1. Investigation reports were also gathered for the same year in Ramsey County (St. Paul) and Anoka County, a suburban county adjoining both of the others with a much smaller population than either. Divorces and investigation reports from these counties are not systematically analyzed in this article for the reasons stated infra pp. 739–45.Google Scholar

The investigations described and analyzed in this article were conducted by a professional staff employed by a division of the Hennepin County Department of Court Services, the social service adjunct agency of the county's district court. One division of the department conducts presentence investigations for criminal cases; the division whose work is examined here, the Domestic Relations Division, conducts custody and financial investigations and provides custody and visitation counseling and supervision in divorce cases. See also infra pp. 736–38 & appendix 1.Google Scholar

2. Cases without temporary hearings tended to be (a) those in which middle-class spouses and their lawyers agreed to interim custodial and financial arrangements and the husband was not expected to disappear or default on his financial obligations; (b) cases in which there was little prospect of contest (or support) because the husband, either missing or not a resident of Minnesota, could not be served with process personally. These two types of cases continue to this day to be less likely to involve a temporary hearing.Google Scholar

3. In this article the investigator will be designated a caseworker. Most caseworkers whose reports are analyzed here were trained as social workers; one was a Ph.D. psychologist. See infra p. 739.Google Scholar

The case numbers (H-1, H-2, etc.) that identify cases in this article were derived by arranging all custody investigation reports alphabetically by county and numbering them consecutively. To impede identification of the divorcing couple, I have reported less than complete information about the family. I do not believe that failing fully to report the facts unduly distorts the conclusions 1 reached about individual reports or the custody investigation enterprise. That the children whose custody was at stake in the 1970 Hennepin County reports are now 15 years older should help to shield them from the limited invasion of their privacy that this analysis of the problem of custody investigations requires. I have tried to disguise individual caseworkers by using personal pronouns interchangeably in referring to staff members without regard to their sex.Google Scholar

4. Seldom do the 1970 reports indicate whether the investigation was ordered because the spouses took the initiative by stipulating to it or because the referee or judge took the initiative and requested the litigants' consent. At the time, lawyers believed that the Supreme Court of Minnesota had ruled that an investigation could not be ordered unless both spouses approved. The legal situation at the time was in fact fairly unclear. See Stanford v. Stanford, 266 Minn. 250, 123 N.W.2d 187 (1963); Oltmanns v. Oltmanns, 265 Minn. 377, 121 N.W.2d 779 (1963); Gumphrey v. Gumphrey, 262 Minn. 515, 115 N.W.2d 353 (1962).Google Scholar

A requirement that both Litigants consent may not always assure that an investigation is truly voluntary. See, e.g., Withrow v. Withrow, 212 La. 427, 31 So. 2d 849 (1947) (trial judge stated that under the evidence neither spouse should get custody and requested permission to order an independent investigation; on plaintiff's refusing consent, custody was awarded to defendant). See Comment, 24 U. Chi. L. Rev. 349 (1957). See also infra note 13.Google Scholar

5. For an indication of the varieties of information custody investigation reports contain, see Foote, Levy, & Sander, Cases and Materials on Family Law 944–76 (3d ed. 1985).Google Scholar

6. See, e.g., Wis. Stat. Ann. § 767.045 (1981); Haugen v. Haugen, 82 Wis. 2d 411, 262 N.W.2d 769 (1978); Bahr v. Bahr, 72 Wis. 2d 145, 240 N.W.2d 162 (1976).Google Scholar

7. See Inker & Perretta, A Child's Right to Counsel in Custody Cases, 55 Mass. L.Q. 229 (1970).Google Scholar

8. See Landsman & Minow, Lawyering for the Child: Principles of Representation in Custody and Visitation Disputes Arising from Divorce, 87 Yale L.J. 1126, 1176–80(1980). See also infra note 13 and accompanying text.Google Scholar

9. The Domestic Relations Division of the Department of Court Services developed custody resolution counseling in 1976, calling the service “mediation counseling” initially but dropping the term “because of the confusion with the way the term mediation has been used in the legal profession.” See D. Olson et al., Custody Resolution Counseling (CRC): Description and Comparison with Custody Study (CS), in Child Custody: Literature Review and Alternative Approaches 97, 99 (mimeographed 1979). For an examination of the similarities and differences between the division's work in custody investigations and in custody resolution counseling, more recently once again referred to as mediation, see infra note 193 and accompanying text and appendix 1.Google Scholar

10. See generally Thompson, J. & McEwen, W., Organizational Goals and Environment: Goal Setting as an Interaction Process, 23 Am. Soc. Rev. 23 (1958); C. Perrow, Organizations: Organizational Goals, 11 Int'l Encyclopedia Soc. Sci. 305–10 (2d ed. 1968). Cf. C. Brager & S. Holloway, Changing Human Service Organizations 49 (1968): “Human Service agencies are vulnerable to program fads and fashions. This may be due to uncertainty about efficacy of service technologies or their limited measurability.” See also infra note 193 and appendix 1.Google Scholar

This is not to suggest that all “conciliators” and “mediators” are simply trying to maintain their incomes—there are undoubtedly many sincere and talented professionals who have come to reject old methods of helping divorcing families and have decided that a new form of help combined with (nonlegal) self-help offers more promise than any other kind of social service. See, e.g., Stier & Hamilton, Teaching Divorce Mediation: Creating a Better Fit Between Family Systems and the Legal System, 48 Alb. L. Rev. 693 (1984); J. Kelly, Mediation and Psychotherapy: Distinguishing the Differences, Mediation Q., No. 1, at 33 (1983); McIsaac, Court-connected Mediation, 21 Conciliation Cts. Rev., Dec. 1983, at 49. But the sincerity of the professionals does not always alter the dangers to clients posed by the nature of bureaucracies.Google Scholar

11. In the last period, 221 investigations were completed. See Hansen & Lawrence, Custodians Recommended by Domestic Relations Personnel in Custody Studies Assigned by the Court from January 1, 1968 to December 31, 1969 (Hennepin County Department of Court Services, mimeographed, undated). The figures for years prior to 1968 and the caseload figures that follow come from the department's annual reports as related to me by Robert B. Wyckoff, Supervisor, Domestic Relations Division, Hennepin County Department of Court Services, December 1, 1981. The department's data are sparse and may not be completely reliable. Although the 1958 and 1959 reports state that 50 and 97 studies, respectively, were conducted in those years, the 1961 report states that 800 investigations were completed in 1960 and 880 in 1961. Later reports indicate 166 investigations in 1962, 595 in 1963, 213 in 1964, and 137 in 1965. Since only 92 investigations were conducted in 1967, according to the department's report, and 84 were conducted in 1970, according to my personal check of the department's files, the 1961–65 data seem suspect. Yet even if some of the department's figures are inaccurate, it is not unfair to say that the volume of the department's custody investigation business was substantial during the 1960s and growing fairly regularly.Google Scholar

In addition to investigations, the division has regularly maintained a fairly extensive responsibility for postdecretal custody and visitation supervision and counseling. In 1957, e.g., when 51 investigations were conducted, the department also supervised 177 divorced couples and their children. Some of the variation in caseload accounting in annual reports may be the product of mixing initial divorce investigation numbers with supervision or counseling case totals. Some investigations are never completed because the parties settle their custody dispute. See infra pp. 778–80. In 1970, there were 6 such cases.Google Scholar

12. These figures were taken from a 1979 Department of Court Services study. See supra note 9. Note that these figures include “custody resolution counseling” cases as well as custody investigations.Google Scholar

In 1970, investigations were conducted in some 4% of the divorce actions with children that terminated in a decree. The 55 investigations conducted in Ramsey County that year represented 5.7% of the divorce actions with children that terminated in a decree. In 1980 in Ramsey County custody investigations were conducted in 7.2% of a considerably smaller by more than 13%) number of divorces with children. No comparable data for 1980 have been gathered for Hennepin County.Google Scholar

13. Landsman & Minow, supra note 8. Two pages of the authors' essay, id. at 1178–79, were devoted to custody investigations. The authors suggested that they and the attorneys whose efforts they were reporting agreed that appointment of counsel for the children was often in the nature of a “second bite at the apple.” The authors concluded that the “child's attorney brings a perspective on the custody question different from that of the Family Relations Officer and can in addition protect the child in court. The attorney's independent judgment is therefore an important contribution.”Id. at 1180. This conclusion was drawn from interviews with the attorneys. Some of the authors' data suggest that attorney representation will not obviate custody investigations. The attorneys’ recommendations, the authors concluded, were more likely to be attuned to the wishes of the child than were those of the family relations officer. Id. at 1179. Moreover, the attorneys were reported to have found the investigators' information about the child to be “misleading and superficial.”Id. Yet in an anecdotal report by one attorney, quoted in a different section of the article, the authors reported:Google Scholar

Id. at 1147. It is certainly possible that in many, even in most, cases the caseworkers may have been more “misleading” as well as more “superficial.” It is sufficient for present purposes to conclude that, whether or not children should have separate representation in divorce cases, the custody investigation seems to be alive and well in Connecticut.Google Scholar

14. The term “custody investigation” was mentioned in 960 West Publishing Company key number paragraphs between 1951 and 1960 and in 3,464 key number paragraphs during the period 1961–70-an increase of 361%. The term was mentioned in 8,227 key number paragraphs during the period 1970–80, an increase of 238% from the previous decade. See also J. Pearson & M. Ring, Judicial Decision-Making in Contested Custody Cases, 21 J. Fam. L. 703,713 (1983) (custody investigations ordered in almost all contested cases in Denver County during 1973–77, in less than 50% of cases in two other counties).Google Scholar

15. See Uniform Marriage and Divorce Act § 405 (1970). For litigation in states that have adopted the act, see In re Kramer, 177 Mont. 61,580 P.2d 439 (1978); In re Waggener, 13 Wash. App. 911, 538 P.2d 845 (1975). The case law is described in Foote, Levy, & Sander, supra note 5, at 922–39. The Uniform Act provided for custody investigations but subjected them to a variety of procedural constraints. In evaluating the positions taken in this article, the reader should consider that the author was the reporter for the Uniform Marriage and Divorce Act.Google Scholar

16. See Symposium: Children, Divorce and the Legal System: The Direction for Reform, 19 Colum. J.L. & Soc. Probs. 105 (1985) quoting the statutory language and the Law Revision Commission's Commentary.Google Scholar

