1 New York Times Co. v. Sullivan, 376 U.S. 254 (1964).Google Scholar

2 The American common law of defamation was drawn largely from seventeenth- and eighteenth-century English law. It provides, in brief, that the defendant becomes liable for damages for making a statement that lowers the plaintiff's reputation in the eyes of a segment of the community. This statement usually relates to the character or morals of the plaintiff, but it may also extend to matters beyond the plaintiff's control, such as illegitimate birth or having been raped.Google Scholar

Until recently, all speakers were strictly liable for the impacts of their statements. For example, if the defendant believed the he was writing about a fictitious person but a real person with the name given to the fictitious person could reasonably have been thought by readers to have been the defamed subject in the story, the real person could sue. Similarly, a false story about a real person led to liability no matter how carefully researched it had been.Google Scholar

Finally, the action has been encumbered with some complex damage rules, including one that sometimes makes it more difficult to sue for oral defamation (slander) than for written defamation (libel) even though the words are the same.Google Scholar

At common law, a defendant who can prove that his statement was true has a complete defense. The other major defenses, called privileges, fall into two groups. The common law has deemed it so important that high government officials feel entirely free to discuss matters that these officials are accorded absolute privileges. This means that they cannot be held liable for defamation even if they knowingly tell lies about other people. See text at note 72 infra. More commonly, the law accords conditional privileges to encourage persons to speak in certain situations in which the lawmakers think it appropriate that rumors and other unverified statements be conveyed. The speaker in such a situation is protected if he honestly, or in some cases, reasonably, believes the truth of what he is saying and the speech occurs in a situation for which a conditional protection has been developed.Google Scholar

As this brief summary barely begins to suggest, the common law was and is quite complex. The intricacies are not essential to the substance of this paper. To learn more about common law defamation-and its recent constitutional developments, see William 1. Prosser, Handbook of the Law of Torts 737-801 (4th ed. St. Paul, Minn.: West Publishing Co., 1971); Restatement (Second) of Torts §§ 558-623 (1977). Recent books devoted exclusively to defamation include Laurence H. Eldredge, The Law of Defamation (Charlottesville, Va.: Bobbs-Merrill, 1978), and Clarence Morris, Modern Defamation Law (Philadelphia: American Law Institute-American Bar Association Committee on Continuing Professional Education, 1978).Google Scholar

3 In New York Times Co. v. Sullivan, 376 U.S. 254 (1964), the Supreme Court for the first time held that constitutional considerations limited the freedom of the states in developing and enforcing defamation law.Google Scholar

In New York Times, the Court ruled that an elected public official suing a newspaper and persons who placed an advertisement in that newspaper must show that the defendants knew that the defamatory statement was false or that they published it with reckless disregard for its truth or falsity. This showing must be made with “convincing clarity” (at 285-86) instead of the usual requirement that civil plaintiffs prove their cases simply by a preponderance of the evidence. No damages whatever may be awarded unless plaintiff can make the required showing. If such a showing is made, the state is free to award whatever damages it wishes to allow, including presumed but unproven general damages to reputation.Google Scholar

4 The ruling of New York Times, id., was extended to appointed public officials in Rosenblatt v. Baer, 383 U.S. 75 (1966), and to “public figures” in Curtis Publishing Co. v. Butts and Associated Press v. Walker, 388 U.S. 130 (1967). The term “public figure,” which plays an important role in the constitutional discussion, has been the subject of several Supreme Court decisions in the last decade. At the outset, the phrase extended to persons who, though not on the public payroll in highly visible positions, were nevertheless important in public life and were the objects of public attention and discussion. Butts involved an athletic director at a major university accused of giving his school's game plan to the coach of an opposing teAm. Walker involved a retired army general, active in opposing desegregation, who was accused of leading students in violence against federal marshals who were enforcing court-ordered desegregation at the University of Mississippi.Google Scholar

Subsequent developments are discussed in notes 7-9 infra. Google Scholar

5 The recent emergence of cultgroups as vigorous and frequent plaintiffs in defamation cases may bring about a change here.Google Scholar

6 Specialization in defamation is difficult in any event because of the small number of cases involved. West states that in 1978 its Key Number System reported “approximately 47,382” cases. The majority were civil, although no precise breakdown was available. Letter from James D. Coates to author, Sept. 21, 1979. Our study identified 171 defamation cases (.36 percent of all cases reported by West). See note 21 infra. Google Scholar

