George A. ROSENBLOOM, Petitioner, v. METROMEDIA, INC.

In a series of cases beginning with New York Times Co. v. Sullivan, 376 U.S. 254, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964), the Court has considered the limitations upon state libel laws imposed by the constitutional guarantees of freedom of speech and of the press. New York Times held that in a civil libel action by a public official against a newspaper those guarantees required clear and convincing proof that a defamatory falsehood alleged as libel was uttered with 'knowledge that it was false or with reckless disregard of whether it was false or not.' Id., at 280, 84 S.Ct., at 726. The same requirement was later held to apply to 'public figures' who sued in libel on the basis of alleged defamatory falsehoods. The several cases considered since New York Times involved actions of 'public officials' or 'public figures,' usually, but not always, against newspapers or magazines.1 Common to all the cases was a defamatory falsehood in the report of an event of 'public or general interest.'2 The instant case presents the question whether the New York Times' knowing-or-reckless-falsity standard applies in a state civil libel action brought not by a 'public official' or a 'public figure' but by a private individual for a defamatory falsehood uttered in a news broadcast by a radio station about the individual's involvement in an event of public or general interest.3 The District Court for the Eastern District of Pennsylvania held that the New York Times standard did not apply and that Pennsylvania law determined respondent's liability in this diversity case, 289 F.Supp. 737 (1968). The Court of Appeals for the Third Circuit held that the New York Times standard did apply and reversed the judgment for damages awarded to petitioner by the jury. 415 F.2d 892 (1969). We granted certiorari, 397 U.S. 904, 90 S.Ct. 917, 25 L.Ed.2d 85 (1970). We agree with the Court of Appeals and affirm that court's judgment.

* In 1963, petitioner was a distributor of nudist magazines in the Philadelphia metropolitan area. During the fall of that year, in response to citizen complaints, the Special Investigations Squad of the Philadelphia Police Department initiated a series of enforcement actions under the city's obscenity laws. The police, under the command of Captain Ferguson, purchased various magazines from more than 20 newsstands throughout the city. Based upon Captain Ferguson's determination that the magazines were obscene,4 police on October 1, 1963, arrested most of the newsstand operators5 on charges of selling obscene material. While the police were making an arrest at one newsstand, petitioner arrived to deliver some of his nudist magazines and was immediately arrested along with the newsboy.6 Three days later, on October 4, the police obtained a warrant to search petitioner's home and the rented barn he used as a warehouse, and seized the inventory of magazines and books found at these locations. Upon learning of the seizures, petitioner, who had been released on bail after his first arrest, surrendered to the police and was arrested for a second time.

Following the second arrest, Captain Ferguson telephoned respondent's radio station WIP and another local radio station, a wire service, and a local newspaper to inform them of the raid on petitioner's home and of his arrest. WIP broadcast news reports every half hour to the Philadelphia metropolitan area. These news programs ran either five or ten minutes and generally contained from six to twenty different items that averaged about thirty seconds each. WIP's 6 p.m. broadcast on October 4, 1963, included the following item:

'City Cracks Down on Smut Merchants

'The Special Investigations Squad raided the home of George Rosenbloom in the 1800 block of Vesta Street this afternoon. Police confiscated 1,000 allegedly obscene books at Rosenbloom's home and arrested him on charges of possession of obscene literature. The Special Investigations Squad also raided a barn in the 20 Hundred block of Welsh Road near Bustleton Avenue and confiscated 3,000 obscene books. Capt. Ferguson says he believes they have hit the supply of a main distributor of obscene material in Philadelphia.' This report was rebroadcast in substantially the same form at 6:30 p.m., but at 8 p.m. when the item was broadcast for the third time, WIP corrected the third sentence to read 'reportedly obscene.' News of petitioner's arrest was broadcast five more times in the following twelve hours, but each report described the seized books as 'allegedly' or 'reportedly' obscene. From October 5 to October 21, WIP broadcast no further reports relating to petitioner.

On October 16 petitioner brought an action in Federal District Court against various city and police officials and against several local news media.7 The suit alleged that the magazines petitioner distributed were not obscene and sought injunctive relief prohibiting further police interference with his business as well as further publicity of the earlier arrests. The second series of allegedly defamatory broadcasts related to WIP's news reports of the lawsuit. There were ten broadcasts on October 21, two on October 25, and one on November 1. None mentioned petitioner by name. The first at 6:30 a.m. on October 21 was pretty much like those that followed:

'Federal District Judge Lord, will hear arguments today from two publishers and a distributor all seeking an injunction against Philadelphia Police Commissioner Howard Leary * * * District Attorney James C. Crumlish * * * a local television station and a newspaper * * * ordering them to lay off the smut literature racket.

'The girlie-book peddlers say the police crackdown and continued reference to their borderline literature as smut or filth is hurting their business. Judge Lord refused to issue a temporary injunction when he was first approached. Today he'll decide the issue. It will set a precedent * * * and if the injunction is not granted * * * it could signal an even more intense effort to rid the city of pornography.'

On October 27, petitioner went to WIP's studios after hearing from a friend that the station had broadcast news about his lawsuit. Using a lobby telephone to talk with a part-time newscaster, petitioner inquired what stories WIP had broadcast about him. The newscaster asked him to be more specific about dates and times. Petitioner then asked for the noon news broadcast on october 21, 1963, which the newscaster read to him over the phone; it was similar to the above 6:30 a.m. broadcast. According to petitioner, the ensuing interchange was brief. Petitioner told the newscaster that his magazines were 'found to be completely legal and legitimate by the United States Supreme Court.' When the newscaster replied the district attorney had said the magazines were obscene, petitioner countered that he had a public statement of the district attorney declaring the magazines legal. At that point, petitioner testified, 'the telephone conversation was terminated * * *. He just hung up.' Petitioner apparently made no request for a retraction or correction, and none was forthcoming. WIP's final report on petitioner's lawsuit—the only one after petitioner's unsatisfactory conversation at the station—occurred on November 1 after the station had checked the story with the judge involved.8

In May 1964 a jury acquitted petitioner in state court of the criminal obscenity charges under instructions of the trial judge that, as a matter of law, the nudist magazines distributed by petitioner were not obscene. Following his acquittal, petitioner filed this diversity action in District Court seeking damages under Pennsylvania's libel law. Petitioner alleged that WIP's unqualified characterization of the books seized as 'obscene' in the 6 and 6:30 p.m. broadcasts of October 4, describing his arrest, constituted libel per se and was proved false by petitioner's subsequent acquittal. In addition, he alleged that the broadcasts in the second series describing his court suit for injunctive relief were also false and defamatory in that WIP characterized petitioner and his business associates as 'smut distributors' and 'girlie-book peddlers' and, further, falsely characterized the suit as an attempt to force the defendants 'to lay off the smut literature racket.'

At the trial WIP's defenses were truth and privilege. WIP's news director testified that his eight-man staff of reporters prepared their own newscasts and broadcast their material themselves, and that material for the news programs usually came either from the wire services or from telephone tips. None of the writers or broadcasters involved in preparing the broadcasts in this case testified. The news director's recollection was that the primary source of information for the first series of broadcasts about petitioner's arrest was Captain Ferguson, but that, to the director's knowledge, the station did not have any further verification. Captain Ferguson testified that he had informed WIP and other media of the police action and that WIP had accurately broadcast what he told the station. The evidence regarding WIP's investigation of petitioner's lawsuit in the second series of broadcasts was even more sparse. The news director testified that he was 'sure we would check with the District Attorney's office also and with the Police Department,' but 'it would be difficult for me to specifically state what additional corroboration we had.' In general, he testified that WIP's half-hour deadlines required it to rely on wire-service copy and oral reports from previously reliable sources subject to the general policy that 'we will contact as many sources as we possible can on any kind of a story.'

Pennsylvania's libel law tracks almost precisely the Restatement (First) of Torts provisions on the subject. Pennsylvania holds actionable any unprivileged 'malicious'9 publication of matter which tends to harm a person's reputation and expose him to public hatred, contempt, or ridicule. Schnabel v. Meredith, 378 Pa. 609. 107 A.2d 860 (1954); Restatement of Torts §§ 558, 559 (1938). Pennsylvania law recognizes truth as a complete defense to a libel action. Schonek v. WJAC, Inc., 436 Pa. 78, 84, 258 A.2d 504, 507 (1969); Restatement of Torts § 582. It recognizes an absolute immunity for defamatory statements made by high state officials, even if published with an improper motive, actual malice, or knowing falsity. Montgomery v. Philadelphia, 392 Pa. 178, 140 A.2d 100 (1958); Restatement of Torts § 591, and it recognizes a conditional privilege for news media to report judicial, administrative, or legislative proceedings if the account is fair and accurate, and not published solely for the purpose of causing harm to the person defamed, even though the official information is false or inaccurate. Sciandra v. Lynett, 409 Pa. 595, 600—601, 187 A.2d 586, 588—589 (1963); Restatement of Torts § 611. The conditional privilege of the news media may be defeated, however, by "want of reasonable care and diligence to ascertain the truth, before giving currency to an untrue communication.' The failure to employ such 'reasonable care and diligence' can destroy a privilege which otherwise would protect the utterer of the communication.' Purcell v. Westinghouse Broadcasting Co., 411 Pa. 167, 179, 191 A.2d 662, 668 (1963). Pennsylvania has also enacted verbatim the Restatement's provisions on burden of proof, which place the burden of proof for the affirmative defenses of truth and privilege upon the defendant.10

