Judicially Compelled Disclosure of Researcher’s Data: A Judge’s View


It can be a vexing question for a judge whether to compel an unretained expert to disclose research data for use in litigation. Certainly, judges can sympathize with the stories of those who resist the disclosure: the scholar whose compliance with a subpoena may require her to spend weeks compiling materials for discovery and appearing for depositions to the detriment of her own research; the expert faced with the prospect of being served with discovery subpoenas in more than eighty cases and devoting his retirement to fending off subpoenas; the graduate student whose studies are brought to a halt by involuntary involvement in a grand jury investigation of a restaurant fire; or the researchers required to review and delete identifying information from ninetyseven file drawers of documents. However, judges operate within a system that places a high priority upon obtaining relevant evidence that will aid in the truth-finding process. They are not easily persuaded by claims of “burdensomeness”; they are called upon routinely to compel testimony from reluctant witnesses, and they know that burdens are inherent in any litigation.

JUDICIALLY COMPELLED DISCLOSURE OF RESEARCHERS’ DATA: A JUDGE’S VIEW, Duke University School of Law, Law and Contemporary Problems, Vol. 59: No. 3, Summer 1996.

Researchers engaged in their own truth-finding process are apt to think that the legal system looks for truth in the same way that the scientific community does, but the reality is otherwise. The two systems operate differently. Scientific research tends to be an evolving and uncertain process in which hypotheses are developed, tested, affirmed or discarded, and subjected to re-evaluation over time, whereas litigation demands certainty, even if only temporary certainty. In litigation, individual cases must be decided one way or another, whatever the evolving state of scientific information. Judicial fact-finders, whether judges or juries, must make a decision that determines the outcome of the case in front of them without waiting for science to produce a definitive answer on the issue in dispute.

Conflict between the two systems is inherent in the divergent ways in which they operate. This conflict is exacerbated by the fact that scientists are almost as ignorant of the needs and workings of the legal system as lawyers and judges are of scientific activity. This article will introduce researchers to the manner in which discovery requests are handled within the judicial system and how courts approach disputes over compelled disclosures from unretained experts. Unretained experts faced with discovery subpoenas, as well as the parties seeking to enforce such subpoenas, may find it useful to understand how judges view their disputes.

Part II of the article will examine in detail the paradigm case of compelled disclosure and look at the approach one court has established to balance the demands of the legal system with the legitimate concerns of researchers. Part III will look at why unretained experts have a duty to testify or produce evidence at all and whether such experts fit into any “privilege” category. Part IV will examine how federal courts have treated claims of researcher’s privilege. Part V will highlight the concerns of each of the participants in compelled disclosure cases, namely, judges, researchers, and litigants. Part VI will offer a brief conclusion.


The case of Deitchman v. E.R. Squibb & Sons, Inc., is the paradigmatic case of the conflict between the demands of the legal system and the legitimate concerns of researchers. Squibb and other drug companies were defendants in civil actions brought by plaintiffs who alleged that their mothers took the drug diethylstilbestrol (“DES”) while pregnant, causing plaintiffs to develop adenocarcinoma of the vagina. Squibb served a deposition subpoena upon Dr. Arthur Herbst, Chairman of the Department of Obstetrics and Gynecology at the University of Chicago, seeking every document in the records of a registry he had established of cases of vaginal and cervical adenocarcinoma contracted by women since 1940. Herbst had not treated the plaintiffs or their mothers at any time and had not agreed to serve as an expert witness for any party to the litigation. He had established the registry in 1972 to serve as a centralized repository of data on adenocarcinoma of the genital tract. In soliciting medical records of women throughout the world who were born after 1940 and had contracted this particular form of cancer, Herbst promised that all information turned over to the registry would be kept confidential. As of 1984, the registry had collected more than 500 case files, and Herbst and his colleagues had published more than a dozen articles reporting significant findings based on the data from the registry. Although Herbst refused to become involved personally in any DES litigation, DES victims used his studies in products liability cases against manufacturers of DES in an effort to establish a causal connection between their adenocarcinoma and their mothers’ use of DES.

When Herbst was served with Squibb’s subpoena, he moved quickly to quash it, arguing that the subpoena was unreasonable and oppressive and the risk of destruction of the registry outweighed the burden to Squibb of not obtaining the documents it sought. Herbst maintained that if his promises of confidentiality were breached, his sources would dry up—neither doctors nor women with adenocarcinoma would provide records to him and he would be unable to study the additional cases of clear cell adenocarcinoma expected to occur throughout the 1990s. Numerous epidemiologists and physicians supported his position with affidavits to the same effect. The federal district court agreed with Herbst that his need to retain the confidentiality of the registry was important to society, whereas Squibb’s need for the data was not truly compelling. The district court relied on several factors: The registry documents were not relevant because the major findings on which plaintiffs would rely were those in the 1971 study and the registry was not established until after that study had been made public; Herbst’s studies had been in the public domain for years and had not been challenged by his peers; Herbst would not be a witness at the trial; and Squibb had failed to make any showing of what it hoped to prove by obtaining access to the data in the registry. By contrast, Herbst had shown that forced disclosure of registry data threatened the viability of the registry because doctors would stop reporting cases if Herbst could not guarantee confidentiality. Furthermore, the court found, premature disclosure of data in an ongoing study threatened the study’s validity and usefulness by exposing information before the researchers could test and verify their conclusions.