17. Uniform Marriage and Divorce Act § 405(a) (1970) limited the trial judge's authority to order investigations to “contested proceedings and in other custody proceedings if a parent or the child's custodian so requests.” The New York Law Revision Commission's proposal permits the judge to order a “custody evaluation” at the end of an initial conference “upon motion by any party or the court's own motion,” even if the spouses have agreed as to which of them shall become the child's postdivorce custodian, so long as the report “may aid in [the court's] determination of custody.” See proposed § 242(d)(4), Symposium, supra note 16, at 134–35. For the Law Revision Commission's support of this policy, see id. at 150.Google Scholar

18. Consider Note, 35 A.L.R.2d 629,631 (1954): “There is a growing conviction that persons with specialized training and experience such as social workers are better qualified to determine what is in the best interests of the child than even the best intentioned judge.” The remark has been quoted in the following articles: Note, The Family in the Courts, 17 U. Pitt. L. Rev. 206, 250 (1956); Foster, & Freed, , Child Custody (pt. 2), 39 N.Y.U. L. Rev. 615 (1964); Okpaku, Psychology, Impediment or Aid in Child Custody Cases? 29 Rutgers L. Rev. 1117 n.3 (1976); Gozansky, Court-ordered Investigations in Child Custody Cases, 12 Willamette L.J. 511 (1976). But see Wolf, What Is Wrong with Home Studies, Speech Delivered to the Family Law Section, American Bar Association, New Orleans, La., Aug. 11, 1981: “[T]he ‘home study’ is usually merely one untrained person's opinion based upon individual prejudices superimposed upon a collection of hearsay devoid of any objective standards. … If ever the Rona Barretts of the United States are seeking a new career, the expertise as gossips most certainly could be put to immediate use as Court-appointed investigators conducting ‘home studies.“’.Google Scholar

19. See generally D. Rothman, Conscience and Convenience: The Asylum and Its Alternatives in Progressive America 205–89 (1980); Feld, Juvenile Court Legislative Reform and the Young Offender: Dismantling the “Rehabilitative Ideal,” 65 Minn. L. Rev. 167 (1980). Some of the ideas expressed in this article were also formulated for another purpose by Halem, Divorce Reform: Changing Legal and Social Perspectives 114–41 (1980). Halem's emphasis was on the extent to which nonadversarial techniques were used to prevent or at least to minimize divorce. Kay, A Family Court: The California Proposal, 56 Calif. L. Rev. 1205(1968), recognized the analogy to juvenile court ideology arguments for a family court present and the risks that analogy suggests. For a thoughtful effort to distinguish the “private-dispute-settlement” function of divorce courts from the “child protection” function of juvenile courts, and a recommendation that divorce court judges should not engage in “child protection” endeavors, see R. Mnookin, Child-Custody Adjudication: Judicial Functions in the Face of Indeterminacy, 39 Law & Contemp. Probs. 226, 229, 266 (1975). See also infra pp. 733, 782–90.Google Scholar

20. Platt, A., The Child Savers: The Invention of Delinquency (1969).Google Scholar

21. See, e.g., Feld, supra note 19.Google Scholar

22. For an extensive review of one of the broader recent reform efforts, see Symposium, I.J.A.-A.B.A. Juvenile Justice Standards Project, 57 B.U.L. Rev. 617–726 (1977).Google Scholar

23. Leavell, , Custody Disputes and the Proposed Model Act, 2 Ga. L. Rev. 162, 165 (1968). See also Comment, Alternatives to “Parental Right” in Disputes Involving Third Parties, 73 Yale L. Rev. 151 (1963). The literature review that follows is impressionistic rather than thorough and comes largely from the 1950s and later—the coalescing period of divorce court “child saving.” Earlier cases, authorities, and legislative initiatives are not inconsistent. See generally Halem, supra note 19. For a much more detailed and careful tracing of the shifts in child custody doctrine, see Zainaldin, J., Emergence of a Modern American Family Law: Child Custody, Adoption and the Courts 1796–1851, 73 Nw. U.L. Rev. 1038 (1979). See also Mnookin, supra note 19, at 233–35.Google Scholar

24. The resulting morass is nicely pictured in Mnookin, supra note 19.Google Scholar

25. Rothman, , supra note 19, at 212. See also Rothenberg, The Lawyer's Role in Child Custody Disputes, 23 N.Y.C.B. Bull. 95, 96 (1966).Google Scholar

26. Note, The “Adversary” Process in Child Custody Proceedings, 18 Case W. Res. L. Rev. 1731, 1733 (1967).Google Scholar

27. Gellhorn, W., Children and Families in the Courts of New York City, a Report by a Special Committee of the Association of the Bar of the City of New York 315–16 (1954). It may be unfair to single out Gellhorn's work in this impressionistic survey—but because his report conceptualized the purposes so clearly and addressed the issues so fully, it is by far the best statement as well as defense of the child-saver emphasis in the divorce-custody reform “movement” literature.Google Scholar

28. Id. at 12–15.Google Scholar

29. McKeiver v. Pennsylvania, 403 U.S. 528, 544 (1971).Google Scholar

30. Rothman, supra note 19, at 217.Google Scholar

31. Id.Google Scholar

32. Id. at 217–18.Google Scholar

33. Gellhorn, supra note 27, at 310.Google Scholar

34. For a statement of the “anti-adversary procedures” policy, see Levy, Treatment of Child Custody Problems in the [Texas] Family Code j-1, j-4 to j-5, in Proceedings of the Institute on the Family Code, Family Code Project, Southern Methodist University School of Law (mimeographed 1967). The argument was made to support opposition to routine appointment in custody cases of an attorney or a guardian ad litem for children in uncontested divorce cases. See also infra pp. 778–82.Google Scholar

35. Rothman, supra note 19, at 223.Google Scholar

36. Id. at 216–18.Google Scholar

37. For a report on the strength of the “informality” notion in adult corrections as it focuses on the presentence investigation, and a powerful argument for formal hearing controls, see Fennell & Hall, Due Process at Sentencing: An Empirical and Legal Analysis of the Disclosure of Presentence Reports in Federal Courts, 93 Harv. L. Rev. 1613 (1980). For a discussion of the ties between probation, informality, and the “rehabilitative ideal,” see Rothman, supra note 19, at 216. See also S. Schlossman, Love and the American Delinquent (1977).Google Scholar

38. See Rothman, , supra note 19, at 218–19:.Google Scholar

39. Gellhorn, supra note 27, at 310–11, 314.Google Scholar

40. See Ennis, & Litwack, , Psychiatry and the Presumption of Expertise: Flipping Coins in the Courtroom, 62 Calif. L. Rev. 693 (1973).Google Scholar

41. Litwack, Gerber, & Fenster, , The Proper Role of Psychology in Child Custody Disputes, 18 J. Fam. L. 269, 283, 295–97 (1979). For the article that led to this defense, see Okpaku, supra note 18, at 1117. For an equally vigorous defense, by psychiatrists, of the psychiatric version of a “mental health expertise” approach to custody litigation, see R. Gardner, Family Evaluation in Child Custody Litigation (1982); A. Watson, Children of Armageddon: Problems of Child Custody Following Divorce, 21 Syracuse L. Rev. 55 (1969) (recommending a “psychological best interests of the child” test and appointment of “behavioral science judges” to sit in custody cases with legally trained judges). But see infra pp. 763–78 & 794–95.Google Scholar

42. See Rothman, , supra note 19, at 218. For a similar recommendation for the New York Family Court, see Gellhorn, supra note 27, at 321–28. It is not impossible that the commitment to mental health experts may have been at least in some juvenile courts something of a symbol rather than a reality. Rothman reports on a private investigation of the Cincinnati Juvenile Court conducted for the Rockefeller Foundation by William T. Thomas, a noted sociologist and child guidance advocate: “What disturbed Thomas most was [Judge] Hoffman's ignorance of the principles of psychology and development: the court, after all, was to be deciding the fate of the delinquent on the basis of his treatability, and here the presiding judge really knew very little or nothing about the matter. Hoffman had all the respect in the world for psychology—indeed, too much respect. ‘He believes,’ declared Thomas, ‘in the infallibility of “science,” especially as embodied in the psychiatrist….’ Ultimately, Hoffman's ‘policy is one of “common sense” rather than scientific procedure.’” Rothman, supra note 19, at 239–40.Google Scholar

43. Leavell, supra note 23, at 184–85.Google Scholar

44. See, e.g., Rothenberg, supra note 25, at 98–99: “[W]hen you are not limited in your freedom of action and you do have an opportunity to exercise your own judgment, your failure to accept the proposal from your adversary or from the Court, to call in the [social service agency that conducted custody investigations in New York City]-or, better yet, to initiate such a step yourself-puts in serious question your protestations of solicitude, however eloquent, for the child's well-being.”.Google Scholar

45. Note, Use of Extra-Record Information in Custody Cases, 24 U. Chi. L. Rev. 349, 358–59 (1957). For a more traditional view as to the need for confrontation and cross-examination of the caseworker, see Note, supra note 26, at 1733.Google Scholar

46. See, e.g., Fewel v. Fewel, 23 Cal. 2d 431, 144 P.2d 592 (1943) (praising custody investigations but holding that the investigation in issue should have been excluded and the caseworker subject to cross-examination); Gumphrey v. Gumphrey, 262 Minn. 515, 115 N.W.2d 353 (1962).Google Scholar

The earliest and most formidable of the divorce reformers was Judge Paul Alexander, founder of the first family court, in Toledo, Ohio, and tireless promoter of the reforms described in the text. See Alexander, P., Let's Get the Embattled Spouses Out of the Trenches, 18 Law & Contemp. Prob. 98 (1953).Google Scholar

47. See, e.g., Rothenberg, supra note 25.Google Scholar

48. This is the explanation for the move to compulsory mediation in California given by J. Pearson & N. Thoennes, Mediating and Litigating Custody Disputes: A Longitudinal Evaluation, 17 Fam. L.Q. 497, 514 (1984). The New York Law Revision Commission's mediation proposal allows (indeed, may require) the judge to compel mediation even if the spouses have already agreed as to the proper custodian for the child so long as mediation will “serve the best interests of the child.” See supra note 17.Google Scholar

