In Re: Subpoena

UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA
_________________________________________
)
)
Paul B. Goldberg, )
)
Movant, )
)
v. ) Civil No. 1:15-mc-00825 (APM)
)
Amgen, Inc. )
)
Respondent. )
_________________________________________ )
MEMORANDUM OPINION AND ORDER
This case presents an issue that appears with increasing frequency in the federal courts:
Under what circumstances can a journalist be compelled to testify in a civil case about his or her
First Amendment-protected activities? The present dispute arises out of a shareholder class-action
lawsuit pending in federal court in the Central District of California, In Re Amgen Inc. Securities
Litigation, No. 07-2536 (C.D. Cal.). In connection with that case, Respondent Amgen, Inc., issued
a subpoena to Movant Paul B. Goldberg, a journalist residing in Washington, D.C. Amgen sought
to depose Goldberg regarding an article that he wrote in 2007 about a clinical trial of one of
Amgen’s FDA-approved drugs, Aranesp. The clinical trial was terminated early because of safety
and efficacy concerns. At issue in the California litigation is whether Amgen and four of its former
officers misled investors when they publicly stated that Aranesp was safe for its FDA-approved
uses. The plaintiffs allege that they suffered losses when Goldberg’s article revealed the truth
about Aranesp’s safety and efficacy. Goldberg moved to quash Amgen’s subpoena, claiming that
the information it seeks is protected by the First Amendment reporter’s privilege.
Having considered the parties’ respective arguments and the record evidence, the court
grants Goldberg’s motion to quash. Goldberg’s request for attorney’s fees is denied.
I. BACKGROUND
A. The Cancer Letter Article
The Movant in this matter is Paul Goldberg, a journalist and editor-in-chief/publisher of
The Cancer Letter. Decl. of Paul B. Goldberg [hereinafter “Goldberg Decl.”], ECF No. 1-2, ¶ 1.
The Cancer Letter is a weekly newsletter-style publication that covers events concerning the
development of cancer therapies, cancer research funding, and health care finance, legislation, and
policy. Id. ¶ 2. The Cancer Letter is based in Washington, D.C. Id. ¶ 1.
On February 16, 2007, Goldberg authored and published an article titled “Danish
Researchers Post Long-Awaited Aranesp Results–Ever So Discreetly” [hereinafter “Article”].
Goldberg Decl., Ex. A, ECF No. 1-3. The Article described a study conducted by the Danish Head
and Neck Cancer Group, which the parties have referred to as the “DAHANCA 10” study, about
the efficacy of adding Aranesp to radiation treatment of patients with head and neck cancer. Id.
The study, according to the Article, “showed a significantly inferior therapeutic outcome from
adding Aranesp to radiation treatment of patients with head and neck cancer.” Id. at 1. The study
was suspended in October 2006 because of “potential unexpected negative effects,” id. at 4, and
ultimately was not resumed, id. at 2.
The Article also reported that few were aware of the DAHANCA 10 study: “[E]ven
informed observers have been largely unaware that the Danish study was temporarily stopped on
October 18, 2006, and that the decision not to resume the study was made on Dec. 1, 2006, and
posted on the Web by the principal investigator, Jens Overgaard.” Id. at 1-2. The Article went on
2
to note that Amgen had not announced the study’s results in public disclosures or during its January
25, 2007, conference call discussing its annual financial results. Id. at 2.
The Article also stated that “[s]everal Wall Street sources who monitor Amgen confirmed
that they have been awaiting these results and were not aware of them until hearing about the
closing of the [clinical] trial from this reporter.” Id. The Article did not identify the Wall Street
sources. In a declaration submitted in this matter, Goldberg clarified that he had spoken to two
Wall Street sources before publishing the Article. Goldberg attested that one of them had talked
to him on the condition that the source’s comments would remain confidential. Supplemental
Decl. of Paul B. Goldberg [hereinafter “Goldberg Supp. Decl.”], ECF No. 8-1, ¶¶ 4-5. As for the
other, Goldberg could not recall the identity of the source and thus could not recall whether he had
promised the source confidentiality. Id. ¶ 6.
The Article did not only refer to Goldberg’s Wall Street sources, but identified several
other sources by name. They included: (1) Dr. Michael Henke, a German oncologist, who was
quoted as saying he had found the DAHANCA 10 study “by [using] Google,” Goldberg Decl., Ex.
A, at 2; (2) Dr. David Steensma, an associate professor of medicine and oncology at the Mayo
Clinic, who commented on the validity of the DAHANCA 10 study, id. at 3; (3) Dr. Charles
Bennett, an oncologist at Northwestern University, who said that he had learned about the
DAHANCA 10 study results recently from a European colleague, id.; (4) Dr. Scott Lippman, the
then-chairman of thoracic/head and neck medical oncology at M.D. Anderson Cancer Center, who
stated that the Danish study was consistent with prior studies of similar drugs, id.; and (5)
Dr. Howard Ozer, chief of hematology and oncology and professor of medicine at the University
of Oklahoma Cancer Center, who, like Dr. Henke, opined on the validity of the study, id. at 3-4.