At the close of the evidence, the District Court denied respondent's motion for a directed verdict and charged the jury, in conformity with Pennsylvania law, that four findings were necessary to return a verdict for petitioner: (1) that one or more of the broadcasts were defamatory; (2) that a reasonable listener would conclude that the defamatory statement referred to petitioner; (3) that WIP had forfeited its privilege to report official proceedings fairly and accurately, either because it intended to injure the plaintiff personally or because it exercised the privilege unreasonably and without reasonable care; and (4) that the reporting was false. The jury was instructed that petitioner had the burden of proof on the first three issues, but that respondent had the burden of proving that the reporting was true. The jury was further instructed that 'as a matter of law' petitioner was not entitled to actual damages claimed for loss of business 'not because it wouldn't ordinarily be but because there has been evidence that this same subject matter was the subject' of broadcasts over other television and radio stations and of newspaper reports, 'so if there was any business lost * * * we have no proof * * * that (it) resulted directly from the broadcasts by WIP * * *.' App. 331a. On the question of punitive damages, the judge gave the following instruction:

'(I)f you find that this publication arose from a bad motive or malice toward the plaintiff, or if you find that it was published with reckless indifference to the truth, if you find that it was not true, you would be entitled to award punitive damages, and punitive damages are awarded as a deterrent from future conduct of the same sort.

'They really are awarded only for outrageous conduct, as I have said, with a bad motive or with reckless disregard of the interests of others, and before you would award punitive damages you must find that these broadcasts were published with a bad motive or with reckless disregard of the rights of others, or reckless indifference to the rights of others * * *.'

The jury returned a verdict for petitioner and awarded $25,000 in general damages, and $725,000 in punitive damages. The District Court reduced the punitive damages award to $250,000 on remittitur, but denied respondent's motion for judgment n.o.v. In reversing, the Court of Appeals emphasized that the broadcasts concerned matters of public interest and that they involved 'hot news' prepared under deadline pressure. The Court of Appeals concluded that 'the fact that plaintiff was not a public figure cannot be accorded decisive importance if the recognized important guarantees of the First Amendment are to be adequately implemented.' 415 F.2d, at 896. For that reason, the court held that the New York Times standard applied and, further, directed that judgment be entered for respondent, holding that, as a matter of law, petitioner's evidence did not meet that standard.

Petitioner concedes that the police campaign to enforce the obscenity laws was an issue of public interest, and, therefore, that the constitutional guarantees for freedom of speech and press imposed limits upon Pennsylvania's power to apply its libel laws to compel respondent to compensate him in damages for the alleged defamatory falsehoods broadcast about his involvement. As noted, the narrow question he raises is whether, because he is not a 'public official' or a 'public figure' but a private individual, those limits required that he prove that the falsehoods resulted from a failure of respondent to exercise reasonable care, or required that he prove that the falsehoods were broadcast with knowledge of their falsity or with reckless disregard of whether they were false or not. That question must be answered against the background of the functions of the constitutional guarantees for freedom of expression. Rosenblatt v. Baer, 383 U.S. 75, at 84—85, n. 10, 86 S.Ct. 669, at 675, 15 L.Ed.2d 597 (1966).

Self-governance in the United States presupposes far more than knowledge and debate about the strictly official activities of various levels of government. The commitment of the country to the institution of private property, protected by the Due Process and Just Compensation Clauses in the Constitution, places in private hands vast areas of economic and social power that vitally affect the nature and quality of life in the Nation. Our efforts to live and work together in a free society not completely dominated by governmental regulation necessarily encompass far more than politics in a narrow sense. 'The guarantees for speech and press are not the preserve of political expression or comment upon public affairs.' Time, Inc. v. Hill, 385 U.S. 374, 388, 87 S.Ct. 534, 542, 17 L.Ed.2d 456 (1967). 'Freedom of discussion, if it would fulfill its historic function in this nation, must embrace all issues about which information is needed or appropriate to enable the members of society to cope with the exigencies of their period.' Thornhill v. Alabama, 310 U.S. 88, 102, 60 S.Ct. 736, 744, 84 L.Ed. 1093 (1940).

Although the limitations upon civil libel actions, first held in New York Times to be required by the First Amendment, were applied in that case in the context of defamatory falsehoods about the official conduct of a public official, later decisions have disclosed the artificiality, in terms of the public's interest, of a simple distinction between 'public' and 'private' individuals or institutions:

'Increasingly in this country, the distinctions between governmental and private sectors are blurred. * * * In many situations, policy determinations which traditionally were channeled through formal political institutions are now originated and implemented through a complex array of boards, committees, commissions, corporations, and associations, some only loosely connected with the Government. This blending of positions and power has also occurred in the case of individuals so that many who do not hold public office at the moment are nevertheless intimately involved in the resolution of important public questions. * * *

'* * * Our citizenry has a legitimate and substantial interest in the conduct of such persons, and freedom of the press to engage in uninhibited debate about their involvement in public issues and events is as crucial as it is in the case of 'public officials." Curtis Publishing Co. v. Butts, 388 U.S. 130, 163—164, 87 S.Ct. 1975, 1996, 18 L.Ed.2d 1094 (1967) (Warren, C.J., concurring in the result).

Moreover, the constitutional protection was not intended to be limited to matters bearing broadly on issues of responsible government. '(T)he Founders * * * felt that a free press would advance 'truth, science, morality, and arts in general' as well as responsible government.' Id., at 147, 87 S.Ct., at 1987 (opinion of Harlan, J.). Comments in other cases reiterate this judgment that the First Amendment extends to myriad matters of public interest. In Time, Inc. v. Hill, supra, we had 'no doubt that the * * * opening of a new play linked to an actual incident, is a matter of public interest,' 385 U.S., at 388, 87 S.Ct., at 542, which was entitled to constitutional protection. Butts held that an alleged 'fix' of a college football game was a public issue. Associated Press v. Walker, 388 U.S. 130, 87 S.Ct. 1975, 18 L.Ed.2d 1094 (1967), a companion case to Butts, established that the public had a similar interest in the events and personalities involved in federal efforts to enforce a court decree ordering the enrollment of a Negro student in the University of Mississippi. Thus, these cases underscore the vitality, as well as the scope, of the 'profound national commitment to the principle that debate on public issues should be uninhibited, robust, and wide-open.' New York Times Co. v. Sullivan, 376 U.S., at 270—271, 84 S.Ct., at 721 (emphasis added).

If a matter is a subject of public or general interest, it cannot suddenly become less so merely because a private individual is involved, or because in some sense the individual did not 'voluntarily' choose to become involved. The public's primary interest is in the event; the public focus is on the conduct of the participant and the content, effect, and significance of the conduct, not the participant's prior anonymity or notoriety.11 The present case illustrates the point. The community has a vital interest in the proper enforcement of its criminal laws, particularly in an area such as obscenity where a number of highly important values are potentially in conflict: the public has an interest both in seeing that the criminal law is adequately enforced and in assuring that the law is not used unconstitutionally to suppress free expression. Whether the person involved is a famous large-scale magazine distributor or a 'private' businessman running a corner newsstand has no relevance in ascertaining whether the public has an interest in the issue. We honor the commitment to robust debate on public issues, which is embodied in the First Amendment, by extending constitutional protection to all discussion and communication involving matters of public or general concern, without regard to whether the persons involved are famous or anonymous.12

Our Brother WHITE agrees that the protection afforded by the First Amendment depends upon whether the issue involved in the publication is an issue of public or general concern. He would, however, confine our holding to the situation raised by the facts in this case, that is, limit it to issues involving 'official actions of public servants.' In our view that might be misleading. It is clear that there has emerged from our cases decided since New York Times the concept that the First Amendment's impact upon state libel laws derives not so much from whether the plaintiff is a 'public official,' 'public figure,' or 'private individual,' as it derives from the question whether the allegedly defamatory publication concerns a matter of public or general interest. See T. Emerson, The System of Freedom of Expression 531—532, 540 (1970). In that circumstance we think the time has come forthrightly to announce that the determinant whether the First Amendment applies to state libel actions is whether the utterance involved concerns an issue of public or general concern, albeit leaving the delineation of the reach of that term to future cases. As our Brother WHITE observes, that is not a problem in this case, since police arrest of a person for distributing allegedly obscene magazines clearly constitutes an issue of public or general interest.13

We turn then to the question to be decided. Petitioner's argument that the Constitution should be held to require that the private individual prove only that the publisher failed to exercise 'reasonable care' in publishing defamatory falsehoods proceeds along two lines. First, he argues that the private individual, unlike the public figure, does not have access to the media to counter the defamatory material and that the private individual, unlike the public figure, has not assumed the risk of defamation by thrusting himself into the public arena. Second, petitioner focuses on the important values served by the law of defamation in preventing and redressing attacks upon reputation.

