CHILD CUSTODY REVISITED
ROBERT MNOOKIN* I INTRODUCTION When Professors Katharine Bartlett and Elizabeth Scott proposed a symposium and accompanying issue of Law and Contemporary Problems organized around my article Child-Custody Adjudication: Judicial Functions in 1 the Face of Indeterminacy, I was flattered, thrilled, and most of all surprised: flattered because of the extraordinarily generous ways they characterized the importance of the article to the development of family-law scholarship, thrilled because they suggested a number of leading family-law scholars had agreed to reflect on the article’s impact and contemporary relevance, and surprised because my article was nearly forty years old. It was also with a tinge of trepidation that I approached the prospect of the symposium. For the first fifteen years of my academic career, my central scholarly concerns were children and the law broadly, and issues relating to 2 divorce custody and the foster-care system in particular. But in the last twenty years I have focused nearly exclusively on conflict resolution, negotiation, and alternative dispute resolution, not family law. I was only generally abreast of developments in custody law. I had given no thought to the continuing relevance, if any, of my article. Truth be told, I had not even looked at, much less read, the article in more than twenty years. When my article appeared, Gerald Ford was President. International concerns focused on the Soviet Union and the Cold War. The median household income in the United States was $11,800, and the average cost of a 3 new house was $39,000. I wrote the article on a typewriter. There were no personal computers, or cell phones. There was no world-wide web. I wondered, Would the social, economic, political, technological, and legal changes since 1975 make my analysis seem strangely out-of-date?
Copyright © 2014 by Robert Mnookin. This article is also available at http://lcp.law.duke.edu/. * Samuel Williston Professor of Law and Director, Harvard Negotiation Research Project, Harvard Law School. I would like to thank Ilan Stein for his research assistance. 1. 39 LAW & CONTEMP. PROBS. 226 (Summer 1975). 2. See, e.g., ROBERT H. MNOOKIN & D. KELLY WEISBERG, CHILD, FAMILY, AND STATE: PROBLEMS AND MATERIALS ON CHILDREN AND THE LAW (6th ed. 2009); see also ELEANOR E. MACCOBY & ROBERT H. MNOOKIN, DIVIDING THE CHILD: SOCIAL AND LEGAL DILEMMAS OF CUSTODY (1992); Robert Mnookin, Foster Care—In Whose Best Interest?, 43 HARVARD EDUC. REV. 599 (1973); Robert Mnookin & Lewis Kornhauser, Bargaining in the Shadow of the Law: The Case of Divorce, 88 YALE L.J. 950 (1979). 3. See News and Events of 1975, INFOPLEASE, http://www.infoplease.com/year/1975.html (last visited Aug. 12, 2013).

250

LAW AND CONTEMPORARY PROBLEMS

[Vol. 77:249

I have now reread my article in light of the fascinating contributions to this issue of Law and Contemporary Problems. I have asked myself, Are the article’s two core themes—relating to the indeterminacy of the best-interests standard and differentiating child protection from private dispute resolution— still relevant? To what extent have changes in social norms, technology, and legal doctrine made my analysis unresponsive to contemporary challenges for custody law and policy? With profound gratitude to Professors Bartlett and Scott, I offer here my observations regarding these questions. II THE INDETERMINACY CLAIM The first, most fundamental claim in my article was that the best-interests standard was indeterminate. I exposed the indeterminacy of this custody standard by framing the judge’s decision under the standard “as a problem of rational choice” between custodial alternatives in which the judge was being asked to choose the alternative that would maximize “what is best for a 4 particular child.” Drawing on decision theory, a rational decisionmaker would need to (1) specify alternative outcomes and predict the probability that each outcome might occur, and (2) assign to each outcome a utility measure that 5 integrated his values and allowed comparisons among alternative outcomes. To do this the judge would need considerable information, predictive ability, and 6 some source “for the values to measure utility for the child.” After framing the best-interests standard in this way, I exposed its indeterminacy with two basic arguments: First, “[f]or most custody cases, existing psychological theories simply do not yield confident predictions of the 7 [long-term] effects of alternative custody dispositions.” Second, “even if accurate predictions were possible in more cases, our society today lacks any clear-cut consensus about the values to be used in determining what is ‘best’ or 8 ‘least detrimental.’” None of the authors in this issue express any disagreement with either of these two basic assertions. With respect to the first of these indeterminacy arguments, I claimed, “[P]resent-day knowledge about human behavior provides no basis for the kind 9 of individualized predictions required by the best-interests standard.” I pointed out that there were “numerous competing theories of human behavior, based on radically different conceptions of the nature of man” and that “no consensus 10 exists that any one is correct.” This remains true today.

4. 5. 6. 7. 8. 9. 10.

See Mnookin, supra note 1, at 255. Id. at 256. Id. at 257. Id. at 229. Id. Id. at 258. Id.

No. 1 2014]

CHILD CUSTODY REVISITED

251

Mental-health professionals still lack the capacity to make accurate predictions about the long-term impact of alternative custodial dispositions on children. Indeed, Professors Elizabeth Scott and Robert Emery suggest mentalhealth professionals today continue to use the illusion of scientific expertise to obscure the deficiencies of the best-interests standard, not simply as expert 11 witnesses in trials but throughout the dispute-resolution process. Scott and Emery see this as deeply problematic because such professionals “lack the scientific knowledge to guide them in linking clinical observations or test data to qualitative proxies for best interests or in comparing incommensurable 12 factors to make custody recommendations to the court.” Nonetheless it appears that mental-health professionals’ plans today might play an increasingly 13 important role throughout the dispute-resolution process. Psychologists and mental-health professionals continue to make predictive claims that cannot be justified by social-science research. My second indeterminacy argument also remains as relevant today as it was in 1975. The lack of a social consensus about values still plagues best-interests determinations. Determining what is best for a particular child inevitably involves judgments about the hierarchy of and trade-offs between often competing values. Presumably with the goal of providing greater guidance and more consistent decisionmaking, many states now list criteria that a court is 14 asked to use in applying the best-interests standard. But, as Professor Bartlett demonstrates, “the[] criteria encompass every factor potentially relevant to a child’s welfare,” and “most of the criteria are themselves open-ended and subjective.” In other words, the criteria do little to guide and regulate judicial 15 decisionmaking. I would stand by my general critique of using an indeterminate standard for child-custody adjudication. Using a best-interests standard poses fundamental questions of fairness. It risks violating the “fundamental precept that like cases 16 should be decided alike.” It largely removes the special burden of justification that is characteristic of adjudication, and it involves using the judicial process in a way that is quite uncharacteristic of traditional adjudication: It requires predictions of what will happen in the future and not determinations of what happened in the past. A broad “person-oriented” standard risks that a judge’s
11. See Elizabeth S. Scott & Robert E. Emery, Gender Politics and Child Custody: The Puzzling Persistence of the Best-Interests Standard, 77 LAW & CONTEMP. PROBS., no.1, 2014 at 69, 91–95. 12. Id. at 92. 13. Kimberly Emery and Robert Emery point out that “expert” psychological evaluations by mental-health professionals are now increasingly used as part of the “dispute-resolution funnel” to encourage settlement as part of the dispute-resolution process before a judge is asked to adjudicate a conflict. See Kimberly C. Emery & Robert E. Emery, Who Knows What is Best for Children? Honoring Agreements and Contracts Between Parents Who Live Apart, 77 LAW & CONTEMP. PROBS., no. 1, 2014 at 151, 159–61. 14. Katharine T. Bartlett, Prioritizing Past Caretaking in Child-Custody Decisionmaking, 77 LAW & CONTEMP. PROBS., no. 1, 2013 at 29. 15. Id. 16. See Mnookin, supra note 1, at 263.