3
The Article noted that Goldberg had unsuccessfully tried to reach the DAHANCA 10 study’s
principal investigator, Jens Overgaard, for comments. Id. at 3.
B. Amgen’s Discovery Efforts
According to the plaintiffs in the Amgen Securities case, the Article was the first time that
Amgen’s investors were informed of the DAHANCA 10 study’s results. Amgen Mem., at 1. The
Article was thus a “corrective disclosure,” which revealed the truth about Aranesp’s safety. Id. at
2. Amgen contends that “[o]ne of the ways [it] can defend [itself] is to show that investors or other
stock market participants learned of the DAHANCA 10 termination before the Article was
published on February 16, 2007,” id., and therefore the Article’s publication did not cause their
losses. Amgen argues that Goldberg, as author of the Article, is “uniquely positioned to provide
admissible testimony critical to Amgen’s defense concerning the disclosure of the DAHANCA 10
termination to market participants.” Id.
Before serving Goldberg with a subpoena, Amgen made some effort to discover the
evidence it sought through alternative sources. Specifically, it served document subpoenas on the
four doctors identified in the article who are located in the United States: Dr. Bennett, Dr. Ozer,
Dr. Steensma, and Dr. Lippman. Decl. of Douglas J. Dixon [hereinafter “Dixon Decl.”], ECF No.
2-1, ¶¶ 5-9. Amgen also served a document subpoena on MD Anderson Cancer Center,
Dr. Lippman’s former employer. Id. ¶ 10. Other than obtaining a copy of the Article, none of
these subpoena efforts bore any fruit. Id. ¶¶ 11-14. Moreover, Dr. Steensma informed Amgen
through his counsel that he could not recall any communications with Goldberg before the Article
was published. Id. ¶ 11. And Dr. Ozer told Amgen that he could not recall specifics of the
DAHANCA 10 trial or his quote in the Article. Id. ¶ 12.
4
Before it subpoenaed Goldberg, Amgen did not, however, attempt to depose any of the
U.S.-based physicians mentioned above. Nor did it attempt to obtain any evidence from
Dr. Henke, who resides in Germany. Amgen Mem. at 15 n.8. Amgen also did not try to obtain
evidence from the DAHANCA 10 study’s lead investigator, Jens Overgaard. Mot. Hr’g Tr. 21-
23, Aug. 11, 2015, ECF No. 10.
Furthermore, according to an affidavit submitted by Amgen’s counsel, before Amgen
subpoenaed Goldberg, it did not depose any of the Wall Street analysts who followed Amgen in
2007. See Dixon Decl. ¶ 23. As of July 9, 2015, the date Amgen filed its opposition brief, Amgen
had only noticed depositions of five investment managers and analyst depositions. Id. However,
at oral argument held on August 11, 2015, Amgen’s counsel represented—for the first time—that
Amgen had deposed three analysts two years earlier in October 2013. Mot. Hr’g Tr. 13, 35.
Amgen’s counsel also stated that, since serving Goldberg, Amgen had deposed four investment
managers of the lead plaintiff, Connecticut Retirement Plans and Trust Funds, with another
deposition pending. Id. at 36. Though not entirely clear, these may be the same five investment
managers noticed for depositions as of July 9, 2015.
C. Procedural Background
On May 12, 2015, Amgen served on Goldberg the subpoena at issue in this case. The
parties met and conferred about the subpoena, but were unable to resolve their differences.
Thereafter, on June 25, 2015, Goldberg filed a motion seeking to quash Amgen’s subpoena. In its
opposition brief, Amgen made clear that it does not seek confidential information from Goldberg,
such as the names of his sources, but only non-confidential information about his writing of the
Article. Goldberg nevertheless contends that the non-confidential information sought is protected
by the reporter’s privilege.
5
II. LEGAL STANDARD
As the party asserting the reporter’s privilege, Goldberg bears the burden of establishing
the privilege’s applicability. See Estate of Klieman v. Palestinian Auth., 293 F.R.D. 235, 241
(D.D.C. 2013). Once the privilege’s applicability is established, the burden shifts to Amgen “to
show that on ‘the specific facts of the case,’ its interest outweighs the public interest in protecting
the journalist’s sources and information.’” Id. (citation omitted). In other words, if the privilege
applies, Amgen bears the burden of abrogating it. See Hutira v. Islamic Republic of Iran, 211 F.
Supp. 2d 115, 118 (D.D.C. 2002).
