Countering the Excessive Subpoena for Scholarly Research

I. Introduction

Researchers usually rely on others to provide the data which will become the foundation of their research, and often, in order to proceed, they must assure their sources that their identities will remain confidential. Confidential firsthand reports enable researchers to explore a plethora of important questions of health, public policy, and economics, to name a few. These studies may culminate in published reports that aggregate the data without revealing the identity or identifying details of confidential sources of information. Often, such reports can significantly advance measures to ameliorate a variety of serious problems, given their availability to the public, as well as to legislators or other policymakers.

Unfortunately, regardless of how objective research may be, it is put at risk if it becomes bogged down in the muddy fields of litigation. The average researcher will be ill-prepared for the sudden hurdle of a subpoena or the blocking tactics of an adversary flanked by witnesses of varying levels of expertise. Subpoenas are an omnipresent threat in all kinds of cases, from a products liability case against a cigarette manufacturer, to a criminal prosecution of a prostitute. A subpoena for confidential data is likely to be intimidating to a researcher regardless of his level of confidence in his objective evaluation of the data.

But what makes a subpoena excessive? There is no simple definition, nor bright-line rule; rather, the determination requires weighing various factors: the public interest served by the research project versus the public or private interest that prompted the subpoena; the importance of guaranteeing confidentiality in gaining access to essential information and scholarly research versus an alleged right to know the identity of the confidential sources or to review the research data.

Unfortunately, neither legislatures nor courts have granted researchers an absolute privilege to protect the confidentiality of their research data. The Supreme Court's interpretation of the First Amendment fails to provide such a privilege even for newspaper reporters.¹ Likewise, statutory or judicial rules of evidence ordinarily do not provide an explicit privilege for researchers.² Some even attempt to preclude privileges not expressly provided by statute.³ The most promising means by which to counter an excessive subpoena may be federal statutes which provide for confidentiality in certain circumstances,⁴ privacy provisions in state constitutions or laws,⁵ reporters' shield laws,⁶ or rules authorizing courts to quash or modify subpoenas or issue protective orders.⁷ While a subpoena demanding confidential research data does not automatically lead to the granting of an absolute privilege, a court may exercise its judicial discretion to undertake a balancing test that may tip the scales for confidentiality.

Although some researchers may be willing or eager to serve as witnesses themselves or to recruit knowledgeable experts, others may find the very process of litigation uncongenial or even demeaning, particularly when it involves undue interruption of work in progress. The reluctance to testify may be even greater for litigation in which they are not parties and have no interest.

This essay examines how researchers, research institutions, and their counsel may foresee and effectively counter excessive subpoenas. The best defense against an excessive subpoena comes from being alert to the possibility from the original planning of the research, and remaining alert throughout the process. Taking early precautions and remaining alert allows the researcher to take advantage of existing protections, and enables him to quickly mobilize his defense should he be served with a subpoena.⁸

II. The Planning Stage

A. Identify The Reasons For Confidentiality

Researchers should determine at the outset whether they can obtain the necessary data free from of any guarantee of confidentiality. If not, they should document the reasons requiring confidentiality. In some cases, confidentiality may be essential to protect data sources from an invasion of privacy, from embarrassment or distress, or from criminal prosecution, tax audits or other government investigations, as well as from litigation by others.

Courts and legislatures already protect trade secrets because their value and utility depend on their not being widely known.⁹ Likewise, courts should be receptive to requests to protect essential research information because its value and utility also depends on confidentiality. The researcher who prepares a written memorandum at the inception of the research setting forth the reasons for confidentiality will be well-prepared to persuade a court that the project could not have proceeded without the assurance of confidentiality.

Case law and commentators offer guidance as to when confidentiality should be safeguarded. For example, in the physician-patient context Wigmore's handbook on the Federal Rules of Evidence includes the following requirements for confidentiality:

(1) The communica-tions must originate in a confidence that they will not be disclosed.

(2) This element of confiden-tiality must be essential to the full and satisfac-tory maintenance of the relation between the parties.

(3) The relation must be one which in the opinion of the community ought to be sedulously fostered.

(4) The injury that would inure to the relation by the disclosure of the communica-tions must be greater than the benefit thereby gained for the correct disposal of litiga-tion.¹⁰

> While these requirements are for a different context, they nonetheless demonstrate the factors that are taken into account when determining whether information should be accorded confidentiality protection. Researchers would be well-advised to consider these factors before proceeding with their research.

B. Give Confidentiality Assurances Sparingly

As mentioned previously, researchers should offer assurances of confidentiality only when they are the only avenue to forthright and full disclosure of useful data. If an assurance was not necessary, a court may not be inclined to protect the data from disclosure. Moreover, if a court refuses to uphold such an assurance, researchers and their sponsors may be liable for the ensuing breach of confidentiality.¹¹

Researchers often seek protection from liability by adding a proviso that confidential data will not be disclosed .except as required by law.. Such a proviso may alert the source to the possibility of compelled disclosure and may strengthen the researchers' defense against a claim of liability premised in contract, promissory estoppel or tort in the event of such disclosure. On the other hand, such a proviso could lead the party subpoenaing the data to contend that the possibility of compelled production was anticipated, and that enforcement of a subpoena, therefore, is not inconsistent with the qualified assurance given.¹² Nonetheless, because the proviso is so broad, it does not entirely foreclose the possibility of prolonged litigation against a researcher for disclosure. Additionally, this type of proviso may dissuade some potential research subjects from participating in research studies. Therefore, because the protective effect of the proviso is questionable, researchers should consider excluding it and obtaining federal confidentiality protection where available. In other cases, a confidentiality agreement may be appropriate.