252

LAW AND CONTEMPORARY PROBLEMS

[Vol. 77:249

decision will be based on “unarticulated (perhaps even unconscious) predictions 17 and preferences that could be questioned if expressed.” In this regard, I find the distinction offered by Daniel Kahneman in his 18 recent book Thinking, Fast and Slow relevant both to my deconstruction of the best-interests standard and my critique of its use as an adjudicatory standard. 19 Kahneman’s core idea is that human beings think in two very different ways. One cognitive system (what he calls “System 1” or thinking fast) is fast, instinctive, almost automatic; it draws on emotions and can be subconscious and 20 stereotypic. A second way of thinking, (“System 2”) is slower, more 21 deliberative, and logical. It is effortful, conscious, and calculating. The decision-analytic framework I used in deconstructing the best-interests standard is quintessentially a system-two method of thinking, a paradigmatic 22 example of what Kahneman would characterize as thinking slow. And the product of my decision-analytic framework—the two-prong test—would also be thinking slow, were judges to ever engage in it. By describing the best-interests standard in these terms, I exposed its indeterminacy. But in so doing, I never thought I was actually describing how judges make custody decisions. Indeed, a primary problem of an indeterminate custody standard is that it invites judges to make decisions by “thinking fast”—relying primarily on the basis of intuition, 23 visceral reaction, and gut feeling about to the parties before them. With respect to indeterminacy, I should underscore an important limitation on the scope of my claim: Not all custody disputes are plagued by indeterminacy. Some custody disputes are “easy” to decide because there is a 24 social consensus about “what is very bad.” If one placement exposes a child to substantial short-run risks of serious harm, especially irreversible harm, while 25 another placement poses no such risk, the choice is clear-cut and easy. There is no need to make longer-term predictions or more complicated evaluations of what is likely to happen to the child’s long-run development. However, “to be easy, a case must involve only one claimant who is well known to the child and 26 whose conduct does not endanger the child.” More typically, custody disputes involve more difficult choices because they do not fit this pattern. In many private disputes, the court must choose between two claimants who each offer

17. Id. 18. DANIEL KAHNEMAN, THINKING, FAST AND SLOW (2011). 19. Id. at 415. 20. Id. 21. Id. 22. Id. 23. As Bartlett relates, one court noted that “[o]ften trial judges who see the witnesses in a custody dispute come away with a gut feeling that one parent is a better fit than the other, though it may be difficult to explain the underlying reasons.” Bartlett, supra note 14, at 66 (emphasis added) (internal quotation marks omitted) (citing McKee v. Dicus, 785 N.W.2d 733, 738 (Iowa Ct. App. 2010)). 24. See Mnookin, supra note 1, at 261. 25. Id. 26. Id.

No. 1 2014]

CHILD CUSTODY REVISITED

253

advantages and disadvantages and neither of whom would endanger the child. In child-neglect proceedings, the existing home is typically far from optimal, but placing the child with a foster family unknown to the child can pose serious risks as well. III THE CLAIM CONCERNING TWO FUNCTIONS My second fundamental claim in my article was that child-custody adjudication can be best understood by differentiating between two different 27 judicial functions: child protection and private dispute resolution. The distinction between these two functions remains critical because the implications of indeterminacy are quite different depending on the function at 28 issue. Because the policy goals related to each function are different, the search for a single set of substantive standards and procedures to govern all custody disputes would be as unwise today as it would have been forty years ago. None of the authors would appear to disagree. The child-protection function involves the judicial enforcement of stateimposed minimum standards of parental behavior believed necessary for the protection of the child. Defining the proper scope for “child protection” poses fundamental questions concerning political and moral philosophy about the proper allocation of power and responsibility between the family and the state. The use of an indeterminate legal standard confers broad discretion and can invite inappropriate intrusion into the family. Adjudication implicates the child-protection function most commonly in juvenile-court neglect or dependency proceedings. These occur within an elaborate juvenile-court and social-welfare bureaucracy with responsibility for a foster-care system. The state may retain oversight over the child for an extended period of time—often years. An indeterminate best-interests standard gives too much discretion to social workers, probation officers, and juvenilecourt judges to inappropriately remove children from parental custody. And for those children who are removed, an indeterminate best-interests standard does too little to ensure bureaucratic accountability once such children are in foster care. The procedures and standards for private dispute resolution play a very different role. This function involves resolving disputes between private
27. Id. at 229. 28. The distinction also remains analytically useful in understanding the different strands of contemporary custody law: divorce custody proceedings, guardianship proceedings, juvenile-court neglect proceedings, and involuntary termination of parental rights as part of adoption. I demonstrate that “either function can be involved in a judicial proceeding involving the application of any of the four strands of custody law, and a single case may involve both strands.” Id. at 248–49. A recent Supreme Court decision involved a private custody dispute in the context of an adoption between a Native American biological father and adoptive parents. Adoptive Couple v. Baby Girl, 133 S. Ct. 2552 (2013).

