dissenting. I believe that any fair observer of our child custody opinions, both majority and dissent, would conclude that there is a significant gap between the standard of review we say we employ, and our actions in resolving the cases before us. I fear that gap is growing, and we are moving much closer to de novo review than we are willing to admit. The two cases in which this dissent is filed are, in my judgment, examples of this trend. Although the cases are different, and the issues are different, they share one common element — that is, *457the family court decisions would be affirmed if we actually employed. the standard of review set out in the beginning of the majority opinions . and in our precedents. Thus, I am filing this common dissent to say that I believe the direction in which we are going in appellate review of custody decisions is wrong.
I acknowledge that the pressures that exist to abandon or “fudge” deferential standards of review in custody decisions are real. Custody decisions are critically important to the child or children who will grow up, good or bad, in the new custodial situation. The interests of children in their family and home environment may be the most important interests we seek to protect in our judicial system. Unlike in many areas of the law, most of us have life experiences we draw on in facing custody questions. We are, accordingly, more likely to abandon a deferential posture to rely upon our own knowledge and experience. Finally, the consequences of excessive discretion in custody determinations are troubling, a subject of much scholarly interest in recent years. See generally C. Schneider, Discretion, Rules, and Law: Child Custody and the UMDA’s Best-Interest Standard, 89 Mich. L. Rev. 2215 (1991).
The effect of these pressures is to elevate the desire to produce the “right” answer for the child in the case before us over our normal appellate co-goal of producing fair and predictable rules of law to guide future cases. If the result were to achieve a balance between predictable and fair rules, on the one hand, and judicial discretion on the other hand, I would be less concerned. I think, however, the real result is to substitute our discretion and individual judgment for that of the family court in pursuit of the “right” outcome for the case before us. By this process we are squaring discretion, not containing it.
I can think of no area where the need to contain the exercise of appellate discretion is greater. I won’t rehash the general reasons for deferential' standards of review based largely on the fact that the family court, often aided by a guardian ad litem and professional evaluators, saw and heard the parties, particularly the parents, and we are dealing solely with transcripts. Those reasons should give us pause, but there are additional reasons almost unique to child custody litigation. We know from numerous studies that custody litigation has a tremendous adverse impact on the children who are the subject of that litigation. See A. Schepard, Parental Conflict Prevention Programs and the Unified Family Court: A Public Health Perspective, 32 Fam. L.Q. 95, 102-06 (1998); E. Brandt, The Challenge to Rural States of Procedural Reform in High Conflict Custody Cases, 22 U. Ark. *458Little Rock L. Rev. 357,359-60 (2000). Whatever order the court issues as a result of that litigation, the destructive impact of the litigation itself, and the accompanying adversary contentiousness of the parents, ihay leave the greatest mark on the growth and development of the child. In the cases before us, unless the parties settle after our decision, that litigation will occur at least three times — twice in the family court and once in this Court. I seriously doubt that there is any longer a “right” answer, even if we can discover it. The real need is to stop the contentious litigation as soon as possible, not to discover a better custody order.
Unfortunately, our decisions breed further appeals. See G. Crippen; The Abundance of Family Law Appeals: Too Much of a Good Thing?, 26 Fam. L.Q. 85, 100-01 (1992). There is no predictable rule of law in either of the majority decisions in the cases before us, except with respect to considering the age of a prospective custodial parent and, even there, it is unclear what is the holding of the Court. What there is, instead, is a clear indication that the Supreme Court will substitute its judgment for that of the family court. Thus, the message to any parent who has lost a custody case is to try an appeal to this Court, which may weigh the relevant factors differently.
There are, I believe, three main ways in which the decisions in Cloutier v. Blowers and Spaulding v. Butler are inconsistent with our proper limited role in custody appeals. I discuss my reasons for dissenting from the majority decisions under the headings below.
I. Adoption of Inappropriate Rules
Custody determinations are now governed by statute, 15 V.S.A. § 665. That statute requires that the family court be guided by the best interest of the child, id. § 665(b), and sets out nonexclusive factors in determining the best interest of the child. Because the list is nonexclusive, the court may consider other factors bearing on the best interest of the child. Hansen v. Hansen, 151 Vt. 506, 508, 562 A.2d 1051,1053 (1989). While requiring that the family court consider all the statutory factors, if relevant, the Legislature has prohibited the court from establishing a preference based on the sex of the child, the sex of the parent or the financial resources of a parent. 15 V.S.A. § 665(c). The statute, however, contains no authorization for this Court to add categorical rules that restrict the trial court from determining the best interest of the child. In Cloutier v. Blowers, the majority has done exactly that, interfering with the proper and necessary discretion of the family court.