Indicating that libel litigation is such a small percentage of all litigation—“probably less than 1 percent”—a recent article stated that few defense lawyers have developed national reputations in this area. “Successful plaintiff's attorneys are rarer still. Because of the difficulty in overcoming constitutional protections and the tenacity with which the media will fight through the appellate process, no attorney has been able to make a living only on plaintiff's libel cases—at least up until now.” See Been Defamed? Don't Call on These Guys, Nat'l L.J., Apr. 21, 1980, at 27.Google Scholar

7 Gertz v. Robert Welch, Inc., 418 U.S. 323 (1974). In Gertz, the Supreme Court concluded that “private persons” should not be subjected to the same constitutional rigors in their defamation cases as the New York Times sequence had imposed on public officials and public figures. The rule that emerged in Gertz provided that private persons could recover damages from media defendants whose defamatory falsehoods were due simply to negligence rather than the deliberate or reckless falsity required by New York Times. A plaintiff who proved such negligence could recover “actual injury” damages, which included any out-of-pocket loss suffered plus any proven emotional distress or proven loss of reputation. Any plaintiff who wished to recover for loss of reputation without specific proof that such a loss had occurred would have to meet the New York Times test.Google Scholar

8 Wolston v. Reader's Digest Asr'n, 443 U.S. 157 (1979). In Gertz, 418 U.S. 323, the Court had justified its distinction between public and private plaintiffs on two grounds: (1) a public plaintiff was able to mount a counterattack in the press because of the public interest in the person or the story; and (2) public persons had voluntarily assumed the risk of public attention and comment and the resultant risk of inaccuracies.Google Scholar

In Wolston, the Court stressed the voluntary assumption of risk and concluded that a plaintiff who had become an object of public attention by being called and failing to appear before a grand jury investigating Soviet espionage and his subsequent conviction for contempt of court was still “private” for purposes of defamation law—and thus came within the Gertz rule rather than the New York Times rule. The plaintiff had taken no voluntary steps to enter the debate over Soviet espionage in the United States and had not attempted to influence opinions on this matter. To decide otherwise, said the Court, “would create an ‘open season’ for all who sought to defame persons convicted of a crime.”Id. at 158.Google Scholar

9 Hutchinson v. Proxmire, 443 U.S. 111 (1979). In Hutchinson, a research scientist sued a United States senator for disparaging comments about the scientist's work and the actions of government agencies in funding it. The scientist had not voluntarily undertaken to influence the outcome of any public debate over what types of research government agencies should sponsor, just as Wolston (443 U.S. 157) had taken no part in influencing the debate over the role of Soviet espionage in the United States. Simply applying for government grants did not make the plaintiff a public figure for defamation purposes.Google Scholar

10 Herbert v. Lando, 441 U.S. 153 (1979). In Herbert, the Supreme Court held that a plaintiff who had to meet the burdens of the New York Times rule could attempt, before trial, to learn the motivations of the defendants by asking them why they made certain investigations but not others and why they included certain material in the final product but not other material. The plaintiff also could obtain copies of intraoffice memoranda written by the participants during the preparation of the allegedly defamatory story.Google Scholar

11 The search yielded 67 cases in the Federal Supplement and 10 state trial court cases. Six came from New York, 3 from New Jersey, and I from Delaware.Google Scholar

12 A case involving the futile effort to enjoin a libel, indexed exclusively under Constitutional Law and Injunction, was missed. See Willing v. Mazzacone, 482 Pa. 377, 393 A.2d 1155 (1978), which was known to the author in another context. In other research we have uncovered no overlooked damage case.Google Scholar

13 Some recent intermediate appellate cases in the study might still be on appeal and might be reversed. Less than 10 percent of our cases have involved two appeals, and we expect no surge now. Reversals might run in either direction.Google Scholar

We have coded these cases as though the appeal went directly from the trial court to the highest appellate court. This may yield a few misleading procedural codings: if the intermediate court reversed the trial court only to be reversed by the highest court, this was coded as the appellate court affirming the decision of the trial court. We see no impact on any question discussing content, final result, or any other point discussed in this study.Google Scholar

14 See note 22 infra. Google Scholar

15 In addition to the available federal constitutional claims, important points of state defamation law are often raised in media cases because media may find it easier and faster to obtain dismissals for failures to state a claim rather than to rely on federal privilege. See part V infra. Google Scholar

16 The exceptions were said to occur approximately once every five years, in the experience of two large insurers. See, e.g., Gannett Sets Policy on Libel Cases, Editor & Publisher, July 28, 1979, at 4, reporting that Gannett “has advised its media lawyers that it frowns on out-of-court settlements in libel and invasion of privacy cases.”Google Scholar

