(dissenting).
I would affirm the court of appeals because I believe it was correct in upholding *457a special jury verdict of the trial court which found that:
1. An article published by the Duluth News Tribune was defamatory to Diesen;
2. The implication of the article was substantially false;
3. Diesen demonstrated by clear and convincing evidence that the article was published with reckless disregard for the truth; and
4. The article was published willfully, maliciously and with intent to harm Diesen.
The article was written by appellant, John Hessburg, who was hired by the Duluth News Tribune in February 1981. Hessburg left the paper in September. 1981 after he wrote the article. Hessburg started work on the article in April 1981 after he had met with a group of advocates for battered women, including representatives of the Duluth Women’s Coalition, Kathy Berglund, and her attorney. Berglund herself was a battered woman. Diesen claimed that these same people had met with him the previous fall in 1980 and urged him to pursue a felony charge for an assault on Berglund that occurred earlier. Diesen testified that he had denied the request because, in a statement to the Sheriff of Carlton County, Berglund stated that she could not remember the alleged assailant, Melvin DeFoe, hitting her and that she had been drinking heavily prior to the alleged attack.
Following the April 1981 meeting Hess-burg had with the women, he attempted to investigate every recent assault in Carlton County, but concentrated on 44 domestic assault cases. He charted those 44 cases, but did not state in his article that Diesen had been involved in only ten of them. At the outset, Hessburg referred to his investigation as “the Diesen probe.”
As the majority opinion points out, Hess-burg set up interviews with a number of people: victims of abuse, attorneys, supporters of Diesen and Diesen himself. His interview with Diesen alone took IOV2 hours and was taped by Hessburg; however, over 5 hours of the interview on the tape is not a part of the record in this case. Diesen alleges that portions of the interview favorable to him were erased or were taped over by Hessburg. Some of the allegations made by battered women were not contained in the printed article. The newspaper found some of these battered women not to be credible. Witnesses contacted by Hessburg favorable to Diesen found Hess-burg’s questions misleading and vindictive and called for answers which reflected poorly on Diesen. One of the newspaper’s editors who read Hessburg’s notes after the latter left the newspaper testified that Hessburg had lost his objectivity. The editors and publishers also testified that they knew that the article implied that Diesen had committed malfeasance and misfeasance in office. Moreover, facts were intentionally omitted from the article, and the juxtaposition of facts within the article itself created false implications. Hess-burg’s malicious investigative techniques created negative opinions about Diesen and omitted facts which indicated that Diesen was an intelligent, hard-working, conscientious, and impartial county attorney.
The court of appeals, in a scholarly, thoughtful opinion, cited the cases which it thought were applicable at the time the opinion was issued. However, at that time, two important cases, which will be discussed, had not yet been released: one was by the Eighth Circuit Court of Appeals entitled Price v. Viking Penguin, Inc., 881 F.2d 1426 (8th Cir.1989), cert. denied, — U.S. -, 110 S.Ct. 757, 107 L.Ed.2d 774 (1990); the second was by the United States Supreme Court entitled Harte-Hanks Communications, Inc. v. Connaughton, — U.S. -, 109 S.Ct. 2678, 105 L.Ed.2d 562 (1989). The court of appeals did cite Prosser’s treatise on the law of torts to the effect that the gist of a publication protects a defendant who gets the details wrong, but gets the “gist” right; but also works in reverse to impose liability on a defendant who has the details right, but the “gist” wrong. Diesen v. Hessburg, 437 N.W.2d 705, 709-10 (Minn.App.1989). The court goes on to indicate that the defendant who juxtaposes a series of facts so as to imply a defamatory connection between them or creates a defam*458atory implication by omitting facts may be held liable. Id. That is what we have in this case.
I. Falsity
The trial court in the present case granted judgment notwithstanding the verdict based on the “Janklow / Oilman opinion analysis.” Under this analysis, an opinion cannot be false because “there is no such thing as a false idea.” Price, 881 F.2d at 1431 n. 1 (quoting Gertz v. Robert Welch, Inc., 418 U.S. 323, 339-40, 94 S.Ct. 2997, 3006-07, 41 L.Ed.2d 789 (1974)). This same test, however, was used by the Minnesota Court of Appeals to reverse the trial court and reinstate the jury’s verdict. Diesen, 437 N.W.2d at 711-12. The majority opinion of this court now employs the same test to reverse the court of appeals. These varying results highlight the test’s subjectivity and justify closer scrutiny of the Janklow / Oilman test.
The opinion/fact test followed in Price v. Viking Penguin, Inc., 881 F.2d 1426, 1431-32 (8th Cir.1989), cert. denied, — U.S. -, 110 S.Ct. 757, 107 L.Ed.2d 774 (1990),1 was first articulated .in Ollman v. Evans, 750 F.2d 970, 975-79 (D.C.Cir.1984) (en banc), cert. denied, 471 U.S. 1127, 105 S.Ct. 2662, 86 L.Ed.2d 278 (1985), and was adopted, as modified, by the Eighth Circuit Court of Appeals in Janklow v. Newsweek, Inc., 788 F.2d 1300, 1302 (8th Cir.), cert. denied, 479 U.S. 883, 107 S.Ct. 272, 93 L.Ed.2d 249 (1986) (“[W]e choose here to adopt the four factors suggested in Judge Starr’s scholarly opinion [in Oilman ], and to expand them, for reasons we will explain, to include elements of the concurrence by Judge Bork.”). This test substantially changes the traditional New York Times Co. v. Sullivan, 376 U.S. 254, 84 S.Ct. 710, 11 L.Ed.2d 686 (1963), analysis by creating an extremely broad opinion privilege for defendants in public official libel actions. The substantial change in the law results from the fact that the applicability of this opinion privilege, unlike the issue of whether there is actual malice, is a question of law and, therefore, answered at the summary judgment stage. By preventing the case from getting to a jury, this test creates a separate constitutional threshold that, in addition to the actual-malice standard, virtually immunizes public official libel defendants.
