Young v. Young, [1993] 4 S.C.R. 3

Irene Helen Young Appellant

James Kam Chen Young Respondent

and

W. Glen How Respondent

and

Watch Tower Bible and Tract Society of Canada Respondent

and

The Attorney General of Canada,

the Attorney General for Ontario,

the Attorney General of Quebec,

the Attorney General of Manitoba,

the Attorney General of British Columbia,

the Law Society of British Columbia and

the Seventh‑day Adventist Church in Canada Interveners

Indexed as: Young v. Young

File No.: 22227.

1993: January 25, 26; 1993: October 21.

Present: La Forest, L'Heureux‑Dubé, Sopinka, Gonthier, Cory, McLachlin and Iacobucci JJ.

on appeal from the court of appeal for british columbia

Family law ‑‑ Custody ‑‑ Access ‑‑ Best interests of the child -- Access parent insisting on instructing children on religion -- Custodial parent and children objecting to religious instruction -- Court ordering that access parent discontinue religious activities with children -- Scope of "best interests of the child" -- Whether or not "best interests of the child" equivalent of absence of harm -- Whether or not restriction on access in best interests of the children.

Family law ‑‑ Children ‑‑ Best interests of the child ‑‑ Access parent insisting on instructing children on religion -- Custodial parent and children objecting to religious instruction -- Court ordering that access parent discontinue religious activities with children -- Scope of "best interests of the child" -- Whether or not "best interests of the child" equivalent of absence of harm -- Whether or not restriction on access in best interests of the children.

Family law -- Property and financial awards -- Lump sum payment -- Family debts -- Principles governing reallocation of property.

Constitutional law ‑‑ Charter of Rights ‑‑ Freedom of religion -- Freedom of expression -- Divorce Act requiring that orders concerning children only take into account "the best interests of the child" -- Access parent insisting on instructing children on religion -- Custodial parent and children objecting to religious instruction -- Court ordering that access parent discontinue religious activities with children -- Whether or not access restriction infringing freedom of religion -- Whether or not access restriction infringing freedom of expression -- Divorce Act, R.S.C., 1985, c. 3 (2nd Supp.), ss. 16(8), 17(5) -- Canadian Charter of Rights and Freedoms, s. 2(a), (b).

Courts -- Costs -- Principles governing awards of costs on solicitor‑client basis.

Torts -- Maintenance -- Religious society carrying cost of action -- Common religious action -- Whether or not tort of maintenance.

Appellant's and respondent's separation was marked by a protracted series of court battles. Appellant was awarded custody of the couple's three daughters and respondent was granted access subject to court imposed restrictions arising from appellant's objection to his religious activity with the children. Respondent was ordered not to discuss the Jehovah's Witness religion with the children, take them to any religious services, canvassing or meetings, or expose them to religious discussions with third parties without appellant's prior consent. Organized religion was not important to appellant although she wanted the children to be raised within the United Church.

The two older daughters liked their father but came to dislike his religious instruction to the extent that it was damaging his relationship with them and was contributing to the stress the children were experiencing in adjusting to their parents' separation.

The trial judge also made orders for the distribution of property and for costs. The respondent's interest in the matrimonial home was ordered transferred to the appellant because any remaining interest in the house, after respondent paid what was already owing to appellant, was to be transferred in the form of lump sum maintenance. Respondent was found responsible for debts incurred by the appellant for the support of herself and the children pending maintenance and for a debt made to a family corporation. Costs were awarded on a solicitor‑client basis against respondent, his lawyer and a religious society not a party to the proceedings.

Respondent appealed. The Court of Appeal set aside the limitations on religious discussion and attendance, on the ground that it was in the best interests of the children that they come to know their non‑custodial parent fully, including his religious beliefs, unless the evidence established the existence of or the potential for real harm or the child did not consent to being subject to the access parent's views or practices. The Court of Appeal also altered the division of property and the awards of costs made by the trial judge. Appellant appealed these rulings to this Court.

Four constitutional questions queried (1) whether ss. 16(8) and 17(5) of the Divorce Act (requiring that judicial decisions regarding custody and access be made "in the best interests of the child") denied the Charter guarantees of freedom of religion, of expression and of association (s. 2(a), (b), and (d)), and if so, (2) were they justified under s. 1; (3) whether ss. 16(8) and 17(5) violated the equality guarantee of the Canadian Charter of Rights and Freedoms (s. 15(1)), and (4) if so, were they justified under s. 1. The Court considered the requirements of the "best interests of the child" and whether this standard infringed the guarantees of freedom of religion and expression under the Charter. A main consideration was unrestricted access by a non‑custodial parent and the conditions necessary to curtail that access.

Held (L'Heureux-Dubé J. dissenting in the result): The appeal should be allowed in part.

The issues should be decided as follows:

The test regarding access is the best interests of the child (L'Heureux-Dubé J., La Forest and Gonthier JJ., and Iacobucci and Cory JJ.). McLachlin J. suggests that in cases such as this harm is usually an important element in determining the best interests of the child. Sopinka J. would recognize a threshold element of harm.
Sections 16(8) and 17(5) of the Divorce Act do not violate ss. 2(a), (b), (d) or 15(1) of the Charter. L'Heureux-Dubé J. (and La Forest and Gonthier JJ.) found the Charter to be inapplicable. McLachlin J. found the impugned legislation did not violate the Charter. Cory and Iacobucci JJ. agreed that there was no Charter violation. Sopinka J. found that the Charter applied and could only be overridden in limited circumstances.
The restrictions on access should be removed (L'Heureux-Dubé J. and La Forest and Gonthier JJ. dissenting).
The judgment dealing with property and financial matters and the award of costs should be varied (L'Heureux-Dubé J. dissenting).

Best Interest of the Child, Charter Considerations and Access

Per L'Heureux‑Dubé J.: The power of the custodial parent is not a "right" with independent value granted by courts for the benefit of the parent. Rather, the child has a right to a parent who will look after his or her best interests and the custodial parent a duty to ensure, protect and promote the child's best interests. That duty includes the sole and primary responsibility to oversee all aspects of day-to-day life and long-term well‑being, as well as major decisions with respect to education, religion, health and well‑being. The non‑custodial parent retains certain residual rights over the child as one of his or her two natural guardians.

Child placement decisions should safeguard the child's need for continuity of relationships, reflect the child's (not the adult's) sense of time, and take into account the law's inability to supervise interpersonal relationships and the limits of knowledge to make long‑range predictions. This need for continuity generally requires that the custodial parent have the autonomy to raise the child as he or she sees fit without interference with that authority by the state or the non‑custodial parent. A custody award is a matter of whose decisions to prefer, as opposed to which decisions to prefer. Courts cannot make the necessary day‑to‑day decisions which affect the best interests of the child. Once a court has determined who is the appropriate custodial parent, it must presume that that parent will act in the best interests of the child.

Decisions are made according to the best interests of the child without the benefit of a presumption in favour of either parent. The Act envisages contact between the child and each of his or her parents as a worthy goal which should be in the best interests of the child. Maximum contact, however, is not an unbridled objective and must be curtailed wherever the welfare of the child requires it.

The right to access is limited in scope and is conditioned and governed by the best interests of the child. The legislation makes it quite explicit that only the best interests of the child as it is comprehensively understood should be considered in custody and access orders. The role of the access parent is that of a very interested observer, giving love and support to the child in the background. He or she has the right to know but not the right to be consulted. Access rights recognize that the best interests of the child normally require that the relationship developed with both parents prior to the divorce or separation be continued and fostered. The right to access and the circumstances in which it takes place must be perceived from the vantage point of the child. Wherever the relationship to the non‑custodial parent conflicts with the best interests of the child, the furtherance and protection of the child's best interests must take priority over the desires and interests of the parent.

As the ultimate goal of access is the continuation of a relationship which is of significance and support to the child, access must be crafted to preserve and promote that which is healthy and helpful in that relationship so that it may survive to achieve its purpose. Sources of ongoing conflict which threaten to damage or prevent the continuation of a meaningful relationship should be removed or mitigated. Notwithstanding a general concern about the vulnerability of access rights to the caprices of a vengeful custodial parent, courts should not be too quick to presume that the access concerns of the custodial parent are unrelated to the best interests of the child. Courts should also not be blind to issues, such as financial support, which form part of the broader context in which these rights are exercised. The access parent has no obligation to exercise those rights and cannot be forced to comply with such an order even if that contact has been determined to be in the child's best interest.

The best interests of the child cannot be equated with the mere absence of harm: it encompasses a myriad of considerations. Courts must attempt to balance such considerations as the age, physical and emotional constitution and psychology of both the child and his or her parents and the particular milieu in which the child will live. One of the most significant factors in many cases will be the relationship that the child entertains with his or her parents. Since custody and access decisions are pre‑eminently exercises in discretion, the wide latitude under the best interests test permits courts to respond to the spectrum of factors which can both positively and negatively affect a child. What may constitute stressful or damaging circumstances for one child may not necessarily have the same effect on another.

The most common presumption now governing the best interests test is the primary caregiver presumption. It explicitly restores the values of commitment and demonstrated ability to nurture the child and recognizes the obligations and supports the authority of the parent engaged in day to day tasks of childrearing.

The order of the trial judge is not subject to the Charter. Even if it were, the best interests test is nevertheless value neutral and does not, on its face, violate any Charter right. Its objective, the protection of a vulnerable segment of society, is completely consonant with the Charter's values. Broad judicial discretion is crucial to the proper implementation of the legislative objective of securing the best interests of the child. Such discretion in a legislative provision does not of itself give rise to an inference of Charter infringement. It cannot be considered in the absence of an examination of the legislative objectives and must be rationally tied to those objectives.

The standard for finding a legislative provision unconstitutional because of vagueness is high. The provisions need only permit the framing of an intelligible legal debate with respect to the objectives contained in the legislation. The best interests test is not so uncertain as to be incapable of guiding a consideration of the factors relevant to custody and access determinations. The fact that it must be applied to the facts of each case does not militate in favour of its unconstitutionality.

The vagueness of a legislative provision cannot be examined in the abstract but must be considered within the context of the particular legislative objectives in question, bearing in mind that some objectives will require a panoply of judicial remedies for their meaningful fulfilment. Among the factors with which courts should be concerned when the vagueness of a law is at issue are: (a) the need for flexibility and the interpretive role of the courts, (b) the impossibility of achieving absolute certainty, a standard of intelligibility being more appropriate and (c) the possibility that many varying judicial interpretations of a given disposition may exist and perhaps coexist.

The custodial parent need not show harm in order to restrict access to the children by the non‑custodial parent. There is no rationale for defining the best interests of the child with the absence of harm. Nothing in the Act mandates or even suggests that "real danger of significant harm to the child" be the sole consideration in matters of custody and access. Indeed, the harm test would require courts to ignore the very factors which are set out in the Act and invert the basic focus of the inquiry into custody and access. The welfare of children is put at considerable risk if the prospect of harm becomes the sole prerequisite for restrictions on access. The best interests of the child is not simply the right to be free of demonstrable harm; it is the positive right to the best possible arrangements in the circumstances of the parties. The harm test cannot meet the legal system's primary goal in divorce situations ‑‑ minimizing the adverse effects of children. This goal requires a vision of the best interests of the child that is more than neutral to the conditions under which custody and access occur. Judges must exercise their discretion to prevent harm to the child rather than merely identify or establish its presence after the damage is done.

Expert evidence should not be routinely required to establish the best interests of the child. Expert testimony, while helpful in some circumstances, is often inconclusive and contradictory because such assessments are both speculative and may be affected by the professional values and biases of the assessors themselves. Experts are not always better placed than parents to assess the needs of the child. The person involved in day to day care may observe changes in the child that could go unnoticed by anyone else and normally has the best vantage point from which to assess the interests of the child. The custodial parent, therefore, will often provide the most reliable and complete source of information to the judge on the needs and interests of that child. The importance of the evidence of children in custody and access disputes, too, must be emphasized.

Restrictions on access do not necessarily prevent children from coming to know their parents in meaningful ways. Interpreting the goal of maximum contact as requiring unrestricted access may defeat the Act's objective if the pre‑eminence of unlimited "knowledge" results in the ultimate destruction of the relationship. In this case, the purpose of the restrictions was to ensure that the children will continue to know their father "at all".

Freedom of religion and freedom of expression are public in nature and encompass the freedom of the individual from state compulsion or restraints. The state's role in custody and access decisions does not transform the essentially private character of parent‑child interchanges into activity subject to Charter scrutiny. Legitimate questions may arise about the role of the state, and hence the application of the Charter in regulating other aspects of family law. A valid purpose can hardly be served, however, by importing the discourse of freedom of expression and religion into orders made in the resolution of custody and access disputes. Once the best interests test itself has been found to accord with Charter values, the trial judge's order itself is not subject to further constitutional review, as the necessary state infringement of religious rights required to sustain a Charter challenge is not present. The principles enunciated in Dolphin Delivery apply as custody and access matters are essentially private in nature and there exists no state action to be impugned.

Decisions regarding custody and access must not be based on the parents' faith. The religion of the parties, however, may be relevant as one of the circumstances to be assessed in the determination of the best interests of the child. Where there is conflict over religion, the court is not engaged in adjudicating a "war of religion" and the religious beliefs of the parties themselves are not on trial. Rather, it is the manner in which such beliefs are practised together with the impact and effect they have on the child which must be considered. In all cases where the effects of religious practices are at issue, the best interests of the child must prevail.

Ordinarily, the exposure of a child to different religions or beliefs may be of value to the child. Where religion becomes a source of conflict between the parents or is the very cause of the marriage breakdown, it is generally not in the best interests of the child and may in some circumstances be very detrimental for the child to be drawn into the controversy over religious matters. Where there is conflict over religion, courts must secure the longstanding authority of the custodial parent to make decisions over religious activities. This ensures that stress occasioned by such issues does not become a continuing and ultimately destructive feature in the life of the child after divorce.

Freedom of religion is not an absolute value. Here, powerful competing interests must also be recognized, not the least of which, in addition to the best interests of the children, are the freedoms of expression and religion of the children themselves.

Respondent's religious beliefs and practices and his general rights of access were not threatened. The restrictions were aimed at reducing the area of conflict which had arisen on account of the respondent's behaviour with his children during access and the effects of that behaviour on their best interests. Much of the stress the children were experiencing was related to their resistance to becoming involved in their father's religious practices. The restrictions were to further the best interests of these children by removing the source of conflict, particularly as the ultimate purpose of the restrictions was to preserve the relationship between the respondent and his children. Evidence supported the conclusion that the respondent would not respect the wishes of the children without an order to do so.

Per La Forest and Gonthier JJ.: Agreement was expressed for the reasons of L'Heureux‑Dubé J. holding that access be determined on the basis of what is in the best interests of the child and for her resolution of the constitutional issue.

Per Iacobucci and Cory JJ.: The best interests of the child standard does not violate in ss. 2(a), (b), (d) and 15 of the Canadian Charter of Rights and Freedoms substantially for the reasons given by L'Heureux-Dubé and McLachlin JJ. No opinion was expressed on the questions of whether a Charter infringement, if found, would be so trivial as not to warrant Charter protection and of whether or not the Charter applies to judicial orders made in custody or access proceedings.

For many of the reasons advanced by L'Heureux‑Dubé J., access to children should be determined on the basis of what is in the best interests of the child. Expert evidence, while sometimes helpful, is not always necessary to establish the best interests of the child; that question can be determined normally from the evidence of parties themselves and the testimony, where appropriate, of the children concerned.

The matters of the children's attending religious services with the respondent and accompanying him on his proselytizing activities were resolved by the respondent's undertaking to respect his children's wishes in this regard. The order forbidding the respondent from discussing his religion was not supported by a proper application of the best interests of the child test. Indeed, curtailment of explanatory or discursive conversations between a parent and his or her child should only be rarely ordered.

Per McLachlin J.: The Divorce Act mandates that, on matters of access, the ultimate test in all cases is the best interest of the child. This is a positive test, encompassing a wide variety of factors, including the desirability of maximizing contact between the child and each parent if compatible with the best interests of the child. The custodial parent has no "right" to limit access. The judge must consider all factors relevant to determining what is in the child's best interests. The risk of harm to the child, while not the ultimate legal test, may also be a factor to be considered. This is particularly so where the issue is the quality of access ‑‑ what the access parent may say or do with the child. In such cases, it will generally be relevant to consider whether the conduct in question poses a risk of harm to the child which outweighs the benefits of a free and open relationship which permits the child to know the access parent as he or she is. The judge must act not on his or her personal views but on the evidence.

The legislative provision for the "best interests of the child" does not limit and therefore does not violate the Charter right to religious and expressive freedom. Religious expression not in the best interests of the child is not protected by the Charter because the guarantee of freedom of religion is not absolute and does not extend to religious activity which harms or interferes with the parallel rights of other people. Conduct not in the best interests of the child, even absent the risk of harm, amounts to an "injury" or intrusion on the rights of others and is clearly not protected by this Charter guarantee. "Injure" in this context is a broad concept. To deprive a child of what a court has found to be in his or her best interests is to "injure", in the sense of not doing what is best for the child. A child's vulnerability heightens the need for protection and any error should be made in favour of the child's best interests and not in favour of the exercise of the alleged parental right. An additional factor which may come into play in the case of older children is the "parallel right" of others to hold and manifest beliefs and opinions of their own.

The ambit of freedom of expression is broader than that of freedom of conscience and religion because even harmful expression may be protected. Some forms of harmful expression, however, are not constitutionally protected: violence or threats of violence or a direct attack on the physical integrity and liberty of another. Criminal conduct is an indication, although not a conclusive one, that expressive conduct is constitutionally unprotected.

A prima facie case for protection under the guarantee of freedom of expression can be made out here. The expression challenged does not take the form of the non‑protected categories of expression. The harm done, if any, is of a psychological nature.

A purposive approach to Charter interpretation requires that associated rights ‑‑ religious freedom and freedom of expression ‑‑ be interpreted in a consistent and coherent manner. The ambit of a particular right or freedom, moreover, cannot be defined in the abstract but rather should be defined in the context of the particular activity in question.

The teaching of religious beliefs and practices to one's children, while it has an expressive aspect, is predominantly religious. In seeking to reconcile the rights of freedom of religion and freedom of expression in this context, it is the religious aspect which must dominate. Reading the two guarantees together, the limits of the guarantee of freedom of expression should govern in the context of religious instruction of children.

The custodial parent does not have the "right" to determine limits on access. The only question to be considered, where limitation of access is in issue, is what is in the best interests of the child. The custodial parent's obligation to make certain basic decisions as to how the child is educated (which may extend to religious matters) does not automatically mean that religious contacts with the access parent of a different faith are to be excluded. The failure of the child to consent to instruction on the part of the access parent does not necessarily preclude such instruction's being in the child's best interests.

The benefits which might enure to the children from coming to know their father as he was ‑‑ as a devoutly religious man devoted to the Jehovah's Witness faith ‑‑ were not considered at trial and no reference was made to Parliament's instruction that a child have as much contact with both parents as is compatible with his or her best interests. The question of whether there was any evidence of a risk of harm to the children which might offset the benefit of full access to their father's values, including those related to religion, was not adequately considered. While access may be limited in some circumstances on grounds unrelated to harm, where the issue is whether entirely lawful discussions and activities between the access parent and the child should be curtailed, the judge should enquire into whether the conduct poses a risk of harming the child. The evidence did not establish that harm was being caused to the children.

The order restricting respondent's access was unnecessary given his undertaking and the order enjoining him from preventing blood transfusions was unnecessary from a practical point of view. Parents should not make disparaging comments about the other parent's religion, but the matter might best be left to the parents' good sense.

Per Sopinka J.: While the "best interests of the child" test is the ultimate determination in deciding issues of custody and access, it must be reconciled with the Charter. General language in a statute which, in its breadth, potentially confers the power to override Charter values must be interpreted to respect those values. Here, the best interests test must be interpreted to allow the Charter right to freedom of religious expression to be overridden only if its exercise would occasion consequences that involve more than inconvenience, upset or disruption to the child and incidentally to the custodial parent.

The long‑term value to a child of a meaningful relationship with both parents is a policy that is affirmed in the Divorce Act. Each parent, therefore, can engage in those activities which contribute to identify the parent for what he or she really is. The access parent is not expected to act out a part or assume a phony lifestyle during access periods. The policy favouring activities that promote a meaningful relationship is not displaced unless there is a substantial risk of harm to the child.

The best interests of a child are more aptly served by a law which recognizes the right of that child to a meaningful post‑divorce relationship with both parents. The "rights" must be distributed between the custodial and the access parent so as to encourage such a relationship. The traditional notion of guardianship giving the custodial parent the absolute right to exercise full control over the child, even when the other parent is exercising his or her right of access, is at odds with this concept.

"Harm", in this context, connotes an adverse effect on the child's upbringing that is more than transitory. The impugned exercise by the access parent must be shown to create a substantial risk that the child's physical, psychological or moral well‑being will be adversely affected. Exposure to new experiences and ideas may upset children and cause them considerable discomfort but these experiences are not necessarily in the long‑term best interests of the child. Similarly, conflict between parents on many matters including religion is not uncommon, but in itself cannot be assumed to be harmful unless it produces a prolonged acrimonious atmosphere.

Risk of substantial harm must be shown if religious expression is to be restricted in applying the best interests of the child test. The statutory test in s. 16(10) of the Divorce Act does not constitute a limitation on freedom of religious expression. This freedom does not extend to protect conduct which is harmful to others. However, the concept of harm should not be expanded to reach the conclusion that anything which is not in the best interests of the children is injurious within the meaning of s. 2(b) and thus not protected by the Charter.

Financial Considerations and Costs

Per McLachlin J.: A judge, to fix lump sum maintenance, must fix it in a sum certain with reference to the principles applicable to such an award. The goal of conveying the entire interest in the matrimonial home to the wife did not support an award of lump sum maintenance; more was required. There was ample basis here for an order under s. 51 of the Family Relations Act awarding appellant a greater portion of the family assets because respondent, for a considerable time, paid little or nothing for the support of the family. The debts incurred during this period can serve as a consideration supporting reduction of respondent's interest in the family property. The equity in the home was not substantial and respondent was permitted to retain other assets.

The trial judge, in effecting a de facto reapportionment of the interest in the family assets to do justice, as was permitted on the facts and the law, did not expressly allude to the factors for reallocation. This omission should not result in a new trial being ordered because of the length and cost of the current litigation. The result achieved by the trial judge should be endorsed because the evidence was capable of supporting an order for reallocation of the parties' interest in the family assets to the extent required to give the appellant the entire interest in the matrimonial home.

The money owed by the family's jewelry corporation was not, in law, a debt for which respondent was personally liable. Only the corporation was liable. The debt appellant incurred to support herself and the children before she applied for maintenance is similarly unenforceable against respondent as a debt, although it could be taken into consideration in an order for reduction of his interest in the family assets.

Solicitor‑client costs are generally awarded only where there has been reprehensible, scandalous or outrageous conduct on the part of one of the parties. The facts that an application has little merit and that part of the cost of the litigation may have been paid for by others do not justify awarding solicitor‑client costs.

No order for costs should have been made against respondent's barrister. Costs are awarded as compensation for the successful party, not to punish a lawyer. Any member of the legal profession might be subject to a compensatory order for costs if it is shown that repetitive and irrelevant material, and excessive motions and applications, characterized the proceedings in which he or she was involved, and that the lawyer acted in bad faith in encouraging this abuse and delay. The courts have jurisdiction to make such an award, often under statute and as part of their inherent jurisdiction to control abuse of process and contempt of court. The proceedings here, despite their length and acrimonious progress, did not fall within these characterizations. Courts, moreover, must be extremely cautious in awarding costs personally against a lawyer, given the duties upon a lawyer to guard confidentiality of instructions and to bring forward with courage even unpopular causes. A lawyer should not be placed in a situation where his or her fear of an adverse order of costs may conflict with these fundamental duties of his or her profession.

Since the Society did not appear as a party, the costs awarded against it must be taken to be the equivalent of an award for the tort of maintenance. A person must intervene "officiously or improperly" to be liable for the tort of maintenance. Provision of financial assistance to a litigant by a non‑party will not always constitute maintenance. Funding by a relative or out of charity must be distinguished from cases where a person wilfully and improperly stirs up litigation and strife. The society's support was "out of charity and religious sympathy" and so did not constitute maintenance. It did not put forward respondent in an attempt to escape liability for costs. Its interest in the constitutional issue was insufficient to distinguish it from interveners who appear on constitutional cases and who have never been liable for costs.

Per La Forest and Gonthier JJ.: The reasons of McLachlin J. were agreed with on the property and monetary issues and on the principles governing costs.

Per Iacobucci and Cory JJ.: The reasons of McLachlin J. were agreed with on the property, monetary and costs issues.

Per L'Heureux-Dubé J. (dissenting): The reasons of the trial judge on the issues of maintenance, division of property and costs were agreed with.

Cases Cited

By L'Heureux‑Dubé J.