*459In Cloutier, the family court held that because all other factors were in balance, it had to give critical weight to the age of the parents who sought custody. The majority rejects this approach although its ground is unclear. It holds either that (1) the relationship between the age of the proposed custodian and the best interest of the child can be considered only if there is evidence, presumably expert evidence, to support such consideration; or (2) the age of the custodian may not be considered because it results in discrimination based on age.
The first alternative is contrary to our precedents, intended to support the discretion of family court judges. In Harris v. Harris, 149 Vt. 410, 546 A.2d 208 (1988), the mother, who did not prevail in the custody dispute in the trial court, argued that the trial judge could not consider that she was living out of wedlock with a man, without expert testimony to show the effect on the best interest of the child. We rejected that argument as follows:
While the expert testimony would have been helpful in this case, we agree with the trial court that the evidence fell in an area where the court could evaluate it without expert testimony. Such evaluation was expected under the language of § 665(b)(7). We concur with the Supreme Court of Kentucky which, facing a similar statute and similar evidence, said:
A trial judge has a broad discretion in determining what is in the best interests of children when he makes a determination as to custody. In many instances he will be able to draw upon his own common sense, his experience in life, and the common experience of mankind and be able to reach a reasoned judgment concerning the likelihood that certain conduct or environment will adversely affect children. It does not take a child psychologist or a social worker to recognize that exposure of children to neglect or abuse in many forms is likely to affect them adversely. Many kinds of neglect or abuse or exposure to unwholesome environment speak for themselves, and the proof of the neglect or abuse or exposure is in itself sufficient to permit a conclusion that its continuation would adversely affect children.
We also think the trial court is not precluded from consideration of circumstances where the neglect, abuse, or environment has not yet adversely affected the *460children but which, in his discretion, will adversely affect them if permitted to continue. In other words, a judge is not required to wait until the children have already been harmed before he can give consideration to the conduct causing the harm.
Krug v. Krug, 647 S.W.2d 790, 793 (Ky. 1983). Accordingly, we hold that the trial court did not err in accepting the evidence and relying on it in the custody determination.
Id. at 416-17, 546 A.2d at 212-13. We have reiterated the right of the family court judge to use common sense and common and life experience in making custody determinations. See Payrits v. Payrits, 171 Vt. 50, 53, 757 A.2d 469, 472 (2000); Bissonette v. Gambrel, 152 Vt. 67, 70, 564 A.2d 600, 601 (1989).
Ironically, we applied exactly this principle to the issue of whether the court could consider the age of the proposed custodian in making a child custody decision. In Miles v. Farnsworth, 121 Vt. 491, 494-95,160 A.2d 759, 761 (1960), we held that the trial court could consider the “infirmatives of advanced years” of the custodian as a factor in determining the custody of the child.
This is an exceptional case. We can take judicial notice that the average life .expectancy of an American male is 74 years.* Centers for Disease Control and Prevention, United States Life Tables, 1998, National Vital Statistics Reports, Feb. 7, 2001, at 2. Thus, in this case, there is a substantial chance that father will be unable to provide guidance to the child up until he reaches the age of majority of 18 years. As the majority of courts, including this Court, have held, see Phelps v. Phelps, 446 S.E.2d 17, 22 (N.C. 1994) (‘We conclude that a trial court should ... be allowed to consider a parent’s age and its potential effect on the welfare of the child as a factor in its determination of what is in the best interest of the child.”); Albright v. Albright, 437 So. 2d 1003, 1005 (Miss. 1983) (court may consider numerous factors including the “physical and mental health and age of the parents”); Ex parte Devine, 398 So. 2d 686, 696 (Ala. 1981) (court should consider numerous factors including “the characteristics of those seeking custody, including age, character, stability, mental and physical health”), the family court should be able to consider the age of a prospective custodian as it bears on the best interests of the child.
