The University of Notre Dame Du Lac appeals from an order granting enforcement of an administrative subpoena duces tecum issued by the Equal Employment Opportunity Commission ("EEOC") requiring the University to produce the personnel files of certain faculty members in the University's Department of Economics. We reverse and remand.

On May 21, 1980, Oscar T. Brookins filed a charge with the EEOC alleging that the University had unlawfully discriminated against him on the basis of race in denying him a tenured position in the University's Economics Department. In his EEOC charge, Brookins alleged:

"I began employment with [the University of Notre Dame] in August, 1974 as assistant professor of Economics. I received a one-year terminal contract on May 13, 1980 which denied me tenure as a permanent professor.

"I feel that I have been discriminated against because of my face [sic], black, in that:

(1) no black professor has ever received tenure from the Economics Department.

(2) the University refused to state to me in writing why I was being denied tenure.

(3) my publications are of greater or equal quality and of greater or equal number than most tenured professors within the Department of Economics.

(4) the Chairman of the Department of Economics has made the statement to my wife that he doesn't believe that races should be mixed. My wife is white.

(5) Chairman and Dean have refused to identify external and internal reviewers of my performance. Further, they have refused to allow me to read these reviewer's comments.

(6) those persons and committees reviewing me for tenure have used student's teacher-course evaluations as a measure of my competency to teach.

(7) the Department of Economics has refused to provide copies of its member's vitaes to me. According to the University's rules, I have a right to see the vitaes."In investigating Brookins' charge of unlawful racial discrimination, the EEOC sent the University an eight-part questionnaire demanding inter alia the following: (1) "Describe in detail your tenure practices in the University of Notre Dame"; (2) "[I]dentification by name and race, of persons responsible for reviewing the qualifications of candidates for tenured positions"; (3) "Identify by name, race, and job title the individual(s) making up the tenure committees for the economics department from January 1, 1980 to June 1, 1980"; (4) Construct a chart detailing the name, race and employment history of each member of the University Economics Department; (5) "Furnish copies of the personnel records of the Charging Party [Brookins]."

The University complied with each of the requests for information and documents with the exception of the EEOC's request that the University furnish copies of the personnel records of the charging party, Brookins. In a February 6, 1981 letter to the EEOC, the University's assistant general counsel offered to allow an EEOC investigator to visit the University to personally review Brookins' personnel files, but refused to permit the EEOC to make copies of the file on the grounds that the file was "so voluminous, and because so much of it is of such a confidential nature ...."

On February 19, 1981, the EEOC demanded from the University not only the personnel files of Brookins but also the personnel files of all teaching personnel in the Economics Department. In response to the EEOC's latest demand for the personnel files of the Economics Department, the University offered to produce the files, subject to the EEOC signing an agreement requiring that they maintain the confidentiality of the information contained in the personnel files. The EEOC refused to sign the nondisclosure agreement, and then the EEOC issued an administrative subpoena duces tecum requiring the University to provide "copies of the complete personnel records of [the] charging party Oscar T. Brookins, and all other teaching personnel in the Economics Department for the period January 1, 1980 to the present." After being served with this subpoena duces tecum, the University filed with the EEOC a petition to revoke or modify the subpoena. The district director of the Indianapolis district office of the EEOC denied the University's petition to revoke or modify the subpoena and ordered the University to produce the documents. The University appealed to the full commission in Washington, which denied the University's appeal but modified the language of the subpoena to require production of the following evidence:

"Copies of the complete personnel records of charging party, Oscar T. Brookins, and all other teaching personnel in the Economics Department for the period January 1, 1980 to the present are requested regardless of whether such teaching personnel were fired by respondent before or after January 1, 1980, or left respondent's employ after January 1, 1980."

The University did not comply with the subpoena and, on March 16, 1982, the EEOC filed an application in district court for an order to show cause why the subpoena should not be enforced.