Confidentiality agreements may vary depending on the source and the extent of the data. An agreement with a chemical company involved in an environmental clean-up or an insurance company involved in mass tort litigation may provide more rules governing confidential data and subpoenas than a short form of consent and confidentiality assurance that might be used in a study of mentally ill homeless persons or elderly medical patients. Such an agreement might require notification if a subpoena is served or the use of best efforts by the researcher to resist production of confidential data, it might limit the .except as required by law. proviso to a court order, not merely a subpoena, and it might provide for return or destruction of the data at the conclusion of the study.

C. Obtain Federal Confidentiality Protection, If Available

In the area of public health, federal law offers two sources of confidentiality protection. First, there are federal statutory provisions that limit the disclosure and use of information obtained in the course of research supported or conducted by the Public Health Service.¹³ For example, one provision protects information obtained through activities carried out or supported by the Agency for Health Care Policy and Research:

[N]o information, if an establishment or person supplying the information or described in it is identifiable, obtained in the course of activities undertaken or supported under this subchapter [VII] may be used for any purpose other than the purpose for which it was supplied unless such establishment or person has consented (as determined under regulations of the Secretary) to its use for such other purpose. Such information may not be published or released in other form if the person who supplied the information or who is described in it is identifiable unless such person has consented (as determined under the regulations of the Secretary) to its publication or release in other form.¹⁴

This provision automatically grants confidentiality protection to all projects that fall within its scope. This provision is interpreted as giving researchers no discretion regarding disclosure of the protected data, regardless of who seeks it.¹⁵

Second, some public officials have the authority to grant confidentiality protection under certain circumstances.¹⁶ For instance, one such provision gives the Secretary of Health discretion to grant federal confidentiality certificates for a range of both publicly and privately funded research projects:

[T]he Secretary may authorize persons engaged in biomedical, behavioral, clinical or other research (including research on mental health, including the use and effect of alcohol and other psychoactive drugs) to protect the privacy of individuals who are the subject of such research by withholding from all persons not concerned with the conduct of such research the names and other identifying characteristics of such individuals. Persons so authorized to protect the privacy of such individuals may not be compelled in any Federal, State, or local civil, criminal, administrative, legislative or other proceedings to identify such individuals.¹⁷

To obtain a federal confidentiality certificate, a researcher must submit an application including details about the individuals having major responsibilities in the project, the research protocol, and various assurances.¹⁸ Upon receipt, the Secretary can issue a certificate, deny it and state the reasons therefor, or request additional information from the applicant.¹⁹ Once armed with a confidentiality certificate under federal law, a researcher in federal or state court, or in other proceedings, has the discretion to refuse to divulge the identity or identifying details of the individual source(s) who furnished the data upon assurance of confidentiality. The substance of the information disclosed, however, is not confidential--only the identity of the source is--and it may be aggregated with comparable data in a public report.

These two statutory provisions do overlap to some extent. While logically one might assume that confidentiality protection is enhanced when these provisions are used in combination, this is not the case. According to federal authorities, a confidentiality certificate will weaken the protection that qualifying projects receive under the self-executing statutory grant of confidentiality.²⁰ There are two explanations for this result: First, federal confidentiality certificates give the researcher discretion to disclose the protected data. In contrast, under a statutory grant of confidentiality, the researcher may disclose the protected data only if the research subject contents after notice. Second, a federal confidentiality certificate only protects the names or other identifying characteristics of the research subjects. In contrast, a statutory grant of confidentiality protects all data obtained in the course of activities falling under the scope of the statute.²¹ It is important for researchers to have knowledge of these differences so they may determine what information is protected, and when they may legally disclose protected information.

These public health statutes illustrate the variety of federal confidentiality protection which may be available to resourceful researchers. Thus, they are well-advised to investigate possible sources of statutory protection before resorting to home-spun confidentiality assurances.

II. Research in Progress

A. Unlink the Names and Identifying Details of Sources from Confidential Data and Safeguard the Data

It is an elementary precaution to immediately unlink the names and other identifying details of the study participants. The researcher should safeguard the identifying names and details and their linkage to the other data by keeping them in restricted areas or locked files, and in some cases, by destroying them.²² This data should be safeguarded until they are aggregated for publication in a report wherein no ordinary reader could identify any study participant.²³

In sum a research institution and researcher should adhere to a confidentiality plan from beginning to end. Then courts, which already protect trade secrets when their possessors have taken reasonable steps to maintain their secrecy,²⁴ should prove amenable to protecting confidential research that has been carefully safeguarded via restricting access to only those persons directly engaged in research of the data.

B. Comply with the Requirements of Your Institutional Review Board

Research projects that use confidential data from individuals are ordinarily screened by the research institution's institutional review board (.IRB.), sometimes called the human subjects protection committee.²⁵ Such a review ordinarily ensures compliance with requisite forms and procedures for obtaining the informed consent of study participants and protecting their privacy. Should litigation nonetheless ensue, the researcher could then demonstrate compliance with the requirements of its IRB.

