem and higher rates of substance abuse disorders and are at risk for anxiety and depression (Baker, 2005; Baker & Chambers, 2011; Baker & Verrocchio, 2015, 2016; Ben Ami & Baker, 2012; Bernet et al., 2015; Harman et al., 2022; Miralles et al., 2021; Verrocchio et al., 2016). While alienated children may appear to excel in some areas of their life (e.g., school), they often develop psychosocial problems and have difficulties forming adult relationships (Warshak, 2015). Alienated parents also report many mental health issues including anxiety, depression, post-traumatic stress symptoms, and suicidality (Baker & Verrocchio, 2016; Balmer et al., 2018; Harman et al., 2019b; Kruk, 2015; Sher, 2015; Tavares et al., 2020). These outcomes are comparable to outcomes experienced by victims of other forms of CCA (Stewart & Vigod, 2017; Zlotnick et al., 2006).

Parents that use PABs often levy multiple false allegations of abuse towards the alienated parent (Harman & Lorandos, 2021; Hines et al., 2015; Lorandos, 2013; Lowenstein, 2012), which may be considered a form of legal and administrative aggression (Harman et al., 2018; Hines et al., 2015). Likewise, parents use PABs to make the child believe their other parent never loved them, abandoned them, and is unsafe (Harman et al., 2021). Moreover, many alienated parents report leaving their relationship with the alienating parent because that parent was abusive towards them (Lee-Maturana et al., 2020; Rowlands et al., 2023). The use of PABs to harm a parent and their relationship with the child often coincides with the use of other abusive behaviors towards the child, such as child physical abuse and neglect (Baker, 2007; Baker & Ben Ami, 2011; Clawar & Rivlin, 2013). Indeed, different types of abuse often co-occur, such that someone who abuses their romantic partner is also likely to abuse their children (Slep & O’leary, 2005) and children often experience multiple types of abuse rather than one (e.g., Chan et al., 2021; Finkelhor et al., 2007; Turner et al., 2010). If an alienating parent is more likely to have other substantiated allegations of abuse against them compared to the alienated parent, this finding would lend empirical support to PABs being a form of CCA.

Abuse Claims and Parental Alienation

Some critics of PA scholarship have argued that the concept is merely a legal defense created by abusive fathers who are attempting to gain or maintain custody of their children (e.g., Meier et al., 2019; Neilson et al., 2019).The studies used to support these arguments have been subjected to considerable criticism for conceptual, methodological, and statistical issues (e.g., Harman & Lorandos, 2021), such as presuming that all allegations of abuse are valid (e.g., Webb et al., 2021). Harman and Lorandos (2021) tested the Meier teams conclusory statements and failed to find empirical evidence that courts were placing children in the custody of active abusive parents. Other critiques of PA have argued against the notion that PABs may be comparable to other abusive behaviours and Mercer (2021) argues there is not enough evidence to conclude that they are. The goal of the current research is to examine whether parents that use PABs may be more likely to be abusive than their victims. This would provide more empirical support for the position that PABs are a category of abusive behaviours.

A “call to action” has been issued by critics of PA scholarship to change domestic violence and child abuse laws such that PA is taught to mental health and legal professionals as a concept misused in court by abusive fathers and research on the topic declared inadmissible in court (e.g., Meier et al., 2019). But there has been a lack of transparency regarding the methods and statistical models used by Meier et al. (2019), so it has not been possible for scientists evaluate their statements that PA is in fact being “misused” in court (see Harman & Lorandos, 2021). As to admissibility, Lorandos (2020a) found that in almost 1,200 trial and appellate records in the United States between 1985 and 2018, the concept of PA was found to be “material to the proceedings, probative of important facts, relevant to the court’s deliberations, admissible, and worthy of discussion” (Lorandos, 2020a, p. 330). Given admissibility standards are dependent upon the scientific status of the field, it is important to note that Harman et al. (2022) conclude at the end of the largest systematic review of the scientific literature to date that that “it is no longer tenable to dismiss the field as lacking in scientific status.”