The University objected to the EEOC subpoena on three grounds. First, the University argued that the personnel files in question contained peer review evaluations which were made with the assurance and expectation that the evaluations would remain confidential, and therefore the peer review evaluations were protected from disclosure by a qualified academic privilege.1 The University contended that pursuant to the qualified academic privilege, it should be permitted to delete the names and any and all identifying information of the academicians participating in the peer review process before giving the files to the EEOC. Second, the University objected to the breadth of the EEOC subpoena which demanded production of the personnel files of all faculty members having taught in the Economics Department since January 1, 1980. The University maintained that the files of only those faculty members eligible for tenure during the same period Brookins was eligible for tenure were relevant to Brookins' charge of unlawful discrimination. The University also argued that the EEOC should be required to execute a nondisclosure agreement as a condition precedent to the University's release of any of the faculty personnel files in question.

The district court, 551 F. Supp. 737, rejected the University's contention that it should be allowed to delete the names and identities of academicians participating in the tenure peer review process, finding that the EEOC's need for the information was "substantial," without adequately explaining its reasoning, and that the University's argument against disclosure of the identities was "premature." The court reasoned that under 42 U.S.C. § 2000e-8(e), the EEOC was prohibited from disclosing any information in the files to anyone, including the charging party, Brookins, until such time as a lawsuit was actually filed, and therefore, "any damage that might result from disclosure would not occur at this stage of the EEOC investigation but rather at its conclusion." The district court agreed with the University's argument that the subpoena was overly broad, and limited the subpoena's scope to the personnel files of those Economics Department faculty members eligible for tenure beginning in August 1974 (when the charging party Brookins began working in the Notre Dame Economics Department) until the date of the court's order (August 5, 1982). By limiting the scope of the subpoena the EEOC was precluded from obtaining the files of those Economics Department faculty members granted tenure prior to Brookins eligibility for tenure.

Finally, the district court rejected the University's contention that the faculty files, in whatever form they were produced, should be subject to a protective order to prevent disclosure of the information in the files by the EEOC. The University's request would require the EEOC to obtain a nondisclosure agreement from the charging party prior to releasing it to them. The court held that the University could not condition its release of the requested information upon the EEOC's signing of a nondisclosure agreement because to do so would be contrary to "the plain language and spirit of Title VII and EEOC v. Associated Dry Goods Corp., [449 U.S. 590, 101 S. Ct. 817, 66 L. Ed. 2d 762 (1981) ]."

The University appeals from the district court's finding that (1) the University could not, prior to producing the files, delete from the files the names and identifying features of academicians participating in the tenure peer review process; and (2) the University could not condition the release of the files on the execution of a nondisclosure agreement between the EEOC and the University. On this appeal, neither party challenges the district court's decision to limit the scope of the EEOC subpoena to only those faculty members eligible for tenure between August of 1974 and August of 1982.

Rule 501 of the Federal Rules of Evidence authorizes federal courts to recognize new evidentiary privileges in certain circumstances. Specifically, Rule 501 provides:

"Except as otherwise required by the Constitution of the United States or provided by Act of Congress or in rules prescribed by the Supreme Court pursuant to statutory authority, the privilege of a witness, person, government, state, or political subdivision thereof shall be governed by the principles of the common law as they may be interpreted by the courts of the United States in the light of reason and experience."

The Supreme Court has stated that the purpose of Rule 501 is to "provide the courts with the flexibility to develop rules of privilege on a case-by-case basis ... and to leave the door open to change." Trammel v. United States, 445 U.S. 40, 47, 100 S. Ct. 906, 910, 63 L. Ed. 2d 186 (1980). See also 2 J. Weinstein & M. Berger, Weinstein's Evidence p 501 (1980).

A determination on whether to recognize an evidentiary privilege requires a balancing of competing policies. As this court stated in Memorial Hospital for McHenry County v. Shadur, 664 F.2d 1058 (7th Cir.1981):

"[I]n deciding whether the privilege asserted should be recognized, it is important to take into account the particular factual circumstances of the case in which the issue arises. The court should 'weigh the need for truth against the importance of the relationship or policies sought to be furthered by the privilege, and the likelihood that recognition of the privilege will in fact protect that relationship in the factual setting of the case.' "

Id. at 1061-62 ( quoting Ryan v. Commissioner of Internal Revenue, 568 F.2d 531, 543 (7th Cir.1977), cert. denied, 439 U.S. 820, 99 S. Ct. 84, 58 L. Ed. 2d 111 (1978)). See also Branzburg v. Hayes, 408 U.S. 665, 710, 92 S. Ct. 2646, 2671, 33 L. Ed. 2d 626 (1972) (Powell, J., concurring).