This additional constitutional hurdle unnecessarily diminishes the protection afforded to the public official’s interest in his or her professional reputation by state defamation law. Moreover, because this test is not mentioned in the most recent United States Supreme Court case of Harte-Hanks Communications v. Connaughton, — U.S. -, 109 S.Ct. 2678, 105 L.Ed.2d 562 (1989), or any other United States Supreme Court decision and not even uniformly accepted by the United States Circuit Courts of Appeals, its validity as first amendment law is suspect. I believe that the Harte-Hanks Court deliberately avoided the troublesome fact/opinion dichotomy and the additional uncertainty it creates in favor of a detailed actual-malice analysis.2
In Harte-Hanks, an unsuccessful candidate for the position of municipal judge [Connaughton] brought a federal court diversity libel action against a newspaper. Harte-Hanks, — U.S. at -, 109 S.Ct. at 2681-82. The newspaper published an article which stated that a local woman [Thompson] alleged that Connaughton committed various types of wrongful acts in order to gain her cooperation in exposing corruption on the part of one of the incumbent judge’s key employees. See id. 109 *459S.Ct. at 2692. Following a trial, a jury found, by special verdict, that the publication in question was defamatory, false, and published with actual malice. Id. at-n. 2, 109 S.Ct. at 2682 n. 2.
On appeal, the Sixth Circuit Court of Appeals reviewed the evidence supporting each of the jury’s special verdicts, including the finding that the article was false, and concluded that these findings were not clearly erroneous. Connaughton v. Harte-Hanks Communications, Inc., 842 F.2d 825, 841-44 (6th Cir.1988). With respect to the element of falsity, the court of appeals focused on the falsity of Thompson’s “charges” and stated:
Equally apparent from the jury’s answer to the second special interrogatory is that it considered the published Thompson charges to be false. * * * Thus, upon reviewing the record in its entirety, this court concludes that the jury’s determinations of the operational facts bearing upon the falsity of the article in issue were not clearly erroneous.
Connaughton, 842 F.2d at 841 (quoted in Harte-Hanks, — U.S. at -, 109 S.Ct. at 2683 n. 4).
The court of appeals rejected the newspaper’s assertion that the constitutional privilege for expressions of opinion protected the publication of Thompson’s “dirty tricks” statement. Connaughton, 842 F.2d at 847. In so doing, the court of appeals stated:
It [the newspaper] argued that the conclusion that Connaughton was guilty of “dirty tricks” represented Thompson’s personal opinions of Connaughton’s actions. However, “[ojpinions based on false facts are actionable ... against a defendant who had knowledge of the falsity or probable falsity of the underlying facts." Davis v. Ross, 754 F.2d 80, 86 (2nd Cir.1985) (quoting Hotchner v. Castillo-Puche, 551 F.2d 910, 913 (2nd Cir.), cert. denied sub nom. A.E. Hotchner v. Doubleday & Co., 434 U.S. 834, 98 S.Ct. 120, 54 L.Ed.2d 95 (1977)); accord Cianci v. New Times Publishing Co., 639 F.2d 54, 65 (2nd Cir.1980); see also Lewis v. Time Inc., 710 F.2d 549, 554 (9th Cir.1983). Thompson’s claim that Connaughton resorted to “dirty tricks” to induce her statements was based upon the false factual assertions that Con-naughton had offered Thompson jobs, trips and anonymity. Because the Journal had obvious reasons to doubt the truth of those underlying facts, the references to “dirty tricks” were not constitutionally protected statements of opinion.
Id. at 847 (emphasis added). Based on the foregoing, the court of appeals affirmed the judgment of the district court. Id.
In his dissent, Court of Appeals Judge Guy adopted the newspaper’s argument that the references to “dirty tricks” were constitutionally protected expressions of opinion. Connaughton, 842 F.2d 825, 857 (Guy, J., dissenting) (citing Gertz v. Robert Welch, Inc., 418 U.S. 323, 339-40, 94 S.Ct. 2997, 3006-07, 41 L.Ed.2d 789 (1974) (“There is no such thing as a false idea.”)), and Ollman v. Evans, 750 F.2d 970, 978 (D.C.Cir.1984) (en banc), cert. denied, 471 U.S. 1127, 105 S.Ct. 2662, 86 L.Ed.2d 278 (1985). Judge Guy reasoned that the language of the article indicated that it was Thompson’s opinion and also set forth the basis for Thompson’s belief such that “readers were able to decide for themselves as to whether or not the plaintiff’s actions constituted ‘dirty tricks.’ ” Despite the obvious similarity between Judge Guy’s dissent and the opinion/fact distinction adopted by the Oilman and Janklow courts, the United States Supreme Court did not even discuss the possible existence of constitutional protection for statements of opinion even though it included a one-paragraph summary of the other points made by Judge Guy in his dissent. Harte-Hanks, — U.S. at -, 109 S.Ct. at 2683. Despite this conspicuous omission, the majority of this court chooses to follow the opinion analysis found in Judge Guy's twice-rejected dissent.
For purposes of the case at bar, it is particularly significant that the Harte-Hanks Court observed that each of Thompson’s allegations were “accurately report*460ed.” Id. 109 S.Ct. at 2692.3 If the Janklow /Oilman fact/opinion rule somehow insulates newspapers from liability as long as what they print is not verifiably false, the Court surely would have discussed the rule. It could easily be argued that, since the newspaper in Harte-Hanks accurately published what Thompson said, any false implications flowing from those accurate statements were constitutionally protected opinion.
Instead of adopting the Janklow / Oilman opinion analysis in order to immunize media defendants in these types of cases, the Supreme Court focused on the competing individual interest at stake. The Court stated:
We have not gone so far, however, as to accord the press absolute immunity in its coverage of public figures or elections. If a false and defamatory statement is published with knowledge of falsity or a reckless disregard for the truth, the public figure may prevail.
Harte-Hanks, — U.S. -, 109 S.Ct. at 2696 (citing Curtis Publishing Co. v. Butts, 388 U.S. 130, 162, 87 S.Ct. 1975, 1995, 18 L.Ed.2d 1094 (1967)). Moreover, the court noted:
Of course, the protection of “calculated falsehoods” does not promote self-determination. As we observed in Garrison v. Louisiana, 379 U.S. 64 [85 S.Ct. 209, 13 L.Ed.2d 125] (1964):
“At the time the First Amendment was adopted, as today, there were those unscrupulous enough and skillful enough to use the deliberate or reckless falsehood as an effective political tool to unseat the public servant or even topple an administration. That speech is used as a tool for political ends does not automatically bring it under the protective mantle of the Constitution. For the use of the known lie as a tool is at once at odds with the orderly manner in which economic, social, or political change is to be effected.”