III. DISCUSSION
A. Goldberg’s Invocation of the Reporter’s Privilege
The Court of Appeals has long recognized that, in civil cases, journalists enjoy a qualified
privilege against compelled disclosure of their First Amendment activities. See Carey v. Hume,
492 F.2d 631, 636 (D.C. Cir. 1974); Zerilli v. Smith, 656 F.2d 705, 712 (D.C. Cir. 1981). In Zerilli,
the Court of Appeals explained that:
When striking the balance between the civil litigant’s interest in compelled
disclosure and the public interest in protecting a newspaper’s confidential sources,
we will be mindful of the preferred position of the First Amendment and the
importance of a vigorous press. Efforts will be taken to minimize impingement
upon the reporter’s ability to gather news.
656 F.2d at 712 (citation omitted). As a result, “in the ordinary case the civil litigant’s interest in
disclosure should yield to the journalist’s privilege. Indeed, if the privilege does not prevail in all
but the most exceptional cases, its value will be substantially diminished.” Id. (footnote omitted).
Nevertheless, because the journalist’s privilege is not absolute but qualified, trial courts are
directed to conduct a factual inquiry to decide whether the subpoena’s proponent can overcome
6
the privilege. See id. (stating that the “courts should look to the facts of each case” in deciding
whether the journalist’s privilege applies).
Courts must consider two main factors in deciding whether to compel a journalist to
provide evidence. The first is the “civil litigant’s need for the information.” Id. at 713. “If the
information sought goes to ‘the heart of the matter,’ that is, if it is crucial to his case, then the
argument in favor of compelled disclosure may be relatively strong.” Id. (citations omitted). Yet,
“[o]n the other hand, if the information sought is only marginally relevant, disclosure may be very
difficult to justify.” Id. (footnote omitted) (citation omitted). Thus, as to this factor, the “the court
must examine how important, not just relevant, the reporter’s information is to the party’s case.”
In re Slack, 768 F. Supp. 2d 189, 194 (D.D.C. 2011).
The second factor for the court to consider is the civil litigant’s efforts “to obtain the
information from alternative sources.” Zerilli, 656 F.2d at 713. This factor requires the
subpoenaing party to exercise diligence in obtaining evidence from reasonably available sources
before seeking disclosures from the journalist. “Even when the information is crucial to a litigant’s
case, reporters should be compelled to disclose their sources only after the litigant has shown that
he has exhausted every reasonable alternative source of information.” Id.
1. Applying the Journalist’s Privilege to Non-Confidential Information
Before examining the two factors, the court first addresses a threshold issue raised by
Amgen—its contention that the journalist’s privilege does not apply to the non-confidential
information it seeks from Goldberg. According to Amgen, it does not intend to ask Goldberg to
reveal any confidential sources or information. Rather, because it only wants to elicit non-
confidential evidence—for example, how and when Goldberg learned of the DAHANCA 10
termination; when he spoke to the identified and anonymous sources cited in the Article; and to
7
confirm quotes attributed to identified sources—Amgen contends that Goldberg has failed to
demonstrate that the privilege applies. Amgen Mem. at 10-12. The court disagrees.
Our Court of Appeals has not decided whether non-confidential information is subject to
the reporter’s privilege. But, as Judge Lamberth observed in Hutira v. Islamic Republic of Iran,
“[a]ll of the federal circuit courts of appeal that have addressed this question . . . have concluded
that the privilege for journalists shields both confidential and nonconfidential information from
compelled disclosure.” 211 F. Supp. 2d 115, 120 (2002) (citations omitted). Other judges in this
jurisdiction have come to the same conclusion. See Estate of Klieman, 293 F.R.D. at 242; In re
Slack, 768 F. Supp. 2d at 194; Tripp v. Dep’t of Defense, 284 F. Supp. 2d 50, 54 (D.D.C. 2003).
As Judge Howell observed in Slack, not applying the journalist’s privilege to non-confidential
information would risk “‘wholesale exposure of press files to litigant scrutiny [and] would burden
the press with heavy costs of subpoena compliance . . . [thereby] making journalists appear to be
an investigative arm of the judicial system, the government, or private parties.” Slack, 769 F.
Supp. 2d at 194 (citation omitted) (internal quotation marks omitted). This court finds Slack and
the other cited decisions persuasive and therefore rejects Amgen’s contention that the privilege
does not apply to non-confidential information.
That said, courts have recognized that a party seeking to compel reporter testimony faces
a less weighty burden when seeking non-confidential information. “[W]hen a party seeks to
compel a reporter to testify regarding nonconfidential information, the risk of debilitating a
journalist’s ability to gather information is considerably diminished. Consequently, the showing
needed to overcome a reporter’s privilege when the information sought is nonconfidential is less
demanding than the showing required where confidential materials are sought.” Id. (citations
omitted) (internal quotation marks omitted). In light of Amgen’s concession that it does not seek
8
to elicit confidential information from Goldberg, the court will apply this less rigorous standard to
Amgen’s request to abrogate the privilege.
