Seven years ago the Desnick eye clinic, joined by two of the clinic's surgeons (who are no longer parties), brought this diversity suit against the ABC television network, a producer of ABC's program "PrimeTime Live," and the program's star reporter, Sam Donaldson, seeking damages for a variety of torts allegedly committed by the defendants in connection with a 15-minute program segment that was highly critical of the clinic. We affirmed the district court's dismissal, on the ground of failure to state a claim (Fed. R. Civ. P. 12(b)(6)), of all but the defamation charge. 44 F.3d 1345 (7th Cir. 1995). That charge was based on an accusation in the broadcast that the plaintiffs had tampered with a machine at the clinic called an "auto-refractor," which tests for cataract. The district judge had dismissed the charge on the ground that the accusation had not added significantly to the harm to the plaintiffs' reputation caused by the parts of the broadcast segment that the plaintiffs had not challenged. We reversed because the fact that the plaintiffs had not challenged the other accusations in the broadcast could not be construed as a concession that those other accusations were true. Id. at 1350-51. "Given the obstacles to proving defamation, the failure to mount a legal challenge to a defamatory statement cannot be considered an admission that the statement is true." Id. at 1350. On remand, the district court granted summary judgment for the defendants, without reaching the question whether the accusation of tampering was true or false, on the ground that there was insufficient evidence of "actual malice" to permit the case to go forward. The Desnick clinic has again appealed.

The clinic is conceded to be a "public figure," so that under the Supreme Court's interpretation of the free-speech clause of the First Amendment it cannot maintain a suit for defamation unless it can prove that the defendant acted with "actual malice." This is a term of legal art that means not what it seems to mean but that the defendant either knew that the defamatory statement (here, the accusation of tampering with the auto-refractor) was false or was recklessly indifferent to whether it was true or false. E.g., Masson v. New Yorker Magazine, Inc., 501 U.S. 496, 510 (1991); Milsap v. Journal/Sentinel Inc., 100 F.3d 1265, 1270 (7th Cir. 1996) (per curiam). "Reckless indifference" denotes the same state of mind that must be proved to establish liability for infringement of a federal right under color of state law or for violation of the federal mail fraud statute knowledge by the defendant that there was a high risk of harm to the plaintiff coupled with a failure to take any feasible measure to counter the risk, either by investigating further to see whether there really is a risk and how serious it is or by desisting from the risky activity. See, e.g., Farmer v. Brennan, 511 U.S. 825, 837-38 (1994); Tesch v. County of Green Lake, 157 F.3d 465, 474-75 (7th Cir. 1998); Billman v. Indiana Dept. of Corrections, 56 F.3d 785, 788-89 (7th Cir. 1995); Archie v. City of Racine, 847 F.2d 1211, 1219 (7th Cir. 1988) (en banc); United States v. Dick, 744 F.2d 546, 551 (7th Cir. 1984); Chance v. Armstrong, 143 F.3d 698, 703-04 (2d Cir. 1998); United States v. DeSantis, 134 F.3d 760, 764 (6th Cir. 1998); Keeper v. King, 130 F.3d 1309, 1314 (8th Cir. 1997).

In a defamation case by a public figure, therefore, "the plaintiff must demonstrate that the author 'in fact entertained serious doubts as to the truth of his publication,' . . . or acted with a 'high degree of awareness of . . . probable falsity,'" Masson v. New Yorker Magazine, supra, 501 U.S. at 510 (quoting St. Amant v. Thompson, 390 U.S. 727, 731 (1968), and Garrison v. Louisiana, 379 U.S. 64, 74 (1964), respectively), or, while suspecting falsity, deliberately avoided taking steps that would have confirmed the suspicion. Harte-Hanks Communications, Inc. v. Connaughton, 491 U.S. 657, 692-93 (1989) ("intent to avoid the truth," id. at 693); Eastwood v. National Enquirer, Inc., 123 F.3d 1249, 1251 (9th Cir. 1997); McFarlane v. Sheridan Square Press, Inc., 91 F.3d 1501, 1510 (D.C. Cir. 1996). (For the analog to this "ostrich" or "willful blindness" principle in cases under 42 U.S.C. sec. 1983, see West v. Waymire, 114 F.3d 646, 651 (7th Cir. 1997).) In other words, the defendant must either know that his published statement was probably false or, suspecting that it may be false, deliberately close his eyes to the possibility.

