The following editorial originally appeared in the journal Psychological Science in the Public Interest (Vol. 6, No. 1). This editorial was written in conjunction with the report “A Critical Assessment of Child Custody Evaluations: Limited Science and a Flawed System” by Robert E. Emory, Randy K. Otto, and William T. O’Donohue which appears in the same issue of PSPI.
Emery, Otto, and O’Donohue give us an admirably clear and hard-hitting analysis of the way our current legal system functions in attempting to resolve child custody disputes. These authors adopt the premises that children will fare better (a) if parental conflict is minimal or at least contained, and (b) if children can maintain a good relationship with at least one, and preferably both, parents following the divorce. The authors focus on how procedures for dividing a child’s residential time between two divorcing parents can best further these two goals.
The authors argue, as many others have done, that the adversarial nature of litigation processes seriously exacerbates parent conflict. They provide evidence that mediation and other alternatives to litigation are workable, and they strongly recommend that these alternatives be used more widely to help parents resolve their conflicts without recourse to the courts.
There will be some parents, however, who cannot resolve their disputes even with the help of mediation, and so turn to litigation. Currently, courts are enjoined to award custody in accordance with whatever arrangement will serve “the best interests of the child.” The authors argue that this best-interests standard is itself a major weakness of the current system. They see it as unworkable because it is too vague. Judges, they say, have found it well-nigh impossible to determine which of two contesting parents can best support the children’s long-term well-being. For one thing, families in which one parent is clearly the more suitable custodian have usually been able to come to agreement privately and do not appear in court, so that contestants who do litigate tend to be equally good (or equally bad!) parents on balance. In disputed cases, judges have come to rely heavily on “expert” evaluators to determine the child’s best interests. The authors’ analysis of the weaknesses of these evaluations is sobering indeed. I can only applaud their judgment that standard measures of parents’ and children’s intelligence, personality traits, and emotional states are wholly inappropriate for custody evaluations, and that even the measures and constructs that have been designed specifically to assess child custody arrangements for individual children have no proven validity as predictors of a child’s well-being in the care of one or the other of two disputing parents.
Over and above the difficulty of determining children’s best interests, the authors claim that this standard has another serious flaw: Its vagueness tempts parents to dispute. There is no way they can know in advance who is more likely to win, so individual parents often think they will have a better chance in court than turns out to be justified. What is needed is a less ambiguous standard, one that will discourage a parent from litigating (and thereby dampen conflict) if the chances of succeeding in court can be known in advance to be poor. They propose a standard known as the “approximation rule”: that the postdivorce division of children’s residential time between the two parents should match, as far as possible, the “respective involvement of the parents in childrearing during marriage.” Obviously, this standard “tilts” toward mothers, since in the majority of families mothers have carried the primary responsibility for childrearing.
Is the approximation rule unfair to fathers? In many cases, yes, as the primary role of fathers in traditional and semitraditional families has been to provide economic support. For many fathers, this does not mean that they are any less committed to their children, or any less competent as parents, than mothers who have been doing more of the parenting. Fathers would surely be right to feel aggrieved by losing so much valued residential time with their children following divorce simply because they did less of the day-to-day interaction with children during the marriage. But the child’s residential time is a zero-sum game. As I have argued elsewhere (Maccoby, 1999), it is seldom possible to be equally “fair” to mother, father, and child in divorcing families. And while a tilt toward maternal custody may be unfair to fathers, a tilt toward father custody or even joint custody may be even more unfair to mothers. The approximation rule does serve to keep the door open to maintaining the child’s relationship with both parents, by allotting some residential time (overnight and vacation visits) to the less-involved parent.
The approximation rule apportions the child’s residential time according to the parenting regime that was in place at the time of the divorce. Judges and court evaluators do not have a crystal ball with which to predict how the child’s need for each parent, or each parent’s parenting competence, may change with time. In intact families, the father’s role relative to the mother’s tends to increase with the child’s age. Does this mean that there should be a father-time inflation factor built in to custody arrangements? Or that the courts should periodically reexamine custodial awards, to see whether the arrangement that was suitable for an infant or toddler is still best for a preschooler or grade-schooler? No, surely not. Such a procedure would increase uncertainty and disrupt the stability of existing arrangements. Furthermore, the courts are overburdened as it is, and they are no better equipped than parents themselves are to say what changes in residential arrangements are needed to accommodate not only a child’s growing maturity but also parental remarriage and residential relocation. There seems no viable alternative to leaving these time-based adaptations to the parents themselves, except in cases where one parent is so coercive as to make it imperative for the other parent to go back to court for protection.
The authors, then, argue that a paramount objective of custody adjudications should be maintaining the greatest possible continuity. Of course, there will be exceptional cases in which a parent who carried the primary responsibility for childrearing during the marriage is demonstrably unfit to do so as a single parent and in which the other parent seeks primary custody. How much flexibility should there be in the application of the approximation rule, to take account of such exceptions?
The authors take a very tough position on determination of parental unfitness. They say that a diagnosis of mental illness or substance abuse should only be considered disqualifying for child custody if the parent’s inadequacy is so great that it would justify taking a child away from parents who are not divorcing. But the two cases are not comparable. Removing a child from an intact family on grounds of parental unfitness—and thus subjecting the child to the risks of the foster-care system—is a much more draconian procedure than separating the child from a familiar (though impaired) caretaker and placing him or her in the care of the other parent. I am uneasy about closing the door against determinations of the relative fitness of the two parents as firmly as the authors recommend doing. Professional psychologists can and do make accurate diagnoses of severe mental health problems, though (as the authors note) these should only be considered in custody proceedings if they demonstrably and seriously impair a mother’s or father’s parental functioning. And there are objective ways of identifying a dangerous level of parental unfitness (e.g., several recent DUI citations) without relying on invalid measures. There is a tension between the advantages of allowing flexibility for such cases and the dangers the authors point to: that any weaker definition would open the system once more to all the ills of the best-interests standard. Parents bent on getting sole custody would be tempted to level charges of unfitness against one another, and the court would need once more to rely on unreliable evaluations to determine fitness. It remains to be seen how this tension can be resolved. I suspect that courts cannot escape altogether the task of making some basic determination of relative parental fitness before turning to the approximation standard. However Emery and colleagues have made an excellent case that the bar should be high for such a determination to prevail.
In recommending the approximation standard, the authors are by no means abandoning the principle that custody decisions should support the best interests of the child as opposed to the “rights” of fathers or mothers. In the fathers’ rights movement, there are a few people who claim that fathers have an intrinsic right to the “ownership” of their children over and above the children’s best interests. I infer that Emery and colleagues would not agree, believing instead that children’s best interests should trump parents’ needs and claims of rights if and when the two conflict. But they often don’t conflict: It is better for a child when the parents are satisfied that their own needs and rights have been given some weight in arriving at the custodial arrangement that has been worked out. The authors only contend that the approximation standard will serve children’s—and parents’—interests better than the current “best interests” standard as it has worked out in practice. About this, they are surely right.
- Maccoby, Eleanor E. (1999). The custody of children in divorcing families: Weighing the alternatives. In R.A. Thompson and P.R. Amato (Eds.), The postdivorce family: Children, parenting and society. Thousand Oaks, CA: Sage Publications.