2. The Evidence Amgen Seeks From Goldberg
Amgen contends that it should be permitted to depose Goldberg as to five categories of
information. First, Amgen seeks to have Goldberg confirm that, before publishing the Article, he
spoke to “Wall Street sources who monitor Amgen”—a fact to which Goldberg attested in his
Supplemental Declaration—and describe the circumstances of those conversations, particularly
how long they occurred before the Article’s publication. Mot. Hr’g Tr. 10-12, 17; Amgen Mem.
at 10. Second, Amgen wants Goldberg to verify the statements attributed to Dr. Henke and
Dr. Bennett in the Article, particularly that both were aware of the DAHANCA 10 study before
Goldberg contacted them. Mot. Hr’g Tr. 18-19; Amgen Mem. at 11. Third, Amgen would like to
inquire if anyone else with whom Goldberg spoke was aware of the DAHANCA 10 study before
the article was published. Mot. Hr’g Tr. 21; Amgen Mem. at 10. Fourth, Amgen wants to know
how Goldberg came to learn of the DAHANCA 10 study. Mot. Hr’g Tr. 17. And, finally, Amgen
would like to inquire about Goldberg’s contention, contained in his motion to quash, that
information about the DAHANCA 10 study was “widely available.” Amgen Mem. at 14 (citing
Goldberg Mem., ECF No. 1-1, at 14).
The court considers below the parties’ arguments with respect to each of these subject
matters, evaluating them against the two Zerilli factors: the litigant’s need for the information and
the litigant’s diligence in obtaining the information from alternative sources.
a. Confirming Goldberg’s conversations with Wall Street sources
Amgen argues that it is critical to its defense for Goldberg to confirm that he spoke to two
Wall Street sources before the Article’s publication. Such testimony, Amgen argues, will bolster
9
its defenses that (1) the Article’s disclosure of the DAHANCA 10 study did not cause the
shareholders’ losses, i.e., the “loss causation” element of a securities fraud claim,1 and (2) the
market was aware of the DAHANCA 10 study and thus Amgen was excused from disclosing it,
i.e., its “truth-on-the-market” defense, see In re Apple Computer Sec. Litig., 886 F.2d 1109, 1115
(9th Cir. 1989) (concluding that “the defendant’s failure to disclose material information may be
excused where that information has been made credibly available to the market by other sources”).
In other words, according to Amgen, Goldberg’s testimony about his Wall Street sources would
show that the DAHANCA 10 study was known to market actors and thus the public. Therefore,
the alleged corrective statement contained in The Cancer Letter could not have caused the
shareholders’ loss and Amgen was excused from its alleged non-disclosure.
The court agrees with Amgen that evidence showing that Wall Street sources spoke to
Goldberg about the DAHANCA 10 study before the Article’s publication goes to the “heart” of its
defenses. The decision in Padnes v. Scios Nova, Inc., No. C 95-1693, 1996 WL 539711 (N.D.
Cal. Sept. 18, 1996), illustrates how establishing market actors’ knowledge of an alleged
undisclosed clinical study’s negative results can be used to defend against a claim of securities
fraud. There, the court dismissed plaintiffs’ complaint, based on a truth-on-the-market defense,
where the supporting evidence showed that a clinical study was (as here) published and that market
analysts were aware of the study’s findings. Id. at *8. Admittedly, Goldberg cannot provide the
kind of direct evidence of market actors’ knowledge that was present in Padnes—there, at least
one analyst had read about the study and reported it and others were aware of aspects of the study.
Id. at *7-8. However, the fact that Goldberg alerted market actors to the DAHANCA 10 study
1
“Loss causation” is defined as the “causal connection between the material misrepresentation and the loss.” Dura
Pharm., Inc. v. Broudo, 544 U.S. 336, 342 (2005) (citations omitted).
10
some time before the Article’s publication, in combination with other evidence, could bolster
Amgen’s defenses.
Amgen’s case to compel Goldberg’s testimony, however, founders on the second Zerilli
factor, as Amgen has failed to demonstrate the requisite diligence in seeking evidence from
alternative sources. Amgen argues that it cannot reasonably be expected to depose a large quantity
of “securities analysts covering the biotechnology sector” in the hope of finding the two with
whom Goldberg spoke. Amgen Mem. at 17 (citation omitted). Unfortunately for Amgen, that is
precisely what the law in this circuit requires.
The Court of Appeals’ decision in Lee v. Department of Justice, 413 F.3d 53 (D.C. Cir.