Id. at - n. 34, 109 S.Ct. at 2696 n. 34 (citations omitted).
In summary, the United States Supreme Court, unlike the majority of this court, recognized in Harte-Hanks the minimal contribution that false statements make towards the interests protected by the first amendment. Moreover, the Court rejected the hostility to juries embodied in the Janklow / Oilman analysis by leaving the jury’s finding of falsity in these cases undisturbed. The Court did not, however, indicate in any way that allowing the element of falsity to be found by a jury based on implications flowing from the published statement would impermissibly infringe on first amendment freedoms.
The absence of any meaningful discussion of Oilman or Janklow in Harte-Hanks is perhaps explained by the Court’s discussion of the meaning of the terms “actual malice” and “reckless disregard.” See Harte-Hanks, — U.S. at -, 109 S.Ct. at 2695. The Court stated: “Uncertainty as to the scope of the constitutional protection can only dissuade protected speech — the more elusive the standard, the less protection it affords.” Id.
The above statement supports an inference that the Court was unwilling to add additional complexity to the already complex law of public figure libel by setting forth yet another elusive constitutional standard, namely, the opinion/fact dichotomy, especially in the absence of a showing that the actual-malice standard did not sufficiently protect the competing interests at stake. If nothing else, the 69 pages and 190 footnotes which make up the Oilman decision demonstrate that the fact/opinion dichotomy is an elusive standard.4
*461Other reasons why the United States Supreme Court did not adopt the Janklow /Oilman opinion analysis may be found by considering the criticism directed at this analysis by the judges who dissented in these cases. In Oilman, Judge Scalia wrote a vigorous dissent attacking the reasoning of Judge Bork’s concurrence. See Ollman v. Evans, 750 F.2d 970, 1036-39, (Scalia, J., with whom Wald, J., and Edwards, J., join, dissenting in part). Judge Scalia stated: “[T]o say, as the concurrence does, that hyperbole excuses not merely the exaggeration but the fact sought to be vividly conveyed by the exaggeration is to mistake a freedom to enliven discourse for a freedom to destroy reputation.” Ollman, 750 F.2d at 1036 (emphasis in text). Judge Scalia properly observed that “[ejxisting doctrine provides ample protection against the entire list of horribles supposedly confronting the defenseless modern publicist.” Id. Judge Scalia further stated:
It is difficult to see what valid concern remains that has not already been addressed by first amendment doctrine and that therefore requires some constitutional evolving — unless it be, quite plainly, the concern that political publicists, even with full knowledge of the falsity or recklessness of what they say, should be able to destroy private reputations at will.
Id. at 1037 (emphasis added).
Judge Scalia also objected to Judge Bork’s unreasonable hostility towards juries in first amendment cases. Judge Bork was alarmed by a “dramatic proliferation” of libel actions. See Ollman, 750 F.2d at 996-97 n. 2 (Bork, J., concurring). Judge Bork stated: “The only solution to the problem libel actions pose would appear to be close judicial scrutiny to ensure that cases about types of speech and writing essential to a vigorous first amendment do not reach the jury.” Id. at 997 (emphasis added). Judge Scalia characterized Judge Bork’s approach as “unguided evolution” and noted:
“The principle that the first amendment does not protect the deliberate impugning of character or reputation, in its application to the preexisting phenomenon ' of political controversy, is to be revised to permit ‘bumping’ of some imprecisable degree because we perceive that libel suits are now too common and too successful.”
Id. at 1038 n. 2.
In a separate dissent, Judge Wald also concluded that the existing law concerning actual malice provided sufficient protection for media defendants. Ollman, 750 F.2d at 1032-35 (Wald, J., with whom Edwards, J., and Scalia, J., join, dissenting in part). Judge Wald accused Judge Bork of attempting to “immunize libel defendants from suit.” See id. at 1033. Judge Wald also recognized that Judge Bork’s analysis “represents an unprecedented extension into the fact-opinion doctrine of the distinction between public and private officials for the purposes of defamation suits.” Id. (emphasis added). Judge Wald correctly observed that:
In view of the protections already afforded public debate by the “actual malice” standard, I can see no reason other than a vague, but obviously overpowering, distrust of juries for holding the entire law of libel hostage to this quite subtle distinction [between opinion and fact].
Id. at 1035 n. 2.
The above criticism of the Oilman analysis is echoed in the dissenting opinion in Janklow. Janklow v. Newsweek, Inc., 788 F.2d 1300, 1307 (8th Cir.) (Bowman, J., joined by Ross, J., and Fagg, J., dissenting), cert. denied, 479 U.S. 883, 107 S.Ct. 272, 93 L.Ed.2d 249 (1986). In this dissent, Judge Bowman also focused on the countervailing interests at stake in public figure libel actions and said:
This judicially created limitation on libel actions has, interestingly enough, no apparent relationship to any change that has occurred in either the Constitution or society since the Bill of Rights was ratified in 1791. Libel is still libel. All that has changed is the prevailing judicial perception of where the balance should be struck between libel plaintiffs and libel defendants, who in our time are fre*462quently large media organizations such as Newsweek.
Today’s decision tips the balance still farther in the media’s favor. To the fortress of actual malice, the Court adds a virtually impenetrable outer barrier built upon an extremely broad and elastic definition of opinion. Because opinion is not actionable, see Gertz v. Robert Welch, Inc., 418 U.S. 323, 339 [94 S.Ct. 2997, 3006, 41 L.Ed.2d 789] (1974), the result is that many libel cases (and probably almost all libel cases in which the plaintiff is a public official) will be dismissed before the issue of actual malice is ever reached. I find it hard to believe that this is what Justice Powell had in mind when he penned his famous dictum in Gertz that “[u]nder the First Amendment there is no such thing as a false idea.” Id. Ideas are one thing, but tawdry attacks on character and reputation are another. I do not see any reason to extend absolute protection under the First Amendment to statements that qualify as opinion rather than fact only by means of judicial semantics based on the Oilman factors.
Janklow, 788 F.2d at 1306-07 (emphasis added) (footnotes omitted). In light of the existence of this persuasive criticism of the opinion/fact dichotomy, it is not surprising that the Harte-Hanks Court refused to adopt the Janklow / Oilman opinion analysis as controlling first amendment law. It is readily apparent from the deference the Harte-Hanks Court gave the jury’s findings that the viewpoint expressed by Judge Scalia in his Oilman dissent prevailed.