49. See, e.g., Schephard, Philbrick, & Rabino, , Ground Rules for Custody Mediation and Modification, 48 Alb. L. Rev. 616, 656–57 (1984) (mediation retainer agreement drafted to provide that in case of postdecretal problems the spouses cannot resolve themselves, they must return to mediator who would act as a “gatekeeper” to subsequent arbitration or postdecretal litigation; mediator could thus deter bad faith efforts by a parent to terminate mediation; no parallel controls needed for mediator because “a mediator presumably would not continue her efforts if both parents thought the process had no chance of success”). But see Stier & Hamilton, supra note 10, at 693.Google Scholar

50. See Rothenberg, , supra note 25.Google Scholar

51. See, e.g., Fuhrman v. Fuhrman, 254 N.W.2d 97 (N.D. 1977) (trial court order providing that children were to remain in the family home with each parent living with them in alternate months reversed); Ziegler v. Ziegler, 304 A.2d 13 (D.C. Ct. App. 1973); Mandelstam v. Mandelstam, 458 S.W.2d 786 (Ky.), 39 U. Cin. L. Rev. 785 (1970).Google Scholar

52. See, e.g., Auge v. Auge, 334 N.W.2d 393 (1983) (although parties are entitled to a hearing as to whether custodial parent can remove child from the state, custodial parent is presumptively entitled to remove the child); Patton v. Armstrong, 6 III. App. 3d 998, 286 N.E.2d 351 (1972). See generally Foote, Levy, & Sander, supra note 5, at 916–19. See also B. Lindsley, Ruling Without Bias, Judges' J., Winter 1985, at 19,20: “Our religious outlook may compel us to forbid the reading of the Bible during visitation, as a judge did recently in an eastern state. Maybe he was right; maybe it was overkill.”.Google Scholar

53. See, e.g., Hanson v. Hanson, 187 Neb. 108, 187 N.W. 2d 647 (1971) (supervision of custody by juvenile probation officer approved where custodial mother had postdivorce sexual relations with two men); Niesen v. Niesen, 38 Wis. 2d 599, 157 N.W.2d 660 (1968) (child's taking stepfather's name does not emancipate child so as to terminate divorced father's support obligation); Uniform Marriage and Divorce Act § 408 (1970). See also infra pp. 782–90.Google Scholar

54. See Chambers, D., Making Fathers Pay (1979).Google Scholar

55. See, e.g., Lindsley, supra note 52: “religious outlook may lead to awarding custody to the most consistent churchgoer. It works both ways. How does the right case get before the right judge?” See also H-85 (referee after final default divorce hearing sent memorandum to custody investigation caseworker suggesting that custodial mother should be reported to welfare department for immediate removal of child unless mother immediately seeks “her own independent living quarters away from her parents in what appears to be a distinctly injurious environment”). See infra pp. 783–85.Google Scholar

56. McKeiver v. Pennsylvania, 403 U.S. 528, 544 (1971).Google Scholar

57. See supra note 38.Google Scholar

58. The three cases used as illustrations in this section were the first of each type to appear in Hennepin County when the reports were arranged alphabetically. See supra note 3.Google Scholar

59. To measure the time required to obtain a divorce, an arbitrary decision must be made as to when the action was commenced. Although some lawyers file the divorce complaint formally when it is signed by the plaintiff and served on the defendant, in many cases the complaint is signed long before it is served; or the complaint may be filed in the courthouse long after the defendant is formally served or informally notified of the divorce action. There was no noticeable pattern to the variation from case to case in Hennepin County in 1970. For all cases in the three-county sample, the delay between formal filing and temporary hearing averaged almost four weeks. In this article the formal filing date is treated as the event that commences the divorce action.Google Scholar

60. Uniform Marriage and Divorce Act § 405(c) (1970) requires disclosure of the file to both parties, The recently proposed New York legislation would allow the attorneys to see “underlying data and reports” used in the preparation of the evaluation report only “upon a showing of good cause.” See Symposium, supra note 16, at 136. The limitation was designed to avoid “pro forma requests for the underlying work product, which may be intrusive and obstructive to the evaluation process.”Id. at 153. But see infra p. 795.Google Scholar

61. But see Uniform Marriage and Divorce Act § 405(c) (1970) (right to cross-examine investigator may not be waived prior to hearing). Subsection (b) required that the procedural rules be followed as a condition of the report's admissibility. The Minnesota legislature added a new provision in 1984 apparently designed to make the report admissible despite a violation of subsection (b). See 1984 Minn. Laws ch. 635, 4.Google Scholar

62. The average length of time from filing the complaint to decree for all 629 cases in the three-county sample, consensual as well as contested cases, was 11.3 months; the median length of time was 9.9 months.Google Scholar

63. See Litwack, , Gerber, & Fenster, supra note 41, at 297. See also infra pp. 765–68.Google Scholar

64. See supra note 33.Google Scholar

65. It seems likely that in many appeals from contested cases, “the merits” (which spouse should have been awarded custody of the children) and the evidentiary issues become inextricably intertwined. The common use by appellate judges of doctrines like “waiver” and “harmless error” in custody cases involving investigations certainly suggests covert concerns. For application of waiver, see, e.g., Kern v. Kern, 333 So. 2d 17 (Fla. 1976) (trial court's use of investigation report not reversible error where parties had been notified and could have called caseworker as a witness for cross-examination). As to harmless error, see Theeke v. Theeke, 105 Ill. App. 3d 119, 433 N.E.2d 1311 (1982) (trial court consideration of incompetent custody investigation report not reversible error where other evidence amply supported the custody award); Larson v. Larson, 30 Wis. 2d 291, 140 N.W.2d 230 (1966) (trial judge's refusal to allow parties to cross-examine caseworker was harmless where other evidence sufficiently supported custody order); Dahl v. Dahl, 237 Cal. App. 2d 407, 414–16, 46 Cal. Rptr. 881, 886–87 (1965) (erroneous admission of report and failure to allow mother opportunity to cross-examine report author harmless because of other evidence and “peculiar circumstances” of case). See generally R. Traynor, The Riddle of Harmless Error (1970).Google Scholar

66. The “stipulation” is correctly named because it is a document filed in a lawsuit that specifies the issues the litigants will not put in issue. In fact, the document is like what lawyers in other states would call a separation agreement.Google Scholar

67. For the requirements and the stylistic idiosyncrasies of “consensual perjurious divorce” prior to the “no fault revolution,” see Foote, Levy, & Sander, Anatomy of Divorce, in Cases and Materials on Family Law 943–1010 (2d ed. 1976).Google Scholar

68. See H-1, supra p. 747. At the trial, the impact on the judge of the “behavioral” aspects of the case (here, the wife's adultery) outweighed the recommendation by the caseworker that the wife be given custody of the three younger children. See also infra note 128 and accompanying text. In cases where the report made a specific recommendation, the children were a little more likely to end up in the custody of the recommended parent in “consensual with c/i cases” like H-2 than they were in “contested with c/i cases” like H-1. Of 42 recommendations, 36, or 86%, were “followed” in “consensual with c/i” cases (although only 60% of those recommendations were that the mother be awarded custody); but only 10 of 18 recommendations, 55%, were “followed” by the judge in “contested with c/i” cases (although the percentage of mother recommendations in this category, 55%, was about the same as in the “consensual with c/i” group). In contested cases, recommendations that the temporary custodian be awarded custody were much more likely to be followed than recommendations of other possible custodians (75% of the 8 temporary custodians recommended compared to only 33% of the 8 spouses recommended who were not temporary custodians). Similar proportions held in “consensual with c/i” cases. Overall, more than 75% of the reports recommended that the temporary custodian be awarded permanent custody.Google Scholar

69. Many of the lawyers who were “regulars” in the divorce court no doubt knew or suspected what the statistics would indicate, see supra note 68, and would have guided their clients in accord with their implications. Yet any prediction as to the judge's response to the report always left some of that uncertainty which facilitates bargaining while giving a substantial bargaining “endowment” to the spouse recommended by the caseworker's report. See Mnookin & Kornhauser, Bargaining in the Shadow of the Law: The Case of Divorce, 88 Yale L.J. 950 (1979). The negotiations of unrepresented spouses might be affected even more substantially because they would have no basis at all for predicting how the judge would respond to the report.Google Scholar

That reports are in fact influential in bargaining is indicated by anecdotal information from lawyers as to the most important role played by custody investigations-to allow lawyers to dampen a client's enthusiasm for a custody contest, without offending the client-“If we get an affirmative report, we can go forward.” When the report comes back with the (sometimes expected) recommendation that the other parent be awarded custody, the disappointed client is not likely to blame the lawyer for failing to represent the client with sufficient vigor. See also infra notes 193 & 239 and accompanying text.Google Scholar

70. Although such “hybrids” are not often identified for discrete analysis, settlements in which the judge has played a role are not unknown. See, e.g., Kritzer, , The Judge's Role in Pretrial Case Processing: Assessing the Need for Change, 66 Judicature 228 (1982); Lempert, , More Tales of Two Courts, 13 Law & Soc'y Rev. 91, 98 (1978); Friedman, & Percival, , A Tale of Two Courts, 10 Law & Soc'y Rev. 267 (1976).Google Scholar

There were 51 contested stipulations where no custody investigation was conducted during 1970. Thus, slightly more than 10% of the sample, 2% of all divorces granted, were contested stipulations. I believe that it is likely that contested stipulations with custody investigations involved disputes over the children while those without investigations involved financial disputes; but there is no way from the courthouse file alone to prove or disprove that belief.Google Scholar

71. See Litwak, , Gerber, & Fenster, supra note 41, at 297.Google Scholar

72. It is likely that the judge's personality, his known preferences in custody cases, his commitment to the agency staff and to informal custody adjudication methods will be part of the lawyer's calculus when the client must be advised whether to settle for the custody arrangement recommended by the investigation report. The lawyer's intuition and predictions will inevitably play a large role. None of the “contested” or “contested stipulation” cases in the 1970 three-county sample lacked counsel for both parents.Google Scholar

For an analysis of a more recent sample of custody investigation cases in another jurisdiction that also finds that custody reports are more often utilized for settlement purposes than for litigation, see Ash & Guyer, In the Shadow of Solomon's Sword: The Functions of Psychiatric Evaluation in Contested Custody and Visistation Cases, - J. Am. Acad. Child Psychology - (1986) (forthcoming).Google Scholar

73. See supra note 17 and accompanying text. The quotation is taken from the comments of the New York Law Revision Commission, Symposium, supra note 16, at 105, 150 (1985).Google Scholar