The assistant general attorney for the New York Times reports: “Some publishers have found that a firm no settlement policy for libel and privacy lawsuits pays for itself by discouraging potential plaintiffs from commencing or continuing costly litigation.” Although this policy may cost more in particular cases, “the fact that the publisher is willing to incur this expense is probably the most persuasive factor to a plaintiff with limited resources.” But such a policy refers only to monetary settlements and “should not automatically preclude the settlement of claims where the plaintiff is seeking a vindication of his reputation rather than damages.” Katherine P. Darrow, Pre-Complaint Phase: Deterring and Handling Claims, in Richard N. Winfield (chairman), Libel Litigation 11, 30 (New York: Practising Law Institute, 1979). The largest out-of-court settlement in recent years involved the payment of $600,000 and a front- page apology running four paragraphs, by the Sun Francisco Examiner to Synanon and its founder. N.Y. Times, July 3, 1976, at 24, col. 1.Google Scholar

Discussion with one of the very few large writers of newspaper and broadcast libel insurance revealed that it has settled only a handful of cases before trial. All but one involved less than $10,000.Google Scholar

This posture may change if other states follow the lead of an Idaho trial court that ordered all defenses struck when the newspaper refused to reveal its confidential sources. The case, Sierra Life Ins. Co. v. Magic Valley Newspapers, Inc., 4 Media L. Rep. 1689 (1977, 1978). is on appeal in the Idaho Supreme Court. See N.Y. Times, Apr. 6, 1978, at A17, col. 1, and The Quill, May 1978, at 11.Google Scholar

Another approach is to hold that when a defendant refuses to identify “his sources of information upon a valid order of the court, there shall arise a presumption that the defendant had no source. This presumption may be removed by disclosure of the sources a reasonable time before trial.” Downing v. Monitor Pub. Co., 415 A.2d 683 (N.H. 1980). The court observed that the use of contempt citations “in no way aids the plaintiff in proving his case.”Google Scholar

Compare the action of some courts in upholding state law privileges against disclosure of sources in defamation cases. Although the defendant may lose the use of arguments that rely on such a source, the plaintiff may not proceed as though no such source existed. See, e.g., Steaks Unlimited, inc. v. Deaner, 623 F.2d 264 (3d Cir. 1980).Google Scholar

Liability insurance coverage for mass media is discussed in Lawrence Worrall, The Insurance Issue and Identifying Problem Areas of Libel Litigation, in Winfield, supra, at 251. Sample policies are presented at 299. See also note 20 infra. Google Scholar

17 Note that the defendants' success rate of 76 percent in these cases closely approximates the defendants' trial court success rate of 72 percent in our study of appeals. See text following note 31 infra. Google Scholar

18 Handelman v. Hustler Magazine, Inc., 469 F. Supp. 1053 (S.D.N.Y. 1979).Google Scholar

19 See table 25 n.b infra. Google Scholar

20 Even so, a large number of possible defamation cases abort at early stages. The experience of one major newspaper that chose to remain anonymous is suggestive. The state involved has a retraction statute—a statute that provides that a plaintiff who believes he has been defamed by some form of mass media must demand a retraction within a short period of time, typically a few weeks. If the retraction is demanded and denied, the case proceeds normally. If the demand for retraction is honored—or no demand is made—plaintiff may recover for proven pecuniary damages but not damages for presumed but unproven loss of reputation.Google Scholar

This newspaper reported that it received an average of 25 to 30 demands for retraction per year. Perhaps 1 in 5 resulted in a retraction. In 20 years only one suit has followed a retraction. One in 20 of the denials ripens into litigation. Only four cases have reached trial in 20 years. Any plaintiff's judgment is appealed. During the past 20 years, the paper made only one small payment without trial and appeal. The attrition is certainly greater than it would be in a state that had no retraction statute.Google Scholar

A large insurer of media defendants reports that some 25-30 percent of claims are dropped after a denial of liability, and a comparable number are dropped without payment after further pretrial proceedings. Virtually all the rest proceed to motions and some to trial. After an adverse result on motion or at trial, about half the plaintiffs appeal. Virtually all media defendants who lose a case do appeal. See note 16 supra. Google Scholar

21 We are attempting a snapshot of the entire period rather than seeking trends within the period. The following table shows the number of cases per year:Google Scholar

