Plaintiffs in this action are two newspaper-publishing corporations;1 twelve individual journalists;2 and the “Reporters’ Committee for Freedom of the Press,” a legal research and defense fund organization established to protect the interests of the institutional press. The defendants in this case are the American Telephone and Telegraph Company (“AT&T”) and the Chesapeake and Potomac Telephone Company (“C & P”). Plaintiffs brought suit in the District Court for declaratory and injunctive relief, claiming that the First and Fourth Amendments require that prior notice be provided to them before defendants turn over their long distance telephone billing records to Government law enforcement officials. The United States intervened as a party defendant. The District Court denied plaintiffs’ motion for summary judgment and granted the summary judgment motion of defendants and of the United States. With respect to plaintiffs Jack Anderson, Richard Dudman, James R. Polk, David E. Rosenbaum and Knight Newspapers, we reverse the District Court and remand the case. We affirm the District Court’s grant of summary judgment against the other plaintiffs.
I. THE FACTS
A. The Records in Question and Their Usefulness in Law Enforcement
Telephone companies, like other businesses, maintain for billing purposes records of the services they provide their customers. Company billing records for local calls contain only the total number of local message units charged to a customer, but no details concerning specific calls.3 However, company billing records for long distance calls, referred to as “toll calls,” contain, for each long distance call charged to the customer’s number, a notation of the number called and the date, time, and duration of the call.4 Telephone companies retain these “toll-call” records for a period of six (6) months.5 Telephone subscribers are aware that these records are maintained, since they receive copies of them with their phone bills each month.
For at least the past 50 years, state and federal law enforcement officials have used information from telephone company toll-billing records in criminal investigations and prosecutions.6 The telephone companies have generally cooperated with law enforcement efforts and, in response to official requests, have made their business records available for inspection.
The toll-call records are useful investigative tools because the information they contain sometimes provides circumstantial evidence that there has been contact between a subscriber and some other person. There are limits on the usefulness of these records, however. Because they reveal nothing regarding the content of the call, they virtually never provide direct evidence of criminality. Furthermore, it is relatively easy for subscribers to avoid recordation of their long distance calls if they desire to keep their telephonic contacts secret. Records relating to a particular subscriber reflect only long distance calls charged to the subscriber’s number. Thus, they cannot be used to trace calls made to or from the subscriber’s telephone but charged to some other number. Moreover, toll records are *1037not maintained for pay phones or for individual extensions of general business telephones and thus cannot be used to trace calls from such phones. Finally, as already noted, toll-call records are maintained only for six months and are not available for inspection once that period has expired,
Despite these limitations, toll-billing records have become an invaluable law enforcement aid. They have been most suecessfully used in the investigation and prosecution of organized crime and major narcotíes offenses.7 Law enforcement agencies and grand juries currently issue subpoenas and summonses for toll-call records relevant to felony investigations at the rate of approximately 2,000-3,000 each month.8
Defendant AT&T and the associated companies of the Bell Telephone System operate a nationwide telecommunications net-
*1038work. Defendant C & P is an associated company of the Bell System, providing telephone services for the District of Columbia. AT&T has no telephone subscribers and thus does not maintain subscriber billing records. C & P, as an operating company, does maintain such records in the ordinary course of its business.
B. Limitations on Disclosure Under the AT&T Policy
Before 1974 there was no uniform policy within the Bell System governing the release of company billing records to law enforcement officials. In March 1974, however, AT&T adopted throughout the System a formal policy on the release of such records. This policy, which is currently in effect, has three important provisions.
First, the policy prohibits the release of toll-billing records in the absence of a subpoena or summons, valid on its face, issued under the authority of a statute, court, or legislative body.9
Second, the policy requires that subscribers whose toll-billing records have been subpoenaed in civil suits, non-criminal investigations, and non-felony criminal investigations be immediately notified upon receipt of the subpoena, the same day by telephone and, in addition, by letter within 24 hours.10 Both the oral and written notification must include the approximate date on which the company will furnish the toll-billing records.
Third, the policy requires that subscribers whose toll records have been subpoenaed in felony investigations be similarly notified, unless the subpoena is accompanied by a written certification stating (1) that the subpoena or summons was issued pursuant to an official investigation of a suspected felony or an official legislative investigation, and (2) that notification to the subscriber could impede the investigation. Such a certification is effective for 90 days, and must be renewed by further certifications in writing for successive 90-day periods. When a certification period expires, the subscriber is to be notified of the subpoena if he so requests or if he has previously submitted to the company a general request for notification.11
The decision to adopt this policy originated solely with AT&T; no Government official requested or prompted the action.12 AT&T placed the new policy into effect on 1 March 1974.
C. Impact on Journalists of Toll-Record Subpoenas Prior to the New Policy
In December 1973 plaintiffs wrote AT&T demanding assurances that their toll-billing records — and those of other journalists— would not be released to government investigative agencies without prior notice to the journalists concerned.13 Plaintiffs also demanded from AT&T information concerning any past instances of Government access to their toll-billing records.14 Although AT&T did provide this latter information, it refused to give plaintiffs the assurances they demanded.
Finding this response inadequate, plaintiffs filed a complaint in the District Court for the District of Columbia on 27 December 1974, alleging that defendants’ policy regarding the release of toll-call records violated their First and Fourth Amendment rights.15 The complaint sought a judicial declaration that it was unlawful for defendants to release the toll-billing records of journalists to government investigative agencies without prior notification to the journalists concerned. In addition, it prayed for an injunction barring defendants from releasing plaintiffs’ records without such prior notice. The United States intervened as a party defendant.
*1039The record developed by plaintiffs in the course of a massive discovery effort reveals that from January 1971 to March 1974, before AT&T’s new policy went into effect, the Government issued approximately 75,-000-100,000 toll-record subpoenas.16 This reflects the importance of these records as a tool in the investigation of modern crime. The record further reveals, during this same period, only five instances in which the toll-call records of journalists were requested. This reflects the relative impact of this investigative method on journalists. Four of these cases involved grand jury subpoenas; one involved an Internal Revenue Service summons. In each of these five instances, the toll-call records were sought in connection with a felony investigation. One subpoena was issued in the course of an investigation into the suspected theft of Government documents and the receipt of stolen documents, violations of 18 U.S.C. § 641;17 three of the subpoenas were issued in furtherance of investigations into possible violations of the espionage laws of the United States, 18 U.S.C. §§ 792-798;18 and *1040one summons was issued in an investigation concerning possible unauthorized disclosure of income tax return information by an employee of the Internal Revenue Service in violation of 26 U.S.C. § 7213.19
D. Impact on Journalists of Toll-Record Subpoenas After the New Policy
Since AT&T adopted its new toll-record policy in March 1974 — over four years ago- — approximately 100,000-150,000 toll-record subpoenas have been issued by the Government in criminal investigations. The record reflects no instance during this period in which the toll records of a journalist have been sought. Moreover, there is no clear indication in the record that, under AT&T’s new policy, a journalist would not receive timely notice of any future subpoena directed at his records.
The dissent has attempted to expand the issues presented in this case by suggesting that plaintiffs’ toll records are presently subject to subpoena without prior notification in a broad range of civil and administrative proceedings. This is pure speculation, and the propriety of any such practice, if it does exist, is simply not raised in the case at bar. The record indicates only that certain plaintiffs have had their toll records subpoenaed in felony investigations. Indeed, we do not find any evidence in the record that it is Government practice to subpoena toll-call records in non-law enforcement matters. Not only does the record fail to reveal any instance — before or after AT&T’s new policy went into effect —in which a journalist’s toll records have been subpoenaed in connection with something other than a felony investigation, but it even fails to reveal an instance in which any person’s toll-call records have been subpoenaed in connection with a non-law enforcement proceeding. The record evidence strongly suggests that virtually all toll-call record subpoenas are issued in the course of felony investigations. While the dissent grandly enumerates 47 agencies that have subpoena power, the record shows that only four agencies have actually subpoenaed toll records — the Department of Justice, the Internal Revenue Service, the Securities Exchange Commission, and the Department of the Treasury — and these subpoenas appear to have been issued in furtherance of law enforcement investigations.20
*1041Thus, the central issue in this case is not whether plaintiffs are entitled to prior notice of subpoenas issued in civil or administrative proceedings, but rather whether plaintiffs are entitled to prior notice of subpoenas issued in the course of criminal investigations.
On 17 August 1976 the District Court granted defendants’ motion for summary judgment, rejecting plaintiffs’ First as well as their Fourth Amendment claims. This appeal followed.
II. STATE ACTION
Before proceeding to the merits of plaintiffs’ claims, we confront, in the case of AT&T, the threshold question of “state action.” Plaintiffs have maintained, and AT&T has vigorously denied,21 that the latter’s cooperation with the Government in unannounced releases of plaintiffs’ toll-billing records has so enmeshed AT&T in official action as to assimilate its own conduct to that of the Government. The problem ought not detain us, however, because resolution of the governmental-action issue is by no means essential to the outcome of this appeal or to progress of the case on remand. We see no occasion, then, to adjudicate the knotty constitutional question this aspect of the litigation tenders.
In several material respects, governmental action is unquestionably involved in the practice plaintiffs challenge. The Government, through its agencies, itself seeks the toll records; the Government itself shuns notice and judicial pre-clearance thereof; and the Government’s potential utilization of the. acquired records — not AT&T’s — is the consequence feared, by reporter and confidential source alike. Governmental action thus pervades any clandestine turnover to a degree more than ample to call for constitutional scrutiny.