2005), is controlling. There, the plaintiff sued the Department of Justice under the Privacy Act for
allegedly leaking information to the news media about him during an espionage investigation. To
unearth the identity of the leaker, Lee deposed no less than 20 government officials from the
Department of Justice, the Federal Bureau of Investigation, and the Department of Energy. Id. at
56. Only after those efforts failed to identify the leaker did Lee seek to compel testimony from
members of the media, which the district court granted. Id. On appeal, the Court of Appeals
affirmed the order of compulsion and rejected the argument that Lee had not exercised enough
diligence in attempting to uncover the leaker’s identity through alternative sources. Id. at 61.
“[T]he number of depositions necessary for exhaustion,” the court stated, “must be determined on
a case-by-case basis.” Id. A party need not undertake the onerous burden of deposing every
conceivable source to satisfy the exhaustion requirement. Id. But where there is a reasonably
well-defined number of possible alternative sources for the same information sought from the
journalist, the party seeking to compel testimony must demonstrate reasonable efforts to exhaust
those possible sources. See id.11
Here, Amgen’s counsel candidly acknowledged at oral argument that, in 2007, there “were
25 or 26 securities analysts covering Amgen.” Mot. Hr’g Tr. 14. Thus, this is not the kind of case
where the possible alternative sources are indefinite in number or so ill-defined as to make it utterly
unreasonable to expect a party to meet the exhaustion requirement. See Carey, 492 F.2d at 638(stating that where the identifying characteristics of possible alternative sources are “vague,” the
concept of exhaustion of remedies is inapplicable). According to Amgen’s counsel’s affidavit,
Amgen deposed no Wall Street analysts before subpoenaing Goldberg. Dixon Decl. ¶ 23. Such
complete lack of effort at exhausting alternative avenues is plainly insufficient to abrogate the
reporter’s privilege. See Zerilli, 656 F.2d at 715 (refusing to compel reporter’s testimony where
party had not taken any depositions of possible alternative sources, even if discovery would have
been “time-consuming, costly, and unproductive”); Clyburn v. News World Commc’n, Inc., 903
F.2d 29, 35 (D.C. Cir. 1990) (reporter’s privilege upheld where party “utterly failed to pursue
obvious alternative sources of information”); Tripp, 284 F. Supp. 2d at 60-61 (quashing subpoena
where “plaintiff has not even attempted to obtain the information requested from other sources”).
Even if, as counsel represented at oral argument, Amgen deposed three analysts in 2013, Mot. Hr’g
Tr. 35, deposing only three out of 25 or 26 analysts two years ago falls well short of the kind of
effort at exhausting alternative sources that a party must show before compelling the disclosure of
a journalist’s First Amendment activities. That is true, even under the more relaxed standard for
obtaining non-confidential information.
At oral argument, Amgen urged the court to follow the Second Circuit’s decision in
Garland v. Torre, 259 F.2d 545 (2d Cir. 1958), which the Court of Appeals cited favorably in
Carey v. Hume, 492 F.2d at 634-35. Mot. Hr’g Tr. 16. In Garland, the court affirmed an order
compelling a journalist’s testimony after the plaintiff had conducted only three depositions.
12
259 F.2d at 547, 551. But just as the Court of Appeals has not set a minimum number of
depositions necessary to establish exhaustion, see Lee, 413 F.3d at 61, nor has it set a maximum.
The inquiry is fact-specific. Id. In Garland, the court deemed three depositions sufficient because
the plaintiff had identified two possible persons as the source of the defamatory statements at issue
and had deposed three witnesses without success. 259 F.2d at 551. Here, Amgen knows of 25 or
26 analysts who followed the company at the time the DAHANCA 10 study was conducted.
Deposing, at most, only three of them to discover which of them might have spoken to Goldberg
and thus learned about the study before the Article’s publication, is not enough. Deposing a large
number of analysts in the hope of finding the two that spoke to Goldberg might indeed be akin to
looking for a needle in a haystack, as Amgen contends. But the law requires Amgen to have
conducted a thorough search of the hay before deposing Goldberg, which it failed to do.
b. Verification of Dr. Henke’s and Dr. Bennett’s statements
Next, Amgen seeks to have Goldberg verify the statements attributed in the Article to
Dr. Henke and Dr. Bennett, specifically their statements that they were aware of the DAHANCA
10 study before Goldberg spoke to them. As to that subject, Amgen has neither established the
criticality of the evidence sought nor the requisite diligence in obtaining it from alternative sources.
Merely confirming that Dr. Henke and Dr. Bennett were aware of the DAHANCA 10 study
before the Article’s publication provides little weight to Amgen’s defenses. Unlike a market
analyst or other Wall Street observer, both are physicians whose pre-publication knowledge of the
study does little to establish that the market was aware of the DAHANCA 10 results with the
“degree of intensity and credibility” required under Ninth Circuit law to excuse Amgen’s
nondisclosure. See Apple Computer, 886 F.2d at 1116 (stating that “any material information
which insiders fail to disclose must be transmitted to the public with a degree of intensity and
13
credibility sufficient to effectively counter-balance any misleading impression created by the
insiders’ one-sided representations.”)