*461If the majority is intending to compromise on the consideration of age by requiring expert testimony, I think the compromise is unwise. Custody litigation is already extraordinarily expensive for all but the most wealthy parents. Adding more complexity and cost to such litigation is a step in the wrong direction.
The second alternative — that it is always unfair and discriminatory to consider the age of a parent in determining custody — is inconsistent with the fundamental policy of custody adjudication as emphasized in the majority opinion. On another issue, the majority points out that the family court must act based on the best interests of the child, not the interest of the parent. On this issue, however, the majority is violating that principle, acting on the interest of the father even though it is inconsistent with the best interest of the child. Although a few isolated precedents support this approach, the vast majority of decisions hold that the court can consider the age of a prospective custodian in making a custody award where it is in the best interest of the child to do so, as I have set out above. That is the holding of Miles v. Farnsworth. We should reject the categorical rule the majority apparently espouses.
I reiterate my point that categorical rules that restrict family court discretion in determining the best interest of the child are not good policy. Whatever the reasons espoused for them, fundamentally they involve substituting our judgment for that of the family court. In this unusual case where consideration of the remainder of the statutory factors produced no clear choice of custodian, consideration of the age of a parent should be in the court’s discretion.
II. Placing the Family Court Decision in a Bad Light
The Cloutier decision has another aspect that is an indication that we are not seriously applying the limited standard of review that binds us. In reviewing findings of fact, we must “view them in the light most favorable to the prevailing party.” Stickney v. Stickney, 170 Vt. 547, 548, 742 A.2d 1228, 1230 (1999) (mem.). A party challenging conclusions of law must overcome the “great deference” we give to the court’s conclusions, and we must “make all reasonable inferences in support of the court’s judgment.” Bevins v. King, 147 Vt. 203, 206, 514 A.2d 1044, 1046 (1986). Put another way, this Court must “construe [findings] so as to support the judgment, if possible.” Armstrong v. Hanover Ins. Co., 130 Vt. 182, 185, 289 A.2d 669, 671 (1972).
I find no indication in the majority opinion that it gave deference to the family court’s conclusions or that it construed the findings to *462support the judgment. Rather than fairly interpreting the family court findings and rationale in support of its decision, the majority recharacterizes the decision in a way that makes it virtually indefensible. In doing so, it raises a new issue not considered by the family court or argued by the parties. We now know how the majority would have decided the case had it been the family court. We have, however, done a disservice to the parties, the child and the trial court by introducing our issue on appeal.
Although the family court decided the case primarily on the relative age of the parents, it found that at least one additional factor favored the mother as custodian, as follows:
While either could discharge the role of parent adequately, the mother shows more promise for long term stability and consistency of caretaking. The mother has shown she can capably provide for the child’s broad breadth of needs. She has successfully raised a family. She has endured and overcome one of life’s most poignant tragedies with the death of a child. She has focused her life and emotional resources to successfully raise this child.
This emotional investment to this child is so strong that any attempt to deny her the primary role of caretaker would destroy her. Her emotional attachment is not unreasonable. Her need to raise this child should be encouraged for both her good and the child’s good.
Should she lose the primary role of caretaker, this would adversely impact on the child. There is no need of the child’s met by that drastic result. The child can have the best of both parents as the situation permits. Perhaps, they will become more comfortable with each other as parents, now that fear of removal of the child from the mother is gone.
In one page of his brief, the father attacked this additional consideration as focusing on the mother’s needs and not the best interest of the child. I discuss that below; for now, the point is that the father made no additional claims about the court’s rationale.
The majority found this rationale to be inadequate and unsupported by the findings, at least as to the factor contained in § 665(b)(5). The Court describes the family court’s rationale as follows: (1) mother and father, who were not married, decided to have a child to fill a void in mother’s life created by the tragic death of one of her children; (2) mother expected that the father would want little role in the child’s *463life; (3) to mother’s frustration, father took a very active role with the child and attempted to provide for both the child and the mother; (4) as a result, mother accused father of all sorts of misconduct during the custody case, but the family court failed to find that the misconduct had occurred; and (5) mother made the accusations to minimize the role of father in the child’s life. Once it has described the family court’s rationale in this way, it is no surprise that the majority concluded it did not support a finding that mother as a custodian would better foster for the child a positive relationship and frequent contact with the father than father as custodian would foster with the mother. Indeed, it has recast the family court’s rationale so it supports custody in the father, not the mother. Consistent with its description of the family court’s rationale, the majority has added that the family court should have pursued whether mother intended to alienate father from the child through her allegations.