Because the “We do not recognize defamation by implication” statement in Price v. Viking Penguin, Inc., 881 F.2d 1426, 1432 (8th Cir.1989), was made in connection with the Janklow / Oilman opinion analysis, it has been rejected by the United States Supreme Court and should, therefore, be rejected by this court.5 A final point about Price is worth mentioning. After noting that the district court had granted summary judgment for the defendants, the Price court stated: “To reverse this judgment, we must find that Price could show that precise factual statements were false.” Price, 881 F.2d at 1434; Philadelphia Newspapers, 475 U.S. at 775, 106 S.Ct. at 1563 (1985) (citing Philadelphia Newspapers, Inc. v. Hepps, 475 U.S. 767, 775, 106 S.Ct. 1558, 1563, 89 L.Ed.2d 783 (1985)) (emphasis added) (footnote omitted).
Philadelphia Newspapers, Inc. v. Hepps, 475 U.S. 767, 106 S.Ct. 1558, 89 L.Ed.2d 783 (1985) (5-4 decision), involved a libel action involving a private figure who was allegedly defamed concerning a matter of public concern. Id. 475 U.S. at 770-75, 106 S.Ct. at 1560-63. It is important to *463note that, in Philadelphia Newspapers, the court discussed the element of falsity from the standpoint of the validity of Pennsylvania common law requirements. Id. A close review of Philadelphia Newspapers indicates that the Price court misinterpreted Philadelphia Newspapers. The cited portion of Philadelphia Newspapers does not say anything that can reasonably be interpreted as requiring “precise factual statements.” Instead, the Court stated:
Our opinions to date have chiefly treated the necessary showings of fault rather than of falsity. Nonetheless, as one might expect given the language of the Court in New York Times [475 U.S.] at 772-73 [106 S.Ct. at 1562], a public-figure plaintiff must show the falsity of the statements at issue in order to prevail in a suit for defamation.
Philadelphia Newspapers, 475 U.S. at 775, 106 S.Ct. at 1563. Philadelphia Newspapers is silent as to how a plaintiff may prove falsity, that is, whether falsity may be proved by implication.
In a sharp dissent, Justice Stevens focused on the state interest in preventing and redressing injuries to reputation and concluded that the actual-malice standard provided sufficient protection for first amendment interests. See Philadelphia Newspapers, 475 U.S. at 780-90, 106 S.Ct. at 1565-71 (Stevens, J., with whom Burger, C.J., White, J., and Rehnquist, J., join, dissenting). In reminding the Court of the competing interest at stake, Justice Stevens repeated the following statement from an earlier Supreme Court decision:
The right of a [person] to the protection of his own reputation from unjustified invasion and wrongful hurt reflects no more than our basic concept of the essential dignity and worth of every human being — a concept at the root of any decent system of ordered liberty. The protection of private personality, like the protection of life itself, is left primarily to the individual States under the Ninth and Tenth Amendments. But this does not mean that the right is entitled to any less recognition by this Court as a basic of our constitutional system * * *.
Id. 475 U.S. at 781, 106 S.Ct. at 1566 (quoting Rosenblatt v. Baer, 383 U.S. 75, 92-94, 86 S.Ct. 669, 679-80, 15 L.Ed.2d 597 (1966) (Stewart, J.)).
The United States Supreme Court decision in Barte-Hanks focuses on the protection afforded by the actual-malice standard and, in so doing, balances the competing interests at stake in a manner much more consistent with the dissent in Philadelphia Newspapers than with the Janklow /Oilman /Price opinion analysis. Accordingly, this court should follow the approach taken in Harte-Hanks. Under this approach, the jury’s finding of falsity by implication is consistent with Minnesota’s defamation law as set forth in Lewis v. Equitable Life Assurance Society of the U.S., 389 N.W.2d 876 (Minn.1986), and should not be disturbed by this court.
In Lewis, this court discussed the substantive law of defamation in Minnesota. With respect to the element of falsity, we said:
True statements, however disparaging, are not actionable. Since it is true that plaintiffs were fired for gross insubordination, the [defendant] company argues, they cannot maintain an action for defamation. The company contends the relevant statement to consider when analyzing the defense of truth is the one that plaintiffs made to their prospective employers, that is, that they had been fired for gross insubordination. Plaintiffs counter that it is the truth or falsity of the underlying statement — that plaintiffs engaged in gross insubordination— that is relevant.
sfc * * * * *
Requiring that truth as a defense go to the underlying implication of the statement, at least where the statement involves more than a simple allegation, appears to be the better view.
Lewis, 389 N.W.2d at 888-89 (emphasis added). This unambiguous rule recognizing falsity by implication was recently applied by the court of appeals in Karnes v. Milo Beauty and Barber Supply Co., Inc., 441 N.W.2d 565 (Minn.App.), pet. for rev. denied, (Minn., Aug. 15, 1989). In Karnes, *464the court of appeals cited Lewis for the proposition that truth as a defense must go to the underlying implication and then stated:
Viewed in the light most favorable to the verdict, these statements go beyond mere allegations and are conclusory; they could be interpreted by a jury as implying that Karnes was either stealing or allowing stealing to take place. In order to employ truth as a defense, [the defendants] would have to show the truth of the allegations, which they did not do.
Id. at 568 (emphasis added). Because a public official libel plaintiff has the burden of proof as to the issue of the truth or falsity of the published statements, it is clear that, as a matter of Minnesota defamation law, the trial court erred in holding that there could be no falsity by implication and granting JNOV. Accordingly, I would affirm the court of appeals’ decision.
The majority suggests that its decision is “rooted in state defamation law” by indicating that it is somehow consistent with the discussion in Lewis of the qualified privilege protecting certain communications between employers and employees. See Majority Op. at 452. This is an obvious attempt to shield the majority decision from review by the United States Supreme Court. The majority opinion is unquestionably an extension of the Gertz dictum and, as such, rests solely on a determination of federal law (i.e., a pretended federal defense to a state law claim). The only principled way to reconcile Lewis with the majority decision is to recognize that Lems is a state defamation case and the majority’s decision is an example of the Janklow /Oilman /Price first-amendment analysis.
