(dissenting).
I respectfully dissent. In order to affirm, the majority has discovered and applied a legal theory that was not considered by the trial court and was not argued, or even suggested, by any party. The majority’s holding cuts new ground. It ties the pro tanto credit, not against the comparative fault judgment, but against the jury’s finding of Jamie-son’s total damages. This seems a plausible, though not an inevitable, conclusion. The $10,000 judgment already reflects the net due Jamieson under the comparative fault Act. An argument might be made that the majority’s theory would make more sense only if applied in the converse situation, where a comparative fault settlement is being claimed as a pro tanto credit in a dramshop suit.
I would not decide the question. My disagreement with the majority is grounded in the belief that the theory was inappropriate for consideration. Can a litigant inject a new legal theory at the appellate level? Apparently the answer is yes, but only if its purpose is to affirm, not reverse, the trial court. Johnston Equip. Corp. v. Industrial Indem., 489 N.W.2d 13, 16-17 (Iowa 1992); City of Marion v. National Casualty Co., 431 N.W.2d 370, 374 (Iowa 1988). See generally 5 C.J.S. Appeal and Error § 715 (1993); 5 Am.Jur.2d Appeal and Error § 727 (1962).
Is it appropriate for this court, rather than the litigants, to inject the new theory? The answer, at least under our more recent cases, is that it can be appropriate. Kramersmeier v. R.G. Dickinson & Co., 440 N.W.2d 873, 876 (Iowa 1989); Kelly v. Iowa Valley Mut. Ins. Ass’n, 332 N.W.2d 330, 333 (Iowa 1983).
These cases should not cause us to independently suggest and apply the new theory in the present case. Only in rare circumstances should we decide appeals on theories not suggested by litigants in their briefs.1 Certainly we should not do so with a theory that is reasonably subject to debate so that the outcome might be changed if both sides were to be briefed and argued in the customary manner.
I consider the majority’s answer to the controlling question to be debatable. I would prefer to accord Harrison’s counsel the opportunity to at least argue the opposite side. Harrison’s counsel will no doubt be astonished to have prevailed on every point briefed and argued, only to lose because of the ramification injected by this court’s majority.
Harrison contended for a pro tanto credit. In resisting, Jamieson limited his argument *784in district court and on appeal to a claim that he should prevail by applying the proportionate credit rule. Harrison was right; the pro tanto rule applies and the proportional credit rule (or some suggested application of it) does not. This was the end of the dispute as counsel framed it upon submission of the appeal to us. I would save for a case in which it is briefed the question whether Ja-mieson can nevertheless be rescued because the pro tanto rule can be applied in a way he never suggested. The case should be reversed.
McGiverin, C.J., joins this dissent.
. To all other uncertainties faced by litigants who must choose whether to invest in an appeal — or invest in defending against one — there should not be added the fear that this court is highly likely to inject a legal theory of its own choosing, and decide the case on that theory without benefit of the parties’ briefs or arguments. It would be far more prudent, and eminently more fair, to leave it to counsel to frame the issues and for us usually to limit the decision to those issues presented.