Distinguished: Hockey v. Hockey (1989), 1989 4299 (ON SC), 21 R.F.L. (3d) 105; referred to: Anson v. Anson (1987), 10 B.C.L.R. (2d) 357; Andrews v. Andrews, B.C.S.C., June 9, 1983, unreported; Brown v. Brown (1983), 1983 2311 (SK CA), 39 R.F.L. (2d) 396; DeLaurier v. Jackson, 1934 2 (SCC), [1934] S.C.R. 149; R. v. Big M Drug Mart Ltd., 1985 69 (SCC), [1985] 1 S.C.R. 295; In re Agar-Ellis, [1883] Ch. D. 317; R. v. De Manneville (1804), 5 East. 221, 102 E.R. 1054; Re Orr, 1933 110 (ON CA), [1933] O.R. 212; Talsky v. Talsky, 1975 29 (SCC), [1976] 2 S.C.R 292; Re Moores and Feldstein, 1973 535 (ON CA), [1973] 3 O.R. 921; Kruger v. Kruger (1979), 1979 1663 (ON CA), 25 O.R. (2d) 673; Baker v. Baker (1979), 1979 1962 (ON CA), 8 R.F.L. (2d) 236; C.(G). v. V.‑F.(T.), 1987 20 (SCC), [1987] 2 S.C.R. 244; Dussault v. Ladouceur (1987), 1987 876 (QC CA), 14 R.F.L. (3d) 185; Gunn v. Gunn (1975), 1975 1938 (PE PC), 24 R.F.L. 182; Benoit v. Benoit (1972), 1972 1869 (ON CJ), 6 R.F.L. 180 (Ont. Prov. Ct.), rev'd (1972), 1972 1936 (ON CA), 10 R.F.L. 282 (Ont. C.A.); Charlton v. Charlton (1980), 1980 545 (BC SC), 15 R.F.L. (2d) 220; Hewer v. Bryant, [1970] 1 Q.B. 357; Clarke v. Clarke (1987), 1987 2745 (BC SC), 7 R.F.L. (3d) 176; McCahill v. Robertson (1974), 1974 2225 (ON SC), 17 R.F.L. 23; Fougere v. Fougere (1987), 1987 7842 (NB CA), 77 N.B.R. (2d) 381; Dipper v. Dipper, [1980] 2 All E.R. 722; Keyes v. Gordon (1985), 1985 123 (NS CA), 45 R.F.L. (2d) 177; Droit de la famille -‑ 316, [1986] R.D.F. 651; Moge v. Moge, 1992 25 (SCC), [1992] 3 S.C.R. 813; Pierce v. Pierce, 1977 2206 (BC SC), [1977] 5 W.W.R. 572; Gubody v. Gubody, 1955 405 (ON SC), [1955] O.W.N. 548; Sudeyko v. Sudeyko (1974), 1974 2223 (BC SC), 18 R.F.L. 273; Frame v. Smith, 1987 74 (SCC), [1987] 2 S.C.R. 99; Tocco v. Tocco (1977), 1977 2628 (ON CA), 4 R.F.L. (2d) 174; Racine v. Woods, 1983 27 (SCC), [1983] 2 S.C.R. 173; In re McGrath (Infants), [1893] 1 Ch. 143; King v. Low, 1985 59 (SCC), [1985] 1 S.C.R. 87; Beson v. Director of Child Welfare (Nfld.), 1982 32 (SCC), [1982] 2 S.C.R. 716; M. (B.P.) v. M. (B.L.D.E.), (1992), 1992 8642 (ON CA), 97 D.L.R. (4th) 437; R. v. Swain, 1991 104 (SCC), [1991] 1 S.C.R. 933; R. v. Beare, 1988 126 (SCC), [1988] 2 S.C.R. 387; Baron v. Canada, 1993 154 (SCC), [1993] 1 S.C.R. 416; R. v. Morales, 1992 53 (SCC), [1992] 3 S.C.R. 711; R. v. Morgentaler, 1988 90 (SCC), [1988] 1 S.C.R. 30; Irwin Toy Ltd. v. Quebec (Attorney General), 1989 87 (SCC), [1989] 1 S.C.R. 927; Reference re ss. 193 and 195.1(1)(c) of the Criminal Code (Man.), 1990 105 (SCC), [1990] 1 S.C.R. 1123; R. v. Keegstra, 1990 24 (SCC), [1990] 3 S.C.R. 697; Canada (Human Rights Commission) v. Taylor, 1990 26 (SCC), [1990] 3 S.C.R. 892; Committee for the Commonwealth of Canada v. Canada, [1991] 1 S.C.R. 139; Osborne v. Canada (Treasury Board), 1991 60 (SCC), [1991] 2 S.C.R. 69; R. v. Butler, 1992 124 (SCC), [1992] 1 S.C.R. 452; R. v. Nova Scotia Pharmaceutical Society, 1992 72 (SCC), [1992] 2 S.C.R. 606; R. v. LeBeau (1988), 1988 3271 (ON CA), 41 C.C.C. (3d) 163; R. v. Khan, 1990 77 (SCC), [1990] 2 S.C.R. 531; RWDSU v. Dolphin Delivery Ltd., 1986 5 (SCC), [1986] 2 S.C.R. 573; Tremblay v. Daigle, 1989 33 (SCC), [1989] 2 S.C.R. 530; B.C.G.E.U. v. British Columbia (Attorney General), 1988 3 (SCC), [1988] 2 S.C.R. 214; Slaight Communications Inc. v. Davidson, 1989 92 (SCC), [1989] 1 S.C.R. 1038; Hills v. Canada (Attorney General), 1988 67 (SCC), [1988] 1 S.C.R. 513; Canada (Attorney General) v. Mossop, 1993 164 (SCC), [1993] 1 S.C.R. 554; Re Bennett Infants, 1952 293 (ON CA), [1952] O.W.N. 621; Delvenne v. Nabbie (1977), 1977 2270 (MB CA), 4 R.F.L. (2d) 21; Irmert v. Irmert (1984), 1984 ABCA 286, 64 A.R. 342; Harvey v. Lapointe (1988), 1988 8787 (QC CS), 13 R.F.L. (3d) 134; McQuillan v. McQuillan (1975), 1975 1946 (ON SC), 21 R.F.L. 324; Struncova v. Guay (1984), 1984 4817 (QC CS), 39 R.F.L. (2d) 298; Sullivan v. Fox (1984), 1984 4805 (PE PC), 38 R.F.L. (2d) 293; Droit de la famille -‑ 955, 1991 3838 (QC CA), [1991] R.J.Q. 599; Droit de la famille -‑ 353, 1987 398 (QC CA), [1987] R.J.Q. 545; Zummo v. Zummo, 574 A.2d 1130 (1990); Edmonton Journal v. Alberta (Attorney General), 1989 20 (SCC), [1989] 2 S.C.R. 1326; Fraser v. Public Service Staff Relations Board, 1985 14 (SCC), [1985] 2 S.C.R. 455; Adams v. McLeod, 1978 5 (SCC), [1978] 2 S.C.R. 621; Novic v. Novic, 1983 55 (SCC), [1983] 1 S.C.R. 696.

By McLachlin J.

Considered: R. v. Big M Drug Mart Ltd., 1985 69 (SCC), [1985] 1 S.C.R. 295; referred to: R. v. De Manneville (1804), 5 East. 221, 102 E.R. 1054; In re Taylor (1876), 4 Ch. D. 157; In re Agar‑Ellis (1883), 24 Ch. D. 317; Talsky v. Talsky, 1975 29 (SCC), [1976] 2 S.C.R. 292; Kades v. Kades (1961), 35 A.L.J.R. 251; J. v. C., [1970] A.C. 668; Re K. (minors), [1977] 1 All E.R. 647; King v. Low, 1985 59 (SCC), [1985] 1 S.C.R. 87; Irwin Toy Ltd. v. Quebec (Attorney General), 1989 87 (SCC), [1989] 1 S.C.R. 927; R. v. Keegstra, 1990 24 (SCC), [1990] 3 S.C.R. 697; Reference re ss. 193 and 195.1(1)(c) of the Criminal Code (Man.), 1990 105 (SCC), [1990] 1 S.C.R. 1123; RWDSU v. Dolphin Delivery Ltd., 1986 5 (SCC), [1986] 2 S.C.R. 573; Edmonton Journal v. Alberta (Attorney General), 1989 20 (SCC), [1989] 2 S.C.R. 1326; R. v. Sturmer and Town of Beaverton (1912), 1912 588 (ON SCDC), 25 O.L.R. 566; Goodman v. The King, 1939 21 (SCC), [1939] S.C.R. 446; Newswander v. Giegerich (1907), 1907 33 (SCC), 39 S.C.R. 354.

By Sopinka J.

Referred to: Slaight Communications Inc. v. Davidson, 1989 92 (SCC), [1989] 1 S.C.R. 1038; R. v. Big M Drug Mart Ltd., 1985 69 (SCC), [1985] 1 S.C.R. 295.

Statutes and Regulations Cited

AAct to Amend the Law relating to the custody of Infants, S. Prov. Can. 1855, c. 126, s. 1.

Act to amend the Law relating to the Custody of Infants (Talfourd's Act) (U.K.),

2 & 3 Vict., c. 54.

Canadian Charter of Rights and Freedoms, ss. 1, 2(a), (b), (d), 7, 15(1), 24, 26, 32.

Civil Code of Lower Canada, art. 30.

Civil Code of Quebec, arts. 543, 569, 570, 647.

Convention on the Rights of the Child, Can. T.S. 1992 No. 3, Art. 3(1).

Divorce Act, R.S.C., 1985, c. 3 (2nd Supp.), ss. 15(2), 16(1), (2), (4), (5), (6), (8), (9), (10), 17(5).

Family Relations Act, R.S.B.C. 1979, c. 121, ss. 51, 52.

Guardianship of Infants Act, 1886 (U.K.), 49 & 50 Vict., c. 27.

Guardianship of Infants Act, 1925 (U.K.), 15 & 16 Geo. 5, c. 45.

Guardianship of Minors Act, 1971 (U.K.), 1971, c. 3, s. 1.

Lov 8 April nr 7 om barn og foreldre, s. 34 (Law about Children and Parents, of April 7, 1981, no. 7, s. 34).

Supreme Court of Judicature Act, 1873 (U.K.), 36 & 37 Vict., c. 66, s. 25(10).

Authors Cited

Abella, Rosalie Silberman. "Family Law in Ontario: Changing Assumptions" (1981), 13 Ottawa L. Rev. 1.

Bala, Nicholas. "Assessing the Assessor: Legal Issues" (1990), 6 C.F.L.Q. 179.

Benians, Robin. "Preserving Parental Contact: a Factor in Promoting Healthy Growth and Development in Children". In Jo Tunnard, ed., Fostering Parental Contact: Arguments in Favour of Preserving Contact Between Children in Care and Their Families. London: Family Rights Group, 1982.

Boyd, Susan B. "Child Custody Law and the Invisibility of Women's Work" (1989), 96 Queen's Q. 831.

Boyd, Susan B. "Potentialities and Perils of the Primary Caregiver Presumption" (1991), 7 C.F.L.Q. 1.

Boyd, Susan B. "Women, Men and Relationships with Children: Is Equality Possible?". In Karen Busby, Lisa Fainstein and Holly Penner, eds., Equality Issues in Family Law: Considerations for Test Case Litigation. Winnipeg: Legal Research Institute of the University of Manitoba, 1990.

Burrett, Jill F. Child Access and Modern Family Law. Melbourne: Law Book Co., 1988.

Canada. Department of Justice. Custody and Access: Public Discussion Paper. Ottawa: The Dept., 1993.

Canada. Department of Justice. Evaluation of the Divorce Act, 1985 -- Phase I: Collection of Baseline Data. Ottawa: The Dept., 1987.

Canada. Department of Justice. Evaluation of the Divorce Act -- Phase II: Monitoring and Evaluation. Ottawa: The Dept., 1990.

Canada. Law Reform Commission. Report. Family Law. Ottawa: The Commission, 1976.

Canada. Law Reform Commission. The Custody, Care and Upbringing of Children of Divorcing Spouses. By Richard Gosse. Ottawa: The Commission, 1973.

Caplan, Paula J., and Jeffery Wilson. "Assessing the Child Custody Assessors" (1990), 27 R.F.L. (3d) 121.

Chambers, David L. "Rethinking the Substantive Rules for Custody Disputes in Divorce" (1984), 83 Mich. L. Rev. 477.

Deech, Ruth. "The Unmarried Father and Human Rights" (1992), 4 J. Child L. 3.

Delorey, Anne Marie. "Joint Legal Custody: A Reversion to Patriarchal Power" (1989), 3 C.J.W.L. 33.

Dickens, Bernard M. "The Modern Function and Limits of Parental Rights" (1981), 97 L.Q. Rev. 462.

Drakich, Janice. "In Search of the Better Parent: The Social Construction of Ideologies of Fatherhood" (1989), 3 C.J.W.L. 69.

Drakich, Janice. "In Whose Best Interests? The Politics of Joint Custody". In Bonnie Fox, ed., Family Bonds and Gender Divisions: Readings in the Sociology of the Family. Toronto: Canadian Scholars' Press, 1988.

Ehrcke, Ann. "Limiting Judicial Discretion in Custody Proceedings on Divorce" (1987), 6 Can. J. Fam. L. 211.

Ellison, Edythe S. "Issues Concerning Parental Harmony and Children's Psychosocial Adjustment" (1983), 53 Amer. J. Orthopsychiat. 73.

Elster, Jon. "Solomonic Judgments: Against the Best Interest of the Child" (1987), 54 U. Chi. L. Rev. 1.

Emery, Robert E. "Interparental Conflict and the Children of Discord and Divorce" (1982), 92 Psych. Bull. 310.

Fineman, Martha Albertson. "Dominant Discourse, Professional Language and Legal Change in Child Custody Decisionmaking" (1988), 101 Harv. L. Rev. 727.

Fineman, Martha Albertson. The Illusion of Equality: The Rhetoric and Reality of Divorce Reform. Chicago: University of Chicago Press, 1991.

Finnbogason, Eve, and Monica Townson. The Benefits and Cost‑effectiveness of a Central Registry of Maintenance and Custody Orders. Ottawa: Status of Women, 1985.

Folberg, Jay. "Joint Custody". In Rosalie S. Abella and Claire L'Heureux‑Dubé, eds., Family Law: Dimensions of Justice. Toronto: Butterworths, 1983.

Furstenberg, Frank F., Jr., and Andrew J. Cherlin. Divided Families: What Happens to Children When Parents Part. Cambridge, Mass.: Harvard University Press, 1991.

Gélinas, Louis, et Bartha Maria Knoppers. "Le rôle des experts en droit québécois en matière de garde, d'accès et de protection" (1993), 53 R. du B. 3.

Goldstein, Joseph, Anna Freud and Albert J. Solnit. Beyond the Best Interests of the Child. New York: The Free Press, 1979.

Goldstein, Joseph. "In Whose Best Interest?". In Rosalie S. Abella and Claire L'Heureux‑Dubé, eds., Family Law: Dimensions of Justice. Toronto: Butterworths, 1983.

Grassby, Miriam. "Women in Their Forties: The Extent of Their Rights to Alimentary Support" (1991), 30 R.F.L. (3d) 369.

Hetherington, E. Mavis, Martha Cox and Roger Cox. "Effects of Divorce on Parents and Children". In Michael E. Lamb, ed., Nontraditional Families: Parenting and Child Development. Hillsdale, N.J.: Lawrence Erlbaum & Associates, 1982.

Holmes, Sheila M. "Imposed Joint Legal Custody: Children's Interests or Parental Rights?" (1987), 45 U.T. Fac. L. Rev. 300.

Johnston, Janet R., Marsha Kline and Jeanne M. Tschann, "Ongoing Postdivorce Conflict: Effects on Children of Joint Custody and Frequent Access" (1989), 59 Amer. J. Orthopsychiat. 576.

LaFave, LeAnn Larson. "Origins and Evolution of the `Best Interests of the Child' Standard" (1989), 34 S.D.L. Rev. 459.

L'Heureux‑Dubé, Claire. "La garde conjointe, concept acceptable ou non?" (1979), 39 R. du B. 835.

Maidment, Susan. Child Custody and Divorce. London: Croom Helm, 1984.

Mnookin, Robert H. "Child-Custody Adjudication: Judicial Functions in the Face of Indeterminacy" (1975), 39 Law & Contemp. Probs. 226.

Munro, Karen M. "The Inapplicability of Rights Analysis in Post‑Divorce Child Custody Decision Making" (1992), 30 Alta. L. Rev. 852.

Payne, Julien D. Payne on Divorce, 2nd ed. Toronto: Butterworths, 1988.

Payne, Julien D., and Brenda Edwards. "Co‑operative Parenting After Divorce: A Canadian Perspective" (1989), 11 Advocates' Q. 1.

Perry, Debra. Access to Children Following Parental Relationship Breakdown in Alberta. Calgary: Canadian Research Institute for Law and the Family, 1992.

Richardson, C. James. Court‑based Divorce Mediation in Four Canadian Cities: An Overview of Research Results. Ottawa: Department of Justice Canada, Family Law Research, 1988.

Roth, Allan. "The Tender Years Presumption in Child Custody Disputes" (1976‑77), 15 J. Fam. L. 423.

Rutter, Michael. Maternal Deprivation Reassessed. Harmondsworth, England: Penguin Books, 1981.

Rutter, Michael. "Protective Factors in Children's Responses to Stress and Disadvantage". In Martha Whalen Kent and Jon E. Rolf, eds., Primary Prevention of Psychopathology: Social Competence in Children, vol. III. Hanover, N.H.: University Press of New England, 1979.

Ryan, Judith P. "Joint Custody in Canada: Time for a Second Look" (1986), 49 R.F.L. (2d) 119.

Schneider, Carl E. "Religion and Child Custody" (1992), 25 U. Mich. J.L. Ref. 879.

Smart, Carol, and Selma Sevenhuijsen. Child Custody and the Politics of Gender. London: Routledge, 1989.

Smith, Lucy, and Peter Lødrup. "The Child in the Divorce Situation -- Factors Determining the Custody Question and the Use of Experts in Custody Cases in Norway". In Ian F. G. Baxter and Mary A. Eberts, eds., The Child and the Courts. Toronto: Carswell, 1978.

Syrtash, John Tibor. Religion and Culture in Canadian Family Law. Toronto: Butterworths, 1992.

Toope, Stephen J. "Riding the Fences: Courts, Charter Rights and Family Law" (1991), 9 Can. J. Fam. L. 55.

Wallerstein, Judith S., and Sandra Blakeslee. Second Chances: Men, Women and Children a Decade After Divorce. New York: Ticknor & Fields, 1989.

Wallerstein, Judith S., and Joan Berlin Kelly. Surviving the Breakup: How Children and Parents Cope with Divorce. New York: Basic Books, 1980.

Weisman, Norris. "On Access After Parental Separation" (1992), 36 R.F.L. (3d) 35.

Weiss, Martin, and Robert Abramoff. "The Enforceability of Religious Upbringing Agreements" (1991), 25 J. Mar. L. Rev. 655.

Weitzman, Lenore J. The Divorce Revolution. New York: The Free Press, 1985.

Zarowny, Steven M. "The Religious Upbringing of Children After Divorce" (1980), 56 Notre Dame Law. 160.

APPEAL from a judgment of the British Columbia Court of Appeal (1990), 1990 3813 (BC CA), 50 B.C.L.R. (2d) 1, 75 D.L.R. (4th) 46, allowing an appeal in part from a judgment of Proudfoot J. (1989), 1989 8903 (BC SC), 24 R.F.L. (3d) 193. Appeal allowed in part, L'Heureux-Dubé J. dissenting in the result. Sections 16(8) and 17(5) of the Divorce Act did not violate ss. 2(a), (b), (d) or 15(1) of the Canadian Charter of Rights and Freedoms.

Lorne N. MacLean and Fred C. Lowther, for the appellant.

W. Glen How, Q.C., and Sarah E. Mott‑Trille, for the respondent James K. C. Young.

Gordon Turriff, for the respondent W. Glen How.

John M. Burns and Linda How, for the respondent Watch Tower Bible and Tract Society of Canada.

Brian Evernden, for the intervener the Attorney General of Canada.

Michel Y. Hélie, for the intervener the Attorney General for Ontario.

Monique Rousseau and Isabelle Harnois, for the intervener the Attorney General of Quebec.

Shawn Greenberg, for the intervener the Attorney General of Manitoba.

Written submission only for the intervener the Attorney General of British Columbia.

Gerald D. Chipeur and Karnick Doukmetzian, for the intervener the Seventh‑day Adventist Church in Canada.

The reasons of La Forest and Gonthier JJ. were delivered by

La Forest J. -- I have had the advantage of reading the reasons of my colleagues. I am in agreement with the reasons of Justice L'Heureux-Dubé that the issue of access should be determined on the basis of what is in the best interest of the child. I also agree with her on the constitutional issue.

On the property and monetary issues, I am in agreement with the reasons of Justice McLachlin. I also agree with her on the principles that should govern costs and accordingly with the order of the Court of Appeal respecting the costs at trial. However, since the appellant was largely successful in the final result, I would allow her costs on this appeal and the Court of Appeal.

Accordingly, I would allow the appeal and restore the order of the trial judge on all matters except the monetary issues, which I would dispose of in the manner proposed by McLachlin J. I would award the appellant her costs at trial as set forth by the Court of Appeal, as well as costs on a party and party basis in this Court and in the Court of Appeal.

The following are the reasons delivered by

L'Heureux‑Dubé J. (dissenting in result) -- I have had the advantage of reading the opinion of my colleague, Justice McLachlin. With great deference, I disagree both with her reasons and the result she has reached. Since I do not characterize the issue quite as my colleague does, I will pursue my own analysis.

The main issue in this case, in my view, concerns access by a non‑custodial parent to his children whose custody was granted to the other parent. More precisely, this Court must determine whether curtailment of access is warranted in the circumstances of this case. The focus of the inquiry is the standard applicable to such a determination. According to the Court of Appeal, the test is one of harm to the children. I disagree. In my view, the only applicable test is the best interests of the children, assessed from a child‑centred perspective, a test which is mandated by the Divorce Act, R.S.C., 1985, c. 3 (2nd Supp.) (the "Act"), as well as provincial legislation and which is universally applied and constitutionally sound. While the respondent, the father in this case, raises the issues of freedom of religion and expression and the infringement of his guarantees under s. 2(a) and (b) of the Canadian Charter of Rights and Freedoms (the "Charter") due to the trial judge's order restricting access to his children, these questions simply do not arise in the circumstances of this case. If they do, I agree with my colleague that there is no infringement of the Charter.

On the related issues of maintenance, division of property and costs, I agree with the trial judge for the reasons that she expressed at length, given her privileged vantage point and her findings of fact. Accordingly, I will deal only with the main issue: access.

Facts and Judgments

Since the facts leading to the dispute between the parties have been set out by my colleague, I will only point out those most relevant to my subsequent discussion.

The parties, who had two children when they first separated in February, 1987, resumed cohabitation and finally separated in August, 1987, shortly after the birth of their third child. It appears from the evidence that the breakdown of their marriage was due, to a great extent, to religious differences between the parties. The respondent became involved, unbeknownst to the appellant, in the Jehovah's Witnesses faith some two years before the separation. The children had been baptized in the United Church, and the appellant, an Anglican, apparently did not favour any involvement for herself or the children in her husband's new faith. At the time, the older children were six and eight years of age.

Upon separation, it was common ground between the parties that the appellant should have custody of the children and they have remained at all times since in the care of their mother. On July 12, 1988, the appellant brought an application for divorce and sought spousal and child support, interim and permanent custody of the children and an order enjoining the respondent from inculcating the children in the Jehovah's Witness faith or involving them in church activities. The respondent brought a cross‑application for interim joint custody with care and control to the appellant and a declaration that the restrictions sought by his wife violated his and the children's rights under the Charter.

In the interim order of August 16, 1988, Scarth L.J.S.C. dismissed the respondent's motion for a declaration and order pursuant to s. 24 of the Charter. He granted custody to the appellant, awarded her $3,500 a month in spousal and child support and specified the following restrictions on the respondent's access:

a)the Respondent will not take the said children of the marriage to the Respondent's church without the Petitioner's consent and if the Petitioner withholds consent, the Respondent may make application to this court;

b) the Respondent will not suggest that the three children of the marriage will have to separate from the Petitioner if she does not convert to the Jehovah's Witnesses [sic] faith, or that the Petitioner will be punished by God for failing to adhere to that faith;

c) the Respondent will not deprecate the Petitioner's religion in the presence of the three infant children of the marriage;

d) the Respondent is enjoined from preventing the said infants from having blood transfusions in the event blood transfusions are required and the Petitioner shall be solely responsible for all health care decisions and choices relative to the said infant children of the marriage.

The respondent appealed the order and instituted a number of further applications in which he sought to vary the interim custody order and the terms and conditions imposed on access. He also applied to set the dates and times of access and sought overnight access. The appeal of the interim order was dismissed by the B.C. Court of Appeal on September 23, 1988 and the applications for interim and permanent custody of the children and variation of the restrictions were rejected by Scarth L.J.S.C. on October 7, 1988. The respondent then appealed the order of October 7, 1988 and sought a declaration that both the order and s. 16(1), (2), (6) and (8) of the Act violated his Charter rights. This appeal was dismissed on November 15, 1988. Prior to trial, the respondent also unsuccessfully contested the amount of support to be paid and applied to list the matrimonial home for sale and have the arrears of support accumulated to that date cancelled.

At trial (1989), 1989 8903 (BC SC), 24 R.F.L. (3d) 193, on the issue of custody and access, Proudfoot J. (now J.A.) found that the respondent's attempts to argue loss of freedom of religion were but a guise under which he was attempting to gain control of the children's religious upbringing. She held that the respondent would not be denied custody or access because of his religious beliefs and that his freedom of religion had not been infringed. Rather, custody and access would be decided only according to the best interests of the children. She found that the appellant did not oppose access by the respondent, and that the difficulty was created solely by their religious differences. Distinguishing the decision of the Divisional Court of the Supreme Court of Ontario in Hockey v. Hockey (1989), 1989 4299 (ON SC), 21 R.F.L. (3d) 105, and relying on Anson v. Anson (1987), 10 B.C.L.R. (2d) 357 (Co. Ct.), and Andrews v. Andrews, B.C.S.C., June 9, 1983, unreported, Proudfoot J. found that both the right and the responsibility to determine a child's education, health care and religion rest with the custodial parent. Citing the decision of the Saskatchewan Court of Appeal in Brown v. Brown (1983), 1983 2311 (SK CA), 39 R.F.L. (2d) 396, she found that the Charter has not affected the general principle enunciated in DeLaurier v. Jackson, 1934 2 (SCC), [1934] S.C.R. 149, that the father's rights in respect of the religious upbringing of his child must yield to the welfare of the child.

Proudfoot J. noted that the original custody order had been made with the respondent's consent and that all the experts agreed that custody should remain with the appellant. The children had indicated in letters written to the judge that they wished to remain with their mother. The evidence was clear that they were opposed to any involvement in the respondent's religious activities and that there had been total disregard for their wishes. The reports of both court experts indicated that the two older children's relationship with their father was deteriorating, that they were under increasing pressure regarding family issues and needed to settle the conflict so they could proceed with their lives. Proudfoot J. found that there was no doubt on the evidence that the religious conflict was causing a problem for the children and that it should be resolved. She granted the appellant custody and guardianship of the children with sole responsibility for their religious upbringing, health care and education and granted the respondent access subject to the following conditions:

a) that the Respondent shall not discuss the Jehovah's Witness religion with the children and shall not take any of the children to any religious services, canvassing or meetings, without the written consent of the Petitioner and shall not expose the children to religious discussions with a third party or parties without the written consent of the Petitioner;

b) that the Respondent is enjoined from preventing any of the three infant children of the marriage from having blood transfusions in the event blood transfusions are required;

c)and that the Respondent shall not make any adverse remarks with respect to the religious beliefs of the Petitioner....

She further declared that neither the custody order nor the access restrictions infringed the respondent's freedom of religion under the Charter and made other incidental findings as to support, the division of matrimonial property and costs, with which I agree as already indicated.

On appeal (1990), 1990 3813 (BC CA), 50 B.C.L.R. (2d) 1, the respondent abandoned his custody claim but sought an order for joint guardianship of the children, unrestricted access and a declaration that the restrictions on access violated his and the children's Charter rights. Upon reviewing the rights of custodial parents and the historical development of access, Wood J.A. for the majority (Cummings J.A. concurring on this issue) noted that the concept of access did not receive statutory recognition until the Act. Wood J.A. found that the terms of s. 16 of the Act, when taken as a whole, indicate an intention by Parliament to foster substantial changes in the law of custody and access and to facilitate a meaningful and continuing relationship with the children of the marriage and the access parent. In his opinion, the word "contact" within s. 16(10) should be given a broad meaning, and a proper construction of the Act requires the distribution of "rights" between the custodial and the access parent. He found that an order for custody under s. 16(1) of the Act necessarily gives the custodial parent the power to determine the religious upbringing of the child, but does not give that parent the right to prevent the other from sharing his or her views with the child, whether that takes the form of discussions, observances or other related activities. The right of the access parent to share those beliefs is subject only to two limitations: (a) the unwillingness of the child to participate and (b) the power of the court to restrict the access if exposure to the beliefs or practices is, or is likely to be, harmful to the well-being of the child. Such harm must be "real psychological or physical harm", established on a civil standard of proof and must be distinguished from the general emotional distress which every child experiences upon divorce.