Despite these research findings, some domestic violence and child abuse advocacy groups have pushed for changes to local and state-level laws to suppress the dissemination of scientific evidence on the problem to decision-makers (Harman & Kruk, 2022). If successful, these initiatives could have a negative impact on families affected by PA, particularly if the parent using PABs has been abusing the children in other ways.

The purpose of the current study is to empirically examine whether alienating parents use abusive behaviors that span multiple categories of abuse, because evidence of co-occurring victimization may provide support for the theory that PABs are sometimes part of a pattern of behavior used by a coercively controlling abusive parent to maintain power and control in the family (Harman & Kruk, 2022; Joshi, 2020). This would provide further support for the need to better understand and recognize PABs in family court cases. We also examined whether alienating parents are more likely to lodge unsubstantiated claims of abuse towards an alienated parent than vice versa. This finding could indicate that unsubstantiated allegations of abuse may be used by an alienating parent to deflect attention away from their own abusive behaviors and could provide additional evidence that alienating parents use legal and administrative aggression (e.g., filing false claims of abuse) to harm the target of their aggression (Harman et al., 2018; Hines et al., 2015).

Current Study

We tested two Open Science Framework preregistered hypotheses, https://osf.io/uz2pq/?view_only=9dd31927fe194322aeaf00ff53e06f01.

Our first hypothesis was:

  • H1: When there is a finding of PA, there is a higher likelihood of finding a substantiated claim of abuse against the alienating parent than the alienated parent.

Making false allegations of abuse against a party is another form of abuse called legal and administrative aggression (Hines et al., 2015), and can be used by alienating parents to obtain a custody advantage during parenting time disputes (Baker & Fine, 2014; Hines & Douglas, 2016; López et al., 2014). Therefore, we also wanted to examine whether alienating parents with substantiated findings of other forms of abuse would also be more likely to make unsubstantiated claims of abuse against the alienated parent than the alienated parent would make against them. Our second hypothesis was:

  • H2: When there is a finding of PA, there is a higher likelihood of finding an unsubstantiated claim of abuse against the alienated parent than vice versaFootnote 1.

Method

The data for this paper were collected and analysed for other projects (Lorandos, 2020a; Harman & Lorandos, 2021). A more detailed description of the data collection procedure can be found in an open access paper (https://doi.org/10.1037/law0000301) or on the Open Science Framework, https://osf.io/j9bh5/. All study materials, statistical output, and data are available on the OSF page.

Selection of Cases

A sample of appellate cases in which PA was known (identified by the court or a professional) or alleged (alleged, but not found by an expert or the court) were drawn from the ALLSTATES Westlaw database. The search query was: Among domestic violence (alienat! /s (mother father son daughter parent!)) & DA(aft 12 -31-1984 & bef 01-01-2019). In plain English, the query searched for: (1) any word fragment that contained ‘alienat’ (which could include alienate, alienated, alienating, or alienation); (2) the ‘alienat’ word fragment appeared within the same sentence as with one of these words: ‘mother,’ ‘father,’ ‘son,’ ‘daughter’ or the root word fragment ‘parent;’ and (3) the case was released and available between January 1, 1984 and December 31, 2018.

Our sample for this study only contained cases from the original dataset where PA was found to have occurred either by an independent evaluating expert or by the court after reviewing the evidence presented in the case (n = 492; 245 father alienating parent; 247 mother alienating parent). More detailed descriptive information regarding the process of classifying cases as ones where PA was found versus alleged, descriptive details about the about the appellate cases, and including details about all variables coded (e.g., coding of alienating parent, coding of custody outcomes) may be found in the original article (Harman & Lorandos, 2021).

