dissenting:
Ronald Watkins (“Watkins”) argues that he should be allowed to interpose a qualified privilege to block the plaintiffs from examining evidence which they claim is critical to their case. Despite the fact that (1) Watkins’ source, Leonard Shoen (“Shoen”), was a paid, nonconfidential informant who never insisted that the evidence he provided be kept secret; (2) much of what Shoen told Watkins is now a matter of public record; and (3) the plaintiffs have been frustrated in their efforts to obtain this information in any other way, the majority concludes that Watkins’ assertion of this qualified privilege should act as an absolute bar to the plaintiffs. I must respectfully dissent.
The majority adopts a three-prong test for determining whether, and under what circumstances, the assertion of a journalist’s qualified privilege must yield to a discovery request for 'nonconfidential information possessed by the journalist. Specifically, a party seeking discovery must now show that the evidence sought is (1) practically unavailable from any other source; (2) noncumulative; and (3) clearly relevant to an important issue in the litigation. See maj. op. at 415-416.
Against these three factors we must consider
the threat of administrative and judicial intrusion into the newsgathering and editorial process; the disadvantage of a journalist appearing to be an investigative arm of the judicial system or a research tool of government or of a private party; the disincentive to compile and preserve non-broadcast material; and the burden on journalists’ time and resources in responding to subpoenas.
Shoen v. Shoen (“Shoen I”), 5 F.3d. 1289, 1294-95 (9th Cir.1993) (quoting United States v. La Rouche Campaign, 841 F.2d 1176, 1182 (1st Cir.1988) (internal quotations omitted)).
Without discussing the first prong of this test as it applies to the facts of the instant appeal, the majority holds that the plaintiffs have failed to carry their burden of proof with respect to the second and- third prongs. Turning first to the third prong, the majority cites to New York Times Co. v. Sullivan, 376 U.S. 254, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964) and Greenbelt Coop. Pub. Ass’n v. Bresler, 398 U.S. 6, 90 S.Ct. 1537, 26 L.Ed.2d 6 (1970) as support for their conclusion that Shoen’s statements to Watkins “are not clearly relevant to an important issue in this litigation[,]” maj. op. at 417, because all twenty-nine of the alleged libels occurred prior to Watkins’ first interview.
Neither Sullivan nor Bresler stands for the proposition that evidence is somehow rendered irrelevant by the mere passage of time. Indeed, to hold as the majority does on this point would effectively preclude the admission of any evidence that does not coincide precisely with the time of allegedly defamatory act(s). Moreover, Shoen’s com*419ments to Watkins on the subject of the murder of his daughter-in-law, Eva Berg Shoen (“Eva”), have a direct bearing on the plaintiffs’ claim of actual malice: The whole defamation action arose out of Shoen’s accusation that the plaintiffs were involved in Eva’s death, which was the subject of the in-depth interviews between Watkins and Shoen. I have no doubt that, if we knew the contents of those interviews, we would know the extent of Shoen’s knowledge of the details of Eva’s death. If Shoen had no such knowledge, the trier of fact could conclude that he likewise had no knowledge of the details of that event when he published the allegedly defamatory statements. If, on the other hand, Shoen knew any of the details as of September 8, 1991, it would be left only to determine when he acquired that knowledge and from what source. In short, I can think of no more relevant evidence available to any party that would help the trier of fact to determine whether, at the time Shoen published the allegedly defamatory statements, he may have known they were false or acted with reckless disregard for the truth.
Turning from the third prong to the second, the majority concludes in the alternative that the evidence sought would be cumulative, anyway, because there is other evidence showing that Shoen bore considerable ill will against the plaintiffs. However, it is not enough under Sullivan to prove mere ill will on Shoen’s part; rather, the plaintiffs must prove the extent of his knowledge. Moreover, while it is true that there is evidence of Shoen’s animosity other than that possessed by Watkins, there obviously can be no better evidence, nor any evidence so clearly pertinent to the question of Shoen’s mind-set, as it bears on his allegations of the plaintiffs’ complicity in Eva’s murder. Accordingly, any evidence of Shoen’s actual knowledge of the details of Eva’s murder is noncumulative, and the requested material is practically unavailable, despite the plaintiffs’ exhaustion of all reasonable alternative sources.
-It was pointed out in Shoen I that the absence of confidentiality may be considered in the balancing of competing interests as a factor that serves to diminish both the journalist’s and the public’s interests in nondisclosure. 5 F.3d at 1295-96. Thus, any consideration of the public’s right to the free flow of information in the instant ease must be tempered by two facts: First, the trial court offered to allow Watkins to redact any material that might suggest the existence and identity of a confidential source; and, second, Shoen is to have a share in future royalties on the book and any possible movie deal.
I find this second fact to be very important, and think it is fair to say that, in a case involving a nonconfidential source who is being paid by the witness, disclosure may only tend to drive up the price, rather than shut down the “free” flow, of information. With respect to the facts of this particular- case, and aside from their agreement to share the book and movie proceeds, Watkins — whom Shoen refers to as a “novelist,” see Shoen I, 5 F.3d at 1296 — owes no professional ethical duty to Shoen.
Finally, and almost as an aside, the majority rejects the plaintiffs’ contention that Watkins’ tapes and notes would be invaluable impeachment material because, “[wjhether Leonard Shoen stated falsely during the deposition that he loved his sons is collateral; the important issue is whether he had actual malice at the time he made the allegedly defamatory statements.” Maj. op. at 418. However, the plaintiffs are not particularly interested in proving that Shoen lied at his deposition about the degree of his affection for them; rather, they seek to prove the existence of actual malice as defined by Sullivan.
In light of the above, I believe that the plaintiffs have more than carried their burden with respect to all three prongs of the test enunciated here, and have thereby made a showing sufficient to overcome Watkins’ assertion of the journalist’s qualified privilege against disclosure of the requested material. I would therefore affirm the district court’s order of contempt.