One can find pieces of the majority’s rationale in the family court decision, but not in the way the majority has put them together. For example, the family court rejected mother’s allegations that father abused her and misused drugs and speculated that she might have made such accusations to minimize his role in the child’s life. It never suggested that the mother was trying to alienate the child from the father. Indeed, it found that the parents had successfully co-parented in the past and tried hard to induce them to develop a new co-parenting arrangement. This case is a far cry from Renaud, v. Renaud, 168 Vt. 306, 309, 721 A.2d 463, 465 (1998), in which the family court expressly found that “mother had undermined the child’s relationship with father by filing excessive and baseless abuse allegations.” We are not faithful to the limited standard of review if we construe the family court decision in a way to make it least defensible.
If the family court made any mistake here to induce the majority’s response, it was a labeling mistake. The family court’s rationale better fits factor three, 15 V.S.A. § 665(b)(3), or an independent factor, than factor five, id. § 665(b)(5). Under the proper standard of review, a labeling mistake does not justify the majority’s response of creating a new and indefensible analysis. '
We do, I believe, have to address directly the argument that father did make on appeal — that the family court rendered its decision based on the interests of the mother, rather than the interests of the child. I again stress that this is an exceptional case. The parties, who ■ never expected to marry, planned this child to fill a void in mother’s life caused by the death of another child. Thus, in explaining its rationale, *464the family court stressed the effect an adverse custody decision would have on the mother. If the court had gone no further, I would agree with appellant father that we could not affirm the court’s conclusion.
But the family court did go further and relate the mother’s loss to the effect on the child. Essentially, the court concluded that making the father the primary custodian would have such an effect on the mother that it would destroy her ability to be an effective parent, adversely affecting the child. It concluded that the reverse custody situation would not have this effect. Its reasoning is focused primarily on the effect of its decision on the best interests of the child. In these circumstances, I would affirm the family court’s conclusion as consistent with the statutory mandate of § 665(b) under our deferential standard of review.
III. Reevaluation of the Relevant Factors
It is impossible to read the trial court and majority decisions in Spaulding v. Butler without concluding that these decisions simply give different weight to the various factors bearing on custody and, as a result, reach different conclusions. This would be entirely understandable and healthy if the decisions came from two different trial judges. It is not appropriate if one of the decisions comes from an appellate court purportedly issued under a limited standard of review.
In two respects, the circumstances of Spaulding, and the family court decision, make it more likely that this Court will abandon its limited role and decide the case de novo. First, the family court faced a choice between two flawed parents. For either, it is easy to state why that parent should not be a primary custodian; it is far harder to find and weigh positive skills and conduct which warrant confidence that a child in that parent’s custody will have positive and nurturing parenting in a safe and secure environment. Unfortunately, there is no third option; the trial judge was forced to make a disquieting and unpleasant choice, almost on the basis of the least damage to the child.
Second, the trial judge, to her credit, did not sugarcoat the negative history, skills, characteristics or motives of either parent. I say “to her credit” because the decision reinforces that the court had no illusions that it could assure the child a good home and that the court struggled to find the best result. But that result has become harder to affirm because the flaws of the custodial parent, as explicitly contained in the findings, are so difficult to accept.
Indeed, on the surface, it appears much easier to accept the majority analysis because it details the father’s flaws and largely ignores, or *465explains, the mother’s flaws. In a truncated fashion, the majority quotes the family court’s analysis of the mother’s flaws in its decision on whether there were changed circumstances and then largely ignores that analysis in looking at the best interest of the child. Let me repeat, in full version, what the family court found about the negative factors bearing on the mother as a custodian:
Specifically, during [the] time [between September 1994 and April 1997] there was a pattern of poor care of a severe and painful diaper rash, the severity of which was unnecessary as shown by the improvement in the rash during the weekend visitations with Mr. Spaulding; a pattern of Nathan appearing for visits with Mr. Spaulding with a number of bruises and marks on his body; a lead test indicating lead poisoning of Nathan with no information that the problem had been taken care of; a pattern of biting behavior between Michael and Nathan that had not improved despite the work of Ms. Butler with Meredith McCartney in therapy; delays in Nathan’s developmental milestones as indicated by testing through Triple E and Stepping Stones; and severe bite marks on Nathan from Michael on April 11, 1997. . . . [The evidence] shows a child living in an environment with a pattern of neglect sufficient to result in measurable developmental delays and physical injuries. The change is substantial in that it resulted in Nathan lagging behind in his development and suffering painful personal injury despite remedial efforts. It is unanticipated in that one would never expect parental care to result in such an impact on a child.
