This is an appeal by Samuel Popkin from an order of the district court holding him in civil contempt for refusing to answer certain questions propounded to him by a federal grand jury in Massachusetts. An assistant professor of government at Harvard University, Popkin has written numerous articles on the war in Indochina.1 He contends first that he should not be forced to respond without a demonstration by the government of the relevance of both the general inquiry and the specific questions. He urges also that he should be excused from answering those questions by virtue of a scholar's First Amendment privilege not to divulge his sources of information insofar as those sources are confidential and supply him with information relating to his field. Finally, he claims that he need not testify until the government has disclaimed the use of illegal electronic surveillance.

The grand jury which issued a subpoena to Popkin is the same as that involved in United States v. Doe (Mike Gravel, United States Senator, Intervenor), 455 F.2d 753 (1st Cir. 1972), cert. granted, 405 U.S. 916, 92 S. Ct. 1243, 30 L. Ed. 2d 785 (Feb. 22, 1972 [hereinafter Gravel]).

In ordering Popkin to testify, the district court found that the grand jury is "engaged in an inquiry into alleged violations of Chapter 37 [Espionage and Censorship] and Sections 2314 [Transportation of stolen goods etc.] and 2315 [Sale or receipt of stolen goods etc.] of Title 18, United States Code", with its primary focus the dissemination of the classified Defense Department study known as the Pentagon Papers.

Popkin was first subpoenaed to appear before the grand jury on August 19, 1971. Before that date, he moved to quash the subpoena and for disclosure of electronic surveillance, which motions were denied. The grand jury did not, however, ask him to testify at that time. He received a second subpoena to appear before the grand jury on October 13. Having refused to testify on October 14, he was ordered to reappear on October 27. On that date, he filed motions, supported by his own affidavit and those of twenty-four other social scientists, for an order to protect him from inquiry as to "information obtained by him in his capacity as a scholar, author and teacher", for a transcript of his testimony, and for disclosure of surveillance. After these motions were denied on October 28, Popkin was granted immunity from prosecution and ordered to testify. Again excused, he was not subsequently required to appear until January 18, 1972. His renewed motions for a protective order and for a transcript were denied. Upon his refusal to answer three questions, the government sought and obtained a contempt order on March 21. Popkin purged himself of contempt by answering these three questions on March 27, then answered further questions but, relying on his asserted First Amendment privilege as a scholar, refused to answer a number of other questions. Motions for a protective order and for disclosure were again denied on March 29, and he was held in contempt. He now appeals from the March 29 contempt order.

The district court, without opinion, based its contempt order on Popkin's refusal to answer nine questions, of which the government now presses seven. In order to understand the scope of Popkin's refusal to testify, it may be useful first to sketch the relevant parts of his testimony in response to the questions he did answer. He did state, among other things, that he had never seen a copy of the Pentagon Papers other than those in mass distribution, that he had not discussed with Daniel Ellsberg the possibility of releasing a copy of the Pentagon Papers to Neil Sheehan (of the New York Times), that he had no knowledge other than from public sources as to how various newspapers had obtained copies of the Pentagon Papers, and that he "was never given definitive information that someone possessed" the Pentagon Papers in Massachusetts. In addition, Popkin admitted having an opinion that certain unnamed persons had possession of the Pentagon Papers in Massachusetts, explaining that this opinion was formed on the basis of conversations "with numerous persons about numerous decisions and documents, I see now that a lot of those decisions and documents are in the Pentagon Papers".

The questions Popkin refused to answer represent three lines of inquiry which, while overlapping, we will consider separately. One line of inquiry attempted to identify the participants in the Pentagon Papers study or, more precisely, persons having knowledge as to who participated in the study. This inquiry began with the question: "Who are the persons you interviewed in order to acquire this knowledge of who participated in the Pentagon Papers study?" (chronologically, question 2). This was followed, after Popkin had declined to answer and had indicated in response to subsequent questions that it had never been his research object to discover who participated in the study, by the colloquy (question 3):

"Q. Who are those persons to whom you are referring with regard to conversations in the District of Massachusetts which led you to the knowledge of who are the participants in the Pentagon Papers study?"

******

* * *

"Q. Would you please name them?

A. I respectfully and regretfully decline [etc.]."

The second line of inquiry is more explicit and is self-explanatory. Popkin refused to answer the question "Did Daniel Ellsberg ever discuss with you the content or existence of the Pentagon Papers between January 1, 1971 and June 13, 1971?" (question 4).