Before turning to our discussion of the appropriate balance to be arrived at in this case it is important to emphasize that the University is seeking only to assert a qualified privilege, rather than an absolute privilege. The University does not argue that the personnel files of the Economics Department faculty members who were involved in the relevant tenure decisions should be shielded in their entirety from disclosure. Rather, the University merely asserts that it should be allowed to delete from the files the names and any and all identifying features of the academicians who participated in the respective tenure peer reviews.

We in America have an interest of the highest magnitude in fostering academic excellence. Freedom of thought and expression is critical if this excellence is to be achieved in our educational system. In his plurality opinion in Regents of the University of California v. Bakke, Justice Powell stated:

"Academic freedom, though not a specifically enumerated constitutional right, long has been viewed as a special concern of the First Amendment. The freedom of a university to make its own judgments as to education includes the selection of its student body. Mr. Justice Frankfurter summarized the 'four essential freedoms' that constitute academic freedom:

'It is the business of a university to provide that atmosphere which is most conducive to speculation, experiment and creation. It is an atmosphere in which there prevail "the four essential freedoms" of a university--to determine for itself on academic grounds who may teach, what may be taught, how it shall be taught, and who may be admitted to study.' Sweezy v. New Hampshire, 354 U.S. 234, 263 [77 S. Ct. 1203, 1218, 1 L. Ed. 2d 1311] (1957) (concurring in result).

"Our national commitment to the safeguarding of these freedoms within university communities was emphasized in Keyishian v. Board of Regents, 385 U.S. 589, 603 [87 S. Ct. 675, 683, 17 L. Ed. 2d 629] (1967):'Our Nation is deeply committed to safeguarding academic freedom which is of transcendent value to all of us and not merely to the teachers concerned. That freedom is therefore a special concern of the First Amendment .... The Nation's future depends upon leaders trained through wide exposure to that robust exchange of ideas which discovers truth "out of a multitude of tongues, [rather] than through any kind of authoritative selection.' United States v. Associated Press, 52 F. Supp. 362, 372 [1943]."

438 U.S. 265, 312, 98 S. Ct. 2733, 2759-60, 57 L. Ed. 2d 750 (1978) (emphasis added). See also Wieman v. Updegraff, 344 U.S. 183, 197, 73 S. Ct. 215, 221, 97 L. Ed. 216 (1952) (Frankfurter, J., concurring) ("the functions of educational institutions in our national life and the conditions under which alone they can adequately perform them are at the basis of ... limitations upon State and National power"). Similarly, the decisions of this court also recognize the constitutional dimensions of academic freedom. "Basic academic decisions such as the determination as to the make up of the faculty and who may be a student on the first day of classes, have long been regarded as among the essential prerogatives and freedoms of the university administration." Martin v. Helstad, 699 F.2d 387, 397 (7th Cir.1983) (concurring opinion). See also Dow Chemical Co. v. Allen, 672 F.2d 1262, 1274-76 (7th Cir.1982) (scholarly research subject to First Amendment protections for academic freedom).

It is clear that the peer review process is essential to the very lifeblood and heartbeat of academic excellence and plays a most vital role in the proper and efficient functioning of our nation's colleges and universities. The process of peer evaluation has evolved as the best and most reliable method of promoting academic excellence and freedom by assuring that faculty tenure decisions will be made objectively on the basis of frank and unrestrained critiques and discussions of a candidate's academic qualifications. See Johnson v. University of Pittsburgh, 435 F. Supp. 1328 (W.D.Pa.1977). Moreover, it is evident that confidentiality is absolutely essential to the proper functioning of the faculty tenure review process. The tenure review process requires that written and oral evaluations submitted by academicians be completely candid, critical, objective and thorough in order that the University might grant tenure only to the most qualified candidates based on merit and ability to work effectively with colleagues, students, and the administration. For these reasons, academicians who are selected to evaluate their peers for tenure have, since the inception of the academic tenure concept, been assured that their critiques and discussions will remain confidential. Without this assurance of confidentiality, academicians will be reluctant to offer candid and frank evaluations in the future.