The Data File

Two excel spreadsheets were used for this analysis. The first spreadsheet was the original excel spreadsheet created for Harman and Lorandos (2021), which lists the details for each appellate case. The second excel spreadsheet contained all claims of abuse made in each appellate case. This file included a number of variables, including (a) the case number, (b) the sequential number for each abuse claim for that case, (c) the type of abuse alleged, (d) who the allegation of abuse was made against, (e) who the alleged victim(s) of the abuse was/were, (f) who made the allegation of abuse, (g) the outcome of the abuse investigation, (h) any information about the abuse claim provided in final court judgment within the appellate case, and (i) whether the parent accused of abuse lost a significant amount (> 20%) of parenting time.

The Codebook and Additional Data Coding

As part of the original study, a codebook was created, and the variables are described in detail in Harman and Lorandos (2021). We updated the coding for the type of abuse that was claimed to have occurred so that this variable could be tested in our analysis. We also created a new variable, which was whether the claim of abuse was cited in the appellate case as being substantiated, unsubstantiated, or unclear. Finally, we coded whether the allegation was toward the alienating parent or the alienated parent.

Updated Coding for Claims of Abuse

A detailed, fillable pdf form was completed by coders for every claim of abuse made by any party described in the appellate report. The type of abuse for each allegation was recorded (domestic violence, child physical abuse, child sexual abuse, neglect, or “other”). Abuse was classified as “other” if it did not fit into one of the categories of abuse but referred to a serious allegation where a third-party authority became involved (e.g., assault on a co-worker). We updated the coding of this variable to include categories for child emotional abuse, multiple types of child abuse, and multiple child abuse and domestic violence. In the original database, the party that was making the claim and who the claim was made about were coded, and this was updated to include a numerical code for claims of abuse toward third parties (stepparents, extended family). This coding was originally completed by the first author and re-coded by a second coder who was not made aware of the study’s hypotheses. Inter-rater reliability was assessed by calculating Cohen’s Kappa as our codes were categorical. Inter-rater reliability was high, k = 0.91. All coding discrepancies were resolved by the first author by going through the data and identifying differing codes and explanations for how these were resolved. Rationales as to why a particular code was assigned were recorded in the database. Most discrepancies were due to the first author only coding one party and the second coder using codes for multiple parties. In many of the cases, specific incidents of abuse involved multiple parties being accused (e.g., a parent and stepparent). For analytic purposes, we only used coding for one party (the parent accused of abuse). If a third party (e.g., a grandparent) was the only person a claim of abuse was made about, that coding for the abuse was used in the analysis.

We created new variables to indicate whether the accused abusive party was the alienating or alienated parent, and whether the claim of abuse was only made towards the parents because our hypotheses were focused on comparing alienating and alienated parents and not third parties. When available, the exact wording used to describe the outcomes of the claims (e.g., false allegation, substantiated, or unsubstantiated) were entered. Claims made against the parents were coded as substantiated, unsubstantiated, or unclear. Two columns were critical in the coding of abuse claims: “Outcome of Investigation” and “Final Court Judgment.” The outcome of investigation column refers to whether there was a finding by the original party investigating the claim (e.g., police, criminal court, child protective services) and the final court judgement column refers to whether there was a finding regarding the allegation of abuse in the family court case. This coding was originally completed by the first author and re-coded by a second coder who was not aware of the study’s hypotheses. The inter-rater reliability was high, k = 0.82. All coding discrepancies were resolved by the first author by going through the data and identifying differing codes. Many of these were simple errors (e.g., substantiation of abuse was clearly noted but claim was coded as not substantiated) or cases where it was unclear (e.g., would say “charged” under outcome of investigation but “jury declined to indict” under final court ruling), these were coded as unclear (given a code of 3 and removed from the analyses). Detailed comments for the resolution of all coding discrepancies can be found in the abuse claims spreadsheet on the OSF page, https://osf.io/uz2pq/?view_only=9dd31927fe194322aeaf00ff53e06f01.