We have recognized the force of petitioner's arguments, Time, Inc. v. Hill, supra, 385 U.S., at 391, 87 S.Ct., at 543—544, and we adhere to the caution expressed in that case against 'blind application' of the New York Times standard. Id., at 390, 87 S.Ct., at 543. Analysis of the particular factors involved, however, convinces us that petitioner's arguments cannot be reconcided with the purposes of the First Amendment, with our cases, and with the traditional doctrines of libel law itself. Drawing a distinction between 'public' and 'private' figures makes no sense in terms of the First Amendment guarantees.14 The New York Times standard was applied to libel of a public official or public figure to give effect to the Amendment's function to encourage ventilation of public issues, not because the public official has any less interest in protecting his reputation than an individual in private life. While the argument that public figures need less protection because they can command media attention to counter criticism may be true for some very prominent people, even then it is the rare case where the denial overtakes the original charge. Denials, retractions, and corrections are not 'hot' news, and rarely receive the prominence of the original story. When the public official or public figure is a minor functionary, or has left the position that put him in the public eye, see Rosenblatt v. Baer, supra, the argument loses all of its force. In the vast majority of libels involving public officials or public figures, the ability to respond through the media will depend on the same complex factor on which the ability of a private individual depends: the unpredictable event of the media's continuing interest in the story. Thus the unproved, and highly improbable, generalization that an as yet undefined class of 'public figures' involved in matters of public concern will be better able to respond through the media than private individuals also involved in such matters seems too insubstantial a reed on which to rest a constitutional distinction. Furthermore, in First Amendment terms, the cure seems far worse than the disease. If the States fear that private citizens will not be able to respond adequately to publicity involving them, the solution lies in the direction of ensuring their ability to respond, rather than in stifling public discussion of matters of public concern.15

Further reflection over the years since New York Times was decided persuades us that the view of the 'public official' or 'public figure' as assuming the risk of defamation by voluntarily thrusting himself into the public eye bears little relationship either to the values protected by the First Amendment or to the nature of our society. We have recognized that '(e)xposure of the self to others in varying degrees is a concomitant of life in a civilized community.' Time, Inc. v. Hill, supra, 385 U.S., at 388, 87 S.Ct., at 542. Voluntarily or not, we are all 'public' men to some degree. Conversely, some aspects of the lives of even the most public men fall outside the area of matters of public or general concern. See n. 12, supra; Griswold v. Connecticut, 381 U.S. 479, 85 S.Ct. 1678, 14 L.Ed.2d 510 (1965).16 Thus, the idea that certain 'public' figures have voluntarily exposed their entire lives to public inspection, while private individuals have kept theirs carefully shrouded from public view is, at best, a legal fiction. In any event, such a distinction could easily produce the paradoxical result of dampening discussion of issues of public or general concern because they happen to involve private citizens while extending constitutional encouragement to discussion of aspects of the lives of 'public figures' that are not in the area of public or general concern.

General references to the values protected by the law of libel conceal important distinctions. Traditional arguments suggest that libel law protects two separate interests of the individual: first, his desire to preserve a certain privacy around his personality from unwarranted intrusion, and second, a desire to preserve his public good name and reputation. See Rosenblatt v. Baer, 383 U.S., at 92, 86 S.Ct., at 679 (Stewart, J., concurring). The individual's interest in privacy—in preventing unwarranted intrusion upon the private aspects of his life—is not involved in this case, or even in the class of cases under consideration, since, by hypothesis, the individual is involved in matters of public or general concern.17 In the present case, however, petitioner's business reputation is involved, and thus the relevant interests protected by state libel law are petitioner's public reputation and good name.

These are important interests. Consonant with the libel laws of most of the States, however, Pennsylvania's libel law subordinates these interests of the individual in a number of circumstances. Thus, high government officials are immune from liability—absolutely privileged—even if they publish defamatory material from an improper motive, with actual malice and with knowledge of its falsity. Montgomery v. Philadelphia, 392 Pa. 178, 140 A.2d 100 (1958). This absolute privilege attaches to judges, attorneys at law in connection with a judicial proceeding, parties and witnesses to judicial proceedings, Congressmen and state legislators, and high national and state executive officials. Restatement of Torts §§ 585—592. Moreover, a conditional privilege allows newspapers to report the false defamatory material originally published under the absolute privileges listed above, if done accurately. Sciandra v. Lynett, 409 Pa. 595, 187 A.2d 586 (1963).

Even without the presence of a specific constitutional command, therefore, Pennsylvania libel law recognizes that society's interest in protecting individual reputation often yields to other important social goals. In this case, the vital needs of freedom of the press and freedom of speech persuade us that allowing private citizens to obtain damage judgments on the basis of a jury determination that a publisher probably failed to use reasonable care would not provide adequate 'breathing space' for these great freedoms. Reasonable care is an 'elusive standard' that 'would place on the press the intolerable burden of guessing how a jury might assess the reasonableness of steps taken by it to verify the accuracy of every reference to a name, picture or portrait.' Time, Inc. v. Hill, 385 U.S., at 389, 87 S.Ct., at 543. Fear of guessing wrong must inevitably cause self-censorship and thus create the danger that the legitimate utterance will be deterred. Cf. Speiser v. Randall, 357 U.S. 513, 526, 78 S.Ct. 1332, 1342, 2 L.Ed.2d 1460 (1958).

Moreover, we ordinarily decide civil litigation by the preponderance of the evidence. Indeed, the judge instructed the jury to decide the present case by that standard. In the normal civil suit where this standard is employed, 'we view it as no more serious in general for there to be an erroneous verdict in the defendant's favor than for there to be an erroneous verdict in the plaintiff's favor.' In re Winship, 397 U.S. 358, 371, 90 S.Ct. 1068, 1076, 25 L.Ed.2d 368 (1970) (Harlan, J., concurring). In libel cases, however, we view an erroneous verdict for the plaintiff as most serious. Not only does it mulct the defendant for an innocent misstatement—the three-quarter-million-dollar jury verdict in this case could rest on such an error—but the possibility of such error, even beyond the vagueness of the negligence standard itself, would create a strong impetus toward self-censorship, which the First Amendment cannot tolerate. These dangers for freedom of speech and press led us to reject the reasonable-man standard of liability as 'simply inconsistent' with our national commitment under the First Amendment when sought to be applied to the conduct of a political campaign. Monitor Patriot Co. v. Roy, 401 U.S. 265, 276, 91 S.Ct. 621, 627, 28 L.Ed.2d 35 (1971). The same considerations lead us to reject that standard here.

We are aware that the press has, on occasion, grossly abused the freedom it is given by the Constitution. All must deplore such excesses. In an ideal world, the responsibility of the press would match the freedom and public trust given it. But from the earliest days of our history, this free society, dependent as it is for its survival upon a vigorous free press, has tolerated some abuse. In 1799, James Madison made the point in quoting (and adopting) John Marshall's answer to Talleyrand's complaints about American Newspapers, American State Papers, 2 Foreign Relations 196 (U.S. Cong. 1832):

"Among those principles deemed sacred in America, among those sacred rights considered as forming the bulwark of their liberty, which the Government contemplates with awful reverence and would approach only with the most cautious circumspection, there is no one of which the importance is more deeply impressed on the public mind than the liberty of the press. That this liberty is often carried to excess; that it has sometimes degenerated into licentiousness, is seen and lamented, but the remedy has not yet been discovered. Perhaps it is an evil inseparable from the good with which it is allied; perhaps it is a shoot which cannot be stripped from the stalk without wounding vitally the plant from which it is torn. However desirable those measures might be which might correct without enslaving the press, they have never yet been devised in America." 6 Writings of James Madison, 1790—1802, p. 336 (G. Hunt ed. 1906) (emphasis in original).

This Court has recognized this imperative: '(T)o insure the ascertainment and publication of the truth about public affairs, it is essential that the First Amendment protect some erroneous publications as well as true ones.' St. Amant v. Thompson, 390 U.S. 727, 732, 88 S.Ct. 1323, 1326, 20 L.Ed.2d 262 (1968). We thus hold that a libel action, as here, by a private individual against a licensed radio station for a defamatory falsehood in a newscast relating to his involvement in an event of public or general concern may be sustained only upon clear and convincing proof that the defamatory falsehood was published with knowledge that it was false or with reckless disregard of whether it was false or not.18 Calculated falsehood, of course, falls outside 'the fruitful exercise of the right of free speech.' Garrison v. Louisiana, 379 U.S. 64, 75, 85 S.Ct. 209, 216, 13 L.Ed.2d 125 (1964).

Our Brothers HARLAN and MARSHALL reject the knowing-or-reckless-falsehood standard in favor of a test that would require, at least, that the person defamed establish that the publisher negligently failed to ascertain the truth of his story; they would also limit any recovery to 'actual' damages. For the reasons we have stated the negligence standard gives insufficient breathing space to First Amendment values. Limiting recovery to actual damages has the same defects. In the first instance, that standard, too, leaves the First Amendment insufficient elbow room within which to function. It is not simply the possibility of a judgment for damages that results in self-censorship. The very possibility of having to engage in litigation, an expensive and protracted process, is threat enough to cause discussion and debate to 'steer far wider of the unlawful zone' thereby keeping protected discussion from public cognizance. Speiser v. Randall, 357 U.S., at 526, 78 S.Ct., at 1342. Cf. Blonder-Tongue Laboratories, Inc. v. University of Illinois Foundation, 402 U.S. 313, 334—339, 91 S.Ct. 1434, 1445—1448, 28 L.Ed.2d 788 (1971). Too, a small newspaper suffers equally from a substantial damage award, whether the label of the award be 'actual' or 'punitive.'