22 In 36 cases, multiple defendants were sued. To avoid giving such cases undue prominence, we coded each according to the major defendant. Twenty-eight, 9 involving only media defendants and 19 involving only nonmedia defendants, presented no coding problem here. Of the 8 involving at least 1 from each group, 6 were coded as media cases. Two were coded as nonmedia cases because the roles of the media were peripheral to the issues before the court. Thus, we have understated the total number of media involvements by 2.Google Scholar

The only other similar study of reported defamation litigation appeared after this study had begun. Geoffrey Palmer, Defamation and Privacy Down Under, 64 Iowa L. Rev. 1209 (1979) (a study of defamation litigation in Australia and New Zealand). Of 82 cases analyzed, 65 (79 percent) were brought against media defendants.Google Scholar

23 Only 39 cases discussed which party had brought the case to the federal courts. Of these, 29 were brought by plaintiffs and 10 by defendants. Twenty of the cases, 14 brought by plaintiffs and 6 by defendants, involved media defendants. The remaining 19 involved nonmedia defendants.Google Scholar

24 Considering only appeals, 76 (90 percent) of the 84 newspaper appeals were in state court as were all 6 radio appeals. The figures were lower for the more national media. Magazine publishers were in state court in only 9 (56 percent) of their 16 appeals, book publishers in 6 of their 10 appeals, and television defendants, including networks, in 16 (80 percent) of their 20 appeals. One miscellaneous media appeal involved a claim against a law reporter system for statements in an opinion it published. Lowenschuss v. West Publishing Co., 542 F.2d 180 (3d Cir. 1976).Google Scholar

National media were probably in state courts as often as they were because plaintiffs joined local stringers or other local defendants and thus avoided the requisite diversity. See e.g., Time, Inc. v. Firestone, 424 U.S. 448 (1976).Google Scholar

Although about 90 percent of nonmedia appeals were in state courts this rate varied among the larger categories of plaintiffs from 12 (75 percent) of 16 professionals and 36 (84 percent) of 43 managerial business plaintiffs to 18 (95 percent) of 19 law enforcement personnel, 48 (96 percent) of 50 business employees, and all 18 public school personnel plaintiffs. Public employees were likely to be suing their employers. Lower level business employees were likely to be suing local management, although managerial plaintiffs were more likely to be suing officials at the home office.Google Scholar

All large categories of nonmedia defendants were sued in state courts aproximately 90 percent of the time, except that past employers were sued in state court 98 percent of the time (44 out of 45 cases) and creditors were sued in state courts in all 13 cases.Google Scholar

25 The following table lists the states with the largest numbers of appellate decisions (including federal appeals) separated by media and nonmedia appeals:Google Scholar

Several larger states, such as Pennsylvania (4), Ohio (6). New Jersey (U), and Massachusetts (10) are missing. Pennsylvania and Massachusetts do rank high with I 1 and 10 appeals respectively. New Jersey provides only 5 cases and Ohio 3. Massachusets has a retraction statute; the other three do not. Of those states listed, California, Florida, Michigan, Oregon, and Washington have retraction statutes. Even if the presence of retraction statutes were relevant, it should affect only the bringing of media cases. Thus, California's, Florida's, and Oregon's low percentages of media appeals, as opposed to the expected 30 percent, may be due to the existence of a retraction statute, but the Michigan and Washington figures do not support this. (Although Texas has a statute on the subject, Tex. Rev. Civ. Stat. Ann. art. 5431 (Vernon 1958), it provides only that evidence of retraction is admissible and may affect damages.)Google Scholar

26 Plaintiffs made only three of these motions.Google Scholar

27 “Others,” such as summary judgment going to some issues but not to liability or motions addressed to discovery questions, account for the rest. The number of codable cases may vary because of the unavailability of data, especially on cross-tabulations. This will often be signaled by use of “codable” in the text.Google Scholar

In other situations, as in the text, the subdivisions will not add up to the whole. This will be true because of “other” or a variety of small categories that are not reported in detail.Google Scholar

28 The coded state law issues are listed in table 23.Google Scholar

29 There was no variation on this point between state and federal courts. Among media appeals, 12 (50 percent) of the 24 federal appeals and 56 (49 percent) of the 114 state appeals involved questions raised on motions for summary judgment. The Supreme Court was “constrained to express some doubt” about the idea that summary judgment is the appropriate stage for deciding Times cases. Hutchinson v. Proxmire, 443 U.S. 111, 120 n.9 (1979).Google Scholar