The third line of inquiry, finally, related to his conceded opinion as to possession of the Pentagon Papers, beginning with the question: "[W]hat is the opinion as to persons you believe possessed a copy of the Pentagon Papers in Massachusetts prior to June 13, 1971" (question 1). The line of inquiry was then temporarily dropped, but was later resumed by repetition of the question (question 5). Subsequently, he refused to answer the question "Would you please name them, those persons who furnished you information which caused you to form an opinion as to persons you believe possessed a copy of the Pentagon Papers in Massachusetts prior to June 13, 1971?" (question 6) and the similar, but somewhat more focussed, question "Who was the conversation with discussing the documents or decisions indicating that person had knowledge or possession of the Pentagon Papers?" (question 7).

We first discuss appellant's concept that, at least when a grand jury inquiry impinges on the First Amendment rights of a witness, a preliminary showing of relevancy is required. Despite the long-standing doctrine of Blair v. United States, 250 U.S. 273, 282, 39 S. Ct. 468, 471, 63 L. Ed. 979 (1919), that a grand jury witness is "not entitled to urge objections of incompetency or irrelevancy" appellant relies on such legislative inquiry cases as Watkins v. United States, 354 U.S. 178, 77 S. Ct. 1173, 1 L. Ed. 2d 1273 (1957) and Scull v. Commonwealth of Virginia, etc., 359 U.S. 344, 79 S. Ct. 838, 3 L. Ed. 2d 865 (1959).

We are in effect being asked to make new law on the basis of the assumption that grand jury and legislative inquiries function in so similar a fashion that they should be treated alike. The argument, so considered, is not without initial appeal. But a legislative investigation has as its object the resolution of a previously identified problem of a general nature. The power, while broad, is limited to advancing the general resolution. To the extent that individuals can supply information related to the legislative task, they must respond. When the broad brush seeks not only to paint the scene but to detail individual portraits unnecessary to that scene, it ventures too far. The concept of relevance to the legislative purpose is necessary to preserve the boundary between the legislative and the individual, and can ordinarily be enforced by reference to statements of purpose and prior testimony which are already matters of public record.

A grand jury generally has a much smaller object in view, but one in which society has intense interest. At least when inquiring into specific crimes, a grand jury paints with a small brush and necessarily with painstaking detail. Its purpose is to see if there is probable cause to believe that particular crimes have been committed by particular persons, although further crimes and perpetrators may be discovered in the course of taking evidence. In either case, the number of facts which would help to identify the crime and the criminal may well be few, and the relevance of one fact may only become apparent after another is uncovered. Thus while the public product of a grand jury is a limited number of indictments, its processes may include the pursuit of numerous strands of evidence, many of which may uitimately be of no use, but none of which can safely be abandoned. From another perspective, even where a grand jury is working on the clear hypothesis of a prosecutor, its success may often depend upon its ability to offer to witnesses a guarantee of secrecy and to suspects the protection of secrecy unless probable cause is found to indict. Any meaningful statement of relevance in response to a witness's request would entail the naming of suspects and the description of prior testimony. Even if a present witness has no wish for secrecy, disclosure would injure other witnesses and suspects. We therefore reject appellant's argument based on the asserted similarity of this grand jury to a legislative hearing.

Appellant also attempts to draw support for a rule of relevance from Fourth Amendment cases, such as Oklahoma Press Publishing Co. v. Walling, 327 U.S. 186, 209, 66 S. Ct. 494, 90 L. Ed. 614 (1946). As to this, we deem sufficient Judge Hamley's reference in United States v. Weinberg, 439 F.2d 743, 749 (9th Cir. 1971), to the following language of Judge Zirpoli in In re Grand Jury Witnesses Sherrie Bursey and Brenda Joyce Presley, 322 F. Supp. 573, 576 (N.D.Cal.1970):

"The grand jury does not need to have probable cause to investigate; rather its function is to determine if probable cause exists. And if probable cause is not required to investigate, it follows that probable cause is not required to make the preliminary showing necessary to call a witness whose testimony may shed light on criminal activity. . . ."

We conclude that, even if the questions put to appellant have not been demonstrated to be relevant to the grand jury's inquiry, this would not justify his refusals to answer. His invocation of the First Amendment adds nothing to his relevancy claim in the context of a grand jury investigation. Whether he has presented a First Amendment claim strong enough to be treated as a constitutional privilege is a separate question.

