This appeal arises from an action for libel brought by John J. Ryan, former Vice-President and General Manager for North Carolina of Southern Bell Telephone Company. Ryan claimed that he had been libelled in a book written by defendant John Brooks and published by defendant Harper & Row in February 1976. Entitled Telephone: The First Hundred Years, the book was commissioned by AT&T as a corporate history to be published during the centennial year of the telephone's existence. Only one sentence of the 345-page volume mentioned Ryan; it purported to summarize Ryan's account, released to the press in January 1975, of Southern Bell's political activities during his tenure. Ryan took issue with Brooks' choice of words in that sentence; and a federal jury agreed that the sentence was false and defamatory of Ryan. They awarded him $5,000 actual and $150,000 punitive damages, against Brooks and Harper & Row jointly. Defendants raise several issues on this appeal, including a thorny question of the correct statute of limitations to be applied under controlling North Carolina law.1 But we find one issue dispositive, and reach only that: whether there was sufficient evidence that the defendants published the allegedly defamatory matter with "actual malice" under the standard of New York Times v. Sullivan, 376 U.S. 254, 84 S. Ct. 710, 11 L. Ed. 2d 686 (1964).2 Upon our own careful review of the evidence, see id. at 285, 84 S.Ct. at 728, we conclude that there was insufficient evidence of malice to sustain a judgment against defendants, while affording them the full protection required by the First Amendment. We therefore reverse the judgment and remand with instructions to enter judgment n. o. v. for defendants.

* Before reaching the dispositive questions of law, a fuller discussion of the factual background of this case is necessary. John Ryan held the North Carolina Vice-President position for Southern Bell from 1964 to 1973, when higher-ups asked him to retire. In late 1974, rumors of corruption at a high level of the Southern Bell operation began to circulate, sparked by news of the suicide of the Texas Vice-President of Southwestern Bell and his suicide note detailing charges of illegal political contributions and wiretapping by the company. Ryan testified that he was immediately contacted by newspaper reporters in Charlotte, N.C., who were investigating the possible existence of similar political activity in Southern Bell's Charlotte headquarters. At first he refused to talk with them, but soon became concerned when it appeared that Charlotte Observer reporters were getting information from other Southern Bell sources that Ryan had operated a political "slush fund" out of his Charlotte office, funded by proceeds from false expense vouchers. Fearing that he would be made to appear the scapegoat in forthcoming news stories that would not accurately describe Southern Bell's political fund and his role in operating it, Ryan decided to go on record about the fund.

In the interview with Observer reporters Marion Ellis and Howard Covington, published in a copyrighted article on January 15, 1975, and in his testimony at the trial below, Ryan described the operation of Bell's political fund as follows. Each Southern Bell department head, upon his promotion to that position, was informed that he would receive a $2,500 annual raise, but that the money was not his; it was to be paid back to the company's political fund in cash payments of $100 a month. Ryan testified that as it was explained to him when he became a department head, he had the option to decline the raise, but if he took it, he was "committed permanently" to the $100 a month payments. In addition to the inevitable pressure that a rising corporate executive would feel to accept the raise and not appear to be unwilling to do his part for the company, there was a positive incentive. The company increased the raise by the amount of extra income tax that would be owed, and the total increase resulted in a larger salary base and thus a larger pension upon retirement. No department head, to Ryan's knowledge, had ever refused this "raise."

As director of Bell's North Carolina operation, Ryan collected the salary kickbacks from the eight or nine department heads under him, and gave it out in cash to state political candidates of both parties, in hopes that Bell would have the ear of whoever was elected. In 1972, the peak year for this activity, Ryan said that he contributed a total of $28,000 to candidates for governor, senator and congressman in North Carolina. Ryan's story made headlines in the Observer on January 15, 1975, and related articles received extensive coverage over a wide area for some months thereafter.

At about this time, John Brooks, an experienced writer about business matters, was working under contract to AT&T and Harper & Row on the book Telephone. He learned of Ryan's allegations through several secondary sources, and felt they should be included in a section discussing recent scandals within the corporation. Brooks summarized the Ryan story in the following sentence found on page 310 of the published volume:

But in January 1975, John J. Ryan, a former vice-president for Southern Bell Telephone Company in charge of service for North Carolina, who had been forced to retire in June 1973 on grounds of "unsatisfactory performance", said in a series of local newspaper interviews that over a period of years he had made political contributions to candidates who were expected to be favorable to the interests of Southern Bell, deriving the funds from salary kickbacks extorted from leading Southern Bell executives that had been concealed by the use of false vouchers.

