concurring specialty-
I concur in the results.
An established saying, source unknown, is: “The exception proves the rule.” “Proves here does not mean confirms but tests; that which does not conform to the rule compels us to examine the rule.” John Wilson: The Cheats, “To the Reader” (1664).
When a trial court makes findings of fact wholly from documentary evidence, that part of Rule 52(a), N.D.R.Civ.P., which says, “and due regard shall be given [by appeals courts] to the opportunity of the trial court to judge of [sic] the credibility of the witnesses” cannot, under ordinary rules of grammar, be interpreted to mean that when a trial court has no opportunity to judge the credibility of witnesses, the appeals courts should then give “no” regard to the findings of fact which the trial court made.
This perceived exception to the rule turns out to be no exception at all. There is nothing in the rule which implies that “findings of fact shall not be set aside unless clearly erroneous if the trial court had an opportunity to judge credibility of witnesses.” Justice does not compel that we create an exception as a legal fiction.
Dolajak v. State Auto. & Cas. Underwriters, 252 N.W.2d 180 (N.D.1977), was a unanimous opinion which I signed without reluctance. I have cited it without reluctance in my concurring opinion in Matter of Estate of Koch, 259 N.W.2d 655, 664 (N.D.1977); and in the unanimous opinions which I authored, Butts Feed Lots v. Board of Cty. Commissioners, 261 N.W.2d 667, 669 (N.D.1977); Malarchick v. Pierce, 264 N.W.2d 478, 479 (N.D.1978); and First Nat. Bank & Trust Co. of Bismarck v. Hart, 267 N.W.2d 561, 564 (N.D.1978). I have also signed opinions authored by my various colleagues, such as E. E. Bach Millwork Co. v. Meisner & Co., 228 N.W.2d 904 (N.D.1975), which said that we still give deference to the trial court’s findings; Olson v. North Dakota Dist. Court, Etc., 271 N.W.2d 574 (N.D.1978); and Olson v. Peterson, 288 N.W.2d 294 (N.D.1980), which cited and applied Do-lajak.
I often insert dicta into opinions that I write, and I do not refuse to cite or sign opinions authored by others because they contain dicta. But I feel no compulsion to treat dicta as binding precedent. I commend the Chief Justice for his extraordinary efforts and functioning as a trial judge, reaching the same conclusions as were *253reached by Temporary Judge Anderson in this case.
The conflicting testimony relied upon by the appellant and pointed out in the majority opinion, only by remote inference in this case, creates a dispute of facts. Undoubtedly, cases will come to this court where the factual dispute is real and direct. In this case Krohnke did not attack any specific finding of fact. See Sorenson v. Olson, 235 N.W.2d 892 (N.D.1975). The fact that the testimony is transcribed and the trial judge does not hear it, but only reads it, does not eliminate the need for some weighing in order to resolve the conflicts. In every case, the trial judge should do the weighing and appellate judges should leave it that way unless they are convinced that the trial judge was clearly erroneous in making a specific finding. Many theoretical “what if” questions come to mind. E. g., what if a witness cannot speak but writes answers when questioned on the stand? I, for one, am not sure how the majority opinion will be used to answer that question.
In both trial courts and appellate courts, questions of fact sometimes become questions of law, but not unless reasonable men can draw but one conclusion therefrom. Schatz v. Jerke, 199 N.W.2d 908 (N.D.1972). Without just setting up “straw men” for the purpose of knocking them down, consider for a moment what would logically follow if, in trial courts, all documentary evidence be interpreted as raising questions of law only. At best it would confound the practice of law insofar as using, in jury trials, any deposition testimony, answers to interrogatories, or even exhibits with words on them, for fear of opening the door to a claim that a waiver of jury trial is thereby made. After studying the entire transcript, the Chief Justice does not say that reasonable men could not draw a different conclusion than he drew. If that were his statement, then he should also declare the appeal frivolous, as counsel for Lemer suggests, and allow costs and attorney fees under Rule 38, N.D.R.App.P.
I look at the findings of fact prepared by Judge Anderson and, without any regard for his opportunity to judge credibility of witnesses, I conclude that the findings of fact which he prepared are not clearly erroneous. I think the majority opinion ought to say that. My conclusion is not reached after a trial de novo but based upon the appendix, the briefs of counsel, and the oral argument made before us.