Moreover, Amgen has not sought to obtain the information from the obvious alternative
sources—the doctors themselves. See Hutira, 211 F. Supp. at 117, 121-22 (quashing subpoena
where party sought reporter’s testimony “to confirm the accuracy of certain statements and
quotations in his article” where party had not “attempt[ed] to contact the other individuals
discussed in the article” before seeking to depose the reporter); see contra NLRB v. Mortensen,
701 F. Supp. 244, 249-50 (ordering reporters to testify only after the statements’ sources denied
making them).2 If confirming Dr. Bennett’s pre-Article knowledge of the study is as critical as
Amgen claims, it would have been a simple matter to serve him with a subpoena for testimony.
Instead, Amgen only served him with a document subpoena, which of course could not provide
Amgen with the admissible testimonial evidence it seeks.
Amgen also did not seek evidence from Dr. Henke. Admittedly, the burden with regard to
Dr. Henke is far greater. As Dr. Henke is a resident of Germany, Amgen would have been required
to proceed under The Hague Convention to secure evidence from him. But unfortunately for
Amgen, there is no foreign evidence exception to the exhaustion requirement. Litigation in this
case commenced at least two years ago3 and, in that time, Amgen has not made any effort to secure
admissible evidence from Dr. Henke. Nor has it shown that obtaining evidence from Dr. Henke
would be particularly burdensome or difficult to obtain under an international convention.
Cf. Estate of Klieman, 293 F.R.D. at 243 (finding that party was not required to attempt securing
2
The courts in two out-of-jurisdiction cases cited by Amgen, see Amgen Mem. at 15, likewise compelled a reporter’s
testimony to confirm published statements only after the sources denied making them. See SEC v. Seahawk Deep
Ocean Tech., Inc., 166 F.R.D. 268, 271 (D. Conn. 1996); In the Matter of the Application of Michael Waldholz, No.
87 CIV 4296, 1996 WL 389261, at *2 (July 11, 1996).
3
See In re Amgen Sec. Litig., 07-cv-2536, ECF No. 349 (C.D. Cal. 2013) (scheduling order showing that discovery
commenced at least as of August 16, 2013).
14
testimony from Al-Aqsa and Fatah leaders located in the West Bank). A journalist’s privilege
should be overridden only as “a last resort.” In re Slack, 768 F. Supp. 2d at 197. In the case of
both Dr. Henke and Dr. Bennett, Amgen has failed to pursue evidence about their knowledge of
the DAHANCA 10 study from the doctors themselves.
c. Other individuals’ knowledge of the DAHANCA 10 study
Amgen also wants to know whether Goldberg spoke to anyone else who was aware of the
DAHANCA 10 study’s results before the Article was published. Presumably, Amgen would like
to ascertain this information to establish that public knowledge of the DAHANCA 10 study was
widespread.
But Amgen has not shown how it expects to obtain critical evidence by asking Goldberg
the open-ended question whether he spoke to anyone else who knew about the study before the
Article’s publication. In fact, asking the question would be pointless. Through his counsel,
Goldberg already has represented that the only sources who knew about the study before he
published the Article were Dr. Henke and Dr. Bennett. See Decl. of Jenny L. Colgate [hereinafter
“Colgate Decl.”], Ex. 6A, ECF No. 1-4. Thus, deposing Goldberg on that subject would be futile.
Amgen nevertheless argues that it should be able to depose Goldberg because questioning might
refresh his recollection about others with whom he spoke. Mot. Hr’g Tr. 18. But Amgen offers
no reason for the court to believe that Goldberg’s recall will be any different during a deposition.
The law disfavors compelling a journalist’s testimony other than in “the most exceptional cases.”
Zerilli, 656 F.3d at 712. It, therefore, would be incompatible with the privilege to allow a party to
depose a journalist based on the mere hope that questioning might jog his memory.4
4
Notably, Amgen has not sought discovery from the one person that might have information about how widespread
knowledge was about the DAHANCA 10 study before the Article’s publication—the study’s principal investigator,
Jens Overgaard. Mot. Hr’g Tr. 21-22. According to Amgen, it has concluded that Overgaard would have little to
offer other than that the DAHANCA 10 study was posted on a website. Id. at 22. If Amgen has concluded that
15
d. How Goldberg learned of the DAHANCA 10 study
Next, Amgen would like to learn how Goldberg learned of the DAHANCA 10 study.