The real thrust of Brothers HARLAN's and MARSHALL's position, however, is their assertion that their proposal will not 'constitutionalize' the factfinding process. But this clearly is not the way their test would work in practice. Their approach means only that factfinding will shift from an inquiry into whether the defamatory statements were knowingly or recklessly uttered to the inquiry whether they were negligently uttered, and if so, to an inquiry whether plaintiff suffered 'actual' damages. This latter inquiry will involve judges even more deeply in factfinding. Would the mere announcement by a state legislature that embarrassment and pain and suffering are measurable actual losses mean that such damages may be awarded in libel actions? No matter how the problem is approached, this Court would ultimately have to fashion constitutional definitions of 'negligence' and of 'actual damages.'

Aside from these particularized considerations, we have repeatedly recognized that courts may not avoid an excursion into factfinding in this area simply because it is time consuming or difficult. We stated in Pennekamp v. Florida, 328 U.S. 331, 335, 66 S.Ct. 1029, 1031, 90 L.Ed. 1295 (1946), that:

'The Constitution has imposed upon this Court final authority to determine the meaning and application of those words of that instrument which require interpretation to resolve judicial issues. With that responsibility, we are compelled to examine for ourselves the statements in issue and the circumstances under which they were made to see whether or not they * * * are of a character which the principles of the First Amendment, as adopted by the Due Process Clause of the Fourteenth Amendment, protect.' (Footnote omitted.)

Clearly, then, this Court has an 'obligation to test challenged judgments against the guarantees of the First and Fourteenth Amendments,' and in doing so 'this Court cannot avoid making an independent constitutional judgment on the facts of the case.' Jacobellis v. Ohio, 378 U.S. 184, 190, 84 S.Ct. 1676, 1679, 12 L.Ed.2d 793 (1964). The simple fact is that First Amendment questions of 'constitutional fact' compel this Court's de novo review. See Edwards v. South Carolina, 372 U.S. 229, 235, 83 S.Ct. 680, 683, 9 L.Ed.2d 697 (1963); Blackburn v. Alabama, 361 U.S. 199, 205 n. 5, 80 S.Ct. 274, 279, 4 L.Ed.2d 242 (1960).

Petitioner argues that the instructions on punitive damages either cured or rendered harmless the instructions permitting an award of general damages based on a finding of failure of WIP to exercise reasonable care. We have doubts of the merits of the premise,19 but even assuming that instructions were given satisfying the standard of knowing or reckless falsity, the evidence was insufficient to sustain an award for the petitioner under that standard. In these cases our 'duty is not limited to the elaboration of constitutional principles; we must also in proper cases review the evidence to make certain that those principles have been constitutionally applied.' New York Times Co. v. Sullivan, 376 U.S., at 285, 84 S.Ct., at 728. Our independent analysis of the record leads us to agree with the Court of Appeals that none of the proofs, considered either singly or cumulatively, satisfies the constitutional standard with the convincing clarity necessary to raise a jury question whether the defamatory falsehoods were broadcast with knowledge that they were false or with reckless disregard of whether they were false or not.

The evidence most strongly supporting petitioner is that concerning his visit to WIP's studio where a part-time newscaster hung up the telephone when petitioner disputed the newscaster's statement that the District Attorney had characterized petitioner's magazines as obscene. This contact occurred, however, after all but one of the second series of broadcasts had been aired. The incident has no probative value insofar as it bears on petitioner's case as to the first series of broadcasts. That portion of petitioner's case was based upon the omission from the first two broadcasts at 6 and 6:30 p.m. on October 4 of the word 'alleged' preceding a characterization of the magazines distributed by petitioner. But that omission was corrected with the 8 p.m. broadcast and was not repeated in the five broadcasts that followed. And we agree with the analysis of the Court of Appeals that led that court, and leads us, to conclude that the episode failed to provide evidence satisfying the New York Times standard insofar as it bore on petitioner's case based upon the broadcasts on and after October 21 concerning petitioner's lawsuit:

'Only one broadcast took place after this conversation. It is attacked on the ground that it contains an inaccurate statement concerning plaintiff's injunction action in that it stated that the district attorney considered plaintiff's publications to be smut and immoral literature. The transcript of the testimony shows that plaintiff's own attorney, when questioning defendant's representative concerning the allegedly defamatory portion of the last broadcast, said that he was not questioning its 'accuracy'. Furthermore, his examination of the same witness brought out that defendant's representative confirmed the story with the judge involved before the broadcast was made. We think that the episode described failed to provide evidence of actual malice with the requisite convincing clarity to create a jury issue under federal standards.' 415 F.2d at 897.

Petitioner argues finally that WIP's failure to communicate with him to learn his side of the case and to obtain a copy of the magazine for examination, sufficed to support a verdict under the New York Times standard. But our 'cases are clear that reckless conduct is not measured by whether a reasonably prudent man would have published, or would have investigated before publishing. There must be sufficient evidence to permit the conclusion that the defendant in fact entertained serious doubts as to the truth of his publication.' St. Amant v. Thompson, 390 U.S., at 731, 88 S.Ct., at 1325. Respondent here relied on information supplied by police officials. Following petitioner's complaint about the accuracy of the broadcasts, WIP checked its last report with the judge who presided in the case. While we may assume that the District Court correctly held to be defamatory respondent's characterizations of petitioner's business as 'the smut literature racket,' and of those engaged in it as 'girlie-book peddlers,' there is no evidence in the record to support a conclusion that respondent 'in fact entertained serious doubts as to the truth' of its reports.

Affirmed.

Mr. Justice DOUGLAS took no part in the consideration or decision of this case.

Mr. Justice BLACK, concurring in the judgment.

I concur in the judgment of the Court for the reasons stated in my concurring opinion in New York Times Co. v. Sullivan, 376 U.S. 254, 293, 84 S.Ct. 710, 733, 11 L.Ed.2d 686 (1964), in my concurring and dissenting opinion in Curtis Publishing Co. v. Butts, 388 U.S. 130, 170, 87 S.Ct. 1975, 1999, 18 L.Ed.2d 1094 (1967), and in Mr. Justice Douglas' concurring opinion in Garrison v. Louisiana, 379 U.S. 64, 80, 85 S.Ct. 209, 218—219, 13 L.Ed.2d 125 (1964). I agree of course that First Amendment protection extends to 'all discussion and communication involving matters of public or general concern, without regard to whether the persons involved are famous or anonymous.' Ante, at 44. However, in my view, the First Amendment does not permit the recovery of libel judgments against the news media even when statements are broadcast with knowledge they are false. As I stated in Curtis Publishing Co. v. Butts, supra, '(I)t is time for this Court to abandon New York Times Co. v. Sullivan and adopt the rule to the effect that the First Amendment was intended to leave the press free from the harassment of libel judgments.' Id., at 172, 87 S.Ct., at 2000.

Mr. Justice WHITE, concurring in the judgment.

* Under existing law the First Amendment is deemed to permit recoveries for damaging falsehoods published about public officials or public figures only if the defamation is knowingly or recklessly false. But until today the First Amendment has not been thought to prevent citizens who are neither public officials nor public figures from recovering damages for defamation upon proving publication of a false statement injurious to their reputation. There has been no necessity to show deliberate falsehood, recklessness, or even negligence.

The Court has now decided that the First Amendment requires further restrictions on state defamation laws. Mr. Justice BRENNAN and two other members of the Court would require proof of knowing or reckless misrepresentation of the facts whenever the publication concerns a subject of legitimate public interest, even though the target is a 'private' citizen. Only residual areas would remain in which a lower degree of proof would obtain.

Three other members of the Court also agree that private reputation has enjoyed too much protection and the media too little. But in the interest of protecting reputation, they would not roll back state laws so far. They would interpret the First Amendment as proscribing liability without fault and would equate non-negligent falsehood with faultless conduct. The burden of the damaging lie would be shifted from the media to the private citizen unless the latter could prove negligence or some higher degree of fault. They would also drastically limit the authority of the States to award compensatory and punitive damages for injury to reputation.

Mr. Justice BLACK, consistently with the views that he and Mr. Justice DOUGLAS have long held, finds no room in the First Amendment for any defamation recovery whatsoever.

Given this spectrum of proposed restrictions on state defamation laws and assuming that Mr. Justice BLACK and Mr. Justice DOUGLAS will continue in future cases to support the severest of the restrictions, it would seem that at least five members of the Court would support each of the following rules:

For public officers and public figures to recover for damage to their reputations for libelous falsehoods, they must prove either knowing or reckless disregard of the truth. All other plaintiffs must prove at least negligent falsehood, but if the publication about them was in an area of legitimate public interest, then they too must prove deliberate or reckless error. In all actions for libel or slander, actual damages must be proved, and awards of punitive damages will be strictly limited.

For myself, I cannot join any of the opinions filed in this case. Each of them decides broader constitutional issues and displaces more state libel law than is necessary for the decision in this case. As I have said, Mr. Justice BRENNAN would extend the privilege enunciated in New York Times Co. v. Sullivan, 376 U.S. 254, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964), to publications upon any 'subject of public or general interest.' See ante, at 43. He would thereby extend the constitutional protection to false and damaging, but non-malicious, publications about such matters as the health and environmental hazards of widely used manufactured products, the mental and emotional stability of executives of business establishments, and the racial and religious prejudices of many groups and individuals. All of these are, of course, subjects of real concern, and arguments for placing them within the scope of New York Times are by no means frivolous.