II. Actual Malice
Actual malice requires that the statements be made with “reckless disregard for the truth.” Harte-Hanks, — U.S. at -, 109 S.Ct. at 2685. This standard is a subjective one. Id. at -, 109 S.Ct. at 2696. The plaintiff may prove the defendant’s state of mind through circumstantial evidence. Id. at -, 109 S.Ct. at 2686. “There must be sufficient evidence to permit the conclusion that the defendant in fact entertained serious doubt as to the truth of his publication,” Harte-Hanks, — U.S. at -, 109 S.Ct. at 2696 (quoting St. Amant v. Thompson, 390 U.S. 727, 731, 88 S.Ct. 1323, 1325, 20 L.Ed.2d 262 (1968)), or made the false publication with a “high degree of awareness of probable falsity.” Id. (quoting Garrison v. Louisiana, 379 U.S. 64, 74, 85 S.Ct. 209, 215, 13 L.Ed.2d 125 (1964)). In a case involving the reporting of a third party’s allegations, “recklessness may be found where there are obvious reasons to doubt the veracity of the informant or the accuracy of his reports.” Harte-Hanks, — U.S. at -, 109 S.Ct. at 2696 (quoting St. Amant, 390 U.S. at 732, 88 S.Ct. at 1326) (emphasis added). In St. Amant, the United States Supreme Court also stated:
The defendant in a defamation action brought by a public official cannot, however, automatically insure a favorable verdict by testifying that he published with a belief that the statements were true. The finder of fact must determine whether the publication was indeed made in good faith. Professions of good faith will be unlikely to prove persuasive, for example, where a story is fabricated by the defendant, is the product of his imagination, or is based wholly on an unverified anonymous telephone call.
St. Amant, 390 U.S. at 732, 88 S.Ct. at 1326.
“The question whether the evidence in the record in a defamation case is sufficient to support a finding of actual malice is a question of law.” Harte-Hanks, — U.S. at -, 109 S.Ct. at 2694. In discussing the role of a reviewing court, the Harte-Hanks Court said:
In determining whether the constitutional standard has been satisfied, the reviewing court must consider the factual record in full. Although credibility determinations are reviewed under the clearly erroneous standard because the trier of fact has had the “opportunity to observe the demeanor of the witnesses,” the reviewing court must “examine for itself the statements in issue and the *465circumstances under which they were made to see ... whether they are of a character which the principles of the First Amendment ... protect.”
Harte-Hanks, — U.S. at -, 109 S.Ct. at 2696 (citations omitted).
In applying the above principles, the Harte-Hanks Court analyzed the trial court’s jury instructions, the jury’s answers to the special interrogatories, and the facts not in dispute and concluded that the jury must have rejected (1) the testimony of the newspaper’s witnesses as to why the key informant was not contacted; (2) the testimony of the newspaper’s editor as to why he did not listen to the tape containing crucial information; and (3) the testimony of those newspaper employees who testified that they believed that the third party’s allegations were substantially true. Id. 109 S.Ct. at 2697. Based on this reasoning, the Harte-Hanks Court concluded that the newspaper acted with actual malice. Id.
The Harte-Hanks Court focused on several key facts: First, that Thompson’s [the person quoted in the article] most serious charge was highly improbable and inconsistent with facts known to the newspaper. Harte-Hanks, — U.S. at -, 109 S.Ct. at 2697-98. The “hesitant, inaudible and sometimes unresponsive and improbable tone of Thompson’s answers to various leading questions raise obvious doubts about her veracity.” Id. at-, 109 S.Ct. at 2698 (emphasis added). Finally, the Court noted that no one interviewed the people (the primary informant and the police) most likely to confirm or refute Thompson’s charges. Id. at -, 109 S.Ct. at 2693, 2698. The Court then stated:
Accepting the jury’s determination that petitioner’s explanations for these omissions were not credible, it is likely that the newspaper’s inaction was a product of a deliberate decision not to acquire knowledge of facts that might confirm the probable falsity of Thompson’s charges. Although failure to investigate will not alone support a finding of actual malice, the purposeful avoidance of the truth is in a different category.
Harte-Hanks, — U.S. at -, 109 S.Ct. at 2698 (citation omitted) (emphasis added).
In the present case, Kathy Bergland is in a position somewhat analogous to that of Thompson in Harte-Hanks in that the gravamen of the misfeasance- or malfeasance-in-office charge is based on Diesen’s failure to prosecute effectively the man who repeatedly abused Berglund. In the present case, special interrogatory No. 4 stated:
Did plaintiff prove by clear and convincing evidence that the defendants published the articles with reckless disregard as to the truth or falsity of the implication of said articles?
Answer: “Yes”
Berglund testified at length at trial. Berg-lund admitted under cross-examination that, following the second battering incident, she told Sheriff Twomey: “After I got out of my car the next thing I remember is coming to on a pile of rocks in my yard.” Also during Berglund’s cross-examination, the following exchange took place:
Q And the statement that you gave to Mr. Hessburg about Melvin Defoe tackling you and causing you to fall, that was the first time you told anybody that version of the story correct?
A I don’t know. I didn’t talk to anybody else about it between that time, I don’t think.
The jury also heard considerable testimony from Diesen regarding the case of Kathy Berglund and the reasons for his actions. Diesen testified that he did not bring felony charges against Defoe because of a number of considerations, including questions concerning Berglund’s prior inconsistent statements as to whether Defoe broke her jaw and the way a defense attorney would use these statements. On cross-examination, Diesen testified that, in his opinion, Berglund was intoxicated at the time and fell down while running into the house.
The jury also listened to Police Officer Randelin say that he discussed the Berg-lund case with Hessburg at great length. Officer Randelin said that he became involved in the case because of reports of child neglect and because Kathy Berglund *466had been drinking and passing out in bars. Officer Randelin also testified about the cycle of violence/reconciliation between Berglund and Defoe and stated: “The Berglunds had been involved in domestics many, many, times; and Kathy never ever, ever followed through.”