254

LAW AND CONTEMPORARY PROBLEMS

[Vol. 77:249

individuals—typically two parents—who cannot agree on how to share responsibility for the child. The state is providing a forum for resolving conflicts that the parties have not been able to resolve on their own. Once the dispute is resolved, the judge is typically out of the picture. The state retains no broad authority over the child or the family. None of this is meant to suggest that we should be unconcerned about the use of an indeterminate standard for private dispute resolution. Its use raises important questions concerning fairness because of the breadth of the largely unreviewable discretion it gives to a judge. Moreover, an indeterminate standard makes it difficult to predict the outcome of litigation. This uncertainty might encourage more litigation and make the dispute-resolution process more protracted and costly. Neither my analysis of the two functions nor my critique of the best-interests standard was based on the Constitution. As Professor Emily Buss correctly points out, I gave the Supreme Court’s jurisprudence “only the slightest of 29 nods” in reaching my conclusions. Nor did my article focus on the constitutional dimensions of parental rights, or the outer limits of a state’s authority to intervene into the family in the exercise of its parens patriae 30 power. It pleases me that Professor Buss identifies what she sees as “the striking correlation” between my “child-focused legal prescriptions and the parent-focused developments in constitutional law” that has developed since 31 my article appeared. I can hardly claim credit for prescience. But I suspect part of the explanation lies in what I identified as the “starting point” for my policy analysis. In my article I expressed a preference for what I characterized as “family autonomy”: A starting point that “assumes that power and responsibility for children generally ought to be vested in private hands—essentially the family— 32 except in cases where government rule can be justified.” I cited Pierce v. 33 Society of Sisters in support of this position. Supreme Court decisions since my article certainly suggest that the Constitution itself would impose substantive and procedural limitations on the state’s power to remove children from
29. See Emily Buss, An Off-Label Use of Parental Rights? The Unanticipated Doctrinal Antidote for Professor Mnookin’s Diagnosis, 77 LAW & CONTEMP. PROBS., no. 1, 2014 at 1, 2. 30. As Professor Buss notes, in 1975 there were only a handful of Supreme Court cases that touched upon the constitutional basis for parental rights. Id. at 2. She identifies Meyer, Pierce, Prince, Yoder, and Stanley. Id. at 3. The first edition of my casebook, which appeared in 1978, did, among other things, focus on the constitutional dimensions of the allocation of power and responsibility for children. See ROBERT H. MNOOKIN, CHILD, FAMILY AND STATE: PROBLEMS AND MATERIALS ON CHILDREN AND THE LAW (1st ed. 1978). Now in its sixth edition, the amount of constitutional material in the casebook has expanded considerably over the years. Compare MNOOKIN & WEISBERG, supra note 2, with MNOOKIN, supra. 31. See Buss, supra note 29, at 2. 32. See Mnookin, supra note 1, at 266. The other two starting points were (1) state paternalism, which assumes that the state has primary responsibility for children and ought to exercise full control over their lives, except where delegation of the family is justified, and (2) agnosticism, which rests on no preference and instead approaches individual policy issues on their own merits. 33. Id. at 266–67 & n,188 (citing Pierce v. Society of Sisters, 268 U.S. 510, 534–35 (1925)).

No. 1 2014]
34

CHILD CUSTODY REVISITED

255

parental custody. Indeed, with respect to the child-protection function, Professor Buss hints that the requirement of “reasonable efforts” before 35 removal might have a constitutional underpinning. With respect to private dispute resolution, as Buss points out there is no constitutional impediment to my stated preference for a psychological parent’s claim over that of a genetic 36 parent who lacks any substantial prior connection to the child’s life. IV CHANGING GENDER ROLES AND CUSTODY STANDARDS Claims relating to gender roles and gender equity have long been central to family-law policy debates. In my article I traced the prior evolution of custody standards from a paternal preference, through fault-based standards, to a 37 maternal preference. Indeed, in 1975 I reported that the maternal preference was rapidly “being displaced by a formal insistence on a neutral application of 38 the best-interests standard.” I attributed the trend to “no fault divorce, the changing social conception of the appropriate sex roles, and the women’s 39 movement.” The contributions to this issue of Law and Contemporary Problems suggest that today the indeterminate best-interests standard is still the dominant 40 standard. But during the intervening years, advocates worked with limited
34. See Troxel v. Granville, 530 U.S. 57 (2000) (holding that in light of constitutional right of parents to rear their children, Washington state allowing any third party to petition state courts for child-visitiation rights over parental objections is unconstitutional); Santosky v. Kramer, 455 U.S. 745 (1982) (holding that the Due Process Clause of the Fourteenth Amendment requires more than a preponderance-of-the-evidence standard in any procedure in which a state seeks to terminate the rights of parents in light of evidence that the child is neglected). 35. See Buss, supra note 29, at 19. 36. Id. at 27. Relevant in this regard is the recent Supreme Court decision rejecting a statutory claim brought under the federal Indian Child Welfare Act of 1978 (ICWA), 25 U.S.C. §§ 1901–1963 (2006), by a biological father who sought to block an adoption. See Adoptive Couple v. Baby Girl, 133 S. Ct. 2552 (2013). In Adoptive Couple v. Baby Girl, the South Carolina Supreme Court had read the ICWA to deny an adoption that would have otherwise have been valid under state law on the basis of an objection by the Native American biological father, who had had no previous role in the child’s life. Id. at 2556. In a majority opinion for the Court, after “assuming for sake of argument” that the biological father was a “parent” under the ICWA, Justice Alito chose to read the statutory provisions narrowly, and therefore rejected the father’s claim. Id. at 2557. In his dissent, Justice Scalia wrote, The Court’s opinion, it seems to me, needlessly demeans the rights of parenthood. It has been the constant practice of the common law to respect the entitlement of those who bring a child into the world to raise the child. We do not inquire whether leaving the child with his parents is ‘in the best interests of the child.’ It sometimes is not; he would be better off raised by someone else. But parents have their rights, no less than children do. This father wants to raise his daughter, and the statute amply protects his right to do so. Id. at 2572 (Scalia, J., dissenting). 37. See Mnookin, supra note 1. 38. Id. at 235. 39. Id. at 235–36. 40. As earlier noted, many states now identify factors that courts are to take into account in determining best interests, but Professor Bartlett demonstrates these changes do little to make the standard more determinate. Bartlett, supra note 14, at 30.

256

LAW AND CONTEMPORARY PROBLEMS

[Vol. 77:249

success to displace the best-interests standard with new custody standards, which were much more determinate. In courts and state legislatures a gender 41 war over custody standards has marked the last forty years. The main 42 contenders have been (1) presumptions in favor of joint custody and (2) presumptions that would base custody on how parents allocated child-rearing 43 responsibilities before they separated. Fathers’ rights groups have relentlessly pressed for joint-custody presumptions, and women’s groups have vigilantly 44 resisted. The approximation standard has been vigorously opposed by fathers’ 45 groups, while supported by the elite American Law Institute (ALI). One of the puzzles explored in this issue of Law and Contemporary Problems is why 46 women’s groups have not championed this alternative. To put the battles over divorce-custody standards in perspective, I have found it helpful to look at the extent to which the division of responsibilities between fathers and mothers in two-parent households has evolved. A. Evolving Gender Roles and the Division of Family Responsibilities Within a household with children and two parents living together, there are three major functions that must be performed: (1) providing for the economic support of the family, (2) providing day-to-day care for the children, and (3) performing the household chores to keep the family enterprise going. With respect to all three functions, there have been dramatic changes. Forty years ago, the “traditional” middle-class American arrangement involved a considerable degree of gender-role differentiation: The father was the breadwinner responsible for working outside the home to earn income for the family, and the mother was primarily responsible for child rearing and housework. She had no paid work in the labor market outside the home. After World War II and through 1960 this traditional allocation was clearly dominant. By the time my article appeared this traditional allocation was already breaking down. Today it is the exception and not the rule. In 1960, only 27.6% of married women with children under eighteen participated in the labor