Code: Substantiated Abuse Claims

An abuse claim was coded as substantiated only if the notes in the data file indicated the claim was substantiated, founded, or the accused was described as guilty in criminal or family court. However, if one of these outcomes was noted in the “Outcome of Investigation” column, but then ruled as false, unsupported, or unsubstantiated in the “Final Court Judgment” column, it was coded as unsubstantiated. If there was a legal outcome for the parent (probation, jail time), it was coded as substantiated unless it was later disproved (under “Final Court Judgment”). If a restraining or protection order was granted, but it was unclear whether the claim was substantiated, this claim received a separate code because protection orders are frequently granted in the interests of being cautious. Coding these claims separately allowed us to examine the data with and without these cases. If there was a protection order granted and there was also evidence the claim was substantiated, then these cases were conservatively coded as “substantiated.”

Code: Unsubstantiated Abuse Claims

An abuse claim was coded as unsubstantiated if the outcome of trial was an acquittal or not guilty, even if the accused was charged. If the appellate opinion noted that there was not enough evidence to support the allegation, the claim was false, or the claim was withdrawn, then the claim was coded as unsubstantiated. For example, if the court ordered supervised visits and anger management classes for a parent, but later found the child denied or recanted the allegations to a therapist, this case was coded as “unsubstantiated.”

Code: Unclear Abuse Claims

Abuse claims were coded as “unclear” if they were not mentioned in the court ruling for both outcome of investigation and final court judgment columns or were inconclusive. If the appellate opinion stated there was no verdict or was unclear about outcome, the case was coded as unclear because it could not be determined whether the claim was substantiated or not. If there was a vague description of an outcome, but it was unclear whether that outcome meant the abuse was founded (e.g., “mother’s boyfriend was a bad influence on child”), it was coded as unclear.

We found a total of 1,126 separate claims of abuse in the cases where PA was found to have occurred. After removing claims for cases where the alienating party was not a parent (14), there were 1,112 claims of abuse. Of these claims, 224 were substantiated (20.14%), 524 were unsubstantiated (47.12%), 349 were unclear (31.38%), and 15 (1.35%) were cases involving a protection order that was granted, but it was unclear whether the claim of abuse was substantiated.

Analytic Approach

All analyses were performed in R Studio using R version 4.0.2. The abuse claims data were nested such that multiple claims were nested within families, so we used mixed modeling with the glmer function to model a random intercept for each family. All syntax, data, and output for these analyses are available on OSF. Full frequencies of abuse type by whether they were substantiated or not and the gender of the parent accused may be found in Table 3.

Results

Descriptive Analyses

Our exploratory descriptive analyses indicated that across the 492 cases where PA was found to have occurred, 41.46% did not involve any claims of other forms of abuse at all, 25% involved claims of abuse against the father, 22.36% involved claims made by both parties, and 11.18% involved claims against the mother. These proportions indicate that when there are allegations of abuse (58.54% of cases), it is relatively common to have both parents making abuse claims (38.19% of these cases).

Hypothesis 1

Our first hypothesis was that parents found to have alienated their children would be more likely to have a substantiated allegation of abuse against them than the parent who was alienated. Abuse claims was a binary outcome and was coded such that all substantiated claims of abuse received a 1 and all unsubstantiated or unknown claims received a 0. To test this hypothesis, we ran a logistic regression with a logit link function. Substantiated claims of abuse were modeled as a function of the gender of the alienating parent, whether the claim of abuse was against the alienating parent or not, as well as the interaction between our independent variables. Our independent variables were categorical, so we used effect coding such that accused parents and alienating fathers were coded as 1 and accusing parents and alienating mothers were coded as -1.

