Opinion
KLINE, P. J.Monte Mentry (father) appeals from that portion of a restraining order which prohibits him from engaging the children of the dissolved mar*262riage in any religious activity, discussion, or attendance during visitations and from providing them with articles, publications, or other religious material, while they are in his presence. We reverse due to the absence of evidence of harm to the children and our resulting belief that the order represents an unwarranted intrusion into family privacy.
The Facts
The facts are undisputed. The marriage between the father and Betty Mentry (mother) was dissolved in 1979. They had two children, Sherry and Steven, who at that time were seven and six years old, respectively. During the marriage, both parents were observant members of the Church of the Latter-Day Saints (Mormon). The wife separated from the church at about the same time she separated from her husband, and joined a different church.1 The interlocutory decree, which awarded the father visitation rights, made no mention of the religious issue but in April 1980, when he sought to expand his visitation rights, the mother countered by seeking an order enjoining him from requiring the children to engage in any religious activities other than those approved by her. In her declaration in support of the order, the wife stated that during a session with the court conciliator the father had previously agreed to such a limitation but that he had violated the agreement.
At the hearing on the order to show cause, the father, the mother and Warren Weiss, the court conciliator, each presented different versions of the agreement the father allegedly violated. The court made clear its view, however, that the propriety of an order such as that requested was to be determined under the rule prescribed in In re Marriage of Murga (1980) 103 Cal.App.3d 498 [163 Cal.Rptr. 79], which provides that “. . . a court will not enjoin the noncustodial parent from discussing religion with the child or involving the child in his or her religious activities in the absence of a showing that the child will be thereby harmed.” (Id. at p. 505.)
On the question of harm, Mr. Weiss testified that in his opinion such conduct upon the part of appellant would confuse the children and thus would be harmful to them:
*263“Q. And at one time do you recall if they reached—first of all, did you have a suggestion or a recommendation on the subject of religion as concerns these two parties?
“A. [Mr. Weiss] Well, in the general context I suggested to them that they not burden the children with conflicting notions as to the issue of religion.
“Q. Why did you suggest that they not burden the children with the conflict of religion?
“A. Well, religion was only one of a number of conflicts the children were having relative to the parents concerning their polarity as people, as human beings. And religion was one more thing they didn’t need to have a conflict about.
“Q. Were you concerned that it would actually confuse the children as well?
“A. Certainly as to that confusion.
“Q. Thank you. And in your—
“The Court: If it would confuse them, do you think that would be harmful to the children?
“The Witness: Yes, I do.”
It is clear from cross-examination, however, that Mr. Weiss had neither seen nor talked with the children at any time and that his information derived entirely from conversations with the parents. As the following colloquy on cross-examination demonstrates, his views on the question of harm to the children were essentially conjectural.
“Q. And that to state that there may be harm would be a generalization only depending on whether you [Weiss] had analyzed these children and the facts in this particular case?
“A. Yes.
“Q. And so there is also a possibility, then, Mr. Weiss, there would not be harm?
“A. There is that possibility.
“Q. And that even if there were confusion, there may not be harm?
“A. May not be, that is correct.
*264“Q. That hasn’t been delved into by your department or anyone else you know of?
“A. That is right.”
The mother testified that on one occasion she learned from Sherry that the father had taken her to a Mormon movie, but told her not to tell her mother; and that on another occasion she found Sherry doing a crossword puzzle in a Mormon publication, and Sherry did not know what some of the words meant. She also testified that Steven said, at times when he was in distress, angry, or discouraged with his mother: “I am a Mormon. I want to be a Mormon. I want to be like my Daddy. ...” Finally, she testified that the children had been confused about doctrinal differences between the two religions.