For Mr. Justice MARSHALL and Mr. Justice HARLAN, Mr. Justice BRENNAN's opinion is both too severe and too limited. They would make more sweeping incursions into state tort law but purportedly with less destructive weapons. They would permit suit by some plaintiffs barred under Mr. Justice BRENNAN's opinion, but would require all plaintiffs to prove at least negligence before any recovery would be allowed.

I prefer at this juncture not to proceed on such a broad front. I am quite sure that New York Times Co. v. Sullivan was the wiser course, but I am unaware that state libel laws with respect to private citizens have proved a hazard to the existence or operations of the communications industry in this country. Some members of the Court seem haunted by fears of self-censorship by the press and of damage judgments that will threaten its financial health. But technology has immeasurably increased the power of the press to do both good and evil. Vast communication combines have been built into profitable ventures. My interest is not in protecting the treasuries of communicators but in implementing the First Amendment by insuring that effective communication which is essential to the continued functioning of our free society. I am not aware that self-censorship has caused the press to tread too gingerly in reporting 'news' concerning private citizens and private affairs or that the reputation of private citizens has received inordinate protection from falsehood. I am not convinced that we must fashion a constitutional rule protecting a whole range of damaging falsehoods and so shift the burden from those who publish to those who are injured.

I say this with considerable deference since all my Brethren have contrary views. But I would not nullify a major part of state libel law until we have given the matter the most thorough consideration and can articulate some solid First Amendment grounds based on experience and our present condition. As it is, today's experiment rests almost entirely on theoretical grounds and represents a purely intellectual derivation from what are thought to be important principles of tort law as viewed in the light of the primacy of the written and spoken word.

This case lends itself to more limited adjudication. New York Times Co. v. Sullivan itself made clear that discussion of the official actions of public servants such as the police is constitutionally privileged. 'The right of free public discussion of the stewardship of public officials' is, in the language of that case, 'a fundamental principle of the American form of government.' 376 U.S., at 275, 84 S.Ct., at 723. Discussion of the conduct of public officials cannot, however, be subjected to artificial limitations designed to protect others involved in an episode with officials from unfavorable publicity. Such limitations would deprive the public of full information about the official action that took place. In the present case, for example, the public would learn nothing if publication only of the fact that the police made an arrest were permitted; it is also necessary that the grounds for the arrest and, in many circumstances, the identity of the person arrested be stated. In short, it is rarely informative for a newspaper or broadcaster to state merely that officials acted unless he also states the reasons for their action and the persons whom their action affected.

Nor can New York Times be read as permitting publications that invade the privacy or injure the reputations of officials, but forbidding those that invade the privacy or injure the reputations of private citizens against whom official action is directed. New York Times gives the broadcasting media and the press the right not only to censure and criticize officials but also to praise them and the concomitant right to censure and criticize their adversaries. To extend constitutional protection to criticism only of officials would be to authorize precisely that sort of thought control that the First Amendment forbids government to exercise.

I would accordingly hold that in defamation actions, absent actual malice as defined in New York Times Co. v. Sullivan, the First Amendment gives the press and the broadcast media a privilege to report and comment upon the official actions of public servants in full detail, with no requirement that the reputation or the privacy of an individual involved in or affected by the official action be spared from public view. Since respondent Metromedia did nothing more in the instant case, I join in holding its broadcasts privileged. I would not, however, adjudicate cases not now before the Court.

Mr. Justice HARLAN, dissenting.

The very facts of this case demonstrate that uncritical acceptance of the Pennsylvania libel law here involved would be inconsistent with those important First and Fourteenth Amendment values we first treated with in an analogous context in New York Times Co. v. Sullivan, 376 U.S. 254, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964). However, as the plurality opinion implicitly recognizes, only an undiscriminating assessment of those values would lead us to extend the New York Times rule in full force to all purely private libels. My Brother BRENNAN's opinion would resolve the dilemma by distinguishing those private libels that arise out of events found to be of 'public or general concern' from those that do not, and subjecting the former to full-scale application of the New York Times rule.

For the reasons set forth in Part I of by Brother MARSHALL's dissent, I cannot agree to such a solution. As he so well demonstrates, the principal failing of the plurality opinion is its inadequate appreciation of the limitations imposed by the legal process in accommondating the tension between state libel laws and the federal constitutional protection given to freedom of speech and press.

Once the evident need to balance the values underlying each is perceived, it might seem, purely as an abstract matter, that the most utilitarian approach would be to scrutinize carefully every jury verdict in every libel case, in order to ascertain whether the final judgment leaves fully protected whatever First Amendment values1 transcend the legitimate state interest in protecting the particular plaintiff who prevailed. This seems to be what is done in the plurality opinion. But we did not embrace this technique in New York Times, supra. Instead, as my Brother MARSHALL observes, we there announced a rule of general application, not ordinarily dependent for its implementation upon a case-by-case examination of trial court verdicts. See also my dissent in Time, Inc. v. Pape, 401 U.S. 279, 293, 91 S.Ct. 633, 641, 28 L.Ed.2d 45 (1971). Nor do I perceive any developments in the seven years since we decided New York Times, supra, that suggest our original method should now be abandoned. At least where we can discern generally applicable rules that should balance with fair precision the competing interests at stake, such rules should be preferred to the plurality's approach both in order to preserve a measure of order and predictability in the law that must govern the daily conduct of affairs and to avoid subjecting the press to judicial second-guessing of the newsworthiness of each item they print. Consequently, I fully concur in Part I of Mr. Justice MARSHALL's dissent.

Further, I largely agree with the alternative proposals of that dissent. I, too, think that, when dealing with private libel, the States should be free to define for themselves the applicable standard of care so long as they do not impose liability without fault; that a showing of actual damage should be a requisite to recovery for libel; and that it is impermissible, given the substantial constitutional values involved, to fail to confine the amount of jury verdicts in such cases within any ascertainable limits. However, my reasons for so concluding are somewhat different than his, and I therefore reach a different result than he does with respect to the tolerable limits of punitive damages.

* I think we all agree on certain core propositions. First, as a general matter, the States have a perfectly legitimate interest, exercised in a variety of ways, in redressing and preventing careless conduct, no matter who is responsible for it, that inflicts actual, measurable injury upon individual citizens. Secondly, there is no identifiable value worthy of constitutional protection in the publication of falsehoods. Third, although libel law provides that truth is a complete defense, that principle, standing alone, is insufficient to satisfy the constitutional interest in freedom of speech and press. For we have recognized that it is inevitable that there will be 'some error in the situation presented in free debate,' Time, Inc. v. Hill, 385 U.S. 374, 406, 87 S.Ct. 534, 551, 17 L.Ed.2d 456 (1967) (opinion of this writer), a process that needs 'breathing space,' NAA CP v. Button, 371 U.S. 415, 433, 83 S.Ct. 328, 338, 9 L.Ed.2d 405 (1963), to flourish, and that 'putting to the pre-existing prejudices of a jury the determination of what is 'true' may effectively institute a system of censorship.' Time, Inc. v. Hill, supra, at 406, 87 S.Ct., at 551.

Moreover, any system that punishes certain speech is likely to induce self-censorship by those who would otherwise exercise their constitutional freedom. Given the constitutionally protected interest in unfettered speech, it requires an identifiable, countervailing state interest, consistent with First Amendment values, to justify a regulatory scheme that produces such results. And, because the presence of such values dictates closer scrutiny of this aspect of state tort law than the Fourteenth Amendment would otherwise command, it may well be that certain rules, impervious to constitutional attack when applied to ordinarly human conduct, may have to be altered or abandoned where used to regulate speech. Finally, as determined in New York Times, the constitutional interest in tolerance of falsehood as well as the need to adjust competing societal interests, prohibits, at a minimum, the imposition of liability without fault.

The precise standard of care necessary to achieve these goals is, however, a matter of dispute as is the range of penalties a State may prescribe for a breach of that standard. In analyzing these problems it is necessary to begin with a general analytical framework that defines those competing interests that must be reconciled. My Brother MARSHALL's opinion, I think, dwells too lightly upon the nature of the legitimate countervailing interests promoted by the State's libel law and, as a result, overstates the case against punitive damages. Because we deal with a set of legal rules that treat truth as a complete defense it strikes, I think, somewhat wide of the mark to treat the State's interest as one of protecting reputations from 'unjustified invasion.' Post, at 78. By hypothesis, the respondent here was free to reveal any true facts about petitioner's 'obscure private life.'2

Given the defense of truth, it is my judgment that, in order to assure that it promotes purposes consistent with First Amendment values, the legitimate function of libel law must be understood as that of compensating individuals for actual, measurable harm caused by the conduct of others. This can best be demonstrated by postulating a law that subjects publishers to jury verdicts for falsehoods that have done the plaintiff no harm. In my view, such a rule can only serve a purpose antithetical to those of the First Amendment. It penalizes speech, not to redress or avoid the infliction of harm, but only to deter the press from publishing material regarding private behavior that turns out to be false simply because of its falsity. This the First Amendment will not tolerate. Where the State cannot point to any tangible danger, even knowingly erroneous publication is entitled to constitutional protection because of the interest in avoiding an inquiry into the mere truth or falsity of speech. Moreover, such a scheme would impose a burden on speaking not generally placed upon constitutionally unprotected conduct—the payment of private fines for conduct which, although not conformed to established limits of care, causes no harm in fact.