The book Telephone was published in February 1976. By May 1977, its one hardcover and two paperback editions had sold approximately 195,000 copies, of which AT&T had purchased 150,000, and the company's stockholders 30,000 through a coupon purchasing plan.

Ryan first learned about the statement when he read a book review of Telephone by Marion Ellis in the Observer. Ellis pointed out that the book contained "one major error," in that Ryan "has never acknowledged the existence of a bogus voucher scheme." After reading the book, Ryan engaged counsel and demanded that Harper & Row retract the statement. He particularly objected to the use of the word "extorted," and to the mention of a false voucher plan for concealment of the kickbacks. Harper & Row and Brooks did eventually agree to change the statement about Ryan,3 but the new version could not be inserted until the second paperback edition was issued in August 1976. The publishers refused Ryan's demand to delete all mention of him from the book, however, and he filed the complaint that began this suit on May 2, 1977. He named AT&T, Harper & Row, and Brooks as defendants, but AT&T settled before trial and the case proceeded against the publisher and author.

At the trial, John Brooks testified that his contract with AT&T gave him complete control over the content of the corporate history, and that he was given almost complete access to AT&T files. Brooks said that he had never heard of Ryan or of any misconduct in Southern Bell until he read an article in the New York Times on February 9, 1975 about federal investigations of AT&T subsidiaries' "secret political contributions." The article contained an account of Ryan's revelations to the Observer, stating that Ryan "said that he had run a political fund made up of salary kickbacks from an average of six to eight of his immediate subordinates in the North Carolina operation." Though that article introduced him to the problems at Southern Bell, Brooks testified that his most important source for this paragraph of his book was a June 23, 1975 article in Business Week magazine. That story discussed the Southwestern Bell scandals, and then noted that Southern Bell had

tried quietly to oust the head of its telephone operations in North Carolina, John J. Ryan. But Ryan suddenly rebelled and rattled off a detailed story to the Charlotte Observer describing a companywide system of forced political contributions and false vouchering that he and others had engaged in, allegedly with the knowledge and approval of higher-ups.

Brooks' third source of information about the Ryan affair was an issue of the "AT&T Management Report," an internal publication of AT&T, dated January 23, 1975, which Brooks had pulled from the corporate files in February or March of that year. It contained a statement of Mr. L. E. Rast, President of Southern Bell, in response to Ryan's allegations. Rast stated that Ryan had been "forced to retire" in June 1973, that the existence of a bogus voucher plan, whose proceeds had been diverted to political contributions, had been discovered then, and that it had been promptly stopped.

Brooks testified that though none of these sources contained the exact words that he used in his book, he felt that his sentence was a "fair synthesis" of the information he had gathered. He believed that force was inherent in any salary kickback operation, and the Business Week article referred to "forced political contributions." Thus Brooks chose the word "extorted," which, as he confirmed by checking his dictionary, means "to obtain by force." Business Week and Rast's statement both mentioned a false vouchering plan as described by Ryan, and the Times article spoke of "secret" political contributions. Hence Brooks' description of the kickbacks as "concealed by the use of false vouchers." The author acknowledged that he could have checked the accuracy of these secondary source materials by obtaining and reading the Charlotte Observer interview with Ryan to which they referred. But he did not think it necessary because he had found the sources he used to be reliable in the past; moreover, he would not ordinarily spend the time to track down a primary source for a passage that was such a small part of the total work. Ryan's allegations, Brooks testified, were incidental to his history of the telephone company, but he felt they should be included in the interests of objectivity, as a counterpoint to AT&T executives' statements that the Southwestern Bell was an isolated case.

In New York Times v. Sullivan, 376 U.S. 254, 84 S. Ct. 710, 11 L. Ed. 2d 686 (1964), the Supreme Court first set out the standard of proof required in a libel action brought by a public official, in order to safeguard the freedom of speech and press provided by the First and Fourteenth Amendments. In that case an elected official in Montgomery, Alabama brought suit against the New York Times, alleging that he had been libelled in an advertisement published in the newspaper. The advertisement was a plea for aid to the civil rights movement in the South, and included statements, some of which were false, about police action against student demonstrators and against Martin Luther King, Jr. The Court reversed an award of damages to the plaintiff after a state court trial at which the court instructed the jury that the statements were "libelous per se," and that they could award damages if the statements were published by defendants and related to the plaintiff. Justice Brennan wrote for the Court that the defense of truth was not adequate protection for would-be critics of official conduct:

The constitutional guarantees require, we think, a federal rule that prohibits a public official from recovering damages for a defamatory falsehood relating to his official conduct unless he proves that the statement was made with "actual malice"-that is, with knowledge that it was false or with reckless disregard of whether it was false or not.