It is evident, too, that AT&T’s role in this controversial scenario, however collaborative, is but complementary and secondary to that of the Government. As the Government is a defendant in the case, any relief that plaintiffs conceivably might secure can, and in our view should, be exacted from it. Were the propriety of advance notice of the demand for a particular reporter’s toll records ever established, certainly the Government could give it. Just as clearly, were ex parte judicial approval to become a prerequisite in any circumstances to release of the records, the Government necessarily would be the initiating party. And surely there is no impediment to enforcement against the Government of any narrower injunctive remedy to which plaintiffs may become entitled.22
Moreover, in the event that injunctive relief eventuates in this litigation, either as an original matter or to effectuate a declaratory judgment,23 an order directed against the Government will bind AT&T as well if only it is apprised of it. Federal Civil Rule 65(d) provides that
[ejvery order granting an injunction . is binding . . . upon the parties to the action, their officers, agents, servants, employees, and attorneys and upon those persons in active concert or participation with them who receive actual notice of the order by personal service or otherwise.24
*1042It is perfectly obvious that the Government cannot gain access to any toll-billing records save through AT&T. By silently delivering those records at the Government’s behest, AT&T is plainly in “active concert or participation” with the Government in the activity precipitating plaintiffs’ grievance.25 And if AT&T thus aided the breach of a decree enjoining the Government’s unannounced and unapproved acquisition of particular toll-billing records, it would subject itself to appropriate sanctions.26 It matters not that absent a cause of action against the Government itself AT&T might entirely avoid plaintiffs’ suit27 — a matter we do not address. AT&T’s amenability to the injunction would arise in consequence of a command properly leveled at the Government, and to assure that it will be truly meaningful.
Because, then, plaintiffs can secure directly from the Government any injunctive relief obtainable from AT&T and since obedience to any injunction running against the Government can be exacted from AT&T on a nonconstitutional basis,28 we perceive no compulsion to ascertain whether AT&T’s conduct is constitutionally the equivalent of governmental action for purposes of this case. We adhere, then, to the settled principle that courts are not to indulge in constitutional rulings absent “strict adjudicative necessity,” 29 which we are unable to discern here.30
III. PLAINTIFFS’ FOURTH AMENDMENT CLAIM
Plaintiffs contend that AT&T’s policy in certain felony cases of releasing toll-call records to investigators without prior notice to the subscriber violates their Fourth Amendment rights.31 This claim clearly lacks any merit.
The Fourth Amendment strikes a balance between the individual citizen’s interest in conducting certain affairs in private and the general public’s interest in subjecting possible criminal activity to intensive investigation. It strikes this balance by securing for each individual a private enclave — a “zone” bounded by the in*1043dividual’s own reasonable expectations of privacy.32 So long as the individual acts within this “zone of privacy”, his activities are shielded from unreasonable Government investigation;33 any attempt by the Government to search for or seize evidence from within this zone must be based on “probable cause” and usually must be preceded by a judicial determination that “probable cause” exists.34 However, the protections afforded the privacy interests of the individual by the Fourth Amendment are necessarily limited by the public’s interest in effective law enforcement; the Fourth Amendment does not insulate all personal activity from official scrutiny. Just as it creates “zones of privacy”, it also demarcates appropriate areas for investigation, in the sense that it allows the Government relatively free access to evidence located outside the individual’s “zone of privacy.” 35
Every individual must from time to time reach beyond his private enclave, draw other people into his activities, and expose his activities to public view.36 In any normal life, even in pursuing his most private purposes, the individual must occasionally transact business with other people. When he does so, he leaves behind, as evidence of his activity, the records and recollections of others. He cannot expect that these activities are his private affair. To the extent an individual knowingly exposes his activities to third parties, he surrenders Fourth Amendment protections, and, if the Government is subsequently called upon to investigate his activities for possible violations of the law, it is free to seek out these third parties, to inspect their records, and to probe their recollections for evidence 37
In a sense, then, the Fourth Amendment carries with it both a promise and a warning. It promises each individual that there is a zone in which he may conduct his affairs in private, shielded from unwarranted investigative scrutiny, and yet it warns each individual that, once he projects his activities beyond this private enclave, the Government is free to scrutinize them for law enforcement purpose.
These principles, consistently adhered to by the Supreme Court, are the basis for the well-settled rule that a person has no expectation of privacy in the busi*1044ness records of a third party and, therefore, has no interest protected by the Fourth Amendment in such records.38 Accordingly, it has been consistently held by the Supreme Court39 and the Courts of Appeals40 that a person has no Fourth Amendment basis for challenging subpoenas directed at the business records of a third party, and, hence, has no right to notice of such subpoenas.
In the most recent Supreme Court case of United States v. Miller;41 the Treasury Department presented to banks at which respondent maintained accounts grand jury subpoenas requiring the production of “all records of [respondent’s] accounts.” The banks promptly complied, though no notice of the subpoenas had been given respondent. Upon being charged with conspiracy to defraud the United States of tax revenues, respondent made a pre-trial motion to suppress the bank documents, urging that the subpoenas were defective. The District Court denied respondent’s motion but was reversed by the Court of Appeals which *1045held that the Government had violated respondents’ Fourth Amendment rights. The Supreme Court reversed the Court of Appeals, holding that the District Court had properly denied respondent’s motion to suppress since respondent possessed no Fourth Amendment interest in the bank documents.42 The Court further concluded that, since respondent had no interest that could be vindicated by a challenge to the subpoena, respondent had no right to notice of the subpoenas.43
The Court’s decision highlights the principle that “what a person knowingly exposes to the public ... is not a subject of Fourth Amendment protection”: 44
Respondent urges that he has a Fourth Amendment interest in the records kept by the banks because they are merely copies of personal records that were made available to the banks for a limited purpose and in which he has a reasonable expectation of privacy . . . . But in Katz the Court . . . stressed that “[wjhat a person knowingly exposes to the public ... is not a subject of Fourth Amendment protection.”
Even if we direct our attention to the original checks and deposit slips, rather than to the microfilm copies actually viewed and obtained by means of the subpoena, we perceive no legitimate “expectation of privacy” in their contents.
The depositor takes the risk, in revealing his affairs to another, that the information will be conveyed by that person to the government. This Court has held repeatedly that the Fourth Amendment does not prohibit the obtaining of information revealed to a third party and conveyed by him to government authorities, even if the information is revealed on the assumption that it will be used only for a limited purpose and the confidence placed in the third party will not be betrayed.
Since no Fourth Amendment interests of the depositor are implicated here, this case is governed by the general rule that the issuance of a subpoena to a third party to obtain the records of that party does not violate the rights of a defendant, even if a criminal prosecution is contemplated at the time the subpoena is issued.45
The general rule that a person has no Fourth Amendment basis for challenging subpoenas directed at the business records of third parties has been directly applied to toll-billing records maintained by telephone companies.46 A telephone subscriber is fully aware when he places a long distance call that the telephone company will make a record of the call, that the record is the company’s property, and that the Government has ready access to the record for law enforcement purposes. On this basis the courts have uniformly held that subscribers have no Fourth Amendment basis for challenging Government inspection of their toll records, since subscribers, like bank depositors, have taken the risk in revealing their affairs to third parties that the information will be conveyed by that person to law enforcement officials, either voluntarily or in response to compulsory process.
Accordingly, in upholding a felony conviction based in part on evidence derived from toll-call records, the Ninth Circuit in United States v. Fithian 47 stated:
The introduction into evidence of telephone company records of calls from appellant’s residence did not violate his Fourth Amendment rights. The expectation of privacy attaching to telephone conversations relates to the content of *1046the conversations themselves and not to the fact that a conversation took place. No one justifiably could expect that the fact that a particular call was placed will remain his private affair when business records necessarily must contain this information.48
The courts have invariably taken this position.49 These precedents soundly dispose of plaintiffs’ Fourth Amendment claim.
IV. PLAINTIFFS’ FIRST AMENDMENT CLAIMS50
Plaintiffs contend that, as journalists, they are entitled under the First Amendment to prior notice of toll-call-record subpoenas issued in the course of felony investigations, even if citizens in general have no such right.51 Plaintiffs base this claim on two separate and distinct theories.
The first theory relates to the impact of good faith toll-call-record subpoenas on plaintiffs’ First Amendment rights and the need for judicial balancing before such records are released to Government investigators. Plaintiffs develop this theory as follows:
(1) The First Amendment guarantees journalists the freedom to gather information from clandestine sources.
(2) Because toll-call records may disclose the identity of a clandestine source, this *1047freedom is abridged whenever the Government gains access to a journalist’s toll records, even where access is gained in the course of a good faith felony investigation.
(3) In order to determine whether this infringement on First Amendment rights is justified, the Government’s investigation “interests” must be judicially balanced in each case against the journalist’s First Amendment “interests”.
(4) A journalist, therefore, must receive prior notice of a toll-record subpoena so that he may challenge the subpoena and thus prompt the requisite judicial balancing before the records are released.
The second theory relates to the impact of bad faith toll-call-record subpoenas on plaintiffs’ First Amendment rights and the need for judicial screening before such records are released to Government investigators. Plaintiffs develop this theory as follows:
(1) Government investigators have in the past subpoenaed journalists’ toll-call records, not in furtherance of good faith felony investigations, but as part of politically-motivated efforts to interfere with their news-gathering activities.
(2) Such bad faith subpoenas abridge journalists’ First Amendment right to gather information from clandestine sources.
(3) In order to protect journalists from such infringements in the future, it is necessary for the judiciary to superintend investigations of journalists in order to screen out bad faith subpoenas.
(4) A journalist, therefore, must receive prior notice of a toll-call-record subpoena so that he may challenge the subpoena and thus prompt the requisite judicial screening before the records are released.
Common to both theories is the proposition that journalists have a right under the First Amendment to gather information from clandestine sources. However, there is a clear distinction between the two theories, regarding both the type of Government action which supposedly abridges this right and the kind of judicial action necessary to remedy the abridgment. According to the first theory, plaintiffs’ rights are abridged by any and all toll-call-record subpoenas, and the appropriate remedy is a judicial balancing of legitimate interests of the Government against those of the journalist. According to the second theory, plaintiffs’ rights are abridged by bad faith toll-call-record subpoenas and the appropriate remedy is a judicial screening of all subpoenas in order to winnow those issued for illegitimate reasons.52 Each theory involves separate and distinct issues.