In determining whether appellant may refuse to answer any or all of these questions, we are presented, as in Gravel, with a confrontation of constitutional power and constitutional privilege, heightened by similar overclaiming on each side. Here also there is little direct precedent to guide us, and though the Supreme Court may soon pronounce law that bears on the balance between the power of the grand jury and First Amendment privileges,2 we may not await its decision.3

The government argues that the scholar's privilege is a creature not to be found in the province of jurisprudence; that the closest analogue, a reporter's privilege, recognized by the Ninth Circuit in Caldwell v. United States, 434 F.2d 1081 (9th Cir. 1970), cert. granted, United States v. Caldwell, 402 U.S. 942, 91 S. Ct. 1616, 29 L. Ed. 2d 109 (1971),4 has been otherwise uniformly rejected by the courts; and that even if Caldwell was correctly decided, appellant falls outside its compass, his relationship with his sources not being so intimate as in Caldwell, nor covered by the special canopy of free press, and the grand jury being engaged in a specific rather than a general investigation. Appellant discounts the lack of precedent as stemming from the past absence of efforts to invade a scholar's privilege. He claims to serve a public interest fully as vital as that served by a reporter, which interest would be as grievously imperilled by a forced disclosure of his sources, absent a showing of compelling need by the government.

In Gravel, supra, we held that the Speech and Debate Clause was sufficient to block a grand jury from inquiring of a Senator and his aides about the receiving and preparation for legislative publication, the publication, and any official republication of the Pentagon Papers. Our decision, we recognized, could substantially frustrate a grand jury, for the transactions embargoed from inquiry might encompass a crime. But we thought the result compelled by the Speech and Debate Clause, which is not involved here.

For perspective, it is important to recognize what has not been and could not successfully be argued here. Appellant could not, for example, cite his discomfort in being asked to testify about others. Although the discomfort is real, it is shared by all grand jury witnesses. Indeed, a witness called before a grand jury investigating organized crime may legitimately fear for life and limb. Nor can appellant stand on a claim, however justified, that his livelihood as a researcher is threatened. His privilege, if it exists, exists because of an important public interest in the continued flow of information to scholars about public problems which would stop if scholars could be forced to disclose the sources of such information. Appellant is a political scientist. As is true of other behavioral scientists, his research technique rests heavily on inquiry of others as to their attitudes, knowledge, and experience. Often such inquiry is predicated on a relationship of confidence. The question posed to us is whether this relationship gives a scholar a constitutionally based privilege not to testify which others do not possess.

Although both parties have cast their arguments in these broad terms, the substantiated rationale of appellant and the questions put by the grand jury do not, in our opinion, quite meet head on. The overwhelming majority of appellant's affidavits from other scholars lay stress on the importance of a two-way communication between participants in decision making (or those affected) and scholars. The asserted importance of non-disclosure of a network of sources lies in the necessity for a continued flow of inside information to the sphere of scholarly reporting, assessment, and criticism, a flow made particularly important by the selective and self-serving release of information by other, particularly higher officials.5 In short, the thrust of the asserted privilege bears on the right of the scholar not to disclose the identities of his "contacts and sources", those officials and non-governmental actors within the purview of his specialty whose actions, knowledge, and views give him the primary data for his work.6 The reason for the claimed privilege lies not in the importance of protecting the officials and other sources per se but in the importance of preserving the flow of their communications via scholars to the public domain.

This underlying rationale falls short of immunizing a scholar from testifying about conversations with those who are not his sources. Of course a scholar may also be an official or such a participant in an activity that he is also a sensitive source. But to the extent that a scholar qua scholar is asked about statements made to him by other scholars we do not conceive of him as in any different position from that of a doctor asked about his conversations with other doctors, or a lawyer about his talks with other lawyers. Nor is there reason to believe that scholars, as opposed to public officials, will lose their jobs and thus their usefulness as sources if the contents of their conversations are revealed to a grand jury. And while we acknowledge that scholars customarily discuss their work with colleagues and in doing so may perhaps violate confidences, a privilege which would give comprehensive protection to such collateral discussions would make scholars a uniquely privileged class in the broadest sense.

We therefore conclude that the two questions seeking the names of persons interviewed who gave him knowledge of participants in the Pentagon Papers study should be answered, at least to the extent that the persons were not government officials or other participant-sources. One of the questions, relating to persons in Massachusetts, seems not to be directed either to foreign affairs, intelligence, or defense officials in government or to any person, in or out of the government, in Vietnam. Were it to become clear that a question delved into confidential relationships with such persons, we are not sure that the grand jury would see fit to pursue its inquiry. If it did, however, would at least have a factual setting enabling us to address the issue less theoretically.