There was a significant main effect for abuse claims such that it was significantly more likely to find a substantiated claim of abuse against alienating parents as opposed to alienated parents, b = 1.49, SE = 0.15, z(1028) = 9.96, p < 0.001, Odds Ratio = 4.44:1. This result means that alienating parents had an 81.62% greater probability of having a substantiated abuse claim against them than an alienated parent. There was no significant main effect for gender, b = 0.17, SE = 0.16, z(1028) = 1.08, p = 0.28, Odds Ratio = 1.19:1Footnote 2.

Our main effect was qualified by a significant interaction between the gender of the alienating parent and whether the claim of abuse was against the alienating or alienated parent, b = -0.27, SE = 0.13, z(1028) = -2.11, p = 0.03, Odds Ratio = 0.76:1. To understand this interaction, we created two dummy coded variables to represent fathers and mothers, with the target group coded as 0 for each dummy variable. In our model for fathers, we found that fathers who alienated their children were significantly more likely to have a substantiated claim of abuse against them than mothers who were targets of PA, b = 1.22, SE = 0.18, z(1028) = 6.81, p < 0.001, Odds Ratio = 3.39:1. In our model for mothers, we found that mothers who alienated their children were also significantly more likely to have a substantiated claim of abuse against them than fathers who were targets of PA, b = 1.76, SE = 0.21, z(1028) = 8.23, p <. 001, Odds Ratio = 5.83:1. This effect was larger for mothers than the model for fathers. As indicated by Table 1, this finding is likely driven by the fact that alienating fathers were more likely to have unsubstantiated or unclear abuse claims against them (n = 122) than alienating mothers (n = 81).

Table 1 Frequency table for hypothesis 1 (N = 1033)
Full size table

We also conducted two post-hoc analyses on specific forms of abuse. The first analysis excluded claims of emotional abuse because PA may have been considered emotional abuse by the individuals investigating the allegations. This analysis resulted in a nearly identical pattern of results, with the odds ratio for our main effect of alienating parents being more likely to have a substantiated claim of abuse compared to alienated parents (4.61:1). This finding means that alienating parents had 82.18% greater likelihood of having a finding of non-emotional abuse against them than an alienated parent. There was no longer an interaction between the gender of the parent and whether they were the alienating or alienated parent, p = 0.24, Odds Ratio = 0.85:1. The second analysis used only claims of domestic violence, and we found a similar pattern of results, with alienating parents having 3.27 greater odds (76.60% greater likelihood) of having a substantiated claim of domestic violence perpetration compared to alienated parents. We did not find an interaction between the gender of the parent and whether they were the alienating or alienated parent, p = 0.27, Odds Ratio = 1.30:1.

Hypothesis 2

Our second pre-registered hypothesis was that when there is a finding of PA and a substantiated claim of abuse against the alienating parent, an unsubstantiated claim of abuse against the alienated parent is more likely to be found compared to the alienating parent. Our analyses focused only on allegations made by a parent against the other parent. Whether there was an unsubstantiated claim of abuse or not was a binary outcome and was coded such that all unsubstantiated claims of abuse received a 1 and all substantiated or unknown claims received a 0. For ease of interpretation, we also re-coded our alienating parent variable so that 1 represented whether the claim was against the alienated parent and − 1 represented whether the claim was against the alienating parent. To test this hypothesis, we ran a multilevel logistic regression with a logit link function and a random intercept for each case. Unsubstantiated claims of abuse were modeled as a function of the gender of the parent making the allegation and whether the claim of abuse was against the alienated parent or the alienating parent, as well as the interaction between our independent variables. This model now only included abuse claims made by one parent and allowed us to also see if there are gender differences with respect to parents that make unsubstantiated allegations of abuse.