41. See Scott & Emery, supra note 11, at 76–83. 42. See id. at 77–80; see also Bartlett, supra note 14, at 30. 43. I refer here to both the primary-parent standard and the approximation standard. See Richard Neely, The Primary Caretaker Parent Rule: Child Custody and the Dynamics of Greed, 3 YALE L. & POL’Y REV. 168, 169 (1984) (describing the primary-parent standard); Elizabeth S. Scott, Pluralism, Parental Preference and Child Custody, 80 CALIF. L. REV. 615, 617 (1992) (describing the approximation standard); see also PRINCIPLES OF THE LAW OF FAMILY DISSOLUTION: ANALYSIS AND RECOMMENDATIONS § 2.08 (2002); Bartlett, supra note 14, at 32; Scott & Emery, supra note 11, at 76. Although very similar, the two are not identical. A court applying the primary-parent standard would award custody to a parent who had done more, such that he or she was “primary.” A court applying the approximation standard would allocate time after divorce in a way that approximated how the parents shared responsibilities during the marriage. 44. See Susan Dominus, The Fathers’ Crusade, N.Y. TIMES MAG., May 8, 2005, at 26, 50. 45. PRINCIPLES OF THE LAW OF FAMILY DISSOLUTION: ANALYSIS AND RECOMMENDATIONS § 2.08 (2002). 46. See Scott & Emery, supra note 11, at 70–71.

No. 1 2014]
47

CHILD CUSTODY REVIS U SITED

257

force. By 1975, when my article was written that perce n entage had risen to 48 012 eached nearly 70%. y 44.9%; by 20 it had re The incr reases are especially str e riking for w women with young child dren. In 1960, the pa articipation rate for marr ried women w children under six y with n years old was only 18 8.6%; by the time my ar e rticle appear in 1975 t red that percent tage had 49 increased to 36.7%; by 2013 it was 63.9%. o 2

Source: PEW RESEARCH CTR., MODERN PARENTHO : ROLES H OOD CONVERG AS THEY BALANCE WORK AND FAMILY 1 0 (2013). GE A

OF F

MOMS

AND

DADS

A very recent Pew Research re r eport demon nstrates, as the charts f from the pending more time on c report excer rpted below show, that fathers are sp f childcare 50 and on hous sework. Ba ased on time-allocation s surveys goin back to 19 ng 965, Pew
47. See Elizabeth Waldm man, Labor Forc Statistics from a Family Per ce m rspective, MONT THLY LAB. REV., De 1983, at 16, 18. ec. 48. The pr recise number was 68.5%. See id.; Letter from Steven Hipp Economist, Bureau of w e m ple, Labor Statistics Dep’t of Lab s, bor, to author (June 10, 2013 ) (on file with author); see al Arleen lso Leibowitz & Jacob Alex Klerm man, Explaining Changes in Mar rried Mothers’ E Employment Ove Time, 32 er Y 5) l ers labor force DEMOGRAPHY 365, 365 (1995 (“Among all married mothe of children under age 6, l participation (L LFP) rates doubl between 197 and 1990 . . . f led 70 from 30% to 59% %.”). 49. See Ne Release, U.S Bureau of La ews S. abor Statistics, E Employment Ch haracteristics of Familes— f 2013, at 2 (Apr. 25, 2014), avail . lable at http://ww ww.bls.gov/new ws.release/pdf/fam mee.pdf. 50. See PEW RESEARCH CTR., MODERN PARENTHOOD: ROLES OF MOM AND DADS CONVERGE MS AS THEY BALA ANCE WORK AND FAMILY 27 (2 D 2013).

258

LAW AND CONTEMPORARY PROBLEMS

[Vol. 77:249
51

concluded, “Fathers have nearly tripled their time with children since 1965.” A large gender gap in time spent with children remains: “[M]others spend about 52 twice as much time with their children as fathers do.” (In 2011, mothers spent 53 about 13.5 hours per week, compared with 7.3 hours for fathers). There is a similar trend for housework. “Fathers’ time spent doing household chores has more than doubled since 1965 (from an average of about 54 four hours per week to about 10 hours). Mothers’ time doing housework has decreased significantly over the same timeframe (from 32 hours per week to 55 18). The Pew report concludes,
Fathers now spend more time engaged in housework and child care than they did half a century ago . . . . Fathers have by no means caught up to mothers in terms of time spent caring for children and doing housework chores, but there has been some 56 gender convergence in the way they divide their time between work and home.

51. 52. 53. 54. 55. 56.

Id. See id. at 28. Id. Id. at 32. Id. at 33. Id. at 3.

No. 1 2014]

CHILD CUSTODY REVISITED

259

Source: PEW RESEARCH CTR., MODERN PARENTHOOD: ROLES OF MOMS CONVERGE AS THEY BALANCE WORK AND FAMILY 27–28 (2013).

AND

DADS

These trends provide fuel for both sides of the gender wars that plague divorce policy. On the one hand, fathers’ groups point to the fact that today 57 married women work in proportions nearly equal to men. They can also claim 58 that during marriage fathers are now spending more time with the children. But women’s advocates can show that during marriage mothers earn less than
57. See Jocelyn Elise Crowley, Adopting “Equality Tools” from the Toolboxes of their Predecessors: The Fathers’ Rights Movement in the United States, in FATHERS’ RIGHTS ACTIVISM AND LAW REFORM IN COMPARATIVE CONTEXT 79, 86–87 (Richard Collier & Sally Sheldon eds., 2006). 58. Id.

260

LAW AND CONTEMPORARY PROBLEMS
59

[Vol. 77:249

fathers and have far greater child-rearing responsibilities. The legal standards governing both the financial and custodial arrangements should reflect these realities. B. The Joint-Custody Presumption I have long opposed presumptions in favor of joint physical custody. About three years after my child-custody adjudication article was published, I wrote Bargaining in the Shadow of the Law: The Case of Divorce with Lewis 60 Kornhauser. Our core insight was that the legal standard used by courts to 61 adjudicate disputes also provides a backdrop for out-of-court bargaining. We suggested that a joint-custody standard created a dilemma: Although it had some good characteristics as a background rule for private ordering outside of court it was inappropriate as a standard for a judge in court to impose. The advantages of the joint-custody rule from the perspective of bargaining are that it does not disadvantage the relatively risk-averse parent, reduces the scope for strategic behavior, and imposes lower transaction costs. Nevertheless, it would probably be disastrous to impose joint custody on the parties that could not agree to it themselves, because joint custody normally requires a very high degree of parental cooperation. Without such cooperation, the substantial contact both parents would have with the child, and necessarily with each other, would create endless possibilities for antagonism between the parents, with predictably detrimental effects on the child’s well-being. Ordering joint custody 62 might be very much like carrying out Solomon’s threat to cut the child in half. In Dividing the Child, Eleanor Maccoby and I rejected a presumption in favor of joint physical custody because of our deep concern that the use of joint custody in cases where there was substantial parental conflict would harm 63 children.
We do not think it good for children to feel caught in the middle of parental conflict, and in those cases where the parents are involved in bitter dispute, we believe a presumption for joint custody would do harm. Our study suggests that in a number of cases in which families today adopt joint physical custody, there has been substantial legal conflict. To the extent that this custody arrangement is the result of encouragement by mediators, or judges for that matter, we think it is unwise. We wish to note, however, that joint custody can work very well when parents are able to cooperate. Thus we are by no means recommending that joint custody be denied to 64 parents who want to try it.