IV. After the Subpoena Arrives

A. Consult with Your Management and Counsel Immediately

Researchers unfamiliar with court procedures and subpoenas may be jolted when served with a subpoena. They may fear intrusion upon their work, the vitiation of the assurances of confidentiality they extended, or a public array of their records and computer databases in a courtroom. Subpoenas are often phrased in extraordinarily broad and demanding terms that might further alarm researchers who are unaware that sweeping subpoenas are common and mean "just about as much as the asking price for a rug in an Oriental bazaar."²⁶ Such broad subpoenas are inconsistent with Rule 45.²⁷ Researchers might rashly contemplate either destroying, or concealing the data demanded or, conversely, divulging it in detail. However, the belief or hope that such behavior could stave off an appearance in court or at a deposition is unrealistic. Fear, destruction, concealment or the divulgence of confidential data are not merely inappropriate, but also self-defeating. Instead, the researcher should report the subpoena promptly to the appropriate officer (or other designated person) within the research institution who will usually then contact legal counsel.

B. Make Timely Service of Written Objections

A researcher's first line of defense after receiving a subpoena is to make written objections to the inspection or copying of any or all of the researcher's records and data. Federal Rule of Civil Procedure 45 requires that the procedure for making an objection be set forth in the subpoena.²⁸ Specifically, the objections must be served upon the party designated in the subpoena within fourteen days after service of the subpoena, or before the time specified for compliance if such time is less than 14 days after service.²⁹ Additionally, if records and data are being withheld because they are privileged, the claim of privilege must be made expressly, and supported by a description of the withheld material sufficient to enable the demanding party to contest the claim.³⁰

Once the researcher has objected, the party serving the subpoena may neither inspect nor copy the researcher's records and data unless the court that issued the subpoena orders to production of the documents.³¹ Once the researcher makes an objection, the party serving the subpoena can move for an order to compel production.³² Even if the court issues the order to compel production, it has a duty to protect a nonparty researcher from "significant expense" resulting from compliance with the subpoena.³³ Thus, a timely objection serves two purposes: (1) it protects the researcher's data from disclosure for a period, and (2) it lays the foundation for a claim for compensation to the researcher in the event that the court orders compliance with the subpoena.

C. Negotiate an Acceptable Limitation of the Subpoena, or Move to Quash or Modify it

Lawyers should consider the possibility of negotiating with the party subpoenaing the data. Negotiations can often result in substantial limitations on the subpoena--disclosure of data may be limited by scope, type, or time, and, moreover, limited to nonconfidential data. Negotiations may also serve to facilitate convenient arrangements for the appearance of the witness or in some cases for no appearance. Negotiation can be an effective, inexpensive, and muted way to counter intrusive subpoenas. Even if unsuccessful, reasonable attempts to negotiate may improve the researcher's stance before the court. This is reflected in the 1991 amendment to Federal Rule of Civil Procedure 45 which enlarges the protections afforded to those persons required to assist the court.³⁴ Rule 45 gives the court discretion to quash or modify a subpoena in order to protect unretained experts.³⁵ Undoubtedly, a court will be more sympathetic toward a researcher who makes a reasonable attempt to resolve the discovery conflict without court intervention.

Further, a researcher's attempts to negotiate will be considered by the court when assessing whether to impose sanctions under Rule 45.³⁶ Under Rule 45, an attorney issuing a subpoena has a duty to avoid imposing undue burden or expense on the person subpoenaed.³⁷ Breach of this duty is punishable by sanctions, which may include lost earnings and reasonable attorney fees.³⁸

Most importantly, negotiating in good faith is a prerequisite for obtaining a protective order under Federal Rule of Civil Procedure 26.³⁹ The moving party must certify both that a good faith attempt was made to resolve the discovery conflict without court intervention, and that there is good cause for a protective order.⁴⁰ Even if the moving party cannot persuade opposing parties to discuss the conflict, he must note any efforts to arrange such a discussion on the certificate.⁴¹ A court.s protective order can either condition or preclude discovery, depending on what means are necessary to protect the researcher and subjects from annoyance, embarrassment, oppression, or undue burden or expense.⁴²

If negotiations fail to limit the subpoena, the opposing party has the option of moving to quash or modify the subpoena.⁴³ Researchers and their counsel should not hesitate to file such a motion if the subpoenaing party remains obdurate against all efforts to negotiate. They may be able to persuade a court that the subpoena is unreasonably broad and that a limited response, not involving production of confidential data, would be sufficient.

In Richards of Rockford, Inc. v. Pacific Gas & Electric,⁴⁴ for example, the court addressed the question of whether the plaintiff's interest in obtaining discovery outweighed the public's interest in promoting confidential research. The plaintiff in Rockford deposed a professor who had investigated Pacific Gas & Electric ("PG & E") employees as part of his research into organizational structure and decision-making. At his deposition, the professor refused to disclose names of the PG & E employees.⁴⁵

The court found for the professor, but on the basis of Federal Rule of Civil Procedure 26, explicitly stating that, "The result reached here is not based upon any privilege; rather it is founded upon the court's supervisory powers over discovery."⁴⁶

Despite its failure to recognize a researcher.s privilege, the court affirmed the importance of maintaining confidentiality. It recognized that "[c]ompelled disclosure of confidential information would without question severely stifle research into questions of public policy, the very subjects in which the public interest is the greatest."⁴⁷

Federal courts may also protect confidential research consistent with Federal Rule of Evidence 501.⁴⁸ Even though Rule 501 does not expressly provide for a researcher's privilege, it leaves room for a case-by-case recognition of privileges.