Goldberg, through his counsel, however, already has told Amgen that he does not know how he
first became aware of the study and that he is not aware of any documents or other information
that would help refresh his recollection. See Colgate Decl., Ex. 6A. Thus, compelling Goldberg’s
testimony is unlikely to reveal what Amgen seeks. Without some reason to believe that Goldberg’s
memory could be refreshed in a deposition, the court declines to compel Goldberg to testify.
Even if Goldberg could recall when and how he learned about the study, his answer is not
likely to bolster Amgen’s defenses. Indeed, Amgen’s counsel conceded at oral argument that, if
Goldberg learned of the DAHANCA 10 study from a Google search or similar public source, it
would be “less useful to us.” Mot. Hr’g Tr. 18. Unless Goldberg learned about the study from a
direct market participant, such as a Wall Street analyst, the manner by which he learned of the
study would do little to establish that the market was aware of the DAHANCA 10 results with the
degree of intensity and credibility necessary to sustain Amgen’s defenses. Further, the record
provides little reason to believe that Goldberg’s source was a market participant. After all,
Goldberg has attested that he spoke with only two Wall Street sources and he contacted them to
find out if they were aware of the study and whether its results were relevant. Goldberg Supp.
Decl., ¶ 4. He also has stated, through counsel, that the only two people with whom he spoke
before publishing the Article who knew about the study were Dr. Henke and Dr. Bennett. Colgate
Decl. Ex. 6A. Thus, based on the record evidence, Amgen has failed to show that asking Goldberg
how he learned of the DAHANCA 10 study is likely to produce evidence that goes to the heart of
its defenses.
Overgaard would have little to offer to show the market’s knowledge of the study, it is hard to believe that Goldberg
would be able to offer any more.
16
e. The Study was “Widely Available”
Last, Amgen wants to ask Goldberg about the basis for an assertion made in his moving
papers that the DAHANCA 10 study was “widely available.” Amgen Mem. at 10 (citing Goldberg
Mem. at 14); Mot. Hr’g Tr. 21. But here Amgen is grasping at straws. The full statement at issue
is as follows: “However, because the information [about the DAHANCA 10 study] was already
on Google, it was widely available.” Goldberg Mem. at 14. As the complete text makes clear,
Goldberg asserted that the DAHANCA 10 study was “widely available” because it was posted on
the internet. It was not an expression of Goldberg’s personal belief. Therefore, Amgen has failed
to show that questioning Goldberg about his statement would produce critical evidence.
***
Finally, the court declines Amgen’s request to compel Goldberg to invoke the journalist’s
privilege on a question-by-question basis. Amgen Mem. at 11 n.5. Admittedly, the law generally
disfavors blanket assertions of privilege. See, e.g., United States v. Thornton, 733 F.2d 121, 125-
26 (D.C. Cir. 1984) (stating with regard to the Fifth Amendment privilege, “[u]sually, a trial court
cannot speculate whether all relevant questions would or would not tend to incriminate the witness;
accordingly, the court normally requires the privilege to be asserted in response to specific
questions”); United States v. Legal Servs. for N.Y.C, 100 F. Supp. 2d 42, 46 (D.D.C. 2000) (citation
omitted) (stating that “‘[b]lanket assertions of attorney-client privilege in response to a subpoena
duces tecum are strongly disfavored’”). However, courts in this jurisdiction have not required
journalists to invoke the journalist’s privilege on a question-by-question basis before obtaining
protection from the court. See, e.g., In re Slack, 768 F. Supp. 2d at 192 (quashing subpoena even
though reporter did not invoke privilege on a question-by-question basis); In re Subpoena to
17
Jeffrey Goldberg, 693 F. Supp. 2d 81, 83 (D.D.C. 2010) (same); Tripp, 284 F. Supp. 2d at 53-54(same); Hutira, 211 F. Supp. 2d at 117 (same).
There would be little purpose served here in requiring Goldberg to invoke the privilege on
a question-by-question basis. Amgen does not dispute that Goldberg is a journalist or that the
journalist’s privilege attaches to his work on the Article. Thus, this is not a case in which a
deposition would help develop foundational facts that would elucidate the privilege’s application.
Nor is Goldberg likely to provide any substantive answers that go to the heart of Amgen’s defenses.
Other than basic questions about his background, Goldberg has indicated that he will invoke the
journalist’s privilege with respect to any question related to the Article. The court sees no need to
require Goldberg to sit for a deposition in which he would invoke the journalist’s privilege to every
substantive question posed.