Officer Randelin also testified about the Chip Martin case referred to in the newspaper articles. Officer Randelin described in detail how he investigated the alleged stabbing incident in that case and how Mrs. Martin ultimately admitted that there had been no knife attack. Officer Randelin also said: “[H]e (Mrs. Martin’s attacker) did not stab her; and Hessburg knew that because we talked about that case; and that’s not in here [referring to the newspaper article].” The jury also heard Officer Randelin’s testimony concerning Hess-burg’s suggestive interviewing techniques as follows:
It just seemed like — I just got the opinion, my opinion, that he [Hessburg] was out to nail Don and I didn’t know why. Any time I tried to tell him something about, hey, that’s not the way it is, he would just come back at me and accuse me of being friends with Don * * *. But we got into Don’s honesty; we got into my feelings about Don; Don’s weaknesses, Don’s strengths; and we seemed to forget about everybody else in the system, the judges, the probation department, the law enforcement officers, social services, the counselors, everybody else that gets involved in domestic abuse cases; and he seemed to just forget all that, and it was all concentrated on Don.
(Emphasis added.) On the witness stand, Hessburg flatly denied discussing the Chip Martin case with Officer Randelin (“It wasn’t discussed, period.”). This denial, however, was promptly and totally impeached by Diesen’s counsel through the use of Hessburg’s previous deposition testimony which showed that Hessburg had discussed the Chip Martin case with Officer Randelin.6
The jury also heard the defendant’s expert’s statement that this was a “textbook model” of an investigation that the majority finds so authoritative. The jury, however, also observed a withering cross-examination during which the expert admitted that he had not listened to the tapes of Hessburg’s interviews and that:
1. An investigative reporter must be cautious when dealing with special interest groups.
2. An investigative reporter should not begin an investigation with a preconceived story because he would be likely to reach a preconceived conclusion.
3. The reporter must remain objective throughout the investigation because, otherwise, the materials in the files would be one sided.
4. In conducting an investigation, it’s important to have opinions from “informed sources” on both sides of the issue.
5. A judge, a prosecutor’s fellow prosecuting attorney, and law enforcement officials would be “informed sources” on what a prosecuting attorney is doing.
6. He [the expert] was not aware of any judge or fellow prosecuting attorneys of whom Hessburg asked their opinions of Diesen.
7. He did not know if the people whom Hessburg interviewed were “informed sources” or not.
8. From his [the expert’s] review of the transcript of the interview, he did not recall Hessburg asking Diesen for the names of people who might be informed sources.
9. If an investigative reporter conducted an unfair investigation, that could result in an unfair article being printed.
10. If an investigative reporter lost his objectivity during the investigation, that could result in a gathering of in*467formation that was slanted towards one side.
In addition, the defendant’s expert admitted that it would be poor journalism that could result in an inaccurate article:
11. For a reporter to lead sources to the answers he wants;
12. To omit significant facts from an article;
13. To become deeply emotionally involved in a story.
Finally, the defendant’s expert admitted that he was aware that the newspaper was concerned that Hessburg had lost his objectivity during this investigation and had become emotionally involved in this story. The expert agreed that if a reporter lost his objectivity, it could poison the whole series of stories. In light of this testimony, the jury’s answers to the special interrogatories, and the fact that the jury awarded the plaintiff $780,000 dollars (including $500,-000 in punitive damages), the jury must have rejected the defendant’s expert’s opinion and believed the assertions underlying plaintiff’s counsel’s leading questions on cross-examination.
The jury also heard testimony that, during his marathon interview7 with Diesen on August 21,1981, Hessburg made a thinly veiled accusation that Diesen and/or Twomey destroyed the Greensky file. Hessburg told Diesen that he was going to publish information about the disappearance of the Greensky file and said, “The only logical conclusion any reader will make is that you or Terry [Sheriff Twom-ey] removed the file and destroyed it. It’s a felony to destroy public records.”
On August 28, 1981, not long after this marathon interview, Diesen wrote to Mr. Daly, the executive editor of the newspaper, the second of a series of letters expressing concern about the accuracy of the information Hessburg said he was preparing to publish. In this letter, Diesen expressed particular concern about the publication of material which was false and defamatory to him. Daly also received a letter dated August 27, 1981, from Sheriff Twomey indicating that he believed that the newspaper was in the process of a character assassination of Donald Diesen.
The jury heard that, sometime after receiving Diesen’s and Sherriff Twomey’s letters, Georgia Swing, the newspaper’s city editor, was sent to the Carlton County Courthouse to verify the accuracy of Hess-burg’s reports. Swing could not remember the dates when she went to the courthouse and made no notes of her findings. Swing said that she made two trips to the courthouse and spent a total of about 3 hours checking files.
Swing said that she could not remember having copies of the unpublished articles with her on her trips, but said that she had the disposition chart which listed the cases Hessburg had looked into. Swing said that she did not check all 44 of the cases. Swing stated that she was interested in the Berglund file because “[t]hat was the focus of one of his [Hessburg’s] stories.” Nevertheless, Swing said that she never contacted Kathy Berglund, Sherriff Twomey, Sergeant Randelin, or Don Diesen about the Berglund case. Swing said that she could not remember looking at the Chip Martin file. Swing admitted that she did not prepare any kind of written summary of the files for her superiors.
Swing also contacted an attorney who had represented Jennifer Greensky’s parents in 1973 in connection with the alleged rape of Jennifer. Swing reported information regarding the Greensky file to Buster, her supervisor. Swing said that she discussed the decision not to publish the Greensky story with Buster. Under cross-examination, Swing said that she was in support of the decision not to publish the Greensky story in part because “there was a lot of reason to be skeptical of what Jennifer Greensky said * * *.” Buster said that he listened to parts of the Green-sky tape and began to form doubts about the Greensky matter.
*468With respect to the tape of the Diesen interview, Buster said: “I listened to parts of it. The whole interview was — John ran out of tape. The whole interview was not there; and I did not listen to all the pieces that actually were there.” (Emphasis added.) As to why the whole interview was not there, the jury heard Diesen say that the tape contained discussion of cases in a different order than the interview was actually conducted and that a lot of cases discussed during the interview were not on the tapes.