Discussion
In re Marriage of Murga, supra, 103 Cal.App.3d 498, is the only California decision that defines a standard for determining whether a custodial parent may enjoin the noncustodial parent from discussing religious subjects with the child or from involving the child in the noncustodial parent’s religious activities.2 In holding that the noncustodial parent may not be so enjoined “in the absence of a showing that the child will be thereby harmed” (103 Cal.App.3d at p. 505), the court relied in part on the public policy of this state “to assure minor children of frequent and "continuing contact with both parents after the parents have separated or dissolved their marriage, . . .” (Civ. Code, § 4600, subd. (a).) Given the very general nature of this legislative intendment and the dearth of pertinent California cases, the court also looked to the case law of other states. It found that “. . . in the majority of American jurisdictions that have considered the question, the courts have refused to restrain the noncustodial parent from exposing the minor child to his or her religious beliefs and practices, absent a clear, affirmative showing that these religious activities will be harmful to the child.” (Murga, supra, 103 Cal.App.3d at pp. 504-505, citing, as examples, Munoz v. Munoz (1971) 79 Wn.2d 810 [489 P.2d 1133]; Paolella v. Phillips (1960) 27 Misc.2d 763 [209 N.Y.S.2d 165]; Wojnarowicz v. Wojnarowicz (1958) 48 N.J.Super. 349 [137 A.2d 618]; Boergerv. Boerger (1953) 26 N.J.Super. 90 [97 A.2d 419]; see Annot., Divorce—Visitation Rights (1963) 88 A.L.R.2d 148, 217-219; Annot., Custody of Child-Religion as Factor (1959) 66 A.L.R.2d 1410.) It was also noted in Murga that adoption of the rule followed in the majority of other states “would be consonant with the rule consistently followed by our courts that custody decisions will not be governed by the religious tenets or practices of parents absent a clear showing *265that the parent’s religious practices would be harmful to the child.” Murga, supra, 103 Cal.App.3d at page 505. (See, e.g., In re Marriage of Urband (1977) 68 Cal.App.3d 796, 797-798 [137 Cal.Rptr. 433]; and Cory v. Cory (1945) 70 Cal.App.2d 563, 571 [161 P.2d 385]; see also, Note, The Religious Upbringing of Children After Divorce (1980) 56 Notre Dame Law. 160.)
The best recent synopsis of the case law of the jurisdictions in which the courts have addressed the question whether or how to accommodate diverse religious practices of parents, living apart, in the upbringing of minor children is set forth in Felton v. Felton (1981) 383 Mass. 232 [418 N.E.2d 606, 22 A.L.R. 4th 961]. In that case, as here, a trial court entered an order prohibiting a father with visitation rights from instructing the children in his religion. In a notable opinion by Justice Kaplan, the Supreme Judicial Court reversed, holding that the evidence was insufficient to support such a disposition. Relying upon Murga and numerous cases from other states, the court summarized the law as follows: “The parents together have freedom of religious expression and practice which enters into their liberty to manage the familial relationships. [Citations.] But the ‘best interests’ of the child are to be promoted, and when the parents are at odds, the attainment of that purpose may involve some limitation of the liberties of one or other of the parents. [Citations.] However, harm to the child from conflicting religious instructions or practices, which would justify such a limitation, should not be simply assumed or surmised; it must be demonstrated in detail. [Citations.] [f] If the dominating goal of the enterprise is to serve a child’s best interests . . . then it might be thought to follow that a policy of stability or repose should be adopted by which the child would be exposed to but one religion (presumably that of the custodial parent) at whatever cost to the ‘liberties’ of the other parent. The law, however, tolerates and even encourages up to a point the child’s exposure to the religious influences of both parents although they are divided in their faiths. This, we think, is because the law sees a value in ‘frequent and continuing contact’ of the child with both its parents [citation] and thus contact with the parents’ separate religious preferences. There may also be a value in letting the child see, even at an early age, the religious models between which it is likely to be led to choose in later life. And it is suggested, sometimes, that a diversity of religious experiences is itself a sound stimulant for a child. [Citation.] In all events, the question that comes to the courts is whether, in particular circumstances, such exposures are disturbing a child to its substantial injury, physical or emotional, and will have a like harmful tendency for the future. [Citation.] The critical literature warns against perverting a quest for the child’s best interests into one for the psychic comfort of the parents—a warning against overvaluing the parents’ constitutional liberties. [Citation.] A warning is equally in order against depriving a parent of all connection with the child, or connection on the religious plane, out of an exaggerated fear of injury to the child. [Citations.] It *266is often said that if accommodation appears necessary, that form should be sought which intrudes least on the religious inclinations of either parent and is yet compatible with the health of the child.” (Felton v. Felton, supra, 418 N.E.2d at pp. 607-608, italics added, fns. omitted.)