As to freedom of religion, Wood J.A. was of the view that the issue was whether or not the access parent's fundamental freedom of religion under s. 2(a) of the Charter is infringed by the long‑standing common law rule which would give the custodial parent, as guardian of the child of the marriage, the absolute right to control the extent to which the access parent can expose that child to his or her religious beliefs. After considering the nature of freedom of religion set out in R. v. Big M Drug Mart Ltd., 1985 69 (SCC), [1985] 1 S.C.R. 295, Wood J.A. held that an important element of the right to teach and disseminate one's beliefs must be the right to share those beliefs with one's children. In his view, it would be a rare case in which a court would be required to determine whether or not the exposure to conflicting religious beliefs or the beliefs themselves was threatening real physical or psychological harm to the child. Before embarking on such an inquiry, a court would require the evidence of an independent expert.

Applying these principles to the facts of the case, Wood J.A. concluded at p. 109 that:

No such harm was diagnosed or anticipated by either of the expert witnesses which the learned trial judge found to be credible. Nor did the evidence establish that the conflict in the religious beliefs of the parents was causing, or likely to cause, such harm.

The evidence went no further than to establish that the children were under some stress, the cause of which was "multi‑factorial", but which obviously included some component related to the religious dispute between the parents. As I have attempted to show such evidence would not even justify the court embarking on any inquiry into the religious beliefs of a parent. It is therefore axiomatic that it would not support any interference with Mr. Young's fundamental freedom of religion under s. 2(a) of the Charter, in the form of the restrictions imposed.

Nonetheless, Wood J.A. held that the respondent was bound to respect the girls' wishes not to accompany him to his place of worship or on his proselytization efforts. Finding no reason to disbelieve the respondent's willingness to do so, Wood J.A. held that there was no basis for the imposition of any restrictions on his rights of access. In the result, he struck out the restrictions on access, found the declaration that the appellant have sole responsibility for the religious upbringing, health care and education of the children unnecessary and struck out the order that neither party make adverse remarks with reference to the religious beliefs of the other.

Southin J.A., dissenting on this issue, after a comprehensive review of parental authority, concluded that the custodial parent today enjoys the "full plenitude" of parental powers historically enjoyed by a guardian even during the time that the child is with the access parent. In her view, nothing in either the Act or the B.C. Family Relations Act, R.S.B.C. 1979, c. 121, was intended to limit the long‑standing rule that the custodial parent is entitled to "lay down the law" concerning all matters relating to the children even during the time of access. In her view, the word "contact" in s. 16(10) of the Act did not contemplate a right in anyone granted access to disregard the determinations of the custodial parent, as it is a principle of statutory interpretation that (at p. 32) "if Parliament intends to set at naught a long‑standing concept of the common law, it does so plainly and not by woolly words capable of more than one interpretation." With respect to the Charter, Southin J.A. held that the rights within s. 2 must not be construed in such a way as to deny the existence of parental rights, which are preserved by s. 26 of the Charter. She concluded that the appellant may control the children's activities but that she may not control the conversations which take place during the time of access. In her view, the appellant was entitled to establish that it would not be in the best interests of the children to be exposed to the religious observances of the respondent. In the result, she would have ordered that the respondent not take the children to any religious service or have the children accompany him in any religious activity without the written consent of the Petitioner.

Relevant Statutory Provisions

In the present case, the Court is concerned with the statutory criteria which guide courts in matters of custody and access. A proper starting point, then, is to spell out those provisions of the Divorce Act, which are relevant to a discussion of custody and access:

(1) A court of competent jurisdiction may, on application by either or both spouses or by any other person, make an order respecting the custody of or the access to, or the custody of and access to, any or all children of the marriage.

(4) The court may make an order under this section granting custody of, or access to, any or all children of the marriage to any one or more persons.

(5) Unless the court orders otherwise, a spouse who is granted access to a child of the marriage has the right to make inquiries, and to be given information, as to the health, education and welfare of the child.

(6) The court may make an order under this section for a definite or indefinite period or until the happening of a specified event and may impose such other terms, conditions or restrictions in connection therewith as it thinks fit and just.

(8) In making an order under this section, the court shall take into consideration only the best interests of the child of the marriage as determined by reference to the condition, means, needs and other circumstances of the child.

(9) In making an order under this section, the court shall not take into consideration the past conduct of any person unless the conduct is relevant to the ability of that person to act as a parent of a child.

(10) In making an order under this section, the court shall give effect to the principle that a child of the marriage should have as much contact with each spouse as is consistent with the best interests of the child and, for that purpose, shall take into consideration the willingness of the person for whom custody is sought to facilitate such contact.

. . .

(5) Before the court makes a variation order in respect of a custody order, the court shall satisfy itself that there has been a change in the condition, means, needs or other circumstances of the child of the marriage occurring since the making of the custody order or the last variation order made in respect of that order, as the case may be, and, in making the variation order, the court shall take into consideration only the best interests of the child as determined by reference to that change. [Emphasis added.]

Custody

Leaving aside for the moment the constitutional questions stated by Lamer C.J., I wish to focus on the matter of custody and access. While the issue of custody is no longer in dispute, the parties having agreed that the appellant should retain custody of their three children, it is still necessary, to set the stage fully, to examine the rationale underlying the granting of custody.

At common law, the right to custody of children was originally incontestable and reposed with the father to the exclusion of any claims of the mother. The father possessed the full bundle of rights which are an inherent feature of the relationship between parent and child. This right reflected the dominant ideology of the family before this century and was composed of several interrelated concepts. The patriarchal family was the preferred, indeed the only recognized family structure. Fathers were regarded as possessing a natural right to their children, which flowed from the belief that they knew best about the interests of their children and that those interests were best served by upholding their proprietary rights (see S. Maidment, Child Custody and Divorce (1984), at pp. 5‑6; In re Agar-Ellis, [1883] Ch. D. 317). Thus, the sole concern in custody matters was for the rights of the father with little regard for the effect of such decisions on the child. As the famous case of R. v. De Manneville (1804), 5 East. 221, 102 E.R. 1054, illustrates, under such a regime, a mother could be denied custody and even access to a child still breast‑feeding notwithstanding that the undisputed cause of the marriage breakdown was the cruelty of the father.

Since the 19th century, along with a gradual move toward equality of the sexes in custody presumptions, the focus has shifted from the rights of the parents to those of the child. The first statutory inroads were made in England with the passage of Talfourd's Act in 1839 (An Act to amend the Law relating to the Custody of Infants (U.K.) , 2 & 3 Vict., c. 54) which allowed courts to make an order of access to the mother, and even to award her the custody of children of "tender years", those under the age of seven. However, fathers still retained the presumptive right to custody, a right which would only be disturbed where there was serious reason to believe that his behaviour would be detrimental to the child. In contrast, mothers who engaged in adulterous behaviour were deemed to have threatened the family structure and were presumed unfit to protect the interests of the child and routinely denied custody.

The best interests of the child gained ascendancy as the proper focus of custody decisions at the same time as courts moved toward the equality of women in custody decisions. The power of courts to rule in the best interests of the child was originally found in the equitable parens patriae jurisdiction of the Courts of Chancery. Although this power was at first only exercised in respect of the property rights of the child, the concept of the best interests of the child was gradually expanded to include the emotional, physical and spiritual welfare of the child. Changes in the social landscape affecting both the nature of work and the family and growing recognition of the needs of children began to inform custody decisions. With the merger of the courts of equity and common law in the Supreme Court of Judicature Act, 1873 (U.K.), 36 & 37 Vict., c. 66, the ability of courts to overrule any common law parental right where found to be in the best interests of the child was confirmed, as s. 25(10) of the Act specifically provided that "In questions relating to the custody and education of infants the rules of equity shall prevail". In England, Parliament specifically recognized as important a consideration of the welfare of the child in the Guardianship of Infants Act, 1886 (U.K.), 49 & 50 Vict., c. 27. Upon the passage of the Guardianship of Infants Act, 1925 (U.K.), 15 & 16 Geo. 5, c. 45, the welfare of the child became the "the first and paramount consideration". The welfare, or what is now called the best interests, of the child thus prevailed over any potential rights that parents previously held at common law.

A number of rules were developed by courts, presumably predicated on the best interests test, which made the outcome of custody matters dependent upon assumptions as to parents' suitability. The most important of these rules was the tender years doctrine, which was described by an Ontario court in 1933 as the common sense proposition that, in general, children under the age of seven needed the care of their mothers (Re Orr, 1933 110 (ON CA), [1933] O.R. 212 (C.A.)). A related presumption favoured the placement of older children with the parent of the same sex. The Guardianship of Infants Act, 1925, granted women equal rights to custody over their children in England. In Canada, statutes providing for equality in entitlement to custody were passed in some provinces as early as 1923, although the process of legislative entrenchment was not completed until 1977 (see R. S. Abella, "Family Law in Ontario: Changing Assumptions" (1981), 13 Ottawa L. Rev. 1, at p. 13). The net effect of both legislative enactments and judicially developed presumptions was that women were granted custody of children in the majority of cases.

Although the welfare principle long predates the 1970s, it was at that time that the emerging focus on the rights of children began to supplant the tender years doctrine, as courts adopted the principle expressed by de Grandpré J. in Talsky v. Talsky, 1975 29 (SCC), [1976] 2 S.C.R 292, at p. 293, that "the paramount consideration in custody matters is the welfare of the infants". Concurrent with this shift was the dissociation of notions of matrimonial fault from assumptions about parenting ability. Following the Law Reform Commission of Canada's report on Family Law in 1976, the best interests test was entrenched as the sole criterion for the determination of custody and access in the Divorce Act. Since then, not only has the best interests of the child become the paramount consideration, legislation has dictated that it be the sole consideration.

Despite these changes over time with respect to who is regarded as the appropriate custodial parent, the nature and scope of custody itself have remained relatively constant. The chief feature of such orders was, and still is, the implied, if not explicit, conferral of parental authority on the person granted custody. The long-standing rule at common law is that an order of custody entails the right to exercise full parental authority. In the case of a sole custody order, that authority is vested in one parent to the exclusion of the other.

The power of the custodial parent is not a "right" with independent value which is granted by courts for the benefit of the parent, but is designed to enable that parent to discharge his or her responsibilities and obligations to the child. It is, in fact, the child's right to a parent who will look after his or her best interests. Indeed, courts have recognized that there is no magic to the parental tie and will, when the best interests of the child warrant, grant custody to a third party. As Dubin J.A. (now C.J.O.) held in Re Moores and Feldstein, 1973 535 (ON CA), [1973] 3 O.R. 921, at p. 928:

. . . it is the duty of the Court to view all the circumstances relevant to what is in the interest of the child, including a consideration as to whether the evidence disclosed that the child would benefit from the tie of a child to its mother.

It has long been recognized that the custodial parent has a duty to ensure, protect and promote the best interests of the child. That duty includes the sole and primary responsibility to oversee all aspects of day to day life and long‑term well‑being, as well as major decisions with respect to education, religion, health and well‑being. This is reflected in the decision of the Ontario Court of Appeal in Kruger v. Kruger (1979), 1979 1663 (ON CA), 25 O.R. (2d) 673, at p. 677, in which Thorson J.A. stated:

In my view, to award one parent the exclusive custody of a child is to clothe that parent, for whatever period he or she is awarded the custody, with full parental control over, and ultimate parental responsibility for, the care, upbringing and education of the child, generally to the exclusion of the right of the other parent to interfere in the decisions that are made in exercising that control or in carrying out that responsibility.

See also Baker v. Baker (1979), 1979 1962 (ON CA), 8 R.F.L. (2d) 236 (Ont. C.A.).

A parallel situation exists with respect to the exercise of parental authority on separation or divorce under the Civil Code of Quebec. Article 647 of the C.C.Q. grants equal rights to parents in the exercise of authority over their children, stating that "[t]he father and mother have the rights and duties of custody, supervision and education of their children". However, art. 569 requires courts to decide questions of custody, maintenance and education of children "in their interest and in the respect of their rights", and art. 570 preserves only the right of the non‑custodial parent to surveillance over such matters. As Beetz J., writing for this Court, held in C.(G.) v. V.‑F.(T.), 1987 20 (SCC), [1987] 2 S.C.R. 244, at p. 280, the effect of a custody award is to remove the right of the non‑custodial parent to exercise his or her parental authority.

The rationale for granting this authority to the custodial parent is described by Gendreau J.A. in Dussault v. Ladouceur (1987), 1987 876 (QC CA), 14 R.F.L. (3d) 185 (Que. C.A.), at p. 191, as follows:

[translation] It goes without saying that as a general rule, when custody of a child is granted to one parent, he or she then exercises all the attributes of parental authority and the other parent does not normally interfere in the custodial parent's way of doing things, except in the exercise of his or her supervising role. The child's unity of development is preserved in this manner and any scatterings or dislocations that might be harmful to him or her are avoided.

See also Gunn v. Gunn (1975), 1975 1938 (PE PC), 24 R.F.L. 182 (P.E.I.S.C.); Benoit v. Benoit (1972), 1972 1869 (ON CJ), 6 R.F.L. 180 (Ont. Prov. Ct.), reversed on other grounds (1972), 1972 1936 (ON CA), 10 R.F.L. 282 (Ont. C.A.). Similarly, under the B.C. Family Relations Act, the custodial parent has the full bundle of rights over the child (Charlton v. Charlton (1980), 1980 545 (BC SC), 15 R.F.L. (2d) 220 (B.C.S.C.); Anson v. Anson, supra).

As Professor Julien Payne has observed in Payne on Divorce (2nd ed. 1988), at p. 144, it has, at times, been thought, both in Canada and elsewhere, that custody and guardianship of a child were different in scope. While the case law is not consistent on this point, the term "custody" has been used in two senses. When used in its wider sense, custody is akin to the concept of guardianship and encompasses the full bundle of parental rights. In its narrower sense, the term refers only to the right to physical custody or day to day care and control of the child (see Hewer v. Bryant, [1970] 1 Q.B. 357; Anson v. Anson, supra, at p. 363). However, as Payne notes at pp. 144-45, at this point in time, the wider concept of custody is well entrenched in Canadian law:

In the context of Canadian divorce proceedings, the case law tends to support the conclusion that, in the absence of directions to the contrary, an order granting "sole custody" to one parent signifies that the custodial parent shall exercise all the powers of the legal guardian of the child. The non‑custodial parent with access privileges is thus deprived of the rights and responsibilities that previously vested in that parent as a joint custodian of the child.

This view is reflected in the decision of Huddart Co. Ct. J. in Anson v. Anson, supra, at p. 366:

There seems to be no limit other than that of the imagination of the bar and bench to the variety of orders that can be devised to ensure the best interests of a child are served by the distribution of custodial rights and duties between the interested adults. The law reports resound with examples. Key, however, is where the residual authority lies. It seems too late to suggest that the term custody be confined to physical care and control with guardianship of the person being the residual concept, including general supervision of the upbringing of the child. The wide meaning is entrenched in federal legislation and in the case law here and elsewhere. [Emphasis added.]

See also Clarke v. Clarke (1987), 1987 2745 (BC SC), 7 R.F.L. (3d) 176 (B.C.S.C.).

Southin J.A., the dissenting judge on this issue in the Court of Appeal, dealt at p. 31 with the question of custody under the B.C. provincial legislation in the following terms:

I can find nothing in the 1978 Act which warrants an inference that the Legislature intended by it to cut down the powers of guardians as those powers have long been understood. It follows that, if the order giving custody to Mrs. Young is sustained, she has under the Family Relations Act the full right to lay down the law concerning all matters relating to the children even during the time they are with their father for the purpose of his exercise of access.

The non‑custodial parent retains certain residual rights over the child as one of its two natural guardians, among which is the right to apply to the court for variation of custody and access terms. Various other entitlements have been recognized at common law, including, subject to the best interests of the child, the right to access, the right to contest the child's adoption, the right to claim guardianship of the person of the child upon the death of the custodial parent and the right to succeed to the child's property among others (see Maidment, supra, at p. 34; Dussault v. Ladouceur, supra; Law Reform Commission of Canada, report on Family Law, supra; Law Reform Commission of Canada, The Custody, Care and Upbringing of Children of Divorcing Spouses, 1973).

The traditional decision‑making power of the custodial parent recognized by law is intimately connected to the welfare of the child, as the need for a secure and constant source of parental responsibility in the life of the child is well understood among those who are knowledgeable in the psychology of children. J. Goldstein, A. Freud and A. J. Solnit in Beyond the Best Interests of the Child (1979) identified three imperatives that must govern child placement decisions such as custody arrangements. Such decisions should: safeguard the child's need for continuity of relationships, reflect the child's, not the adult's, sense of time, and take into account the law's inability to supervise interpersonal relationships and the limits of knowledge to make long‑range predictions (supra, at pp. 31, 40 and 49). The need for continuity generally requires that the custodial parent have the autonomy to raise the child as he or she sees fit without interference with that authority by the state or the non‑custodial parent, as it is the inability of the custodial parent to protect those interests sufficiently which poses the real threat to the welfare of the child. A custody award can thus be regarded as a matter of whose decisions to prefer, as opposed to which decisions to prefer. In the words of B. M. Dickens, "The Modern Function and Limits of Parental Rights" (1981), 97 L.Q. Rev. 462, at p. 473 "the identification of who is allowed to decide is no less important in theory, and more important in practice, than how choice is to be exercised between options". (See also J. Goldstein, "In Whose Best Interest?", in R. S. Abella and C. L'Heureux‑Dubé, eds., Family Law: Dimensions of Justice (1983), at p. 124.)

As Goldstein, Freud and Solnit stress, an important function of the law on divorce or separation is to reinforce the remainder of the family unit so that children may get on with their lives with as little disruption as possible. Courts are not in a position, nor do they presume to be able, to make the necessary day‑to‑day decisions which affect the best interests of the child. That task must rest with the custodial parent, as he or she is the person best placed to assess the needs of the child in all its dimensions. In my view, this was eloquently expressed by Weatherston J. (later J.A.) in McCahill v. Robertson (1974), 1974 2225 (ON SC), 17 R.F.L. 23 (Ont. S.C), at pp. 23‑24:

A child must know where its home is and to whom it must look for guidance and admonition and the person having custody and having that responsibility must have the opportunity to exercise it without any feeling by the infant that it can look elsewhere. It may be an unfortunate thing for the spouse who does not have custody that he or she does lose a great deal of the authority and indeed to some extent the love and affection of the child that might otherwise be gained, but this is one of the things which is inherent in separation and divorce. The parents cannot have it both ways. As I say, in my view, it is vitally necessary that the child know where its home is, to whom it is responsible and that there be no doubt in the mind of the child as to that. Within those limits, the parent who does not have custody should, of course, have access to the child under terms which are as reasonably generous as possible, but without interfering with that basic responsibility on the parent having custody. [Emphasis added.]

Once a court has determined who is the appropriate custodial parent, it must, indeed it can do no more than, presume that that parent will act in the best interests of the child. As the New Brunswick Court of Appeal stated in Fougere v. Fougere (1987), 1987 7842 (NB CA), 77 N.B.R. (2d) 381, at p. 384:

Once courts award custody they must, in our view, support the custodial parent in that parent's reasonable efforts to bring up the children, including the right of the custodial parent to decide questions relating to the religious upbringing of the children.

As has been widely observed by those studying the nature and sources of changes in family institutions, popular notions of parenthood and parenting roles have undergone a profound evolution both in Canada and elsewhere in the world in recent years (see generally M. A. Fineman, The Illusion of Equality: The Rhetoric and Reality of Divorce Reform (1991); L. J. Weitzman, The Divorce Revolution (1985); J. Drakich, "In Whose Best Interests? The Politics of Joint Custody", in B. Fox, ed., Family Bonds and Gender Divisions: Readings in the Sociology of the Family (1988); C. Smart and S. Sevenhuijsen, Child Custody and the Politics of Gender (1989), and S. B. Boyd, "Women, Men and Relationships with Children: Is Equality Possible?", in K. Busby, L. Fainstein and H. Penner, eds., Equality Issues in Family Law: Considerations for Test Case Litigation (1990)). One of the central tenets of this new vision is that child care both is no longer and should no longer be exclusively or primarily the preserve of women. Society has largely moved away from the assumptions embodied in the tender years doctrine that women are inherently imbued with characteristics which render them better custodial parents (for a discussion, see D. L. Chambers, "Rethinking the Substantive Rules for Custody Disputes in Divorce" (1984), 83 Mich. L. Rev. 477). Moreover, both economic necessity and the movement toward social and economic equality for women have resulted in an increase in the number of women in the paid workforce. Many people have tended to assume that a natural result of this change would be the concurrent sharing of household and childcare responsibilities with spouses, companions and, of course, fathers. In addition, the increased emphasis on the participation of fathers in the raising of children and financial support after divorce gave rise to claims by fathers and fathers' rights groups for legislative changes that would entitle them to the benefit of neutral presumptions in custody decisions.

A corollary of the acceptance of neutral parenting roles is the notion that children after divorce need to maintain contact with both parents. It is now widely assumed to be self‑evident in the child's best interests to ensure the non‑custodial parent's involvement in the life of the child. One result of these changes has been the emergence of joint custody awards which are predicated explicitly on equality of parental responsibilities and the belief that children's interests are served by maximizing the involvement of both parents in decisions concerning the child.

The custody provisions of the Act reflect, to some degree, this evolving view of parental roles. Under the best interests test, courts no longer automatically grant custody according to the tender years doctrine. Instead, decisions are made according to the best interests of the child without the benefit of a presumption in favour of the mother, or, for that matter, the father. Subsection 16(4) of the Act, for the first time, specifies that a court may make joint custody orders. Subsection 16(10) of the Act contains the "friendly parent" rule which directs courts when granting custody to take into account the willingness of the parent seeking custody to maximize the contact of the other parent with the child of the marriage. It is clear, then, that the Act envisages contact between the child and each of its parents as a worthy goal which should be, all other things being equal, in the best interests of the child. This is a value which courts have always recognized in generally granting generous access to the non‑custodial parent.

Wood J.A., however, has interpreted the Act as reflecting significant changes to the traditional law of custody and access. More specifically, he expressed the view that s. 16(10) of the Act indicates a desire on the part of Parliament to enlarge the concept of access beyond its traditional scope. Therefore, according to him, the traditional decision‑making power of custodial parents has been curtailed by s. 16 which, taken as a whole, requires a distribution of rights between the parents.

To that end, Wood J.A. relies on the views expressed about the nature of custody in the English decision of Dipper v. Dipper, [1980] 2 All E.R. 722, as approximating the philosophy found in the Act. In that case, the English Court of Appeal held at p. 733:

. . . it [is] a fallacy which continues to raise its ugly head that, on making a custody order, the custodial parent has a right to take all the decisions about the education of the children in spite of the disagreements of the other parent. That is quite wrong. The parent is always entitled, whatever his custodial status, to know and be consulted about the future education of the children and any other major matters. If he disagrees with the course proposed by the custodial parent he has the right to come to the court in order that the difference may be determined by the court.

The "right to know" is, of course, embodied in the Act, but the "right to be consulted" and to have disagreements aired in Court is not. The only ground set out in the Act on which to contest a decision of the custodial parent is to show that the decision is contrary to the best interests of the child. Courts in Canada have never adopted the view that the custodial parent's decisions are subject to the approval of the non‑custodial parent, and that such disagreements must be resolved in court.

Furthermore, in my view, there is nothing in the Act which mandates such a reversal in the philosophy of custody as held up to now. The Law Reform Commission of Canada, report on Family Law, supra, on which the Act is based, defines custody at p. 46 in the following manner:

4.4 "Custody" has several legal meanings. It stands for the whole collection of legal powers (many of which connote parental obligations as much as "rights") of fathers and mothers over their children: the power to raise and control the child, to determine the nature and amount of the child's education, to determine his or her religious upbringing, to administer the child's property, to grant or withhold consent to the marriage of an under‑age child, to apply to the courts on his or her behalf, and so on. In its narrower sense, it means simply "care and upbringing" or, to use a more vivid but unfortunate phrase, "possession".

4.5 Where custody is granted to one parent, the courts afford the other the right to "reasonable access" ‑‑ usually worked out between the parents, but where they cannot agree, as directed by the judge.

This description of custody mirrors the traditional elements of parental authority and in no way suggests a blending of the rights of the access parent with those of the custodial parent. While the Act specifies that access is a value to be preserved, the objective in s. 16(10) that a child have "as much contact with each spouse" is immediately qualified by the proviso "as is consistent with the best interests of the child" (emphasis added). Thus, it is clear that maximum contact is not an unbridled objective, and that it must be curtailed wherever the welfare of the child requires it. The best interests of the child remain the prism through which all other considerations are refracted.

Professor Payne, in Payne on Divorce, supra, at p. 144, shares the view that the access provisions in the Act do not confer any authority on the non‑custodial parent to participate in the major decisions of a child's life. Rather, s. 16(5) of the Act provides that "[u]nless the court orders otherwise, a spouse who is granted access to a child of the marriage has the right to make inquiries, and to be given information, as to the health, education and welfare of the child". It does not mandate that the custodial parent consult with the access spouse when making such decisions with respect to the child. (See also Anson v. Anson, supra). Custody as defined in s. 2(1) of the Act "includes care, upbringing and any other incident of custody". Payne states at pp. 145‑46:

The provisions of the Divorce Act, 1985, and particularly the definitions of "custody" and "accès" in subsection 2(1) of the Act, apparently preclude Canadian courts from reverting to a narrow definition of custody. The word "includes" in the definition of custody necessarily implies that the term embraces a wider range of powers tha[n] those specifically designated in subsection 2(1).... The opinions expressed in Kruger v. Kruger, supra, have thus been statutorily endorsed by the Divorce Act, 1985. Consequently, in the absence of a successful application to vary an unqualified sole custody disposition with respect to all or any of the incidents of custody, the non‑custodial spouse with access privileges is a passive bystander who is excluded from the decision‑making process in matters relating to the child's welfare, growth and development. This remains true notwithstanding that subsection 16(10) of the Divorce Act, 1985 provides that the court shall promote "maximum contact" between the child and the non‑custodial parent to the extent that this is consistent with the best interests of the child. [Emphasis added.]

I could not agree more, given the wording of the Act itself and the generally accepted view of custody both before and since its enactment. The Act neither suggests nor requires the division of parental responsibilities between the custodial and access parent. If Parliament had intended such a result, it would have used much clearer and less ambiguous language.

It goes without saying that I do not share the assertion of my colleague that "the proposition, put to us in argument, that the custodial parent should have the right to forbid certain types of contact between the access parent and the child, must fail" (p. 119). The proposition is not one of "rights"; it is one of duty and obligation to the child's best interests. For example, a custodial parent, aware of sexual or other abuse by the non‑custodial parent would be remiss in his or her duty to the child not to cut off access by the abuser immediately, with or without a court order. One cannot stress enough that it is from the perspective of the child's interests that these powers and responsibilities must be assessed, as the "rights" of the parent are not a criterion.

The arguments in favour of increased authority over the child by the access parent are closely related to those which support a presumption in favour of joint custody (see J. Folberg, "Joint Custody", in R.S. Abella and C. L'Heureux‑Dubé, eds., supra, at p. 185). They rest on the premise that the relationship of authority and obligation that existed between each of the parents and the child during the marriage should and can continue, despite the fact that the parents may no longer be willing or able to cooperate on its exercise. While joint custody may remain an ideal solution in proper cases, particularly when parents are willing and able to cooperate, such premises are often based on illusion rather than reality and may, in the words of Thorson J.A., amount to "a triumph of optimism over prudence" (Kruger v. Kruger, supra, at p. 681).