30 See note 39 infra. Google Scholar

31 Recall note 2 supra and see note 32 infra. Google Scholar

32 See text at notes 60 and 72 infra. CrossRefGoogle Scholar

33 This problem is particularly apparent in New York. See table 25 n.b infra. Google Scholar

34 The role of constitutional privilege is discussed in the text at note 77 infra. As indicated in the text at note 28 supra, constitutional privilege is less likely to play a role at the stage of the motion to dismiss than at later stages. Although the numbers in table 4 are small, the fact that the affirmance rates are closest for motions to dismiss may suggest the impact of constitutional developments.Google Scholar

35 The posttrial situation is analyzed in greater detail in tables 8 and 9 infra. Google Scholar

36 The figures are drawn from table 5.Google Scholar

The data on defendants' success year by year was curious. In 1976, media won 17 of 31 (55 percent) of their appeals. This rose to 22 of 36 (61 percent) in 1977 and to 38 of 50 (76 percent) in 1978 but then dropped dramatically to 6 of 21 (29 percent) in the first part of 1979 before Wolston (443 U.S. 157) and Hutchinson (443 US. 111). We cannot explain the 1979 drop, which is significant at the p < .05 level using a t-test accounting for multiple hypotheses. We have no reason to believe that decisions rendered during the first half of a calendar year should be any less favorable to media defendants than those rendered later that year.Google Scholar

Among the nonmedia appeals, no discernible trend emerged. The defendants' success rate was 59 percent in 1976, 55 percent in 1977, 63 percent in 1978, and 62 percent in the first part of 1979.Google Scholar

37 See text following notes 53 and 57 infra. CrossRefGoogle Scholar

38 Indeed, the larger media groups were fairly consistent across all stages. The following table shows the percentage of media appeals disposed of at each stage for the 137 codable appeals:Google Scholar

39 Much as looking at all cases in the study hid differences between media and nonmedia defend-ants, looking at nonmedia appeals as a group hides great differences in stage of disposition. The following table shows for each category of defendants the percentage of appeals decided at each stage for the 282 codable appeals:Google Scholar

40 See text at note 3 supra. Google Scholar

41 Patterns of suit are discussed in the text following note 59 infra. Google Scholar

42 As table 6 shows, two of these involved verdicts for zero dollars. The plaintiff appealed one of these and the defendant the other.Google Scholar

Five media and 29 nonmedia trials that reached appeal never reached verdict because the judges dismissed the cases.Google Scholar

43 When juries did decide for plaintiffs, the identifiable awards in the nonmedia cases yielded more extreme highs and lows than did the media cases. But both seem to have a center in the $25,001-$100,000 area.Google Scholar

Analysis of the cases withdrawn from the jury will not explain the disparity. In nonmedia appeals, of the 29 dismissals of jury trials before verdict, 20 were affirmed and 9 reversed. See table 9. Even if we assume that jurors would have decided all 29 of these cases for the defense, plaintiffs would have obtained favorable verdicts in 72 (65 percent) of 110 cases still above the judges' rate of 45 percent. The disparity in media cases also persists after this type of adjustment.Google Scholar

44 Among nonmedia cases, 71 percent of the jurors' awards exceeded $10,000. The comparable figure for judges was 22 percent. By another measure, the judges' median award is in the $1,001-$10,000 category, and the jurors' is in the $25,001-$100,000 category.Google Scholar

45 Rosen v. Reed, 351 So. 2d 1284 (La. App. 1977).Google Scholar

46 No annual trend for plaintiffs' victories appeared. Among media plaintiffs, two won each of the first three years and one won in early 1979, suggesting a rate of one each six months. Although the base of appeals was growing each year, the numbers are too small to analyze. Among nonmedia plaintiffs, the success rate was 15 percent in 1976, 10 percent in 1977 and 1978, and 12 percent in early 1979.Google Scholar

47 The relationship between awards rendered and judgments affirmed in nonmedia appeals is indicated in the following table. The “award” column includes jury actions before rulings by a trial judge. Although “affirm” generally involves the affirmance of a trial judgment that upheld the jury award, there are two exceptions. In one case a judgment notwithstanding verdict was overturned and the verdict reinstated. In the other, the trial judge increased the award, but the appellate court returned it to the level awarded by the jury.Google Scholar

Recall that one plaintiff received his award for the first time at the appellate level.Google Scholar

The zero verdict that was affirmed was the one plaintiff had appealed. The reversal occurred in the case defendant had appealed. The “affirmed” column includes several cases in which the trial judge or appellate court reduced the “award.” See table 8 n.b supra. These were treated as affirmances in the row that reflects the original award. The most significant were cases in which an award of $200,000 was reduced to $75,000 and one in which $225,000 was reduced to $35,000. Note that this exaggerates the affirmances in the $100,000-$250,000 category. Reallocating these two would produce a generally decreasing percentage of affirmances as the size of the award increases.Google Scholar