Conversely, I think that where the purpose and effect of the law are to redress actual and measurable injury to private individuals that was reasonably foreseeable as a result of the publication, there is no necessary conflict with the values of freedom of speech. Just as an automobile negligently driven can cost a person his physical and mental well-being and the fruits of his labor, so can a printing press negligently set. While the First Amendment protects the press from the imposition of special liabilities upon it, '(t)o exempt a publisher, because of the nature of his calling, from an imposition generally exacted from other members of the community, would be to extend a protection not required by the constitutional guarantee.' Curtis Publishing Co. v. Butts, 388 U.S. 130, 160, 87 S.Ct. 1975, 1994, 18 L.Ed.2d 1094 (1967) (opinion of this writer). A business 'is not immune from regulation because it is an agency of the press. The publisher of a newspaper has no special immunity from the application of general laws. He has no special privilege to invade the rights and liberties of others.' Associated Press v. NLRB, 301 U.S. 103, 132—133, 57 S.Ct. 650, 656, 81 L.Ed. 953 (1937). That the damage has been inflicted by words rather than other instrumentalities cannot insulate it from liability. States may legitimately be required to use finer regulatory tools where dealing with 'speech,' but they are not wholly disabled from exacting compensation for its measurable adverse consequences. If this is not so, it is difficult to understand why governments may, for example, proscribe 'misleading' advertising practices or specify what is 'true' in the dissemination of consumer credit advertisements.

Nor does this interest in compensating victims of harmful conduct somehow disappear when the damages inflicted are great. So long as the effect of the law of libel is simply to make publishers pay for the harm they cause, and the standard of care required is appropriately adjusted to take account of the special countervailing interests in an open exchange of ideas, the fact that this may involve the payment of substantial sums cannot plausibly be said to raise serious First Amendment problems. If a newspaper refused to pay its bills because to do so would put it out of business, would the First Amendment dictate that this be treated as a partial or complete defense? If an automobile carrying a newsman to the scene of a history-making event ran over a pedestrian, would the size of the verdict, if based upon generally applicable tort law principles, have to be assessed against the probability that it would deter broadcasters from news gathering before it could pass muster under the First Amendment? However, without foreclosing the possibility that other limiting principles may be surfaced by subsequent experience, I do think that since we are dealing, by hypothesis, with infliction of harm through the exercise of freedom of speech and the press to which the Constitution gives explicit protection, recoverable damages must be limited to those consequences of the publication which are reasonably foreseeable. The usual tort rule seems to be that once some foreseeable injury has been inflicted, the negligent defendant must compensate for all damages he proximately caused in fact, no matter how peculiar were the circumstances of the particular plaintiff involved. W. Prosser, The Law of Torts § 50 (3d ed. 1964). However, our cases establish, I think, that, unless he has knowledge to the contrary, a speaker is entitled to presume that he is addressing an audience that is not especially susceptible to distress at the specter of open, uninhibited, robust speech. Cohen v. California, 403 U.S. 15, 91 S.Ct. 1780, 29 L.Ed.2d 284. See also Brandenburg v. Ohio, 395 U.S. 444, 89 S.Ct. 1827, 23 L.Ed.2d 430 (1969); Butler v. Michigan, 352 U.S. 380, 77 S.Ct. 524, 1 L.Ed.2d 412 (1957). Thus, I think the speaker should be free from a duty to compensate for actual harm inflicted by his falsehoods where the defamation would not have caused such harm to a person of average sensibilities unless, of course, the speaker knew that his statements were made concerning an unusually sensitive person. In short, I think the First Amendment does protect generally against the possibility of self-censorship in order to avoid unwitting affronts to the frail and the queasy.

Of course, it does not follow that so long as libel law performs the same compensatory function as civil law generally it is necessarily legitimate in all its various applications. The presence of First Amendment values means that the State can be compelled to utilize finer, more discriminating instruments of regulation where necessary to give more careful protection to these countervailing interests. New York Times, supra, and Curtis Publishing Co., supra, established that where the injured party is a 'public figure' or a 'public official,' the interest in freedom of speech dictates that the States forgo their interest in compensating for actual harm, even upon a basis generally applicable to all members of society, unless the plaintiff can show that the injurious publication was false and was made 'with 'actual malice'—that is, with knowledge that it was false or with reckless disregard of whether it was false or not.' New York Times, supra, 376 U.S., at 279, 84 S.Ct., at 726. Tacitly recognizing that it would unduly sacrifice the operative legitimate state interests to extend this rule to all cases where the injured party is simply a private individual, the plurality opinion would nevertheless apply it where the publication concerned such a person's 'involvement in an event of public or general concern.' Ante, at 52. I would not overrule New York Times or Curtis Publishing Co. and I do agree, as indicated above, that making liability turn on simple falsity in the purely private libel area is not constitutionally permissible. But I would not construe the Federal Constitution to require that the States adhere to a standard other than that of reasonable care where the plaintiff is an ordinary citizen.

My principal concern with the plurality's view, of course, is that voiced by my Brother MARSHALL. However, even if this objection were not tenable, unlike the plurality, I do think there is a difference, relevant to the interests here involved, between the public and the private plaintiff, as our cases have defined these categories, and that maintaining a constitutional distinction between them is at least as likely to protect true First Amendment concerns as one that eradicates such a line and substitutes for it a distinction between matters we think are of true social significance and those we think are not.

To begin, it does no violence, in my judgment, to the value of freedom of speech and press to impose a duty of reasonable care upon those who would exercise these freedoms. I do not think it can be gainsaid that the States have a substantial interest in encouraging speakers to carefully seek the truth before they communicate, as well as in compensating persons actually harmed by false descriptions of their personal behavior. Additionally, the burden of acting reasonably in taking action that may produce adverse consequences for others is one generally placed upon all in our society. Thus, history itself belies the argument that a speaker must somehow be freed of the ordinary constraints of acting with reasonable care in order to contribute to the public good while, for example, doctors, accountants, and architects have constantly performed within such bounds.

This does not mean that I do not agree with the rule of New York Times, supra, but only that I deem it inapplicable here. That rule was not, I think, born solely of a desire to free speech that would otherwise have been stifled by overly restrictive rules, but also rested upon a determination that the countervailing state interests, described above, were not fully applicable where the subject of the falsehood was a public official or a public figure. For me, it does seem quite clear that the public person has a greater likelihood of securing access to channels of communication sufficient to rebut falsehoods concerning him than do private individuals in this country who do not toil in the public spotlight. Similarly, our willingness to assume that public personalities are more impervious to criticism, and may be held to have run the risk of publicly circulated falsehoods concerning them, does not rest solely upon an empirical assertion of fact, but also upon a belief that, in our political system, the individual speaker is entitled to act upon such an assumption if our institutions are to be held up, as they should be, to constant scrutiny. And, at least as to the 'public official,' it seems to be universally the case that he is entitled to an absolute immunity for what he may utter in response to the charges of others. Where such factors are present the need to provide monetary compensation for defamation appears a good deal more attenuated. Finally, in light of the plurality opinion's somewhat extravagant delineation of the public interest involved in the dissemination of information about nonpublic persons, it bears emphasizing that a primary rationale for extending the New York Times rule to public figures was the desire to reflect, in the constitutional balance, the fact that 'in this country, the distinctions between governmental and private sectors are blurred,' Curtis Publishing Co., supra, 388 U.S., at 163, 87 S.Ct., at 1996 (opinion of Warren, C.J.), and to treat constitutional values as specially implicated where important, albeit nonofficial, policy and behavior were the subjects of discussion. At the very least, this tends to diminish the force of any contention that libelous depictions of non-public persons are often likely to involve matters of abiding public significance.

I cannot agree that the First Amendment gives special protection to the press from '(t)he very possibility of having to engage in litigation,' ante, at 52 (opinion of BRENNAN, J.). Were this assertion tenable, I do not see why the States could ever enforce their libel laws. Cf. my Brother BLACK's opinion, ante, at 57. Further, it would certainly cast very grave doubts upon the constitutionality of so-called 'right-of-reply statutes' advocated by the plurality, ante, at 47, n. 15, and ultimately treat the application of any general law to a publisher or broadcaster as an important First Amendment issue. The notion that such an interest, in the context of a purely private libel, is a significant independent constitutional value is an unfortunate consequence of the plurality's single-minded devotion to the task of preventing self-censorship, regardless of the purposes for which such restraint is induced or the evils its exercise tends to avoid.

It is, then, my judgment that the reasonable care standard adequately serves those First Amendment values that must inform the definition of actionable libel and that those special considerations that made even this standard an insufficiently precise technique when applied to plaintiffs who are 'public officials' or 'public figures' do not obtain where the litigant is a purely private individual.

There remains the problem of punitive damages.3 No doubt my Brother MARSHALL is correct in asserting that the specter of being forced to pay out substantial punitive damage awards is likely to induce self-censorship. This would probably also be the case where the harm actually caused is likely to be great. But, as I indicated above, this fact in itself would not justify construing the First Amendment to impose an arbitrary limitation on the amount of actual damages recoverable. Thus, as my Brother MARSHALL would apparently agree—since he, too, proposes no limitation on actual damages—one cannot jump from the proposition that fear of substantial punitive damage awards may be an important factor in inducing self-censorship directly to the result that punitive damages cannot be assessed in all private libel cases. A more particularized inquiry into the nature of the competing interests involved is necessary in order to ascertain whether awarding punitive damages must inevitably, in private libel cases, serve only interests that are incompatible with the First Amendment.