I found fascinating the article by Bruce Smyth, Richard Chisholm, Bryan Rodgers, and Vu Son, reporting on their valuable research on the Australian

59. See Lyn Craig, Does Father Care Mean Fathers Share?: A Comparison of How Mothers and Fathers in Intact Families Spend Time with Children, 20 GENDER & SOC’Y 256, 270 (2006). 60. See Mnookin & Kornhauser, supra note 2. 61. Id. 62. Id. at 980 (internal citations omitted). 63. See Mnookin, supra note 1, at 150. 64. See MACCOBY & MNOOKIN, supra note 2, at 284–85 (1992).

No. 1 2014]

CHILD CUSTODY REVISITED
65

261

experience with legislation encouraging “shared-time parenting.” Their study 66 is intriguing and their empirical findings are important and very valuable. I would offer the following five observations. First, the Australian legislation 67 was the result of lobbying by fathers’ groups. Second, the Australian law on its face does not create a presumption in favor of joint physical custody. Instead, the Australian presumption of “equal, shared parental responsibility” is akin to a presumption in favor of what in the United States would be characterized as 68 joint legal (as opposed to physical) custody. Third, since the reforms of 2006, shared-time parenting is not being imposed frequently by courts in contested cases. Figure 4 in Smyth, Chisholm, Rodgers, and Son’s article suggests that in the most recent time period only three percent of litigated cases resulted in court orders of equal time. As Smyth, Chisholm, Rodgers, and Son report this decline “seems appropriate . . . because fully adjudicated cases tend to exhibit 69 entrenched, high levels of parental conflict.” Fourth, equal-time arrangements were the result of negotiated settlements in between fourteen and nineteen 70 percent of litigated cases. Smyth, Chisholm, Rodgers, and Son did not find any evidence, however, that the new law was having a substantial effect in this 71 regard. Fifth, “shared parenting” is being adopted most frequently by “cooperative parents,” typically middle class persons with elementary school– 72 aged children. In our California study we also found that shared parenting was rarely chosen by divorcing parents with either very young children or teenage 73 children. C. The Approximation Standard Were I in a state legislature I would vote in favor of the ALI proposals and prefer the approximation standard to the indeterminate best-interests standard for the resolution of private disputes. What explains the “puzzling persistence”

65. See Belinda Fehlberg et al., Legislating for Shared Time Parenting After Separation: A Research Review, 25 INT’L J.L. POL’Y & FAM. 318 (2011); accord Bruce Smyth, Richard Chisholm, Bryan Rodgers & Vu Son, Legislating for Shared-Time Parenting After Parental Separation: Insights from Australia?, 77 LAW & CONTEMP. PROBS., no. 1, 2014 at 109. 66. Fehlberg also suggests that Australia provides a broad range of family-support services to divorcing couples that I would hope American states might consider. Fehlberg et al., supra note 65, at 326–27. . 67 See Patrick Parkinson, Changing Policies Regarding Separated Fathers in Australia, in THE ROLE OF THE FATHER IN CHILD DEVELOPMENT 578, 580 (Michael E. Lamb ed., 5th ed. 2010). 68. Janet R. Jeske, Issues in Joint Custody & Shared Parenting: Lessons from Australia, BENCH & B. MINN., Dec. 13, 2011, at 20, 21. 69. Smyth et al., supra note 65, at 141. 70. Smyth et al., supra note 65, at 137 fig.5. 71. In our study, there were 166 cases in which the decree provided for joint custody. But nearly all of these were the result of a negotiated or mediated resolution. Of the 933 cases in our study, a judge decided only fourteen, which is 1.5%. Of these adjudicated cases that a judge decided, only four resulted in a joint-custody award. 72. See MACCOBY & MNOOKIN, supra note 2, at 239. 73. Id. at 247.

262

LAW AND CONTEMPORARY PROBLEMS
74

[Vol. 77:249

of the best-interests standard? Scott and Emery suggest, correctly in my view, that the best-interests standard offers lawmakers safe shelter in circumstances when either of the two more determinate alternatives—joint custody or the approximation standard— are sure to provoke hostile fire from either mothers’ advocates or fathers’ 75 groups. But a puzzle remains. What explains the failure of women’s advocates to more actively champion the approximation standard given that, at least under current parental-role allocation, it more often than not results in the 76 mother being recognized as the primary custodial parent? Part of the answer lies in the data about the continued evolution of gender roles during marriage and the fact that the best-interests standard is more “neutral.” In my article, I underscored that “[s]ex-based rules have been tried historically and are now being discarded (correctly in my view) because they reflect value judgments and sexual stereotypes that our society is in the process 77 of rejecting.” The approximation standard is gender neutral on its face. It does not explicitly provide for a maternal preference. Those fathers who during marriage spent considerable time with the children would have that prior commitment honored. Nonetheless, given the data concerning the actual allocation of childrearing responsibilities in most marriages, a legal realist might claim that the 78 approximation standard amounts to a maternal preference “in disguise.” Although dressed up in gender-neutral garb, in actual operation it would typically result in something akin to a maternal preference. This explains the vehement political opposition of fathers’ groups. That the approximation standard can be seen as a disguised maternal preference also explains the ambivalence some feminists perhaps feel. Many women are deeply troubled by the extent to which gender-role differentiation persists in our society. The roles of women within the household—especially with regard to parenting—and opportunities in the labor market are connected. During marriage the mother in most American families today has much greater 79 child-rearing responsibility than the father. Many feminists aspire for a world in which fathers and mothers share child-rearing responsibilities much more 80 equally. Reinforcing the current reality—which, as the data shows, is a long way from equal distribution—sacrifices that dream and might perpetuate the stereotype that mothers should be primarily responsible for childrearing.
74. See Scott & Emery, supra note 11, at 69. 75. Id. at 82–83. 76. Of course, should gender roles continue to evolve, resulting in more equal sharing of parental responsibilities, the approximation standard would itself evolve away from a de facto maternal preference. 77. Mnookin, supra note 1, at 284. 78. See MACCOBY & MNOOKIN, supra note 2, at 284–85. 79. Id. 80. Catherine I. Bolzendahi & Daniel J. Myers, Feminist Attitudes and Support for Gender Equality: Opinion Change in Women and Men, 1974–1998, 83 SOC. FORCES 759, 769 (2004).