For example, in Trammel v. United States⁴⁹ the Supreme Court considered whether to modify the marital privilege as it existed at common law. In reaching its conclusion, the Court explained the intended scope of Rule 501: "In rejecting the proposed Rules and enacting Rule 501, Congress manifested an affirmative intention not to freeze the law of privilege. Its purpose rather was to provide courts with the flexibility to develop rules of privilege on a case-by-case basis.. ⁵⁰ This rationale is reinforced by Judge Weinstein who, in his treatise on evidence, reads Rule 501's "reason and experience" requirement as calling for a balancing of interests.⁵¹

The Court furthered the recognition of a researcher.s privilege under Rule 501 in In Re Grand Jury Subpoena Dtd. January 4, 1984.⁵² In that case, a Ph.D. candidate underwent questioning during a police investigation of a fire at "Le Restaurant."⁵³ The graduate student had been observing workers at Le Restaurant for his dissertation, "The Sociology of the American Restaurant."⁵⁴ He refused to disclose any of his notes, stating only that he had promised "many" of his sources confidentiality and that the research was necessary for his thesis.⁵⁵

The court reversed an order quashing the subpoena and remanded the case so that the researcher could make

a threshold showing consisting of a detailed description of the nature and seriousness of the scholarly study in question, of the methodology employed, of the need for assurances of confidentiality to various sources to conduct the study, and of the fact that the disclosure requested by the subpoena will seriously impinge upon that confidentiality.⁵⁶

Although the court acknowledged that there was leeway in Rule 501 for a researcher's privilege, it emphasized that the privilege was not absolute.⁵⁷ Moreover, the court required the researcher to "make a good faith designation of those portions of his work arguably covered by the scholar's privilege and permit in camera inspection and redaction by the court."⁵⁸

Under the recently amended Rule 45(c)(3) of the Federal Rules of Civil Procedure:

(A) On timely motion, the court by which a subpoena was issued shall quash or modify the subpoena if it . . .

(iii) requires disclosure of privileged or other protected matter and no exception or waiver applies, or
(iv) subjects a person to undue burden.

(B) If a subpoena

(i) requires disclosure of a trade secret or other confidential research, development, or commercial information, or
(ii) requires disclosure of an unretained expert's opinion or information not describing specific events or occurrences in dispute and resulting from the expert's study made not at the request of any party, . . . the court may, to protect a person subject to or affected by the subpoena, quash or modify the subpoena, or if the party in whose behalf the subpoena is issued shows a substantial need for the testimony or material that cannot be otherwise met without undue hardship and assures that the person to whom the subpoena is addressed will be reasonably compensated, the court may order appearance or production only upon specified conditions.⁵⁹

The incompleteness of a research project may also bear on a court's decision to quash or modify a subpoena. For example, the court in Application of American Tobacco Co.,⁶⁰ while accepting the concept of a researcher's privilege, confined the privilege to unpublished works. The court recognized that a researcher whose work was still in progress should be protected against the loss of time, opportunity, and/or academic freedom that might ensue from a premature breach of confidentiality.⁶¹ Moreover, the court deemed it important that premature disclosure in this case would have denied the doctors the opportunity of first publication of their studies.⁶²

In contrast, the court had reservations about protecting published findings. It reasoned that published research must undergo public scrutiny to ensure its credibility and hence should not heedlessly be protected from such scrutiny.⁶³

In Dow Chemical Co. v. Allen,⁶⁴ the court discussed various factors to be weighed with respect to subpoenas involving published findings. Dow involved a request for research data regarding the toxic effects of an herbicide, TCDD. At the time of the request, the researcher had not completed his project.⁶⁵ The Dow court acknowledged that the researcher's "interest in academic freedom may properly figure into the legal calculation of whether forced disclosure would be reasonable."⁶⁶ The Court premised its conclusion on the long history of support for academic freedom evidenced by Supreme Court precedents--precedents which create a formidable barrier to intrusion on research. ⁶⁷

The court also reaffirmed earlier holdings that had factored a researcher's third party status into the weighing process, noting that a researcher's distance from the litigation tends to make compliance with a subpoena more burdensome.⁶⁸

Finally, the court spelled out various harmful effects that could ensue from premature disclosure:

that public access to the research data would make the studies an unacceptable basis for scientific papers or other research; that peer review and publication of the study was crucial to the researchers credibility and careers and would be precluded by whole or partial public disclosure of the information; that loss of the opportunity to publish would severely decrease the researchers' professional opportunities in the future; and that even inadvertent disclosure of the information would risk total destruction of months or years of research."⁶⁹

Robert O'Neil, noting "that the researcher enjoys at least minimal constitutional protection," has offered "some practical desiderata" for such cases, including "the potential utility of alternative sources of information;" the necessity for "proof of the probative value of the information;" "the relationship of the researcher to the litigation;" whether the proceeding is a routine civil suit or one of "those rare criminal cases in which information held only by a third party might be critical to a suspect's defense;" "the effects of disclosure upon persons other than the subpoena respondent;" "the status of the research," e.g., whether the demand is "for data supporting a finished publication" or "for raw material in progress;" "the effects of research findings or results upon eventual publication;" the "principal investigator's reasonable expectations;" and "the contribution of each decision to transcendent principles of free inquiry and the advancement of knowledge."⁷⁰

D. Seek an Adequate Protective Order

If a researcher is unable to limit a subpoena to nonconfidential data through negotiations or court order, a court may nonetheless be receptive to a researcher's claim that at the very least he should be granted a protective order for his research.⁷¹ Some of the benefits of a protective order are as follows: It can protect data from access by anyone other than the attorneys, experts, and others working directly on the case; it can limit the use of the data to the court proceeding alone; it can provide for confidentiality monitoring during the use of the data; it can assure the return of the data when the proceedings are funished; and it can require all persons utilizing the data to give assurances of confidentiality and provide for a contempt of cour charge in the event of breach.