My views on Rule 52(a), N.D.R.Civ.P., are not original, in spite of what my colleagues might think. We are just 30 years late in considering the impacts of the matter. A number of judicial scholars have written about this subject. I think William Farnum White aptly covered the apparently on-going dispute in his book, “Winning In Court on the Law of Facts,” Prentice-Hall, Inc. (1972), pages 181-183, as follows:
“The federal appellate courts seem to more readily set aside findings as being clearly erroneous where the evidence does not involve ‘live witnesses’ or where the findings are predicated upon opinions of experts. It is for this reason the character of the record on appeal should be carefully considered in deciding whether you have a fair chance to reverse findings of fact on appeal.
“Fact Review in the ‘Paper Case’: Because Rule 52(a) contains the admonishment that ‘due regard shall be given to the opportunity of the trial court to judge credibility of the witnesses’ it was quickly urged that in the case presented upon written documents, or on depositions or on stipulations of fact that the appellate court is free to set aside findings even if they not be found to be clearly erroneous.
“On the one hand, we find what has been called the Frank view as expressed by Judge Frank in Orvis v. Higgins, (2 Cir., 1950) 180 F.2d 537 at 539:
‘(I)f (the trial judge) decides a fact issue on written evidence alone, we are as able as he to determine credibility, and so we may disregard his findings.’
“On the other hand, we find the view of Judge Clark, who drafted Rule 52(a) to the effect that ‘clearly erroneous’ in Rule 52(a) is the criteria to be applied to all findings of fact irrespective of the nature of the evidence upon which the finding is predicated.
“After itself having gone in both directions on the Frank view and the Clark *254view of Rule 52(a), the Court of Appeals for the Ninth Circuit in a thorough opinion adopted the Clark view as the most sound in Lundgren v. Freeman, (9 Cir., 1962) 307 F.2d 104. In doing so, it believed that the Supreme Court in C.I.R. v. Duberstein, (1960) 363 U.S. 278 [80 S.Ct. 1190, 4 L.Ed.2d 1218] strongly suggested that the Clark view was the more appropriate. It also took the position that the Clark view of Rule 52(a) would encourage litigants to limit appeals to cases where they were firmly convinced that the trial court had erred on facts rather than upon the idea in a ‘paper case’ they had a hope that the appellate court might second guess the trial court in their favor.
“It is only by limiting an appellate court to setting aside those findings which are found to be clearly erroneous can the traditional province of the trial court to find facts be preserved. Ability to see and appraise ‘live witnesses’ is not the only reason for the rule. Litigation upon undisputed facts only arises because the parties with good reason draw different inferences from the same evidence to arrive at different ultimate conclusions. The trial court’s experience with the mainsprings of human conduct is just as sound as that of the appellate court and for that reason should be fully respected. [Footnote omitted.]
“While the Clark view of the ‘paper case’ appears the most sound of the two discussed, it just might happen that ultimately an in-between view will be accepted; that is to say, that while the ‘paper case’ will not be free of the ‘clearly erroneous’ rule it will be relaxed or modified to some unspecified extent. This idea was expressed in Cardelis v. Refineria Panama, S. A., (5 Cir., 1967) 384 F.2d 589. That a ‘paper case’ should be treated with less reverence than a ‘live witness’ case is suggested in the second paragraph of Note 16 to United States v. General Motors, (1966) 384 U.S. 127 at 141 [86 S.Ct. 1321 at 1328, 16 L.Ed.2d 415] where the Supreme Court uses the ‘paper case’ record to justify its holding that certain findings of fact were clearly erroneous:
‘Moreover, the trial court’s customary opportunity to evaluate the demeanor and thus the credibility of the witnesses, which is the rationale behind Rule 52(a) (see United States v. Oregon State Med. Soc., 343 U.S. 326, 331-332 [72 S.Ct. 690, 694-695, 96 L.Ed. 978]), plays only a restricted role here. This was essentially a “paper case.” It did not unfold by the testimony of “live” witnesses. Of the 38 witnesses who gave testimony, only three appeared in person. The testimony, of the other 35 witnesses was submitted either by affidavit, by deposition, or in the form of an agreed-upon narrative of testimony given in the earlier criminal proceeding before another judge. A vast number of documents were also introduced, and bear on the question for decision.
‘In any event, we resort to the record not to contradict the trial court’s findings of fact, as distinguished from its conclusionary “findings,” but to supplement the court’s factual findings and to assist us in determining whether they support the court’s ultimate legal conclusion that there was no conspiracy.’ ”