The importance of confidentiality to decision-making processes is recognized throughout the American institutional, administrative and judicial processes. The Supreme Court recognized the importance of confidentiality, albeit in the jury deliberation context, when Justice Cardozo wrote that "[f]reedom of debate might be stifled and independence of thought checked if jurors were made to feel that their arguments ... were to be freely published to the world." Clark v. United States, 289 U.S. 1, 13, 53 S. Ct. 465, 77 L. Ed. 993 (1933). In its adoption of Exemption 5 to the Freedom of Information Act (5 U.S.C. § 552(b)(5)) which protects the decision-making process of governmental agencies, Congress recognized that " 'frank discussion of legal or policy matters' in writing might be inhibited if the discussions were made public; and that the 'decisions' and 'policies formulated' would be poorer as a result." N.L.R.B. v. Sears, Roebuck & Co., 421 U.S. 132, 150, 95 S. Ct. 1504, 1516, 44 L. Ed. 2d 29 (1975) (quoting S.Rep. No. 813, p. 9). Just as a limited executive privilege is necessary for the executive branch of our government to function properly, and as confidential judicial and jury deliberations are essential to preserve the integrity of those processes, confidentiality is equally critical in the faculty tenure selection process, in order that only the best qualified educators become the "lifeblood" of our institutions of higher learning.

There are, however, factors weighing in favor of disclosure of some of the contents of the peer review files. If an academic freedom privilege could be used to totally prohibit disclosure of tenure review records, the privilege could be used as a shield to hide evidence of discrimination. The integrity of the truth-seeking process would be impaired and Congress' goal of eradicating discrimination would be frustrated. Cf. Garland v. Torre, 259 F.2d 545, 549 (2d Cir.), cert. denied, 358 U.S. 910, 79 S. Ct. 237, 3 L. Ed. 2d 231 (1958) (noting that the protection of the First Amendment "is not absolute" and must give way in some cases "to the paramount public interest in the fair administration of justice"). Likewise the EEOC's investigatory powers must not be frustrated in order that it might substantiate meritorious claims and conciliate when appropriate, and reject non-meritorious claims. Moreover, the important interests of academic excellence might well be frustrated if tenure decisions were allowed to be made on other than lawful grounds.

It is clear that both parties before us have significant and substantial interests at stake. After weighing the respective interests, we recognize in this case a qualified academic freedom privilege protecting academic institutions against the disclosure of the names and identities of persons participating in the peer review process thereby reaffirming long-standing policies of academic institutions.2 In so holding, we join other courts in recognizing a limited academic freedom privilege in the context of challenges to college or university tenure decisions. For example, in Gray v. Board of Higher Education, City of New York, 692 F.2d 901 (2d Cir.1982), the court recognized a qualified academic freedom privilege, but held that under the circumstances,3 the plaintiff's need for the privileged information outweighed the college's interest in keeping the information confidential. See also McKillop v. Regents of the University of California, 386 F. Supp. 1270 (N.D.Cal.1975); Zaustinsky v. University of California, 96 F.R.D. 622 (N.D.Cal.1983). But see In re Dinnan, 661 F.2d 426 (5th Cir.1981) (the interest in disclosure invariably weighs more heavily because confidentiality is not an important interest in the peer review context). We recognize the qualified privilege here because it is likely that "the privilege will in fact protect [the] relationship in the factual setting of [this] case." Memorial Hospital for McHenry County v. Shadur, 664 F.2d at 1061.4

Having determined that the district court erred in refusing to recognize a qualified privilege protecting against the disclosure of the identities of the academicians participating in the peer review process, we remand this case to the district court and direct the court to issue an order implementing the procedures that follow. Before producing the personnel files sought by the EEOC, the University should be permitted to redact the name, address, institutional affiliation, and any other identifying features (e.g., publications, professional honors received, or any other material which could be used to identify the particular academician) of the reporting scholar from the evaluations found in each of the files. After completing the redaction procedures outlined above, the University should, since it volunteered to do so originally, produce the redacted files and, in addition, the original unredacted files to the district court.5 The district court shall then review the files, in camera, to determine whether the redactions are reasonably necessary to prevent disclosure of the identity of the scholar or scholars providing the evaluation. Should the court determine that the redactions are reasonably necessary to prevent disclosure of the identity of the evaluating scholar, then the redacted files should be turned over to the EEOC and the unredacted files returned to the University.