Unlike in other jurisdictions, the Act contains no presumption in favour of joint custody. Nor have Canadian courts generally accepted the view that joint custody is to be preferred in judicial orders as to custody and access. Beside the fact that the term "joint custody" is a misnomer (see C. L'Heureux‑Dubé, "La garde conjointe, concept acceptable ou non?" (1979), 39 R. du B. 835), courts have demonstrated a strong hesitation to making joint custody orders, since in those custody disputes which end up before the courts, the animosity between parents does not generally favour such dispositions. Thorson J.A. observed in Kruger v. Kruger, supra, at p. 678 that:

Above all, [joint custody] requires a willingness by both parents to work together to ensure the success of the arrangement. Such a willingness must be sincere and genuine; by its very nature it is not something that can be imposed by a Court on two persons, one or both of whom may be unwilling or reluctant to accept it in all its implications. Like marriage itself if it is to succeed, it is an arrangement that has to be worked out by two persons who are determined, or their own will and in good faith, to make it work. [Emphasis in original.]

When parents are willing and able to share parenting responsibilities, they usually do so by agreement, which courts generally uphold (Kruger v. Kruger, supra; Baker v. Baker, supra; Keyes v. Gordon (1985), 1985 123 (NS CA), 45 R.F.L. (2d) 177 (N.S.C.A.); Dussault v. Ladouceur, supra; Droit de la famille -‑ 316, [1986] R.D.F. 651 (Que. C.A.)). Courts are also reluctant to interfere with shared parenting arrangements that have survived for a period of time after parental separation or divorce (see J. D. Payne and B. Edwards, "Co‑operative Parenting After Divorce: A Canadian Perspective" (1989), 11 Advocates' Q. 1, at pp. 13‑15; S. M. Holmes, "Imposed Joint Legal Custody: Children's Interests or Parental Rights?" (1987), 45 U.T. Fac. L. Rev. 300).

But the reality of divorce and the circumstances of the parties cannot easily be dismissed. When implementing the objectives of the Act, whether considering joint custody or fashioning access orders, courts, in my view, must be conscious of the gap between the ideals of shared parenting and the social reality of custody and childcare decisions. Despite the neutrality of the Act, forces such as wage rates, job ghettos and socialization about care giving still operate in a manner that cause many women to "choose" to take on the care giving role both during marriage and after divorce. Hence, "we remain a good distance away from the kind of equality which would make decisions between mothers and fathers as to who is going to leave the workforce to care for young children truly gender neutral" (S.B. Boyd, "Women, Men and Relationships with Children: Is Equality Possible?", supra, at p. 76; S. B. Boyd, "Child Custody Law and the Invisibility of Women's Work" (1989), 96 Queen's Q. 831). Moreover, research uniformly shows that men as a group have not yet embraced responsibility for childcare. The vast majority of such labour, both before and after divorce, is still performed by women, whether those women work outside the home or not, and women remain the sole custodial parent in the majority of cases by mutual consent of the parties (see Evaluation of the Divorce Act -- Phase II: Monitoring and Evaluation, at p. 99; J. Drakich, "In Search of the Better Parent: The Social Construction of Ideologies of Fatherhood" (1989), 3 C.J.W.L. 69). Nor does a joint custody order in most cases result in truly shared custody. Rather, in day‑to‑day practice, joint custody tends to resemble remarkably sole custody, as in many such orders care and control remain with one person (C. J. Richardson, Court-based Divorce Mediation in Four Canadian Cities: An Overview of Research Results (1988), at p. 35; S.B. Boyd, "Women, Men and Relationships with Children: Is Equality Possible?", supra, at p. 72; M. Fineman, "Dominant Discourse, Professional Language and Legal Change in Child Custody Decisionmaking" (1988), 101 Harv. L. Rev. 727). The person who has performed the primary care giving role within the marriage as a rule retains that obligation upon separation and divorce. Thus, the lived experience of childcare for both women and children after divorce has changed much less than we might suppose or wish. As J. D. Payne and B. Edwards advise, "[i]n light of such empirical data, which encompass the Divorce Act of 1968 and the Divorce Act, 1985, popular assumptions that public attitudes towards sexual stereotyping and parenting roles after divorce have undergone radical change must be viewed with caution" ("Co‑operative Parenting After Divorce: A Canadian Perspective", supra, at p. 8. See also F. Furstenberg and A. J. Cherlin, Divided Families: What Happens to Children When Parents Part (1991), at p. 74).

Since in the present case, the respondent requested joint custody at trial but abandoned this claim on appeal and since his claim for joint guardianship also failed, it is not necessary to labour the point. One can only note that the trial judge made a finding of fact that the respondent desired custody simply in order to control the religious upbringing of the children.

Support for the decisions of the custodial parent in the discharge of his or her responsibilities remains crucial if the child is to flourish. The conferral of decision‑making authority on the custodial parent acknowledges and reflects the actual day-to-day reality of this task. It is well documented that the single parent bearing primary responsibility for the care of children often labours under increased social, economic and emotional burdens that far exceed those of the intact family. That burden is acutely felt by the children, who typically experience a myriad of hardships in this situation (see M. Grassby, "Women in Their Forties: The Extent of Their Rights to Alimentary Support" (1991), 30 R.F.L. (3d) 369). These may include a reduction in the standard of living, fewer resources for discretionary activities such as sports, artistic and leisure activities, frequent changes of residence, disruptions of friendships and often less parental attention, as the custodial parent is commonly required to devise new ways of fulfilling continuing needs with diminished resources. In the face of the enduring tasks and obligations of the custodial parent, courts must be wary of any expansion of the traditional rights of the non‑custodial parent. The clear risk is that such changes will simply reduce the decision‑making power of the custodial parent, without any parallel reduction in the responsibilities. This creates no benefit, and may amount to pure disturbance, both to the women who, as a rule, provide the primary care and to the children who must function within such a regime.

The perception that upholding the authority of the custodial parent emphasizes the rights of the parent at the expense of the interests of the child misconceives the problem. It is precisely to ensure the best interests of the child that the decision‑making power is granted to the custodial parent, as that person is uniquely situated to assess, understand, ensure and promote the needs of the child. This, of course, is also consistent with the fact that the custodial parent bears both the legal obligation to care for the child and the legal liability for the child's acts.

As a practical matter, the unending litigation occasioned by disputes over parental authority exacts a financial cost that few custodial parents can bear. It is no secret that the vast majority of such parents are women. As emphasized in Moge v. Moge, 1992 25 (SCC), [1992] 3 S.C.R. 813, women are often severely disadvantaged economically upon separation and divorce. Both women and, of necessity, the children under their care often suffer a dramatic fall in the standard of living upon divorce. The cost of ongoing litigation can only exacerbate this decline by diverting whatever scarce resources exist away from day‑to‑day needs of the remaining family.

As I stated at the outset, the custodial parent remains the decision‑maker in all aspects of the child's life. It is to avoid the spectre of the child as the field upon which the battle of competing parental rights is played out that the law confirms the authority of the custodial parent. This policy serves two functions: it precludes such contests entirely and it provides the necessary support to the parent who bears the responsibility for the child. The wisdom of this approach lies in recognizing the ease with which the interests of the child could be obscured or forgotten were courts to get into the business of parcelling out jurisdiction over the emotional, spiritual and physical welfare of the child between parents who no longer agree. This brings us to the heart of this case, a consideration of the nature of access.

Access

As indicated previously, the Act specifies the powers of courts in matters of access. Under s. 16(6), the trial judge possesses broad discretion to "impose such other terms, conditions or restrictions in connection therewith" as he or she thinks "fit and just". However, that discretion is to be exercised strictly from the perspective of the interests of the child, as s. 16(8) requires courts to "take into consideration only the best interests of the child of the marriage as determined by reference to the condition, means, needs and other circumstances of the child". Section 16(5) confers on a parent granted access, again subject to the discretion of the judge, the right to be informed about the child. Finally, s. 16(10) directs courts to give effect to the principle that "a child of the marriage should have as much contact with each spouse as is consistent with the best interests of the child".

I am in agreement with my colleague that s. 16(10) indicates that Parliament has expressed its opinion that contact with each parent is valuable. On the other hand, it must also be recognized that the goal of maximum contact is not absolute and that access may be restricted where there is evidence that such contact would otherwise conflict with the best interests of the child. This limitation on maximum contact is both abundantly clear on a straightforward reading of the section and consistent with the spirit of the Act and its focus on the best interests of the child. In my view, the analysis may and should stop at this point. However, the Court of Appeal adopts quite a different test, that of "real psychological or physical harm" as the proper criterion which should govern access. Given McLachlin J.'s assertion, at p. 127, that:

Thus, for the majority in the Court of Appeal, the prime concern was the best interests of the child.

I find it useful to contrast such analysis with what, in fact Wood J.A., for the majority of the Court of Appeal, said in the following extract:

at p. 98:

The right of the access parent to share his or her religious beliefs with the child flows from the provisions of s. 16(10) of the Divorce Act of 1985. In my view it is subject to only two limitations:

(a) the unwillingness of the child to participate in such sharing, and

(b) the powers of the court to restrict that right on the grounds that exposure to the religious beliefs or practices of the access parents is, or is likely to be, harmful to the well being of the child.

The second [limitation] also seems straightforward, but it contains a complexity that requires discussion. Obviously if it can be established, on a civil standard of proof, that exposure to conflicting religious doctrines is causing the child psychological harm, or that by engaging in the observance of, or activities associated with, the religious beliefs of the access parent, the child is suffering real psychological or physical harm, the well being of that child would clearly require the court to intervene by way of an order which would eliminate the potential for such harm to continue. In those cases where such harm could be anticipated on a balance of probabilities, there would be grounds for an order that would avoid the problems before it begins.

But, in the exercise of the court's duty to protect the well being of the child, care must be taken to ensure that real harm of the sort, and arising in the manner, just described is distinguished from the general emotional distress which every child experiences when confronted with both the reality of divorce and the turmoil which characterizes the post-divorce relationship of many ex-spouses. The former can properly be addressed by judicial intervention. The latter is inevitable, and in most cases lies beyond the influence of any order of the court.

at p. 103:

Obviously, if the exercise by each [parent] of their individual freedom of religion, caused or threatened the probability of real psychological or physical harm to their children, state sanctioned intervention to preserve those interests would be justified under s. 1 of the Charter, but such action would only be justified to the extent that its purpose was accomplished in the manner* least intrusive to the fundamental freedoms of each parent.

While I do not share the doubt which Southin J.A. has expressed concerning the existence of the parens patriae jurisdiction of the Supreme Court of this province, I do agree that any interference with the exercise of parental judgement, on what constitutes the best interests of their children, could only be justified on the basis of evidence establishing the probability of real harm, either physical or psychological. The jurisdiction to interfere, of course, exists in connection with any matter which creates, or threatens such real harm, and does not depend upon the harm originating in any disagreement between the parents. Thus any such harm occasioned by a religious doctrine or practice in which both parents joined would equally justify state sanctioned interference with their freedom of religion.

at pp. 104-5:

Of necessity, however, there are limitations to the exercise of that freedom. From what has been said so far, the obvious limitations are:

(a) the consent of the child to participate in the process of teaching or dissemination, and

(b) the power of the court to restrict the right of the access parent to teach and disseminate his or her religious beliefs to the child on the grounds that exposure to the religious beliefs or practices of the access parent is, or is likely to be, harmful to the well being of the child.

at p. 105:

. . . I do not believe that it can be invoked as a rule which would oblige the court to shrink from its duty to determine whether or not the exposure to conflicting religious beliefs, or the content of a religious doctrine, is threatening real harm to the welfare of a child of the marriage.

But it will be a rare case where such an issue could legitimately arise. As has been noted, it could arise in those few cases where it could credibly be asserted that exposure either to conflicting religious doctrines, or to the doctrines of the religious belief of one of the parents, is causing or is likely to cause the child real harm of a physical or psychological nature.

In such cases the evidence which will establish such harm will not come from a critical evaluation of the religious beliefs of either parent. That sort of evidence takes its force exclusively from the personal value judgments of either person whose evaluation is offered or the judge who hears it. Rather the threshold evidence, which will justify the court embarking on an inquiry into real physical or psychological harm caused by a religious doctrine, will be that of an independent expert whose opinion is based upon generally accepted, objective, scientific criteria for the diagnosis and evaluation of the harm alleged.

at pp. 107-8:

It reinforces my view that the following principles of law are applicable to the issue of access restrictions which this case presents:

(g) Mr. Young's fundamental freedom of religion under s. 2(a) of the Charter, to teach and disseminate his religious beliefs to his children, is limited only to the extent that the exercise of that freedom causes, or threatens the probability of, real physical or psychological harm to the children.

at p. 108:

In my view both [Huddart J. and Proudfoot J.] correctly perceived that no issue of real harm, either physical or psychological, arose in this case. If any such issue had presented itself, it would have been manifest from the evidence of the expert witnesses who examined the children on a number of occasions, and who undoubtedly would have identified the existence of, or the potential for, real harm to the children arising from their father's religious beliefs, if such existed. In the absence of any such evidence, there was no basis upon which the court could legitimately entertain any inquiry into the substance of his religious beliefs.

At no time did the learned trial judge refuse to admit any evidence relevant to the issue of the children's best interests. Her decision not to admit any evidence related to the tenets of the Jehovah's Witness faith was the natural and irresistible result of the fact that the dispute between these two parents had its genesis in the intolerance of each for the religious beliefs of the other. Theirs was a dispute between beliefs, not a dispute over whether real harm would befall their children as a result of the religious beliefs of one or the other.

and, finally, at p. 109:

I have reviewed the evidence which was offered in support of those restrictions. In my view it fell far short of establishing that any real harm would befall the children from Mr. Young's exercise of his fundamental freedom to teach and disseminate his religious beliefs to them. No such harm was diagnosed or anticipated by either of the expert witnesses which the learned trial judge found to be credible. Nor did the evidence establish that the conflict in the religious beliefs of the parents was causing, or was likely to cause, such harm.

The evidence went no further than to establish that the children were under some stress, the cause of which was "multi-factorial", but which obviously included some component related to the religious dispute between the parents. [Emphasis (by underlining) added.]

This, in my view, leaves absolutely no doubt that the only basis for the Court of Appeal's reversal of the orders of the trial judge, which the court could not have said more clearly, was the absence of harm, which, in their view, was not established in evidence. We are far from the best interests of children test. Affirming the Court of Appeal's judgment on this point, my colleague, while repeating profusely that the best interests of children is the sole test, is in fact applying the harm test, which is the only basis on which the Court of Appeal reversed the trial judge who, in their opinion, did not apply that test.

Neither logic, necessity nor the terms of the Act mandate such an equation. As I will discuss later, the best interests of the child encompasses more than the absence of harm and I cannot agree that this objective compels the conclusion that there must be evidence of "real and significant harm to the child" before contact with the access parent can be limited in any manner.

In contrast to custody, there is little jurisprudence or doctrine on the nature of access rights, probably because access has never been understood to confer entitlements on the non‑custodial parent other than those which are specified in court orders. The fundamental principle, however, that has always been recognized by courts is that the right to access is limited in scope and is conditioned and governed by the best interests of the child. Anson v. Anson, supra, at p. 368, in my view, properly circumscribes the right of access at this stage in time as follows:

Access can thus be regarded as a form of temporary possession with the powers granted by an access order being limited to those necessary to ensure the well‑being of the child during the visitation periods. Access confers no right in the parent to influence the upbringing of the child ‑‑ that is for the parent with custody or guardianship. . . . However, under the Divorce Act, 1985 s. 16(5), a parent granted access has a right to make inquiries and to be given information as to the health, education and welfare of the child, unless otherwise ordered by a court.

Thus, the role of the access parent is "that of a very interested observer, giving love and support to [the child] in the background" (Pierce v. Pierce, 1977 2206 (BC SC), [1977] 5 W.W.R. 572 (B.C.S.C. in chambers), at p. 575. See also Gubody v. Gubody, 1955 405 (ON SC), [1955] O.W.N. 548 (H.C.), and Sudeyko v. Sudeyko (1974), 1974 2223 (BC SC), 18 R.F.L. 273 (B.C.S.C.)).

In the words of Wilson J., "there is considerable support for the view that access as a distinct juridical concept is purely a creature of statute" (Frame v. Smith, 1987 74 (SCC), [1987] 2 S.C.R. 99, at p. 130). Apart from legislative provisions, there are no pre‑existing or parallel rights of access at common law.

The Civil Code of Quebec for its part recognizes no right of access but states simply:

570 Whether custody is entrusted to one of the spouses or to a third person, the father and mother retain the right of watching over the maintenance and education of the children, and are obliged to contribute thereto in proportion to their means.

However, art. 30 of the Civil Code of Lower Canada provides that "[i]n every decision concerning a child, the child's interest and the respect of his rights must be the determining factors". Article 543 of the Civil Code of Quebec obliges courts "to see to the interests of the child, at all stages of divorce proceedings". Article 569 directs courts to decide matters of custody, maintenance and education of the children, in their interest and in the respect of their rights. Thus, in both civil and common law jurisdictions, access exists only to the extent it is provided by statute and where it is in the best interests of the child.

One can find in both case law and doctrine discussion and debate as to whether access is more properly understood as a right of the parent or of the child. (See Weitzman, supra, at p. 230; R. Deech, "The Unmarried Father and Human Rights" (1992), 4 J. Child L. 3; and Maidment, supra, at p. 40. See also Frame v. Smith, supra). Some have regarded access as an entitlement which belongs to the parent absent any compelling reason to disallow it, such as the safety of the child (Tocco v. Tocco (1977), 1977 2628 (ON CA), 4 R.F.L. (2d) 174 (Ont. C.A.)). However, as is consistent with the focus on the best interests of the child in the Act, the more perceptive view at this time both in the literature and case law is that access is more properly regarded as a right of the child, not a right of the parent. As Bernard Dickens has stated, the modern function of parental rights is not to enforce powers of custody and control parents enjoy over their children, but to permit parents to discharge their duties to their children (supra, at p. 471). This view accords with the decision of this Court in Racine v. Woods, 1983 27 (SCC), [1983] 2 S.C.R. 173, in which Wilson J. summarized the proper focus in the following words, at p. 185:

... it is the parental tie as a meaningful and positive force in the life of the child and not in the life of the parent that the court has to be concerned about. As has been emphasized many times in custody cases, a child is not a chattel in which its parents have a proprietary interest; it is a human being to whom they owe serious obligations.

Some years later, in Frame v. Smith, supra, at p. 132, Wilson J., although dissenting on another point, clearly said that:

The access right has become the child's right, not the parent's right, and it would be a regressive step to recognize today a cause of action in the parent based on an outmoded concept of parental rights in children....

Access rights exist in recognition of the fact that it is normally in the interests of the child to continue and foster the relationship developed with both parents prior to the divorce or separation. This being said, the right to access and the circumstances in which it takes place must be perceived from the vantage point of the child. Wherever the relationship to the non‑custodial parent conflicts with the best interests of the child, the furtherance and protection of the child's best interests must take priority over the desires and interests of the parent.

Wood J.A. expressed the general concern that access rights remain vulnerable to the caprices of a vengeful custodial parent. In my view, courts should not be too quick to presume that the access concerns of the custodial parent are unrelated to the best interests of the child. Although the myth is that the custodial parent typically attempts to obstruct access of the other parent, studies in Canada, England and the United States indicate that the problem tends to be quite the reverse (see Richardson, supra; Weitzman, supra, at p. 230; and Maidment, supra, at p. 68). At the outset, it must not be forgotten that a parent who is granted access has no obligation to exercise those rights, and neither the child nor the custodial parent can force an access parent to comply with such an order, even though access might have been determined to be in the best interests of the child.

In Canada, the Evaluation of the Divorce Act -- Phase II: Monitoring and Evaluation, supra, at p. 111, discloses that access is denied to non‑custodial parents by courts in only 2.4 percent of cases. However, those access rights which are granted by courts are often not exercised, and over the long run there is a process of gradual disengagement of many non‑custodial parents from active involvement with their children, particularly if they remarry or become part of new family constellations (see N. Weisman, "On Access After Parental Separation" (1992), 36 R.F.L. (3d) 35). The authors of the Evaluation report at pp. 114‑15 that over 40 percent of parents granted access either never see their children or see them no more than once a month. Moreover, those parents who exercise access do so less frequently than in the previous study (Evaluation of the Divorce Act, 1985 -- Phase I: Collection of Baseline Data (1987)), and the level and involvement of access parents is low and appears to be worsening over time. By contrast, many custodial parents express concern about the effects on their children of fathers' absence and the most commonly voiced complaint concerning access is that such rights are not exercised, or are exercised irregularly (Richardson, supra, at p. 36). Thus, as the authors of the Evaluation conclude, any lack of contact between non‑custodial parents and their children can rarely be attributed to a denial of access (pp. 134‑35). (Similar results were recently reported in D. Perry, Access to Children Following Parental Relationship Breakdown in Alberta (1992).)

Furthermore, when considering the role of access parents with respect to their children, courts should not be blind to the broader context in which custody and access are exercised. A vast number of non‑custodial parents are in default of their most serious obligations to their children, as is the respondent in this appeal: the responsibility to provide economic support. In 1985, E. Finnbogason and M. Townson reported in The Benefits and Cost- Effectiveness of a Central Registry of Maintenance and Custody Orders (1985), that all existing studies indicated default rates in child support payments of over 50 percent, and the Law Reform Commission estimated the true rate to be as high as 75 percent. (See also A. M. Delorey, "Joint Legal Custody: A Reversion to Patriarchal Power" (1989) 3 C.J.W.L. 33, at p. 43, and Drakich, "In Whose Best Interests? The Politics of Joint Custody", supra). The most pressing issue in child custody and access is the burden borne by women and children as a consequence of the failure of men to accept their responsibilities to their children. As Ruth Deech noted, supra, at p. 6, "[t]he reality is the call on fathers to pay more, not to enjoy more rights." There is a certain irony in the claim to greater contact and control on the part of access parents in the face of such widespread neglect of children's basic needs.

Where there is a genuine problem with access, the non‑custodial parent is not without recourse in any case. This stems from the statutory directive to facilitate access where it is in the child's best interests and the role of the judge as the arbiter of those interests in the case of a dispute between the parents. Generally, courts will grant liberal access to the non‑custodial parent and usually this is consistent with the best interests of the child. Parents will also normally respect their children's wishes and best interests with regard to access. Most of the time, parents are able to reach an agreement as to custody and access without the intervention of courts, sometimes with the help of psycho‑social services attached to courts or elsewhere. This, of course, is to be encouraged as it is in the best interests of children to minimize conflict in this area. Unfortunately, however, disagreements between parents do reach the courts and in such cases the judge must draw the line. That line must always be drawn in favour of the best interests of the child, from a child‑centred perspective.

The difficulty, of course, always lies in determining what does constitute the best interests of the child, which brings us to the next issue.

Best Interests of the Child

The issue of the best interests of the child raises considerations at a number of different levels. First, the content of the rule must be discussed. Second, its constitutionality must be determined. Finally, the appropriate standard for its application must be established. All of these issues form the basis of the respondent's arguments in support of the Court of Appeal's ruling as to access.

Content of the Rule

As I discussed in the context of custody, the statutory recognition of the best interests of the child in the Act marks the culmination of a long evolution of the focus in custody and access decisions. From the era in which the enforcement of paternal rights was at the forefront of the courts' concern, through a time where the merits of each parent's claims were weighed, we have now reached the point where the sole question to be determined in custody and access arrangements is the best interests of the child. Indeed, the legislation makes it quite explicit that only the best interests of the child as they are comprehensively understood should be considered in such adjudications. This evolution was recognized by Wilson J. in Frame v. Smith, supra, at p. 132:

At first the courts were much more comfortable assessing the competing claims of parents than they were in trying to decide what was in the best interests of children. But over time the best interest of children increasingly became an important concern of the court and today it is the paramount concern. ... In light of these developments it can be said with some assurance that the concept of "parental rights" has fallen into disfavour. Parental responsibilities yes, but rights no.

The courts have moved inevitably and steadfastly in this direction since the 19th century. In In re McGrath (Infants), [1893] 1 Ch. 143, at p. 148, the English Court of Appeal determined that a court should have unfettered scope to consider the best interests of the child, even where it interfered with the father's rights with respect to religion, holding:

The dominant matter for the consideration of the Court is the welfare of the child. But the welfare of a child is not to be measured by money only, nor by physical comfort only. The word welfare must be taken in its widest sense. The moral and religious welfare of the child must be considered as well as its physical well‑being.

This Court has followed suit and unremittingly endorsed the position that in the face of any threat to the welfare of the child, parental interests must give way. As long ago as DeLaurier v. Jackson, supra, at pp. 153-54, Crocket J. held:

If the general welfare of the child requires that the father's rights in respect of the religious faith in which his offspring is to be reared, should be suspended or superseded, the courts in the exercise of their equitable jurisdiction have undoubted power to override them, as they have the power to override all other parental rights, though in doing so they must act cautiously. . . .

Due consideration is, of course, to be given in all cases to the father's wishes but if the court is satisfied in any case upon a consideration of all the facts and circumstances, as shown by the evidence, that the father's wishes conflict with the child's own best interests, viewed from all angles ‑‑ material, physical, moral, emotional and intellectual as well as religious ‑‑ then the father's wishes must yield to the welfare of the child.

The Court reached the same conclusion in King v. Low, 1985 59 (SCC), [1985] 1 S.C.R. 87, finding that such a result was consistent with both modern authority in other jurisdictions and previous decisions of this Court in Racine v. Woods, supra, and Beson v. Director of Child Welfare (Nfld.), 1982 32 (SCC), [1982] 2 S.C.R. 716. The nature of the judicial determination of the best interests was described by McIntyre J. in King v. Low at p. 101 as follows:

. . . the dominant consideration to which all other considerations must remain subordinate must be the welfare of the child. This is not to say that the question of custody will be determined by weighing the economic circumstances of the contending parties. The matter will not be determined solely on the basis of the physical comfort and material advantages that may be available in the home of one contender or the other. The welfare of the child must be decided on a consideration of these and all other relevant factors, including the general psychological, spiritual and emotional welfare of the child. It must be the aim of the Court, when resolving disputes between rival claimants for the custody of a child, to choose the course which will best provide for the healthy growth, development and education of the child so that he will be equipped to face the problems of life as a mature adult. Parental claims must not be lightly set aside, and they are entitled to serious consideration in reaching any conclusion. Where it is clear that the welfare of the child requires it, however, they must be set aside. [Emphasis added].

These words express the important notion that courts must be directed to create or support the conditions which are most conducive to the flourishing of the child. They are, in my view, as applicable in the context of access arrangements as they are in the determination of custody.

A determination of the best interests of the child encompasses a myriad of considerations, as child custody and access decisions have been described as "ones of human relations in their most intense and complex form". In contrast to most issues that come before the courts, such decisions are "person‑oriented" rather than "act-oriented" and require an evaluation of "the whole person viewed as a social being" (L. LaFave, "Origins and Evolution of the `Best Interests of the Child' Standard" (1989), 34 S.D.L. Rev. 459; R. H. Mnookin, "Child‑Custody Adjudication: Judicial Functions in the Face of Indeterminacy" (1975), 39 Law & Contemp. Probs. 226). Courts are required to predict the happening of future events rather than to assess the legal import of past acts and judge the effect of various relationships on the best interests of the child, all the while weighing innumerable variables without the benefit of a simple formula.

In making a determination as to the best interests of the child, courts must attempt to balance such considerations as the age, physical and emotional constitution and psychology of both the child and his or her parents and the particular milieu in which the child will live. These considerations, which underlie the common law, are clearly set out in art. 30 of the Civil Code of Lower Canada:

30 In every decision concerning a child, the child's interest and the respect of his rights must be the determining factors.