Results indicated a main effect for abuse claims such that there was a significantly greater likelihood of finding an unsubstantiated abuse claim against the alienated parent compared to an abusive parent who was also found to have alienated their children, b = 1.82, SE = 0.21, z(714) = 8.47, p < 0.001, Odds Ratio = 6.17:1Footnote 3 This result indicates that there is an 86.05% greater likelihood of finding an unsubstantiated abuse claim against an alienated parent as compared to an alienating parent who had a substantiated finding of abuse. As with each of the other analyses, we did not find a main effect for gender of the parent making the allegation of abuse, b = -0.37, SE = 0.20, z(714) = -1.86, p = 0.06, Odds Ratio = 0.69:1. We did not find an interaction between the gender of the parent and whether they were the alienating or alienated parent, b = 0.08, SE = 0.22, z(714) = 0.37, p = 0.71, Odds Ratio = 1.08:1. Full frequencies can be found in Table 2. Of all the unsubstantiated allegations made by a parent in our dataset (606), 84.32% were made by alienating parents with 58.42% made by mothers and 25.91% made by fathers. Table 3 provides a full breakdown of all abuse claims by category of abuse, whether they were substantiated or not, and whether the accused was the mother or father.

Table 2 Frequency table for hypothesis 2 (N = 719)
Full size table
Table 3 Frequency table for type of abuse (N = 1034)
Full size table

Discussion

The primary objective of this study was to examine whether parents who have been found to have alienated a child from their other parent are more likely to have used other forms of abuse (e.g., domestic violence, child physical abuse) than alienated parents. We found that alienating parents, regardless of gender, had an over 81% greater probability of having a substantiated claim of abuse against them, compared to the parents found to have been alienated from their children. In other words, regardless of whether the alienating parent was a mother or a father, they were more likely to have a substantiated claim of abuse against them than the alienated parent. This finding directly contradicts claims made by critics of PA research that it is the parent (fathers specifically) who alleges being a victim of PA who is the abusive parent (e.g., Meier et al., 2019). We found that it is the alienating parent who is more likely to have other findings of abuse against them, not the alienated parent.

With our second pre-registered hypothesis, we tested whether alienating parents would be more likely to levy unsubstantiated allegations of abuse against the alienated parent than the other way around. Our reasoning was based on evidence that many abusive people tend to deny and deflect responsibility for their own behaviors (Avieli, 2021). We found that parents alienated from their children had an over 86% greater probability of having unsubstantiated claims of abuse made against them than the parent alienating the children. This finding indicates that false, or exaggerated claims of abuse or neglect may be a strategy used by alienating parents to make it appear the other parent is abusive.

Limitations and Future Directions

A limitation of this sample is that PA is very difficult to prove in family court, and given the controversy that surrounds the concept, some professionals are hesitant to make a finding of PA. In contrast, the weight for substantiating other claims of abuse in family court is lower than criminal court due to differences in burden of proof (Lorandos, 2020b). Nevertheless, it is important to note that in our data set there were often many third parties involved with determining serious forms of abuse, such as child sexual abuse. Allegations were often investigated by the police, CPS, child abuse specialists, and these findings were then also considered by experienced trial court judges.

Related to this, another limitation may be the complexity of PA as a concept and in turn the challenge of developing a legal definition of PA that may be used consistently across family court professionals. Although PA as a concept is complicated, awareness regarding PA has been increasing in the family court system over the last thirty-eight years and a remarkable consensus concerning the concept and its sequelae has developed (Lorandos, 2020b). Indeed, recent research suggests that there is nearly 80% agreement among custody evaluators regarding the defining features of PA (Bernet et al., 2022). As such, we can be reasonably confident that court professionals are using similar criteria in making their findings of PA, though there may be considerable variability across families in the types of PABs being used. Importantly, scientists and clinicians study and treat many issues that may have legal implications but for which clear legal definitions do not exist, including post traumatic stress disorder (Blotcky et al., 2022) and child sexual abuse (Haugaard, 2000; Mathews & Collin-Vézina, 2019), and this does not negate the importance of considering these issues when determining legal outcomes.