Even treating all reduction cases as affirmances, the median affirmed award is in the $10,001–$25,000 category. The median award is in the next higher category. Note that no award over $250,000 was affirmed.Google Scholar

In the media appeals, the plaintiff who won in a trial to the judge won $10,000. In the others, awards of $214,000 and $50,000 were affirmed; an award of $132,500 was reduced to $32,500; an award of $85,000, reduced to $15,000 by the trial judge, was affirmed; an award of $75,000 was reduced to $45,000 because of credit for a payment from a nonmedia defendant; and an award of $88,000, reduced to $75,000 by the trial judge, was modified and affirmed.Google Scholar

48 The base of 25 is derived from the 11 defense jury verdicts and the 13 defense judge verdicts that were affirmed and the 1 defense jury verdict that was reversed.Google Scholar

49 Although defense judgments were affirmed overwhelmingly without regard to trier of fact, the data from tables 8 and 9 suggest that, without regard to the winner, judge verdicts are harder to overturn on appeal than are judgments entered on jury verdicts. Perhaps this can be explained by the types of errors that are unique to jury-tried cases, such as erroneous charges to the jury or inadequate opportunity to use the voir dire process.Google Scholar

50 This includes the three nonmedia cases in which the trial court reduced the award. In all three, the appellate courts reinstated the jury awards and affirmed. Also included is the one nonmedia case in which the trial judge increased the punitive award. The appellate court returned the award to the jury's amount and affirmed.Google Scholar

51 State law may be more demanding before permitting punitive damages or may bar them entirely. See Wheeler v. Green, 286 Or. 99, 593 P.2d 777 (1979) (punitive damages in a defamation case are barred by the Oregon constitution). Four states (Louisiana, Massachusetts, Nebraska, and Washington) reject punitive damages in all cases. Victor B. Levit, Punitive Damages: Yesterday, Today and Tomorrow, 1980 Ins. L.J. 257.Google Scholar

52 Our focus will be on the generic subject of defamation. Libel formed the basis for the complaint in 316 of the 534 cases; slander in 99. Both were involved in another 49 cases. The rest were silent on the matter. Four defendants disputed plaintiff's claim that libel, rather than slander, was involved. In 14 cases, libel plaintiffs were required to prove special damages. As might be expected, slander was almost exclusively a nonmedia phenomenon 98 of 99 cases. When the claim was slander, relatively few of the cases discussed whether it was slander per se 19 found slander per se, 16 rejected the per se claim. Nine plaintiffs, including two after trial, lost their cases for failure to allege or show special damages. In short, not enough differences emerged between libel and slander to warrant treating them separately.Google Scholar

53 Note that after the 54 corporate plaintiffs are deducted from the 534 cases, 480 cases remain. As indicated, only 363 of those were codable because many of the opinions that went off on procedural and other nonsubstantive issues detailed few facts. The percentages here are based on 363 rather than 480.Google Scholar

54 Compare the suits by public officials with data in a study of defamation cases in Australia and New Zealand between 1969 and 1978. The 86 cases coded revealed that 13 (16 percent) were brought by elected public officials and candidates, and 12 (14 percent) were brought by appointed public of ficials and those seeking such offices. Palmer, supra note 22, at 1215.Google Scholar

55 Professionals suing media reached trial at a high rate. Although the overall rate of reaching trial for codable plaintiffs was 28 percent, professionals reached trial in 7 (47 percent) of their 15 cases, and 2 of them finally prevailed. Other groups tended to reflect the general pattern of stages of disposition.Google Scholar

56 The other numerically significant winning groups were business employees (6 of 50) and professionals (4 of 29).Google Scholar

57 The remainder included I5 present employers, 9 co-employees, 10 unions or officers, 8 credit reporting agencies, 14 creditors, 3 neighbors, 30 corporations otherwise unidentified, 30 others, and 37 not available.Google Scholar

58 Although the success rate for government defendants was near the general average, 60 percent, when we consider stages of disposition against parties we find that government defendants, at 25 percent, were far below the overall nonmedia trial rate of 42 percent. The trial rate varied from 17 percent for suits by miscellaneous state and local employees and owners and managers, to 33 percent for cases brought by school personnel, to six (60 percent) of the ten cases brought by law enforcement personnel. Although reaching trial frequently, law enforcement personnel, as noted, have had little success.Google Scholar