Unfortunately, plaintiffs have quite confounded the two, repeatedly carrying over considerations relevant to one in an effort to bolster the other. This totally ignores the principle the Supreme Court specifically noted in Branzburg v. Hayes53 that First Amendment challenges to good faith investigative action and First Amendment challenges to bad faith investigative action “pose wholly different issues for resolution.”54 This opinion follows the Supreme Court, analyzing each of plaintiffs’ theories separately.
Before proceeding, however, it is important to note that the dissent apparently does not grasp the significance of this bifurcated analysis. It suggests that all it is required to do here is to declare that plaintiffs are entitled to prior notice of toll-record subpoenas. Apparently, it believes that it can make this pronouncement in vacuo, without addressing whether or not there is actually a need for case-by-case judicial balancing or screening. In the dissent’s view, these latter issues are not immediately rele*1048vant and can be left to future resolution. Yet, this is placing the cart before the horse. The existence of a right to prior notice is predicated precisely on the supposed need for case-by-case judicial balancing or screening. Prior notice is merely a procedural device, the only function of which is to trigger judicial supervision of subpoenas — supervision that necessarily involves either the balancing or screening functions of the court.
Accordingly, if it can be demonstrated that there is no need for case-by-case balancing and no warrant for case-by-case screening, then the rationale for providing prior notice evanesces. We conclude first, that there is no need for case-by-case balancing in criminal investigations and, second, that no justification for case-by-case screening has been made thus far in this case.
A. Plaintiffs’ “Balancing” Theory
In treating plaintiffs’ first theory, two issues must be addressed: first, whether Government access to toll-call records in the course of a good faith felony investigation actually “abridges ” a “freedom ” guaranteed plaintiffs under the First Amendment; and second, if so, whether prior judicial balancing on a case-by-case basis is an appropriate judicial response.
1. No Abridgment of Any Special Journalistic Right or Privilege by Good Faith Subpoenas
a. Relationship of First Amendment to Good Faith Investigation
Plaintiffs contend that the First Amendment guarantees “journalists” the right to gather information from secret sources, and that this right implies a further right to maintain the secrecy of those sources.55 They assert that these rights are abridged whenever the Government gains access to toll-call records, or presumably to any other third-party information which might disclose the identity of their secret sources.56 Thus, plaintiffs, who have neither a property nor a privacy interest in the business records of the defendant telephone companies,57 claim nevertheless to have a “First Amendment interest” in these records. This “interest”, according to plaintiffs, entitles them to bar Government access to these records even during good faith felony investigations.
The enormity of this claim can best be comprehended by considering the following fact situation which, though hypothetical, is patterned after the five actual instances of toll-call-record subpoenas presented in this case.
Suppose that an employee in the Internal Revenue Service’s regional office in Atlanta, Georgia, decides that he is going to make public embarrassing information from the income tax return of a highly prominent citizen. He knows that it is a felony for an IRS employee to disclose tax return information to an unauthorized person,58 and that it is also a felony for any person to whom such information has been disclosed thereafter to publish it.59 Therefore, the employee naturally desires to avoid detection and plans to act cautiously. He places a long distance telephone call to a journalist in Washington, D. C., informing him of his designs, and the journalist says he will consider the matter. Later, the journalist calls the employee and sets up a meeting in Washington. On the agreed date, the employee takes a plane to Washington, and registers in a local hotel. Later, he takes a taxicab from the hotel to a rendezvous at the journalist’s own home. He is seen by *1049one of the journalist’s neighbors as he enters the journalist’s home. During his meeting with the journalist, the employee asks for and receives the journalist’s assurances that he will not disclose the employee’s identity even under compulsion. On this basis, the employee delivers the information to the journalist who subsequently publishes it.
When the information appears in the journalist’s column, it is obvious to Government law enforcement officials that at least one and possibly two felonies have been committed, and appropriate officers set about investigating the suspected crimes. From the nature of the published information itself, the investigators ascertain that the unauthorized disclosure was made from the Atlanta office. The investigatory problem becomes linking an employee at that office with the Washington journalist.
There are two ways in which this can be done. The most direct way is to have the grand jury subpoena the journalist and compel him to disclose his source. The Supreme Court in Branzburg v. Hayes60 expressly held that this may be done — that the journalist may not refuse to disclose his source. According to the Court, the journalist may be required to testify in any and all good faith criminal investigations —there is no case-by-case consideration given to a claim of privilege. Good faith investigation interests always override a journalist’s interest in protecting his source.61
There is also a more indirect way for the investigators to proceed. They can seek out third-party information connecting the journalist with an employee in the Atlanta office. Here, there are at least five sets of third-party business records which, to varying degrees, provide evidence of such a link: (1) the journalist’s toll-call records, (2) the toll-call records of employees in the Atlanta office, (3) the airline records for Atlanta to Washington flights, (4) the Washington hotel records, and (5) the taxicab company’s records. There are also at least two witnesses whose personal recollections .may provide evidence of such a link: (1) the taxicab driver, and (2) the journalist’s neighbor.
Plaintiffs’ position is that the journalist in this hypothetical situation has “First Amendment interests” in these five sets of business records and the testimony of these two witnesses, and that his “interests” may outweigh the Government’s need for this evidence. Thus, according to plaintiffs the Government would be required to notify the journalist before it could seek access to any of these records or witnesses, in order to give the journalist the opportunity to vindicate his supposed First Amendment right to maintain the secrecy of his sources. In short, plaintiffs claim that journalists have the unprecedented privilege of suppressing the records and testimony of third parties to whom they and their sources have carelessly revealed incriminating information.
In our view, plaintiffs’ position is based on erroneous propositions. First, the so-called right of journalists to gather information from secret sources does not include a right to maintain the secrecy of sources in the face of good faith felony investigations. Second, Government access to third-party evidence in the course of a good faith felony investigation in no sense “abridges” plaintiffs information-gathering activities.
Branzburg v. Hayes is dispositive on the first point. In Branzburg journalists had established confidential relationships with certain sources. Through grand jury subpoenas ad testificandum, the Government sought to compel the journalists to betray these confidences and to disclose the identity of their sources. The journalists moved to quash the subpoenas, contending that the First Amendment gave rise to a testimonial privilege, that is, a right to protect the secrecy of their sources. The journalists contended, just as plaintiffs here contend, that if journalists were compelled to identi*1050fy their sources, informants would refuse to furnish information in the future, and that this would interfere with the free flow of information protected by the First Amendment.
The Court expressly rejected this claim, holding that journalists had no special First Amendment right to maintain the secrecy of their sources in the face of a good faith felony investigation.62 The Court found that the possibility that a source might refuse or be reluctant to furnish information to a journalist out of fear that his identity might be revealed was at best a “burden” on the First Amendment right to gather news.63 The Court further stated that “[i]t is clear that the First Amendment does not invalidate every incidental burdening of the press . . .,” 64 and that the “burden” of disclosing the identity of a source was not constitutionally suspect and therefore did not require a privileged position for newsmen.65
Branzburg involved good faith subpoenas ad testificandum directed at the journalists themselves. It is logically inescapable that if, as held, journalists have no right to resist such subpoenas, then they certainly have no right to resist good faith subpoenas duces tecum directed at a third-party’s business records. Not only is the logic inescapable, but, in fact, the Court in Branzburg did specifically uphold the propriety of identifying sources through such independent means. The Court stated:
[I]f the authorities independently identify the informant, neither his own reluctance to testify nor the objection of the newsman would shield him ., whatever the impact on the flow of news or on his future usefulness as a secret informant.66
Thus, both logic and the Court’s plain statement clearly refute plaintiffs’ claim that “journalists” have a special right to resist good faith subpoenas directed at third-party records.
There is another reason that plaintiffs’ reliance on Branzburg v. Hayes, for the principle that journalists have a right to protect the secrecy of their sources from identification by third parties, is so ironic. The journalists in Branzburg did not assert an absolute testimonial privilege, but asserted, inter alia, that the Government should be required to seek out third party sources of evidence before being permitted to compel the journalists themselves to identify their sources. In other words, the journalists in Branzburg sought protection only from being themselves compelled to disclose their sources; they did not claim any right to prevent third parties from identifying their sources; indeed, they insisted that the Government be forced to seek out third parties from whom they could obtain the necessary evidence. It is curious, then, that plaintiffs here rely on Branzburg for the proposition that the First Amendment entitles them to reach out and suppress the testimony of third parties whom they have injudiciously made witting of their secrets.67
*1051It is thus clear from Branzburg and related cases that the freedom to gather information guaranteed by the First Amendment is the freedom to gather information subject to the general and incidental burdens that arise from good faith enforcement of otherwise valid criminal and civil laws that are not themselves solely directed at curtailing the free flow of information.68 The broad scope of acceptable government investigation, so necessary to the secure enjoyment of all liberties, unavoidably places a burden on all citizens. It is difficult, though not impossible, to establish absolutely secret contacts with other people. The freedom that “journalists” enjoy with respect to their news gathering is subject to this burden. The First Amendment does not guarantee plaintiff “journalists,” or other citizens, a special right to immunize themselves from good faith investigation simply because they may be engaged in gathering information. Thus, the Government’s good faith inspection of defendant *1052telephone companies’ toll call records does not infringe on plaintiffs’ First Amendment rights, because that Amendment guarantees no freedom from such investigation.