The Court of Appeals for the Seventh Circuit reversed the district court’s decision to quash the subpoena. Although the standard of review of such a decision is the highly deferential one of abuse of discretion, the court of appeals concluded that the district court had erred in denying Squibb’s entire request.18 In reaching this conclusion, it viewed the facts differently from the district court, finding the following: Squibb had shown the plaintiffs intended to rely on more than just the 1971 study and, in fact, would be using all of the studies Herbst had published since 1971; Squibb was defending many cases in which the plaintiffs were relying on Herbst’s published studies; Herbst and the registry were the “sole monitors” for investigating a possible correlation between prenatal DES exposure and adenocarcinoma of the vagina or cervix; and Herbst’s views would be crucial to the verdict, whether or not he testified to them personally. Moreover, the court found that Squibb’s ability to defend itself was impaired by the denial of access to the registry data upon which Herbst was basing his conclusions because it could not effectively crossexamine the witnesses who relied on Herbst’s conclusions without knowing how Herbst had arrived at them.

The court of appeals observed that the trial court had proceeded erroneously on the premise that either the entire subpoena had to be obeyed or Squibb was entitled to nothing. Rather, the trial court should have recognized Squibb’s critical need for the material sought and fashioned a protective order that would guard Herbst and the registry against any loss of confidential information and unreasonable financial and temporal costs. The court referred the matter back to “the district judge to hear the parties and to fashion as inventive an order as the necessities of this unique case dictate, one which allows Squibb the least necessary amount of information to avoid a miscarriage of justice without doing needless harm to Dr. Herbst or his registry.”

A. Science and Litigation—An Uneasy Fit?

As in Deitchman, disputes over compelled discovery often arise in high stakes litigation in which one party is willing to pay whatever it takes to explore every possible line of defense because the price of losing is so high. In Deitchman, Squibb and the other drug companies that had manufactured DES faced multiple litigation on a scale that threatened their very existence. They recognized the force of Dr. Herbst’s studies and believed their only option was to find flaws in the data underlying the studies and use those flaws to diminish the significance of the research results in the eyes of the jurors.

Scientists might question this approach. Their training suggests that the proper way to test the results of a particular experiment or study is to design a similar study to try to reproduce the results of the first study, rather than to scrutinize the raw materials of someone else’s research. However, the scientific goal is not to achieve a particular result; it is the disinterested one of proving or disproving a hypothesis about some fact. Unlike the scientist, the litigant begins with the goal of achieving a particular result: a jury verdict in its favor. The litigant must persuade a jury of lay persons of the truth of his position. Because juries find scientific studies persuasive, trial lawyers defend against them by arguing that they are flawed and unreliable. The lawyers look for potential biases in the researcher’s approach and possible errors in execution or analysis.

Given the divergent goals of trial lawyers and scientists, it is unlikely that most courts will abandon the usual tools of cross-examination and instead adopt the methods scientists use to test the validity of scientific studies. The two courts that have adopted this approach have not yet convinced their colleagues of the correctness of their position. In the Deitchman case, for example, the court of appeals gave short shrift to the district court’s observations that Herbst’s studies had withstood peer review and for that reason his data should be shielded from discovery. The appellate court criticized the lower court’s finding that Squibb’s need for the data was speculative because Herbst’s study had been subjected to the scrutiny of the medical profession for more than a decade and “nothing in the record indicated anything other than that Herbst’s conclusions had been fully corroborated.” The appellate court pointed out that no one had publicly reviewed the actual core data behind Herbst’s studies or identified the basis and evidence upon which he classified a patient’s “exposure” or “non-exposure” to DES. It noted that;

a study of this sort may have a number of different, but inadvertent, biases present …. Squibb, for example, has presented evidence that not all physicians in the medical community report all their clear cell adenocarcinoma cases to the registry. This fact could very well make the statistical basis for Dr. Herbst’s published conclusions inaccurate or incomplete.

In re Snyder,  in part because the court was unwilling “to substitute the adversarial process of the judicial search for truth for the epistemological standards set by other disciplines,” the district court granted a motion to quash one of the subpoenas directed to Richard Snyder, the author of a study on the crash experience of utility vehicles. In the court’s view, “the validity of opinions formed and expressed in the context of disciplines other than the law should be tested by the relevant discipline’s requirements for validity or acceptability.” This decision was not appealed, and whether the court of appeals would have agreed with it remains unknown. In my view, such agreement is unlikely in light of the contrary holding in Deitchman. So long as the rules authorize broad discovery, the federal courts are far more apt to allow litigants to attack scientific studies through the usual methods of cross-examination that depend for their effectiveness upon extensive discovery of the underlying data and the possible biases of the scientist.

It is still too early to tell whether the Supreme Court’s opinion in Daubert v. Merrell Dow Pharmaceuticals, Inc.,  will affect the lower courts’ views on the breadth of cross-examination of expert witnesses and the concomitant scope of pre-trial discovery. Under Daubert, the trial judge has a “gatekeeping responsibility” to ensure that scientific evidence adduced at trial is “not only relevant but reliable.” The judge is to make an initial assessment whether the reasoning or methodology underlying the testimony is scientifically valid and whether it applies to the facts in issue. In doing so, the judge can consider whether the theory or technique can be or has been tested, whether it has been subjected to peer review and publication, what its known or potential rate of error is, and whether it has achieved general acceptance within the relevant scientific community.

Although in Daubert, the Supreme Court placed emphasis upon scientifically accepted methods of testing research results, it did not suggest that such methods would be the sole means of challenging a theory or a study before the jury. Rather, the Court focused on the methods that will get the studies or theories before the jury, where presumably the usual scope of crossexamination will be permitted. It seems unlikely that Daubert’s emphasis on scientifically approved methods of validating scientific evidence will lead to any limitation on the amount of pre-trial discovery litigants will undertake. Indeed, the Court’s explicit delegation of gatekeeping responsibility to the trial judge is likely to spur litigants to seek discovery that will not only help persuade a jury of the inadequacy of a particular study result but will convince the trial court that the result does not meet Daubert’s standards for admissibility.