Thus, the family court was forced to choose between a parent who exhibited “a pattern of neglect sufficient to result in measurable developmental delays and physical injuries” and a parent who had a history of abuse and attempts to alienate the child from the other parent, at best an unenviable choice.
Where the trial court decision is candid and balanced about the nature and difficulty of this choice, I cannot say the same about the majority decision here. It builds to its ultimate conclusion that the family court could not find within its discretion that it was in the child’s best interest to award custody to a father who “was a batterer, a liar, and . . . [who] consciously and deliberately alienated Nathan from mother.” 172 Vt. at 481, 782 A.2d at 1178. The equivalent criticism of the majority is that it cannot find as a matter of law that *466custody must be awarded to a mother who, when she had custody, engaged in “a pattern of neglect sufficient to result in measurable development delays and physical injuries.” I think we trivialize the difficulty and complexity of child custody adjudication with this kind of analysis.
The essence of this dissent is that the kind of difficult choice presented by Spaulding must be made by the judge who heard the evidence and viewed the parents as they testified and otherwise participated in the merits hearing. To the extent we have an appellate role, we should exercise it sparingly, and not as we are doing here, to second-guess the considered choice of the family court judge and substitute our own judgment. Because that is the fundamental point of the dissent and captures what I believe is wrong in the majority decision in Spaulding, I will not belabor my specific disagreements with that decision beyond three additional points.
I do not understand the remand in this case, other than for the evaluation of changed circumstances. The Court has said that the family court could not reach the decision it did based on its findings and that the findings are supported by the evidence. Under that analysis, the family court has no choice but to award custody to the mother. To the extent the majority is trying to suggest that it did anything other than substituting its judgment for the trial judge, that suggestion is illusory. Indeed, the worst outcome we could have for the child is further extensive litigation to reopen the findings and conclusions of the family court.
A good deal of the majority’s analysis is based on its conclusion that any risk that the child would be unsafe if placed with the mother “is speculative at best.” Id. at 481, 782 A.2d at 1178. I can describe that conclusion only as incredible. The family court detailed the past harm to the child while in mother’s custody and concluded that there had been “a pattern of neglect sufficient to result in . . . physical injuries.” How is it speculative that the identical pattern will reoccur when custody is again transferred to the mother?
Third, we need to be careful in how we define parental alienation and in administering a rule that a parent cannot benefit from alienating a child from the other parent. No parent who believes that the other parent is the cause of physical, sexual or extreme emotional abuse of the child is promoting contact between the child and that parent. The family court justifiably criticized the father for being too quick to go to the police, SRS or the courts, but some of father’s allegations of abuse were confirmed in earlier court proceedings and *467formed the basis of the abuse prevention order transferring custody to father and limiting mother to supervised visitation. Unless the majority is prepared to say that father somehow defrauded the family court, a conclusion with no support in the findings, it is inappropriate to call a court-ordered custody situation a “successful' attemptf] at alienation.” Id. at 480, 782 A.2d at 1177.
In conclusion, I believe that in both Spaulding and Cloutier the majority has embarked on the pursuit of a more just and correct custody determination that will injure the exact children it is attempting to protect. If we actually accorded the family court “broad discretion in determining the best interests of the child,” as both decisions state we do, the only consistent conclusion would be to affirm the custody decisions in both cases. Reluctantly, I must dissent from the reversals in both cases.
I am authorized to state that Justice Morse joins in this dissent.
Having reached 59 years of age, father’s remaining life expectancy is approximately twenty more years. Centers for Disease Control and Prevention, United States Life Tables, 1998, National Vital Statistics Reports, Feb. 7,2001, at 2.