Id. at 279-80, 84 S.Ct. at 726.

The Court then reviewed the evidence under the newly-announced standard and concluded that the proof of actual malice "lacks the convincing clarity which the constitutional standard demands," and could not sustain the judgment. Id. at 285-86, 84 S.Ct. at 728-29. Specifically, the Court found that the failure of Times employees to check the accuracy of its advertisement against news stories in the newspaper's own files could at most support a finding of negligence, but was "constitutionally insufficient to show the recklessness that is required for a finding of actual malice." Id. at 288, 84 S.Ct. at 730.

Subsequently, in Curtis Publishing Co. v. Butts and Associated Press v. Walker, 388 U.S. 130, 87 S. Ct. 1975, 18 L. Ed. 2d 1094 (1967), the Court extended the New York Times constitutional safeguards to defendants in libel actions by individuals who, though not "public officials," are "public figures" and are "involved in issues in which the public has a justified and important interest." Id. at 134, 87 S.Ct. at 1980. Justice Harlan announced the result in a plurality opinion in which he formulated a slightly different standard of "constitutional malice"4 from the New York Times version. His rule would have allowed a public figure who is not a public official to

recover damages for a defamatory falsehood whose substance makes substantial danger to reputation apparent, on a showing of highly unreasonable conduct constituting an extreme departure from the standards of investigation and reporting ordinarily adhered to by responsible publishers.

Id. at 155, 87 S.Ct. at 1991. This rule did not receive the endorsement of a majority of the Court, however; four Justices joined Chief Justice Warren's opinion which concurred in the result reached by Justice Harlan, but on the basis of an extension of the same New York Times rule to libel actions by public figures as that applied to public officials. Later pronouncements by the Court in this area confirm that the New York Times standard has remained as the applicable principle in both types of cases.5

In the Butts case a majority of the Court agreed that the verdict against the defendant should be upheld under either the Harlan or the New York Times standard. Wally Butts, then athletic director of the University of Georgia, had sued the Saturday Evening Post over an article it published that accused Butts of conspiring to "fix" a football game between the University of Georgia and the University of Alabama. The Post story was based only on information from one George Burnett, who had accidentally overheard a telephone conversation between Butts and the Alabama coach, during which he claimed Butts divulged the secrets of Georgia's game strategy. The Post published the story, which they knew to be highly damaging to Butts, without independent verification although they knew that Burnett had been convicted on bad check charges. No one checked Burnett's notes of the conversation, talked with the person who supposedly was with him at the time, or screened the films of the game to see if his information was accurate. The writer assigned to the story was not a football expert and no one knowledgeable in the sport was consulted. Id. at 157-58, 87 S.Ct. at 1992-93. Testimony indicated that the Post wanted to change its image and increase revenues by instituting a new policy of "sophisticated muckraking." Thus they were eager to publish this expose, and did so with little initial investigation, which they failed to supplement even after they were notified by Butts and his daughter that the story was absolutely untrue. Id. at 169-70, 87 S.Ct. at 1998-99 (Warren, C.J., concurring). This evidence, which Justice Harlan found ample to support an award of damages under his standard, id. at 158, 87 S.Ct. at 1993, also satisfied Chief Justice Warren that the Post had exhibited the degree of reckless disregard for the truth contemplated in New York Times. Id. at 170, 87 S.Ct. at 1999.6