Consideration may be given in particular to the child's age, sex, religion, language, character and family surroundings, and the other circumstances in which he lives. [Emphasis added.]

Probably one of the most significant factors in many cases will be the relationship that the child entertains with his or her parents. This must necessarily encompass such considerations as the strength of the emotional ties and the role of the person who has provided primary care in the life of the child.

Goldstein, Freud and Solnit's Beyond the Best Interests of the Child, supra, while perhaps lacking in empirical data, remains an influential analysis of the psychological needs of children following divorce. The authors emphasize, among other factors, the importance of continuity in the child's relationships and conclude that the major focus of custody decisions should be to preserve and protect the relationship between the child and his or her psychological parent. As Weisman reports, supra, at p. 47, there is some consensus in favour of this view in recent research. Goldstein later went as far as to say that, while continuing access is normally in the best interests of the child, this is the case only where there is agreement between parents and no court order to effect access is required (Goldstein, "In Whose Best Interests?", in R.S. Abella and C. L'Heureux‑Dubé, eds., supra). While the Act places the emphasis on the best interests of the child rather than the wishes of the custodial parent, Weisman observes at p. 62, supra, that research indicates that "the child's relationship with the custodial parent may well be the most important factor affecting long‑term outcome. Judges should therefore exercise caution before intervening in any manner that may introduce added stress to this relationship".

Custody and access decisions are pre‑eminently exercises in discretion. Case by case consideration of the unique circumstances of each child is the hallmark of this process. This Court recognized in Moge v. Moge, supra, in the context of spousal support decisions, that the discretion vested in the trial judge is essential to effect the very purposes outlined in the Act. The wide latitude under the best interests test permits courts to respond to the spectrum of factors which can both positively and negatively affect a child. Such discretion also permits the judge to focus on the needs of the particular child before him or her, recognizing that what may constitute stressful or damaging circumstances for one child may not necessarily have the same effect on another.

While the best interests test provides the focus and perspective from which to assess custody and access decisions, the test has nonetheless been subject to evaluation and criticism from various sources. The most common concern is that it is essentially indeterminate and fails to provide the necessary direction and criteria with which to make custody and access decisions (Mnookin, supra). Some commentators have observed that the broad scope of the test may permit the best interests of the child to become hostage to parental rights rhetoric (K. M. Munro, "The Inapplicability of Rights Analysis in Post‑Divorce Child Custody Decision Making" (1992), 30 Alta. L. Rev. 852). Others note that, devoid of presumptions, the test becomes a competition between psychological experts and lawyers concerning the prevailing psychological notions of child rearing (Fineman, The Illusion of Equality: The Rhetoric and Reality of Divorce Reform, supra). Women in particular have become concerned that the importance of women's work in caring for young children can be devalued under the best interests test, as judges consider such factors as the superior financial position of most fathers when deciding custody matters (S. B. Boyd, "Potentialities and Perils of the Primary Caregiver Presumption" (1991), 7 C.F.L.Q. 1). Others claim that the test permits cultural bias in custody and access decisions and is a recipe for the abuse of judicial discretion (S. J. Toope, "Riding the Fences: Courts, Charter Rights and Family Law" (1991), 9 Can. J. Fam. L. 55, at p. 67). See also A. Ehrcke, "Limiting Judicial Discretion in Custody Proceedings on Divorce" (1987), 6 Can. J. Fam. L. 211; J. P. Ryan, "Joint Custody in Canada: Time for a Second Look" (1986), 49 R.F.L. (2d) 119; Department of Justice of Canada, Custody and Access: Public Discussion Paper (1993).

In my view, without minimizing these concerns and risks, it is crucial to distinguish between the criticism of the best interests test and the best interests of the child as an objective. Despite concerns about the application of the test, no one has questioned the best interests of the child as the proper focus of custody and access determinations. In the words of Maidment, supra, at p. 161:

. . . the welfare of the child holds the central position in the resolution of custody cases. Whatever the practice, the law places the child protection or nurturance philosophy at the centre of the decision‑making process, and whatever interpretations are given to the welfare principle, nowadays the pre‑eminence of the child‑centred approach receives universal support.

Furthermore, it is significant that no other test has been broadly endorsed to replace it. At the end of the day, the test has been described as "the least detrimental available alternative for safeguarding the child's growth and development" and most of the suggested guidelines or presumptions are in fact designed to give content to the test in furtherance of the welfare of the child. (See Goldstein, Freud and Solnit, supra, at p. 53. See also Mnookin, supra.)

It is also worthwhile to recall that the best interests test initially fulfilled its purpose because the presumptions such as the tender years doctrine provided the necessary content on which to base custody and access decisions (Fineman, The Illusion of Equality: The Rhetoric and Reality of Divorce Reform, supra, at p. 84). Courts not only presumed that mothers were naturally better custodial parents, but also relied on the parent who had since the beginning provided primary care to the child to continue to act in its best interests. As this Court noted in Talsky v. Talsky, supra, at p. 293, the tender years doctrine was not so much a rule of law as "purely . . . a principle of common sense".

With the removal of such presumptions however, courts are now faced with competing claims concerning the best interests of the child. In the result, a number of presumptions have been suggested to govern the best interests test in custody decisions, the most common of which is the primary caregiver presumption. (See Boyd, "Potentialities and Perils of the Primary Caregiver Presumption", supra). This presumption has developed in response to the trend toward joint custody, and the resultant erosion in both the financial resources and the authority of the parent with day‑to‑day obligations for child rearing. Boyd expresses the concerns underlying this presumption as follows at p. 6:

. . . a misplaced application of "equality" is evident in legislative and judicial trends toward entrenching fathers' rights in child custody law, such as joint custody and access enforcement. The importance of ensuring "equal" fatherly input into children's lives after family breakdown even if they were not "equally" involved before, has been elevated to arguably too high a level in recent years. The form which this trend often takes -‑ retaining an active and full role for both father and mother in making decisions concerning the child (joint legal custody) -‑ potentially inhibits the autonomy and exercise of discretion by the parent with physical care of the child.

However, as she points out, one of the principal rationales for endorsing this presumption is not to supplant the best interests of the child as the ultimate objective but to ensure that those interests are protected. The presumption explicitly restores the values of commitment and demonstrated ability to nurture as well as continued psychological parenting to the child, factors which may be overshadowed by other considerations in custody disputes. The primary caregiver presumption thus recognizes the obligations and supports the authority of the parent engaged in day-to-day tasks of child rearing.

In assessing all the relevant considerations, courts must be careful that the ideals of parental sharing and equality do not overcome the lived reality of custody and access arrangements and that the child's needs and concerns are accommodated and not obscured by abstract claims of parental rights. This is not to say that the parent's interests may not coincide with the child's interests or that a court may never validly take a parent's interests into consideration. However, to further the best interests of the child, a recognition of the close relationship between the needs of the child and the needs of the remaining family unit of which he or she is a part is essential. As Abella J.A. held in M. (B.P.) v. M. (B.L.D.E.), (1992), 1992 8642 (ON CA), 97 D.L.R. (4th) 437, for the majority, at p. 459:

[The best interests of the child] by no means excludes the parental perspective. The needs of children and their parents are obviously inextricable, particularly between children and the parent on whom they depend for their day-to-day care, where only one parent has this primary responsibility. The structure of an environment that fits the child's interests would undoubtedly be reinforced if the economic and emotional needs, especially of custodial parents, were factored in, given the symbiosis of their sense of well‑being.

Moreover, as Weisman, supra, concludes at p. 36:

. . . access issues often present the most intractable problems in all family law. On occasion, fairness and enforceability are impossible to achieve, and the most for which one can hope is the least detrimental alternative for the child.

In the final analysis, when considering the best interests of the child, it may be a virtue rather than a sin for the judge to be pragmatic and take into consideration both the strengths and shortcomings of the parties and the limits of human nature. To do so is only to recognize that, in some circumstances, the perfect may be the enemy of the good. Courts must always be mindful that not only the benefit but also the real cost and burden of all custody and access arrangements ultimately falls on the children themselves.

Constitutionality of the Best Interests Test

It would seem to be self‑evident that the best interests test is value neutral, and cannot be seen on its face to violate any right protected by the Charter. Indeed, as an objective, the legislative focus on the best interests of the child is completely consonant with the articulated values and underlying concerns of the Charter, as it aims to protect a vulnerable segment of society by ensuring that the interests and needs of the child take precedence over any competing considerations in custody and access decisions.

Nonetheless, in his claim that the access order infringes his freedom of religion under the Charter, the respondent challenges the constitutionality of the best interests test on a number of grounds. First, he alleges that the discretion conferred on judges under the best interests test violates the Charter. Second, he claims that the test is so vague that it cannot be said to be "prescribed by law" and, accordingly, any violation of his freedom of religion, cannot be justified by the test under s. 1 of the Charter. Leaving aside for the moment the alleged infringement of freedom of religion of the respondent and his children under the Charter, in my opinion, both of these contentions must fail.

As I observed in R. v. Swain, 1991 104 (SCC), [1991] 1 S.C.R. 933, this Court has previously found that the mere existence of broad judicial discretion in a legislative provision is not in itself sufficient to attract the application of the Charter, nor can the presence or scope of such discretion alone give rise to an inference that Charter rights are thereby infringed. One such instance was the decision in R. v. Beare, 1988 126 (SCC), [1988] 2 S.C.R. 387, where the issue arose in the context of a challenge to the wide discretion of the police to fingerprint suspects. In that case, La Forest J., writing for the Court, stated at p. 410:

The existence of the discretion conferred by the statutory provisions does not, in my view, offend principles of fundamental justice.

Indeed, this Court recently held in Baron v. Canada, 1993 154 (SCC), [1993] 1 S.C.R. 416, that the presence of the residual discretion of a judge may itself be a constitutional requirement in some instances, in order that there be a means of balancing the competing rights and interests of the individual and the state.

In R. v. Morales, 1992 53 (SCC), [1992] 3 S.C.R. 711, Gonthier J., dissenting, observed at p. 754 that:

. . . the identification of a measure of discretion conferred by means of a legislative provision cannot alone provide the basis for a constitutional evaluation of that provision. Nor can the identification of possible parameters of that discretion, for a discretion which is referred to as being fettered can be one which is limited not only by appropriate constraints but also by those which are inappropriate or unsuitable. The more important issue which remains, therefore, is what kind of discretion is conferred, and the capacity of the words of the legislative provision to support the type of reasoning which the matter under adjudication requires. [Emphasis added.]

Thus, discretion cannot be considered in the absence of an examination of the legislative objectives, and the important question is whether the presence of such discretion can be rationally tied to those objectives. For the reasons which I shall delineate hereinafter when considering the vagueness of the best interests test, it is clear that broad judicial discretion is crucial to the proper implementation of the legislative objective of securing the best interests of the child.

This Court has had occasion to consider the concept of vagueness of legislative provisions in the context of the Charter on a number of occasions in recent years (see R. v. Morgentaler, 1988 90 (SCC), [1988] 1 S.C.R. 30; Irwin Toy Ltd. v. Quebec (Attorney General), 1989 87 (SCC), [1989] 1 S.C.R. 927; Reference re ss. 193 and 195.1(1)(c) of the Criminal Code (Man.), 1990 105 (SCC), [1990] 1 S.C.R. 1123; R. v. Keegstra, 1990 24 (SCC), [1990] 3 S.C.R. 697; Canada (Human Rights Commission) v. Taylor, 1990 26 (SCC), [1990] 3 S.C.R. 892; Committee for the Commonwealth of Canada v. Canada, [1991] 1 S.C.R. 139; Osborne v. Canada (Treasury Board), 1991 60 (SCC), [1991] 2 S.C.R. 69; R. v. Butler, 1992 124 (SCC), [1992] 1 S.C.R. 452; R. v. Nova Scotia Pharmaceutical Society, 1992 72 (SCC), [1992] 2 S.C.R. 606). While these cases all dealt with vagueness in the criminal context, a number of principles may be derived from this jurisprudence which are useful in the present context. The underlying concern is that a legislative provision be capable of providing a framework or guide within which judicial decisions can be made. The threshold for constitutional violation due to vagueness in legislative provisions is relatively high, and it is not necessary that such provisions carry a precise technical meaning or provide certainty as to the result (R. v. Butler, supra; Osborne v. Canada (Treasury Board), supra, and R. v. Nova Scotia Pharmaceutical Society, supra). Rather, the standard requires that the provisions permit the framing of an intelligible legal debate with respect to the objectives contained in the legislation.

The doctrine of vagueness was defined most recently by Gonthier J. for this Court in R. v. Nova Scotia Pharmaceutical Society, supra, at p. 643 as follows:

. . . a law will be found unconstitutionally vague if it so lacks in precision as not to give sufficient guidance for legal debate.

The question which arises, then, under the vagueness requirements of constitutional validity, is whether the best interests test is so uncertain as to be incapable of guiding a consideration of the factors relevant to custody and access determinations. In my opinion, it is not.

A similar question arose concerning the concept of the "public interest" in the recent decision of this Court in R. v. Morales, supra, where Gonthier J., while dissenting on this issue, set out a framework for considering the constitutional validity of broad legislative concepts which are nonetheless designed to encompass a specific set of concerns and interests. He described "public interest" in the following manner at p. 751:

. . . a notion which has traditionally been recognized as affording a means of referring to the special set of considerations which are relevant to those legal determinations concerned with the relationship of the represented private interest or interests and the broader interest of the public.

The same type of characterization is appropriate to the best interests test. The "best interests of the child" can be regarded as the term employed to refer to the spectrum of considerations encompassed by the needs of the child, as distinct from those of any other party, in the determination of custody and access disputes. The fact that it must be applied to the facts of each case does not militate in favour of its unconstitutionality. Rather, this feature is part and parcel of what makes decisions in the best interests of the child possible at all.

As its widespread adoption and application suggest, the best interests of the child standard cannot be regarded as unconstitutionally vague. On the contrary, the best interests test is universally recognized as the foundation of modern family law around the world and is legislatively entrenched in both common law and civil jurisdictions in the United States, Australia and Europe. Moreover, the need to make the best interests of the child the primary consideration in all actions concerning children, including legal proceedings, is specifically recognized in international human rights documents such as the United Nations Convention on the Rights of the Child, Can. T.S. 1992 No. 3, Art. 3(1). In my view, this amply demonstrates both the enduring value of the best interests test as a legal norm capable of meaningful application and the broad recognition of the interests of children in the field of human rights.

It is important to recall that the vagueness of a legislative provision cannot be examined in the abstract but must be considered within the context of the particular legislative objectives in question, bearing in mind that some objectives will require a panoply of judicial remedies for their meaningful fulfilment. As Beetz J. observed in R. v. Morgentaler, supra, at p. 107, "[f]lexibility and vagueness are not synonymous". Nor should the vagueness of a provision be considered solely with reference to the words of the statute. The Ontario Court of Appeal held in R. v. LeBeau (1988), 1988 3271 (ON CA), 41 C.C.C. (3d) 163, at p. 173:

It has been recognized that the void for vagueness doctrine is not to be applied to the bare words of the statutory provision but, rather, to the provision as interpreted and applied in judicial decisions.

See also Reference re ss. 193 and 195.1(1)(c) of the Criminal Code (Man.), supra, at p. 1157.

This Court held recently in R. v. Nova Scotia Pharmaceutical Society, supra, at p. 627 that, among the factors with which courts should be concerned when the vagueness of a law is at issue are:

. . . (a) the need for flexibility and the interpretive role of the courts, (b) the impossibility of achieving absolute certainty, a standard of intelligibility being more appropriate and (c) the possibility that many varying judicial interpretations of a given disposition may exist and perhaps coexist. . . .

Child custody and access decisions engage all of these considerations, as the best interests of the child remain, in the words of Abella J.A., "as fluid as each child's circumstances" (M. (B.P.) v. M. (B.L.D.E.), supra, at p. 459). As I have discussed earlier in these reasons, when fashioning custody and access orders, courts are concerned with the design of future living arrangements which may be affected by a multitude of different factors. In establishing this modus vivendi, courts must consider not only the circumstances which might ideally be in the best interests of the child, but must also weigh the capacity of the parties to fulfil and endure the conditions which are set for custody and access. Thus, family law statutes, and child custody provisions in particular, probably best exemplify the type of legislation in which broad judicial discretion is necessary to fulfil the legislative purpose effectively.

Furthermore, the best interests test has been judicially applied for decades, and the case law provides a large body of jurisprudence to which courts may resort when considering factors relevant to the best interests test.

In my opinion, the prophetic caution of Gonthier J. in R. v. Nova Scotia Pharmaceutical Society, supra, at p. 642, is particularly apposite in the circumstances of this case:

One must be wary of using the doctrine of vagueness to prevent or impede State action in furtherance of valid social objectives, by requiring the law to achieve a degree of precision to which the subject‑matter does not lend itself.

It should not be assumed that a grant of discretion is an invitation to exercise personal prejudice, as it is well established in the jurisprudence that discretion in every instance must be exercised judicially and in conformity with the objectives and standards of the legislation. The application of the best interests test, if done in an individual case according to irrelevant or improper criteria, remains subject to the normal process of review on appeal.

Having analyzed the content of the best interests of the child standard prescribed in the Act in matters of custody and access and determined its constitutionality, what burden of proof must the custodial parent meet to obtain an order restricting access?

Evidence

One might have thought that the clear and unequivocal language of the Act makes the best interests of the child the only consideration in matters of custody and access, and that the only evidence required for an order limiting access would be evidence indicating that the exercise of access in a particular manner is in the best interests of the child. This is the test which was applied in this case by all of the judges, including those at the Court of Appeal, who were seized of the matter at the interim stage, as well as the trial judge. The Court of Appeal, however, relying in particular on the decision of the Ontario Supreme Court in Hockey v. Hockey, supra, would elevate the test to one of harm which the custodial parent must show in order to restrict access to the children by the non‑custodial parent. I say at the outset that, even under such a severe test, the appellant has established that restrictions are required. However, the question is whether harm is the test. My answer is a clear no for the reasons that follow.

As I noted above, the Court of Appeal has equated the alleged absence of harm in this case with the best interests of the child. However, no rationale is proposed for so defining the best interests of the child. Moreover, nothing in the Act mandates or even suggests that "real danger of significant harm to the child" be the sole consideration in matters of custody and access.

Subsection 16(8) of the Act directs that the best interests of the child are to be determined "by reference to the condition, means, needs and other circumstances of the child". There is no mention of harm at all in this list of considerations, and nothing supports the inference that the best interests of the child not only can but also should be encompassed by the simple concept of absence of harm, thus making harm the controlling factor in custody and access decisions. On the contrary, the adoption of the harm test would require courts to ignore the very factors which are set out in the Act and which have guided the courts up until now in custody and access decisions. While it is self‑evident that if the exercise of access by the non‑custodial parent harms the child, access must be curtailed; even in the absence of harm the best interests of the child may also warrant such curtailment.

The harm test by which the Court of Appeal proposes to qualify the best interests test inverts the basic focus of the inquiry into custody and access. Under the best interests test, courts must consider how to best foster the child's overall development and protect the child from conflict and the disruptive effects of divorce at a vulnerable point in his or her life. In contrast, the harm test essentially requires a court to determine how much conflict and stress a child should be required to endure in order that the parent's wishes may prevail. Once the pendulum swings in that direction, it is difficult to control. If not this much stress, why not a little more? If one type of conflict, why not that? Moreover, it may be difficult if not impossible, even for an expert, to determine with any degree of accuracy whether the child has been harmed or will be harmed in the long run. However, the most serious deficiency of the harm test is the following. While the effects of custody and access decisions always remain uncertain to some degree, the harm test places any risk of miscalculation in the degree of stress or conflict occasioned by such decisions squarely on the back of the child, depriving the child of any presumption in his or her favour. Obviously this cannot be correct from the perspective of the interests of the child. To wait until harm has occurred to correct the situation is not only to waive the benefit of prevention, but also to increase the possibility of error. Instead of minimizing the risks, the harm test would maximize them.

A discussion of the research data on the long-term effects of divorce, as well as custody and access decisions, on children is helpful at this point to assess the consequences of adopting the harm standard as regards the welfare of children and the importance of adhering to the best interests of the child as the overarching goal in custody and access determinations. While this remains an area of social science that has yet to be comprehensively researched, studies of both the effects of divorce and the role of conflict in the subsequent family life indicate that children often suffer more extensively than is generally acknowledged on divorce, and that those who must endure continuing conflict after divorce stand at serious risk of harm down the road. The resounding message is that courts must pay more, not less, attention to the needs of children on divorce.

I acknowledge at the outset the limits in applying such research to the wider population, as the studies to date have tended to focus on groups within a particular social class or locale. Furthermore, the conclusions of some studies, for example those by J. S. Wallerstein and J. B. Kelly, Surviving the Breakup: How Children and Parents Cope with Divorce (1980), have been used to support a number of different propositions concerning the best interests of the child on divorce. (See Fineman, The Illusion of Equality: The Rhetoric and Reality of Divorce Reform, supra, at p. 118.) In addition, notions concerning the optimum child rearing conditions on the breakdown of the marriage are subject to a degree of difference and controversy between professionals working in the field. This is illustrated, for example, in the contrasting conclusions reached by Goldstein, Freud and Solnit, supra, and those of Wallerstein and Kelly, supra.

Nonetheless, a number of conclusions about the effects of divorce on children emerge with remarkable consistency in all of the major studies and psychological literature on children after divorce. One of the most important of these is the role of conflict in the welfare of the child. Along with the quality of the relationship with the custodial parent and the ability to maintain contact with the non‑custodial parent, there is substantial evidence that continuing conflict is the most important factor affecting the ability of children to readjust to the new family situation after divorce (Weisman, supra, at pp. 47‑48). It appears that, above and beyond the disruption caused by divorce or separation itself, it is the discord and disharmony within the family which are most damaging to children in the aftermath of divorce.

Two of the major studies on divorce and the effects of conflict on the emotional and psychological well‑being of children forcefully make the point that the ability to reduce conflict is crucial to the welfare of children. The study of Wallerstein and Kelly, supra, although limited to white middle class families in California without previous clinical histories who volunteered to participate in the study, is the most comprehensive long‑term study on the effects of divorce on children. A ten‑year follow up of the same subjects is reported in J. S. Wallerstein and S. Blakeslee, Second Chances: Men, Women and Children a Decade After Divorce (1989). Among the most significant findings of these two reports is that separation or divorce cannot be regarded as a discrete event to be dealt with once and for all, but is most often the beginning in a continuum of disruptive events in the life of the child. The stress resulting from changing family structures and reduced financial support following divorce in many cases continues to be experienced by children long after the divorce is final. While many children do adjust in such situations, the notion that children automatically can and do "get used to" new family situations in a relatively short period of time is not borne out by the results of the study. By all such indicators as success at educational endeavours and the later ability to establish stable personal and professional lives, children of divorce appear in general to be subject to more stresses than their counterparts in intact families. While the authors found considerable evidence that divorce was beneficial to the parents, there was no comparable evidence regarding children. The majority of such children neither experienced relief at the time of the divorce nor felt it had resulted in an improvement in their lives five years later. Moreover, Wallerstein and Blakeslee reported in the later study that problems and conflicts in some children resulting from divorce did not manifest themselves until much later, particularly in the case of girls who in the earlier study had generally appeared to cope better with divorce than boys. Rather, it appears that the long-term effects of divorce cannot always be predicted from the reactions of children at the outset. (See Wallerstein and Blakeslee, supra, at p. 15.)

At the five‑year point, Wallerstein and Kelly concluded that, while no factor could in every case be associated with a good outcome, the extent to which conflict between the parents had been resolved was the single most important factor in the well-being of the child. Following interviews ten years after their initial contact with the subjects, Wallerstein and Blakeslee again identified prolonged hostility between the parents as the single most destructive outcome for the children on divorce. Children subject to fierce legal battles between parents appear to be the most vulnerable group, and ongoing litigation is consistently identified as detrimental to the welfare of children (Wallerstein and Blakeslee, supra, at p. 196). Wallerstein and Kelly also determined, as have other researchers, that children generally fare best when they are able to maintain a continuing relationship with both parents. However, an equally important corollary to this conclusion is often ignored. That is, continued contact may only be in the best interests of the child where parents are not adversarial and where interaction between the child and the access parent is not beset by conflict. Where conflict cannot be resolved or minimized, the detriment of continued contact may outweigh the benefit, as forced cooperation between hostile parents may lead to further litigation and conflict, which itself extends and increases the difficulties faced by children. (See also Weitzman, supra, at pp. 253‑54.)

These findings are largely corroborated in another U.S. study of the impact of divorce on children. E. M. Hetherington, M. Cox and R. Cox in "Effects of Divorce on Parents and Children" in M. E. Lamb, ed., Nontraditional Families: Parenting and Child Development (1982), in a study of white middle class pre‑school children in the custody of their mothers, focused on the effects of conflict on children by comparing its effects in both divorced and intact families. The role of conflict in the welfare of children is highly visible in this study, as the authors determined that two years after divorce, children in low conflict divorced families actually fared better than those in high conflict intact families. Other studies which have come to similar conclusions about the role of conflict in the adjustment of the child after divorce include: E. S. Ellison, "Issues Concerning Parental Harmony and Children's Psychosocial Adjustment" (1983), 53 Amer. J. Orthopsychiat. 73; R. E. Emery, "Interparental Conflict and the Children of Discord and Divorce" (1982), 92 Psych. Bull. 310; M. Rutter, "Protective Factors in Children's Responses to Stress and Disadvantage", in M. W. Kent and J. E. Rolf, eds., Primary Prevention of Psychopathology: Social Competence in Children, vol. III (1979); J. R. Johnston, M. Kline and J. M. Tschann, "Ongoing Postdivorce Conflict: Effects on Children of Joint Custody and Frequent Access" (1989), 59 Amer. J. Orthopsychiat. 576. For further discussion, see also Boyd, "Women, Men and Relationships with Children: Is Equality Possible?" supra; Drakich, "In Whose Best Interests? The Politics of Joint Custody", supra; Weitzman, supra, at pp. 248, 254; LaFave, supra, at p. 489; J. F. Burrett, Child Access and Modern Family Law (1988); Weisman, supra.

Thus, while most research discloses that continued contact with both parents after divorce is normally in the best interests of the child, that finding cannot be separated from a consideration of the degree of conflict to which the child will be subject (Weisman, supra). Ironically, unrestricted access may, in some circumstances, cause the continuation of the very stresses from which the parties sought relief when they divorced, and the desire to maintain the pre‑existing roles of each parent may in fact result in no ending at all. (See Fineman, The Illusion of Equality: The Rhetoric and Reality of Divorce Reform, supra, at p. 164.)

Bearing in mind these concerns, in particular the fact that harm to a child may only become perceptible at a much later date, it is evident that the welfare of children is put at considerable risk if the prospect of harm becomes the sole prerequisite for restrictions on access. In my view, the proper approach to access was stated by Abella J.A. in M. (B.P.) v. M. (B.L.D.E.), supra, at p. 461.

It is not a question of what standard should be used to deprive a parent of access, it is a question of what standard should be used in deciding what form of access, if any, should be ordered. The answer is clear from the statute: the standard is the child's best interests.