Reviewing the evidence presented in the trial court, the Massachusetts high court noted particularly the “failure of proof about [the child’s] physical and emotional condition or about any causal connections between her visits with her father and that condition, such as it may have been. General testimony by [the mother] that the child was upset or confused . . . will not suffice. [Citations.] . . . Thus there is clear error, for lack of foundation in the record, in the judge’s findings of a ‘deleterious effect’ on the children and an ‘undermining’ of the custodial relationship by reason of the father’s religious instruction or practice.” (418 N.E.2d at p. 610.)
Here, too, the evidence is manifestly insufficient. Boiled down, it simply consists of testimony by the mother concerning her distress at appellant’s religious activities with the children, and speculative testimony by a counselor who had never seen or interviewed the children. There was evidence that the boy had social adjustment problems in school for which he had received psychological counseling, but no evidence relating these problems, or periodic stomach aches which both children assertedly suffered, to conflict over religion. There is no evidence of any other physical, or emotional, problems. Indeed, the mother conceded that aside from the stomach aches and some aggressive behavior of the son, “the children are basically physically and mentally well adjusted” and are both “normal... in every aspect.” The record contains no competent evidence that the children will in the future be harmed by the parental religious conflict. Respondent has thus failed to meet her burden of “a clear affirmative showing that these religious activities will be harmful to the children].” (In re Marriage of Murga, supra, 103 Cal.App.3d at pp. 505-506.)
The foregoing finding would be sufficient to dispose of this case. There are, however, additional considerations implicit in the Murga rule that also affect our decision; considerations that focus as much upon the proper reach of judicial authority as upon the facts of any particular case.
Murga, Felton and most of the cases they rely upon reflect a salutary judicial disinclination to interfere with family privacy without the evidentiary establishment of compelling need. (See also In re Carmaleta B. (1978) 21 Cal.3d 482, 489,495-496 [146 Cal.Rptr. 623, 579 P.2d 514] and cases there cited.) This attitude, which has been maintained against the urgings of some commentators who favor expansion of the legal grounds for intervention and a stronger *267judicial role in the supervision of parents,3 is predicated, inter alia, on the recognition that a “. . . court cannot regulate by its processes the internal affairs of the home. . . . The vast majority of matters concerning the upbringing of children must be left to the conscience, patience, and self restraint of father and mother. No end of difficulties would arise should judges try to tell parents how to bring up their children.” (People ex rel. Sisson v. Sisson (1936) 271 N.Y. 285, 287 [2 N.E.2d 660, 661], see also, Kilgrow v. Kilgrow (1958) 268 Ala. 475 [107 So.2d 885] and Lynch v. Uhlenhopp (1956) 248 Iowa 68 [78 N.W.2d 491].) Persuasive arguments have been advanced in support of this noninterventionist approach and of the need to consider family privacy and parental autonomy relevant factors in the assessment of the best interests of the child.4 As stated by the most eminent interdisciplinary team to recently address the subject, our legal system . . does not have the capacity to supervise the fragile, complex interpersonal bonds between child and parent” because it does not possess the “resources [or] the sensitivity to respond to a growing child’s everchanging needs and demands.” (Goldstein, Freud & Solnit, Before The Best Interests of the Child (1979) at pp. 11-12.)5
The rationale that supports judicial respect for family privacy does not lose its force upon the dissolution of marriage where, as here, a family relationship—even a disharmonious one—continues between the former spouses in connection with the rearing of their minor children.6 The concept of family *268privacy embodies not simply a policy of minimum state intervention but also a presumption of parental autonomy.7 Many of the purposes served by this presumption become more important after dissolution than they were before. One such purpose, for example, is to diminish the uncertainties and discontinuities that can afflict the parent-child relationship whenever third parties (lawyers as well as judges) episodically intrude through an ill-equipped adversarial process in which decisions are subject to reconsideration and eventual appellate review. Such uncertainties and discontinuities are of course more likely and serious dangers after separation or dissolution than before. Another purpose of the presumption that need only be served after separation or dissolution is to advance the public