Among the few patterns large enough to say much about, the only other significant deviation from the 42 percent nonmedia trial rate appeared in cases brought against past employers in which 16 (53 percent) of the 30 went to trial, presumably because of fact issues concerning the conditional privilege.Google Scholar

59 The other codable losers in nonmedia appeals were co-employees (2), credit-reporting companies (l), corporations (2), and other (1).Google Scholar

60 Although we also investigated the earlier status of plaintiffs when the defamation referred to an earlier period in the plaintiff's life, this involved only 29 cases 13 media and 16 nonmedia and produced groups too small to be revealing.Google Scholar

61 We dropped 8 media and 18 nonmedia appeals here because they involved combinations of charges that we could not identify further, but which might have included one of the three.Google Scholar

62 Accusations that plaintiff held unpopular ideas accounted for only two cases in each group.Google Scholar

63 The following chart discloses the media distribution:Google Scholar

64 The plausibility of this allocation is strengthened when we note that the plaintiffs' success rate in the business moral failing group (3 of 14 = 21 percent) closely matches the success rate in the combined group (2 of 11 = 18 percent). Also, the defense success rate of 45 percent in the combined charge is closer to the 50 percent rate of moral failing in business than to the rate for incompetence.Google Scholar

65 This number derives from the 14 single charges of moral failing in business, plus the 2 single charges that cannot be identified, plus the 11 combined charges and the 2 combined charges involving moral failings that could not be further identified, for a total of 29.Google Scholar

66 It is not obvious where the combined charges of moral failing and incompetence should be allocated. We have allocated them to incompetence because it is the larger group (62 to 35 or 39) and the one that comes closer to approximating the success rates of the combined charges.Google Scholar

67 One case involved an accusation of incompetence in trade or profession with an accusation of sexual misbehavior. The other involved a charge of noncriminal moral failing combined with a serious physical, emotional, or intellectual failing.Google Scholar

68 See note 66 supra. Google Scholar

69 But this would not explain why the media losses cluster here. Recall the explanation of privileges in note 2 supra. One example here is a conditional privilege for present employers to discuss among themselves the conduct of their employees. Another is the conditional privilege of past employers to respond to inquiries about former employees made by prospective new employers.Google Scholar

70 For this section, the combined charges involving crime are being allocated to the single charge of crime. In media cases the other combined charge is being allocated to moral failings. In nonmedia cases that category is allocated to incompetence. See notes 64 and 66 supra. Google Scholar

71 The record libel privilege, which exists in all states, protects a defendant who has accurately reported what happened in official government proceedings or, in some cases, in open meetings of private groups. Recently one court created a constitutional privilege for fair reports of public disputes, whether or not they occurred in government meetings. In this respect, the court expanded upon the record libel privilege. Edwards v. National Audubon Soc'y, Inc., 556 F.2d 113 (2d Cir.), cert. denied, 434 U.S. 1002 (1977). The Audubon approach was limited to public figure plaintiffs in Dixson v. Newsweek, Inc., 562 F.2d 626 (10th Cir. 1977), and rejected outright in Dickey v. CBS, Inc., 583 F.2d 1221 (3d Cir. 1978). See also Kathryn Dix Sowle, Defamation and the First Amendment: The Case for a Constitutional Privilege of Fair Report, 54 N.Y.U.L. Rev. 469 (1979).Google Scholar

72 Throughout this section, “asserted” and “raised” refer to situations in which the court's opinion reveals that the particular claim was made. It is, of course, quite likely that in many cases the defendants made claims that were not discussed, either because they were too weak to warrant discussion, because some other defense disposed of the case, or because such a defense was premature given the stage of litigation.Google Scholar

73 Both types of privilege were asserted in 20 cases.Google Scholar

74 A miscellany of “other” state defenses were pleaded in 13 percent of the media and 14 percent of the nonmedia cases.Google Scholar

75 Combining the numbers in the last two notes to table 20 we find that 56 of the 112 rulings recognizing conditional privilege occurred after trial and an equal number before trial.Google Scholar

Of the 52 defendants who received absolute privileges, 27 got them on motions to dismiss, 19 on motions for summary judgment, and only 6 after trial. The 27 rejections were equally divided among the three stages. Thus, 46 of 52 got the privilege before trial and 18 of the 27 who were denied learned the result without trial.Google Scholar