In Farnsworth v. Proctor & Gamble Co.,⁷² the plaintiffs brought a products liability action against a tampon manufacturer claiming that they had been injured by Toxic Shock Syndrome ("TSS"). In connection with the action, Proctor & Gamble ("P & G") served a discovery request on the Center for Disease Control ("CDC"), a nonparty.⁷³ P & G sought the names and addresses of women who had participated in the CDC's TSS studies.⁷⁴ The CDC researchers gathered sensitive information from female participants, but the CDC did not give its research subjects any confidentiality assurances.⁷⁵ The CDC produced all of its TSS-related research documents but refused to disclose the identities of its study participants.⁷⁶ P & G sought to compel disclosure, alleging that the information was necessary to discover biases in the studies' methodology.⁷⁷ The CDC refused to comply with the request on the ground that source disclosure would inhibit its ability to conduct future studies.⁷⁸

The circuit court held that the lower court's protective order was permissible under Federal Rule of Civil Procedure 26(c).⁷⁹ It explicitly stated that the confidentiality of sources could be maintained even absent a recognized researcher's privilege, finding that "Rule 26(c) gives the district court discretionary power to fashion a protective order. The decision does not depend upon a legal privilege."⁸⁰

In addition, the court deemphasized the need for express assurances of confidentiality, finding that confidentiality can be protected even absent an agreement. "Even without an express guarantee of confiden-tiality there is still an expectation, not unjus-tified, that when highly personal and potentially embarrass-ing information is given for the sake of medical research, it will remain private."⁸¹

Courts will weigh the the competing interest when determining whether to grant an order protecting confidential sources. A protective order, by deterring lengthy proceedings, lessens the risks to researchers. They may be spared the lost time and opportunity costs that result from protracted hearings,⁸² the compulsory disclosure of information that might jeopardize future research by "reducing, rather than increasing, the production of information useful to the resolution of lawsuits,"⁸³ and the concomitant risk of losing the cooperation of research subjects and even scientific colleagues.⁸⁴

In U.S. v. Private Sanitation Industry Assoc. of Nassau/Suffolk, Inc.,⁸⁵ the government brought a civil RICO action against several defendants claiming that the defendants were part of a criminal organization that controlled the garbage industry in parts of New York. Defendants served a deposition subpoena on Peter Reuter, a nonparty economist who had authored a RAND study eight years prior to the litigation entitled Racketeering in Legitimate Industries: A Study in the Economics of Intimidation. Reuter's report may have contributed to the government's decision to pursue a RICO action against defendants, and the government considered using the report as substantive evidence in the case. Defendants claimed that their sole motivation for deposing Reuter was to counter the government's use of Reuter's report.

As counsel for Reuter, I moved for a protective order. The district court granted the motion for a protective order conditioned upon the government's assurance that it would not use the Reuter report offensively.⁸⁶ If the government later decided to use the report, the court required it to give defendants sixty days advance notice.⁸⁷ At that point, Reuter would be obligated to respond to the defendants. written interrogatories. If interrogatories did not yield satisfactory responses, then the court could order a limited deposition. To date, Reuter has not been required to submit to any discovery.

The district court in Private Sanitation noted several competing factors in its analysis:

I'm fully aware . . . that if scholars could not conduct research, which would essentially involve the receipt of information that is closely guarded, many public purposes would be disserved. . . . I understand that public policy. I also understand that a defendant who the government is seeking to deprive of a good portion of his wealth also has certain rights that have to be balanced against that scholar's rights. And I would say that the best formula would be one that accommodates both, if such a formula is available.⁸⁸

Dr. Reuter is not a party to this litigation. In balancing equities, one of the factors that are [sic] thrown into the scales is the fact that a person who is not involved in a dispute cannot be compelled, expert or not, to expend time and treasure to serve as a minimally-paid witness, expert or otherwise.⁸⁹

There is a distinction between the deference I have to show a criminal defendant and the deference to be shown a civil defendant. Liberty is still a higher value in the value scales than the mere loss of money.⁹⁰

The court's approach reflects the complicated weighing process that accompanies the issuance of a protective order.

In their deliberations on a protective order, courts may also consider the status of the subpoenaed party. It might be a private research institution that is privately funded, a private research institution that is publicly funded, or a federally mandated research institution. Courts may be disinclined to intercede on behalf of the last category.⁹¹ There is some support for the proposition, however, that a private institution's receipt of federal funds does not reduce its confidentiality protections.⁹²

Once a court issues a protective order ensuring confidentiality, it may impose sanctions against any party that attempts to violate confidentiality.⁹³

Whatever balance of competing interests a protective order may represent, it still may not settle all matters. It still involves a divulgence of confidential data to persons who were not intended recipients for purposes that were not intended. Moreover, it transfers some control over the data from the researcher to the court, with the attendant risk that a busy court will not have time for the requisite diligence to protect data adequately. There also remains the lingering possibility that if the court does not hold an evidentiary hearing or call for and review thorough affidavits at the outset, it may not be sufficiently informed of the interests of the litigants or of the public, to balance them adequately.⁹⁴

V. When Disclosure Becomes Necessary

A. Seek Recovery for the Costs of Compliance with a Subpoena when Possible and Appropriate

Courts may take steps to ease the financial burden on a nonparty researcher incurred from compliance with a subpoena, even when such an accommodation might not be provided to a party.⁹⁵ Federal Rule of Civil Procedure 45 contains two provisions that allow nonparty researchers to recover the costs of compliance.⁹⁶ First, researchers will be "reasonably compensated" where a subpoena requires disclosure of information and opinions obtained through a study which was not conducted at the request of any party.⁹⁷ Second, if the court issues an order to compel production, the court has a duty to protect a nonparty researcher from "significant expenses" connected with the production of documents.⁹⁸ Some courts merge these two provisions, even though they provide distinct and separate grounds for recovery.⁹⁹