No. 1 2014]

CHILD CUSTODY REVISITED

263

There is also a fairness argument. After all, as David Chambers argued, there was no evidence, other than for very young children, that after divorce a fit father could not learn to be as good a primary parent as a mother who had 81 previously occupied that role. Although there is some power to this argument, as Chambers recognized, it does not make the approximation standard unfair to 82 fathers. During the marriage, a father who has played a substantial role in child rearing would at the time of divorce bargain against a backdrop that considered that reality. As Eleanor Maccoby and I pointed out,
A woman who has served as the primary parent, after all, has already largely developed and demonstrated the skills to care for the child on an everyday basis. While her post-divorce role as custodial 83 parent would require change, she has much less to learn in most cases than the father.

I remain of the view that “her experience as well as his inexperience strikes us 84 as relevant to the custodial decision.” Moreover, during a time that gender roles are evolving, the approximation standard provides a transitional rule that appeals to my pragmatic instincts. D. The Impact of Law on Social Change Let me add a few words explaining why I believe that even if widely adopted as a replacement to the best-interests standard, the ALI standard would not in actual operation lead to a distribution of outcomes that would be substantially different than presently achieved under the best-interests standard. I am of the view that changes in custody standards are unlikely to have a significant impact on how parents allocate child-rearing responsibilities either before or after divorce. Law more typically reflects and perhaps reinforces the underlying social reality. Changing gender roles poses a macrochallenge that is too 85 formidable for law to affect substantially. Claims relating to gender equality and gender equity have been central to family-law policy debates for decades. As Eleanor Maccoby and I found in our empirical study of California divorce, gender-role differentiation with respect to parenting that follows divorce rests largely on differential roles that parents 86 occupied before the separation. Unless family law can modify predivorce roles, I remain doubtful that custody standards can have much greater impact on the postdivorce division of responsibilities: “[M]ost divorcing families couples would still end up allocating 87 primary child-rearing responsibilities to mothers.” I think this would be true under a best-interests standard. I also think it would be true under the
81. See David L. Chambers, Rethinking the Substantive Rules for Custody Disputes in Divorce, 83 MICH. L. REV. 477, 511 (1984). 82. Id. 83. MACCOBY & MNOOKIN, supra note 2, at 283–84. 84. Id. 85. Id. at 280. 86. See MACCOBY & MNOOKIN, supra note 2, at 279. 87. Id.

264

LAW AND CONTEMPORARY PROBLEMS

[Vol. 77:249

approximation standard. Indeed, Smyth, Chisholm, Rodgers, and Son’s article suggests that it remains true in Australia, notwithstanding Australia’s 88 presumption in favor of shared parental responsibility. This is not to say custody standards are of no practical importance. A standard might affect outcomes in some cases, and a custody standard might have some impact on the mother’s bargaining power with respect to economic issues. The approximation standard might increase the bargaining position of women and decrease the risk that fathers could strategically use a custody claim under the vague best-interests standard as a bargaining chip to reduce economic 89 support. But as we found in Dividing the Child, even under the best-interests standard, at least where there are reasonably precise support guidelines, there is no evidence that a father’s strategic use of custody conflict leads to less financial 90 support for women. E. Custody Conflicts Involving Same-Sex Households Nancy Polikoff correctly points out that in my article I did not anticipate the way “legal” and “cultural” changes might disrupt the assumed definition of 91 natural parent.” There is not a word in my article about custody disputes involving same-sex couples and I did not anticipate the remarkable changes in 92 93 public attitudes towards same-sex adoption and same-sex marriage. Nor is there a discussion of conflicts that can arise today where a child has been conceived through new assisted reproductive technologies or carried to term through a surrogacy arrangement. Since 1975 extraordinary technological developments have created radically new possibilities in terms of human 94 95 reproduction. In vitro fertilization did not exist in 1975. These developments
88. See Smyth et al., supra note 65, at 137 fig.5, 138. 89. See MACCOBY & MNOOKIN, supra note 2, at 284 (discussing the primary-parent standard in this regard). 90. Id. at 11. 91. See Nancy D. Polikoff, From Third Parties to Parents: The Case of Lesbian Couples and Their Children, 77 LAW & CONTEMP. PROBS., no. 1, 2014 at 195, 195–96. 92. Surveys suggest that the percentage of Americans now supporting adoption by same-sex couples has more than doubled in the last eighteen years. See Susan Page, Attitudes Toward Gays Changing Fast, Poll Finds, USA TODAY, Dec. 6, 2012, at A1. 93. In 1996 when the Defense of Marriage Act was signed, it was estimated that only one-quarter of the American public supported same sex marriage. Public Opinion of Same-Sex Marriage in the United States, WIKIPEDIA, http://en.wikipedia.org/wiki/Public_opinion_of_same-sex_marriage_in_the_ United_States (last modified June 6, 2014). A May 9, 2013 Washington Post–ABC News poll found that 55% of Americans support same-sex marriage, 40% oppose, and 5% are undecided. The same poll suggests that one-third of those who now support same-sex marriage say they once held the opposite view and have changed their opinion. GARY LANGER, POLL FINDS MAJORITY ACCEPTANCE OF GAYS FROM THE B-BALL COURT TO THE BOY SCOUTS (2013), available at http://www.langerresearch.com/ uploads/1144a19GayRights.pdf. Media reports covering same-sex parenting have become increasingly mainstream. For example, a recent Vermont Public Radio audio report detailed a same-sex couple’s efforts to have the nonbiological mother formally adopt their child. See Abigail Mnookin, Second Parent Adoption, VT. PUB. RADIO (Aug. 8, 2013, 3:43 PM), http://digital.vpr.net/post/mnookin-secondparent-adoption. 94. See Polikoff, supra note 91, at 209–11.