Moreover, it is clear that Government access to defendants’ toll-call records in no sense “abridges” plaintiffs’ news-gathering activities within the meaning of the First Amendment. Not every Government action that affects, has an impact on, or indeed inhibits First Amendment activity constitutes the kind of “abridgment” condemned by the First Amendment. Historically considered, freedom of the press means primarily, although not exclusively, immunity from prior restraints or censorship,69 but the guarantee also affords protection from the imposition of post-publication sanctions and punishments.70 Additionally, in recent years, the Supreme Court has found in a number of cases that constitutional violations may arise from the deterrent, or “chilling”, effect of governmental action that falls short of a direct prohibition against the exercise of First Amendment rights.71 Yet not every Government action that has an inhibiting or constrictive impact on' First Amendment activity is said therefore to have an impermissible “chilling effect.” The constrictive impact must arise from th(e present or future exercise, or threatened exercise, of coercive power.72
As the Supreme Court stated in Laird v. Tatum:73
In none of these [previous “chilling effect”] cases, however, did the chilling effect arise merely from the individual’s knowledge that a governmental agency was engaged in certain activities or from the individual’s concomitant fear that, armed with the fruits of those activities, the agency might in the future take some other and additional action detrimental to that individual. Rather, in each of these cases, the challenged exercise of governmental power was regulatory, proscriptive, or compulsory in nature, and the complainant was either presently or prospectively subject to the regulations, proscriptions, or compulsions that he was challenging.
In the instant case, the Government action challenged by plaintiffs is the mere gathering of evidence from third parties. As the Supreme Court expressly stated in Branzburg, such investigation imposes no restrictions, prior restraint, or sanctions on journalists’ information-gathering.74 Furthermore, in this case, unlike Branzburg, there is no effort to compel the journalists themselves to disclose the identity of their sources. Plaintiffs in this case are neither presently or prospectively subject to “regulation, proscription or compulsion” as a result of the Government action they here challenge. Consequently, it cannot be asserted that Government access to defendants’ toll-call records “abridges” plaintiffs’ First Amendment news-gathering rights.
*1053In sum, then, the two propositions underlying plaintiffs’ first theory are erroneous. The First Amendment does not guarantee journalists the right to preserve the secrecy of their sources in the face of good faith criminal investigation. Moreover, Government inspection of third-party records, while it may inhibit plaintiffs’ news-gathering activity, does not impermissibly abridge such activity. Hence, journalists in this context have no “First Amendment interest” in third-party records which disclose the identity of a secret source and, consequently, have no First Amendment right to notice of subpoenas directed at such records.75 In other words, the First Amendment does not guarantee a journalist, or any other citizen, the freedom to collect information immune from good faith criminal investigation by means which accord with Fourth and Fifth Amendment protections. Nor does it insulate a journalist, or any other citizen, from the general and subjective inhibitions that naturally arise from the prospect of such investigation — inhibitions to which all citizens are subject in every field of endeavor.
b. Relationship of First, Fourth, and Fifth Amendments76
The issues raised by plaintiffs’ first theory may be viewed from a somewhat different perspective. I believe that plaintiffs’ claims pose fundamental questions concerning the interplay of First, Fourth and Fifth Amendment protections. Essentially, plaintiffs contend that particular good faith investigative actions inhibit, and thereby infringe upon, their information-gathering activities. Specifically, plaintiffs complain that the investigative action has “intruded *1054into”, “invaded” or “penetrated” activity which is secret or private.
As already demonstrated, to the extent plaintiffs have exposed their activities to third parties they had no protectable privacy interests in them under the Fourth Amendment.77 The crux of plaintiffs’ argument, then, is that they have a right to establish relationships in secret under circumstances in which the Fourth Amendment ensures no secrecy and that they have a right to engage in furtive action free from investigative scrutiny within an arena ordinarily open to such scrutiny. In short, plaintiffs claim that the zone of privacy guaranteed them by the Fourth and Fifth Amendments does not give them adequate privacy within which to conduct clandestine collection and that they, apart from other citizens, must have a broader zone of privacy-
The question thus presented is this: Does the First Amendment afford plaintiffs this extra margin of privacy by imposing substantive or procedural limitations on good faith criminal investigative action above and beyond the limitations imposed by the Fourth and Fifth Amendments? I believe the answer must be no.
The Supreme Court has repeatedly emphasized that one of the main reasons for adoption of the Fourth Amendment was to provide citizens with the privacy protection necessary for secure enjoyment of First Amendment liberties.78 First Amendment values permeate the Fourth Amendment. To a somewhat lesser extent, the same can be said of the Fifth Amendment.79 In my view, the guarantees of the Fourth and Fifth Amendments achieve their purpose and provide every individual with sufficient protection against good faith investigative action for the full enjoyment of his First Amendment rights of expression. To the extent an individual insists that he must shield himself from the prospect of good faith investigation and operate in secrecy in order to exercise effectively particular First Amendment liberties, he must find that shield and establish that secrecy within the framework of Fourth and Fifth Amendment protections. This is not to say that the First Amendment never gives rise to any privacy-type interests apart from those secured by the Fourth and Fifth Amendments. It does mean, however, that such interests are overridden in criminal cases by the public’s interest in effective law enforcement investigation at least insofar as they go beyond protections already afforded by the Fourth and Fifth Amendments.
In several cases the Supreme Court has dealt with situations in which law enforcement investigative action undertaken in good faith has had an impact on First Amendment activity.80 In these cases involving “a convergence of First and Fourth Amendment values”, the Court has held that First Amendment interests are to be safeguarded by strict adherence to Fourth Amendment standards.81 Significantly, the Court has not concluded that the First Amendment interests themselves give rise to additional protections against good faith investigative action above and beyond those afforded by the Fourth Amendment.82 Nor *1055has the Court concluded that persons engaging in certain types of First Amendment activity have broader Fourth Amendment rights than other citizens.
The proposition that the First Amendment offers no procedural or substantive protections against good faith criminal investigative activity beyond that afforded by the Fourth and Fifth Amendments is directly supported by the Supreme Court’s recent decision in Zurcher v. Stanford Daily83 Indeed, this seems to be the tacit basis of the decision. In that case police had searched the offices of the Stanford Daily newspaper pursuant to a warrant issued on probable cause to believe that there were photographs on the premises that would identify the perpetrators of certain felonies. The members of the Daily staff were not themselves involved in the criminal activity under investigation. After the search, the Daily and members of its staff brought a civil action seeking declaratory and injunctive relief under 42 U.S.C. § 1983 against the police. The complaint alleged that the search of the Daily’s office had deprived the newsmen under color of state law of rights secured them by the First, Fourth, and Fourteenth Amendments. The journalists asserted, inter alia, that where an innocent object of a third-party search is a newspaper, the First Amendment requires that evidence in the newspaper’s possession be obtained by a subpoena duces tecum rather than by a search warrant unless the police make a clear showing that the evidence would be destroyed. The district court granted the journalists declaratory relief, and the court of appeals affirmed. The Supreme Court reversed.
Significantly, the Supreme Court’s decision analyzes the protections available to the journalists entirely within the framework of the Fourth Amendment. Observing that the Fourth Amendment was adopted largely in response to a history of conflict between the Crown and the press, the Court reaffirmed the principle that “[wjhere the materials sought to be seized may be protected by the First Amendment, the requirements of the Fourth Amendment must be applied with ‘scrupulous ex*1056actitude’ ”.84 Thus, the Court held that the protections to be afforded the journalists were to be found in rigorous application of Fourth Amendment standards. Implicit in this decision is the conclusion that, in this type of case, the existence of First Amendment “interests” does not give rise to any substantive or procedural protections above and beyond those afforded by the Fourth Amendment. The Court stated:
Neither the Fourth Amendment nor the cases requiring consideration of First Amendment values in issuing search warrants, however, call for imposing the regime ordered by the District Court. Aware of the long struggle between Crown and press and desiring to curb unjustified official intrusions, the Framers took the enormously important step of subjecting searches to the test of reasonableness and to the general rule requiring search warrants issued by neutral magistrates. They nevertheless did not forbid warrants where the press was involved, did not require special showings that subpoenas would be impractical, and did not insist that the owner of the place to be searched, if connected with the press, must be shown to be implicated in the offense being investigated. Further, the prior cases do no more than insist that the courts apply the warrant requirements with particular exactitude when First Amendment interests would be endangered by the search.85
Justice Powell’s concurring opinion is even more explicit:
If the Framers had believed that the press was entitled to a special procedure, not available to others, when government authorities required evidence in its possession, one would have expected the terms of the Fourth Amendment to reflect that belief. As the opinion of the Court points out, the struggle from which the Fourth Amendment emerged was that between Crown and press. The Framers were painfully aware of that history, and their response to it was the Fourth Amendment. Hence, there is every reason to believe that the usual procedures contemplated by the Fourth Amendment do indeed apply to the press, as to every other person.
. [Considerations such as these are the province of the Fourth Amendment. There is no authority either in history or in the Constitution itself for exempting certain classes of persons or entities from its reach.86
Circuit court cases also support the proposition that the First Amendment affords no procedural or substantive protection against good faith criminal investigative activity beyond that afforded by the Fourth and Fifth Amendments. Most analogous to the case at hand are the so-called “mail cover” cases. A “mail cover” is an investigative technique involving the screening and analysis of the envelope exteriors of all mail addressed to a particular recipient. By this means, investigators can determine, for each item of mail, the city from which the mail was sent, the date on which it was sent, and, if a return address is provided, the name and address of the sender. No warrants are required to initiate a “mail cover”. Thus, “mail covers” provide essentially the same information concerning mail communication as toll-call record inspections provide concerning long distance telephone communication. Criminal defendants have attempted to suppress evidence derived from mail covers on the grounds that such warrantless mail surveillance violated their Fourth Amendment rights.
