MEMORANDUM OF OPINION AND ORDER
SCHWARZER, District Judge.In this Title VII action plaintiff, a tenured faculty member of the University of California at Santa Cruz, seeks to prove that she was denied promotion to full professor and otherwise discriminated against because of her sex. Plaintiff has twice moved to compel discovery of peer evaluations which were written and submitted with an express expectation of confidentiality.1 Her first motion was denied in February 1981 by Magistrate Woodruff on the authority of McKillop v. Regents of the University of California, 386 F.Supp. 1270 (N.D.Cal.1975). Judge Weigel denied plaintiff’s motion for reconsideration.
On August 6, 1982, plaintiff again requested defendant to produce her personnel file in addition to the files of seven comparable males. The University again objected, and plaintiff filed a second motion to compel. Plaintiff based the renewal of her motion on these “changes in law and fact”: (1) the intervening Ninth Circuit decision of Lynn v. Regents of the University of California, 656 F.2d 1337 (9th Cir.1981), had, according to plaintiff, “implicitly repudiated” McKillop; (2) defendant’s responses to interrogatories had made it clear that the material in the confidential files was the “heart” of the University’s defense.
Magistrate Woodruff granted plaintiff’s second motion, relying on Lynn. On December 3, 1982, he entered an order compelling production of the files, subject to certain protective restrictions.2 Defendants move for reconsideration of that order.
I. The Lynn Case
The first question to be decided is whether Lynn requires affirmance of the magistrate’s decision.
The plaintiff in Lynn had been denied access to her tenure review file pursuant to a protective order. At trial the University submitted the file to the court for an in *624camera inspection after the plaintiff had introduced into evidence the tenure committee’s minority report. The Ninth Circuit held that the receipt and review of the file by the trial court without disclosure to the plaintiff violated due process. “In view of [that] holding,” the court noted, it “need not decide the question whether tenure review files, and more particularly peer evaluations, are privileged in academic Title VII cases generally.” 656 F.2d at 1346. The court went on, however, to “provide the district court with some guidance” on that question for the proceedings on remand. Id It did so by expressing agreement with the decisions of the Fourth and Fifth Circuits in Jepsen v. Florida Board of Regents, 610 F.2d 1379, 1384 (5th Cir.1980), and Keyes v. Lenoir Rhyne College, 552 F.2d 579, 581 (4th Cir.) (dictum), cert. denied, 434 U.S. 904, 98 S.Ct. 300, 54 L.Ed.2d 190 (1977), and stating:
The University claims that Lynn was denied tenure because of deficient scholarship. Since its view of Lynn’s ability is based, in large part, on the content of the tenure review file, including peer evaluations, the University is defending, in essence, on the ground that its tenure decision with respect to Lynn was based on nondiscriminatory peer evaluations. Under Jepsen and Keyes, disclosure of the evaluations would be required.
656 F.2d at 1347.
The facts of this case appear to be covered by the language in Lynn, for there is little question that the University’s defense is based “in whole or in part” on the plaintiff’s confidential evaluation. Lynn, supra, 656 F.2d at 1342 n. 15. Since both the majority and the concurring opinions in Lynn make it clear, however, that the court did “not decide the question whether tenure review files, and more particularly peer evaluations are privileged . . .,” id. at 1346, this Court is not bound by the Lynn court’s expression of its view. Moreover, considering the delicacy and importance of this issue, the Ninth Circuit presumably will wish to give full consideration to all of its ramifi1 cations before articulating a definitive rule. This is particularly true in light of the post-Lynn decision of the Second Circuit in Gray v. Board of Higher Education, 692 F.2d 901, 30 FEP Cases 297 (1982).
II. The Scope of the Duty to Produce Confidential Review and Evaluation Files
No absolute privilege protects academic files from disclosure. The University’s claim must be judged in accordance with Rule 501 of the Federal Rules of Evidence, which leaves questions of privilege to case-by-case adjudication “in the light of reason and experience.” See Trammel v. United States, 445 U.S. 40, 47-48, 100 S.Ct. 906, 910-11, 63 L.Ed.2d 186 (1980). When universities have asserted in academic Title VII cases that confidential files or votes are privileged, the courts have “balanced the university’s interest in confidentiality, i.e., in maintaining the effectiveness of its tenure review [or promotion] process, and the need which Title VII plaintiffs have for obtaining peer evaluations in their efforts to prove discriminatory conduct.” Lynn, supra, 656 F.2d at 1347.
Although courts have struck this balance with varying results,3 there can be no question that the materials in issue here are eligible for treatment as privileged confidential communications. The criteria of eligibility, as stated by Wigmore, are these:
*625(1) The communications must originate in a confidence that they would not be disclosed.
(2) This element of confidentiality must be essential to the full and satisfactory maintenance of the relation between the parties.
(3) The relation must be one which in the opinion of the community ought to be sedulously fostered.