This is the criminal sense of recklessness, Farmer v. Brennan, supra, 511 U.S. at 839-40; Hill v. Shobe, 93 F.3d 418, 421 (7th Cir. 1996); Hemmings v. Gorczyk, 134 F.3d 104, 108 (2d Cir. 1998) (per curiam), or, if a little broader, is so only by a hair, West v. Waymire, supra, 114 F.3d at 650-52, whereas in tort cases the term sometimes denotes little more than gross negligence. Farmer v. Brennan, supra, 511 U.S. at 836 n. 4; Duckworth v. Franzen, 780 F.2d 645, 652 (7th Cir. 1985); In re New York City Asbestos Litigation, 678 N.E.2d 467 (N.Y. 1997) (per curiam); W. Page Keeton et al., Prosser and Keeton on the Law of Torts sec. 34, p. 213-14 (5th ed. 1984). Negligence, the standard in defamation suits brought by private rather than public figures, does not require proof of a state of mind at all, but only that the defendant failed to exercise the care that a reasonable person in his position would have exercised. The contrast with recklessness in the strong sense in which the term is used to denote the standard in constitutional, mail-fraud, and public-figure defamation cases is stark. "Reckless conduct [in a public-figure defamation case] 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." St. Amant v. Thompson, supra, 390 U.S. at 731. We may assume that the defendants were careless in having failed to investigate the auto-refractor accusation further; but there is no evidence that they actually believed the accusation to be false and so the question is whether the clinic has presented enough evidence of recklessness to defeat summary judgment.

The accusation of tampering was made originally by Paddy Kalish, an optometrist who had worked for the Desnick eye clinic for two years. Kalish claimed that technicians employed by the clinic tampered (at the clinic's direction) with the auto-refractor in order to produce false diagnoses of cataract. A symptom of cataract is that one's normal eyesight becomes severely degraded when there is a lot of glare. The auto- refractor tests for this symptom as follows. First it inspects the patient's eyes without glare. The machine automatically adjusts for whatever correction the patient requires, so that if the patient's vision has been corrected to 20/20 that is what the eye chart in the machine will report. Then the glare function is activated. If the patient does not have a cataract, his vision will still register as 20/20, but if he does have a cataract, the glare will degrade his vision, and if it degrades it to 20/50 or worse this is an indication that he needs surgery to remove the cataract. According to Kalish, the glare created by the machine can be amplified to degrade the patient's vision even if the patient does not have a cataract. In an interview with Donaldson that was videotaped (only part of which was used in the broadcast), Kalish first tested Donaldson to ascertain that without tampering his corrected vision was 20/20 even with the glare produced by the (untampered- with) machine. In other words, Donaldson did not have a cataract. Then Kalish explained that the glare could be intensified by removing the housing of a part of the machine and unscrewing the "glare bulb" exposed by that removal, covering the bulb with a piece of scotch tape, "painting" the tape with a magic marker, and reinstalling the bulb. The dimming of this bulb caused by the tampering is detected by a glare detector in the machine, and the glare detector reacts by sending more power to the glare bulb, and Kalish said that the net effect is to create increased glare in the patient's field of vision. After several failed attempts to degrade Donaldson's vision, Kalish with the aid of a friend of his, a technician formerly employed by him, was able to degrade Donaldson's vision to 20/40.

There is nothing, so far, to indicate any recklessness on the part of ABC in crediting Kalish's accusation. It is true that Kalish and the technician needed several attempts to degrade Donaldson's vision, but this is not surprising or suspicious. The machine was new to the technician, and Kalish had not done the tampering of the eye clinic's machine himself. Nor is it significant that Donaldson's vision was not degraded to the 20/50 level, given the technician's lack of tampering experience and the fact that Donaldson was younger than the patients at the clinic, almost all of whom were Medicare- eligible and thus 65 or older. The technique of tampering was odd--dimming the glare bulb in order to signal the glare detector to restore its brightness doesn't seem a likely recipe for a net increase in glare. The obvious way to achieve this end would be to mask the detector, causing it to send additional power to the undimmed glare bulb. The manufacturer of the auto-refractor acknowledged that the machine could be caused to register false positives in this way, and it is likely that Kalish, who had not done the actual tampering himself but merely observed it, mistakenly thought that the tamperer covered the bulb rather than the detector, which sits beside it in the machine.

The accusation of tampering was corroborated by the fact that ABC's investigation of the Desnick clinic turned up evidence of unneeded surgery, alteration of patients' records to show they needed cataract surgery when they didn't, diagnoses by clinic surgeons of cataract in testers (ABC "undercover agents") with normal eyesight, and statements by former employees of the clinic that almost everyone failed the glare test. The plaintiff points out correctly that these "facts" have not yet been established, because the district judge dismissed the suit before determining their truth. But that is irrelevant. All that matters is that ABC was not reckless in stating these as facts, facts establishing a pattern of herding elderly patients into unneeded cataract surgery, or in making the further charge that a diagnostic machine had been tampered with to produce false positives.

The defendants knew more than the facts we have summarized so far, however, and it is on the "more" that the plaintiff pitches its contention that a jury could infer that they knew there was a high probability that Kalish's accusation, at least, was false. But neither singly nor in combination do the additional facts that the defendants knew permit such an inference.