The courts have uniformly rejected this argument.87 Just as courts have held *1057that a person’s expectations of privacy concerning telephone communication attach only to the contents of the conversation and not to the fact the communication was made, so also the courts have held that a person may reasonably expect privacy only with respect to the contents of an envelope and not with respect to information knowingly exposed to third parties on the envelope’s exterior. Anticipating plaintiffs’ argument in this case, individuals have argued that, even though “mail covers” may not violate the Fourth Amendment, they nevertheless violate the First Amendment by “chilling” free communication through the mails. There is no doubt that communication through the mails is a First Amendment right guaranteed by both the free speech and press guarantees88 and is “almost as much a part of free speech as the right to use our tongues.”89 Nevertheless, the courts have held that “mail covers” violate no First Amendment right.90
The principle that the First Amendment offers no procedural or substantive protection from good faith criminal investigation beyond that afforded by the Fourth and Fifth Amendments is also reflected in cases involving informants and undercover agents.91 Law enforcement investigators regularly collect information from informants and undercover agents concerning the activities, conversations, and associations of individuals and groups.92 No warrant is required when Government investigators solicit information from an informant or place an undercover agent. This practice was challenged in a series of Supreme Court cases. In Lopez v. United States,93 defendant in a bribery trial objected to the introduction into evidence of tape recordings of an incriminating conversation between himself and an IRS agent in defendant’s private office. The recordings were made by a tape recorder in the agent’s pocket. Justice Brennan argued, in dissent, that such covert investigative practices threatened First Amendment liberties: “[Tjhere is a grave danger of chilling all private, free, and unconstrained communication if secret recordings, turned over to law enforcement officers by one party to a conversation, are competent evidence of any self-incriminating statements the speaker may have made.”94 Implicitly rejecting this position, the majority held that the recordings of defendant’s conversation were properly admitted into evidence and that the secret recording of defendant did not violate defendant’s constitutional rights.
Similarly, in United States v. White,95 Justice Douglas argued that a criminal suspect’s First Amendment rights were violated when his conversations with an informant were electronically monitored by *1058Government agents through a transmitter installed on the informant:
Monitoring, if prevalent, certainly kills free discourse and spontaneous utterances. Free discourse — a First Amendment value — may be frivolous or serious, humble or defiant, reactionary or revolutionary, profane or in good taste; but it is not free if there is surveillance.
. [M]ust everyone live in fear that every word he speaks may be transmitted or recorded and later repeated to the entire world? I can imagine nothing that has a more chilling effect on people speaking their minds and expressing their views on important matters.96
The Court implicitly rejected this First Amendment argument in upholding the constitutionality of employing “wired” informants,97 stating: “Inescapably, one contemplating illegal activities must realize and risk that his companions may be reporting to the police.”98
Finally, cases involving physical surveillance support the principle that the First Amendment affords no protection against good faith criminal investigative activity beyond that afforded by the Fourth and Fifth Amendments. Courts have held that physical surveillance consistent with Fourth Amendment protections and in connection with a bona fide law enforcement investigation does not violate First Amendment rights, even though it may be directed at communicative or associative activities and even though it may inhibit such activities.99 Thus, in Oaks v. United States,100 the Ninth Circuit upheld the conviction of a leader of a tax rebellion group for failure to file an income tax return, rejecting defendant’s contention that IRS surveillance had violated his First Amendment rights. IRS undercover agents had infiltrated the Tax Rebellion Committee of which defendant was a member, attended various meetings of the Committee, and submitted reports on the activities of the Committee and its members, including defendant. The Court stated:
Nor do we find that the surveillance of the meetings of the Tax Rebellion Committee by the undercover agents was improper. The meetings were open to the public and the participants were openly advocating the willful violation of Internal Revenue laws. The risk of surveillance of meetings of this type must be assumed. No interest legitimately protected by the First and Fifth Amendments is involved.101
The principle is clear. To the extent individuals desire to exercise their First Amendment rights in private, free from possible good faith law enforcement investigation, they must operate within the zone of privacy secured by the Fourth Amendment. When individuals expose their activities to third parties, they similarly expose these activities to possible Government scrutiny. The mere prospect that such investigation may occur or, indeed, the actual conduct of such investigation does not “chill” or otherwise abridge First Amendment rights, even though it may give rise to *1059subjective inhibitions for those who desire to avoid the prospect of investigation altogether.
Plaintiffs in this case insist that the mere fact of a relationship between themselves and certain sources must be secret. If this is so, then plaintiffs and their sources have the burden of establishing that secrecy within the limits of Fourth and Fifth Amendment protections. This may be difficult, but it is not impossible. All citizens when they choose to act surreptitiously are put to some inconvenience; they must lower their voices, put little or nothing in writing, speak and meet outside the presence of third parties; these simple precautions are commonly used, whether the ends be lofty or evil. Plaintiffs are no exception simply because they are “journalists.”
The plaintiffs’ and the dissent’s position is that the First Amendment does afford additional protections against good faith criminal investigation beyond those provided by the Fourth and Fifth Amendment. In skeletal form, their position is that when particular investigative action “implicates” First Amendment activities, then its utility must be judicially balanced against the First Amendment “interests” it may affect before it may be put into effect. However, it is clear from a moment’s reflection that this position is wholly untenable.
In Katz v. United States, the Supreme Court observed that all investigative action intrudes upon privacy to some extent.102 The same can be said for the relationship between investigation and First Amendment activity; that is, all investigative action affects or “implicates” First Amendment activity. After all, the First Amendment is not the personal preserve of “journalists.” It covers almost all forms of expression; it covers associative activities; it covers religious activities. Each citizen has First Amendment rights and each one’s rights are as precious as the other’s. Every time law enforcement officers arrest an individual, they place restraints on First Amendment activity. Does this mean that before an arrest can be made there must be a hearing at which the State’s “interest” in making the arrest is judicially balanced against the consequent inhibitions on First Amendment rights? Each time law enforcement officers place a suspect under physical surveillance or make inquiries and collect information on a suspect, they “implicate” First Amendment rights. Does this mean that before the police can take this fundamental investigative action there must be a hearing at which the particular utility of these acts is judicially balanced against the First Amendment “interests” they may affect? Each time the Government solicits information from an informant or places an undercover agent it “implicates” First Amendment rights. Does this mean that before such action is taken there must be a hearing at which the Government’s “interest” in taking these actions is judicially balanced against the First Amendment activities “implicated”? And each time the police obtain a warrant for more intrusive investigation, such as for wiretaps or searches, First Amendment rights are “implicated.” Does this mean that in addition to determining the probable cause required by the Fourth Amendment the issuing magistrate must also balance First Amendment interests? Of course, the practical consequence of such a regime would be the complete and absolute stultification of law enforcement.
If this is indeed what the First Amendment means, then I am surprised we have not heard of it sooner. Think of all those unfortunates in prison who would be free men today if the dissent had its way and if they but had the foresight to anticipate these plaintiffs and base their evidentiary challenges on First rather than Fourth Amendment grounds.
The only way in which plaintiffs and the dissent could avoid this denouement is to limit these First Amendment protections only to some individuals; or perhaps only to *1060some species of First Amendment activities; or perhaps only to some individuals in some activities. In other words, under plaintiffs' approach the courts would have to decide that certain individuals’ First Amendment rights are more important than those of others, and that certain First Amendment activities are more important than others. Presumably, the more significant individuals and the more significant activities would have greater protection from good faith investigation.
How would this gradation be made without doing violence to the First Amendment itself? If the First Amendment protects “journalists” from third-party subpoenas, how about ministers, priests and rabbis? How about authors, teachers, social workers and publishers? How about labor unionites, Democrats, Republicans and sorority sisters? If the First Amendment shields “news gathering” from good faith investigation, how about other kinds of First Amendment activity? How about all forms of associational activity? Are the associational activities of organized crime shielded from good faith investigation? If the police issue third-party subpoenas for information bearing on the “membership” of a suspected organized crime “family,” must prior notice be given the “family” so that it can challenge the subpoena in order to preserve the “confidentiality” of its members? Why dismiss this out of hand? Are there no genuine associational rights “implicated” in such a case?
In Branzburg, the Supreme Court recognized that it would be improper, as well as impracticable, to define the categories of persons entitled to insist on the balancing procedures sought in that case:
Sooner or later, it would be necessary to define those categories of newsmen who qualified for the privilege, a questionable procedure in light of the traditional doctrine that liberty of the press is the right of the lonely pamphleteer who uses carbon paper or a mimeograph just as much as of the large metropolitan publisher who utilizes the latest photo-composition methods. . . . The informative function asserted by representatives of the organized press in the present cases is also performed by lecturers, political pollsters, novelists, academic researchers, and dramatists. Almost any author may quite accurately assert that he is contributing to the flow of information to the public, that he relies on confidential sources of information, and that these sources will be silenced if he is forced to make disclosures before a grand jury.103
In sum, the approach urged by plaintiffs and the dissent must inevitably result in one of two consequences. Either ail people at all times will have the First Amendment balancing protection against good faith investigation, in which ease law enforcement will be completely throttled. Or, certain people at certain times will be entitled to such protection, in which event first the police and then the courts will be obliged in each case to delineate the individuals and species of activity entitled to protection, and this, in turn, will necessarily require distinguishing “real” from “sham” First Amendment claims. In my view, neither approach is acceptable, in theory or in practice.