This analysis also disposes of the question asking whether Ellsberg discussed with appellant the content or existence of the Pentagon Papers between January 1 and June 13, 1971. There is no representation that Ellsberg during that period was a government official, or otherwise a sensitive source, or even that Popkin had established a confidential relationship as to these matters with Ellsberg when the latter was a public official.

What this analysis does not dispose of is the remaining group of questions seeking appellant's opinion concerning who might have possessed the Pentagon Papers generally or in Massachusetts and the persons with whom he talked who gave him a basis of that opinion. Appellant had candidly testified that, while he had no knowledge of possession in Massachusetts, he had formed opinions over the years as to persons he thought had had access to the Papers. This kind of inquiry, at least in the present posture of this case, does not appeal to the author of this opinion. Appellant in his brief has asserted that he knows of no case where a witness has been held in contempt for refusing to give an opinion. Nor do I. In the long run, the quest for opinions would not be a useful investigative tool. If appellant were forced to answer, scholar-sleuths would in the future think long and hard before admitting to an opinion, and grand juries would be without workable means for forcing them to do so.

Beyond this, and more to the point, the previously stated basis for his opinion lay in matching the content of conversations he had had with the content of what subsequently publicly appeared, and deducing that the source was the Papers. Appellant testified that many people had publicly admitted having had access to the Papers or copies thereof. It is reasonable to assume that much of the work forming the basis of appellant's judgment is as well known to government experts as to appellant. In this connection, note the statement, in the Attorney General's instructions regarding subpoenas to the press, that "The Department of Justice does not consider the press 'an investigative arm of the government"' (quoted in Levin v. Marshall, 317 F. Supp. 169, 173-174 (D.Md. 1970). Nor has the government here shown that it cannot obtain the information it seeks other than by procuring Popkin's opinion.

The generality of the opinion questions here, the apparent basis for the opinion being pursued, and the idea of using one scholar to speculate about the sources of others' work, without any showing of strong need therefor, are repugnant to me. In my view, even apart from constitutional claims, we should exercise our supervisory power to state that in this circuit scholars ought not to go to prison for refusing to give their opinions or beliefs based on casual and retrospective reflections on similarities of content. But while my brothers agree that the opinion questions were improper, their objection is a narrower formal one. See infra. Accordingly, at present we simply disapprove the questions as asked.

In addition to appellant's relevancy and First Amendment justifications for refusal to answer, he asserts that he is under no further duty to testify until the government makes a satisfactory disclosure that no illegal electronic surveillance has tainted the questioning. We have traced the history of appellant's attempts to force disclosure, the last and most formal and specific motion being filed on March 28, the day after his last grand jury appearance, and specifying as the ground for his claim that he had been asked his telephone number.

It may be, as a majority of the Third Circuit held in In the Matter of Egan, 450 F.2d 199 (3rd Cir. 1971), cert. granted United States v. Egan, 404 U.S. 990, 92 S. Ct. 531, 30 L. Ed. 2d 541 (Dec. 14, 1971), that a grand jury witness has standing in a civil contempt proceeding to raise as a defense the prohibition in 18 U.S.C. Sec. 2515 against the introduction of evidence based upon electronic surveillance, or, as a majority of a District of Columbia Circuit panel held in In re Evans, 452 F.2d 1239 (D.C. Cir. 1971), that a grand jury witness has standing under 18 U.S.C. Sec. 2518(10) (a) to make a motion to suppress, or that he has standing under 18 U.S.C. Sec. 3504(a) to make the government affirm or deny the use of illegal electronic surveillance. On the other hand, it may be that he has standing under none of these sections. United States v. Gelbard, 443 F.2d 837 (9th Cir. 1971), cert. granted, Gelbard v. United States, 404 U.S. 990, 92 S. Ct. 529, 30 L. Ed. 2d 540 (Dec. 14, 1971). We do not reach the question because we find that Popkin, even if he has standing, has failed to make sufficient claim that the issuance of the subpoena or the asking of questions is the product of illegal surveillance.