In Butts as well as in other decisions applying the New York Times standard, the Court confirmed that the "constitutional malice" requirement announced there contemplated much more than a failure to exercise ordinary care in verifying statements about public officials or public figures.7 Thus, in Garrison v. Louisiana, 379 U.S. 64, 74, 85 S. Ct. 209, 216, 13 L. Ed. 2d 125 (1964), the Court stated that "only those false statements made with the high degree of awareness of their probable falsity demanded by New York Times may be the subject of either civil or criminal sanctions." See also Rosenblatt v. Baer, 383 U.S. 75, 84, 86 S. Ct. 669, 675, 15 L. Ed. 2d 597 (1966). In St. Amant v. Thompson, 390 U.S. 727, 88 S. Ct. 1323, 20 L. Ed. 2d 262 (1968), the Court stressed the stringent evidentiary standard necessary to prove reckless conduct. In that case Thompson, a deputy sheriff, sued St. Amant, a candidate for public office, for defamation of the sheriff in a televised speech, in which St. Amant had quoted another person's statement that Thompson had taken bribes. The Louisiana Supreme Court had upheld the jury verdict for the plaintiff, finding sufficient evidence of reckless conduct under the New York Times standard. The record revealed that St. Amant had no personal knowledge of Thompson's activities, but relied solely on a source whose reputation he did not know; and he had failed to verify the information with those who might have known the facts. Nevertheless, the United States Supreme Court reversed, holding that these facts failed to prove St. Amant's reckless disregard for the accuracy of his statements. After reviewing the Court's previous decisions in the area of libel, Justice White wrote for the Court:

These 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. Publishing with such doubts shows reckless disregard for truth or falsity and demonstrates actual malice.

390 U.S. at 731, 88 S. Ct. at 1325. Justice White then noted some examples of reckless conduct that would indicate bad faith publication and negate the New York Times privilege. They included publication of a completely fabricated story, or of one based entirely on an unverified anonymous telephone call; or publication where there are obvious reasons to doubt the veracity of the informant. Id. at 732, 88 S.Ct. at 1326. St. Amant's mere failure to investigate, however, did not prove bad faith since there was no evidence that he was aware of the probable falsity of the statements about Thompson. Id. at 732-33, 88 S.Ct. at 1326.

" 'Reckless disregard,' " Justice White noted in St. Amant, "cannot be fully encompassed in one infallible definition. Inevitably its outer limits will be marked out through case-by-case adjudication ...." 390 U.S. at 730, 88 S.Ct. at 1325. Viewed in light of the foregoing precedent, we think the evidence in this case was insufficient to bring John Brooks' actions within those outer limits of reckless conduct marked out in Supreme Court cases. Assuming that the use of the words "extortion" and "false vouchers" rendered the sentence false and defamatory, there is clearly no evidence that Brooks knew they were false. The only question is whether he actually doubted their accuracy but left them unchanged, without further investigation. There is nothing in the record to indicate that Brooks had any such doubts. He relied on two secondary sources which he had used in the past and which have an excellent reputation. He had no reason to doubt the accuracy of their accounts of Ryan's Observer interview. The reliability of the third source, the internal Management Report of AT&T, is more questionable, but Brooks used nothing from it that was not also found in his other sources. It simply served to corroborate the existence of the false vouchering system reported in Business Week. Even if the three sources together should have tipped Brooks to the existence of a dispute between Ryan and Southern Bell executives, as Ryan argues they must have, he would still have no reason to suspect that the Times and Business Week had not reported Ryan's statements accurately.

Clearly it would have been better journalistic practice to have verified the accuracy of these secondary sources by reading the original account in the Charlotte Observer. But we cannot say that the failure to do so amounted to more than mere negligence. We recognize that the book was not "hot news," and a more thorough investigation should be expected in these circumstances than in the preparation of a news story under deadline pressure. Nevertheless, the sentence was such a small part of the whole work that the author might understandably feel three sources to be sufficient. Certainly where there was no reason to doubt the accuracy of the sources used, the failure to investigate further, even if time was available, cannot amount to reckless conduct. See Vandenburg v. Newsweek, Inc., 507 F.2d 1024, 1026-28 (5th Cir. 1975).

Nor can the fact that Brooks changed the words of his sources create a jury issue on the question of malice. The historian's job is not to copy statements exactly as written in a secondary source, but to interpret and rework them into the whole. Though "extorted" was an unfortunate choice of words because of its criminal connotations, it does also mean simply "obtained by force." Since Ryan's testimony indicated that the contributions to the fund were not entirely voluntary, the word was not really off the mark. In Time, Inc. v. Pape, 401 U.S. 279, 91 S. Ct. 633, 28 L. Ed. 2d 45 (1971), the Court considered a defamation claim arising from a magazine writer's omission of the word "alleged" when citing a report of a certain incident of police brutality. The Court reasoned that omission of the word was perhaps due to a misconception, but was nevertheless an interpretation drawn from the report as a whole; to permit the malice issue to go to the jury because of it "would be to impose a much stricter standard of liability on errors of interpretation or judgment than on errors of historic fact." Id. at 290, 91 S.Ct. at 639. We think this reasoning applies here, and would not find proof of malice in Brooks' use of slightly stronger language than his source's. See also Orr v. Argus-Press Co., 586 F.2d 1108 (6th Cir. 1978).