policy of this state to assure minor children of frequent and continuing contact with divorced or separated parents and “to encourage [such] parents to share the rights and responsibilities of child rearing in order to effect this policy.” (Civ. Code, § 4600, subd. (a).) Coercive judicial intervention perceived to be on the side of one parent against the other will in many instances have the opposite effect; that is, it will discourage the judicially disfavored parent who has lost control from continuing to maintain a meaningful role in the upbringing of his or her child. For these reasons, among others, considerations of family privacy and parental autonomy should continue to constrain the exercise of judicial authority despite the fact that the family is no longer intact; indeed, such considerations more often than not gain force because the family is no longer intact.
In short, for purposes of resolving family disputes such as that here presented, separation of the spouses or dissolution of the marriage has more formal than substantive significance. Proposed interventions in the privacy of the family that would not conceivably be entertained by the courts during marriage (such as the order here appealed from) are not suddenly tenable simply because the parents have become separated or divorced. A judicial approach to the legally nonintact family as nonetheless residually cohesive is warranted in custody and visitation disputes not simply because such disputes often invite the *269perilous interposition of personal values8 and are otherwise generally unsuited for judicial disposition but as well for the affirmative reason that such an approach is consistent with legislative intent to encourage the maintenance of postmarital family ties. (In addition to Civ. Code, § 4600, see Civ. Code, §§ 4600.5, subd. (a), which creates a presumption of joint custody in certain instances, and 4601, which creates a presumption that reasonable visitation rights shall be awarded.)
Recognizing that the adversary process can be destructive to the parent-child relationship and that a court order is not an effective way to influence parental attitudes and control behavior, the Legislature has enacted a framework for the mandatory mediation of contested custody and visitation disputes. (Civ. Code, § 4607. See also, Code Civ. Proc., §§ 1760, 1768.) The system of “private ordering”9 contemplated by this statutory framework, which is much more likely than judicial fiat to acceptably reconcile conflicting desires of the parents and thereby serve the best interests of the child, deserves strong judicial deference and should be pursued more fully than appears to have been accomplished in this case prior to issuance of the order appealed from.
Mediation does not resolve every dispute, and there are of course numerous situations in which despite its disabilities forceful judicial intervention is necessary and appropriate. Consistent with the rule in Murga, we emphasize only that in cases such as the one before us, in which the best interest of the child must be adjudicated upon the basis of debatable value judgments, the decision to intervene must not only be conditioned upon a clear affirmative showing of harm or likely harm to the child10 and the exhaustion of mediation but must *270concomitantly reflect the considered judgment that such harm presents a graver problem than those attendant upon coercive intervention in family privacy.11
None of these criteria have been met in this case. In our view enforcement of the decree of the trial court—accepting the very questionable assumption that as a practical matter it is enforceable—would more likely exacerbate than diminish the conflict complained of.
While there are, as we readily acknowledge, cases in which the dangers of judicial intervention are outweighed by the danger of doing otherwise, this case is not among them.
The order appealed from is reversed.
Girard, J.,* concurred.
As noted by our dissenting colleague, the reason for the separation (and as well for the mother’s departure from the Mormon church) related to the mother’s belief that the father had molested the daughter and, more particularly, her claim that when she brought this to the attention of a Mormon bishop he responded that “this was not a moral issue.” We believe these matters are here irrelevant because the claimed events which never have been factually established, allegedly occurred more than two years prior to the hearing below, because at the time of separation and continuously thereafter the mother never objected to the father having visitation rights, because the mother did not assert in the hearing below that father’s visitation rights should be restricted for any reason other than the religious conflict, and because the record is bereft of any evidence whatsoever that either the mother or the trial court believed there was any present or fiiture possibility of molestation by the father.