B. High Stakes Science

Because of the legal system’s heavy reliance on scientific evidence and the difficulty of retreating from this reliance, cases such as Deitchman illustrate that courts will rarely grant expert witnesses all the protection they seek. Deitchman is a warning to researchers that they cannot rely on a total exemption from litigation if their research happens to be critical to the issues in litigation.

Deitchman was a high stakes case in terms of money. It was also a high stakes case in another respect: the risk of serious harm to a significant research study. Not only did the district court and the court of appeals agree that Herbst’s concern for the confidentiality of the registry was well-founded, even Squibb appeared to concede that the loss of confidentiality would affect the registry adversely and that “all society would be poorer … because a unique and vital resource for learning about the incidence, causes, and treatment of adenocarcinoma would be lost.” Despite these undisputed serious concerns about the potential harm, the court of appeals was not persuaded that Herbst and his registry should be exempted from discovery. A major factor in its decision was that the plaintiffs were relying on Herbst’s studies; therefore, according to the court, it was unfair to Squibb and the other drug companies to withhold all the registry data. Instead, the appropriate response was to accommodate the confidentiality concerns with a carefully drawn protective order. The court’s decision demonstrates that researchers can take little comfort from the case if they believe a court will relieve them of the obligation to testify or produce records upon a showing that the success and even the continuation of their research is at risk.

It is noteworthy that neither the district court nor the court of appeals considered denying plaintiffs the use of the studies as a way of leveling the playing field. When, as in Deitchman, the plaintiffs would not have a case without the scientific study at issue, it is almost inconceivable that any court would deny plaintiffs the use of the study as a means of protecting the research, at least in the absence of a showing that allowing inquiry into the particular study raises a reasonable danger of exposing matters subject to the state secrets privilege. In my experience, a researcher should not count on a court accepting any lesser showing as a reason to deny plaintiffs the use of a study on which they have built their entire case, as in Deitchman.

Deitchman is illustrative of the legal system’s ever-increasing reliance on scientific research. The Deitchman plaintiffs would not have known of the existence of a redressable injury had it not been for the scientific research that linked a disease (adenocarcinoma) to a particular product (DES). Such reliance on scientific evidence is becoming increasingly common as trial work becomes more specialized and complex. New prosecutorial tools and evolving theories of criminal defense or civil negligence involve an expanding range of expert knowledge. Without expert witnesses, criminal defense lawyers cannot assert a defense of post-traumatic stress syndrome or prove that the defendant’s DNA does not match the samples found at the crime scene, patent lawyers cannot establish that a competitor’s patent is invalid, and personal injury lawyers cannot explain how an accident occurred, prove the extent of the plaintiff’s injuries, show that a product was manufactured defectively, or even, as in Deitchman, establish a cause of action.

The legal system’s need for research results and expert knowledge can provide a wealth of opportunities for researchers willing to advise parties to litigation or to testify on the subject of their research. However, this need for evidence and advice can be a nightmare for researchers like Herbst who want to stay uninvolved and whose interests lie in protecting their research, maintaining the confidentiality of their research sources, controlling their time and staff resources, and preserving their reputations as disinterested scientists.


The legal system rests on the premise that “‘the public … has a right to every person’s evidence,’ except for those persons protected by a constitutional, common-law or statutory privilege.” Exemptions to this general duty to provide evidence are few and narrowly construed. They extend only to those relationships in which society’s interest in protecting the secrecy of the communication is greater than the “normally predominant principle of utilizing all rational means for ascertaining truth.” Thus, confidential communications between attorney and client, husband and wife, and in some cases, psychiatrist or psychologist and patient are protected because of the belief that confidentiality is essential to the relationship, the community values the relationship, and the injury to the relationship that would result from disclosure of confidential communications is greater than the benefit for the resolution of litigation that would result from disclosure. The Fifth Amendment to the United States Constitution gives individuals the right to decline to testify if their testimony would tend to subject them to prosecution for their own criminal behavior, but no privilege to refuse to testify out of a desire to protect a friend or family member from prosecution or out of fear for their own safety or that of their families. Persons who cannot claim any of these privileges are generally required to testify, whether they want to or not.

In criminal cases, the reason for compelling testimony is obvious. The giving of testimony when called upon is the price that members of society pay for the maintenance of a safe society. Anyone with knowledge of facts relevant to a criminal prosecution has an obligation to testify about those facts, however inconvenient, uncomfortable, embarrassing, or even dangerous it may be to do so.

It is not so obvious why persons can be compelled to testify in private civil litigation. One commentator suggests two theories. The first is that members of society have a reciprocal right to compel others to testify. Under this theory, citizens receive two benefits from the principle of compelled testimony. First, if they testify on behalf of another in one suit, they may reap a reciprocal benefit if they become a party in a different suit. Second, by testifying in another person’s suit, they are helping to protect their own rights as well as the other person’s. The witness’s time is “claimed by the public as a tax paid by him to that system of laws which protects his rights as well as others.”

The second theory is that all members of society have a normative obligation to participate in the judicial system and a concomitant obligation to testify when called. Each citizen has an interest in living in a society that is just and in which law and order prevail. Under this theory, the citizen’s duty to testify is an obligation that runs primarily to the public rather than to the parties. The duty of an expert to supply evidence is the same as that of any other witness. Compelling testimony increases the amount and quality of information that the courts have for making decisions and therefore increases the likelihood that judicial decisions will be accurate.