If, after receiving the redacted files, the EEOC believes that it is necessary to seek more information from the unredacted files, the EEOC must then make a particularized showing before the court as to its need for further disclosure in any one or more of the files produced. In this respect, the EEOC will be required to make a substantial showing of "particularized need" for relevant information, a burden similar to that imposed on a party seeking disclosure of grand jury materials. See Illinois v. Abbott & Associates, Inc., --- U.S. ----, 103 S. Ct. 1356, 75 L. Ed. 2d 281 (1983); In re Grand Jury Proceedings, Miller Brewing Co., 687 F.2d 1079 (7th Cir.1982). Cf. McGraw Hill, Inc. v. Arizona, 680 F.2d 5 (2d Cir.), cert. denied, --- U.S. ----, 103 S. Ct. 215, 74 L. Ed. 2d 171 (1982) (disclosure of journalist's sources improper absent clear and specific showing that information sought was highly material and relevant and unavailable from other sources).

Before determining whether to compel disclosure of materials covered by the qualified privilege, the court must apply a balancing test to determine whether the need of the party seeking disclosure outweighs the adverse effect such disclosure would have on the policies underlying the privilege. Douglas Oil Co. v. Petrol Stops Northwest, 441 U.S. 211, 223, 99 S. Ct. 1667, 1675, 60 L. Ed. 2d 156 (1979); Pittsburgh Plate Glass Co. v. United States, 360 U.S. 395, 400, 79 S. Ct. 1237, 1241, 3 L. Ed. 2d 1323 (1959); In re Grand Jury Proceedings, Miller Brewing Co., 687 F.2d at 1090-91. Under the "particularized need" standard, "a party's need varies in proportion to the degree of access he has to other sources of information he seeks." United States v. Moten, 582 F.2d 654, 663 (2d Cir.1978). A party must conduct thorough and exhaustive discovery to exploit each and every possible source of information prior to seeking those materials protected by the qualified privilege. "Exploratory" searches will not be condoned. Similarly, the mere fact that certain information may be relevant or useful does not establish a "particularized need" for disclosure of information. United States v. Short, 671 F.2d 178 (6th Cir.), cert. denied, 457 U.S. 1119, 102 S. Ct. 2932, 73 L. Ed. 2d 1332 (1982); Miller Brewing, 687 F.2d at 1091. The party seeking disclosure of the privileged information must show a "compelling necessity" for the specific information requested. Douglas Oil Co., 441 U.S. at 222-23, 99 S.Ct. at 1674-75; United States v. Procter & Gamble Co., 356 U.S. 677, 682-83, 78 S. Ct. 983, 986-87, 2 L. Ed. 2d 1077 (1958).

In addition to the foregoing general rules regarding "particularized need," the court will also be required to consider factors unique to the context of faculty tenure denial. Once the basis for faculty tenure denial is determined it may be easier for the EEOC to demonstrate a "particularized need" for the identity of one or more evaluating academicians because then the court could determine whether the material sought was relevant to the discrimination claim and unavailable elsewhere. For example, if the University contends that in fact Brookins was denied tenure based on his peer evaluations and the EEOC is able to make a particularized showing that an evaluation was not supported by legitimate non-discriminatory reasons or that an evaluator is likely to have displayed racial animus against Brookins or other individuals, disclosure of the particular evaluator's identity may be warranted. See Gray, 692 F.2d at 905 and Keyes v. Lenoir Rhyne College, 552 F.2d 579 (4th Cir.1977). We foresee that the identities of the scholars would be released only under the most limited circumstances. Any particularized showing in this regard would have to be supported by specific material contained in the redacted peer review files or probative evidence obtained through other investigatory means such as the thorough and exhaustive discovery mandated above. If, however, the University could demonstrate that its decision was motivated by reasons separate and distinct from those addressed by the peer reviewers, such as unsatisfactory student evaluations as alleged by Mr. Brookins, there would be little or no need for the identities of the peer reviewers. Lynn v. Regents of the University of California, 656 F.2d 1337 (9th Cir.1981). As always, the relevancy of the requested material to the charge under investigation must be established. See 42 U.S.C. § 2000e-8(a). Obviously, the factors we have set forth are not intended to be exclusive, but rather any determination of "particularized need" must be made on a case-by-case basis and any finding of "particularized need" by the district court must be supported by specific findings allowing for a meaningful appellate review, if necessary. The court should be aware that the procedure outlined above is designed to accommodate the competing interests which are at stake and to prevent an EEOC investigation from becoming "a fishing expedition," or exploratory surgery without the required prior medical testing, screening, evaluation and diagnosis.6 Furthermore, the substantial showing of particularized need must be demonstrated to the satisfaction of the court each and every time the EEOC seeks additional privileged material.