74. See Symposium, supra note 16, at 135, 156–57. As to juvenile court probation, see supra note 38.Google Scholar

75. These two terms were first used by Mnookin to distinguish the policy concerns of divorce-custody decision making from those of juvenile court jurisdiction; he also explains carefully why the “starting point” for “private-dispute-settlement” should be “family autonomy.”Id. at 266–68. See Mnookin, supra note 19, at 229. See also infra pp. 782–90.Google Scholar

76. The New York Commission apparently contemplated that the evaluation, or the threat of one, would encourage settlement by making spouses who had not yet reached agreement more reasonable in the negotiations. See Symposium, supra note 16, at 151: “The concept of mediation followed by evaluation should have a synergistic effect on both processes. The prospect for parties undergoing an evaluation which will produce a recommendation to the court should be a catalyst for serious and good faith efforts on their parts to reach agreement in mediation. Moreover, the evaluation process will usually be invoked only when the parties have not been able to arrive at their own custodial arrangements for the child, and thus it is an appropriate time to call upon qualified and expert mental health professionals for their assessment of the situation.”.Google Scholar

77. See, e.g., the authorities cited supra note 41.Google Scholar

78. A fairly typical Ramsey County report, R–3, read in part:Google Scholar

The report went on to discuss in detail plaintiff's assets (a car worth $200), the value of the homestead and plaintiff's and defendant's debts, individually specified and with current and expected monthly payments, and plaintiff's and defendant's specific monthly budgets (including the plaintiff's $10.00 a month to Big Brothers of America and defendant's $5.00 a month to Minnesota Milk).Google Scholar

I was not able to obtain all the Ramsey County investigation reports. The district's trial judges voted to give the study access only to reports where both spouses had agreed in writing to the disclosure. Some reports were obtained with permission; others were obtained from the files of the Ramsey County District Court-which are by law open to the public. The Hennepin County judges voted to authorize the study to have access to all reports and to the underlying documentary files of the Domestic Relations Division.Google Scholar

79. Cf. Watson, supra note 41.Google Scholar

80. The Minnesota Multiphasic Personality Inventory (MMPI), a standard nonprojective personality test, was formulated and standardized at the University of Minnesota. It remains a fairly popular measurement device for psychologists trained at the university. For further information about how the MMPI is used in Hennepin County custody investigations, see infra pp. 775–78.Google Scholar

81. See Satir, V., People Making (1972); V. Satir, Conjoint Family Therapy (1967).Google Scholar

82. See infra pp. 763–78.Google Scholar

83. See supra note 75.Google Scholar

84. There were 84 custody investigation cases in Hennepin County in 1970–53 “consensual with c/i,” 9 “contested stipulations with c/i,” and 22 “contested with c/i.” In 3 cases, two full investigations with reports were conducted. In some of the “aborted” cases, some paperwork indicating the caseworker's activities was found. See infra pp. 771–73. In 8 cases no report was found, although the court file indicated that the temporary hearing referee had ordered a custody investigation. In only 3 cases did a full report fail to contain the caseworker's recommendation. See H-13, discussed infra p. 746.Google Scholar

85. Most of the investigation cases ended with stipulations rather than litigation. See supra p. 730. And there is persuasive evidence that at least some custody investigators were quite aware that their reports were not being prepared for an actual trial. See infra pp. 772 & 780. Where the caseworker expected a stipulation, it would not be unlikely that he would recommend as custodian that person likely to be named in the stipulation-especially if the caseworker was counseling the spouses or seeking to “mediate” (i.e., help to resolve without litigation) their dispute. Since the caseworkers in any event recommended the temporary custodian in more than 70% of the cases, and since in many cases the temporary custodian was probably chosen consensually by the spouses rather than imposed on them at the temporary hearing, the custody awards and the caseworker recommendations should often match.Google Scholar

The caseworkers' “batting averages” in contested cases were very good so long as they recommended the temporary custodian (10 of 12 temporary custodian recommendations followed by the post-trial custody award; only 2 of 8 recommendations of a person other than the temporary custodian were followed in post-trial award). The husband recommendations in contested cases were at least as successful as the wife recommendations so long as the husband recommended was also the temporary custodian. But only 2 of the 5 recommendations of husbands who were not temporary custodians were followed after contests; and none of the 3 recommendations of a noncustodial nonparent were followed after a contest.Google Scholar

86. This generalization does not accurately describe one caseworker, employed by the agency for a short period, who wrote only one of the reports in this sample. Of the ten caseworkers in 1970, seven had been staff members for five years or more and five are still members of the staff. Some of these five may now be occupied primarily with “custody resolution counseling.” See infra notes 193 & 200 and appendix 1. For additional comments about training and its impact on the quality of the reports, see infra notes 129 & 220–40 and accompanying text.Google Scholar

87. Interview respondents were not separated by county of divorce for this analysis. There were, then, 290 spouses from all three counties to whom these questions were relevant. The generalizations in the text and footnotes in this paragraph are based on a small sample of this population, only 51 to 73 total responses from all counties, only 28 to 50 responses from the Hennepin County divorcing population.Google Scholar

88. Of the 52 wives responding to the question, 66% thought that the interview by the caseworker was “fair”; 59.1% of the husbands responding described the interview in the same fashion. While 67.3% of responding wives thought the interviewer was “thorough,” only 47.5% of responding husbands agreed. Since at least 60% of the investigation recommendations favored the wife, it is not likely that the “fair” and “thorough” responses only or even perhaps primarily reflected the caseworker's recommendation. Additional evidence that the spousal ratings of the caseworkers were not simply the product of the recommendation is afforded by controlling for the variable ultimate custodian. Three of 9 responding wives whose husbands ultimately became the custodian nonetheless thought the caseworker's interview was fair and 2 of 8 thought the interview was thorough. Perhaps more important in view of the relatively lower approval by husbands of investigation interviews, 23 of 39 husbands whose wives were ultimately named custodian believed that the caseworker's interview was fair, and 14 of 36 such respondents thought the caseworker's interview was thorough. The responding wives whose husbands were ultimately named custodian were considerably less charitable to the investigator than were the husbands who did not obtain custody. The responses are shown in table 1. At least among these respondents, husband evaluations of the report seem more independent of the ultimate custodial choice than do wife evaluations. See also infra notes 89 & 91.Google Scholar

89. Predictably, the proportions decline when account is taken of the ultimate custodian and of the respondent's rating of the fairness and thoroughness of the caseworker's interview, as shown in table 2. Of the 33 wives who labeled the interview fair, 26 would agree to another investigation; only 8 of the 17 who thought the investigation interview unfair would agree again. For those labeling the interview thorough, 23 of 33 would agree again. Of the 16 wives who said the interview was not thorough, 10 would agree again. The data for husband respondents were similar: 33 of 39 respondents who thought the interview fair (22 of 28 who thought it thorough) would agree again; 14 of the 27 who thought the interview unfair (20 of 31 who thought the interview not thorough) would agree again.Google Scholar

That a large proportion of husbands who thought the interview lacked either fairness or thoroughness would agree to another investigation suggests that many of them believed an investigation to be a necessary prelude to a successful quest for custody. An even more potent reason for some husbands' agreement to an investigation was the desire to check on their children's care-even if there was no plan to litigate the custody issue. Of the husbands who answered the “agree again” question affirmatively, 40% gave as a reason that “it's good to have an outside opinion” or “to see what's good for the children.” Only 19% of the wives answering affirmatively gave those two reasons. Of the wives answering affirmatively, 33% responded: “I have nothing to hide.” These data suggest that a substantial number of custody investigations may have been requested by fathers who wanted to be reassured that their wives would take good care of the children when the parties were divorced. This “father reassurance” function of investigations is explored infra pp. 78–81. I am indebted to Harry Gallaher, American Bar Foundation, for help with these figures and their meaning.Google Scholar

90. I rated each report impressionistically as each of them was read on an arbitrary five-point quality scale. When the responses to the “fairness” and “agree again” questions were cross-tabulated with report quality (as measured arbitrarily), there was no measurable tendency on the part of either husband or wife respondents to rate the investigation fairer or to be more likely to agree again based on the quality of the report. The same was true when the variable for awarded custodian was controlled.Google Scholar

91. There is some evidence to suggest that therapy outcomes vary in relation to the patient's perception of the therapist as understanding and empathic. See generally D. Orlinsky & K. Howard, The Relation of Process to Outcome in Psychotherapy, in Handbook of Psychotherapy and Behavior Change 283, 295 (2d ed. 1978). Hennepin County caseworkers may well have proceeded on such an assumption, leading the divorcing spouses to feel warmly toward them whatever their recommendation in the case.Google Scholar

Empathic relationships between professional and clients may cast a different light on those survey data that fairly consistently report enthusiastic responses by divorcing couples to the efforts of mediators in court settings. Comparative data have not been gathered. Cf. R. Levy, Comment to Pearson and Thoennes, 17 Fam. L. Q. 525 (1984). But see Pearson & Thoennes, Final Report of the Divorce Mediation Research Project (mimeographed, Nov. 15, 1984). In Hennepin County, responses by both husbands and wives to questions about the utility of the investigation to the spouse were overwhelmingly negative.Google Scholar

92. See, e.g., infra pp. 753–78. See generally Mnookin, supra note 19, at 226.Google Scholar

93. There is some unimportant variation in some of the reports. The only systematic and noteworthy stylistic differences were in the investigations that became informal mediations. See infra pp. 778–82.Google Scholar

94. For a review of some of these factors, see Foote, Levy, & Sander, supra note 5, at 850–976. The discussion which follows shows that there was no unanimity among the caseworkers on a number of other “factors” considered relevant to custody by some trial judges and appellate courts. See particularly the discussion of the “maternal preference,”infra pp. 753–54, and extramarital sexual activities, infra p. 758.Google Scholar

95. See, e.g., infra note 128 and accompanying text.Google Scholar

96. See infra pp. 741–43 & 749–50. See generally J. Simon, Basic Research Methods in Social Sciences: The Art of Empirical Investigation 22–26 (1969); I., Kogan, Principles of Measurement, in Social Work Research 87 (N. Polansky ed. 1960).Google Scholar

97. The full reports ranged in length from 1 to 11 single-spaced typewritten pages; they averaged more than 4 1/2 pages.Google Scholar

98. See infra pp. 778–82.Google Scholar

99. See generally R. Nisbett & L. Ross, Human Inference: Strategies and Shortcomings of Social Judgment 66–73, 167–92 (1980). Cf. M. Mahoney, Publication Prejudices, I Cognitive Therapy & Research 161 (1977) (journal peer reviewers biased against reviewed manuscripts whose results were contrary to their own theoretical perspectives).Google Scholar

