concurring in part, dissenting in part.
I agree that the 2001 amendments to ORS 109.119 apply to this case. Where I part company with the lead opinion is its decision to apply those amendments to this case for *27the first time on appeal without remanding to give the parties an opportunity to develop a record that addresses the requirements of the new law. In my view, we should remand the case to the trial court if not out of considerations of due process then at least out of considerations of basic fairness.
In Jones v. General Motors Corp., 139 Or App 244, 264, 911 P2d 1243 (1996), aff'd. on other grounds, 325 Or 404, 939 P2d 608 (1997), we declined to apply for the first time on appeal amendments to the summary judgment standard provided in ORCP 47 C. We explained that “[t]he goal posts have moved. * * * Due process requires that plaintiff be given an opportunity * * * to meet the new evidentiary burdens imposed by the amendment * * 139 Or App at 264; see also State v. Meyers, 153 Or App 551, 559-60, 958 P2d 187 (1998) (declining to apply new evidentiary standard for first time on appeal because “[s]uch a moving of the proverbial goal posts after the contest is over raises serious questions of due process”). In this case, the goal posts certainly have moved from the time that this case was tried. At the time this case was tried, the United States Supreme Court had not even issued its decision in Troxel v. Granville, 530 US 57, 120 S Ct 2054, 147 L Ed 2d 49 (2000), and the 2001 amendments to ORS 109.119 were still a year off. Under Jones, it strikes me that we have an obligation to remand the case.
Having said that, I will confess that I am not certain that Jones was correct in asserting that a remand in such circumstances is required as a matter of federal due process principles. I am not aware of any United States Supreme Court decision holding that the retroactive application of a nonpenal statute violates the Due Process Clause of the Fourteenth Amendment. Moreover, even if Jones is correct in the abstract, it is not clear that it would apply in a case such as this, in which it could be argued that grandparents do not have the sort of liberty interest that triggers due process procedural protections. See, e.g., Graham v. Children Services Division, 39 Or App 27, 30-34, 591 P2d 375, rev den, 286 Or 521 (1979) (“grandparents have no liberty interest * * * and no rights superior to a nonrelative applying for permission to adopt”).
*28Even if federal due process principles do not require a remand, however, we have discretion to remand based on considerations of basic fairness. In that regard, I note that it is not uncommon for the Oregon Supreme Court to remand domestic relations cases for taking additional evidence after the court announces a new principle of law or when it merely finds a record inadequate. In Massee and Massee, 328 Or 195, 970 P2d 1203 (1999), for example, the Supreme Court determined that the presumption of equal contribution to marital assets applies to the appreciation of assets that one spouse brings into the marriage. The question then arose whether the husband in that case actually had rebutted the presumption. In that regard, the court noted that the record was less than adequate, particularly with respect to the wife’s contributions as a homemaker. The court explained that it could resolve the case simply by ruling de novo on the record before it or by remanding it to the Court of Appeals to do the same. But the court rejected both of those approaches, explaining:
“The record in this case lacks evidence on several important facts, including husband’s net worth at the time of marriage and at the time of dissolution, which likely will influence, if not control, any ultimate property division after remand. Resolution of this case thus requires further development of the factual record. Consequently, we remand the case to the trial court for the taking of additional evidence and reconsideration in light of this opinion.”
Id. at 209.
Similarly—and closer to home—in State ex rel Juv. Dept. v. Lauffenberger, 308 Or 159, 777 P2d 954 (1989), the Supreme Court addressed whether the juvenile court erred in applying a “best interests of the child” standard in deciding whether to place a child with the father or with third parties. During the pendency of the appeal, the court issued its decision in Hruby and Hruby, 304 Or 500, 510, 748 P2d 57 (1987), in which it held that, in private custody disputes, in the absence of a “compelling reason” for placing a child with another person, the natural parents have the right to the care and custody of the child. In Lauffenberger, the court then held that the “compelling reason” standard announced in Hruby also applied to juvenile court custody decisions. Lauffenberger, 308 Or at 164-66. The court did not simply *29review the juvenile court’s decision in light of the new standard, however. It remanded to the juvenile court to develop the record under the standard required by Hruby:
“Deciding the case before Hruby, the [juvenile] court made no specific findings with respect to potential harm from parental custody that Hruby called for in ordinary private custody disputes nor with respect to any other factors that might play a role under the juvenile court statute. * * *
“Under the circumstances, * * * we do not order a change of custody. That wrill be for the circuit court to decide on remand and on a new and more complete record, if the father still desires custody.”
Lauffenberger, 308 Or at 167-68.
In light of decisions such as Massee and Lauffenberger, I see no reason, as a matter of simple fairness, not to give the parties to this case an opportunity to develop a record that responds to the changes in the law that have occurred since the original trial. As I have noted, when this case was tried, Troxel had not yet been decided. By the time the trial court issued its letter opinion, the decision had been issued, and the court even mentioned the case in two footnotes. But the fact remains that the parties did not have the opportunity to develop a record that directly responds to the presumption that the Troxel decision established.
Even assuming that Troxel was fairly before the parties at trial, the fact remains that the legislature did not enact the amendments to the statute that the lead opinion now applies until a year later. Thus, in this case, the lead opinion reverses the trial court because the record that the grandparents developed at trial failed to satisfy a statutory standard that did not even exist at the time. To me, that is fundamentally unfair.
I suppose that it could be argued that, if the statute merely “codifies” the Troxel decision, the parties would be in no position to complain that they could not have adequately developed the record. But the argument rests on an assumption that the statute has such a limited effect. That assumption is debatable, to say the least. And we have not yet had occasion to decide the matter. Indeed, to determine whether *30the assumption is well founded would require us, among other things, to interpret the new statute, determine its constitutionality, address the standard of proof that it prescribes, and articulate its relationship to Troxel — all that simply to decide whether to remand the case. It strikes me that the better use of our resources is to address those questions as they arise, and are briefed, in concrete cases.
The lead opinion acknowledges that there has been a change in the legal standard during the course of this case but insists that there is no reason to remand for essentially two reasons.
First, the lead opinion complains that grandparents failed to request the trial court to reopen the record when Troxel first was published. Grandparents prevailed at trial, however. Under those circumstances, it is not clear to me why it was incumbent on grandparents to seek to reopen the record. In any event, the legislature did not amend ORS 109.119 until a year later, long after any party could have asked the trial court to reopen the record. It is that new statute that the lead opinion invokes in reaching its decision in this case.
Second, the lead opinion complains that grandparents do not identify the specific evidence that they wish to add to the record. That is true enough, but of little consequence. In Massee and Lauffenberger, the parties did not even request remands, and the Supreme Court gave no indication that such requests are necessary, much less that parties are required to identify the evidence that needs to be added to the record. In this case, grandparents have made clear that they were not given an opportunity to develop a record with the criteria of the amended statute in mind and wish to be given that opportunity. That seems sufficient to me. Indeed, given the fact that this was such a close case, and given the importance of credibility in evaluating the record, it strikes me that a remand would be appropriate even without the need for additional evidence.
The concurrence similarly argues that a party seeking a remand must identify the sort of evidence he or she would offer on remand. I am aware of no case imposing such *31a requirement, and, as I have noted, Massee and Lauffenberger appear to be quite to the contrary.
In short, we should not decide cases on the basis of standards that did not even exist at the time of trial. We should remand this case as grandparents request so that they may develop a record that responds to Troxel and to the specific demands of the 2001 amendments to ORS 109.119.
Deits, C. J., and Linder, Wollheim, and Brewer, JJ., join in this opinion.