In conjunction with requiring further disclosure after a substantial showing of particularized need for relevant information, the district court may issue any protective order it deems necessary. By following this procedure, judicial and administrative intrusions into the affairs of colleges and universities will be minimized: "[I]t is clear that courts must be vigilant not to intrude into [discretionary academic determinations], and should not substitute their judgment for that of the college with respect to the qualifications of faculty members for promotion and tenure." Kunda v. Muhlenberg College, 621 F.2d 532, 548 (3rd Cir.1980). See also Faro v. New York University, 502 F.2d 1229, 1231-32 (2d Cir.1974). Courts have no more business in substituting their judgment for that of a legitimate peer review determination than they do in determining whether a particular physician or surgeon is qualified to practice in a particular hospital. We believe the procedure outlined above strikes the appropriate balance between the need to preserve the integrity of the peer review process and the need to adequately investigate proper charges of discrimination in higher education.

We now turn to the University's request that the EEOC sign a nondisclosure agreement which would require it to have Mr. Brookins also sign an agreement prior to the EEOC's release of materials to him for litigation purposes. "[I]t is a matter for the district court to issue protective orders permitting a party to keep secret discovered material when 'good cause' is shown." Am. Tel. & Tel. Co. v. Grady, 594 F.2d 594 (7th Cir.1978). See also Jepsen v. Florida Board of Regents, 610 F.2d 1379 (5th Cir.1980) (court has duty to limit use and availability of confidential documents when it orders them produced). Because we have held that the peer review materials are subject to a qualified privilege, it necessarily follows that sufficient "good cause" exists to protect these materials. Furthermore, it does not appear that EEOC procedures are geared towards materials protected by a judicially recognized privilege. Accordingly, we believe that the proper method of handling the University's request is that the district court issue a protective order which assures that the privileged materials are not disclosed to persons not directly involved in this action up to and including the time any private action is filed by the charging party.7 While the names and identifying characteristics contained in the files have been redacted, the files are still highly confidential. This is especially true when one considers the relatively small number of personnel that the files pertain to. Further protective orders may be required if peer reviewer identities are released.

The restrictions and limitations we have placed on the access to the records in question here are necessary not only to preserve the integrity of academic freedom at stake, but also to prevent those who would use this case as authority to invade other confidential relationships from doing so absent the most compelling circumstances.8 Searches of previously sacrosanct places, such as attorneys' offices, physicians' offices, hospital record rooms, etc., are not uncommon and with the ease of collecting and storing information in today's computerized society, Americans have good cause to seek to protect their valued right of privacy from unwarranted intrusions. If we, as the EEOC would have it, allowed unrestricted intrusions into the confidentiality of the peer review process absent the most compelling reasons for disclosure, we would effectively strangle faculty tenure selection procedures by destroying any expectation of privacy held by those participating in the faculty peer review process and thereby discourage honest and candid appraisals of tenure applicants. "Human experience teaches that those who expect public dissemination of their remarks may well temper candor with a concern for appearances and to their own interests to the detriment of the decisionmaking process." United States v. Nixon, 418 U.S. 683, 705, 94 S. Ct. 3090, 3106, 41 L. Ed. 2d 1039 (1974). Secret voting without discussion and reason would be encouraged and lawsuits could conceivably be filed by every unsuccessful tenure applicant in an effort to pierce the qualified academic privilege. Our decision assists in "safe-guarding academic freedom which is of transcendent value to all of us and not merely to the [educators] concerned." Keyishian v. Board of Regents, 385 U.S. 589, 603, 87 S. Ct. 675, 683, 17 L. Ed. 2d 629 (1967).

The decision of the district court is reversed and this case is remanded for further proceedings consistent with this opinion.