Under either the normative or the reciprocal rights theory, the idea of compelling the testimony of occurrence witnesses seems objectively fair because of the randomness of the obligation. Nevertheless, a number of commentators argue that neither of these theories justifies compelling the testimony of an expert who does not wish to be involved in litigation. These commentators base their opinions on major distinctions they draw between occurrence and expert witnesses. Although both types of witnesses may have to spend hours or weeks in deposition, be forced to provide information against their will, be cross-examined, and, because of the broad discovery rules, be required to provide more information than either party needs, ordinary occurrence witnesses do not purposefully acquire the knowledge about which they will testify. Rather, they happen to be present when the event at issue occurs. By contrast, the expert witness invests time and money in acquiring the knowledge that the parties to the lawsuit seek. The information the expert is asked to give is the product of professional endeavors, not information gained by happenstance.

Some commentators argue that “compelling disclosure of the unretained expert may have the paradoxical effect of reducing, rather than increasing, the production of information useful to the resolution of lawsuits.” If litigants can compel researchers to disclose their research against their will or before they are ready to do so, researchers may decide that the costs of becoming an expert in a particular area are far greater than the benefits. In addition, researchers may be deterred from studying areas such as public safety and health in which research might be most beneficial to society and, therefore more needed in court.

It is not clear that these fears of reduced research or production of information are borne out in fact. Although some individual researchers may be deterred from entering a particular field because of its controversial and public nature, others may be attracted to it for the same reason. In the 1980s, when large numbers of lawsuits were brought against Merrell Dow Pharmaceuticals alleging that its prescription anti-nausea drug had caused birth defects in children of mothers who had ingested the drugs, research in the drug and its effects expanded. According to one scholar, the litigation did not deter researchers from studying Bendectin; to the contrary, it had the opposite effect.

Legal needs gave shape and direction to the epidemiological study of teratogenic effects: “The volume and sophistication of studies specifically on Bendectin was in large part the result of the litigation.” Researchers found that the interest in Bendectin made it easier to get studies and articles published and to get funding. Moreover, the government offered research underwriting through the Federal Drug Administration and the parties to the litigation funded research of their own.

Professor John Wigmore rejects the idea of giving expert witnesses any special protections. In his authoritative text on evidence, he sets out five reasons for not excusing experts from the public duty of testifying and for treating them like any other witness:

  1. the expert is giving testimony, not providing a professional service;
  2. the hardship on the expert in losing income-producing time is no greater than in the case of any other witness;
  3. the expert becomes desirable as a witness only by accident, not because of the litigation itself;
  4. it is impractical to attempt to distinguish between kinds of experts, or between what is opinion and what is fact;
  5. and no one will refrain from becoming an expert because of fear of being called to testify.

As Wigmore notes, it is difficult to distinguish between different kinds of experts or between opinions and facts. Commentators have suggested that such lines be drawn and have criticized the courts for not doing so, but they have not suggested a principled way to differentiate between experts who should be free of any obligation to testify or produce documents against their will and experts who should be required to obey a discovery subpoena.


A. Federal Rules Governing Discovery

The federal courts have never recognized a constitutional or common law privilege equivalent to the Fifth Amendment or the attorney-client privilege that would give a researcher an automatic exemption from participating in litigation. In Kaufman v. Edelstein, an antitrust action against IBM in which the court allowed the government to compel testimony from two individuals with experience in the computer industry, the court explained:

We can find no justification for a federal rule that would wholly exempt experts from placing before a tribunal factual knowledge relating to the case at hand, opinions already formulated, or even, in the rare case where a party may seek this and the witness feels able to answer, a freshly formed opinion, simply because they have become expert in a particular calling.

The rules that govern the discovery of factual information in civil cases in the federal courts support the general principle that all have a duty to testify. Rule 26(b)(1) of the Federal Rules of Civil Procedure provides that “parties may obtain discovery regarding any matter … which is related to the subject matter involved in the pending action … . The information sought need not be admissible at the trial if the information sought appears reasonably calculated to lead to the discovery of admissible evidence.” Under this rule, parties may compel testimony from persons who are strangers to the action, including expert witnesses voluntarily retained by a party and expert witnesses who are neither parties to the action nor retained by a party. However, Rule 45(c)(3)(B)(ii), as amended in 1991, authorizes courts to quash or modify a subpoena that “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.” In cases in which the requester shows “a substantial need for the testimony or material that cannot be otherwise met without undue hardship,” the court may enforce the subpoena upon receiving assurances that the person subpoenaed will be reasonably compensated, and may order that certain specified conditions attend the required appearance or production. The advisory committee explains that this provision is intended to protect the intellectual property of nonparty witnesses and notes its approval of the accommodation of competing interests exemplified in Wright v. Jeep Corporation, in which the court held that “the solution is not to cover-up [research data] because disclosure is too burdensome but to use the tools available to lessen the burden and to permit the information to become available.” In Wright, the court suggested that these tools might include a reasonable fee for testifying that included a professional fee as well as the cost of producing the documents, remuneration for inconvenience and a charge for a portion of the expenses of the original research. The court added that in the case before it, it would consider any proposals submitted by the researcher or the Jeep Corporation to impose additional conditions that would lessen the burden of compliance.

It is worth noting that Rule 45(c)(3)(B)(ii) excepts disclosure of information “not describing specific events or occurrences in dispute,” to distinguish it from testimony that could be characterized as factual in nature. “Factual” testimony is treated separately even when such factual testimony includes facts the expert acquired in conducting a study or in preparing expert testimony in another proceeding. It is probable that courts will read this exception as applying to information acquired by researchers who study the actual business or activity of persons who become embroiled in litigation. Although the information is gathered for the purpose of drawing general conclusions about the activity or the business, it would not receive special treatment under Rule 45(c)(3)(B) to the extent it consists of observations about specific events or occurrences in dispute. Commentator David Siegel suggests that the committee was more concerned with protecting experts’ opinion testimony than their observations of facts, both because factual observations are not thought of generally as intellectual property and because the expert may be the only person available by the time of trial who can testify about the facts.