100. In some of the cases, caseworkers kept a diary of all their contacts and interviews; the dictation included reports on the parents' conflicts and the progress of the investigation. In very few cases was the dictation complete. Dictation was much more common concerning postdecretal activities when the trial judge had ordered custody counseling or visitation supervision. See also infra note 240 and accompanying text.Google Scholar

101. This remark is quoted from the caseworker's running dictation. See supra note 100. Unless otherwise indicated, all quotations in this article are from the investigation reports, not from the caseworker's tile dictation.Google Scholar

102. The mother's amenability to counseling and agreement to be counseled may well have played an important, if unprovable and unnoted, role in the caseworker's decision to support the mother as postdecretal custodian: “The counselor's impression of Mrs. H-7 was of a young woman somewhat upset over the failure of two marriages, beginning to have some serious reservations about herself as a marriage risk. The counselor made a strong recommendation that she seek professional help from a social agency in examining herself as to why she is unable to adapt to the stresses and strains of marriage, She has indicated she will do so once the pending case is resolved.” See also infra pp. 769–70.Google Scholar

103. The psychiatrist noted the caseworker's delay and the fact that the investigation report was due shortly. He expressed the “hope that my letter has not arrived too late.” It may well be that the case worker planned to supplement his report orally with this information about the mother-especially if he planned to testify at the trial. But to guide the lawyers preparing for the trial, a written addendum to the report should have been prepared. In “consensual with c/i” cases, in any event, in the absence of supplementation, the caseworker's report is likely to influence bargaining, and thus the custody award, without any opportunity for the parties to learn about information discovered by the caseworker after the report is written.Google Scholar

104. It is certainly possible that in the H-7 case the mother was “really” more concerned about the child than the father and therefore more likely to provide whatever the child might need to prosper in the future; any neutral expert might have been able to discover that the father was “really” seeking only to cause trouble for the mother; the father's day-care prospect may have appeared (to CW/3 or to any other observer) to be unable to cope effectively with another child; the great-grandmother might have been completely qualified and physically competent to provide child care; the caseworker's counseling of the mother may have had such a great initial impact on the mother's sociopathy as to make a prediction of future maturity and child-care excellence appropriate; a “maternal presumption,” applied in this as well as other cases (see infra note 143 and accompanying text) may wisely avoid more decision-making problems than it produces. The possibilities are endless-but the “bottom line” is the same: it is difficult if not impossible to second-guess the caseworker's judgment without having another full investigation done by another caseworker (who may have similar or quite different foibles). See also infra pp. 777–79. Yet failures to report relevant information seem inappropriate even if the recommendation and the outcome are both qaite correct. See also infra pp. 749–52.Google Scholar

The trial judge awarded custody to the mother.Google Scholar

105. See Mnookin, , supra note 19, at 262. See also id. at 227.Google Scholar

106. See generally H. Clark, The Law of Domestic Relations in the United States 584–96 (1968); Foote, Levy, & Sander, supra note 5.Google Scholar

107. See, e.g., H-14 (CW/l) (mother's epilepsy); H-22 (CW/3) (mother's mental illness and frequent hospitalizations); H-11 (CW/2) (husband a paraplegic living with his mother); H-82 (CW/8) (mother had series of venereal disease episodes; now in remission).Google Scholar

108. See, e.g., H-51 (CW/6) (description of series of psychotic episodes); H-53 (CW/2) (hospitalization for emotional problems). But see H-7 (CW/3), supra pp. 742–44 (mother's hospitalization mentioned but not fully reported).Google Scholar

109. See, e.g., H-29 (CW/7), infra, p. 747 (both parents have jobs and either one could maintain a household for the children); H-69 (CW/6) (woman with serious depression improved sufficiently by employment to be awarded custody of her children).Google Scholar

110. See, e.g., H-79 (CW/6); H-35 (CW/5;) (“her relationships with her father and mother were at times stormy and at the same time apparently overprotective and controlling. Two sisters … are divorcees, a situation which may indicate some early childhood emotional deprivations and lack of guidance for all three daughters”); H-75 (CW/7) (“family dynamics” emphasis).Google Scholar

111. See, e.g., H-69 (CW/6) (mother's compulsive cleanliness resulted in father and children retreating to basement to live).Google Scholar

112. See infra pp. 758–59.Google Scholar

113. See, e.g., H-79 (CW/6), discussed infra pp. 745–46.Google Scholar

114. See, e.g., H-29 (CW/7), discussed infra pp. 747–49.Google Scholar

115. See, e.g., H-13 (CW/6), discussed infra p. 746.Google Scholar

116. See, e.g., H-29 (CW/7), discussed infra pp. 747–48 (need of 13-year-old daughter for discipline); H-63 (CW/5) (minority race family with nine children; caseworker checks with school social worker on children's adjustment in school).Google Scholar

117. See, e.g., H-47 (CW/7); H-69 (CW/6); H-79 (CW/6), discussed infra pp. 745–46.Google Scholar

118. See, e.g., H-7 (CW/3), discussed supra pp. 741–44.Google Scholar

119. See, e.g., H-69 (CW/6) (discussing father's bowling and fishing); H-l (CW/9) (mother's interest in civil rights causes); H-13 (CW/6), discussed infra p. 746 (mother's out-of-the-home activities).Google Scholar

120. See, e.g., H-39 (CW/6) (13-year-old daughter's “chief interests were her girlfriends, clothes, records and boys”).Google Scholar

121. See, e.g., H-13 (CW/6), discussed infra p. 746; H-39 (“her relationship with [the younger child] seemed to be quite amiable, although it was apparent the girls were looking to each other for more support than usual while talking with me …. With much vehemence, she described her dislike of, and bitterness toward, her mother”).Google Scholar

122. See H-29 (CW/7), quoted infra pp. 747–49. But see H-13 (CW/6) (children's desire to live with father noted but no custodial recommendation made). See also H-36 (“he … told me, with no such questioning on my part, that he loved both his father and mother and didn't know which he would like to live with”); H-38b (child asked by caseworker how he would respond if asked his preference in court).Google Scholar

123. See, e.g., H-79 (CW/6), quoted infra pp. 745–46.Google Scholar

124. See generally H. Clark, supra note 106, at § 17.4 (1968); Annot., 98 A.L.R.2d 926, 928 (1964).Google Scholar

125. This case also illustrates how difficult it would be either to second-guess the caseworker or to characterize the reports for quantification. See supra pp. 739–41. Is this a report that favors “the child's wishes,” one that respects “family autonomy,” or an illustration of the sophisticated application of notions of “family dynamics”? See also supra note 104.Google Scholar

126. The caseworker also relied on MMPI test results for both spouses. See infra p. 773. The affirmative reasons noted for preferring the father included some obvious makeweights. See infra pp. 763–64.Google Scholar

127. Report quality seems also to decline in two other circumstances: where the contesting parents' claims are closely balanced and where inconsistent but equally valued parenting or behavioral styles are in direct conflict. Given the “indeterminate” quality of custody decision making, see supra note 105, these are likely to be the most difficult cases to resolve-and thus the most likely to produce reports in which the caseworker indulges in the most vigorous (and therefore one-sided) advocacy for his recommendation. But see H-13 and H-79, discussed supra pp. 745–49.Google Scholar

It is possible that the surprisingly extensive attention paid by the caseworkers to the mother's extramarital sexual behavior may be the product of the pressures caseworkers feel in “close” cases to find a “right” answer by appealing to facts thought likely to influence the judge. See infra p. 761. Social psychologists might explain such phenomena in terms of “cognitive dissonance.” See generally L. Festinger, A Theory of Cognitive Dissonance (1957).Google Scholar

128. This discussion assumes the relevance to the custody decision of extramarital sexual activities of both spouses. The issue as presented by the reports is discussed at infra pp. 758–59. Even if such matters as venereal disease were considered irrelevant, the issue as it was presented in H-l might be pertinent because the way the husband dealt with this medical problem raised questions about his character and integrity.Google Scholar

The judge awarded custody of all the children but the teenage daughter, who had expressed a desire to stay with her mother, to the father. According to a memorandum filed by the judge with the divorce decree, the fatherGoogle Scholar

129. See also H-80 where the same caseworker failed fully to support his own recommendation that the mother be awarded custody of all the children. The caseworker does mention some previous physical and sexual abuse by the father but fails to report a specific warning by a local psychiatric clinic that the husband “could become abusive to his wife and possibly the children under stress.” Instead, the report emphasizes the wish of the middle child to be with her mother and the wisdom of refusing to split the children's custody: “Although it is not [the child's] responsibility to decide the matter of custody, I feel that she has made a decision that must be recognized and consideration given to it.”.Google Scholar

The tendency to suppress facts may be strengthened by the caseworkers' view of themselves as counselors and “mediators” rather than detectives. See infra pp. 763–64. for the suggestion that this tendency requires that the spouses' attorneys have access to the investigator's underlying file as a matter of right, see infra note 240 and accompanying text.Google Scholar

130. Imposing personal values was by no means universal. In H-13, supra, pp. 746–47, e.g., CW/6 carefully disclosed her biases and the quandaries they produced and left the ultimate value choice to the judge.Google Scholar

131. As to “indeterminacy,” see Mnookin, supra note 19, at 227. See also Note, Mandatory Mediation, California Civil Code Section 4607, 33 Emory L.J. 749 (1984). For an unusually forthright acknowledgment of this truism, see H. Fain, Our Child Custody Laws and Policies-Are ‘They in Need of Revision or Change? Proceedings, Family Law Section, American Bar Association 29 (1963).Google Scholar

In subsequent sections of this article, a number of factors bearing on either judges' or caseworkers’ decisions in custody cases are discussed separately because they seem so salient (or at least often appeared so) in the reports: the “maternal presumption,” the “disqualification for sexual conduct,” and “emotional and relational criteria.” See infra pp. 753–78. Each of these criteria, as well as their components, is at least arguably only one more illustration of the imposition of a particular decision maker's personal and particularized values on a divorcing population. But many of the criteria discussed below are almost always considered relevant by, and often enjoy widespread (if not universal) acceptance among, a large body of child development and social service professionals. That kind of acceptance is not likely for many (if any) of the personal values reviewed in this paragraph of the text.Google Scholar