2. Case-by-Gase Judicial Balancing — an Unprecedented, Unnecessary, and Unworkable Remedy
Even if it is assumed arguendo that plaintiffs have established an abridgment of a First Amendment “interest,” this does not necessarily mean that the judicial response sought by plaintiffs is appropriate. Plaintiffs contend that there must be a case-by-case balancing of the Government’s investigative interests against their First Amendment interests as journalists whenever the Government seeks access to their toll-call records in the course of a good faith felony investigation.104 We know of no authority supporting the proposition that any such balancing must be performed on a *1061case-by-case basis, and neither plaintiffs nor the dissent cites any.105
It is at this point that plaintiffs’ and the dissent’s confusion between the balancing function of the court and its screening function becomes evident. Apparently the source of this confusion is a single paragraph in the Branzburg decision. The Court in Branzburg determined that good faith criminal investigation interests always override a journalist’s interest in preserving the secrecy of his sources. The Court observed that there was no question but that the particular subpoenas in question were issued in good faith and it held that the First Amendment offers no protection from such good faith investigative efforts. The Court explicitly rejected a case-by-case balancing approach, which absolutely refutes the plaintiffs’ and the dissent’s position on good faith investigations. At the very end of its decision, the Court commented on the possible impact of bad faith subpoenas on journalists’ First Amendment rights:
Finally, as we have earlier indicated, news gathering is not without its First Amendment protections, and grand jury investigations if instituted or conducted other than in good faith, would pose wholly different issues for resolution under the First Amendment. Official harassment of the press undertaken not for purposes of law enforcement but to disrupt a reporter’s relationship with his news sources would have no justification.106
By these comments, the Court was merely indicating that journalists would have an effective remedy if bad faith harassing subpoenas were employed against them. The remedy suggested was a screening remedy, and it was available on a case-by-case basis because the Court would necessarily be called upon to enforce the subpoenas on a case-by-case basis. Thus, the Court did not contemplate case-by-case balancing at all, for if subpoenas were issued in bad faith, there would be no legitimate Government interest to balance against the journalist’s interests; rather it would be a question of screening out bad faith subpoenas.107
*1062The competing interests assertedly at stake in this case can easily be balanced and resolved once and for all. Indeed, if one reads Branzburg, as plaintiffs do, as involving the balancing of interests virtually identical to those here, then it is significant that the Court balanced these interests but once. The question presented in the instant case is simply whether the public interest in ensuring the effectiveness of good faith felony investigations is sufficient to override the burden on newsgathering which is said to result from permitting Government inspection of toll-call records which might disclose the identity of journalists’ secret sources. In our view, Branzburg dictates an outcome in favor of law-enforcement “interests.”
In the first place, the burden placed on journalists by Branzburg is substantially more onerous than the burden that would be placed on them by being subject to toll-call-record inspections. There is no way for journalists to circumvent the rigors of Branzburg. They can take every secrecy precaution and be as discrete as possible, and still be directly compelled to disclose their secrets. In contrast, the real burden that the possibility of toll-call-record inspections places on journalists is mere inconvenience. Journalists and sources can easily take steps to avoid detection. It is therefore highly questionable that existing accessibility of toll-call records deters or inhibits potential sources any more than they are already inhibited by the Branzburg rule.
In the second place, it remains unclear how often and to what extent informers are actually deterred from furnishing information by the mere fact that toll-call records may be subpoenaed by the Government. The record does not provide even one genuine instance of such inhibition.108 More significantly, the plaintiffs’ own affidavits indicate that sources generally are apprehensive about the circumspection of the journalists themselves, and cooperate only on the basis of the journalist’s personal assurance of confidentiality; there is no indication that sources are especially inhibited by the contingency of toll-call-record subpoenas any more than they are by the prospect of third-party detection in general.
Finally, as the Branzburg Court pointed out, it is difficult to take seriously claims that toll-record subpoenas will appreciably constrict the flow of information to journalists.109 Journalists’ records have been subject to such subpoenas for 50 years, and during this time, “the press has flourished” 110 and so has its use of confidential sources.111 Indeed, plaintiffs’ own affidavits aver that the number of sources using telephones has steadily and dramatically increased in recent years, belying the notion that the existing accessibility of toll-call records has curtailed the flow of information. It is therefore clear that “existing constitutional rules have not been a serious obstacle to either the development or retention of confidential news sources by the press”.112 In the face of these considerations, if a balance must be struck, then it *1063must be struck in favor of the general public’s interest in effective law enforcement investigations.
Under the approach urged by plaintiffs and the dissent, courts will be required to enter into a balancing effort in each case to determine whether a particular journalist’s “interest” in suppressing toll-call-record data outweighs the Government’s “interest” in gathering such evidence. The Court in Branzburg was similarly requested to adopt a case-by-case balancing approach with respect to journalists’ claims of testimonial privilege.113 The Court stated: “We are unwilling to embark the judiciary on a long and difficult journey to such an uncertain destination.”114 The Court realized that administration of such a case-by-case approach “would present practical and conceptual difficulties of a high order.”115
One problem, as the Court in Branzburg recognized, is that a case-by-case balancing approach will require courts to devise criteria for estimating the relative exigency of a particular good faith investigation:
Thus, in the end, by considering whether enforcement of a particular law served a “compelling” governmental interest, the courts would be inextricably involved in distinguishing between the value of enforcing different criminal laws. By requiring testimony from a reporter in investigations involving some crimes but not in others, they would be making a value judgment that a legislature had declined to make, since in each case the criminal law involved would represent a considered legislative judgment, not constitutionally suspect, of what conduct is liable to criminal prosecution. The task of judges, like other officials outside the legislative branch, is not to make the law but to uphold it in accordance with their oaths.116
A second problem is that a court will have to devise criteria for estimating the relative importance of certain information-gathering activities.117 Are a novelist’s interests in protecting his sources less compelling than a newspaper’s? Is a small-town reporter’s interest in protecting his sources as compelling as a national newspaper reporter’s? Is the secrecy of a source who provides information on international matters more important than the secrecy of a source who only provides information on national or local affairs? In short, courts will be placed in the position of assigning different “values” to various exercises of First Amendment rights.
Finally, these mystical distinctions and calculations would become even more obscure in the context of the ex parte hearing suggested by the dissent. How will the court in an ex parte hearing come to know the “quantum” of “press” interests in any particular case? How will the Government necessarily know? How is the court going to balance anything if it does not know what to put on one side of the scale?
In sum, then, where the Government subpoenas a journalist’s toll-call records in the course of a criminal investigation, there is no need for a case-by-case balancing of the Government’s interests against those of the journalist. If any balancing is necessary at all, it has already been done and the balance has been struck in favor of the Government. Therefore, plaintiffs’ theory that journalists are entitled to notice of third-party subpoenas in order to secure case-by-case balancing must fail. Since there is no need for case-by-case balancing, there is no need for prior notice.
B. Plaintiffs’ “Screening” Theory.
In treating plaintiffs’ second theory, two issues must be addressed: first, whether plaintiffs’ First Amendment rights would be abridged by defendant telephone companies’ compliance with toll-call-record subpoenas issued in bad faith as part of an effort to harass plaintiffs and interfere *1064with their sources of information; and second, if so, whether prior judicial screening of toll-call-record subpoenas is an appropriate remedy.
1. Abridgment of First Amendment Rights by Bad Faith Subpoenas
When used in good faith, investigative techniques such as physical surveillance, the gathering of third-party information through interrogation or subpoena, the collection of information from informants and the placement of undercover agents are all proper police activities that violate no constitutional rights of the suspects involved. However, all investigative techniques are subject to abuse and can conceivably be used to oppress citizens and groups, rather than to further proper law enforcement goals. In some cases, bad faith use of these techniques may constitute an abridgment of the First Amendment rights of the citizens at whom they are directed, be they “journalists” or less exalted citizens.118
Plaintiffs contend that, in the past, the Government has issued subpoenas for their toll-call records, not in furtherance of bona fide felony investigations, but in order to harass plaintiffs in their journalistic information-gathering activities. Plaintiffs assert that these bad faith subpoenas violated their First Amendment rights. Whether or not these particular allegations are true, there can be no doubt that, as a general proposition, such bad faith action would constitute an abridgment of a journalist’s First Amendment rights at least in some cases, if not in every case.119 In Branzburg, the Supreme Court specifically stated that the First Amendment protected news-gathering from such abusive assaults:
[N]ews gathering is not without its First Amendment protections, and grand jury investigations if instituted or conducted other than in good faith, would pose wholly different issues for resolution under the First Amendment. Official harassment of the press undertaken not for purposes of law enforcement but to disrupt a reporter’s relationship with his news sources would have no justification.120
Thus, while the First Amendment does not immunize the information-gathering activities of a journalist or any other citizen from good faith law enforcement investigation, it does protect such activities from official harassment. Unlike good faith investigation to which all citizens are subject, official harassment places a special burden on information-gathering, for in such cases the ultimate, though tacit, design is to obstruct rather than to investigate, and the official action is proscriptive rather than observatory in character. Thus, plaintiffs’ initial premise that bad faith toll-record subpoenas may operate as an abridgment of their First Amendment rights is correct.
2. Prior Judicial Screening — an Equitable Remedy Unjustified Factually on this Summary Judgment Record
Plaintiffs have not sought damages for the past instances of bad faith harassment which they allege. Rather, plaintiffs have sought extraordinary prospective relief by which they are to be protected from future subpoenas — a procedural remedy by which plaintiffs must be notified of any subpoena directed at their toll-call records so that they may prompt a judicial determination as to whether or not the subpoena has been issued in connection with a bona fide investigation. In short, plaintiffs seek an ongoing judicial audit of future government investigations in order to screen out bad faith subpoenas.
Such extraordinary anticipatory relief cannot be justified merely on the ground that toll-record subpoenas might possibly be abused in the future so as to effect an infringement of plaintiffs’ First *1065Amendment rights. If the mere possibility of future government misconduct were sufficient to warrant such prophylactic relief, then the courts would be called upon to superintend virtually all investigative activity. Much more than the mere possibility of future official misconduct is needed to justify this type of judicial intervention.