B. Academic Freedom—A Source of a Privilege?

In a number of cases, courts have discussed the possibility of a constitutional protection against the compelled disclosure, but no court has held that such a protection exists. In Dow Chemical Co. v. Allen,  the Court of Appeals for the Seventh Circuit suggested that the First Amendment might offer protection against a “judicially authorized intrusion” into the scholarly research activity of university researchers. The court was addressing a researcher’s challenge to a subpoena issued by the Environmental Protection Agency at the request of the Dow Chemical Company. The subpoena sought information about ongoing studies at the University of Wisconsin involving dietary ingestion by rhesus monkeys of the chemical 2,3,7,8-tetrachlorodibenzo-p-dioxin.81 Dow wanted the information for use in cancellation hearings before the agency on certain uses of two herbicides manufactured by Dow. The First Amendment issue was raised in the district court but not reached in granting the researchers’ motion to quash. The court of appeals chose to discuss the issue, characterizing Dow’s efforts to subpoena ongoing research as a threat of “substantial intrusion into the enterprise of university research … capable of chilling the exercise of academic freedom,” which would “inevitably tend to check the ardor and fearlessness of scholars … .” The court did not go so far as to hold that the researchers were entitled to an evidentiary privilege, but concluded that where a researcher’s interest in academic freedom is involved, “the interests of the government must be strong and the extent of intrusion carefully limited.” The court defined academic freedom as “the right of the individual faculty member to teach, carry on research, and publish without interference from the government, the community, the university administration, or his fellow faculty members.” It added that “we think it clear that whatever constitutional protection is afforded by the First Amendment extends as readily to the scholar in the laboratory as to the teacher in the classroom.” In Deitchman v. E.R. Squibb & Sons, Inc., the same court acknowledged its previous comments in the Dow Chemical case to the effect that academic freedom might be implicated in the premature release of Herbst’s registry data or in unguarded disclosure of such data, but held that concerns of academic privilege must yield “if to enforce them would produce a miscarriage of justice.

In re Grand Jury Subpoena,  the Court of Appeals for the Second Circuit was less receptive to the idea of a First Amendment privilege. The court reversed a district court decision quashing a grand jury subpoena for a graduate student’s journal kept in preparation for his doctoral dissertation on the sociology of the American restaurant. The student had testified before the grand jury but resisted the grand jury’s request for his journal. The district court found a limited scholar’s privilege analogous to a news reporter’s privilege and quashed the subpoena on the ground that the grand jury had not shown a compelling need for the journal that overrode the scholar’s privilege. The court of appeals found the facts pertaining to the alleged privilege too sparse to support a finding of privilege in the absence of any testimony from scholars about the nature of the researcher’s work or its “role in the scholarly literature of sociology” or about methodology and the need for confidentiality. The court found that:

there is thus no evidence of a considered research plan, conceived in light of scholarly requirements or standards, contemplating assurances of confidentiality for certain parts of the inquiry. Finally, and even more astonishingly, the researcher has not established that all of the materials he seeks to keep from the grand jury in fact are covered by the privilege he asserts.

In Richards of Rockford, Inc. v. Pacific Gas & Elec. Co., 96 the District Court for the Northern District of California considered the confidentiality of research sources as a basis for denying a motion to compel disclosure. A nonparty university professor was subpoenaed to produce documents concerning confidential interviews of the defendant’s employees that were undertaken as part of a research project. The court refrained from deciding whether a constitutional privilege existed that would protect researchers from disclosing the confidentiality of their sources, although it noted “that society has a profound interest in the research of its scholars, work which has the unique potential to facilitate change through knowledge.” The court compared society’s interest in scholarly research with the public’s interest in news gathering, concluding that cases addressing discovery requests for news reporters’ sources provided a useful framework for balancing the interests of the researcher and the party seeking discovery.

In still another case, Kennedy v. Connecticut Department of Public Safety,  a federal magistrate judge rejected a motion to quash a subpoena calling upon a researcher to disclose all material relating to her study of the integration of women into the Connecticut State Police. The court was unpersuaded by the researcher’s assertion that production would be burdensome and that plaintiffs had alternate sources of information. The court allowed her to redact the names of persons to whom she had promised confidentiality, but required that she do the redactions at her own expense.

The decided cases give little evidence that a researcher’s privilege is emerging or that if one did, it would offer researchers any more protection than they have now. At most the privilege would be conditional and would still require balancing the interests of the party seeking discovery against the researcher’s asserted need to withhold the information. Even in those cases in which the courts have considered First Amendment concerns relating to academic freedom comparable to a reporter’s privilege, they have been unwilling to find that a privilege outweighed disclosure when the information is important to the essential elements of the case, has significant probative value, and cannot be obtained through another source or another expert witness. In Richards of Rockford, for example, where the court applied the balancing test used in cases involving the news reporter’s privilege, the factors the court considered were “the nature of the proceeding, whether the deponent is a party, whether the information sought is available from other sources, and whether the information sought goes to the heart of the claim.” The court held that the societal interest in protecting the confidential relationships between academic researchers and their sources outweighed the interests of this litigant and the public in obtaining the research data. The court reached this conclusion because the factual issues of the litigation could be proved independently of the research data, making the data largely supplementary.