8 Wigmore, Evidence § 2285 at 527 (McNaughton rev. 1961). Each criterion is satisfied by the materials involved here. First, the evaluations were submitted with an expectation of confidentiality. Second, confidentiality is a prerequisite to the effectiveness of a peer evaluation system of faculty selection and promotion. See McKillop v. Regents of the University of California, 386 F.Supp. 1270, 1276 (N.D.Cal.1975); see also Gray v. Board of Higher Education, 92 F.R.D. 87, 92-93 (S.D.N.Y.1981); Note, Preventing Unnecessary Intrusions on University Autonomy: A Proposed Academic Freedom Privilege, 69 Cal.L.Rev. 1538, 1551-52 (1981). Third, the peer review system is thought to be worth fostering because it “has evolved as the most reliable method for assuring promotion of the candidates best qualified to serve,” Johnson v. University of Pittsburgh, 435 F.Supp. 1328, 1346 (W.D.Pa.1977); see Kunda v. Muhlenberg College, 621 F.2d 532, 547-48 (3d Cir.1980).
To say that the material may qualify for treatment as privileged does not dispose of the matter, however. The decision whether to protect material against disclosure because of its confidential character turns on a fourth criterion, stated by Wigmore as follows:
The injury that would inure to the relation by the disclosure of the communication must be greater than the benefit thereby gained for the correct disposal of the litigation.
8 Wigmore, supra. See Jepsen v. Florida Bd. of Regents, supra, 610 F.2d at 1384-85; Keyes v. Lenoir Rhyne College, supra, 552 F.2d at 581.
This issue cannot be resolved in the abstract but only in the particular litigation context in which it arises. See Herbert v. Lando, 441 U.S. 153, 169-71, 99 S.Ct. 1635, 1645-46, 60 L.Ed.2d 115 (1979); Gray, supra, 30 FEP Cases at 300. Moreover, a claim of relevance must be closely scrutinized to ensure that the substantial interest in confidentiality is “weighed carefully in striking the proper balance.” Herbert v. Lando, 441 U.S. at 180, 99 S.Ct. at 1651 (Powell, J., concurring).
In approaching this task, the first consideration should be that a plaintiff claiming discriminatory treatment ought not to be entitled to a general inquisition into the University’s files merely on the strength of having filed a complaint. Thus, plaintiff must first demonstrate, at some stage of discovery, his or her ability to establish a prima facie case. See Lynn, supra, 656 F.2d at 1340-41.4 If it appears at that stage that plaintiff can probably make a prima facie case, the defendant must disclose the reasons for its actions and supporting evidence. To the extent those reasons are based on materials in the University’s files which it regards as confidential, some disclosure must then be required which is adequate to enable plaintiff to attempt to meet the burden of proving those reasons to be pretextual. In the first instance, that disclosure should be in the form of “a written statement of the reasons for [defendant’s] decision, including a comprehensive summary in writing of the substance of confidential documents in such records.”5 If after that production it appears that the material produced does not fully reflect the basis for the University’s action or for its defense of the case, the Court must determine the extent to which the University's position implicates other *626confidential and undisclosed material. Whether additional production should then be required, or whether an appropriate preclusion order should be imposed on the University, will turn on a final balancing of the relative prejudice and injury to the parties in the light of the record at that time.
To be meaningful, this procedure must be pursued during the discovery and pretrial stage. Cf. Gray, supra, 30 FEP at 300. There is no reason why the parties cannot be required to disclose their litigating positions and evidence in the course of discovery and pretrial proceedings sufficiently to make this procedure feasible. The Court has adequate powers under Rules 16 and 26(f) of the Federal Rules of Civil Procedure to define the issues and to control discovery in conformity with such a definition. The procedure outlined should provide reasonable protection to the University’s interest in confidentiality while precluding it from unfairly hiding behind that confidentiality in conducting its defense.
In accordance with the foregoing discussion, it is ordered that:
(1) The magistrate’s order of December 3, 1982, is vacated;
(2) Inasmuch as it appears likely that plaintiff will be able to prove a prima facie case, defendant shall produce for plaintiff within ten days of this order:
(a) a written statement of the reasons for its actions complained of here; and
(b) a comprehensive summary of the substance of the confidential documents in plaintiff’s personnel review file.
The Court will entertain such proceedings as may be appropriate following production of the above materials.
IT IS SO ORDERED.
. Pursuant to University policy, departmental chairmen notify all prospective evaluators that their letters “will be held in confidence and the candidate will not be given access” to them. Academic Personnel Manual § 220-80(c).
. The files are to be available only to the parties, their experts, and their attorneys. Documents from the files may be used at depositions only if the witness wrote the document, is otherwise familiar with it, or would have had access to it independently under University regulations. Deposition transcripts shall be sealed in relevant part and “shall not be filed unless required in any further proceeding herein.”
. Courts have taken four different positions:
(1) The interest in disclosure invariably weighs more heavily because confidentiality is not an important interest in this context. In re Dinnan, 661 F.2d 426 (5th Cir.1981).
(2) The interest in disclosure weighs more heavily when the university’s defense is based on the content of documents which the plaintiff seeks to discover. Lynn; Jepsen; Keyes.
(3) The interest in disclosure weighs more heavily only when the plaintiff was given no contemporaneous statement of reasons for the denial of tenure or promotion. Gray v. Board of Higher Education, 692 F.2d 901 (2d Cir.1982).
(4) The interest in confidentiality outweighs the interest in disclosure. McKillop v. Regents of the University of California, supra.
. As Lynn shows, the elements of a prima facie case do not require proof of a discriminatory intent.
. In this case plaintiff is entitled to such a statement and summary in any event under the terms of the University’s Academic Personnel Manual, § 220-80(i).