It is fundamental that in order to obtain the kind of equitable relief sought in this case, a plaintiff “must show not only that there is an imminent threat of harm but also that the threatened harm is irreparable.” 121 A party invoking equitable intervention in the criminal investigative process has a particularly heavy burden. Only the most extraordinary circumstances warrant anticipatory judicial involvement in criminal investigations. Even where federalism concerns are absent, the fundamental concept of separation of powers dictates judicial restraint. The powers of criminal investigation are committed to the Executive branch. The balance between the Executive and Judicial branches would be profoundly upset if the Judiciary assumed superintendence over the law enforcement activities of the Executive branch upon nothing more than a vague fear or suspicion that its officers will be unfaithful to their oaths or unequal to their responsibility. As the Supreme Court stated in Laird v. Tatum:
Carried to its logical end, this approach [of judicial supervision of military intelligence activities] would have the federal courts as virtually continuing monitors of the wisdom and soundness of Executive action; . . . it is not the role of the judiciary, absent actual present or immediately threatened injury resulting from unlawful governmental action.122
Since plaintiffs in this case contend that their First Amendment rights are violated by bad faith subpoenas issued for harassment purposes, then in order to justify anticipatory relief each individual plaintiff must show (1) that there is an imminent threat that the Government will subpoena his toll records in bad faith, (2) that such subpoena will cause him substantial and irreparable harm, and (3) that his remedy at law is inadequate. And, still, a grant of anticipatory relief is a matter of equitable discretion and must depend in each case on an assessment of all relevant circumstances. In order to meet this burden, plaintiffs contend that the past instances of toll-record subpoenas reflected in the record establish a pattern of bad faith harassment; that this pattern indicates the imminence of future bad faith subpoenas; that such subpoenas will cause irreparable injury; and that, hence, they are entitled to equitable relief to prevent the recurrence of irreparable injury which is clear and imminent.
This case comes before us on cross motions for summary judgment. Under Rule 56(c) of the Federal Rules of Civil Procedure, summary judgment is to be rendered only when there is “no genuine issue as to material fact” and “the moving party is entitled to judgment as a matter of law.” As this Court recently stated:
In assessing the motion, all “inferences to be drawn from the underlying facts con*1066tained in [the movant’s] materials must be viewed in the light most favorable to the party opposing the motion.” Indeed, “the record must show the movant’s right to [summary judgment] ‘with such clarity as to leave no room for controversy’, and must demonstrate that his opponent ‘would not be entitled to [prevail] under any discernible circumstances’.”
Summary judgment is unavailable if it depends upon any fact that the record leaves susceptible of dispute. Facts not conclusively demonstrated, but essential to the movant’s claim, are not established merely by his opponent’s silence; rather, the movant must shoulder the burden of showing affirmatively the absence of any meaningful factual issue. That responsibility may not be relieved through adjudication since “[t]he court’s function is limited to ascertaining whether any factual issue pertinent to the controversy exists [and] does not extend to [the] resolution of any such issue.” 123
Applying these principles to the case at bar, it is clear that the District Court was correct in granting defendants’ motion for summary judgment with respect to ten of the 15 plaintiffs, namely, The Reporters’ Committee for Freedom of the Press, Dow Jones & Co., Inc., Marquis W. Childs, Emmett Dedmon, Morton Mintz, Bruce Morton, John Pierson, Richard Salant, Daniel Schorr, and Frederick Taylor. These plaintiffs failed to adduce any evidence from which it could be inferred that their toll-call records had ever been subpoenaed in bad faith in the past or that their records were in imminent danger of being subpoenaed in bad faith in the future. Therefore, the District Court’s summary judgment against these ten plaintiffs is affirmed.
However, the remaining five plaintiffs have adduced evidence that their toll-call records were inspected by the Government in the past. Viewing this evidence in the light most favorable to plaintiffs, it is at least a possible inference that the Government acted in bad faith in issuing these past subpoenas. Also, it is possible that, upon further proof, the circumstances of these past inspections might indicate the imminence of further abuse. Though it is a close question, it appears that these five plaintiffs have adduced just enough evidence to withstand defendants’ motion for summary judgment. Therefore, the District Court’s action in granting summary judgment against these five is reversed.
Furthermore, it is clear that the District Court acted properly in denying plaintiffs’ motion for summary judgment. At this stage, the five remaining plaintiffs have failed to lay an adequate foundation for the exceptional remedy they seek. Since plaintiffs have relied on the occurrence of past abuses to justify equitable intervention, then the critical factual issue in the case is whether there indeed has been official abuse in the past. Only when this factual issue has been resolved in favor of plaintiffs does the remedial issue arise as to whether the past misconduct is sufficient to warrant the extraordinary relief sought. Yet, this critical factual issue is in genuine dispute. The defendant United States emphatically claims that the five past subpoenas directed at plaintiffs’ toll-call records were all issued in connection with bona fide felony investigations, i. e., there has been no pattern of official abuse in the past which could possibly justify the anticipatory relief sought by plaintiffs. Affidavits, depositions, and answers to interrogatories all present information supporting this contention. Plaintiffs, on the other hand, assert that in these past instances “the Government had, at most, only a highly tenuous law enforcement justification for seeking to scrutinize plaintiffs’ news-gathering activities.”124 Therefore, this case must be remanded to the District Court. If the remaining plaintiffs determine to pursue this suit, then the issue of bad faith must be resolved in the District Court. If on remand the District *1067Court finds no record of past abuse, and no other basis to anticipate impending future misconduct, then plaintiffs will have failed on their claim for equitable relief. If, however, the District Court finds that there has been a pattern of abuse in the past, then the appropriate judicial response must be determined at that time.
In sum; then, while we accept the notion that otherwise legitimate investigative techniques may be abused in such a way as to abridge the First Amendment rights of the persons at whom they are directed, and while we also accept the proposition that harassing toll-record subpoenas can abridge journalists’ First Amendment rights, we believe that there is a genuine dispute in this case as to whether past toll-record subpoenas were in bad faith, and therefore any equitable relief predicated on a finding of past official abuse is premature.
It must be stressed that even if the remaining plaintiffs establish, and the District Court finds, that there have been past instances of abuse, it does not necessarily mean that each plaintiff will be entitled to prior notice of future subpoenas. As already stated, in order to obtain the kind of anticipatory relief sought in this case, each individual plaintiff must show not only that he personally faces an imminent threat of harm but also that the threatened harm is irreparable. In addition, each plaintiff must show that his remedy at law is inadequate. As the record now stands, plaintiffs have not made a sufficient showing as to any of these requirements.
First, there is little indication in the record that plaintiffs have suffered, are suffering, or will suffer substantial and irreparable harm. Pláintiffs have alleged that substantial injury occurs to their journalistic activities by disclosure of their telephone billing records. Plaintiffs have attempted to elevate that allegation to the level of an uncontroverted fact, by stating:
There can be no question but that serious and irreparable jury [sic] is done to a reporter’s First Amendment rights when the Government obtains unsupervised access to his toll records.125
However, the information developed in the course of discovery so far provides no clear evidence that either past instances of such disclosure or the prospect of future disclosure have caused such injury. To support an allegation that disclosure of toll-billing records “drastically curtails” plaintiffs’ ability to gather news, individual plaintiffs provided only their conclusions of what would, could, or might happen if sources could not be guaranteed confidentiality. In each case plaintiffs were merely speculating.126
Even as to past injury, no plaintiff has demonstrated that he has ever lost a source as a result of a bad faith subpoena of his telephone billing records. The United States directed interrogatories to each plaintiff inquiring as to loss of sources from toll-record inspection. Each plaintiff responded that the inquiry was not applicable.
Two plaintiffs did indicate that information from confidential sources ceased at times coincident with the securing of telephone billing records by federal investigators. In other words, plaintiffs have provided only two instances over a five-year period to support their claim that disclosure to others of toll-call records effects a “drastic curtailment” of their journalistic functions.
Even in these two instances, however, the plaintiffs failed to show that the injury resulted from federal law enforcement access to telephone toll-billing records. In the first instance, plaintiff David E. Rosenbaum stated:
In one instance, I stopped contacting one source after learning of an Internal Revenue Service subpoena for my telephone records because of concern that our relationship might jeopardize his career.127
Clearly this unilateral decision by Mr. Rosenbaum to terminate a reporter-source relationship is not sufficient evidence to support a contention that a source has been chilled.
*1068In the second instance, plaintiff Jack Anderson responded that in the course of a federal law enforcement investigation in which toll-billing records were utilized, some of his sources were questioned. He further stated that: “With only one exception, I never received any information from any of the sources again.”128 However, in responding to inquiry as to whether the refusal of any source to provide further information could be attributed to this disclosure he responded, “I do not know whether these sources stopped or refused to provide further information, or whether they simply never again had any information which they wished, to provide to me.” 129 This conclusion certainly does not support an allegation of injury as to his ability to gather and disseminate news as a result of governmental access to toll-billing records.
As exemplified by the foregoing references to the record, it is clear that plaintiffs have thus far failed to provide sufficient factual support for their assertion that unsupervised toll-record subpoenas result in substantial and irreparable injury to their news-gathering activities. At this stage, that assertion remains nothing more than speculation.130
Moreover, plaintiffs have so far failed to demonstrate an “imminent threat of harm.” In order to meet the imminency requirement, plaintiffs in general have usually been required to show a pervasive pattern of past abuse such as will indicate a continuing program of misconduct.131 Courts have frequently denied anticipatory relief when plaintiffs have failed to establish such a pattern.132 In Long v. District of Columbia,133 the proprietor of a jewelry store had reported to the police that plaintiff, a “suspicious person,” was in his store. On the basis of this report, the police initiated an investigation, went to the store and frisked plaintiff. Plaintiff “sought to enjoin the police from stopping and frisking individuals in any manner not in accord with their constitutional rights, alleging the actions complained of . . . to be typical and to be violative of constitutional rights.”134 In denying relief this Court said:
Considerations of policy dictate that the courts act cautiously in granting injunctions against police action. A court should not bind the hands of the police on the mere possibility that certain conduct may be repeated. To do so would unnecessarily involve the courts in police matters and dictate action in situations in which discretion and flexibility are most important. In order for a court to grant an injunction, there should be a showing that there is a substantial risk that future violations will occur.