In Dow Chemical Co. v. Allen,  several factors tipped the balance in favor of protecting university research efforts from intrusion. First, probative evidence from the studies sought would not be available to either side for months or even years. Second, Dow would not be confronted by information from the studies in an adversary proceeding. Finally, the data from the studies at issue had no bearing on the studies that would be used against Dow. In Deitchman v. E.R. Squibb & Sons, Inc.,  however, a different panel of the same court of appeals that decided Dow concluded that interference with scientific research was justified when the studies at issue would be used in numerous trials against defendants, the studies formed the very basis for the bringing of the plaintiffs’ cases, the studies were unique and no other expert had access to the data that had been compiled, and the defendants would be prejudiced if they could not cross-examine the plaintiffs’ experts on the data underlying the studies.

In reality, amended Rule 45(c)(3)(B) provides researchers with as much protection as a conditional privilege would. The rule rests on the assumption that researchers’ opinions should not be the subject of a subpoena. The subpoena will be enforced only if the requester can show a substantial need for the information and no alternative methods of obtaining it. Even then, the court has discretion to modify the subpoena and to impose conditions upon production.

C. Property Rights in Research or Specialized Knowledge

Claims to proprietary interests in the fruits of research efforts have failed to persuade courts to exempt researchers from testifying. However, courts have been willing to recognize such claims to the extent of requiring parties who subpoena such research to compensate the researchers for any time they expend in collecting and organizing the data and occasionally to pay a portion of the costs of the study. No court has addressed the damage resulting from the researcher’s loss of the exclusive opportunity to mine the data for scientific insights that might bring scholarly recognition, tenure, or other academic advancement. It is an interesting question whether courts will recognize this lost opportunity cost and, if so, how they will quantify its value.

Most courts reject the idea that experts have property rights in their opinions that exempt their opinions from disclosure. One court has ruled that “to clothe all such expert testimony with privilege solely on the basis that the expert ‘owns’ his knowledge free of any testimonial easement would be to seal off too much evidence important to the just determination of disputes.” An important aspect of the amended Rule 45(c)(3)(B)(ii) is its recognition that certain knowledge does constitute intellectual property. This does not mean that the owner of such property is exempt from ever having to testify or produce the information or opinion, but it does mean that the owner is entitled to have the person seeking the subpoena show both substantial need for the information or opinion and the lack of any reasonable alternative for obtaining it. In addition, the new rule codifies what had become the standard judicial practice of requiring the person seeking the information to pay the reasonable costs of obtaining it.

The amended rule is a practical compromise between important competing interests. It recognizes the economic value of expert opinions based upon research and study. In shifting the burden to the requester to show substantial need, the drafters have told the courts that they are to give special consideration to requests for expert opinion. This may be helpful for researchers, but it is no panacea. For the court, the amendment adjusts the analysis it must apply, but does not eliminate the difficulty of trying to accommodate the conflicting concerns of scientists and litigants.


A. The Judicial Task

In deciding an unretained expert’s motion to quash a subpoena, the initial step must be to determine the purpose for which the expert is being asked to testify or to produce documents and other materials. Federal Rule of Civil Procedure 45(c)(3)(b) applies only to the extent that the expert is being asked about opinions or information resulting from the expert’s study and not about knowledge of specific events in dispute. An expert may move to quash a subpoena asking for factual testimony or information regarding specific events, but must do so under subsection (3)(A) of Rule 45(c), which applies to all subpoenaed persons, not just experts, and requires a showing of undue burden. A significant difference between the two provisions lies in the placement of the burden of proof. The person proceeding under subsection (3)(A) bears the burden of persuading the court that compliance with the unmodified subpoena will be unduly burdensome. At the outset, the movant must set out the manner and extent of the burden and the injurious consequences of compliance. By contrast, movants proceeding under subsection (3)(B) need show only that they are unretained and are being asked to testify or produce materials that they have prepared independently of any party’s request and that do not describe specific events in issue.

Once the initial showing is made by the movant proceeding under (3)(B), it is up to the requester to show the substantial need for the information and why it is necessary to obtain it from the movant. Factors bearing on this question include the following:

  1. the needs of the case;
  2. the possibility that the witness is a unique expert or whether the discovery could be obtained from another source;
  3. the degree to which the discovery sought is necessary to enable the parties to prepare an adequate case or defense;
  4. the amount in controversy;
  5. the limitations on the parties’ resources;
  6. the importance of the issues in the litigation;
  7. and whether the discovery is cumulative or duplicative.

Assuming that the requester can make a showing of substantial need, the court will have to balance that need against the burden imposed upon the unretained expert.

At this point, it is critical to determine the exact nature of the researcher’s concerns because they bear upon both the balancing test and the aptness of protective measures that might be taken to accommodate those concerns. In striking this balance, the judge may confront a number of questions. For example, is the researcher facing large numbers of subpoenas, of which this is just one? Does the movant object to the excessive burden of this one subpoena because of the time and resources that will have to be devoted to compliance? Is the researcher’s primary concern for the confidentiality of the subjects of the study? Will release of the subpoenaed data endanger an entire study because of the preliminary, untested state of the data? Would ongoing research be interrupted if working documents had to be turned over or the principal researcher diverted from the work? Would the release of research studies and results deprive them of significant value to the researcher, as might be the case if the researcher has not yet had a chance to analyze the data, if the information could lead to a patentable invention or commercial product, or if disclosure would reveal unique testing or production methods the researcher does not want to reveal? Does the researcher’s concern derive from the scope of the subpoena, such as its demand for personal notes, false leads, diary entries, or correspondence between researchers?