---

¹ See Branzburg v. Hayes, 408 U.S. 665 (1972) (holding that reporters are not exempt from the duty to appear before a grand jury and answer relevant questions). See also In re Grand Jury Proceedings (James Richard Scarce), 5 F.3d 397 (9th Cir. 1993), cert. denied, 114 S.Ct. 685 (1994) (citing Branzburg for the position that there is a reporters.s privilege, and denying Scarce.s claim that a scholar.s privilege is analogous).

² See, for example, the absence of such a privilege in CAL. EVID. CODE §§ 900-1063 (West 1996).

³ See, e.g., CAL. EVID. CODE § 911 (West 1996). See also University of Pennsylvania v. EEOC, 493 U.S. 182 (1989). Cf. Jaffee v. Redmond, No. 95-266, 1995 U.S. LEXIS 3879 (June 13, 1996).

⁴ See, e.g., 42 U.S.C. §§ 241(d); 242m (d); 299a-l(c).

⁵ See, e.g., CAL. CONST. ART. I, § 1.

⁶ See, e.g., id. § 2; CAL. EVID. CODE § 1070; see Delaney v. Superior Court, 50 Cal.3d 785, 789 P.2d 934 (1990) (holding that the shield law does not require showing by a newperson that information was obtained in confidence).

⁷ See, e.g., FED. R. CIV. P. 45(c); FED. R. CIV. P. 26(c). See David D. Siegel, Federal Subpoena Practice Under The New Rule 45 of The Federal Rules of Civil Procedure, 139 F.R.D. 197, 231-36 (1992).

⁸ This essay makes no distinction between researchers and research institutions. It is a fair assumption that confidentiality assurances made by a researcher are authorized by the supervising research institution and that their interests in protection and nondisclosure are shared and do not conflict in any material way. If they arrive at a final juncture of deciding whether to risk contempt for violating a court order, it is possible, but by no means certain, that the individual researcher and the institution might view the issues differently. In that event each should be counseled separately.

⁹ See, e.g., UNIF. TRADE SECRETS ACT, CAL. CIV. CODE § 3426.1(d)(1); RESTATEMENT (THIRD) OF UNFAIR COMPETITION §§ 39-45 (1995).

¹⁰ 8 JOHN H. WIGMORE, EVIDENCE § 2285, at 52, (McNaughton rev. 1961 & Supp. 1991) (emphasis in original).

¹¹ See Cohen v. Cowles Media Co., 501 U.S. 663 (1991) (holding that the First Amendment does not bar a promissory estoppel action for breach of assurance of confidentiality); G. Michael Harvey, Confidentiality: A Measured Response to the Failure of Privacy, 140 U. PA. L. REV. 2385 (1992) (arguing for the creation of an action in tort for disclosure of private facts against the source of disclosure).

¹² See, e.g., Atlantic Sugar Ltd. v. U.S., 85 Cust. Crt. 128, (1980) (the court compelled disclosure of a nonparty's answer the questionnaire, noting that persons who responded to an International Trade Commission questionnaire were informed that the information would not be disclosed "except as required by law," ).

¹³ E.g., 42 U.S.C. §§ 242m(d), 299a-1(c) (19 ).

¹⁴ 42 U.S.C. § 299a-1(c) (19 ).

¹⁵ Memorandum from Susan Greene Merewitz, Senior Attorney in the Office of General Counsel of the Department of Health and Human Services, to John P. Fanning, Senior Health Policy Advisor 1 Jan. 30, 1995 (on file with____) [hereinafter Merewitz memo].

¹⁶ 42 U.S.C. § 241(d) (19__) (giving Secretary of Health discretion to grant federal confidentiality certificates for biomedical, behavioral, clinical, and other research); 21 U.S.C. § 872 (c) (giving Attorney General discretion to authorize grants of confidentiality for educational and research programs directly related to the enforcement of laws under the Attorney General's jurisdiction).

¹⁷ 42 U.S.C. § 241(d).

¹⁸ See 42 C.F.R. § 2a.
Researchers should be aware that persons entering the project after the expiration date of the confidentiality certificate are not protected from disclosure. However, researchers may submit a written request for an extension of the expiration date. Upon approval, the Secretary of Health will issue an amended confidentiality certificate. 42 C.F.R. § 2a.6(c).

¹⁹ 42 C.F.R. § 2a.6(b).

²⁰ See Merewitz memo, supra note 16, at 2.

²¹ See Merewitz memo, supra note 16, at 2-3.

²² See Paul Nejelski & Lindsey M. Lerman, A Researcher-Subject Testimonial Privilege: What to Do Before the Subpoena Arrives, WIS. L. REV. 1085, 1096-98 (1971).
I note but do not discuss in this article the important and separate question of whether and how studies based on confidential data can be verified or tested reliably without compromising confidentiality.

²³ It is possible in some circumstances that an insider who is highly knowledgeable and astute, for example an industrial participant in a confidential study of an industry, might infer that certain data reported in the aggregate possibly relates to a particular participant. Holding researchers and research institutions to the exacting standard that it must be impossible for even the most informed and astute reader to extract individual data would be counterproductive because it would lead to dilution of reported data that could render tables and conclusions so general that they would be meaningless. I am not aware of any case on the point.