Another limitation of this study is that the sample contained U.S. appellate cases, some of which had minimal detail compared to what is available in a trial record. Often, a full history of an appellate case and specific details about claims of abuse for all of them was not possible to obtain. This may be why many of the allegations in our dataset were coded as unclear (31.38%). Many family court cases are not appealed and when they are, an appeal is often based on alleged errors made by the trial judge. Moreover, many families cannot afford to appeal a trial ruling. As such, there is a question as to what extent these findings could be generalized to all family court cases. A related research team is currently collecting data using trial level cases from Canada, as Canadian trial courts provide more detailed descriptions of the cases and will allow us to re-test our hypothesis to see if the findings reported here replicate.

A challenge in studying the relationship between PABs and other forms of abuse is that PABs themselves may be classified as other forms of abuse. In one of the cases examined, a mother was found to have engaged in sexual misconduct towards her children after trying to convince the children they were being sexually abused by their father. In this case, the mother was accusing the father of using PABs, so she was not coded as the alienating parent in our dataset. However, the courts found she was the parent that was using PABs. Moreover, one of the PABs she was using could be classified both as a PAB and as child sexual abuse. Although we tried to remove any abuse claims that seemed to be a PAB, it is still possible that there was some overlap in our sample.

This research builds on work of other scholars (e.g., Harman & Lorandos, 2021) indicating that allegations of abuse, when founded, affect child custody decisions and interventions that are ordered to assist the family. More research examining the way allegations of abuse leveraged by both alienating and alienated parents are investigated and litigated in child custody matters would be very useful.

Implications

The current research has important implications for scholarship and practice related to domestic violence, judicial systems, and the treatment of alienated children. We document that parents found to have alienated their children were more likely to have a substantiated claim of abuse against them than parents found to have been alienated, regardless of gender. These findings also suggest that alienating parents are more likely than alienated parents to engage in legal and administrative aggression by making false allegations of abuse. Judges and third-party experts (e.g., custody evaluators) may be aware of this behavior but do not always take action to deter parents from making these accusations in family court. When a false allegation has been made, particularly when these reach a level of severity that the child is affected, family courts should recognize these allegations are a form of legal and administrative aggression and be willing to take appropriate actions to protect the falsely accused party and the children. It also important to note that many claims of abuse that are unsubstantiated may be valid, but they lack enough evidence to make a finding, or they may be valid concerns that turned out to be unsubstantiated (Trocmé & Bala, 2005). The threshold for determining a false claim should be high enough to ensure that parents with good intentions and valid concerns are not deterred from coming forward.

These findings also support the need for more research on treatment for parental alienation across all levels of severity. Individual therapy without removing a severely alienated child from the abusive parent has been found to be contraindicated and often makes the child’s condition worse (Clawar & Rivlin, 2013; Warshak, 2015). Likewise, intensive intervention programs designed to treat a more general resist/refuse dynamic are not effective for some families (Saini, 2019) because of a failure to differentiate cases where there has been severe parental alienation versus other forms of family conflicts (e.g., loyalty conflicts). Effective interventions can be applied when needed for families where parental alienation is found to have occurred, and prevention and early invention for all forms of abuse will minimize the need for more intensive and structured interventions (Niemelä et al., 2019). Failing to recognize the psychological abuse that the alienated child is being subjected to in severe cases of parental alienation, and that they may also be subjected to other forms of abuse, leaves the child unprotected.

Conclusion

Mental health and legal professionals have been studying PA for decades using rigorous methodological standards and detection tools (Bernet et al., 2020; Harman et al., 2022). Indeed, the development of a 5-factor model for the detection of PA has been important for ensuring that cases where children have legitimate reasons for refusing contact are not mistakenly determined to be instances of PA (Bernet & Greenhill, 2022). Despite the misinformation perpetuated by some scholars, this research documents that it is the alienating parents who are more often found to be the abusive ones, while it more likely the alienated parents who are the victims of false abuse claims. The evidence provided in this study suggests that PABs are just another name for the coercively controlling use of children as an IPV strategy. We argue that children and their alienated parents need to be protected from this form of abuse.