Assuming that the researcher can show that production of information will entail significant and unusual costs and inconvenience, the court must balance the need for the information against the hardships of production, keeping in mind which person has the burden of persuasion. In the end, the balancing usually comes down to this: If a party seeking discovery shows that the discovery consists of unique information of significant probative value at the heart of the case, this showing will outweigh almost any claim of burdensomeness or breach of confidentiality. This is particularly true when one side will be relying on some aspect of the researcher’s information, as were the plaintiffs in Deitchman.  If such a showing is made by the party seeking discovery, the court must then consider the need for a protective order to ameliorate the burden upon the researcher.

The scope and format of a protective order will depend in large part upon the exact nature of the burden on the movant as well as the specific needs of the request. For researchers facing large numbers of subpoenas that threaten their ability to work on their scientific endeavors (or enjoy their retirement), the court may consider an order that will require the requester to preserve the materials produced and the transcript of the deposition and make them available to other litigants. Litigants in subsequent trials would be required to show that the existing materials are insufficient before being permitted to subpoena additional materials or testimony from the researcher. For the researcher who objects to the burden of compliance with one subpoena because of the time and resources such compliance would require, the court will have to inquire into the resources available to the researcher, the amount of time that will be required to comply with the subpoena and the possibility that temporary workers can be used to collect and organize the subpoenaed information. The court must determine whether the subpoena should be modified or whether the burden can be offset by financial reimbursement for the work of production. For the researcher who fears interruption of an ongoing project, it may be necessary to fashion an order that will ensure that neither the work of gathering and organizing data nor the temporary loss of the data or other materials prevents the researcher from continuing the project.

If the researcher’s primary objection is the confidentiality of the requested information, a number of steps may be taken to meet that objection. If confidentiality of research subjects is at issue, the court should determine the nature of the promise made to the subjects (not all such promises require total confidentiality), as well as any potential harm to the study if confidentiality is not maintained. If the court is convinced that confidentiality must be maintained, it can order that the subjects’ names be redacted. Depending on the situation, additional redaction of identifying information may be ordered. Other protective options include strictly limiting the dissemination of subpoenaed material, turning it over to an independent third party for review, or restricting its use by certain named experts under a protective order governing future use.

If the researcher is concerned that the data are untested and still preliminary, the court must consider the extent to which the requester can legitimately say that it has a substantial need for such information. It may be, as in Dow Chemical,  that no party intends to use the preliminary data in litigation, making the requester’s need for the data questionable. If the researcher’s concern about the preliminary state of the data is actually a desire that she have the first opportunity to analyze the data, the court might consider the possibility that she be given a short period of exclusive access to the data, particularly because reimbursement for the costs of collecting and organizing the data may fall far short of the economic value to the researcher of analyzing the data for the first time. If such a compromise is impractical because of the immediacy of the requester’s need for the data or the length of time the researcher would need to have exclusive access, some form of protective order might preserve the value of the data. If other economic concerns are at stake, such as plans for patenting the results of studies or experiments, protective orders can be fashioned to take these concerns into account.

If the researcher’s concern is directed at having to turn over personal writings such as his notes or diary entries, the court can modify the subpoena to eliminate these requests. Of course, this assumes the requester has not shown that the need for these particular materials is so substantial as to outweigh the burden upon the researcher.

The availability of protective orders and the option of modifying subpoenas do not imply that the basic question of compelling disclosure is an easy one. Trying to balance the needs of litigants against the needs of researchers is made particularly difficult by the need to accommodate the often incompatible differences between the scientific world and the legal system. Rarely is the question so straightforward as it was in Karp v. Cooley,  where the only question was whether a party could subpoena Dr. Michael DeBakey, a noted surgeon, to testify in a medical malpractice suit about the work of the defendant, another noted heart surgeon. DeBakey was able to show that he lacked knowledge of the specific artificial heart used by the defendant and could not supply any information that would link the death of plaintiff’s decedent to his alleged lack of informed consent. This showing made it easier for the court to rule that the subpoena was unnecessary. Far more frequent is the case where the testimony of the unretained expert will be neither so clearly irrelevant nor so obviously available from other experts.

Although DeBakey was subpoenaed only for his testimony and not for production of documents or other materials, his motion to quash the subpoena was analyzed in the same manner as a subpoena directed primarily at the production of research data and other information. This was an appropriate way to address the issue. The same factors apply whether the requester wants the expert to come to court to give an opinion or wants the expert to produce information. The court determines whether the expert resisting the subpoena is proceeding under subsection (3)(A) or (3)(B), whether he or she has made the requisite showing under the applicable subsection, and then determines the exact nature of the burden and weighs the showing of substantial need against that of burdensomeness. It may be that a single deposition or a one-time appearance in court will not pose a threat of any kind to a researcher’s work. On the other hand, it may be that such a limited request, unaccompanied by any demand for data, could be satisfied by any number of possible witnesses.

B. Recommendations for Researchers

The previous discussion should alert researchers seeking to quash an onerous subpoena to the need to make the particular facts of their situation known to the judge. Researchers cannot assume that the judge will know anything about the milieu in which researchers work, about their resources or lack thereof, about what disruption of a particular study might mean, or about alternative sources of the same information. Researchers must educate the judge about these matters if they want them taken into consideration.

The researcher should keep in mind that trial judges are extremely busy. They do not handle cases one at a time, as one might custom-build a piece of fine furniture, but operate more like a short-handed fire department, faced with fires breaking out continually and inadequate resources to respond. Busy trial judges do not welcome discovery disputes, not only because of the time that such disputes require for hearing and resolution, but also because of the improbability of getting an accurate sense of the actual nature of the dispute. It is inherent in the nature of litigation that judges can know only a small portion of what is really going on in a case, even in those cases that go to trial. The situation is intensified in discovery disputes, which arise before the judge has had any chance to become familiar with the facts of the case. The judge may have an instinctive feeling that one side is trying to use the discovery process either to make the litigation so expensive it will force the other side to settle or to undermine the value of certain information. Only rarely, however, does the judge have a means of testing the validity of those instinctive feelings.