We recognize that the New York Times standard is a difficult one for libel plaintiffs to meet, and that its application may sometimes produce harsh results. But in the effort to balance private rights to protection of reputation against the First Amendment rights of writers and publishers to print information on matters of interest to the public, the courts have, as they must, favored the latter. A standard that allowed liability to attach for mere negligence "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. 374, 389, 87 S. Ct. 534, 542, 17 L. Ed. 2d 456 (1967). There is no social value, on the other hand, in the knowing publication of false information, and no danger that attaching liability for such publications will cause the press to steer so far from the danger zone that free expression will be stifled. Adjudications in this area, therefore, have carefully avoided upholding damage awards unless the defendant's conduct was so reckless in publishing material of obviously doubtful veracity that it approaches the level of publishing a knowing, calculated falsehood. Aside from the Butts case, there is only one instance in which the Supreme Court has upheld a finding of liability under the New York Times standard.8 And Butts, we think, is distinguishable from this case on its facts. There the Post's story was based almost entirely on a source previously unknown to the defendants, whom they knew to have a criminal record. Obvious avenues for checking the facts and allegations in the story were ignored, even though the editors had actual warning that much of it was false. The critical difference in this case is that Brooks relied on two reputable sources, and there was nothing to indicate to him that their information, or his synthesis of it, was not wholly accurate. Moreover, his reason for including the sentence about Ryan was his desire to make the book a balanced, objective history of AT&T. Certainly the Southern Bell controversy was worthy of mention in a history of the telephone company, and a rule of liability that could cause authors to avoid such controversial topics for fear of damage judgments would ultimately reduce the flow of information to the public.

Our reading of this case is fully in line with the decisions in other circuits that have applied the New York Times standard. As long as the sources of the libelous information appeared reliable, and the defendant had no doubts about its accuracy, the courts have held the evidence of malice insufficient to support a jury verdict, even if a more thorough investigation might have prevented the admitted error. See Dickey v. CBS Inc., 583 F.2d 1221, 1227 (3d Cir. 1978); Hotchner v. Castillo-Puche, 551 F.2d 910, 912-14 (2d Cir. 1977); Grzelak v. Calumet Publishing Co., 543 F.2d 579, 583 (7th Cir. 1975); Vandenburg v. Newsweek, 507 F.2d at 1026-28; Drotzmann's, Inc. v. McGraw-Hill, Inc., 500 F.2d 830, 834 (8th Cir. 1974). In two cases in which the evidence of malice was found to be sufficient, by contrast, the facts indicated strongly that the challenged allegations had been completely fabricated by the writer. See Carson v. Allied News Co., 529 F.2d 206, 213 (7th Cir. 1976); Davis v. Schuchat, 510 F.2d 731, 735-36 (D.C. Cir. 1975).

The trial court below submitted this case to the jury with the instruction that they must find the element of actual malice on the part of defendants in order to find them liable. The court's definition of malice, however, was too broad, in that it included Justice Harlan's standard from Butts, which has since been rejected by a majority of the Court, and which may have confused the jury by indicating that a finding of negligence would suffice for malice.9 Defendants, however, did not object to the inclusion of this definition. They did request the court, after it had finished the instructions but before the jury retired, to reopen the instructions and include the language of the St. Amant case, which the court refused. Though we do not say it was error for the court to refuse to include this belated offering, we disagree with its conclusion that the St. Amant language would be inappropriate to the circumstances of this case. The St. Amant language has been cited frequently in subsequent decisions of the Court, including very recent ones,10 and it has never been limited to suits brought by public officials.

We need not decide, however, whether the instructions were so prejudicial as to constitute "plain error" and a basis for reversal. The defendants have properly challenged the sufficiency of the evidence, and on that basis we have followed the Supreme Court's lead in First Amendment cases and have conducted an independent examination of the record as a whole to determine whether the judgment constitutes "a forbidden intrusion on the field of free expression." Beckley Newspapers Corp. v. Hanks, 389 U.S. 81, 82, 88 S. Ct. 197, 198-199, 19 L. Ed. 2d 248 (1967). We find the evidence of actual malice insufficient to present a jury question in light of the constitutional standard, and we therefore reverse.

REVERSED.

WIDENER, Circuit Judge, concurs in the result.