No. 1 2014]

CHILD CUSTODY REVISITED

265

have brought to court disputes that do not fit comfortably or neatly into existing 96 legal categories. Polikoff’s article is centrally concerned with the sort of custody dispute that might arise between two mothers who while living in a lesbian relationship have decided “to bring a child into their relationship and to raise that child together, 97 with each of them acting as a parent.” How should the law respond? Let me first deal with the easy case. Suppose the second mom who is not biologically related to the child, with the consent of the biological mother, goes through a state-sanctioned adoption process and becomes an adoptive parent. As Polikoff presumably recognizes, a court applying the test I proposed in my article would not differentiate the rights of an adoptive parent from those of a 98 biological or “natural” parent. In other words, if the second mom had adopted the child she would in my view stand on equal legal footing with the biological mother. But suppose the second mom does not adopt the child. Under the approach suggested in my article, what would her rights be? For private dispute resolution, I proposed three intermediate rules that I claimed were not plagued by indeterminacy: (1) custody should not be awarded to a claimant whose limitations or conduct would endanger the health of a child under the minimum standards of child protection, (2) the court should prefer a psychological parent over any claimant, including a “natural” or biological parent, who, from the child’s perspective, is not a psychological parent, and (3) subject to the first two rules, “natural parents” (by which I meant biological parents) should be 99 preferred over others. The second mom would surely qualify as a “psychological” parent. For some disputes this rule would vindicate the relational interests of the second mom. She would prevail in a custody fight with a nonparent third party—perhaps a collateral relative of some sort—who is not a psychological parent. It would also cover a case in which the biological mom had somehow become a stranger to the child. Polikoff’s concern, however, is my response to what I see as a harder case— a custody fight between two moms, both of whom are psychological parents, but only one of whom is a biological or adoptive parent. The application of my third intermediate rule, with its preference for a “natural parent” would seem to place a biological mother who is also a psychological parent on a superior footing to a second mom who is not biologically related to the child but is a psychological parent.

95. Id. at 211–19. 96. See, e.g., Johnson v. Calvert, 851 P.2d 776 (Cal. 1993); In re Baby M, 537 A.2d 1227 (N.J. 1988). 97. See Polikoff, supra note 91, at 196. 98. See id. at 195 (indicating that “adoptive” and “natural” parents are both included in the Uniform Parentage Act). 99. See Mnookin, supra note 1, at 282–83.

266

LAW AND CONTEMPORARY PROBLEMS

[Vol. 77:249

I must confess I did not have a case like this in mind when I proposed my 100 third intermediate rule that “natural parents should be preferred over others.” I justified this third standard by claiming that
[i]n a dispute where the natural parent poses no danger to the child’s physical health and is viewed by the child as a psychological parent, a preference for that natural parent over a third party gives expression to broadly shared social values about parental responsibilities for the welfare of their offspring and reflects the importance 101 to the child of a sense of lineage.

But when the two parents have raised the child together since infancy, I am unsure there is “a broadly shared social value” that would support the 102 preference. I am sympathetic to Polikoff’s claim that the lesbian partner who has raised the child since birth should be treated as a parent for purposes of the private dispute–settlement function—a parent on equal footing with a biological 103 or “natural” parent. When the two moms are married at the time of the child’s arrival in the household, this would not seem much of a doctrinal stretch. Even if the moms are not married, I suspect some courts will develop a doctrine of de facto adoption that could be applied to Polikoff’s case. For me the most difficult case would be dispute in a state where both marriage and adoption were available to the second mom when the child arrived but neither was pursued. In such circumstances, I am not persuaded the de facto parent should stand on an equal footing with the biological parent. I am not offended by the notion that equal footing should require either marriage or a formal legal adoption by the second mom, assuming state law provides these options. I would apply the same standard to an unmarried heterosexual couple that had raised a child since birth if the child is biologically related to only one parent and has not been adopted by the other. F. Process Implications In my article I suggested that a “primary implication of the indeterminacy of the best-interests standard is that the legal process should encourage the parties 104 themselves to work out their own resolution.” Since the article, there has been extraordinary expansion of the use of a variety of mediation and other methods to facilitate parental negotiation. Scott and Emery are entirely correct when they point out that the indeterminacy of the best-interests standard is an 105 important rationale for mediation. As I and others have noted, the traditional adversarial process used to resolve custody conflicts can in itself often make 106 things worse for the child. It can inflame conflict, and further strain or even destroy a conflicted coparenting relationship. I entirely endorse the notion that
100. 101. 102. 103. 104. 105. 106. Id. Id. Id. See Polikoff, supra note 91, at 219–20. See Mnookin, supra note 1, at 292. See Scott & Emery, supra note 11, at 108. Mnookin & Kornhauser, supra note 2, at 956.

No. 1 2014]

CHILD CUSTODY REVISITED

267

there should be judicial deference to parental agreements. In this regard I 107 completely agree with the ALI’s recommendation. I am moreover very sympathetic with Kimberly Emery and Robert Emery’s suggestion that we eliminate judicial review of consensual parental agreements concerning custody 108 altogether. In her fascinating article, Jana Singer explores “the close connection between changes in substantive child-custody doctrine and changes in custody 109 dispute–resolution processes over the past thirty years.” She is surely correct that the “unmediated best-interests standard, and the ensuing rejection of the sole-custody paradigm, paved the way for the shift from adversarial to 110 nonadversarial resolution in divorce-related parenting disputes.” Indeed, if the function is to facilitate private dispute settlement, then members of the legal system should devote energy to thinking about processes that might encourage and facilitate the resolution of disputes without the need for formal adjudication. What I found especially intriguing was Singer’s claim that the shift from adversarial to nonadversarial dispute resolution has itself had an effect on the substantive standards—to a degree it has “delegalized custody decisionmaking 111 and has diminished the importance of custody as a legal concept.” A striking 112 example relates to the contemporary emphasis on the parenting plan. To frame the parental decision in terms of a parental-planning process—of how to allocate responsibilities for childrearing—rather than a decision about who has custody and who has visitation rights is surely constructive. I share Singer’s view that “collaborative lawyering” also represents a process innovation that probably diminishes the importance of the formal legal custody norms in 113 framing and resolving family disputes.

107. PRINCIPLES OF THE LAW OF FAMILY DISSOLUTION: ANALYSIS AND RECOMMENDATIONS § 2.06 (2002). 108. Emery & Emery, supra note 13, at 168–69. Lewis Kornhauser and I examined the possible justifications for requiring judicial proceedings in cases where there was no dispute and largely found them wanting. See Mnookin & Kornhauser, supra note 2, at 992–96. 109. Jana B. Singer, Bargaining in the Shadow of the Best-Interests Standard: The Close Connection Between Substance and Process in Resolving Divorce-Related Parenting Disputes, 77 LAW & CONTEMP. PROBS., no. 1, 2014 at 177. 110. Id. 111. Id. at 178. 112. See PRINCIPLES OF THE LAW OF FAMILY DISSOLUTION: ANALYSIS AND RECOMMENDATIONS § 2.09(2) (2002); see also MO. REV. STAT. § 452.375(9) (2000); MONT. CODE ANN. § 40-4-234(1) (2000); WASH. REV. CODE § 26.09.181(1) (2001); Carl E. Schneider, On the Duties and Rights of Parents, 81 VA. L. REV. 2477, 2485–86 (1995); Elizabeth S. Scott, Rehabilitating Liberalism in Modern Divorce Law, 1994 UTAH L. REV. 687, 720; Sally Burnett Sharp, Modification of Agreement-Based Custody Decrees: Unitary or Dual Standard?, 68 VA. L. REV. 1263, 1280 (1982). 113. Singer, supra note 109, at 190–91.