132. Even ‘“mainstream’ child development values” are commonly subject to substantial academic and professional disagreement. See, e.g., the discussion of the “maternal presumption,” beginning at infra p. 753, as well as the authorities cited in infra notes 144, 147, & 148. The criticisms in the discussion that follows of caseworkers' use of “mainstream” values does not dispute the relevance of such values or that many professionals, as well as judges, might find them dispositive in individual cases. Rather, the focus is on the propriety of their use under the prevailing doctrinal norms (see, e.g., infra pp. 758–59) and on the selective and (more than occasionally) manipulative use of admittedly relevant doctrines to reinforce a conclusion reached for other reasons (see, e.g., infra pp. 756–57).Google Scholar

133. See, e.g., H-31 (CW/5) (references to personal bankruptcy, which had been arranged two years after marriage, six years prior to divorce proceeding).Google Scholar

134. See, e.g., H-68 (CW/10) (measuring time spent “playing with the children and taking them places” by each parent prior to separation and criticizing both parents for leaving children “in the care of sitters”).Google Scholar

135. See, e.g., H-56 (CW/9).Google Scholar

136. See, e.g., H-36 (CW/8) (wife's taste in clothing and hair color described; husband described as a person with “large physique, who has become too heavy and careless about his physical condition”); H-82 (CW/8) (wife “neatly dressed when she came to my office and has been cooperative”).Google Scholar

137. To my knowledge, in every case in which a person of minority race was involved, either as a litigant or as one involved in some fashion with the child or one of the parents, there was some extraordinary treatment. In H-65, already mentioned, CW/7 sought to bolster her recommendation that the husband be awarded custody by telling the judge that the mother had given birth out of wedlock to a child whose “father … was a Negro.” See supra p. 748. In H-34, the referee had removed the children from the mother at the temporary hearing; CW/6 reported that the wife was living out of wedlock with a black man and recommended that the father be given custody in part because the mother had failed to keep appointments and “by Mrs. H-34′s refusal to let me know her whereabouts, she has, apparently, withdrawn from her pursuit of custody.” In H-9 (CW/5), the report commented that the wife, a young woman who had emigrated from Germany after her marriage to a soldier from Minnesota, was living with an “age 27, young, a divorced negro man. This was verified during an unannounced home visit.” (This was one of the very few unannounced home visits described in the entire sample of reports.) The caseworker emphasized the wife's “independent and defiant nature” and “rebellious attitude” when threatened that she would lose custody of her son “when she became involved in racially mixed relationships,” and reported that the wife “stubbornly stayed with her decision to seek divorce and make friends of her own choosing.” The caseworker also made a full report about allegations of physical abuse of the child by the “stepfather,” including directions as to how to obtain a police report and pictures, despite the fact that an investigation of the allegations by the police had concluded (and the caseworker agreed) that abuse had not occurred. See also H-27 (CW/5) (black family with four young children; father tells caseworker he cannot care for the children and that the mother should have custody; caseworker continues investigation and checks with school social worker as to children's condition and care; report recommends award of custody to mother; despite absence of evidence of neglect, report also recommends that order include compulsory “custodial counseling,” disposition normally reserved for families with relationship or other child care problems). For other examples of juvenile-court-like “child protection-oriented” reports, see infra pp. 782–83.Google Scholar

138. Like all the analyses in this article, another set of reports might lead to different conclusions. But “impressions” of an observer create additional and special risks: such inferences depend to a large extent on the objectivity and “clinical skill” of the observer. Thus, readers should be alert to the possibility that they would characterize these reports and their value emphases in a fashion different from, perhaps inconsistent with, the characterizations I have presented. At the same time, readers tempted to reject my characterizations as “biased” or “distorted,” perhaps to reject the criticisms of custody investigations offered here, should also be willing to acknowledge that it is just such impressionistic reporting and characterization that form the core of the “clinical” style of custody investigations and the reports that are their product. See also infra pp. 763–78.Google Scholar

139. Indeed, in some of the cases the caseworker's apparent effort to impose his own values may well be only an effort to justify an outcome already or thought at a future hearing to be required by a judge or referee. See, e.g., H-65 (CW/7), discussed supra pp. 748–49. See also H-1 (CW/l) and H-61 (CW/3) (report recommended wife despite evidence of extramarital sexual behavior, but judge awarded custody to husband). See also supra note 128.Google Scholar

140. See supra notes 42–45 and accompanying text.Google Scholar

141. See supra notes 68–69 and accompanying text.Google Scholar

142. Cf. Foote, Levy, & Sander, supra note 5, at 886. See also infra notes 216–26 and accompanying text.Google Scholar

143. 280 Minn. 444, 160 N.W.2d 30 (1968).Google Scholar

144. Id. at 447, 160 N.W.2d at 32. The statute was later amended to prohibit any favoritism bated on the gender of the custodian. See 1969 Minn. Laws 1030, § I. The provision was amended to read: “In determining custody, the court shall consider the best interests of the child and shall not prefer one parent over the other soleiy on the basis of the sex of the parent.” Minn. Stat. 518.17 (3) (1984). For general discussions of the “maternal presumption” problem, see D. Chambers, Rethinking the Substantive Standards for Child-Custody Disputes in Divorce, 83 Mich. L. Rev. 477 (1985); Mnookin, supra note 19; Ellsworth & Levy, Legislative Reform of Child Custody Adjudication: An Effort to Rely on Social Science Data in Formulating Legal Policies, 4 Law & Soc'y Rev. 167 (1969). Clark, supra note 106, at 584–85, reports that “today, generally by case law and sometimes by statute, both spouses are given equal right to the custody of their children.” The statutes and judicial standards are collected in Foster & Freed, Life with Father, 1958, 11 Fam. L.Q. 321, 343 (1978).Google Scholar

145. I t may be appropriate to issue another warning that the conclusions reported in the text are made on the basis of my own subjective assessment of some hut by no means all of the comments made by the caseworker in the report. Such inferences, no matter how “educated” or sophisticated, are notably unreliable. See supra note 138. See also supra pp. 741–42.Google Scholar

146. See supra H-22, p. 755.Google Scholar

147. In addition to the authorities cited supra note 144, see Mnookin, & Kornhauser, , supra note 70. Weitzman & Dixon, Child Custody Awards: Legal Standards and Empirical Patterns for Child Custody, Support and Visitation After Divorce, 12 U.C.D. L. Rev. 473 (1979), concluding that the proportion of father-custodians did not increase substantially in either San Francisco or Los Angeles after the California legislature abolished the maternal presumption in 1973, inferred on the basis of interviews that practicing lawyers played a large role in perpetuating the presumption in practice.Google Scholar

148. In addition to the authorities cited supra notes 144 & 147, see Comment, The Tender Years Presumption: Is It Presumably Unconstitutional? 21 U. San Diego L. Rev. 861 (1984). In recent years some feminist authors have argued for a reinvigoration of the maternal presumption. See Klaff, R., The Tender Years Doctrine: A Defense, 70 Calif. L. Rev. 335 (1982); Uviller, R., Father's Rights and Feminism: The Maternal Presumption Revisited, 1 Haw. Women's L.J. 107 (1978).Google Scholar

149. For a review of what the child development literature suggests as sensible substantive standards, rejection of the maternal presumption, but support for a mild preference favoring the predivorce “primary parent,” see Chambers, supra note 144.Google Scholar

150. See Ellsworth & Levy, supra note 144.Google Scholar

151. For examples of cases where the maternal presumption was acknowledged but not followed, see, e.g., Fish v. Fish, 280 Minn. 316, 159 N.W.2d 271 (1968) (trial court did not abuse discretion in awarding custody of son to father because of their close relationship even though the mother was fit and the custody of young children ordinarily should be awarded to the mother); Stoll v. Stoll, 243 Minn. 510, 68 N.W.2d 367 (1955) (maternal presumption not decisive because of mother's mental instability); Haskell v. Haskell, 279 N.W.2d 903 (S.D. 1979) (mother's conduct and failure adequately to supervise child sufficient to overcome statutory preference for mother as custodian; preference operates only when other things are equal and is subservient to best interests of child).Google Scholar

152. See, e.g., Mnookin, supra note 19.Google Scholar

153. See the discussion of H-65 supra, pp. 748–49.Google Scholar

154. See supra note 104.Google Scholar

155. See supra note 143.Google Scholar

156. See, e.g., Bunim v. Bunim, 298 N.Y. 391, 83 N.E.2d 848 (1949). See generally Foote, Levy, & Sander, supra note 5, at 858–60.Google Scholar

157. Reiland v. Reiland, 280 Minn. 444, 447–48, 160 N.W.2d 30, 32–33 (1968).Google Scholar

158. Again a warning: the judgment is impressionistic and “clinical”; it is impossible to say in any of the cases whether the exclusion was decisive. See supra note 105 and accompanying text.Google Scholar

159. See H-24 (CW/S), quoted supra p. 756. Other parts of this report are quoted infra, text at note 168. See also the discussion of H-9 (CW/S), supra note 137.Google Scholar

160. See also H-9 (CW/5), described supra note 137.Google Scholar

161. See also H-65 (CW/7), described supra pp. 748–49.Google Scholar

162. See also supra note 159.Google Scholar

163. See, e.g., H-1, described supra pp. 750–51.Google Scholar

164. See H-65, discussed supra pp. 748–49; H-34 (CW/6), supra p. 760.Google Scholar

165. See H-59 (CW/2).Google Scholar

166. See supra notes 137, 146, 159, & 162.Google Scholar

167. See, e.g., Reiss, How and Why America's Sex Standards Are Changing, in Perspectives in Marriage and the Family 397 (J. Eshleman ed. 1969) (changes are more in attitudes toward sex than in sexual behavior and are marked by a trend toward greater egalitarianism, more open discussion of sex, and greater female acceptance of permissiveness); Smigel & Serden, The Decline and Fall of the Double Standard, 376 Annals 6 (1968).Google Scholar

168. See, e.g., Levy, Uniform Marriage and Divorce Legislation: A Preliminary Analysis 226–27 (1968) (monograph prepared for Committee of National Conference of Commissioners on Uniform State Laws as introduction for drafting Uniform Marriage and Divorce Act). Uniform Marriage and Divorce Act § 402 (1970) contained the following clause: “The court shall nor consider conduct of a proposed custodian which does not affect his relationship to the child.” The comment to the section indicated that the commissioners intended to minimize the impact of extramarital behavior on custody decision making. See Foote, Levy, & Sander, supra note 5, at 859.Google Scholar