²⁴ See, e.g., UNIF. TRADE SECRETS ACT; CAL. CIV. CODE § 3426.1(d)(2).

²⁵ 42 U.S.C. § 289(a); 45 CFR §§ 46.101-.409. Various federal agencies have recently adopted a common Federal Policy for the Protection of Human Subjects. 56 F.R. 28003 (June 18, 1991).

²⁶ Deitchman v. E.R. Squibb & Sons, Inc., 740 F.2d 556, 560 (7th Cir. 1984).

²⁷ E.g., Audiotext Communications Network, Inc. v. U.S. Telecom, Inc., No. 94-2395-GTV, 1995 U.S. Dist. LEXIS 15416, at *18-19 (D. Kan. October 5, 1995) ("Use of too all-encompassing language . . . violates Fed. R. Civ. P. 45").

²⁸ FED. R. CIV. P. 45(a)(1)(D).

²⁹ Id. 45(c)(2)(B).

³⁰ Id. 45(d)(2).

³¹ Id.

³² Id.

³³ Id. 45(c)(2)(B).

³⁴ Id. 45 advisory committee's note.

³⁵ Id. 45(c)(3)(B).

³⁶ See High Tech Med. Instrument., Inc. v. New Image Industries, Inc., No. C-93-4152 SBA(PJH)-ENE, 1995 U.S. Dist. LEXIS 16504, at *3, *6 (N.D. Cal. March 17, 1995) (counsel breached duty under Rule 45 by failing to take reasonable steps to avoid imposing an undue burden on a nonparty).

³⁷ FED. R. CIV. P. 45(c)(1).

³⁸ Id. ³⁹ Id. 26(c).

⁴⁰ Id.

⁴¹ Id. advisory committee's note, subdivision (c).

⁴² Id. 26(c).

⁴³ See, e.g., id. at 45(c); CAL. CODE CIV. PROC. § 1987.1. See also Amendments to the Federal Rules of Civil Procedure, Communication from the Chief Justice of the United States, Comm. on the Judiciary, House of Representatives, 102 Cong., 1st. Sess., H. Doc. 102-77, pp. 11-20, 132-150 (May 1, 1991).

⁴⁴ 71 F.R.D. 388 (N.D. Cal. 1976).

⁴⁵ Id. at 389.

⁴⁶ Id. at 389 n.2. See United States v. Doe, 406 F.2d 328 (1st Cir. 1972); Richard L. Marcus, Discovery Along The Litigation/Science Interface, 57 BROOKLYN L. REV. 381, 400-11 (1991) (considering the possibility of a researcher.s privilege, and concluding that one should not be created).

⁴⁷ Richards, 71 F.R.D. at 390. See Robert M. O'Neil, Scientific Research and the First Amendment: An Academic Privilege, 16 U.C. DAVIS L. REV. 837, 848-49 (1983).

⁴⁸ FED. R. EVID. 501 (privileges in certain actions "shall be governed by the principles of the common law . . . in the light of reason and experience"). See John H. Derrick, Annotation, "Scholar's Privilege" under Rule 501 of Federal Rules of Evidence, 81 A.L.R. FED. 904 (1993).

⁴⁹ 445 U.S. 40 (1980) (holding that the marital privilege vests only in witness spouse who may neither be compelled to testify nor foreclosed from testifying, thereby modifying earlier rule that the testimony of one spouse against the other was barred unless both consent).

⁵⁰ Id. at 47 (citing 120 CONG. REC. 40891 (1974) (statement of Rep. Hungate)).

⁵¹ 2 JACK B. WEINSTEIN & MARGARET A. BERGER, WEINSTEIN'S EVIDENCE §§ 501 [03] (1996).

⁵² 750 F.2d 223 (2nd Cir. 1984).

⁵³ Id. at 224.

⁵⁴ Id.

⁵⁵ Id.

⁵⁶ Id. at 225.

⁵⁷ Id.

⁵⁸ Id. at 226. ⁵⁹FED. R. CIV. P. 45(c)(3).

⁶⁰ 880 F.2d 1520 (2d Cir. 1987). A state court in New York had earlier quashed a similar subpoena. In re R.J. Reynolds Tobacco Co., 518 N.Y.S. 2d 729 (1987).

⁶¹ See id. at 1524, 1528.

⁶² See id.

⁶³ Id. at 1529-30.

⁶⁴ 672 F.2d 1262 (7th Cir. 1982).

⁶⁵ Id. at 1266.

⁶⁶ Id. at 1276-77.

⁶⁷ Id. at 1275.

⁶⁸ Id. at 1277.

⁶⁹ Id. at 1273. See also O'Neil, supra note 47, at 839-40, 850-51.

⁷⁰ O'Neil, supra note 47, at 853-55.

⁷¹ See FED. R. CIV. P. 26(c), 45(c)(3). See also Richard L. Marcus, The Discovery Confidentiality Controversy, 1991 U. ILL. L. REV. 457 (1991) [hereinafter Confidentiality Controversy]; Arthur R. Miller, Confidentiality, Protective Orders, and Public Access to the Courts, 105 HARV. L. REV. 428 (1991) (arguing against the need for reforms geared towards greater public access to discovery); Jacqueline S. Guenego, Trends in Protective Orders Under Federal Rule of Civil Procedure 26(c): Why Some Cases Fumble While Others Score, 60 FORDHAM L. REV. 541 (1991). See also Lloyd Doggett & Michael J. Mucchetti, Public Access to Courts: Discouraging Secrecy in the Public Interest, 69 TEX. L. REV. 643 (1991) (defending Texas Rule of Civil Procedure 76a which establishes a presumption of openness in most civil court records).