Researchers should not resort simply to telling the judge that a discovery request is “burdensome.” Judges are inured to such claims because lawyers make them frequently. The researcher should tell the judge exactly why the request is burdensome, including what the request encompasses, what size office or laboratory the researcher works in, what staff resources are available, how much time it will take to comply with the request, how the compliance will be managed, what other obligations the researcher has and the effect on those obligations if the subpoena is enforced, whether the work can be performed by untrained persons or only by the research staff, whether the materials sought could or could not be made available for review at the researcher’s work place, whether unlimited dissemination will destroy the value of the research, whether production of the materials will interrupt an ongoing project, and what financial costs the researcher is likely to incur in complying with the subpoena.

It is unlikely that the judge will grant any motion to quash in its entirety unless the researcher establishes the existence of realistic, alternative sources for the subpoenaed information, shows that the information subpoenaed does not have the probative value claimed by the requester, or proffers some other persuasive reason for not complying with the subpoena in any form. It is far more likely that the judge will direct the requester and the researcher to negotiate the terms of a modified subpoena that meets the concerns of the researcher and the needs of the requester. As the court noted in Deitchman v. E.R. Squibb & Sons, Inc., an overly broad discovery subpoena is common and “means about as much as the asking price for a rug in an Oriental bazaar. It is normally just a means of opening discussion between discoverer and discoveree.” Rule 45(c)(3)(B) gives courts the additional option of imposing specified conditions upon the researcher’s production or testimony. In addition, Rule 26(c) provides for the entry of an order protecting the expert from annoyance, embarrassment, oppression, or undue burden or expense.Researchers need to understand the strategic importance of the proceedings in the trial court. In all likelihood, the issue will be resolved at that level with no opportunity for appeal. It is rare for such a dispute to be reviewed by an appellate court. The general rule is that only “final orders” can be appealed; orders dealing with subpoenas and other aspects of discovery are “interlocutory” in nature, that is, they are nonfinal orders that do not dispose of the entire case and are therefore not appealable. The general rule foreclosing appeal of discovery matters has a few exceptions “in the limited class of cases where denial of immediate review would render impossible any review whatsoever of an individual’s claims.” Thus, a nonparty witness whose motion to quash a subpoena is denied can obtain appellate review of the denial by refusing to honor the subpoena and being found in contempt.

A second exception to the general rule of non-appealability has been created in proceedings to quash or enforce a subpoena that are brought in a federal district other than the one in which the lawsuit is pending. In those instances, many courts treat a ruling granting a motion to quash as a final order that may be appealed; this is the way in which many of the cases discussed in this article reached the courts of appeals. The situation is different when the motion to quash is denied. Then the resister’s only means of obtaining appellate review is to subject him- or herself to a finding of contempt.

Even researchers who know that appellate review is available should be concerned primarily about what happens in the district court. Not only are appeals time-consuming and expensive, they also cannot make up for botched presentations in the district court. Courts of appeal are constrained from considering facts that were not made part of the record in the lower court, and they will not allow a researcher to put in new facts on appeal that he or she may have omitted from the presentation in the lower court.

Recommendations for Litigants

Parties seeking production of data and other materials from unretained experts face a slightly more difficult task under the amended Rule 45, because they now bear the burden of persuasion unless they can show they are seeking only descriptions of relevant factual events or occurrences. As with researchers, it is critical for requesters to be precise in making the required showing. Requesters should focus on the same things as researchers, only from the opposite perspective, because they are trying to depreciate the significance of the burden the researcher is asserting.

Litigants should be able to explain why the materials they are seeking are of significant probative value on matters going to the heart of their case. They should be able to show why the materials cannot be obtained from any other source and why the case is of sufficient importance to justify intrusion into an unretained expert’s research. Litigants who can show that their opponents intend to rely on the results of certain studies will want to make that known to the court at the outset. If they cannot make this showing, they will face a much more difficult task of showing the importance of the information. Finally, litigants should consider whether the demands of the Daubert case bear on their need for expert information and be prepared to advise the court on this point. It may well be that information developed by unretained experts does not meet the standards for admissibility set out in Daubert.  If this is so, the litigant may be able to bar its use at trial altogether.


In summary, it is unlikely that the courts will recognize an absolute privilege that will exempt researchers from testifying or producing data and documentation. Considerations of fairness to all parties militate against an evidentiary rule that would foreclose a party from access to critical information bearing on its case. With the relatively recent amendments to Rule 45, courts are apt to be receptive to drafting protective orders that will relieve researchers’ concerns about confidentiality and to conditioning enforcement of subpoenas upon the payment of fair and adequate compensation to the subpoenaed researcher. However, researchers who are subpoenaed should not anticipate that courts are familiar with the needs, operations, or resources of the research world. It is up to researchers and their attorneys to educate the courts about these matters and to mount a strong offense if they want to ameliorate the potentially harsh effects of the forced disclosure of research data. Litigants must be prepared to explain thoroughly their need for obtaining scientific materials from unretained experts. The complex rules governing judicially compelled disclosure and the difficulties inherent in weighing requests for information require that those seeking or resisting disclosure be detailed in their requests or responses, forthcoming in explaining their concerns, and willing to compromise.


Click to download the full paper.

More DES DiEthylStilbestrol Resources

Have your say! Share your views