The best interests of the child is not simply the right to be free of demonstrable harm. It is the positive right to the best possible arrangements in the circumstances of the parties. This could not be more clearly indicated than in the report on Family Law, supra, as the Law Reform Commission of Canada begins its recommendations on children and the dissolution of marriage with the following words at p. 63:

Children should have two fundamental rights when their parents' marriage ends:

(a)the right to social and psychological support by having the most suitable arrangements possible in the circumstances made for their custody, care and upbringing; and

(b)the right to economic support.

Without disregarding the disruption inherent in divorce, courts must do whatever is within their power to alleviate rather than disregard, minimize or normalize difficulties, and should beware of adopting a needlessly cavalier attitude where solutions can be seen to exist. The fact that many children experience stress and disruption upon the breakdown of the marriage, in my view, only increases the obligation of courts to focus their attention on the best interests of the child. The primary goal of the legal system on divorce must be to minimize the adverse effects on children. However, this requires a vision of the best interests of the child that is more than neutral to the conditions under which custody and access occur. The harm test clearly cannot meet this objective.

The Court of Appeal concedes that a court could be obliged to determine whether or not either the content of religious beliefs or conflict arising from exposure to conflicting beliefs is threatening "real harm" to the child. However, in the view of the majority, at p. 105, "it will be a rare case where such an issue could legitimately arise". The Court of Appeal also concludes that restrictions on access should only be allowed where harm can be demonstrated by independent experts.

It is clear that the proposed test falls far short of protecting the best interests of the child, as it may well be impossible to demonstrate the high level of harm the test envisions in many cases even where there are real concerns for the welfare of the child. It is unrealistic to expect that the accumulated evidence of ill effects engendered by separation or divorce will be immediate or readily demonstrable to a court as, in the usual case, custody decisions are made shortly after separation. However this does not mean that such concerns should therefore be readily dismissed. Rather, judges must exercise their discretion in order to prevent harm to the child, rather than merely to identify or establish its presence after the damage is done.

I would also point out that to adopt the harm test would be a regression back to the era when the interests of the child were subjugated to those of the parents. Despite the fact that the test purports to further the best interests of the child by promoting unfettered contact between non‑custodial parents and their children, in reality the test subordinates the best interests of children to a presumptive right of the non‑custodial parent to unrestricted access.

Finally, it is important to place this issue in its proper context by noting that the vast majority of access orders contain no restrictions on access. Over 96 percent of divorces are now finalized with either an uncontested hearing or no court hearing at all (Evaluation of the Divorce Act -- Phase II, supra, at p. 44). Of the remainder that are contested, most result in orders for liberal or reasonable access without restrictions. As a practical matter then, the question of restrictions only arises where something in the evidence or the circumstances of the parties suggests a concern or a potential problem. Where those concerns emanate from the custodial parent, that parent will, of course, bear the evidentiary burden of establishing that access should be curtailed. However, the standard that must be met is not the harm standard but evidence that restrictions are in the best interests of the child or, to put it otherwise, that unlimited access is not in the best interests of the child.

I agree with my colleague that expert evidence should not be routinely required to establish the best interests of the child. In my view, it is a modern day myth that experts are always better placed than parents to assess the needs of the child. Common sense requires us to acknowledge that the person involved in day to day care may observe changes in the behaviour, mood, attitude and development of a child that could go unnoticed by anyone else. The custodial parent normally has the best vantage point from which to assess the interests of the child, and thus will often provide the most reliable and complete source of information to the judge on the needs and interests of that child.

Furthermore, it is important to emphasize the importance of the evidence of children in custody and access disputes, and I would not wish to suggest that their testimony alone might not be a sufficient evidentiary basis upon which to restrict access. Courts have increasingly come to accept and understand in the criminal context that children themselves can be a reliable source of evidence to the judge (R. v. Khan, 1990 77 (SCC), [1990] 2 S.C.R. 531). To disregard their evidence when their own interests are directly at issue would, in my opinion, be at odds with this clear evolutionary trend in the law.

Many legal commentators have noted the degree to which custody and access disputes have become a contest between experts, involving increasing amounts of time and money. (See N. Bala, "Assessing the Assessor: Legal Issues" (1990), 6 C.F.L.Q. 179; P. J. Caplan and J. Wilson, "Assessing the Child Custody Assessors" (1990), 27 R.F.L. (3d) 121; L. Gélinas and B. M. Knoppers, "Le rôle des experts en droit québécois en matière de garde, d'accès et de protection" (1993), 53 R. du B. 3; Fineman, The Illusion of Equality: The Rhetoric and Reality of Divorce Reform, supra.) In the absence of clear legal presumptions about the best interests of children, judges have increasingly come to rely on the recommendations of experts to determine custody and access issues, believing that such experts possess objective, scientific knowledge and can in fact "know" what is in the best interests of the child. However, expert testimony, while helpful in some and perhaps many circumstances, is often inconclusive and contradictory (Gélinas and Knoppers, supra, at p. 17). That this should be so is not surprising, since such assessments are both speculative and may be affected by the professional values and biases of the assessors themselves.

Even where such expertise is valuable, there are impediments in such reliance. Assessments may occasion delays in resolving proceedings and may at times constitute a significant disruption in the lives of both parents and children. The cost involved in routinely hiring experts to establish the best interests of the child only increases the expense of custody litigation and is far beyond the resources of most divorcing couples. Furthermore, as Professor Bala, supra, points out, at p. 224, "much of what assessors ultimately recommend may simply be a matter of `common experience and common sense'".

Given these concerns, while the evidence of experts may form a valuable and necessary part of some custody and access decisions, most of the time they are unnecessary to an ordinary determination of the best interests of the child. Nor does the prospect of access restrictions inevitably require resort to expert opinion, as it may be apparent to the judge from the evidence of the parties and often the children themselves that access should only be granted subject to certain conditions.

I am compelled to address the issue of the right of the children to "know" their father, as Wood J.A. suggested at several points that "contact" in s. 16(10) of the Act encompasses the opportunity for the access parent and the child to know each other well and to appreciate each other as individuals. My colleague has also found that the trial judge erred in failing to consider the interest of the children in coming to "know" their father as he is, and is of the view that unrestricted access, even if it causes the deterioration of the relationship, is preferable to the alternative which prevents them from "knowing" him as he is. In my view, the reproaches addressed by my colleague to the trial judge are totally unfounded. The trial judge carefully assessed all the interests at play and not only did she not ignore the best interest of the children involved but also her whole decision is based on their best interest. That is, in fact, what the Court of Appeal reproached her for since, in their view, she should have applied the harm test.

However, to find that unrestricted access is the only available means by which a child may know a parent is to construct a false choice. Restrictions on access, in general or in the particular circumstances of this case, do not necessarily prevent children from coming to know their parents in meaningful ways. In this case, the purpose of the restrictions is to ensure that the children will continue to know their father "at all". The restrictions only affect the area of religious activities involving the children. Many other activities and topics of communication remain open to the respondent and his children. The children already "know" who their father is and are well aware of the significance of his religious beliefs in his life. Should they choose to, it remains open to them to investigate those beliefs further at some later point when they are ready and able to do so in a manner that is non‑threatening to them.

Furthermore, to interpret the goal of maximum contact in such a way as to require unrestricted access may be to defeat the very objective that the Act itself seeks to promote, if the pre‑eminence of unlimited "knowledge" results in the ultimate destruction of the relationship. Once the long-term interest of children in the maintenance of a relationship with the non‑custodial parent is acknowledged, it is difficult to see the logic or the value in this approach. For example, a non-custodial parent, during access, may prevent on personal, philosophical or religious grounds, his or her child from attending school. Restricting access may be in the best interests of the child even if such an order impinges on the child's ability to "know" the parent.

In brief, the content of the rule, its constitutionality and the applicable standard of proof do not favour the respondent's position. The respondent's main argument, however, which I will now discuss, relates to the infringement of his freedom of religion under s. 2(a) of the Charter due to the trial judge's access order.

Freedom of Religion

This Court has held on occasions too numerous to require mention at this point that it will consider both the purpose and the context of a right when determining whether there has been an infringement of the Charter. I must state at the outset that the purposes underlying the protection of religious and expressive freedoms have little if anything to do with regulating activities between family members. Such rights are public in nature and have typically referred to and encompassed freedom of the individual from state compulsion or restraints. Even the most ardent liberal theorists and advocates of rights have never pretended that these fundamental freedoms should provide a basis on which an individual could, for example, claim the protection of the state either to say or refrain from saying something to a spouse or child. The mere fact that the state plays a role in custody and access decisions in formalizing the circumstances of parent‑child interaction does not transform the essentially private character of such interchanges into activity which should be subject to Charter scrutiny.

An examination of our traditional understanding of freedom of religion and expression reveals that those notions are inappropriate and ill‑suited to the family, if only because of the differences in power and development and the nature of rights and obligations between parent and child. While legitimate questions may arise about the role of the state, and hence the application of the Charter, in regulating other aspects of family law, in my view, it is difficult to imagine that any valid purpose is served by importing the discourse of freedom of expression and religion into orders made in the resolution of custody and access disputes.

It follows that, once the best interests test itself has been found to accord with Charter values, the trial judge's order itself is not subject to further constitutional review, as the necessary state infringement of religious rights required to sustain a challenge based on the Charter is no longer present.

As s. 32 dictates, the Charter applies to governments and legislatures. Its purpose is to provide a measure of protection from the coercive power of the state and a mechanism of review to persons who find themselves unjustly burdened or affected by the actions of government. It is not meant to provide a means to regulate the affairs of private citizens. Thus, the sine qua non to any application of the Charter is the presence of state action, whether by legislation or other means.

In RWDSU v. Dolphin Delivery Ltd., 1986 5 (SCC), [1986] 2 S.C.R. 573, this Court first considered the meaning of "government action", and articulated the principle which established the threshold for judicial review under the Charter: the Charter applies to the legislative, executive and administrative branches of government but does not apply to judicial orders made in the resolution of private disputes. The issue arose again in Tremblay v. Daigle, 1989 33 (SCC), [1989] 2 S.C.R. 530, when the appellant sought to uphold an injunction preventing his former girlfriend from obtaining an abortion, relying on, among other grounds, the alleged rights of the foetus under s. 7 of the Charter. The Court held that, as there was no state action to be impugned and the Charter could not be invoked in a civil action between two parties, the principle outlined in Dolphin Delivery provided a complete answer to the Charter argument.

Exceptions to this rule have been recognized in certain specific circumstances. The first such instance arose in B.C.G.E.U. v. British Columbia (Attorney General), 1988 3 (SCC), [1988] 2 S.C.R. 214, when the Court considered the application of the Charter to the order of a judge who, acting on his own motion, issued an injunction restraining picketing outside a court house. Dickson C.J. in that case found the order subject to the Charter, on the basis that the ruling was entirely public in nature and was required to ensure continuing access to the courts and the protection of the rule of law and the Charter rights of others. In Slaight Communications Inc. v. Davidson, 1989 92 (SCC), [1989] 1 S.C.R. 1038, the Court found the order of an adjudicator pursuant to the Canada Labour Code properly subject to Charter review. The Court distinguished the exercise of the arbitrator's powers in that case from judicial decision‑making on the basis that the adjudicator was appointed pursuant to a legislative provision and derived all his powers from the statute.

The present case clearly falls to be decided under the principles enunciated in Dolphin Delivery as custody and access matters are essentially private in nature and there exists no state action to be impugned. In contradistinction to the situation in Slaight Communications, the powers of a judge making an order under the Act cannot be equated with those of an adjudicator who derives his authority from a legislative enactment. The judiciary is a separate branch of the government, and possesses the inherent jurisdiction under its parens patriae power to rule in the interests of the child. This power is both distinct from and long predates the statutory entrenchment of the best interests test. In the result, the Charter will not apply to the order.

As the Court has stressed on previous occasions, Charter values nonetheless remain an important consideration in judicial decision‑making. Courts must strive to uphold Charter values, and preference should be given to such values in the interpretation of legislation over those which run contrary to them (Slaight Communications, supra; Hills v. Canada (Attorney General), 1988 67 (SCC), [1988] 1 S.C.R. 513, at p. 558; Canada (Attorney General) v. Mossop, 1993 164 (SCC), [1993] 1 S.C.R. 554).

There is no question that, had the order been based on the sole fact that the respondent adheres to the Jehovah's Witness faith, the order could not be legitimized, as it has long been a tenet of the common law that courts will not prefer one religion over another in the adjudication of custody disputes. (See In re McGrath (Infants), supra; Re Bennett Infants, 1952 293 (ON CA), [1952] O.W.N. 621 (Ont. C.A.) at p. 624; Delvenne v. Nabbie (1977), 1977 2270 (MB CA), 4 R.F.L. (2d) 21 (Man. C.A.); Irmert v. Irmert (1984), 1984 ABCA 286, 64 A.R. 342 (C.A.); Harvey v. Lapointe (1988), 1988 8787 (QC CS), 13 R.F.L. (3d) 134 (Que. C.A.).) Nor is a court authorized to dictate to a parent his or her religious philosophy (McQuillan v. McQuillan (1975), 1975 1946 (ON SC), 21 R.F.L. 324 (Ont. H.C.)). Decisions regarding custody and access must not be based on the parents' faith, whether or not that religion is that of the majority. However, the religion of the parties may be relevant as one of the circumstances to be assessed along with all the others in the determination of the best interests of the child, as it is in this case.

In instances where there is conflict over religion, it is important to emphasize that the court is not engaged in adjudicating a "war of religion" nor are the religious beliefs of the parties themselves on trial. Rather, as courts have often recognized, it is the manner in which such beliefs are practised together with the impact and effect they have on the child which must be considered (Struncova v. Guay (1984), 1984 4817 (QC CS), 39 R.F.L. (2d) 298 (Que. S.C.); Brown v. Brown, supra; Irmert v. Irmert, supra; Harvey v. Lapointe, supra; Fougere v. Fougere, supra). In all cases where the effects of religious practices are at issue, the best interests of the child must prevail (Sullivan v. Fox (1984), 1984 4805 (PE PC), 38 R.F.L. (2d) 293 (P.E.I.S.C.)). See also C. E. Schneider, "Religion and Child Custody" (1992), 25 U. Mich. J.L. Ref. 879.

In the vast majority of cases, social values such as religion are shared between the parents and pose no threat to the exercise of access. It is when such values clash, whether over religion, education, ethnic traditions, mores or the lifestyles of the parents that problems arise. It is probably safe to say that in the "best case" scenario, courts would prefer to avoid orders which touch on peoples' beliefs and religion, as few issues are more likely to evoke the passions of both the parties and the wider community. However, it must be remembered that courts are only called to adjudicate such issues when the differences between the parties themselves have become irreconcilable. At this point, courts have no choice but to resolve the matter according to the best interests of the child.

Moreover, even if the Charter were to apply to custody and access orders, no infringement of religious freedoms would occur where such orders are made in the best interests of the child. As this Court recognized in R. v. Big M Drug Mart Ltd., supra, at pp. 336-37:

The essence of the concept of freedom of religion is the right to entertain such religious beliefs as a person chooses, the right to declare religious beliefs openly and without fear of hindrance or reprisal, and the right to manifest religious belief by worship and practice or by teaching and dissemination. But the concept means more than that.

Freedom can primarily be characterized by the absence of coercion or constraint. . . . Freedom means that, subject to such limitations as are necessary to protect public safety, order, health, or morals or the fundamental rights and freedoms of others, no one is to be forced to act in a way contrary to his beliefs or his conscience. [Emphasis added.]

Thus, freedom of religion is inherently limited by a number of considerations, including the rights and freedoms of others. While parents are free to engage in religious practices themselves, those activities may be curtailed where they interfere with the best interests of the child without thereby infringing the parent's religious freedoms.

There is no dispute in the case law as regards this principle since, even where the religious rights of the non‑custodial parent in access disputes have been recognized, courts have nonetheless imposed conditions on their exercise where warranted by the interests of the child. As Malouf J.A. for the Quebec Court of Appeal held in Droit de la famille -‑ 955, 1991 3838 (QC CA), [1991] R.J.Q. 599 (C.A.), at p. 606:

Such constraints, provided they are reasonable in their content and application, do not interfere in any way with a particular parent's freedom of religion. They simply reaffirm the principle that the best interest of the child must be protected. They are not inconsistent with the basic freedoms referred to in the Charter.

(See also Droit de la famille -‑ 353, 1987 398 (QC CA), [1987] R.J.Q. 545 (C.A.).)

The only question is the applicable standard for imposing such conditions. As I have indicated above, that standard is not one of harm but must at all times be the best interests of the child.

Both the Court of Appeal and the respondent rely on the decision of Hockey v. Hockey, supra, for the proposition that access orders which restrict the ability of non‑custodial parents to involve their children in religious activities contrary to the wishes of the custodial parent violate freedom of religion. In that case, the custodial parent's solicitor had sent a letter to the access parent which included the following statement: "[t]o ensure that there is no misunderstanding let me state unequivocally that the restrictions of your exercise of access to the children has been precipitated primarily because of your acceptance of the Jehovah's Witness faith" (p. 106). Not surprisingly, the court found the subsequent denial of access by the custodial parent for a period of six months, despite a court order permitting access, to be a form of religious discrimination. The court concluded in those circumstances that, in the absence of any finding that the sharing of beliefs or practices with the children by the non‑custodial parent was contrary to the best interests of the children, the Act must be interpreted in a manner compatible with the religious freedom of the access parent.

It appears that the basis of that particular decision was the absence of any evidence before the court of adverse effects on the children. Moreover, the court in Hockey v. Hockey explicitly recognized that the religious rights of the access parent could be curtailed in the interests of the child. Despite a contrary interpretation by the Court of Appeal in the present case, in my view that case does not support the conclusion that restrictions on religious activities during access cannot be sustained unless harm, or the threat of harm, to the child has been established.

The respondent also invokes the decision of Zummo v. Zummo, 574 A.2d 1130 (Pa. 1990), in support of his submissions that the religious freedom of an access parent may only be curtailed where harm can be conclusively demonstrated. In Zummo v. Zummo, the Pennsylvania Superior Court engaged in a comprehensive historical review of the protection of freedom of religion under the First Amendment to the United States Constitution. On that basis, Kelly J. concluded that limits to the non‑custodial parent's involvement in the religious upbringing of the child could not be sustained in the absence of clear evidence that access without restrictions would pose a substantial threat of present or future harm to the child.

In my view, it is important, when considering the applicability of such a standard in Canadian jurisprudence, to be conscious of the differences in both the history and the constitutional protection of freedom of religion in Canada and the United States. Without delving into the development of freedom of religion in the United States, it is sufficient to note that American courts have strictly interpreted the requirement of non‑interference by the state in any matter touching on religion. In addition, we must recall that under the Charter, a court order pursuant to the litigation of a private issue such as access does not constitute state infringement of religious freedoms.

In Zummo v. Zummo, the court accepted the conclusions of a number of studies which demonstrated that exposure to conflicting religious doctrines should not give rise to a presumption of harm to the child. I am in complete agreement that there is ordinarily nothing wrong with the exposure of a child to different religions or beliefs and that such exposure may be of value to the child. However, it is crucial to acknowledge that there may at times be important differences between intermarried intact families and those in the process of separation or divorce. Indeed, the authors of two of the studies relied on in Zummo v. Zummo later expressly disavowed the application and relevance of their findings where parents were not in agreement or where children were, by virtue of divorce, thrust into two religions. (See M. Weiss and R. Abramoff, "The Enforceability of Religious Upbringing Agreements" (1991), 25 J. Mar. L. Rev. 655, at pp. 714‑15). The missing element is conflict. In intact families, parents are normally in agreement about how to raise their children and have accommodated the presence of two religions. However, where religion becomes a source of conflict between the parents or is the very cause of the marriage breakdown, it is generally not in the best interests of the child and may in some circumstances be very detrimental for the child to be drawn into the controversy over religious matters.

As one commentator has pointed out, it is precisely the cases in which children become embroiled in religious conflict that cast doubt on the wisdom of the decisions which have allowed the religious rights of the access parent to prevail (J. T. Syrtash, Religion and Culture in Canadian Family Law (1992), at p. 90). Where there is conflict over religion, courts must secure the longstanding authority of the custodial parent to make decisions over religious activities. This ensures that stress occasioned by such issues does not become a continuing and ultimately destructive feature in the life of the child after divorce. (See S. M. Zarowny, "The Religious Upbringing of Children After Divorce" (1980), 56 Notre Dame Law. 160, at p. 173; Weiss and Abramoff, supra.)

Freedom of Expression

The respondent also argues that the restrictions on communicating his religious views to his children infringe his rights to expressive freedom. Wood J.A. agreed, although he subsumed this right under the rubric of freedom of religion, holding that an important part of the right to teach and disseminate religious beliefs is the right to share those beliefs with one's children. Southin J.A., for her part, was of the view that the appellant does not have the right to control the conversations which take place between the respondent and his children during access, reasoning that the child has a right to know anyone to whom the court grants access and that a child cannot know anyone unless he or she is able to speak freely to that person.

I have no quarrel with these conclusions as general principles, and would observe that they normally reflect and enhance the very purpose of access itself. In the overwhelming majority of cases, courts do in fact recognize and give effect to them, if only through the absence of any restrictions regarding communication between the child and the access parent. However, it is beyond contention that rights do not exist in a vacuum but are shaped and formed both by the particular context in which they are exercised and the rights of others (Edmonton Journal v. Alberta (Attorney General), 1989 20 (SCC), [1989] 2 S.C.R. 1326). Dickson C.J. in Fraser v. Public Service Staff Relations Board, 1985 14 (SCC), [1985] 2 S.C.R. 455, at p. 463, after acknowledging the centrality and importance of freedom of expression in our society, observed:

But it is not an absolute value. Probably no values are absolute. All important values must be qualified, and balanced against, other important, and often competing, values.

In this case, there are other powerful competing interests which must be recognized, not the least of which, in addition to the best interests of the children, are the freedoms of expression and religion of the children themselves. There is cogent, persuasive evidence, found credible by the trial judge, that the children themselves do not want to discuss religion with their father or be subject to his comments about beliefs which are at odds with their own religious upbringing, whether they take the form of indoctrination, instruction or mere observations. Indeed, the letters written to the trial judge disclose that, not only do they not want it, but also that the prospect of such discussions has so profoundly disturbed the children and coloured the periods of access that they no longer wish to continue to see the respondent according to a schedule.

In such circumstances, it is obviously inadequate merely to invoke freedom of religion or expression of an access parent without considering the effect on the children and their inability to assert their own desires and rights. Rather, both the best interests of the children and a respect for their rights may require restrictions on communication, if only so that the larger interest, maintenance and development of the relationship between the access parent and child, is not frustrated by the means by which it is carried out.

In the result, I would dismiss the respondent's argument that his freedoms of religion and expression have been infringed by the trial judge's order. He remains totally free to practise his religion. I also reject his submission that he has the right to engage his children in religious observances with him where it is determined that such observances are detrimental to their best interests.

In summary, as a matter of statutory interpretation, the Divorce Act mandates that, in decisions of custody and access, the sole consideration be the best interests of the child. The focus must remain at all times on the child, not the needs or interests of the parents, and parental rights play no role in such decisions except in so far as they are necessary to ensure the best interests of the child.

The custodial parent is responsible for the care and upbringing of the child, including decisions concerning the education, religion, health and well‑being of the child. Parental authority rests with the custodial parent, not for his or her own benefit, but in order to enable that parent to discharge effectively the obligations and responsibilities owed to the child.

As set out in the Act, maximum contact between the child and the non‑custodial parent is a worthwhile goal which should be pursued to the extent that it is in the best interests of the child. Generous and unrestricted access, which is the norm, should be favoured except when such access would not be in the best interests of the child. However, ongoing conflict between parents which adversely affects the child must be minimized or avoided, as it is the single factor which has consistently proven to be severely detrimental to children upon separation or divorce.

The best interests of the child must be approached from a child‑centred perspective. It is not simply the right to be free of significant harm. It is the right of the particular child in question to the best possible arrangements in the circumstances of the parties, taking into consideration the wide spectrum of factors which may affect the child's physical, spiritual, moral and emotional well‑being and the milieu in which the child lives.

Where the question of restrictions on access arises, the best interests of the child must be determined by considering the "condition, means, needs and other circumstances of the child" as required by the Act. The totality of these circumstances must be considered. Nothing in the Act suggests that harm should be the controlling factor. To adopt the harm standard would be to invert the focus of the best interests test and place the risk of error on the child, contrary to the objectives of the Act.

Expert evidence, while helpful in some cases, is not routinely required to establish the best interests of the child. That determination is normally possible from the evidence of the parties themselves and, in some cases, the testimony of the children involved.

Freedom of religion and expression are fundamental values protected by the Charter. However, the best interests of the child standard in the Divorce Act does not offend Charter values, but is completely consonant with the underlying objectives of the Charter. The Charter has no application to private disputes between parents in the family context, nor does it apply to court orders in the area of custody and access. While a child's exposure to different parental faiths or beliefs may be of value, when such exposure is a source of conflict and is not in the best interests of the child, such exposure may be curtailed.

This brings us to the evidence in this case which led to the trial judge's order restricting the respondent's access.

Application to the Case

At the outset, it must be remembered that this Court has always emphasized the caution with which an appellate court should interfere with the decision of a trial judge in custody matters. (See Talsky v. Talsky, supra; Adams v. McLeod, 1978 5 (SCC), [1978] 2 S.C.R. 621; Novic v. Novic, 1983 55 (SCC), [1983] 1 S.C.R. 696.) Because the trial judge has the inestimable advantage of observing the parties and assessing the witnesses first hand, such decisions should not be interfered with unless there is a gross distortion of the evidence or misapprehension of the relevant legal principles. As the foregoing discussion demonstrates, Proudfoot J. did not err in law as to the proper standard applicable to the case. She was entirely correct in holding that the only relevant consideration is the best interests of the child. The sole remaining question therefore, is whether there is evidence that the best interests of the children would be jeopardized by the respondent's unrestricted access to his children.

The majority of the Court of Appeal found no evidence of harm which would support the order made at trial. Even if I were to agree that some measure of harm must be demonstrated before the trial judge is entitled to restrict the respondent's religious activities with the children, a view I do not share, the evidence amply demonstrates the harm such religious activity has brought about. While the rationale underlying the right of access and the objective of maximum contact is to permit the relationship to flourish, in this case the contrary result has occurred, as it has antagonized the children.

I would begin by noting that the appellant has always favoured access by the respondent to his children. This is not an iota of evidence that the respondent's general rights of access were threatened. Rather, the only restrictions that were requested were aimed at reducing the area of conflict which had arisen on account of the respondent's behaviour with his children during access.

Although the respondent characterizes this litigation as an attack on his religious freedom, those beliefs are not at risk, as the restrictions place no limits on the respondent's ability to engage in religious practices himself. It is the effect of his practices on the best interests of his three daughters that is in question here.

In reaching her decision about the impact of these practices on the children, the trial judge relied in large measure on the reports of the two witnesses for the court, Donna MacLean, the family court counsellor whom she found to be thorough and highly credible in both her reports and testimony, and Dr. Williams, a psychologist.

In his first report, dated November 28, 1988, Dr. Williams found that the major obstacle to smooth access was the matter of religion. He found that the children were not significantly distressed by the issue at that time, but regarded it as a potential problem in the future. He stated:

I see Natalie as being particularly vulnerable to such issues in light of the dynamics of her personality as well as her age. As such, in my judgement some agreement and/or understanding will need to be arrived at as to the degree of religious dogma to which Mr. Young can expose the children, bearing in mind that such beliefs and related activities are anathema to Mrs. Young.