At a minimum, even in the purely private libel area, I think the First Amendment should be construed to limit the imposition of punitive damages to those situations where actual malice is proved. This is the typical standard employed in assessing anyone's liability for punitive damages where the underlying aim of the law is to compensate for harm actually caused, see, e.g., 3 L. Frumer et al., Personal Injury § 2.02 (1965); H. Oleck, Damages to Persons and Property § 30 (1955), and no conceivable state interest could justify imposing a harsher standard on the exercise of those freedoms that are given explicit protection by the First Amendment.

The question then arises whether further limitations on this general state power must be imposed in order to serve the particularized goals of the First Amendment. The most compelling rationale for providing punitive damages where actual malice is shown is that such damages assure that deterrent force is added to the jury's verdict. If the speaker's conduct was quite likely to produce substantial harm, but fortuitously did not, simple assessment of actual damages will not fully reflect the social interest in deterring that conduct generally. Further, even if the harm done was great the defendant may have unusually substantial resources that make the award of actual damages a trivial inconvenience of no actual deterrent value. And even where neither of these factors obtains, the State always retains an interest in punishing more severely conduct that, although it causes the same effect, is more morally blameworthy. For example, consider the distinction between manslaughter and first-degree murder.

I find it impossible to say, at least without further judicial experience in this area, that the First Amendment interest in avoiding self-censorship will always outweigh the state interest in vindicating these policies. It seems that a legislative choice is permissible which, for example, seeks to induce, through a reasonable monetary assessment, repression of false material, published with actual malice, that was demonstrably harmful and reasonably thought capable of causing substantial harm, but, in fact, was not so fully injurious to the individual attacked. Similarly, the State surely has a legitimate interest in seeking to assure that its system of compensating victims of negligent behavior also operates upon all as an inducement to avoidance of such conduct. And, these are burdens that are placed on all members of society, thus permitting the press to escape them only if its interest is somehow different in this regard.

However, from the standpoint of the individual plaintiff such damage awards are windfalls. They are, in essence, private fines levied for purposes that may be wholly unrelated to the circumstances of the actual litigant. That fact alone is not, I think, enough to condemn them. The State may, as it often does, use the vehicle of a private lawsuit to serve broader public purposes. It is noteworthy that my Brother MARSHALL does not rest his objection to punitive damages upon these grounds. He fears, instead, the self-censorship that may flow from the unbridled discretion of juries to set the amount of such damages. I agree that where these amounts bear no relationship to the actual harm caused, they then serve essentially as springboards to jury assessment, without reference to the primary legitimating compensatory function of the system, of an infinitely wide range of penalties wholly unpredictable in amount at the time of the publication and that this must be a substantial factor in inducing self-censorship. Further, I find it difficult to fathom why it may be necessary, in order to achieve its justifiable deterrence goals, for the States to permit punitive damages that bear no discernible relationship to the actual harm caused by the publication at issue. A rational determination of the injury a publication might potentially have inflicted should typically proceed from the harm done in fact. And where the compensatory scheme seeks to achieve deterrence as a subsidiary by-product, the desired deterrence, if not precisely measured by actual damages, should be informed by that touchstone if deterrence of falsehood is not to replace compensation for harm as the paramount goal. Finally, while our legal system does often mete out harsher punishment for more culpable acts, it typically begins with a gradation of offenses defined in terms of effects. Compare, for example, larceny with murder. It is not surprising, then, that most States apparently require that punitive damages in most private civil actions bear some reasonable relation to the actual damages awarded, Oleck, at § 275, Pennsylvania included, Weider v. Hoffman, 238 F.Supp. 437, 444—447 (MD Pa.1965).

However, where the amount of punitive damages awarded bears a reasonable and purposeful relationship to the actual harm done, I cannot agree that the Constitution must be read to prohibit such an award. Indeed, as I understand it, my Brother MARSHALL's objection to my position4 is not that the interest in freedom of speech dictates eliminating such judgments, but that this result is compelled by the need to avoid involving courts in an 'ad hoc balancing' of 'the content of the speech and the surrounding circumstances,' post, at 86, 85, much like that undertaken today in Part VI of the plurality opinion, the same technique criticized in my dissent in Time, Inc. v. Pape, supra. I find this argument unpersuasive. First, I do not see why my proposed rule would necessarily require frequent judicial reweighing of the facts underlying each jury verdict. A carefully and properly instructed jury should ordinarily be able to arrive at damage awards that are self-validating. It is others, not I, who have placed upon the federal courts the general duty of reweighing jury verdicts regarding the degree of fault demonstrated in libel actions. Further, to the extent that supervision of jury verdicts would be required it would entail a different process from that undertaken where judges redetermine the degree of fault. The defendant's resources, the actual harm suffered by the plaintiff, and the publication's potential for actual harm are all susceptible of more or less objective measurement. And the overriding principle that deterrence is not to be made a substitute for compensation should serve as a useful mechanism for adjusting the equation. Finally, even if some marginal 'ad hoc balancing' becomes necessary, I should think it the duty of this Court at least to attempt to implement such a process before preempting, for itself, all state power in this regard.5

In sum, given the fact that it seems to reflect the majority rule, that most of our jurisprudence proceeds upon the premise that legislative purposes can be achieved by fitting the punishment to the crime, and since we deal here with a precise constitutional interest that may legitimately require the States to resort to more discriminating regulation within a more circumscribed area of permissible concern, I would hold unconstitutional, in a private libel case, jury authority to award punitive damages which is unconfined by the requirement that these awards bear a reasonable and purposeful relationship to the actual harm done. Conversely, where the jury authority has been exercised within such constraints, and the plaintiff has proved that the speaker acted out of express malice, given the present state of judicial experience, I think it would be an unwarranted intrusion into the legitimate legislative processes of the States and an impermissibly broad construction of the First Amendment to nullify that state action.

Because the Court of Appeals adjudicated this case upon principles wholly unlike those suggested here, I would vacate the judgment below and remand the case for further proceedings consistent with the views expressed herein.

Mr. Justice MARSHALL, with whom Mr. Justice STEWART joins, dissenting.

Here, unlike the other cases involving the New York Times1 doctrine, we are dealing with an individual who held no public office, who had not taken part in any public controversy, and who lived an obscure private life.2 George Rosenbloom, before the events and reports of the events involved here, was just one of the millions of Americans who live their lives in obscurity.

The protection of the reputation of such anonymous persons '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.' Rosenblatt v. Baer, 383 U.S. 75, 92, 86 S.Ct. 669, 679, 15 L.Ed.2d 597 (1966) (Stewart, J., concurring). But the concept of a citizenry informed by a free and unfettered press is also basic to our system of ordered liberty. Here these two essential and fundamental values conflict.

* The plurality has attempted to resolve the conflict by creating a conditional constitutional privilege for defamation published in connection with an event that is found to be of 'public or general concern.' The condition for the privilege is that the defamation must not be published 'with knowledge that it was false or with reckless disregard of whether it was false or not.' I believe that this approach offers inadequate protection for both of the basic values that are at stake.

In order for particular defamation to come within the privilege there must be a determination that the event was of legitimate public interest. That determination will have to be made by courts generally and, in the last analysis, by this Court in particular. Courts, including this one, are not annointed with any extraordinary prescience. But, assuming that under the rule announced by Mr. Justice BRENNAN for the plurality, courts are not simply to take a poll to determine whether a substantial portion of the population is interested or concerned in a subject, courts will be required to somehow pass on the legitimacy of interest in a particular event or subject; what information is relevant to self-government. See Whitney v. California, 274 U.S. 357, 375, 47 S.Ct. 641, 648, 71 L.Ed. 1095 (1927) (Brandeis, J., concurring). The danger such a doctrine portends for freedom of the press seems apparent.

The plurality's doctrine also threatens society's interest in protecting private individuals from being thrust into the public eye by the distorting light of defamation. This danger exists since all human events are arguably within the area of 'public or general concern.' My Brother BRENNAN does not try to provide guidelines or standards by which courts are to decide the scope of public concern. He does, however, indicate that areas exist that are not the proper focus of public concern, and cites Griswold v. Connecticut, 381 U.S. 479, 85 S.Ct. 1678, 14 L.Ed.2d 510 (1965). But it is apparent that in an era of a dramatic threat of over-population and one in which previously accepted standards of conduct are widely heralded as outdated, even the intimate and personal concerns with which the Court dealt in that case cannot be said to be outside the area of 'public or general concern.' The threats and inadequacies of using the plurality's conditional privilege to resolve the conflict between the two basic values involved here have been illustrated by the experience courts have had in trying to deal with the right of privacy. See Cohen, A New Niche for the Fault Principle: A Forthcoming Newsworthiness Privilege in Libel Cases?, 18 U.C.L.A.L.Rev. 371, 379—381 (1970); Kalven, Privacy in Tort Law—Were Warren and Brandeis Wrong?, 31 Law & Contemp.Prob. 326, 336 (1966). The authors of the most famous of all law review articles recommended that no protection be given to privacy interests when the publication dealt with a 'matter which is of public or general interest.' Warren & Brandeis, The Right to Privacy, 4 Harv. L.Rev. 193, 214 (1890). Yet cases dealing with this caveat raise serious questions whether it has substantially destroyed the right of privacy as Warren and Brandeis envisioned it.3 For example, the publication of a picture of the body of plaintiff's daughter immediately after her death in an automobile accident was held to be protected. Kelley v. Post Publishing Co., 327 Mass. 275, 98 N.E.2d 286 (1951). And the publication of the details of the somewhat peculiar behavior of a former child prodigy, who had a passion for obscurity, was found to involve a matter of public concern. Sidis v. F-R Pub. Corp. 113 F.2d 806 (CA2 1940).