Buster made notes of his review of the tapes. The jury heard Buster’s testimony concerning these notes. During this testimony, the jury heard that Buster had noted: “John’s question, not only on this tape but on lots of others, lead [sic] the sources to the answers that John wants concerning people’s opinions on Diesen.” (Emphasis added.) Later, Buster noted that “Hessburg is suggesting to subsequent sources the idea of a special panel to probe the way Diesen handles battering cases, though he didn’t suggest it in the first place.” Buster also noted that Kathy Berglund indicated that she reconciled with Defoe after the first battering incident and that “Kathy [Berglund] was the one who recommended help for Defoe rather than jail — recommended to probation officer Mark Zuber.” Buster admitted in front of the jury that he specifically noted: “John’s comment show [sic] that he’s deeply, emotionally involved in the story.” Buster told the jury that he noted that Berglund planned to campaign against Diesen in the 1982 election. Buster also noted that he questioned whether Berglund had a “political ax to grind.” Despite these observations, Buster never contacted any of the other sources that Hessburg had contacted in developing his story with the exception of a brief discussion with Sheriff Twomey.
Sometime in early October 1981, the newspaper received some additional information from Diesen which indicated that the Greensky story was false. Shortly after this, Fortner and Buster made the decision not to publish the Greensky story.
Sometime around the end of October 1981, Hessburg stopped working for the newspaper. The jury considered ambiguous evidence concerning why Hessburg left the newspaper. Hessburg testified that he tendered his resignation on July 29 or 30, 1981, but that it was effective at the end of September 1981. Hessburg said that he left the Duluth paper because he “wasn’t getting along with people the way I thought I should; and it was a matter of, I guess, personal and professional chemistry.” Hessburg said that he worked on the article on his own time for roughly 2 weeks after his resignation became effective at the end of September. By contrast, in a letter from the newspaper’s attorney, Mr. Killen, to the newspaper’s publisher dated October 27, 1981, Killen wrote:
You have also indicated that the versions which we have reviewed have been sharply edited from the original drafts of the reporter who worked on the story. This reporter is no longer with the newspaper and has been terminated, as I understand it, because you questioned his objectivity.
During his testimony at trial, Killen explained this statement by saying:
They [the newspaper] certainly left me with the impression that Mr. Hessburg had been terminated. That impression was later corrected by the newspaper to say, in effect, we didn’t terminate him, he resigned; but we probably wouldn’t have kept him.
Killen’s October 27, 1981 letter framed the basic issue for the newspaper as follows: The real question is whether such an apparent attitude on the part of the reporter has poisoned the whole series of stories.
The jury’s decision in this case was undoubtedly based on its assessment of credibility. The trial court gave the jury detailed instructions regarding credibility and how to judge the testimony of a witness. The following is a summary of the evidence of falsity and actual malice contained in the record in this- case:
A. Berglund Story
1. Omitted fact of reconciliation after first assault.
*4692. Omitted fact that Berglund suggested treatment.
3. Omitted fact that Diesen recommended jail.
4. Published the story without verifying details after suspecting that Berg-lund had a “political ax to grind.”
5. Mischaracterized Diesen’s reasons for not prosecuting, especially prior inconsistent statements of Berglund and other proof problems.
B. Chip Martin Story
1. Omitted fact that Mrs. Martin admitted that there was no knife attack.
2. Editor failed to check the Chip Martin file when, supposedly, double-checking facts.
3. Hessburg was impeached on the witness stand regarding his denial of the conversation with Officer Randelin regarding Martin's admission.
C. Unpublished Stories
1. Failed to interview Berglund, Martin, or other sources likely to be able to confirm or refute their stories even after learning that the Greensky story was false and forming serious doubts about the truth of several other stories and terminating John Hessburg’s employment for reasons related to his loss of objectivity.
2. Published the Berglund and Martin stories after discovering that several of the other stories were doubtful.
D. Verification Efforts
1. Editor who supposedly double-checked accuracy of articles only checked some of the files on a disposition chart listing 44 cases, of which only 10 involved Diesen. Of the 10 that involved Diesen, the editor could only recall checking the Berglund file.
2. Jury heard testimony regarding tape tampering.
3. Editor who listened to tapes (not the same person who checked files) did not listen to all of the tapes in the newspaper’s possession and knew that portions of the interview were missing.
(a) Noted use of leading questions.
(b) Noted reporter’s deep emotional involvement.
(c) Suspected Berglund’s political motives.
(d) Noted that it seemed “strange” that Ms. Greensky could not remember the details of her alleged rape without coaching from the reporter.
(e) Failed to interview Berglund, Martin, Diesen, or other informed primary sources even after noting concerns.
4.The newspaper published the articles after being advised by its attorney that there was a question as to whether Hessburg’s attitude had poisoned the whole series of articles.
E.Credibility
This case involved lengthy testimony, much of it conflicting, by a number of witnesses. The trial court gave the jury careful and detailed instructions regarding credibility and methods of judging testimony.
The above summary of the record demonstrates that the jury considered evidence sufficient to support a conclusion that the newspaper entertained serious doubts or had obvious reasons to doubt the truth of the reports, especially the Chip Martin and Kathy Berglund stories. Accordingly, the majority’s decision reversing the jury’s conclusion as to actual malice unreasonably discounts the value of the jury’s opportunity to observe the demeanor of the witnesses, a factor that is significant in determining whether a defendant had a particular subjective state of mind.
It seems to me that the newspaper's conduct in the present case is precisely the type of conduct the United States Supreme Court condemned in Harte-Hanks. Although this dissent is unusually long, I believe that the legal profession, as well as the general public, should understand fully the consequences of today’s decision by this court.
Diesen was the subject of a deliberate, cold-blooded, malicious “hatchet job” which had the effect of removing him from public office, destroying his professional reputation, and destroying his ability to earn a living as a lawyer. It must be understood *470by all that, with this decision, in Minnesota, a newspaper can publish a story with actual malice that (1) intentionally juxtaposes the facts in order to lead a reader to reach a preconceived defamatory conclusion not warranted by those facts and (2) omits significant facts that would lead to a different conclusion if those facts were not omitted. Yet, says the majority, the newspapers cannot be sued for libel in Minnesota because “there is no such thing as a false idea” and, therefore, we do not recognize falsity by implication.
I doubt that the United States Supreme Court ever intended such a result by its decisions in New York Times v. Sullivan and Gertz v. Robert Welch, Inc. To the contrary, in Harte-Hanks, decided only last year, the court specifically said: “We have not gone so far, however, as to accord the press absolute immunity in its coverage of public figures or elections. If a false and defamatory statement is published with knowledge of falsity or a reckless disregard for the truth, the public official may prevail.” Harte-Hanks, — U.S. at -, 109 S.Ct. at 2696 (citing Curtis Publishing Co. v. Butts, 388 U.S. at 162, 87 S.Ct. at 1995). In Minnesota, however, this is not true any longer. The majority decision virtually closes the door to a public official for any recovery. This is because no editor will set forth all the material facts and expressly charge that a public official is guilty of malfeasance or misfeasance. No, every charge will simply be drafted in the form of an “opinion.”