In his assessment the following year (September 22, 1989), he reported that, although they were continuing to deal with the separation adequately, the children were under increasing pressure regarding family issues. They expressed antagonism to their father, no longer trusted him and generally did not view him in a positive light. They were firmly opposed to attending religious services or being schooled in the tenets of his beliefs. The expert found that they were under stress from the religious issue and ongoing litigation and conveyed a need to get it settled in order to get on with their lives. Dr. Williams found, with respect to the children's emotional and social foundation, that:

. . . I am of the opinion that it is in the best interests of the children that the dispute over tangible and philosophical/religious issues be settled conclusively . . . . nothing has been broken and thus there is no apparent need for repair. If, however, the children are obliged to labor at length under the pressures they are presently experiencing a destabilizing influence ‑‑ particularly in the case of Natalie ‑‑ will likely be exerted.

Donna MacLean, in her report of January 30, 1989 essentially corroborated this finding. In the updated report of April 18, 1989, she noted that the children were unhappy with the access visits and that their relationship with their father was deteriorating because he was making the children feel guilty and uncomfortable by questioning them.

The trial judge also had access to the direct testimony of the two older children who wrote letters to the judge which are most revealing of their state of mind. From the letters, it is evident that both are afraid to disclose their thoughts about the visits to their father for fear that he will make them feel guilty. Adrienne, the eldest, wrote that she did not want to visit him or sleep overnight because she thinks he will trick her into doing things and quiz her about the Jehovah's Witness religion. She expressed frustration regarding access, visitation and the court proceedings, saying "It's driving me crazy." Natalie, the middle child, evinced considerable distress about the family situation, stating that her father makes her feel guilty and makes her cry when she wants to go home from access visits, and that she is afraid to tell him. She repeated that she wanted the court proceedings to be over.

In my opinion, this evidence amply demonstrates the stress the children were under, much of it related to the children's resistance to becoming involved in their father's religious practices. The trial judge can in no way be said to have erred in finding that the best interests of these children were served by removing the source of conflict, particularly as the ultimate purpose of the restrictions was to preserve the relationship between the respondent and his children.

Moreover, there was evidence leading the trial judge to conclude that the respondent would not respect the wishes of the children without an order to do so. As the family court counsellor reported, "it is of concern that Mr. Young does not acknowledge the choice of the older children to remain with their mother". Proudfoot J. found as a fact that the respondent wanted custody in order to control the religious upbringing of the children. In other words, she found that his interest was not simply to share or communicate these beliefs; rather he fundamentally disagreed with allowing the appellant control over religious matters. This finding of fact is difficult to reconcile with the Court of Appeal's ruling that the trial judge was in error since, in their view, the respondent was bound to respect his children's wishes not to attend at his place of worship or accompany him on his proselytization efforts, which is exactly what the trial judge ordered. However, they discarded the evidence that the respondent was not willing to do so voluntarily. The trial judge had made a finding that the respondent could not be relied upon to act in the best interests of the children as regards his religious practices when exercising access. That finding should have been respected by the Court of Appeal.

In my opinion, the fact that the circumstances in this case were found by the Court of Appeal not to meet the harm threshold clearly demonstrates the difficulty, if not the impossibility, of reconciling such a standard with the best interests of the child even where experts testify, as they did in this case.

As the trial judge observed, the respondent was quite unconcerned with the conflict and stress on the children caused by the pursuit of his own religious interests. At the same time as he found the resources to press his claims to religious rights, he apparently did not find the payment of maintenance a top priority. Upon reading the evidence, I cannot disagree with the trial judge that the respondent's concern with religious rights had clearly overtaken the practical realities of parenthood and overshadowed his larger responsibilities to his children in this case.

Since writing these reasons, I have had the opportunity to read the joint reasons of my colleagues Justices Cory and Iacobucci. They invite the following comments.

First, as to the first part of the trial judge's order, contrary to their affirmation, the issue is not moot. It is precisely because the trial judge found, as a fact, that the respondent could not be trusted for the reasons she outlined, that she issued that order. How can a court of appeal reverse those findings of fact, and yet, on the same facts, affirm that such an order is no longer necessary, particularly when the court of appeal has not indicated where the trial judge has erred in her assessment of those facts?

As to the second part of the order, my colleagues assert that the best interests test does not support an order preventing the respondent from discussing religion with his children, on the ground that it is (at p. 110) "difficult to accept that any genuine and otherwise proper discussion between a parent and his or her child should be curtailed by court orders". Otherwise, they suggest, a non-custodial parent who espouses a theory of evolution would be ordered, under the best interests test, not to discuss or explain his or her views to a child in the custody of a parent who is a member of a fundamentalist religion.

However, this scenario has nothing to do with the facts of this case. With respect, to equate the two scenarios is to misapprehend fundamentally the focus of the best interests test. It is not the fact that the parents differ in their fundamental beliefs that warrants the restrictions in this case. It is the finding of fact made by the trial judge, on the basis of evidence she found credible, that continuing conflict over religion, including the respondent's repeated attempts to discuss religious matters with the children against their clearly expressed desires, profoundly disturbed the children and was contrary to their best interests. At the time of trial, the respondent was not engaging in other religious activities with the children, as they had already been curtailed by the interim order almost a year and a half earlier. Therefore, it is precisely these continuing "discussions" that were disturbing the children, causing the deterioration of their relationship with their father and which, therefore, had to be curtailed.

I wish to emphasize two points. As I thought I made clear earlier in my reasons, a child's exposure to different parental faiths or beliefs may generally be of value and even of great value; there is no presumption that such exposure is not in the best interests of the child. Rather, the contrary is true. Second, as I also emphasized earlier, generous and unrestricted access should be favoured except where it is not in the best interests of the child. Hence, restrictions such as the trial judge found to be required in the second part of her order will be rare indeed and there is no reason to suppose that, absent a threat to the best interests of the child, any question of such restrictions will arise. However, it is important to acknowledge that in those rare cases where parents cross the line and engage in conduct which constitutes, in the words of my colleagues at p. 110, "indoctrination, enlistment or harassment", courts have a duty to intervene in the best interests of children. The evidence strongly suggests that this is just such a case and that is the finding of the trial judge. In my view, these findings of fact must be respected absent, as here, an error in the applicable principles.

In the result, I would allow the appeal and restore the order at trial, the whole with costs throughout. I would answer the constitutional questions in the negative.

The following are the reasons delivered by

Sopinka J. -- I have read the reasons of my colleagues Justice L'Heureux-Dubé and Justice McLachlin and I find myself in agreement with McLachlin J. as to the disposition of the appeal and with most of her reasons. While I agree with McLachlin J. that the ultimate determination in deciding issues of custody and access is the "best interests of the child test", it must be reconciled with the Canadian Charter of Rights and Freedoms. General language in a statute which, in its breadth, potentially confers the power to override Charter values must be interpreted to respect those values. See Slaight Communications Inc. v. Davidson, 1989 92 (SCC), [1989] 1 S.C.R. 1038. It cannot be done the other way around and allow the best interests test in its broadest interpretation to read down Charter rights so as to accommodate this interpretation.

In my view, the test in s. 16(10) of the Divorce Act, R.S.C., 1985, c. 3 (2nd Supp.), and the Charter right involved in this case, namely freedom of religious expression, can best be reconciled by interpreting the best interests test to allow the right to be overridden only if its exercise would occasion consequences that involve more than inconvenience, upset or disruption to the child and incidentally to the custodial parent. The long-term value to a child of a meaningful relationship with both parents is a policy that is affirmed in the Divorce Act. This means allowing each to engage in those activities which contribute to identify the parent for what he or she really is. The access parent is not expected to act out a part or assume a phony lifestyle during access periods. The policy favouring activities that promote a meaningful relationship is not displaced unless there is a substantial risk of harm to the child.

In this regard, I agree with Wood J.A. in the Court of Appeal (1990), 1990 3813 (BC CA), 50 B.C.L.R. (2d) 1. In his reasons he states, at p. 96:

I believe the whole of s. 16 of the Divorce Act of 1985, when properly construed, reflects the modern view that the best interests of a child are more aptly served by a law which recognizes the right of that child to a meaningful post-divorce relationship with both parents. That construction in turn requires that the distribution of "rights", between the custodial and the access parent, be such as to encourage such a relationship. And such a construction is inconsistent with the full-blooded traditional notion of guardianship which would give the custodial parent the absolute right to exercise full control over the child even when the other parent is exercising his or her right of access.

"Harm" is a term which in this context connotes an adverse effect on the child's upbringing that is more than transitory. The impugned exercise by the access parent must be shown to create a substantial risk that the child's physical, psychological or moral well-being will be adversely affected. Exposure to new experiences and ideas may upset children and cause them considerable discomfort. Anything from starting school to having to go to bed may evoke a strong emotional response. This does not mean that these experiences are not in the long-term best interests of the child. Similarly, conflict between parents on many matters including religion is not uncommon, but in itself cannot be assumed to be harmful unless it produces a prolonged acrimonious atmosphere.

I would, therefore, go a step further than my colleague McLachlin J. and conclude that what is in the best interests of the child is the generally applicable test, but in its application to restrict religious expression, risk of substantial harm is not only an important factor but also must be shown.

Interpreted in this way, the statutory test in s. 16(10) of the Divorce Act does not constitute a limitation on freedom of religious expression. As my colleague points out, this freedom does not extend to protect conduct which is harmful to others. I would not, however, adopt her expansion of the term "injure" as used in R. v. Big M Drug Mart Ltd., 1985 69 (SCC), [1985] 1 S.C.R. 295, to arrive at the conclusion that anything that is not in the best interests of the child (on the basis of her definition of this test) is subtracted from the Charter right involved.

I would dispose of the appeal and the constitutional questions as proposed by McLachlin J.

The following are the reasons delivered by

Cory and Iacobucci JJ. -- We have read with great interest the excellent reasons of Justice L'Heureux-Dubé and Justice McLachlin. We are in agreement with their conclusions that the best interests of the child standard provided in ss. 16(8) and 17(5) of the Divorce Act, R.S.C., 1985, c. 3 (2nd Supp.), does not violate ss. 2(a), (b) and (d), and s. 15 of the Canadian Charter of Rights and Freedoms substantially for the reasons given by our colleagues. In this respect, however, we wish to refrain from expressing any opinion on McLachlin J.'s discussion of whether, if an infringement of the Charter were found, such an infringement would be so trivial as not to warrant Charter protection. We similarly wish to reserve our views on the question discussed by L'Heureux-Dubé J. of whether or not the Charter applies to judicial orders made in custody or access proceedings.

We agree, again for many of the reasons she advances, with L'Heureux-Dubé J. that the issue of access to children should be determined on the basis of what is in the best interests of the child. In that respect, we also agree with both our colleagues that expert evidence is, while admittedly helpful is some cases, not always necessary to establish the best interests of the child; that question can be determined normally from the evidence of parties themselves and the testimony, where appropriate, of the children concerned.

We note that the majority of the British Columbia Court of Appeal (1990), 1990 3813 (BC CA), 50 B.C.L.R. (2d) 1, held that the matters of the children attending religious services with the respondent and accompanying him on his proselytizing activities were resolved by the respondent's undertaking to respect his children's wishes in this regard. This leaves as the only remaining issue, whether the order forbidding the respondent from discussing his religion is valid. On this point, we agree with McLachlin J. that a proper application of the best interests of the child test does not support such an order. We find it difficult to accept that any genuine and otherwise proper discussion between a parent and his or her child should be curtailed by court orders. Indeed, curtailment of explanatory or discursive conversations or exchanges between a parent and child should be rarely ordered in our view. To take an example, suppose custodial Parent A is a member of a fundamentalist religion and access Parent B is a scientist who espouses the pure Darwinian theory of evolution. We find it unacceptable that Parent B should be ordered, under the rubric of the best interests of the child test, not to discuss or explain his views to his child as opposed to being forbidden from indoctrinating or otherwise undermining the religious choice made by the custodial parent for the child or children involved. Surely the best interests of the child test embraces genuine discussion of religious belief as opposed to indoctrination, enlistment or harassment having the aim or effect of undermining the religious decision made by the custodial parent.

On the property, monetary, and costs issues, we agree with the reasons of McLachlin J. and would dispose of the appeal in the manner proposed by her.

The following is the judgment delivered by

McLachlin J. -- This case raises the question whether a divorced parent, who does not have custody, should be able to offer his children his religious views over the objection of the custodial parent. This issue raises, in turn, the question of the place of the "best interests of the child" standard in the Canadian constitutional system.

The Background

Mr. and Mrs. Young were married in 1974. They had three daughters. In 1987 the Youngs separated. Mrs. Young took custody of the children; Mr. Young had access, subject to court-imposed restrictions following from Mrs. Young's objection to his religious activity with the children.

The separation was marked by a protracted series of court battles. It is unnecessary to detail the many disputes over which the parties joined issue before the case finally came on for trial. Suffice it to say that one of those issues was the disagreement between the parents over which religious activities Mr. Young might appropriately undertake with his daughters. Mrs. Young was brought up in the Anglican church. She expressed a wish that her children be brought up in the United Church, although the evidence suggests that religion, particularly organized religion, does not play an important role in the life of Mrs. Young. Mr. Young converted to the Jehovah's Witness faith two years prior to separation from his spouse. He wished, at the very least, to communicate his faith to his children. To this end, he read bible stories, and discussed his beliefs with his children during his periods of access. He also questioned them about religious matters during these periods.

The evidence shows that the elder two daughters like their father but, as time went on, came to dislike his religious instruction. There was evidence that by exposing his elder children to his religious beliefs, Mr. Young was damaging his relationship with his children and contributing to the stress the children were experiencing in adjusting to their parent's separation. On the other hand, the evidence also established that the children were functioning in an entirely normal fashion, suggesting that neither their mental nor physical health had been adversely affected by the dispute between their parents over Mr. Young's religious instruction, or by that instruction itself.

The trial judge granted custody of the children to Mrs. Young and access to Mr. Young. But Mr. Young's access was again restricted by court order. The order provided that Mr. Young not discuss the Jehovah's Witness religion with the children, not take them to any religious services, canvassing or meetings, and not expose the children to religious discussions with third parties without the prior consent of Mrs. Young. Both parties were ordered not to make adverse remarks about the other's beliefs. Mr. Young was also enjoined from preventing blood transfusions for the children, should the need arise. The basis of the order was the trial judge's finding (1989), 1989 8903 (BC SC), 24 R.F.L. (3d) 193, at p. 211, that "the religious conflict [between the parents] was causing a problem for the children." She wrote at p. 215:

There will be certain restrictions because that is necessary to protect the best interests of these children. That can only be done by putting an end to this religious conflict. The respondent has become so involved in enforcing his rights he has completely overlooked the welfare of the children. The respondent can have a meaningful relationship with his children without promoting his religious beliefs . . . . If the respondent is seriously interested in retaining a relationship with his children, he will have no difficulty abiding by the restrictions I propose to place on him when he has access.

The trial judge also made orders for the distribution of property and for costs which are considered later in these reasons.

Mr. Young appealed. The Court of Appeal (1990), 1990 3813 (BC CA), 50 B.C.L.R. (2d) 1, Southin J.A. dissenting in part, set aside the limitations on religious discussion and attendance, on the ground that it is in the best interests of children that they come to know their non-custodial parent fully, including his or her religious beliefs. The majority concluded, at p. 108, that restrictions should not be placed on the freedom of an access parent to discuss religion with his or her child, or to involve the child in religious activities, unless either "the existence of, or the potential for, real harm" to the child was established on the evidence, or the evidence established that the child did not consent to being subject to the access parent's views or practices. The Court of Appeal also altered the division of property and the awards of costs made by the trial judge. Mrs. Young appeals these rulings to this Court. I will consider each issue in turn.

A. The Limitations on Mr. Young's Access

The Constitutional Validity of the Best Interest of the Child Standard

The Divorce Act, R.S.C., 1985, c. 3 (2nd Supp.), provides that a court shall abide by the following matters in deciding questions of custody and access:

. . .

The courts are to heed the following provision in s. 17 when varying a custody order made under s. 16:

. . .

Two questions arise. First, what does the "best interests of the child" test require? Second, does this test, properly understood, infringe upon the guarantees of freedom of religion, expression, association and equality under the Canadian Charter of Rights and Freedoms? The sections of the Charter relevant to the case at bar read as follows:

The Canadian Charter of Rights and Freedoms guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.
Everyone has the following fundamental freedoms:

(a) freedom of conscience and religion;

(b) freedom of thought, belief, opinion and expression, including freedom of the press and other media of communication;

(d) freedom of association.

(1) Every individual is equal before and under the law and has the right to the equal protection and equal benefit of the law without discrimination and, in particular, without discrimination based on race, national or ethnic origin, colour, religion, sex, age or mental or physical disability.

The answers to the questions set out above depend upon an examination of what is meant by the "best interests of the child" test and the constitutional guarantees which are put in issue.

(a) The Best Interests of the Child Test

The Historical Perspective

The express rule that matters of custody and access should be resolved in accordance with the "best interests of the child" is of relatively recent origin. Under the common law regime of the 18th and 19th centuries the governing principle in a custody dispute was the rule of near-absolute paternal preference: see R. v. De Manneville (1804), 5 East. 221, 102 E.R. 1054; In re Taylor (1876), 4 Ch. D. 157. The rule was defended on pragmatic grounds, including what was thought to be the general interest of children: see In re Agar-Ellis (1883), 24 Ch. D. 317. In truth, the rule probably had more to do with the acceptance of the father's dominant right in all family matters, which in turn found its roots in the notion of the inherent superiority of men over women.

The rule of paternal preference was displaced by a rule establishing in the mother a primary right to custody of a child of tender years: see, for instance, An Act to Amend the Law relating to the custody of Infants, S. Prov. Can. 1855, c. 126, s. 1. Later still there arose a presumption, in many foreign jurisdictions and to a more limited extent in Canada, of maternal preference: Talsky v. Talsky, 1975 29 (SCC), [1976] 2 S.C.R. 292, Kades v. Kades (1961), 35 A.L.J.R. 251 (Aust. H.C.); see also Susan Maidment, Child Custody and Divorce (1984), Robert H. Mnookin, "Child-Custody Adjudication: Judicial Functions in the Face of Indeterminacy" (1975), 39 Law & Contemp. Probs. 226, and Allan Roth, "The Tender Years Presumption in Child Custody Disputes" (1976-77), 15 J. Fam. L. 423. This presumption, like the paternal preference rule, was justified on pragmatic grounds; the welfare of the child was the often cited reason for the presumption. So justified, the presumption carried the seeds of its own demise. Courts increasingly looked behind the preference to focus directly upon what was in the child's interest, which was sometimes found to conflict with a maternal preference.

By the 1970s, a number of western countries had accorded statutory recognition to a "best interests" or "welfare of the child" test. Questions relating to the weight to be given these interests, and the proper means of understanding these interests, remained. In England, the child's welfare is stipulated as the "first and paramount" consideration: Guardianship of Minors Act, 1971 (U.K.), 1971, c. 3, s. 1. English jurisprudence indicates that the child's welfare has, in fact, become the sole consideration: J. v. C., [1970] A.C. 668; Re K. (minors), [1977] 1 All E.R. 647. In Norway, decisions in respect of custody shall "mainly" (or "primarily") consider the interests of the child: Lov 8 April nr 7 om barn og foreldre, s. 34 (Law about Children and Parents, of April 7, 1981, no. 7, s. 34). In practice it appears that other criteria do not simply function as "tiebreakers" where the interests of the child would be equally well served by either parent, but can, in certain cases, determine the issue: see Jon Elster, "Solomonic Judgments: Against the Best Interest of the Child" (1987), 54 U. Chi. L. Rev. 1; also Lucy Smith and Peter Lødrup, "The Child in the Divorce Situation -- Factors Determining the Custody Question and the Use of Experts in Custody Cases in Norway", in Ian F. G. Baxter and Mary A. Eberts, eds., The Child and the Courts (1978).

Some variant of the best interests of the child test is in place in most American states, although in certain states the application of the test is delimited by a renewed concern that the debate be organized by certain presumptions, either a "primary caregiver" presumption, or a presumption in favour of the mother of a child of tender years: see David L. Chambers, "Rethinking the Substantive Rules for Custody Disputes in Divorce" (1984), 83 Mich. L. Rev. 477, and Roth, "The Tender Years Presumption in Child Custody Disputes", supra.

The Wording of the Act

Parliament has adopted the "best interests of the child" test as the basis upon which custody and access disputes are to be resolved. Three aspects of the way Parliament has done this merit comment.

First, the "best interests of the child" test is the only test. The express wording of s. 16(8) of the Divorce Act requires the court to look only at the best interests of the child in making orders of custody and access. This means that parental preferences and "rights" play no role.

Second, the test is broad. Parliament has recognized that the variety of circumstances which may arise in disputes over custody and access is so diverse that predetermined rules, designed to resolve certain types of disputes in advance, may not be useful. Rather, it has been left to the judge to decide what is in the "best interests of the child", by reference to the "condition, means, needs and other circumstances" of the child. Nevertheless, the judicial task is not one of pure discretion. By embodying the "best interests" test in legislation and by setting out general factors to be considered, Parliament has established a legal test, albeit a flexible one. Like all legal tests, it is to be applied according to the evidence in the case, viewed objectively. There is no room for the judge's personal predilections and prejudices. The judge's duty is to apply the law. He or she must not do what he or she wants to do but what he or she ought to do.

Third, s. 16(10) provides that in making an order, the court shall give effect "to the principle that a child of the marriage should have as much contact with each spouse as is consistent with the best interests of the child." This is significant. It stands as the only specific factor which Parliament has seen fit to single out as being something which the judge must consider. By mentioning this factor, Parliament has expressed its opinion that contact with each parent is valuable, and that the judge should ensure that this contact is maximized. The modifying phrase "as is consistent with the best interests of the child" means that the goal of maximum contact of each parent with the child is not absolute. To the extent that contact conflicts with the best interests of the child, it may be restricted. But only to that extent. Parliament's decision to maintain maximum contact between the child and both parents is amply supported by the literature, which suggests that children benefit from continued access: Michael Rutter, Maternal Deprivation Reassessed (1981), Robin Benians, "Preserving Parental Contact: a Factor in Promoting Healthy Growth and Development in Children", in Jo Tunnard, ed., Fostering Parental Contact: Arguments in Favour of Preserving Contact Between Children in Care and Their Families (1982).

Wood J.A. in the Court of Appeal, put the matter as follows at p. 93:

It seems to me that at the very least, by enacting this subsection [s. 16(10) of the Divorce Act], Parliament intended to facilitate a meaningful, as well as a continuing, post-divorce relationship between the children of the marriage and the access parent.

Without limiting the generality of the adjective "meaningful", such a relationship would surely include the opportunity on the part of the child to know that parent well and to enjoy the benefit of those attributes of parenthood which such person has to share. In most cases that would clearly be in the best interests of the child, and the best interests of the child, not parental rights, are the focus of the whole of s. 16 of the Act.

I would summarize the effect of the provisions of the Divorce Act on matters of access as follows. The ultimate test in all cases is the best interests of the child. This is a positive test, encompassing a wide variety of factors. One of the factors which the judge seeking to determine what is in the best interests of the child must have regard to is the desirability of maximizing contact between the child and each parent. But in the final analysis, decisions on access must reflect what is in the best interests of the child.

It follows from this that the proposition, put to us in argument, that the custodial parent should have the right to forbid certain types of contact between the access parent and the child, must fail. The custodial parent's wishes are not the ultimate criterion for limitations on access: see King v. Low, 1985 59 (SCC), [1985] 1 S.C.R. 87, at p. 101. The only circumstance in which contact with either parent can be limited is where the contact is shown to conflict with the best interests of the child.

Risk of harm to the child is not a condition precedent for limitations on access. The ultimate determinant in every case must be the best interests of the child. Many decisions on access may involve no reference to harm. For example, a judge might conclude that it is not in the best interests of a child that he or she see the access parent every day on the ground that this would result in undue disruption to the child's schedule of activities. Again, a judge might conclude that it is in the best interests of the child that he or she move with the custodial parent to a distant location, notwithstanding that this will limit the access of the other parent. Optimum access may simply not be in the best interests of the child for a variety of circumstances.

On the other hand, in some cases the risk of harm may be a factor to be considered in determining what is in the child's best interests. For example, where the limits on access relate to the quality of the access -- what the access parent may say or do with the child -- the question of harm may become highly relevant. Given the interest of the child in coming to know his or her access parent as fully as possible, judges may well be reluctant to impose limits on what the access parent may say or do with the child in the absence of some evidence suggesting that the activity may harm the child. The legal test is not harm; the Divorce Act makes this clear. However, in some circumstances, the risk of harm to the child or the absence thereof may become an important factor to be considered. To this extent I agree with the Court of Appeal that, in determining whether religious discussions and activities between parent and child should be curtailed, it may well behoove the judge to enquire whether the proposed conduct poses a risk of harming the child. In doing so, the judge should bear in mind that conflict between parents over the access issue does not necessarily indicate harm, nor does the objection of the child necessarily impose that conclusion. In some circumstances they may; in some they may not.

I conclude that the ultimate criterion for determining limits on access to a child is the best interests of the child. The custodial parent has no "right" to limit access. The judge must consider all factors relevant to determining what is in the child's best interests; a factor which must be considered in all cases is Parliament's view that contact with each parent is to be maximized to the extent that this is compatible with the best interests of the child. The risk of harm to the child, while not the ultimate legal test, may also be a factor to be considered. This is particularly so where the issue is the quality of access -- what the access parent may say or do with the child. In such cases, it will generally be relevant to consider whether the conduct in question poses a risk of harm to the child which outweighs the benefits of a free and open relationship which permits the child to know the access parent as he or she is. It goes without saying that, as for any other legal test, the judge, in determining what is in the best interests of the child, must act not on his or her personal views, but on the evidence.

(b) The Constitutionality of the Test

The first question is whether the Charter applies. Because of my conclusion later in these reasons that valid orders under the "best interests of the child" standard cannot violate the Charter, I find it unnecessary to decide whether the Charter applies to an action for access under the Divorce Act between two parents. For the purposes of this section, I assume that it does.

The constitutional focus in this case centres on the guarantee of freedom of religion in s. 2(a) of the Charter and the guarantee of freedom of expression in s. 2(b) of the Charter. The guarantees of freedom of association and equality apply only tangentially, if at all, and were not emphasized in argument.

The respondent says that legislative provision for the "best interests of the child" violates his religious and expressive freedom. The argument is that in some cases the "best interests of the child" will require a judge to make an order limiting expressive or religious freedom. Therefore, it is submitted, the test is unconstitutional, unless it can be saved under s. 1.

In my view, this argument cannot stand. The reason is that the guarantees of religious freedom and expressive freedom in the Charter do not protect conduct which violates the best interests of the child test.

Whether Application of the Best Interests Standard Violates the Charter

Does the Charter protect religious expression which is not in the best interests of the child? In my view, the answer to this question is no.

It is established that the guarantee of freedom of religion does not extend to religious activity which harms other people. In R. v. Big M Drug Mart Ltd., 1985 69 (SCC), [1985] 1 S.C.R. 295, this Court held that freedom of religion is not absolute; in particular, it does not extend to conduct that would injure or interfere with the parallel rights of others. Dickson J., (as he then was), after a lengthy historical review of freedom of religion in our society and legal system, concluded at p. 346:

Viewed in this context, the purpose of freedom of conscience and religion becomes clear. The values that underlie our political and philosophic traditions demand that every individual be free to hold and to manifest whatever beliefs and opinions his or her conscience dictates, provided inter alia only that such manifestations do not injure his or her neighbours or their parallel rights to hold and manifest beliefs and opinions of their own. [Emphasis added.]