Section 2518(10) (a) indicates that a motion to suppress evidence obtained by illegal electronics surveillance may be made on any one of three "grounds"7 and that "[s]uch motion shall be made before the trial, hearing, or proceeding unless there was no opportunity to make such motion or the person was not aware of the grounds of the motion." The language of Sec. 3504(a) (1)-"upon a claim by a party aggrieved that evidence is inadmissible because it is the primary product of an unlawful act or because it was obtained by the exploitation of an unlawful act"-carries with it, we think, a similar responsibility on the aggrieved party to make a threshold showing, although the threshold may be considerably lower. See In re Evans, supra, 452 F.2d at 1268 (Wilkey, J., dissenting), and the legislative history detailed there. And while Sec. 2515 contains no comparable requirement, neither does it explicitly provide a remedy. Any implied remedy would, in our view, import a requirement of some showing of basis for suspecting illegal action. Otherwise, the two last cited sections would in effect have enacted an automatic disclosure rule, which we do not think the Congress intended.8

At the same time, we note the statement of Chief Judge Bazelon in In re Evans, supra, 452 F.2d at 1247, that the duty of the government to disclose under Sec. 3504(a) (1) "is triggered . . . by the mere assertion that unlawful wiretapping has been used", and are sensitive to the concern which underlies it.9 We recognize full well that it is difficult, and perhaps impossible, particularly in the early stages of a case when he is called before the grand jury, for a witness to know that he has been the subject of electronic surveillance. We would concede the absurdity of requiring precise information about a surveillance, which, unless poorly done, could not be pin-pointed by the victim. Because of the elusive nature of electronic surveillance, adequate grounds might be gleaned from a number of circumstances, including subject matter, association, coincidence or events.

In the present case, the district court found no such circumstances as would lead it to suspect illegal electronic surveillance. At one stage of the proceedings, on January 18, by way of preliminary questioning, the grand jury asked his name, his address, and his telephone number. The last question was asked again and answered on March 27. We can attach no sinister inference to the fact that the government sought to record these identifying items. The telephone number was in the telephone book. If we drew any inference as to surveillance it might be, in the light of the history of the Pentagon Papers investigation, cf. In re Marx, 451 F.2d 466 (1st Cir. 1971), that the government wanted to be ready to prepare a disclaimer if the court ordered it, which, of course, it never did. Furthermore, the questions asked in the grand jury were of the broadest and most speculative kind, bearing no hallmark that the government had any specific notion of what Popkin might be able to contribute. On this record, we do not think that the district court erred in refusing to compel the government to affirm or deny illegal electronic surveillance.

The government makes a further point. Although Popkin filed motions to the court raising the matter of illegal surveillance, when he refused to answer questions on March 27 he asserted at length the First Amendment privilege which had been the basis of his original affidavit, but made no mention of his allegation or defense of electronic surveillance. The government says this was a waiver, and suggests that the grand jury might accept the claim. While we need not and do not decide whether his actions constituted a waiver, because of his failure adequately to claim illegal electronic surveillance at any time, we believe it behooved him, particularly with the Marx precedent before him, to reassert the surveillance point when he refused to answer.

Affirmed in part, reversed in part.

ALDRICH, Chief Judge (concurring).

We are, to the extent that we are asked to recognize any privilege here at all, exploring very new ground, and while Judge McEntee and I agree with the result, and much of Judge Coffin's opinion, our cast would be somewhat different. A minor difference relates to Judge Coffin's approach to the questions about Popkin's "opinions" as to who had had possession of the Pentagon Papers. Our only objection to those questions is the semantic one that they are badly phrased. What, after all, is meant by an opinion? Had the question been, "Is there anyone who you have reason to believe had possession of the papers in Massachusetts, and what are the reasons?" it would have seemed just the sort of inquiry that might lead to something useful. One cannot expect gold with every stroke of the pick.

Of more significance, we are not so sure on what the court calls "decision-making (and those affected)" sources. A valuable, confidential source may be at a very low, and even unrelated level. If what is sought to be protected is the public interest in information, should not the need of confidentiality be the test, not the position of the source?

This question leads us to a dilemma. What assurance does a court have that there is a need of confidentiality in the particular case? Popkin, if we judge from his oral argument, believes that he should have an all-encompassing mantle in whatever may be his field, so that he can be known as a "safe" man to talk to. We do not read Judge Coffin as going that far; nor would we. But where does one stop?

Perhaps I am old-fashioned, but I was taught that a scholarly study was valuable to the extent that it disclosed its sources. How does Popkin know that he, and hence his public, is not being hornswoggled by a "source"? Is there great public worth in a book, the reference table of which consists of a bare curriculum vitae of the author?

The answer may be yes, and may be no. I am tempted to wonder, though I hope uncharacteristically, if too much is not being asked of the First Amendment. Hearst could consider Walter Winchell so valuable to it that it was willing to agree that, in case of a libel suit, it would pick up the tab and not require him to divulge his source. Is the public so interested in research that the government finds itself with a similar, although diminished in scope, contract of immunity from disclosure with every Ph.D.? If so, we believe it should be in very narrow limits. Happily this case does not call for them to be defined.