In New York Times the Court chose to balance the competing interests by seeming to announce a generally applicable rule. Here it is apparent that the general rule announced cannot have general applicability. The plurality's conditional privilege approach, when coupled with constitutionalizing of the factfinding process,4 see Part VI of Mr. Justice BRENNAN's opinion, results in the adoption of an ad hoc balancing of the two interests involved. The Court is required to weigh the nuances of each particular circumstances on its scale of values regarding the relative importance of society's interest in protecting individuals from defamation against the importance of a free press. This scale may arguably be a more finely tuned instrument in a particular case. But whatever precision the ad hoc method supplies is achieved at a substantial cost in predictability and certainty. Moreover, such an approach will require this Court to engage in a constant and continuing supervision of defamation litigation throughout the country. See Time, Inc. v. Pape, 401 U.S. 279, 293, 91 S.Ct. 633, 641, 28 L.Ed.2d 45 (1971) (Harlan, J., dissenting); Curtis Publishing Co. v. Butts, 388 U.S. 130, 171, 87 S.Ct. 1975, 1999—2000, 18 L.Ed.2d 1094 (1967) (opinion of Black, J.).

Undoubtedly, ad hoc balancing may be appropriate in some circumstances that involve First Amendment problems. See, e.g., Bates v. City of Little Rock, 361 U.S. 516, 80 S.Ct. 412, 4 L.Ed.2d 480 (1960); Tinker v. Des Moines Independent Community School Dist., 393 U.S. 503, 89 S.Ct. 733, 21 L.Ed.2d 731 (1969). But in view of the shortcomings of my Brother BRENNAN's test, defamation of a private individual by the mass media is not one of the occasions for unfettered ad hoc balancing. A generally applicable resolution is available that promises to provide an adequate balance between the interest in protecting individuals from defamation and the equally basic interest in protecting freedom of the press.

As the plurality recognizes here and as was recognized as the basic premise of the New York Times principle, the threat that defamation law presents for the values encompassed in the concept of freedom of the press is that of self- censorship.5 Our notions of liberty require a free and vigorous press that presents what it believes to be information of interest or importance; not timorous, afraid of an error that leaves it open to liability for hundreds of thousands of dollars. The size of the potential judgment that may be rendered against the press must be the most significant factor in producing self-censorship—a judgment like the one rendered against Metromedia would be fatal to many smaller publishers.6

The judgments that may be entered in defamation cases are unlike those that may be entered in most litigation since the bulk of the award is given to punish the defendant or to compensate for presumed damages. Here the jury awarded Mr. Rosenbloom $725,000 in punitive damages.7 This huge sum was given not to compensate him for any injury but to punish Metromedia. The concept of punitive or exemplary damages was first articulated in Huckle v. Money, 2 Wils. 205, 95 Eng.Rep. 768 (K.B. 1763)—one of the general warrant cases. There Lord Camden found that the power to award such damages was inherent in the jury's exercise of uncontrolled discretion in the awarding of damages. See 1 T. Sedgwick, Damages §§ 347—350 (9th ed. 1912). Today these damages are rationalized as a way to punish the wrongdoer and to admonish others not to err. See Morris, Punitive Damages in Tort Cases, 44 Harv.L.Rev. 1172 (1931). Thus they serve the same function as criminal penalties and are in effect private fines. Unlike criminal penalties, however, punitive damages are not awarded within discernible limits but can be awarded in almost any amount. Since there is not even an attempt to offset any palpable loss and since these damages are the direct product of the ancient theory of unlimited jury discretion, the only limit placed on the jury in awarding punitive damages is that the damages not be 'excessive,' and in some jurisdictions, that they bear some relationship to the amount of compensatory damages awarded.8 See H. Oleck, Damages to Persons and Property § 275, pp. 557—560 (1955). The manner in which unlimited discretion may be exercised is plainly unpredictable. And fear of the extensive awards that may be given under the doctrine must necessarily produce the impingement on freedom of the press recognized in New York Times.

In addition to the huge awards that may be given under the label of punitive or exemplary damages, other doctrines in the law of defamation allow substantial damages without even an offer of evidence that there was actually injury. See Montgomery v. Dennison, 363 Pa. 255, 69 A.2d 520 (1949); Restatement of Torts § 621 (1938). These doctrines create a legal presumption that substantial injuries 'normally flow' from defamation. There is no requirement that there be even an offer of proof that there was in fact financial loss, physical or emotional suffering, or that the plaintiff's standing in the community was diminished. The effect is to give the jury essentially unlimited discretion and thus to give much the same power it exercises under the labels of punitive or exemplary damages. The impingement upon free speech is the same no matter what label is attached.

The unlimited discretion exercised by juries in awarding punitive and presumed damages compounds the problem of self-censorship that necessarily results from the awarding of huge judgments. This discretion allows juries to penalize heavily the unorthodox and the unpopular and exact little from others. Such free wheeling discretion presents obvious and basic threats to society's interest in freedom of the press. And the utility of the discretion in fostering society's interest in protecting individuals from defamation is at best vague and uncertain. These awards are not to compensate victims; they are only windfalls. Certainly, the large judgments that can be awarded admonish the particular defendant affected as well as other potential transgressors not to publish defamation. The degree of admonition the amount of the judgment in relation to the defamer's means—is not, however, tied to any concept of what is necessary to deter future conduct nor is there even any way to determine that the jury has considered the culpability of the conduct involved in the particular case. Thus the essence of the discretion is unpredictability and uncertainty.

The threats to society's interest in freedom of the press that are involved in punitive and presumed damages can largely be eliminated by restricting the award of damages to proved, actual injuries. The jury's wide-ranging discretion will largely be eliminated since the award will be based on essentially objective, discernible factors. And the self-censhorship that results from the uncertainty created by the discretion as well as the self-censorship resulting from the fear of large judgments themselves would be reduced. At the same time, soceity's interest in protecting individuals from defamation will still be fostered. The victims of the defamation will be compensated for their real injuries. They will not be, however, assuaged far beyond their wounds. And, there will be a substantial although imprecise and imperfect admonition to avoid future defamation by imposing the requirement that there be compensation for actual damages.

My Brother HARLAN argues that it is unnecessary to go so far. Although he recognizes the dangers involved in failing 'to confine the amount of jury verdicts * * * within any ascertainable limits,' Mr. Justice HARLAN suggests that on a finding of actual malice punitive damages may be awarded if they 'bear a reasonable and purposeful relationship to the actual harm done.' My Brother HARLAN envisions jurors being instructed9 to consider the deterrent function of punitive damages and to try to gear the punitive damages awarded in some undetermined way to actual injury. Apparently, the jury under the supervision of the court would weigh the content of the speech and the surrounding circumstances—inter alia, the position of the plaintiff, the wealth of the defendant, and the nature of the instrument of publication—on the scale of their values and determine what amount is necessary in light of the various interests involved. Since there would be no objective standard by which to measure the jury's decision there would be no predetermined limit of jury discretion and all of the threats to freedom of the press involved in such discretion would remain. The chant of some new incantation will, of course, provide clear authority for a court to substitute its values for the jury's and remake the decision. If this is what my Brother HARLAN envisions, he is merely moving the ad hoc balancing from the question of fault to the question of damages.

I believe that the appropriate resolution of the clash of societal values here is to restrict damages to actual losses. See Hill, The Bill of Rights and the Supervisory Power, 69 Col.L.Rev. 181, 191 n. 62 (1969). Of course, damages can be awarded for more than direct pecuniary loss but they must be related to some proved harm. See Wright, Defamation, Privacy, and the Public's Right to Know: A National Problem and a New Approach, 46 Tex.L.Rev. 630, 648 (1968). If awards are so limited in cases involving private individuals—persons first brought to public attention by the defamation that is the subject of the lawsuit—it will be unnecessary to rely, as both the plurality and to some extent Mr. Justice HARLAN do, on somewhat clusive concepts10 of the degree of fault, and unnecessary, for constitutional purposes, to engage in ad hoc balancing of the competing interests involved.11 States would be essentially free to continue the evolution of the common law of defamation and to articulate whatever fault standard best suits the State's need.12

The only constitutional caveat should be that absolute or strict liability, like uncontrolled damages and private fines, cannot be used.13 The effect of imposing liability without fault is to place 'the printed, written or spoken word in the same class with the use of explosives or the keeping of dangerous animals.' W. Prosser, The Law of Torts § 108, p. 792 (3d ed. 1964). Clearly, this is inconsistent with the concepts of freedom of the press.

Thus in this case I would reverse the judgment of the Court of Appeals for the Third Circuit and remand the case for a determination of whether Mr. Rosenbloom can show any actual loss.