268

LAW AND CONTEMPORARY PROBLEMS

[Vol. 77:249

V THE CHILD-PROTECTION FUNCTION I learned a great deal from Clare Huntington’s wonderful article, The Child114 Welfare System and the Limits of Determinacy. Huntington describes both the legislative overhaul of the child-welfare system—particularly with respect to federal legislation—that has occurred since my article and what is known of its 115 present operation. She encapsulates her findings as follows:
To a remarkable degree, the reforms Mnookin proposed to the child-welfare system are what Congress and the states adopted in the following two decades. And yet reading Mnookin’s article is also a Groundhog Day experience. The problems he described with the child-welfare system nearly forty years ago sound all too familiar 116 today.

To address the deficiencies of the foster-care system, I proposed two core reforms. First, I proposed a much stricter standard for removing a child from 117 parental custody. I suggested that removal only be allowed (1) if there is “an 118 immediate and substantial danger to the child’s health” and (2) if there were “no reasonable means” acceptable to the parents by which the state could 119 protect the child’s health without removal. Second, for those children that were removed I suggested the use of time 120 limits. During an initial period the state should be required to make 121 reunification efforts. Then, after a fixed period of time, if notwithstanding these reunification efforts the child could not be safely returned to his home, the state should focus its efforts on permanency planning, preferably through 122 adoption or guardianship. The Adoption Assistance and Child Welfare Act of 1980 (AACWA) had as its stated goals protecting children within their homes and, for children that had to be removed more promptly, taking action to return them to parental 123 custody. As Huntington points out, the federal legislation requires states to develop somewhat more focused standards for removal and significantly authorized funds both for “preservation services” and for “reunification 124 services.” As Huntington documents, the AACWA did not succeed in 125 She also shows, however, that reducing the foster-care population. notwithstanding the new federal mandates, the states often provided little in the
114. Clare Huntington, The Child-Welfare System and the Limits of Determinacy, 77 LAW & CONTEMP. PROBS., no. 1, 2014 at 221. 115. Id. at 222–29. 116. See id. at 221. 117. See Mnookin, supra note 1, at 278. 118. Id. 119. Id. 120. Id. at 280. 121. Id. 122. Id. 123. Pub. L. No. 96-272, 94 Stat. 500 (codified as amended in scattered sections of 42 U.S.C). 124. Huntington, supra note 114, at 227. 125. Id. at 228.

No. 1 2014]

CHILD CUSTODY REVISITED

269
126

way of preservation services and courts did little to provide for accountability. 127 The Adoption and Safe Families Act of 1997 (ASFA) shifted the focus away from reducing the number of removals and getting children back home 128 more promptly and towards stimulating more timely permanency planning. As Huntington notes, the ASFA set a time limit on family-reunification efforts and conditioned federal funds on permanency planning and substantial state efforts to terminate parental rights for children who had been in foster care for 129 130 more than a given period of time. It also created subsidies for adoption. As a result of these congressional actions all states now require child131 welfare agencies to provide preservation services. But Huntington reports neither the states nor Congress have invested much by way of resources in these 132 services. It is troubling to learn that evaluations of the impact suggest that these services might not have much effect on the maltreatment rates for children. It is nonetheless stunning that Congress allocates twelve times as 133 much for foster-care services as for family services. 134 Regrettably no state has yet adopted a stringent removal standard. Huntington therefore finds that “despite a greater emphasis on family preservation, the underlying removal standard has not changed much since 1975 135 and is not significantly more determinate.” Huntington also reports that the ASFA has resulted in the states incorporating time limits for children in foster care, and placing much more 136 emphasis on permanency planning for children taken from parental custody. These efforts aimed at permanency planning are obviously having an effect. 137 The foster-care population has declined substantially since my article. The primary reason for this is that the average stay in foster care has become on 138 balance shorter. I think this is largely the result of permanency planning. Professor Huntington makes a strong case for reorienting the child-welfare 139 system towards prevention. I would certainly support the sorts of polices she suggests, and particularly think that nurse and family-support programs and
126. Id. 127. Pub. L. No. 105-89, 111 Stat. 2115 (codified as amended in scattered sections of 42 U.S.C.). 128. See Libby S. Adler, The Meanings of Permanence: A Critical Analysis of the Adoption and Safe Families Act of 1997, 38 HARV. J. ON LEGIS. 1, 1 (2001). 129. Huntington, supra note 114, at 229. 130. Id. 131. Id. 132. Id. at 231–32. 133. Id. 134. Id. at 230. 135. Id. 136. Id. at 221–23. 137. See Leroy H. Pelton, Not for Poverty Alone: Foster Care Population Trends in the Twentieth Century, 14 J. SOC. & SOC. WELFARE 37, 43–44 (1987). 138. KATHY BARBELL & MADELYN FREUNDLICH, FOSTER CARE TODAY 13 (2001), available at http://www.docs.lakotalaw.org/ombudsman%20rebuttal%20sources/foster_care_today.pdf. 139. Huntington, supra note 114, at 245–48.

270

LAW AND CONTEMPORARY PROBLEMS

[Vol. 77:249

child-centered early-childhood education are worthy of support. For me her article suggests some interesting questions about how law can be used to provide greater bureaucratic accountability, as well as to elucidate the extent to which services are successful in preventing harm and the degree to which spending money to provide incentives and subsides might be more important than legal constraints. VI CONCLUSION The articles in this issue suggest both continuity and change. For the private dispute–resolution function, the indeterminate best-interests standard remains ascendant notwithstanding my critique. Indeed, for this function the most important reason for the continuing relevance of my old article is the stubborn persistence of the best-interests standard. Moreover, for reasons I articulated in my article, the lack of a social consensus about appropriate gender roles makes it difficult to formulate alternative standards that would be broadly accepted. With respect to process, however, the years since my article have witnessed an explosive growth of mediation and other less adversarial methods of resolving conflicts. With respect to child protection, the law has substantially changed, largely as a result of federal legislation and dollars that have influenced the states. But states have not changed the legal standards governing removal, and I suspect children today are being placed in foster care in circumstances when preventive and preservation services would have allowed them to be protected within their homes. On the other hand, I am gratified by the evidence that my suggestions for time limits after placement have not only been adopted but appear to be having a beneficial effect. To answer the questions I posed at the outset, I believe the article’s two core themes have withstood the test of time remarkably well. And to a surprising degree, the analysis remains relevant to contemporary challenges for custody law and policy.