169. See, e.g., Schiele v. Sager, 174 Mont. 533, 571 P.2d 1142 (1977) (despite legislature's failure to adopt “no-fault” clause of custody provision, supra note 168, in enacting Uniform Marriage and Divorce Act. case authority has adopted precedent similar to the evidentiary rule expressed by the excluded clause); Ouellette v. Ouellette, 246 Md. 604, 229 A.2d 129 (1967) (although custody of children usually awarded to innocent party, mother's adultery is not an absolute bar where maternal custody would be in best interest of children).Google Scholar

170. See, e.g., H-1, described supra pp. 749–50. In some of the cases, a mother's sexual exploits known to the caseworker may not have been mentioned at all. Cf. the discussion of fact suppression in H-l supra. In still other cases, the caseworker may have decided not to make the investigation or ask the questions that would have disclosed sexual exploits. This much is certain: with the exception of H-1. supra, in which evidence of the husband's sexual exploits came to the caseworker by a chance letter from a doctor about a venereal disease, discussion of the husband's behavior in the reports is conspicuously absent. It is possible, of course, that divorcing husbands in Hennepin County during 1970 engaged in substantially less extramarital sexual behavior than divorcing wives; but a more likely hypothesis is that caseworkers were less interested in husbands' than in wives’ exploits.Google Scholar

171. It may well be that the problem posed by the disqualification for sexual exploits is less serious now than it was in 1970. For the suggestion that in the trial courts the disqualification has not been abolished, see Wexler, Rethinking the Modification of Child Custody Decrees, 94 Yale L.J. 757 (1985). See also Pearson & Ring, supra note 14, at 703, 714 (second most frequently mentioned factor in Denver County custody investigation reports during 1973–77 was “sexual misconduct of the contesting parties”). An analogous problem, this one posing difficulties for fathers, may have been created by the recent publicity given to the prevalence of sexual abuse of children by their fathers. See Gordon, C., False Allegations of Abuse in Child Custody Disputes, 135 New L.J. 687 (A.B.A. Special Issue 1985).Google Scholar

172. See supra note 81 and accompanying text.Google Scholar

173. See, e.g., the cases cited supra notes 107 & 108.Google Scholar

174. See generally J. Ziskin, Coping with Psychiatric and Psychological Testimony (2d ed. 1975); Morse, , Failed Explanations and Criminal Responsibility: Experts and the Unconscious, 68 Va. L. Rev. 971, 992 (1982);Gardner, , The Myth of the Impartial Psychiatric Expert—Some Comments Concerning Criminal Responsibility and the Decline of the Age of Therapy, 2 Law & Psychiatry Rev. 99 (1976). See also Litwack, Gerber, & Fenster, supra note 41, quoted in part in the text accompanying note 41.Google Scholar

For an illustration of the difficulties of psychological diagnosis and prediction in the 1970 Hennepin County custody investigation case sample, consider the report of CW/6 in H-69, a contested case in which there was a serious question as to the wife's mental stability:Google Scholar

Exactly one year later, CW/7, writing a new report in response to the mother's motion to modify the custody decree to award the daughter to her, wrote:.Google Scholar

175. See supra notes 39–43 and accompanying text.Google Scholar

176. See supra note 41 and accompanying text.Google Scholar

177. For an illustration of the technique of finding covert meanings that contradict the client's stated position, as well as the ease with which professional colleagues rely on each other's conclusions, see Landsman & Minow, supra note 8, at 1147 (1980), quoted in part in supra note 13, describing a case in which a lawyer contended that the father should be given visitation despite the fact that his client, the child, told him he did not want to visit his father, because the custody investigator reported that the child “really did want to see his father.” But see Pearson & Ring, supra note 14, at 723 (judges interviewed “were especially critical of the private, psychological reports that were extremely jargon-laden and evaluative”).Google Scholar

178. See H-45, quoted supra pp. 768–69.Google Scholar

179. Or perhaps the client's contact with a clinically trained lawyer who deals with the “nonlegal,” the personal and emotional, aspects of the client's divorce. For the contention that lawyers should deal with their clients' personal problems in a therapeutic fashion, see Foote, Levy, & Sander, supra note 5, at 1007–22.Google Scholar

180. See H-24 (CW/5), quoted supra p. 756. See also H-31, quoted infra p. 787.Google Scholar

181. See H-35 (CW/S), quoted supra p. 765.Google Scholar

182. See also H-7 (CW/3) supra pp. 741–43 & note 102.Google Scholar

183. As to continuing counseling and supervision by the Department of Court Services, tee infra pp. 782–90.Google Scholar

184. H-26 (CW/1) (from a memorandum to the referee in Feb. 1970).Google Scholar

185. H-26; memorandum, Sept. 1970, from CW/1 to family court referee.Google Scholar

186. Id. See also H-19 (CW/10):.Google Scholar

187. See generally W. G. Dahlstrom & G. Welsh, An MMPI Handbook: A Guide to Use in Clinical Practice and Research (1960). See also infra note 191.Google Scholar

188. The scale was designed “to identify the personality features related to the disorder of male sexual inversion.” Dahlstrom & Welsh, supra note 187, at 63. Nonetheless, “the Manual for the MMPI also makes it clear that, ‘it should be continually kept in mind that the great majority of persons having deviant profiles are not in the usual sense of the word mentally ill or are they in need of psychological treatment. Having no more information about a person than that he has a deviant profile one should always start with the assumption that the subject is operating within the normal range.”’ Ziskin. supra note 168, at 160.Google Scholar

189. See, e.g., H-61; H-75. See also H-53 (report refers judge to MMPI report in caseworker's file).Google Scholar

190. Id. CW/3, consistent with his adherence to the maternal presumption (see supra p. 156), recommended that the mother be awarded cut tody of the five-year-old daughter. The report's conclusion emphasized the mother's status as temporary custodian and the child's need for stability. The trial judge awarded custody to the father—apparently disqualifying the mother because of extramarital sexual conduct. Two years later the mother regained the child's custody.Google Scholar

191. See, e.g., C. Deinhardt, Personality Assessment and Psychological Interpretation 169 (1983): “The MMPI seems fairly well able to differentiate those who have from those who do not have emotional and adjustment problems, in a wide variety of settings. But there is conflicting evidence regarding its ability to significantly differentiate either within the normal range or within the abnormal range”; Ziskin, supra note 174, at 160: “the test has modest validity for distinguishing one kind of group from another in terms of pathology, as its manual states, ‘A high score on a scale has been found to predict positively the corresponding final clinical diagnosis or estimate in more than 60% of new psychiatric admissions.’ Simple arithmetic tells us that it obviously fails to predict in something near 40% of the cases. Query the usefulness of such predictive inefficiency in a court of law?” W. G. Dahlstrom, Personality Measurement: MMPI, in 12 International Encyclopedia of the Social Sciences 43, 47 (D. Sill ed. 1968): “Thus, it appears that even though the various MMPI indexes are useful when applied to separate and discrete decisions about patients or clients, there remains unresolved the additional problem of integrating the multitude of possible implications of a particular profile and producing a coherent and accurate personological summary of that test subject.”.Google Scholar

192. See infra note 206 and accompanying text.Google Scholar

193. See Olson et al., supra note 9. This monograph, commissioned by the division and funded by a local foundation, contained a study that purported to distinguish custody resolution counseling (CRC) from custody investigations (CS) on the basis of analyses of the responses of 15 caseworkers to a 19–page questionnaire prepared by the research team after a “full day meeting of the counseling and research groups so that they could become acquainted with each other.”Id. The research group concluded that CRC and CS were different enterprises by comparing standard deviations of the caseworkers' averaged responses to a series of questions about a variety of techniques used by the staff in each field of activity. One of the study's comparisons of CRC and CS techniques, this one of counselor “feedback” and interpretation techniques, is quoted at length in appendix 1.Google Scholar

As the excerpt in the appendix shows, the actual differences between averaged responses were quite small. It is remarkable that a survey study, conceived in collaboration with a group of subjects interested in the conclusion that two techniques differ substantially, was able to come only to such modest conclusions about the differences between those techniques.Google Scholar

In a 1984 report on the Domestic Relation Division's custody resolution counseling services, prepared by Cauble, Thoennes, Pearson, & Appleford with the help of a member of the research staff from the 1979 study, no mention was made of the earlier study. Ignoring the training in the 1960s provided the division by Virginia Satir (see supra, text accompanying note 81), the 1984 report (whose interest is “mediation”) gives the impression that the division's custody-resolution counseling activities began in 1975 with a visit by a group of the professional staff to Wisconsin family therapist Carl Whitaker. See Cauble, Thoennes, Pearson, & Appleford, A Case Study: Custody Resolution Counseling in Hennepin County, Minnesota, in Pearson & Thoennes, Final Report of the Divorce Mediation Research Project 3 (mimeographed, Nov. 15, 1984).Google Scholar

194. See, e.g., Kelly, supra note 10, at 33.Google Scholar

195. H-20 (CW/4); H-26 (CW/1), described supra p. 112.Google Scholar

196. See supra pp. 779–80.Google Scholar

197. See supra pp. 778–82.Google Scholar

198. See Ash & Guyer, supra note 72, at —. “While encouraging parents to negotiate a resolution is a form of mediation, the ground rules were quite different from the more usual form of divorce mediation in which the mediator limits his role to being a facilitator who will not formulate his own recommendation to the court in the event mediation is unsuccessful.”Id. at —.Google Scholar

199. Id. at —. The New York Law Revision Commission's proposal contemplates that the threat of an evaluation will facilitate agreements. See supra note 75. Cf. the disputes in the mediation literature on whether the mediator should “preside over” or “facilitate” settlements. See, e.g., the authorities cited supra note 49. For an indication that social service employees “conciliate” by frightening their clients with the “horrors” of court hearings, see Beecher, H., Custody and Counselling Under the Australian Family Law Act—Some Sociological Observations, 18 Australian J. Soc. Issues 98, 100 (1983).Google Scholar

200. H-26. The effort is described and quoted supra pp. 778–82. Some caseworkers whose investigation reports are described and quoted in this article are still members of the department staff-conducting “custody studies” and “custody resolution counseling.”.Google Scholar

201. See Levy, , supra note 92. See also Mnookin, supra note 19, at 276, 288.Google Scholar

202. That bifurcation has been accomplished by the department. See supra note 193. For the suggestion that such bifurcations cannot cure “leakage,” see Beecher, supra note 199, at 105–6.Google Scholar