Miller v. Hedrick (1958) 158 Cal.App.2d 281 [322 P.2d 231], addresses the question whether to restrict visitation rights due to religious differences between the parents, but rather than defining any criteria for making such a determination simply defers to the discretion of the trial court.
See, e.g., Foster & Freed A Bill of Rights for Children (1972) 6 Fam.L.Q. 343, 347 in which it is proposed that “[a] child has a moral right and should have a legal right. . . [t]o receive parental love and affection, discipline and guidance, and to grow to maturity in a home environment which enables him to develop into a mature and responsible adult . . . .”
See Wald, State Intervention on Behalf of “Neglected” Children (1975) 27 Stan.L.Rev. 985, 991-993 and Mnookin, Child Custody Adjudication: Judicial Functions in the Face of Indeterminacy (1975) 39 Law & Contemp. Probs. 226, 266.
An earlier work originally published in 1973 (Beyond the Best Interests of the Child (2d ed. 1979)) by the same authors, frequently cited by the courts (see Crouch, An Essay on the Critical and Judicial Reception of Beyond the Best Interests of the Child (1979) 13 Fam.L.Q. 49), provides authority for the idea that in case of differences between divorced parents regarding the upbringing of their children, the custodial parent should have virtually full command, including restrictions upon visitation by the noncustodial parent such as those imposed by the lower court in this case. (See Goldstein, Freud & Solnit, Beyond the Best Interests of the Child, supra, at pp. 116-133.) This particular position, which is the most controversial of those advanced by the authors, was considered and rejected in Felton v. Felton, supra, 418 N.E.2d at page 607, footnote 2, and by numerous other authorities in the field, e.g., Katkin, Bullington & Levine, Above and Beyond the Best Interests of the Child: An Inquiry Into the Relationship Between Social Science and Social Action (1974) 8 Law & Soc. Rev. 669, 680; Strauss & Strauss, Book Review (1974) 74 Colum.L.Rev. 996, 1002-1005 and Dembitz, Book Review (1974) 83 Yale LJ. 1304, 1310. For analyses of the more recent treatise by Goldstein et al., which also addresses the sort of visitation issue presented in this case, see Wald, Thinking About Public Policy Toward Abuse and Neglect of Children: A Review of Before the Best Interests of the Child (1980) 78 Mich.L.Rev. 645, and Appleton, Growing Up With Goldstein, Freud and Solnit (1980) 58 Tex.L.Rev. 1343.
Indeed, judicial respect for the integrity of the family unit has been accorded in cases in which there never was a marriage and the children were illegitimate. (See, e.g., Stanley v. Illinois (1971) 405 U.S. 645 [31 L.Ed.2d 551, 92 S.Ct. 1208].)
As we noted in In re Marriage of Wellman (1980) 104 Cal.App.3d 992 [164 Cal.Rptr. 148], “the state has no general authority to dictate to parents the manner in which they should rear their children. ‘[Constitutional interpretation has consistently recognized that the parents’ claim to authority in their own household to direct the rearing of their children is basic in the structure of our society. “It is cardinal with us that the custody, care and nurture of the child reside first in the parents, whose primary function and freedom include preparation for obligations the state can neither supply nor hinder.” ’ (Ginsberg v. New York (1968) 390 U.S. 629, 639 [20 L.Ed.2d 195, 203-204, 88 S.Ct. 438].” (Id., at p. 996.) See also Stanley v. Illinois, supra, where the Supreme Court made it plain “that the interest of a parent in the companionship, care, custody, and management of his or her children ‘come[s] to this Court with a momentum for respect lacking when appeal is made to liberties which derive merely from shifting economic arrangements.’ ” (405 U.S. at p. 651 [31 L.Ed.2d at p. 558], quoting from Kovacs v. Cooper (1949) 336 U.S. 77, 95 [93 L.Ed. 513, 526, 69 S.Ct. 448, 10 A.L.R.2d 608] (cone. opn. of Frankfurter, J.). The California Supreme Court has been equally deferential to the fundamental rights of parents. (See, e.g., In re Carmaleta B., supra, at p. 489, and In re B.G. (1974) 11 Cal.3d 679, 698-699 [114 Cal.Rptr. 444, 523 P.2d 244].)