B. Goldberg’s Request for Attorney’s Fees
Having concluded that the subpoena to Goldberg should be quashed, the court turns to
Goldberg’s demand for attorney’s fees. Under Federal Rule of Civil Procedure 45(d)(1), “[a] party
or attorney responsible for issuing and serving a subpoena must take reasonable steps to avoid
imposing undue burden or expense on a person subject to the subpoena.” Should the issuing party
or attorney fail to comply with this duty, the issuing court “must enforce this duty and impose an
appropriate sanction—which may include lost earnings and reasonable attorney’s fees.” Id.But Rule 45(d)(1) does not require the court to impose sanctions merely because a party is
unsuccessful in its subpoena. See Alberts v. HCA Inc., 405 B.R. 498, 502-03 (D.D.C. 2009) (“The
mere fact . . . that a disputed subpoena is ultimately deemed unwarranted does not, standing alone,
demand the imposition of sanctions”); Rendon Grp., Inc. v. Rigsby, 10-mc-164, 2010 WL
4627647, at *1 (D.D.C. Nov. 8, 2010) (“Rule 45(c)(1)”—the predecessor to Rule 45(d)(1)—“does
18
not explicitly require expenses to be paid if a party is not entirely successful in its subpoena.”).
Rather, courts in this jurisdiction have looked to whether the subpoena was issued in bad faith or
for some improper purpose. See Alberts, 405 B.R. at 503 (observing that “blatant abuse of the
subpoena power is a common thread running through decisions in which sanctions have been
awarded under Rule 45(c)(1)”); Rendon Grp., 2010 WL 4627647, at *1 (refusing to impose
sanctions where the moving party’s arguments “were for the most part reasonable and justified”);
Alexander v. FBI, 186 F.R.D. 188, 196-97 (D.D.C. 1999) (declining to impose sanctions where
subpoena was “not overbroad or improper” and where motion to compel was “substantially
justified and not brought in bad faith”). See also Legal Voice v. Stormans Inc., 738 F.3d 1178,
1185 (9th Cir. 2013) (stating that a court may impose sanctions under Rule 45(d)(1) “when a party
issues a subpoena in bad faith, for an improper purpose, or in a manner inconsistent with existing
law”).
Moreover, although our Court of Appeals appears not to have addressed the issue, the
Ninth Circuit has interpreted “undue burden” under Rule 45(d)(1) as “limited to harms inflicted
by complying with the subpoena” and “not to the adjudication of related follow-on issues, such as
whether the subpoenaed information is potentially protected by a privilege.” Mount Hope Church
v. Bash Back!, 705 F.3d 418, 428 (9th Cir. 2012). See also Tiberi v. CIGNA Ins. Co., 40 F.3d 110,
112 (5th Cir. 1994) (stating that Rule 45(c) provides for sanctions against “one issuing a
vexatiously overbroad subpoena” but reversing sanctions award where party engaged in “sufficient
good faith efforts to negotiate reasonable parameters on the subpoena”). In Mount Hope Church,
the Ninth Circuit reversed a sanction imposed after a third-party had succeeded in quashing a
subpoena on First Amendment grounds. See Mount Hope Church, 705 F.3d at 423, 428-29.
The court explained: “Because [the issuing party] Mount Hope and its counsel could reasonably
19
assert that [the] First Amendment privilege did not invalidate the subpoena, they could issue
discovery process based on that reading without fear of sanctions. The Church also had objective
reason to subpoena the information and therefore complied with existing law under 45(c)(1).” Id.
After reviewing the parties’ meet and confer efforts, the court concludes that Amgen did
not fail to comply with its duty to avoid “undue burden” under Rule 45(d)(1). For six weeks,
the parties attempted to resolve this matter in good faith without success. Amgen stated from the
outset that it would not be seeking confidential source information but instead would be relying
on the lesser showing required to secure non-confidential information from a journalist. Colgate
Decl., Ex. 5B. The parties attempted to narrow the issues, but ultimately to no avail. The court
finds nothing on the record to suggest that Amgen issued the subpoena in bad faith or for an
improper purpose.
Goldberg argues that Amgen should be sanctioned because it improperly turned to him first
for information before exhausting other sources. Goldberg Mem. at 15-16. But Amgen did make
some effort to discover the information it seeks before issuing the subpoena to Goldberg. Amgen
issued document subpoenas to the doctors identified in the article and spoke with some of them.
Dixon Decl. ¶¶ 5-14. It also reviewed analyst reports and other public source information for news
of the DAHANCA 10 study. Id. ¶ 15. And, according to counsel’s representation at oral argument,
it deposed three analysts in 2013. Mot. Hr’g Tr. 13, 35. Amgen just did not do enough to overcome
the journalist’s privilege. A party’s mere failure to enforce a subpoena is not sufficient to require
that party to pay attorney’s fees under Rule 45(d)(1).
20
IV. CONCLUSION AND ORDER
For the foregoing reasons, the court grants Goldberg’s motion to quash. His request for
attorney’s fees under Rule 45(d)(1) is denied.
Dated: August 21, 2015 Amit P. Mehta
United States District Judge
21