⁷² 758 F.2d 1545 (11th Cir. 1985).

⁷³Id. at 1546

⁷⁴Id.

⁷⁵Id.

⁷⁶Id.

⁷⁷Id.

⁷⁸Id.

⁷⁹Id. at 1547.

⁸⁰ Id. at 1548. For additional cases discussing the discretion of the court under Rule 26(c), see Buchanan v. American Motors Corp., 697 F.2d 151 (6th Cir. 1983); Kaufman v. Edelstein, 539 F.2d 811 (2d Cir. 1976); In re The Exxon Valdez Re: All cases, Misc. No. 92-0072-RV-C (S.D. Ala.) (protective order filed June 12, 1993). With respect particularly to grand jury subpoenas or government criminal investigations, see supra note 2, and compare Martindell v. Int'l Tel. & Tel. Corp., 594 F.2d 291 (2d Cir. 1979) with In re Grand Jury Proceedings (Billy J. Williams), 995 F.2d 1013 (11th Cir. 1993). With respect to subpoenas by the defendant in a criminal case, see United States v. Cutler, 6 F.3d 67 (2d Cir. 1993); Milo Geyelin, Court Limits Shield of Reporters' Notes in Criminal Trials, Wall St. J., Sept. 28, 1993, at B8.

⁸¹ Farnsworth, 758 F.2d at 1547.

⁸² See Virginia G. Maurer, Compelling the Expert Witness: Fairness and Utility under the Federal Rules of Civil Procedure, 19 GA. L. REV. 71, 102-03 (1985).

⁸³ Id. at 113.

⁸⁴ See David A. Kaplan & Brian M. Cogan, The Case Against Recognition of a General Academic Privilege, 60 J. URBAN L. 205, 224-25 (1983).

⁸⁵ No. CV 89-1848 ILG (E.D.N.Y. July 8, 1994).

⁸⁶ The court declined to find that Reuter was protected from discovery under the reporter's privilege.
"I can't see that Dr. Reuter . . . [has] a privilege -- I'm talking about the balancing of equity and Rule 501. . . . [T]here is no reporter privilege engraved into federal privilege . . . . [I]t's not a privilege that I am compelled to absolutely respect." Transcript of Hearing Before the Honorable A. Simon Chrein, U.S. Chief Magistrate Judge at 24-25:9-1.

⁸⁷Id. at

⁸⁸ Id. at 13-14:21-7.

⁸⁹ Id. at 19-20:21-1.

⁹⁰ Id. at 21-22:24-2.

⁹¹ Joe S. Cecil & Eugene Griffin, The Role of Legal Policies in Data Sharing, in SHARING RESEARCH DATA 148, 150 (Fienberg et al., eds., 1985).

⁹² Forsham v. Harris, 445 U.S. 169 (1980) (researcher's data was not accessible under Freedom of Information Act merely because researcher received federal funds); cf. Board of Trustees of The Leland Stanford Junior University v. Sullivan, 773 F.Supp. 472 (D.D.C. 1991) (in ordering that a government contract be awarded to Stanford without including a provision requiring the approval of a contracting officer or other government official prior to publication or discussion of preliminary research results, the court held that Stanford could "use its own judgment on when and where to publish, notwithstanding that its research is supported with federal funds").

⁹³ Application of American Tobacco Co., 880 F.2d 1520, 1530 (2d Cir. 1989).

⁹⁴ Cf. Confidentiality Controversy, supra note 71, 1991 U. ILL. L. REV. at 481 ("the ability of courts to discern whether discovery materials bear on public safety must be doubted"); Howard Schlossberg, Researcher in Exxon Case Develops System to Thwart Demands for Data, MARKETING NEWS, Sept. 16, 1991, at 7 (quoting John Petterson, who criticized a ruling that compelled him to produce confidential data pursuant to a protective order, "Exxon is important," he said, "We are nothing to the case").

It bears noting that a protective order may weaken the researcher's claim that his intellectual property has been taken unconstitutionally. See Note, Trade Secrets in Discovery: From First Amendment Disclosure to Fifth Amendment Protection, 104 HARV. L.REV. 1330, 1344-45 (1991).

⁹⁵ Wertheim Schroder & Co., Inc. v. Avon Products, Inc., No. 91 Civ. 2287(PKL), 1995 U.S. Dist. LEXIS 79, at *19 (S.D.N.Y. January 9, 1995).

⁹⁶FED. R. CIV. P. 45.

⁹⁷ Id. at 45(c)(3)(B).

⁹⁸ Id. at 45(c)(2)(B).
It bears noting that a prerequisite for the recovery of costs under this provision may be a court order to compel production. Tutor-Saliba Corp. v. U.S., 32 Fed. Cl. 609, 611 (1995). An order to compel production may be issued after the nonparty moves to quash the subpoena or after the party serving the subpoena moves to compel production.

⁹⁹ See Standard Chlorine of Delaware, Inc. v. Sinibaldi, 821 F.Supp. 232, 264-65 (D. Del. 1992) (court awarded nonparty "reasonable production costs" under its duty to protect a nonparty from "significant expense"); Wertheim Schroder, 1995 U.S. Dist. LEXIS at *19 (court cited Rule 45(c)(3)(B)(iii) for the proposition that a nonparty should be compensated where the nonparty would otherwise incur "substantial expense").