As has been noted, “the determination of what is ‘best’ or ‘least detrimental’ for a particular child is usually indeterminate and speculative. For most custody cases, existing psychological theories simply do not yield confident predictions of the effects of alternative custody dispositions. Moreover, even if accurate predictions were possible in more cases, our society today lacks any clear-cut consensus about the values to be used in determining what is ‘best’ or ‘least detrimental.’” (Mnookin, supra, 39 Law & Contemp. Probs., at p. 229. See also, Oster, Custody Proceeding: A Study of Vague and Indefinite Standards (1965) 5 J.Fam.L. 21 and Foster & Freed, Child Custody (1964) 39 N.Y.U.L.Rev. 423.)
On the legitimacy of private ordering and its special advantages in family disputes see Mnookin and Kornhauser, Bargaining in the Shadow of the Law: The Case of Divorce (1979) 88 Yale LJ. 950. See also, King, Child Custody—A Legal Problem? (May/June 1979) 54 State Bar 1. 156; Bahr, An Evaluation of Court Mediation (1981) 2 Jour. of Fam. Issues 39; Pearson, Thoennes & Vander Kool, Mediation of Contested Child Custody Disputes (1982) 11 The Colorado Law. 336 and Note, Non-Judicial Resolution of Custody and Visitation Disputes (1979) 12 U.C. Davis L.Rev. 582.
It is unnecessary and we do not here undertake to broadly define the specific types of harm required to meet the Murga test. We note, however, that one thoughtful commentator has in effect done so, suggesting that in cases such as this “coercive intervention should be permissible only when a child has suffered or is likely to suffer serious physical injury as a result of abuse or inadequate care; when a child is suffering from severe emotional damage and his parents are unwilling to deal with his problems without coercive intervention; when a child is sexually abused; when a child is suffering from a serious medical condition and his parents are unwilling *270to provide him suitable medical treatment; or when a child is committing delinquent acts at the urging or with the help of his parents.” (Wald, supra, 27 Stan.L.Rev. at p. 1008.)
The judicial sensibility that intervention may create a more serious problem than the one it purports to remedy is often exhibited in the pertinent case law. Kilgrow v. Kilgrow, supra, 268 Ala. 763 [107 So.2d 885], and Lynch v. Uhlenhopp, supra, 248 Iowa 68 [78 N.W.2d 491], are factually apposite exemplars. In Kilgrow, which also involved a religious dispute between parents, the trial court entered a mandatory decree regarding the religious training of the child. Reversing this decree, the Alabama Supreme Court declared it “anomalous to hold that a court of equity may sit in constant supervision over a household and see that either parent’s will and determination on the upbringing of a child is obeyed, even though the parents’ dispute might involve what is best for the child.” In support of this conclusion, the court observed that “intervention, rather than preventing or healing a disruption, would quite likely serve as the spark to a smoldering fire.” (268 Ala. at p. 479 [107 So.2d at p. 889].)
Similarly, in Lynch, the Iowa Supreme Court noted that in light of the “constant bickering” of his parents concerning compliance or noncompliance with a divorce decree regarding the child’s religious training, “the real sufferer would be the child.” It concluded that enforcement of the decree “would do much harm” and that, therefore, “[cjourts should be slow to place provisions controlling religious beliefs in decrees, even granting certainty and constitutionality and the consent of the parties.” (248 Iowa 81-82 [78 N.W.2d at p. 499].)
Assigned by the Chairperson of the Judicial Council.