One thing they knew was that the U.S. Attorney had refused to join Kalish's multimillion whistleblower dollar suit against the clinic (a suit that ultimately failed, though for a reason unrelated to Kalish's credibility--that he lacked standing to bring such a suit) because he didn't think that Kalish could be the centerpiece of a credible suit. But he didn't think that because he thought Kalish was lying but because he thought Kalish might not be believed, having worked for the clinic for two years and during his employment having participated in the clinic's unethical practices. That Kalish might not be credible enough to have a good chance of persuading a jury does not mean that he was not credible enough to be a source for a news story. Many a criminal conviction has rested entirely on the testimony of coconspirators despite the requirement in a criminal case of proof beyond a reasonable doubt; a fortiori a broadcaster is entitled to repose confidence in a conspirator unless the circumstances create in the broadcaster's mind a belief that there is a high probability that the conspirator is lying.

The plaintiff in its brief repeatedly urges us to view the "outtakes" of Donaldson's interview of Kalish, that is, the parts of the videotape that were not broadcast; and we have done so. The plaintiff directs us to Donaldson's statement in the outtakes that "this is silly." But the referent is omitted. In one of the failed attempts to rig the machine, the piece of scotch tape was left wedged in it and Kalish suggested that he darken the bulb with the magic marker and then reinsert it. Evidently what Donaldson thought "silly" was attempting to rig the machine in two stages. The fact revealed by the outtakes and much harped on by the plaintiff that Kalish and the technician needed repeated attempts to tamper with the machine successfully has no significance given that the technician was not as knowledgeable about the machine as a Desnick eye clinic technician would have been. Incidentally, the fact that Kalish couldn't do the tampering himself, that he is clumsy and needed the assistance of a technician, has a significance unrecognized by the plaintiff. Given his lack of mechanical aptitude, it is unlikely that Kalish could have made up this method of tampering had he never seen or heard about it, though as we noted earlier he may have been confused as the precise method of tampering that he had observed. The clinic's lawyer speculated at the oral argument of the appeal that maybe Kalish had heard about the method of tampering from someone unrelated to Desnick's clinic. Maybe. But it was not a possibility that ABC was required to entertain seriously, given all the evidence it had that was corroborative of Kalish's accusation.

The plaintiff points out that ABC did not attempt to hunt up the actual technicians employed by the plaintiff who had tampered with the machine. This is true, and maybe that failure was negligent, though it is understandable why ABC might think it an unprofitable quest people are reluctant to admit their misconduct. But negligence is not the applicable standard.

Potentially the best fact for the plaintiff is that, as the defendants well knew (because the plaintiff's lawyer told them), the clinic had sued Kalish in state court on account of the tampering accusation, which Kalish had made on a local television station before the "PrimeTime Live" broadcast, and had won a judgment. The plaintiff argues that the judgment "necessarily encompassed a finding that Kalish had made false statements about Dr. Desnick," but that is incorrect; summary judgment was granted for the plaintiff after Kalish's lawyer failed to make a timely response to a request for admissions. Still, the fact that Kalish had lost a defamation suit based on the identical accusation should have set off warning bells at ABC; and it is conceivable (though we need not decide) that the failure to follow up was reckless, was a case of "intent not to learn the truth." There is nevertheless a fatal flaw in this part of the plaintiff's case: the plaintiff's failure to indicate what following up would have revealed. It is not enough to argue that ABC should have examined the state court record. The plaintiff must show what the record contained that would bear on Kalish's credibility. True, this principle is assumed rather than stated in the cases. See McFarlane v. Sheridan Square Press, Inc., supra, 91 F.3d at 1510; Brown v. Hearst Corp., 54 F.3d 21, 26 (1st Cir. 1995); Perk v. Readers Digest Association, Inc., 931 F.2d 408, 412 (6th Cir. 1991). But there is a compelling analogy to the duty of a party who complains about the exclusion of testimony to show by an offer of proof that the testimony would have been helpful. Fed. R. Evid. 103(a)(2); United States v. Vest, 116 F.3d 1179, 1189 (7th Cir. 1997); Israel Travel Advisor Service v. Israel Identity Tours, 61 F.3d 1250, 1260 (7th Cir. 1995); Faigin v. Kelly, 184 F.3d 67, 86 (1st Cir. 1999).

Suppose that all that a study of the record in the clinic's suit against Kalish would have revealed was a procedural bobble on the part of Kalish's lawyer. It would not be surprising if the suit had not been defended vigorously, for Kalish appears to be a person of modest means and it is entirely possible that the clinic sued him not in the hope of obtaining a collectible judgment but in the hope of silencing him and destroying his credibility. This is speculation; our point is different--it is that without any indication that ABC would have learned that Kalish's accusations were false had it studied the record of the clinic's suit against him, there is no evidence of a causal relation between ABC's alleged recklessness and the injury to the clinic. So far as appears, a study of the record would have brought to light nothing that would have cast any doubt on Kalish's truthfulness. The only aspect of the defendants' conduct that might be considered reckless was also, so far as the record discloses, harmless. Cf. Franks v. Delaware, 438 U.S. 154, 171-72 (1978).

Affirmed.