The next question is whether conduct which is not in the best interests of the child amounts to an "injury" or intrusion on the rights of others within the meaning of this comment. If so, the guarantee of freedom of religion will not protect such conduct.

It is clear that conduct which poses a risk of harm to the child would not be protected. As noted earlier, religious expression and comment of a parent which is found to violate the best interests of a child will often do so because it poses a risk of harm to the child. If so, it is clear that the guarantee of religious freedom can offer no protection. But I think a case can be made that even in cases where a risk of harm may not have been established, the guarantee of freedom of religion should not be understood to extend to protecting conduct which is not in the best interests of the child. I understand "injure" in the passage cited from R. v. Big M Drug Mart Ltd., supra, to be a broad concept. To deprive a child of what a court has found to be in his or her best interests is to "injure", in the sense of not doing what is best for the child. The vulnerable situation of the child heightens the need for protection; if one is to err, it should not be in favour of the exercise of the alleged parental right, but in favour of the interests of the child. An additional factor which may come into play in the case of older children is the "parallel right" of others referred to by Dickson J., "to hold and manifest beliefs and opinions of their own". For these reasons, I conclude that the Charter guarantee of freedom of religion does not extend to protect conduct which is not in the best interests of the child under the Divorce Act.

I come then to freedom of expression. The ambit of this right has been more broadly drawn than freedom of conscience and religion, in that even harmful expression may be protected: Irwin Toy Ltd. v. Quebec (Attorney General), 1989 87 (SCC), [1989] 1 S.C.R. 927; R. v. Keegstra, 1990 24 (SCC), [1990] 3 S.C.R. 697; Reference re ss. 193 and 195.1(1)(c) of the Criminal Code (Man.), 1990 105 (SCC), [1990] 1 S.C.R. 1123. On the other hand, some forms of harmful expression are not constitutionally protected. Violence or threats of violence are not protected: see RWDSU v. Dolphin Delivery Ltd., 1986 5 (SCC), [1986] 2 S.C.R. 573, Keegstra, supra, at p. 733. Nor is expression which takes the form of "direct attacks by violent means on the physical liberty and integrity of another person" protected: Reference re ss. 193 and 195.1(1)(c) of the Criminal Code (Man.), supra, at p. 1186. The fact that conduct has been criminalized by Parliament is an indication, although not a conclusive one, that expressive conduct falls in the latter category: Reference re ss. 193 and 195.1(1)(c) of the Criminal Code (Man.), supra, at p. 1185.

The expression challenged in the case at bar does not take the form of violence or threats of violence, nor does it constitute a direct attack on the physical integrity or liberty of another. The harm done, if any, is of a psychological nature. (In saying this, I do not overlook the fact that some religious practices or forms of expression may involve psychological violence or physical violence.) Thus a prima facie case for protection under the guarantee of freedom of expression can be made out.

This poses a problem. Virtually every form of religious practice is expressive. Interpreting freedom of expression broadly, in the religious context, will nullify the principle that freedom of religion does not protect conduct which injures others or which conflicts with their parallel rights to hold and manifest beliefs of their own.

The answer to this apparent dilemma is found in the admonition of Dickson J. in R. v. Big M Drug Mart Ltd., supra, at p. 344, that a purposive approach to Charter interpretation requires that associated rights -- in this case religious freedom and freedom of expression -- be interpreted in a consistent and coherent manner:

. . . the purpose of the right or freedom in question is to be sought by reference to the character and the larger objects of the Charter itself, to the language chosen to articulate the specific right or freedom, to the historical origins of the concepts enshrined, and where applicable, to the meaning and purpose of the other specific rights and freedoms with which it is associated within the text of the Charter. [Emphasis added.]

This admonition must be coupled with the canon that the ambit of a particular right or freedom cannot be defined in the abstract but should rather be defined in the context of the particular activity in questions: Edmonton Journal v. Alberta (Attorney General), 1989 20 (SCC), [1989] 2 S.C.R. 1326. Taking these two injunctions together, one arrives at the conclusion that the court, focusing on the context of the particular case, must strive to interpret the related rights in a coherent manner.

The conduct with which this appeal is concerned -- the teaching of religious beliefs and practices to one's children -- while it has an expressive aspect, is predominantly religious. In seeking to reconcile the rights of freedom of religion and freedom of expression in this context, it is the religious aspect which must dominate. Reading the two guarantees together, it is my view that the limits of the guarantee of freedom of expression should govern in the context of religious instruction of children.

For these reasons, I conclude that the best interests of the child test under the Divorce Act does not violate the Charter and is constitutionally valid.

Whether the Order of the Trial Judge is Valid Under the Divorce Act

The question is whether the trial judge correctly applied the "best interests of the child" test set out in s. 16(8) and 16(10) of the Divorce Act.

The trial judge took a custody-oriented approach to the question of whether the father should be allowed to discuss and practise his religion with the children. For her the first question (at p. 205) was "who gets custody and guardianship of the children and what flows from that." She concluded that custody confers on the custodial parent the exclusive "right" to make all decisions with respect to the child's education, health care and religion. It followed inevitably from this (at p. 205) that she saw the custodial parent as possessing the "sole responsibility for . . . the religious instruction of the child." This exclusive right in Mrs. Young to determine the child's religious instruction led directly to her conclusion that Mr. Young should be prevented from discussing his religious ideas with the children. Having concluded that the religious conflict "must stop", she ordered it stopped in the manner which the custodial parent desired, without further discussion of whether another alternative might better serve the interests of the children.

This reasoning departs from the best interests of the child test as outlined earlier in these reasons in three respects. First, it places undue emphasis on the wishes of the custodial parent. The custodial parent is viewed as having the "right" to determine limits on access. In my respectful opinion, this rights-based approach is erroneous. The only question to be considered where limitation of access is in issue is what is in the best interests of the child. While the custodial parent has the obligation to make certain basic decisions as to how the child is educated, which may extend to religious matters, this does not automatically mean that religious contacts with the access parent of a different faith are to be excluded. It is not the wishes of the custodial parent that govern terms of access, but the best interests of the child.

Second, the trial judge failed to consider the benefits which might enure to the children from coming to know their father as he was -- that is, as a devoutly religious man devoted to the Jehovah's Witness faith. She made no reference to Parliament's instruction in s. 16(10) that the child shall have as much contact with both parents as is compatible with her best interests.

Thirdly, the trial judge failed to consider adequately whether there was any evidence of a risk of harm to the children which might offset the benefit of full access to their father's values, including those related to religion. While in some circumstances access may be limited on grounds unrelated to harm, in cases such as this, where the issue is whether entirely lawful discussions and activities between the access parent and the child should be curtailed, it behooves the judge to enquire whether the conduct poses a risk of harming the child. While the trial judge alluded to the possibility that if Mr. Young continued to share his religion with the children, he would damage his relationship with the elder two, she failed to allude to any other suggestion of harm and failed to consider whether such harm might outweigh the benefit to be gained from a freer, fuller relationship with their father.

The trial judge's undue emphasis on the "rights" of the custodial parent, coupled with her failure to consider the benefits to be gained from unrestricted contact with the access parent or whether those benefits were offset by a greater risk of harm to the children, may have clouded her appreciation of what was in their best interests. It was therefore open to the Court of Appeal to reconsider the matter and seek to give further guidance as to the factors to be considered in determining what is in the child's best interests in cases such as these.

Southin J.A. in the Court of Appeal took a similar but less absolute view of the rights conferred by custody. The custodial parent, according to her, has the "final say" on all matters of religion, although the "right to know" the access parent and the problems with the enforceability of any order restricting the conversation of the access parent, led her to conclude that the trial judge's restriction in respect of religious conversations could not stand.

The majority of the Court of Appeal, by contrast, saw the matter less in terms of the custodial parent's "right" to decide all religious matters affecting the children, and more in terms of the best interests of the children. Unfettered by notions of the custodial parent's rights, the majority of the Court of Appeal asked simply what was in the best interests of the child. Wood J.A. began by noting at p. 92 that "the provisions of s. 16 of the Divorce Act of 1985 must be applied in a manner which is both consistent with the intentions of Parliament and relevant to the evolving role of parents in the modern structure of today's society". He found s. 16(10) to be particularly important in determining Parliament's intent in respect of access, stating at p. 93:

Viewed in that way, real contact would necessarily include the opportunity for an access parent to whom a religious belief is important to share that belief, at least in a consensual way, with his or her children.

Thus, for the majority in the Court of Appeal, the prime concern was the best interests of the child. An order for custody vests the custodial parent with the power to determine the child's religious upbringing, to the extent that this parent can require the child to observe a faith until the age of discretion; this is in the child's best interests where the parents cannot agree. A custody order does not, however, give the custodial parent the "right" to limit the access parent's ability to share his or her religious views with the child, unless that is shown on the evidence not to be in the child's best interests. Viewed thus, the notions of custody and access unite in a common purpose of promoting the child's best interests.

The majority held that an access parent's conduct with his child may be limited where the evidence shows the existence of, or potential for, real harm to the child, or where the child does not consent to the instruction. On the latter point, Wood J.A. noted that while it may be in the best interests of the child that the custodial parent be allowed to enforce religious practices against the child's wishes, that rationale did not extend to the access parent. Both limitations on the right to access found by the Court of Appeal are grounded in the best interests of the child. The majority also held that expert evidence, based on scientific criteria, is required to show harm. While, for the reasons discussed earlier, I would not agree as a general proposition that in every case the risk of harm to the child must be established to justify limitations on access, in the context of the dispute in this case, I do not think that the majority of the Court of Appeal erred in placing considerable emphasis on the absence on the evidence of a risk of harm to the child.

I do not share the view of the majority of the Court of Appeal that expert evidence is required in all cases. Nor am I convinced that the failure of the child to consent to instruction necessarily precludes the conclusion that such instruction by the access parent is in the best interests of the child. Apart from these caveats, I substantially subscribe to the views expressed by Wood J.A.

The majority of the Court of Appeal held that the evidence did not establish that harm was being caused to the children. On the issue of the children's consent, Wood J.A. concluded that no order was necessary because Mr. Young had confirmed under oath that he would respect his children's wishes with respect to the activities they objected to -- attending services with him and accompanying him on his proselytizing missions. In the end, the majority did not find that it was in the best interests of the children, all factors considered, that even this limited form of religious activity by the access parent be restricted.

If one accepts, as did the majority of the Court of Appeal, that the issue of the children's accompanying Mr. Young to services and in his evangelical efforts is resolved by Mr. Young's undertaking -- and I see no reason not to do so given his record of compliance to date -- the only issue is whether the order forbidding Mr. Young to discuss religion with his children is valid. Apart from the value of getting to know their father and the difficulty of enforcement which led Southin J.A. to reject this restriction, it is questionable that a proper application of the "best interests of the child" test supports it. Conflict between the parents is, in and of itself, not a sufficient basis for assuming that the child's interests will not be served. There was, in this case, evidence that the children were functioning in an entirely normal fashion, and had not been adversely affected by the dispute between their parents or by their father's religious instruction. We are left then with the trial judge's concern that Mr. Young's relationship with his children would deteriorate if he persisted in his religious instruction during his periods of access.

With the greatest respect to the trial judge, this was not on the record in this case a sufficient reason for restricting access. In so far as the possible deterioration of Mr. Young's relationship with his daughters can be considered unfortunate, the alternative -- a relationship which prevents them from knowing him as he is -- is also undesirable. In these circumstances, any perceived harm to the children cannot be said to outweigh the benefits of unrestricted access. In short, this is not a case where the evidence supports the view that the best interests of the children requires further curtailment of the father's communication of religious views and practices than already had been agreed to.

I conclude that the trial judge's orders preventing Mr. Young from discussing religion with his children were not supportable on the evidence. In view of Mr. Young's undertakings, the orders relating to church attendance and proselytizing were unnecessary. The order enjoining the respondent from preventing blood transfusions was unnecessary from a practical point of view. Parents should, of course, not make disparaging comments about the other parent's religion, but the matter might, on the record here, best have been left to the parents' good sense.

B. Financial Matters

The trial judge found that each party initially had a one-half interest in the family home where the wife was living with the three children of the marriage. She held that the husband's half interest was extinguished by the costs ordered against him, half the value of the family car which he kept, arrears of maintenance, and lump sum maintenance over and above the monthly maintenance awarded to the wife. As the trial judge put it, at p. 202, "[a]ny further interest that the [husband] has in the matrimonial home will be transferred to the petitioner in the form of lump sum maintenance to assist in the petitioner's efforts to become self-sufficient."

The trial judge ordered that $80,000 owed to the wife's mother by a corporation, which the parties had formed to begin a jewellery business, was a family debt for which the husband and wife were jointly liable. She also held that the sum of $27,586.57 which the appellant had borrowed to support herself before she obtained an order for interim maintenance was a family debt for half of which the respondent was liable.

The trial judge ordered that the husband pay to the wife monthly maintenance in the sum of $1400, and $400 a month support for each of his children.

Two aspects of these orders are appealed to this Court. The first is the trial judge's determination that whatever interest the husband might have in the matrimonial home, after setting off costs and arrears of maintenance, should be absorbed by an order for lump sum maintenance. The Court of Appeal set this order aside. The power to award lump sum maintenance is found in the Divorce Act, s. 15(2):

. . .

(2) A court of competent jurisdiction may, on application by either or both spouses, make an order requiring one spouse to secure or pay, or to secure and pay, such lump sum or periodic sums, or such lump sum and periodic sums, as the court thinks reasonable for the support of

(a) the other spouse;

(b) any or all children of the marriage; or

Southin J.A. stated at p. 41:

If the learned judge wished to fix lump sum maintenance, she was obliged to fix it in a sum certain and to fix it with reference to the principles applicable to such an award.

I agree that she was obliged to fix the maintenance with reference to the principles applicable to such an award. The trial judge did not express any jurisprudential or evidentiary basis for awarding the lump sum maintenance. The goal was no doubt to achieve the end of conveying the entire interest in the matrimonial home to the wife. But this justification does not support an award of lump sum maintenance; more is required.

That said, the fact remains that the trial judge was convinced that a fair allocation of the assets of the marriage required that the house be transferred entirely to the wife. The British Columbia Family Relations Act, R.S.B.C. 1979, c. 121, permits an unequal division of family assets where the trial judge is satisfied that this is required having regard to various enumerated factors. The relevant sections are ss. 51 and 52 of this Act. Section 51 reads:

Where the provisions for division of property between spouses under section 43 or their marriage agreement, as the case may be, would be unfair having regard to

(a) the duration of the marriage;

(b)the duration of the period during which the spouses have lived separate and apart;

(d)the extent to which property was acquired by one spouse through inheritance or gift;

(e)the needs of each spouse to become or remain economically independent and self sufficient; or

(f)any other circumstances relating to the acquisition, preservation, maintenance, improvement or use of property or the capacity or liabilities of a spouse,

the Supreme Court, on application, may order that the property covered by section 43 or the marriage agreement, as the case may be, be divided into shares fixed by the court. Additionally or alternatively the court may order that other property not covered by section 43 or the marriage agreement, as the case may be, of one spouse be vested in the other spouse.

Section 52 of the Act provides that where a determination of an unequal allocation has been made, the trial judge may, by way of remedy, make an order as to the ownership of property.

In the case at bar there was ample basis for an order under s. 51 awarding Mrs. Young a greater portion of the family assets on the ground that, for a considerable period of time, Mr. Young had paid little or nothing for the support of his family. As will be seen below, I do not agree that the debts incurred during this period can, in law, be attributed to Mr. Young. However, they can serve as a consideration supporting reduction of his interest in the family property. It may also be noted that the equity in the home was not substantial and that Mr. Young was permitted to retain other assets. It is on these grounds that I depart from Southin J.A.'s view that there were insufficient reasons to order a reapportionment. Although the matter is not free from difficulty, I incline to the view that a trial judge, looking at all the circumstances, could have made such an order.

The problem, as noted by Southin J.A., is that the trial judge made no reference to s. 51 of the Family Relations Act or the factors on which reapportionment of the presumed equal interest in family assets can be made. This presents this Court with two alternatives: return the matter for a new trial or make the order which it thinks the trial judge should have made. The Court of Appeal chose the latter, and I think they were right in so doing. In this case it is open to inference that the trial judge concluded that a de facto reapportionment of the interest in the family assets was required to do justice. It also appears that there was, on the facts and the law, a method of achieving this end. The fact that the trial judge did not expressly allude to the factors for reallocation under s. 51, which in other circumstances might be sufficient reason for ordering a new trial, must here be weighed against the fact that this litigation has already gone on far too long and cost far too much. Like the Court of Appeal, I am loath to send the matter back to the trial division for further proceedings. In these circumstances, it seems to me that the better course is to endorse the result achieved by the trial judge on the ground that the evidence is capable of supporting an order for reallocation of the parties' interest in the family assets to the extent required to give Mrs. Young the entire interest in the matrimonial home.

I come finally to the matter of the family debts. The $80,000 owed by the corporation to Mrs. Young's mother cannot in law be characterized as a debt for which Mr. Young is personally liable, as ordered by the trial judge. The only entity liable for repayment is the corporation. There is no legal basis on which either Mr. or Mrs. Young can be said to be legally liable to repay the money borrowed by the company. The debt incurred by Mrs. Young to support herself and the children before she applied for maintenance is similarly unenforceable against Mr. Young as a debt, although it can be taken into consideration in an order for reduction of his interest in the family assets, as suggested above. I agree with Southin J.A., at p. 39, that "[t]he court cannot make a spouse jointly liable to a creditor for a debt of the other spouse, no matter for what purpose it was incurred, or, in the absence of some contractual foundation, make one spouse liable to indemnify the other, either in whole or in part, for a liability of the latter." Accordingly, the paragraphs of the order making Mr. Young liable for one-half of these debts must be struck out.

C. Costs

Costs Against the Respondent

The trial judge ordered solicitor-client costs against the respondent. This award was made on the basis that the custody claim had "little merit", that the respondent attempted to mislead the court, that the respondent was recalcitrant on matters of custody and maintenance and, finally, on the basis that unnecessary proceedings had resulted. The trial judge also referred to the fact that someone else was promoting and paying for the legal action and that repetitive and irrelevant evidence was tendered.

The Court of Appeal, per Cumming J.A., upheld the imposition of solicitor-client costs for four days of the trial and for four days of the interlocutory proceedings concerned with financial issues, on the basis of the husband's non-disclosure of financial information. Otherwise, costs against the respondent were reduced to party-and-party costs.

The Court of Appeal's order was based on the following principles, with which I agree. Solicitor-client costs are generally awarded only where there has been reprehensible, scandalous or outrageous conduct on the part of one of the parties. Accordingly, the fact that an application has little merit is no basis for awarding solicitor-client costs; nor is the fact that part of the cost of the litigation may have been paid for by others. The Court of Appeal meticulously considered all the proceedings in the light of these principles to arrive at its conclusion that only partial solicitor-client costs were justified.

Finding no error in the reasoning or conclusion of the Court of Appeal on this question, I conclude that its order for costs should remain, save to the extent different conclusions on the merits in this Court require that an adjustment be made. As I have made clear, the only respect in which I would vary the order of the Court of Appeal is that instead of ordering lump sum maintenance and a moratorium on the sale of the matrimonial home, I would restore the trial judge's order that the entire interest in the home be conferred on the wife. In my view, this difference does not warrant altering the award of costs against the respondent made below.

Costs Against the Respondent's Counsel

The trial judge ordered solicitor-client costs against counsel for the husband, Mr. How. For the reasons recited above in connection with costs against the respondent, she concluded that the proceedings had been unnecessarily lengthened. She also referred, at p. 216, to the fact that "[c]ounsel for the respondent had a forum and a cause to pursue. Unfortunately, what was in the best interests of the children, their welfare, was totally lost by the respondent and his counsel in these protracted proceedings. . . . The court was subjected to unwarranted abuse, criticism and insult." She made no finding, however, that Mr. How had been in contempt of court.

The Court of Appeal held that no order for costs should have been made against Mr. How. There is no need to repeat that entirely satisfactory analysis. The basic principle on which costs are awarded is as compensation for the successful party, not in order to punish a barrister. Any member of the legal profession might be subject to a compensatory order for costs if it is shown that repetitive and irrelevant material, and excessive motions and applications, characterized the proceedings in which they were involved, and that the lawyer acted in bad faith in encouraging this abuse and delay. It is clear that the courts possess jurisdiction to make such an award, often under statute and, in any event, as part of their inherent jurisdiction to control abuse of process and contempt of court. But the fault that might give rise to a costs award against Mr. How does not characterize these proceedings, despite their great length and acrimonious progress. Moreover, courts must be extremely cautious in awarding costs personally against a lawyer, given the duties upon a lawyer to guard confidentiality of instructions and to bring forward with courage even unpopular causes. A lawyer should not be placed in a situation where his or her fear of an adverse order of costs may conflict with these fundamental duties of his or her calling.

The Court of Appeal found that the trial judge's criticism of Mr. How related to his conduct in bringing the action. Assuming that costs might, in certain circumstances, be imposed for contempt of court, none was found. Accordingly, no order for costs should have been made against Mr. How. I see no error in the conclusion of the Court of Appeal in this regard.

Costs Against Burnaby Unit (Watch Tower Bible and Tract Society)

Since the Watch Tower Bible and Tract Society (the Society) did not appear as a party, the costs awarded against it must be taken to have been premised on the fact that it supported the litigation financially. In effect, this was equivalent to an award for the tort of maintenance: see Re Sturmer and Town of Beaverton (1912), 1912 588 (ON SCDC), 25 O.L.R. 566 (Div. Ct.), at pp. 568-69. To be liable for maintenance, a person must intervene "officiously or improperly": Goodman v. The King, 1939 21 (SCC), [1939] S.C.R. 446. Provision of financial assistance to a litigant by a non-party will not always constitute maintenance. Funding by a relative or out of charity must be distinguished from cases where a person wilfully and improperly stirs up litigation and strife: Newswander v. Giegerich (1907), 1907 33 (SCC), 39 S.C.R. 354.

In this case there was no evidence that the respondent had been induced to allow the Society to use his name in order that it might avoid liability for costs while advancing its own interests, that he would not have advanced his own interests in the absence of its help, that the Society's funding was for other than charitable motives, or that it controlled or directed the proceedings. Its support was "out of charity and religious sympathy" and did not constitute maintenance. The fact that the Society had a common interest with the respondent (as followers of the same religion) did not affect this.

Cumming J.A. so found, and then qualified his position at p. 85 by stating:

I hasten to add that it does not follow that the resources of the Watch Tower Bible & Tract Society can be brought to bear in every dispute between a Jehovah's Witness parent and a non-Jehovah's Witness parent. Once an issue of constitutional law of the kind raised here is settled then, if further litigation of the point between other litigants is supported, another question might arise. It may be that the right to assist without facing an award of costs cannot itself be used by the rich and powerful, no matter how great their interest in the issue, as an instrument of the oppression of those who must fight their battles alone.

I find again that no error has been made in the Court of Appeal's reasoning or conclusion. The evidence established that the respondent paid for a considerable portion of the cost of the proceedings personally, that the dispute was instigated at least in part by the appellant, and that the Society could not be considered to have stirred up the litigation, much less to have done so wilfully or improperly.

One argument, however, was not touched on by the Court of Appeal. This is the argument that the Society was an unnamed party to the litigation and, as such, should properly bear its portion of the costs. I would not discount the possibility that a court might properly hold an unnamed party liable for costs. The rule is that a non-party who has put forward another person in whose name the proceedings are taken cannot escape liability for costs in putting forward another: R. v. Sturmer and Town of Beaverton, supra. However, it seems to me that the evidence here falls short of establishing that the society was a party in this sense. Even on the constitutional issue, it cannot be said that the Society put Mr. Young forward, in effect bringing its own action in his name. The constitutional issue was first raised by Mrs. Young's objection to Mr. Young's communicating his religious beliefs to the children and was validly pursued by Mr. Young in his own interest. The Society's interest in the constitutional issue is insufficient, as I see it, to distinguish it from interveners who appear on constitutional cases and who have never been liable for costs.

Disposition

I would affirm the order of the Court of Appeal, except in respect of the matrimonial home, which should be transferred to Mrs. Young. In the circumstances of this case, I would order that each party bear his or her own costs on this appeal.

I would answer the constitutional questions as follows:

1.Do ss. 16(8) and 17(5) of the Divorce Act, 1985, which provide that judicial decisions regarding custody and access be made "in the best interests of the child", deny the rights and freedoms guaranteed in s. 2(a), (b), and (d) of the Canadian Charter of Rights and Freedoms?

No.

2.If the answer to question 1 is affirmative, are ss. 16(8) and 17(5) of the Divorce Act, 1985, justified as reasonable limits by s. 1 of the Canadian Charter of Rights and Freedoms and therefore not inconsistent with the Constitution Act, 1982?

Given my answer to question 1, it is unnecessary that I answer this question.

3.Do ss. 16(8) and 17(5) of the Divorce Act, 1985, violate the guarantees to equality set out in s. 15 of the Canadian Charter of Rights and Freedoms?

No.

4.If the answer to question 3 is affirmative, are ss. 16(8) and 17(5) of the Divorce Act, 1985, justified as reasonable limits by s. 1 of the Canadian Charter of Rights and Freedoms and therefore not inconsistent with the Constitution Act, 1982?

Given my answer to question 3, it is unnecessary that I answer this question.

Appeal allowed in part, L'Heureux-Dubé J. dissenting in the result. Sections 16(8) and 17(5) of the Divorce Act, 1985 did not violate ss. 2(a), (b), (d) or 15(1) of the Canadian Charter of Rights and Freedoms.

Solicitors for the appellant: MacLean, Nicol & Wong, Vancouver.

Solicitors for the respondent James K. C. Young: W. Glen How & Associates, Halton Hills (Georgetown), Ontario.

Solicitors for the respondent W. Glen How: Douglas, Symes and Brissenden, Vancouver.

Solicitors for the respondent Watch Tower Bible and Tract Society of Canada: W. Glen How & Associates, Halton Hills (Georgetown), Ontario.

Solicitor for the intervener the Attorney General of Canada: The Attorney General of Canada, Ottawa.

Solicitor for the intervener the Attorney General for Ontario: The Attorney General for Ontario, Toronto.

Solicitor for the intervener the Attorney General of Quebec: The Attorney General of Quebec, Ste‑Foy.

Solicitor for the intervener the Attorney General of Manitoba: The Attorney General of Manitoba, Winnipeg.

Solicitor for the intervener the Attorney General of British Columbia: The Attorney General of British Columbia, Victoria.

Solicitors for the intervener the Law Society of British Columbia: Blake, Cassels & Graydon, Vancouver.

Solicitors for the intervener the Seventh‑day Adventist Church in Canada: Milner, Fenerty, Edmonton.

See Erratum [1995] 1 S.C.R. iv

Young v. Young, [1993] 4 S.C.R. 3

Case Summary

Young v. Young, [1993] 4 S.C.R. 3

Judges

Counsel

Parties

Areas of Law

Fact Patterns

Expert Witnesses

Issues

Statutory Provisions 9

Legal Tests 2

Legislation 3