203. In 1984, at the instance of the Department of Court Services, the Minnesota legislature provided that in the absence of written waiver, mediation conferences are entitled to confidentiality even for department custody investigator colleagues of the mediators. See Minn. Stat. Ann. § 518.167 (Supp. 1985). Even if the nondisclosure rule is rigorously followed, the spouses may be too willing to waive the right in response to a request by a custody investigator. See supra notes 190 & 199 and accompanying text.Google Scholar

204. See supra note 199 and accompanying text.Google Scholar

205. See infra pp. 793–94.Google Scholar

206. See Mnookin, , supra note 19, at 229–30. The article contains a powerful argument for distinguishing the child-protection and the private dispute-settlement function and for using as a starting point the notion of family autonomy, an argument that is much more fully articulated than the one offered here. See also id. at 265:.Google Scholar

207. Cf. Smith v. OFFER, 431 US. 316 (1977).Google Scholar

208. As the cases cited supra notes 52–5 indicate, the conceptual distinction between private dispute settlement and child protection may not be as sharp in reality as either Mnookin or I have made it appear. Thus, judges have the authority to refuse to accept a custody disposition the parents have consensually arranged. In fact, they do so infrequently. See Foote, Levy, & Sander, supra note 5, at 882–83. Most commentators urge judges to defer to the spouses' decision. See Mnookin & Kornhauser, supra note 69, at 995 (1979); Mnookin, supra note 19, at 288; Pilpel, & Zavin, , Separation Agreements: Their Function and Future, 18 Law & Contemp. Probs. 33, 35 (1953).Google Scholar

There have been, and no doubt will continue to be, compromises made with the conceptual distinction in the formulation of specific doctrines. In some states, for example, judges can award “joint custody” to both parents despite parental opposition. See Foote, Levy, & Sander, supra, at 887–900. Judges have the authority to choose as a custodian someone other than one of the parents; yet that authority is usually fairly narrowly circumscribed in appellate decisions and sometimes by statute. See generally id. at 900–912. Judges can place limitations on the noncustodian's right to visit the children or even deny visitation completely. But, again, both judicial decisions and legislative provisions have circumscribed the trial judge's authority to limit parental privileges in this fashion. Id. at 916–18. And judges have from time to time imposed limitations on custodial parents' right to exercise control of their children—also within relatively narrow limits imposed by the appellate courts or legislatively. See id. at 919–20. The extent to which trial judges exercise only the powers they have been given (in divorce or in other areas of discretionary authority) is seldom studied systematically. The data provided here will no doubt suggest to some readers that the private dispute settlement-child protection dichotomy is a myth—an ideal we enjoy mouthing in public while we allow government officials to behave in a fashion inconsistent with the ideal. Cf. T. Arnold, The Folklore of Capitalism 207–29 (1937). The recommendations made infra pp. 793–94 assume that the dichotomy should be maintained and strengthened.Google Scholar

209. See supra notes 75 & 76. See generally sec. IB of this article, pp. 718–28.Google Scholar

210. H-85 (CW/10). If the father were a real and viable candidate for custody of the child, an argument could be made that it would be proper for the judge to announce his intentions in advance and allow the spouses to make decisions about their behavior and circumstances accordingly. Thus, the judge might say that he would be disposed to give custody to the mother, but because of concerns about the mother's parents' home he intended to grant custody to the father unless the mother changed her residence. Under these circumstances the mother could make her decision without direct coercion. It is not necessary to decide whether such a policy should be opposed because it allows too many opportunities for subtle judicial coercion in a case like H-85—because there was no other real and viable candidate for custody. See also infra pp. 789–90.Google Scholar

See also H-14 (CW/1) (“emergency” custody investigation ordered at temporary hearing, with report to be delivered in three days, “particularly with respect to the likelihood of any injury to the minor children as a result of the plaintiff's present physical or emotional condition”—where plaintiff-mother was living with and caring for the children but had minor epileptic seizures while asleep as a result of surgery to remove temporal lobe cyst).Google Scholar

211. In some of these cases the caseworkers may not have known that the parents would settle their disagreement about the custodian. But in some cases the caseworker was obviously aware that the case would be settled. See, e.g., notes 210–12 and accompanying text. In 13 of the 21 Hennepin County “contested with custody investigation” cases for which I have records, the caseworker recommended some form of postdecretal involvement by the Department of Court Services. The proportions are similar to those for consensuals.Google Scholar

212. No “severe problems” were mentioned in the report other than those quoted in the text. In fact, the language quoted in the text is the entire report other than a two-sentence paragraph describing the mother's employment situation. See also supra note 137.Google Scholar

213. For other cases in which caseworkers recommended postdecretal counseling despite a report expressly finding that either parent would make a competent custodian, see H-36 (CW/8), H-57 (CW/10), and H-61 (CW/3).Google Scholar

214. These paragraphs provided the only factual information on this case other than a letter to the foster parents, dated three months later, thanking them for their services and indicating that the boys were now living with their mother. A close reading of the quoted text of the report suggests that CW/4's counseling may have played a role in the parents’“mutual agreement” to place the boys in foster care.Google Scholar

215. But see Wallerstein, J. & Kelly, J., Surviving the Breakup: How Children and Parents Cope with Divorce (1980) (children of divorced parents referred to counseling and study project in one upper middle-class county in California and followed longitudinally for five years showed substantial emotional problems resulting from divorce). But see C. Bruch, Parenting at and After Divorce: A Search for New Models, 79 Mich. L. Rev. 708 (1981) (reviewing Wallerstein and Kelly's book and questioning representativeness of sample).Google Scholar

216. See supra notes 126 & 127 and accompanying text.Google Scholar

217. See Pearson & Ring, supra note 14, at 703–22.Google Scholar

218. Id. at 723.Google Scholar

219. See supra note 131 and accompanying text.Google Scholar

220. See generally Doing Good: The Limits of Benevolence (1978) (articles by W. Gaylin, S. Marcus, D. Rothman, and I. Glasser).Google Scholar

221. See supra note 86, indicating that of the ten caseworkers whose investigations and reports are considered in this article, five are still members of the Department of Court Services staff 15 years later.Google Scholar

222. See supra notes 206–16 and accompanying text.Google Scholar

223. See supra notes 16–17 & 73–75 and accompanying text.Google Scholar

224. See supra notes 34, 69, & 201 and accompanying text.Google Scholar

225. See supra notes 6–10 & 193–201 and accompanying text.Google Scholar

226. Very little data are available that would permit fair comparisons of settlement rates between jurisdictions. See Foote, Levy, & Sander, supra note 5, at 872–76. No studies have systematically distinguished contested cases from contested stipulations. For some indication of the disputes about samples to which the claims of mediation advocates have led, see Levy, supra note 91, at 526.Google Scholar

227. “Adversarial” is the term commonly used in the mediation literature to distinguish “traditional” divorces from those that have been subjected to professional mediation. But see Levy, supra note 201.Google Scholar

228. See supra notes 69 & 197–204 and accompanying text. If the agency successfully divorces the mediation from the investigation endeavor, of course, investigations alone will effect fewer settlements. See supra notes 202–4 and accompanying text.Google Scholar

229. See supra note 69.Google Scholar

230. See supra note 217 and accompanying text.Google Scholar

231. For some indication that judges will make use of mental health experts outside the court system if given the opportunity (and if the cost can be covered), see Ash & Guyer, supra note 72.Google Scholar

232. See supra note 41 and accompanying text. Litwack, Gerber, and Fenster believed that “psychological interviews and tests may unearth evidence of a severe, but well-concealed psychological disturbance in a contesting adult,” such as “the belief that all men (but not women) are controlled by the devil”—if the interviewer is “sensitive.” They seemed less enthusiastic about “psychiatric diagnoses.”Id.Google Scholar

233. See supra notes 198–99 and accompanying text.Google Scholar

234. See Uniform Marriage and Divorce Act 5 404(b) (1970): “The court may seek the advice of professional personnel whether or not they are employed on a regular basis by the court.” Although the report must be made available to counsel on request and the professional person may be called for cross-examination, id., the professional person's file of underlying data does not have to be made available as S 405(c) requires in the case of social work investigations.Google Scholar

235. Legislatures committed to “nonadversarial” approaches to divorce litigation may in any event be unwilling to eliminate custody investigations and the bureaucracies that conduct and defend them. Nonetheless, I believe that the recommendation in the text is proper even if “political imperatives” could he ignored.Google Scholar

236. Fanshel, D., Approaches to Measuring Adjustment in Adoptive Parents, in Quantitative Approaches to Parent Selection 18, 25 (Child Welfare League of America, 1962).Google Scholar

237. See supra notes 174–91 and accompanying text.Google Scholar

238. See supra note 205 and accompanying text.Google Scholar

239. In some contexts, it may be appropriate to make the assumption that lawyers will not adequately protect their clients—e.g., when the issue is whether a lawyer should be permitted to waive his juvenile client's Miranda rights. See Feld, B., Criminalizing Juvenile Justice: Rules of Procedure for the Juvenile Court, 69 Minn. L. Rev. 141, 183–90 (1984). Since some divorce litigants may be too easily persuaded to agree to a judge's request for consent to an investigation, the consent provision of the investigation statute could require that the judge formally notify the spouses on the record of their right to refuse.Google Scholar

A lawyer's professional responsibility to a custody claimant client and the client's children warrants more careful analysis than the subject of this article justifies. Nonetheless, note that one of the reasons given above for perpetuating custody investigations is that an investigation will occasionally allow a lawyer to persuade the client to give up a claim to custody. For the lawyer, the line between serving and undermining the client's interests in such situations is not always a clear one.Google Scholar

240. See supra note 100 and accompanying text for a description of the term. The admissibility of the report in the event of a trial might be conditioned on the caseworker's maintenance and disclosure of the running dictation file. Obviously, if the lawyers are to have automatic access to the caseworker's record, it would not be appropriate to require them to prove “good cause” to obtain such access. But see the proposal of the New York Law Revision Commission, supra note 129 and accompanying text.Google Scholar

241. See supra notes 4, 61, & 203.Google Scholar

242. D. Olson et al., Custody Resolution Counseling (CRC): Description and Comparison with Custody Study (CS), in Child Custody: Literature Review and Alternative Approaches (mimeographed 1979). See supra note 193 and accompanying text.Google Scholar

243. Id. at 116–17.Google Scholar

244. Id. at 112–113, 127.Google Scholar