Can it truly be said that this status of the law is better than the pre-New York Times rule? I think not. What led to the New York Times decision was a fear that newspapers would be subject to increasing numbers of lawsuits unless a stricter standard applied to plaintiffs suing newspapers. Instead, the result has been endless confusion in the law and a growth in the number of sensational stories, as well as publications, which pander to the public’s insatiable desire for gossip and sensationalism. If a return to the gre-Times status of the law would result in a more responsible press, it might well be worthwhile to overrule it.
Today, purveyors of scandal routinely blend facts with opinion so that, all too frequently, the stories appearing on the front page are themselves thinly veiled editorials. These “journalists” generate substantial profits by trading on the reputations of public figures. They should be required to absorb the occasional jury verdict in favor of a defamation victim as an ordinary cost of engaging' in this type of business. As interpreted, the New York Times decision unfairly denies public officials a means of protecting their reputations. Thus, it is true once again that giving one element of society a special privilege denies important rights to others. As to the risk of more lawsuits, that threat can be eased by a more careful pre-publication scrutiny of the facts.
I note that, before the New York Times decision, the press was not unduly restricted from publishing stories that were injurious to the reputations of public officials. Cf. Near v. Minnesota, 283 U.S. 697, 719-23, 51 S.Ct. 625, 632-33, 75 L.Ed. 1357 (1931) (5-4 decision) (prohibiting prior restraints). The media was able to divulge that a President of the United States had a child out of wedlock, disclose the Teapot Dome scandal of the 1920’s, the Vicuna Coat stories of the 1950’s and countless other scandals. Effective libel law has long been the primary check on the power of the press. In Near, the Court stated: “Public officers, whose character and conduct remain open to debate and free discussion in the press, find their remedies for false accusations in actions under libel laws providing for redress and punishment, and not in proceedings to restrain the publication of newspapers and periodicals.” Id. 283 U.S. at 718-19, 51 S.Ct. at 631-32 (emphasis added).
The majority’s decision today relieves the press from subsequent punishment and, in so doing, unnecessarily upsets the delicate, but necessary balance between the freedom of the press and the liberty of individuals. Perhaps it is time for the United States Supreme Court to take note of the consequences of the New York Times rule and how lower courts are applying that rule.
*471I thus dissent most vigorously from the majority’s interpretation of the law in New York Times v. Sullivan, Gertz v. Robert Welch, Inc., and Harte-Hanks Communications v. Connaughton and the decision made as a result of those interpretations. I believe that the decision is wrong in its interpretation of first amendment law, wrong in its interpretation of state defamation law, wrong in invading the province of the jury, and wrong as a matter of public policy. For all of these reasons, I would affirm the court of appeals.
. This denial of petition for certiorari is ambiguous. It could be interpreted as endorsing the Price court’s use of the Janklow/OUman opinion/fact distinction, or it could mean that the Supreme Court felt that the outcome of the case would be the same (judgment for defendant) under the actual-malice review suggested by Harte-Hanks.
. The majority opinion incorrectly suggests that this dissent is based on "speculation and dissenting opinions." Majority op. at n. 1. As the following analysis makes abundantly clear, this dissent is premised on the Harte-Hanks majority opinion which, unlike Price, is binding precedent on the constitutional issues presented. In circumventing this binding authority, the majority ignores the elementary rule that a denial of certiorari is not a decision on the merits in an attempt to justify its misplaced reliance on Price.
. The majority opinion places great weight on the paragraph-by-paragraph cross-examination of Diesen wherein no specific false statements were identified. This reliance is misplaced because it was the omission of material facts that made the articles substantially false and, therefore, actionable.
. In his concurrence in Oilman, Judge Bork observed that the United States Supreme Court has not adopted a rule employing the opinion/fact dichotomy, noted the sharp scholarly criticism of the rule, and stated that a "significant minority of jurisdictions” reject the opinion/fact dichotomy as “unworkable." Ollman, 750 F.2d at 1001 n. 6.
. In his concurring/dissenting opinion in this case, Justice Simonett basically adopts the approach offered by Judge Robinson in Oilman. In discussing the fact/opinion dichotomy, Judge Robinson refers to a continuum with statements of pure opinion at one extreme and statements of pure fact at the other. Ollman, 750 F.2d 970, 1021 (Robinson, J., with whom Wright, J., joins, dissenting in part). Judge Robinson says that, along this continuum, are "statements that reflect the author’s deductions or evaluations but are ‘laden with factual content.' " Id. at 1022 (quoting Cianci v. New Times Publishing Co., 639 F.2d at 63) (emphasis added).
Judge Robinson calls these statements “hybrid statements" and says that they "differ from pure opinion in that most people would regard them as capable of denomination as true or false, depending upon what the background facts are revealed to be.” Id.
Judge Robinson recognized that a "defamatory communication may consist of a statement in the form of an opinion, but a statement of this nature is actionable only if it implies the allegation of undisclosed defamatory facts as the basis for the opinion." Id. at 1027 (citing Restatement (Second) of Torts § 566 (1977)). Judge Robinson concluded that this rule should extend to cover hybrid statements accompanied by an incomplete disclosure of background facts. Id.
My basic disagreement with Justice Simonett is that he too unreasonably minimizes the role of the jury. A jury, after hearing the relevant testimony, observing the demeanor of the witnesses, and making credibility assessments, is infinitely better equipped than an appellate court to make a finding of fact as to the element of falsity. In Lewis v. Equitable Life Assurance Society of the U.S., 389 N.W.2d 876 (Minn.1986), this court stated: "[T]he truth or falsity of a statement is inherently within the province of the jury. This court will not overturn a jury finding on the issue of falsity unless the finding is manifestly contrary to the evidence." Id. at 889.
. This portion of Hessburg’s testimony, along with the testimony which indicated that portions of Hessburg's taped interviews were missing and other portions had been rearranged, tends to support the jury's implicit conclusions regarding Hessburg’s credibility.
. The interview began on August 21, 1981, at approximately 1:30 p.m. and ended shortly after midnight.