76 Wheeler v. Green, 286 Or. 99, 593 P.2d 777 (1979) (state constitution bars punitive damages in defamation cases). see also Blubaugh, Ronald, California's Constitution: The Reporter's Forgotten Ally, 55 Cal. St. B.J. 12 (1980).Google Scholar

77 Four cases raised both Times and Audubon privileges. The Audubon privilege is discussed in note 71 supra. The Times case is discussed in note 3 supra and the Gertz case in note 7 supra. Google Scholar

78 Ten cases involved the federal labor law privilege, which provides that a plaintiff claiming to have been defamed during a labor dispute must prove his case by the New York Times standard. Letter Carriers v. Austin, 418 U.S. 264 (1974). A few other cases raised miscellaneous privileges, such as the speech or debate clause, U.S. Const. art I, § 6, which provides that members of Congress are immune from suit for what they say during debate. Analogous privileges protect state legislators.Google Scholar

79 Whether the Constitution requires that nonmedia defendants be given the Times or Gertz privilege is unclear. See Hutchinson v. Proxmire, 443 U.S. 111, 133 n.16 (1979).Google Scholar

80 There is some reason to suspect this from the discovered differences in coverage of corporations and women. Recall the text at note 52 supra. Google Scholar

81 Public plaintiffs lost in 63 percent of the 57 media appeals and in 58 percent of the 31 nonmedia appeals. Plaintiffs held to be private lost 31 percent of the 16 media appeals but none of the 6 nonmedia appeals.Google Scholar

82 In one of the three media cases in the Gertz column the state's standard was more rigorous than that required in Gertz. Dixson v. Newsweek, Inc., 562 F.2d 626 (10th Cir. 1977) (using Colorado law).Google Scholar

83 Only 11 appellate courts decided whether plaintiffs had shown the “actual injury” damages required by Gertz. Six found them adequately proven, and 4 found them adequately alleged. Only 1 court found the showing inadequate.Google Scholar

84 See, e.g., Miami Herald Publishing Co. v. Tornillo, 418 U.S. 241, 258 (1974) (Brennan, J., concurring), and Stanley Ingber, Defamation: A Conflict Between Reason and Decency, 65 Va. L. Rev. 785 (1979).CrossRefGoogle Scholar

85 Of the 191 codable nonmedia defense victories, truth was established in 9 (5 percent). Truth or lack of a showing of falsity controlled in 5 (6 percent) of the 83 media successes.Google Scholar

86 Continental Casualty Company is offering this program through Media/Professional Insurance, Inc. This media special perils policy is a program sponsored by the National Newspaper Association for its members.Google Scholar

87 For this analysis we removed one nonmedia case that was decided by a state court in the group of nine but was based on another state's substantive law. No media case was added to or removed from the study.Google Scholar

88 Florida was also high in total defamation cases. See note 25 supra. Google Scholar

89 Alabama and California also have retraction statutes.Google Scholar

90 The nonmedia analysis is as shown in the following table:Google Scholar

91 Our study found no disproportion of litigation. The ratios of media and nonmedia appeals in the nine states to those in all states were both 17 percent.Google Scholar

92 See, e.g., Montandon v. Triangle Publishing Co., 45 Cal. App. 3d 938, 120 Cal. Rptr. 186, cert. denied, 423 U.S. 893 (1975) ($150,000 general damages plus S1,000 punitive damages); Bindrim v. Mitchell, 92 Cal. App. 3d 61, 155 Cal. Rptr. 29, cert. denied, 444 U.S. 984 (1979) (Justices Brennan, Stewart, and Marshall dissenting) ($80,000 general damages against author and publisher plus $25,000 punitive award against publisher); Alioto v. Cowles Communications, Inc., 430 F. Supp. 1363 (N.D. Cal. 1977) ($350,000 in general damages appeal pending). In addition, we have the $600,000 out-of-court settlement that Synanon obtained from the Sun Francisco Examiner in July 1976. See note 16 supra. Google Scholar

The nonmedia side of California defamation law offers similar examples. Field Research Corp. v. Patrick, 30 Cal. App. 3d 603, 106 Cal. Rptr. 473, cert. denied, 414 U.S. 922 (1973) ($150,000 compensatory damages and equal amount in punitive damages).Google Scholar

93 These figures are based on the entire appellate group. Looking only at those cases in which plaintiffs obtained judgment at the trial level, we found that plaintiffs kept their awards in 7 (37 percent) of 19 media appeals and in 37 (51 percent) of 73 nonmedia appeals.Google Scholar

94 See note 84 supra. Google Scholar

95 Further work on this subject is in progress.Google Scholar