In order to show a substantial likelihood of future conduct, a clear pattern of harassment must be shown. Such a pattern should consist of frequent acts of misconduct by police officers, which acts were known to the superior officers of the police force.135
Although establishing a pattern of past abuse has been the most common way of showing imminency, it is by no means the only way to meet that requirement; the appropriate manner of demonstration de*1069pends on the circumstances of each case.136 Still, the cases involving patterns of past misconduct do illustrate that a clear factual foundation must be laid in order to warrant prospective relief of the kind sought here.
In this case, plaintiffs have alleged only five instances of misconduct over a five-year period.137 Most significantly, there have been no toll-record subpoenas directed at plaintiffs, and — as far as the record indicates — any other journalist, since 1 March 1974, over four years ago. Since that date AT&T has had in effect a new policy providing for notice to subscribers of toll-record subpoenas except in cases where the Government certifies that notice could interfere with an ongoing investigation. Moreover, the Department of Justice has issued guidelines under which certification generally is not to be made in cases involving completed offenses.138 There is no indication that under this new regime notice will be withheld from journalists in the future. It is thus clear that plaintiffs have so far failed to demonstrate that the alleged threat of future harm is, in any sense, clear and imminent — for this must be tested relative to the new AT&T policy effective 1 March 1974, not by practices undeniably abandoned.139
Finally, plaintiffs have not shown that their remedy at law is inadequate. Under AT&T’s new policy, plaintiffs will always be able to discover that a Government subpoena has been issued for their toll-call records. By placing their names on file with their telephone companies, plaintiffs can assure that they will be notified of a certificated subpoena as soon as the 90-day period and subsequent extensions, if any, expire. This guarantees that plaintiffs will have the opportunity to bring post hoe Bivens -type damage suits in any case in which they believe the Government was acting in bad faith. Thus, the inevitability of damage suits would pose a significant deterrent to future Government misconduct, making totally unnecessary any type of equitable relief.
The dissent’s position in this regard is unsound. According to the dissent, a plaintiff must show only two things in order to be entitled to the extraordinary prospective relief sought in this case: first, that he has a First Amendment right, and, second, that there exists an investigative technique that might possibly be used in bad faith so as to abridge that right. Under this approach, once a plaintiff has shown the existence of a right and the mere possibility of its infringement, he becomes entitled to insist that the Judicial branch superintend government investigations in order to prevent this hypothetical abuse.
This position is in direct conflict with the most fundamental tenets of equity and a series of Supreme Court eases, most notably Rizzo v. Goode,140 Allee v. Medrano,141 and Laird v. Tatum.142 These cases reflect the well-established principle that judicial supervision of police activity predicated on the prospect of future abuse must be based on the imminence of future misconduct. The mere possibility of future misconduct is simply not enough.
Beyond the fact that it is utterly in conflict with the authorities, the dissent’s approach would ■ have the most absurd and pernicious practical consequences. Up until this time, it has been the law that a court would not assume prior review of government investigative action simply because it *1070feared future misconduct; it would require a clear showing that such misconduct was a real and imminent prospect. The dissent’s position would completely do away with this threshold. It would require only the existence of a constitutional right and the possibility of its abridgment. It would not base its anticipatory intervention into police investigative action on an actual past abridgment of rights, nor on a present abridgment of rights, nor on a probability of future abridgment of rights, but rather on a mere possibility of future abridgment of rights. This approach has no logical stopping-point. There is no person in the United States who cannot meet these criteria.
Any person can establish the existence of a First Amendment right and of an investigative technique that could possibly be employed in bad faith so as to violate that right. Consider the following example. Suppose that some recognized leader of organized crime in the United States files suit in district court for a declaratory judgment that he must receive prior notice of any physical surveillance conducted against him. He claims (1) that he has a First Amendment right to association and (2) that physical surveillance could be used against him in bad faith, not for any legitimate law enforcement purpose, but rather to harass him in his personal relationships. He alleges, quite reasonably, that the very possibility of abusive surveillance deters people from associating with him. He alleges further that each time he is placed under physical surveillance he suffers irreparable harm to his associational rights because some of his confidential associates are exposed. He moves for summary judgment. Under the dissent’s approach there is no principled reason why this organized crime figure would not be automatically entitled to precisely the same anticipatory remedy sought by the “conceded journalists” in the instant case. Why should he not be able to insist on prior notice, or at least ex parte judicial oversight, of any physical surveillance directed at him ?
It is not normally the role of the courts to hover over law enforcement officers, reviewing, approving, and monitoring each step of a criminal investigation in order to satisfy itself that the officers are acting in good faith. The dissent declares its willingness to engage in this kind of supervision and to assume the administration of criminal investigations based solely on its own suspicion that the officers who are constitutionally charged with law enforcement responsibility will not be equal to the task and will act in bad faith as often as not. The dissent seems surprised that we are willing to permit Executive officials to perform their constitutional function and exercise their discretion without a prior judicial determination that they are acting in good faith. But in our view it is to these officers that the investigative power is committed in the first instance by law, and it is in these officers that it should rest in the first instance until an adequate reason for judicial preemption is established.
C. Summary
Plaintiffs have contended that they have a right to receive prior notice of toll-record subpoenas issued in the course of criminal investigations. They have predicated this right on the supposed necessity for case-by-case judicial balancing (on the theory that good faith subpoenas may abridge their First Amendment rights) and on the supposed necessity for case-by-case judicial screening (on the theory that bad faith subpoenas may abridge their First Amendment rights).
First, by now it is undeniably clear that there is no need for case-by-case balancing where toll-record subpoenas have been issued in criminal investigations. It is undeniably clear because, one, good faith subpoenas do not abridge plaintiffs’ First Amendment rights and, two, any judicial balancing of interests that may be called for need only be performed once, and such balance is to be struck in favor of the public’s interest in effective law enforcement.
Second, it is also undeniably clear that plaintiffs have not yet established their en*1071titlement to a case-by-case screening of toll-record subpoenas. While we have agreed in theory that subpoenas issued in bad faith may in some cases abridge First Amendment rights, all precedent forbids the Judiciary to assume superintendence of criminal investigations merely on the suspicion that Executive officers may act in bad faith in the future, and only permits such judicial intervention when a plaintiff establishes a clear and imminent threat of such future misconduct. Plaintiffs here have so far failed to lay an adequate foundation for such judicial intervention. The five remaining plaintiffs will have the opportunity to meet this heavy burden on remand.
Affirmed in part and Remanded in part.
. Dow Jones & Co., Inc. and The Knight Newspaper Group of Knight-Ridder Newspapers, Inc.
. Jack Anderson, Marquis W. Childs, Emmett Dedmon, Richard Dudman, Morton Mintz, Bruce Morton, John Pierson, James R. Polk, David E. Rosenbaum, Richard Salant, Daniel Schorr, and Frederick Taylor.
. Joint Appendix (J.A.) 71.
. J.A. 70.
. J.A. 87-88. The six (6) month period is prescribed by Federal Communications Commission regulation. See 47 C.F.R. Part 42.
. The record does not indicate precisely when the practice of inspecting toll-call records for law enforcement purposes commenced, but the first court of appeals case involving a challenge to evidence derived from such an inspection was decided in 1929. See Blakeslee v. United States, 32 F.2d 15 (1st Cir. 1929).
. The critical importance of toll-call record evidence to criminal investigations was commented upon in the Affidavit of Donald E. Campbell, Assistant United States Attorney in the District of Columbia (J.A. 272-274):
At the present time, I am Chief of the Major Crimes Division of the United States Attorney’s Office for the District of Columbia. .
Telephone toll records have been not only an invaluable aid in carrying out the investigative responsibility of the Major Crimes Division, but have also been critical evidence in criminal prosecutions developed by this Division. The continued access to such records, without advance notice to subscribers, is absolutely necessary for law enforcement to successfully investigate and prosecute certain criminal activity.
Subscriber and toll information without advance notice to subscribers is critical in any major narcotic, gambling, fencing, loansharking or corruption matter or any other major ongoing conspiratorial or other criminal activity. The upper echelons of the criminal community go to great extremes to insulate themselves from detection by law enforcement. However, these upper echelon individuals must communicate with those individuals on the “street” who are doing their “bidding”, and who are visible to law enforcement. For example, an individual is identified in the D.C. area as a major drug trafficker, and drugs are purchased from this individual by an undercover officer. If law enforcement has conducted a detailed and careful investigation, there will be a telephone on which this trafficker has contacted his out-of-town supplier, because no conspiracy can be carried forward without communication between the conspirators. Experience has shown that a telephone is usually used by one of the parties to the conspiracy to communicate to the principal receiver of a large shipment of drugs, the time, place, and means of arrival of the shipment. Therefore, the drug trafficker is not arrested at this time and a careful review of toll records is made. This may permit law enforcement to identify an individual in another part of the country who is known to law enforcement in that area to be involved in drugs. Then with this information as to the probable identity of the supplier, law enforcement has a chance, by using other investigative techniques, to develop a case against the suppliers. Once probable identity has been established,' physical surveillance then becomes extremely important, informants and undercover officers then know who they are looking for, etc.
In every case where law enforcement considers the use of a wire interception, toll records are not only one of the critical factors in establishing probable cause, but are probably the most revealing factor as to how successful a wire intercept will be. If advance notice is given to the subscriber that law enforcement is interested in his toll records, the effectiveness of wire intercepts will be greatly reduced. Advance notice to subscribers will only serve to frustrate law enforcement in its effort to immobilize the upper echelons of the criminal community and result in more concentration on the lesser figures.
My experience indicates that prosecutors in this Office are extremely cautious in subpoenaing toll records. I know of no case in which toll records were subpoenaed by an Assistant without sufficient information being presented to that Assistant so that he could properly conclude that the individual was involved in some type of criminal conduct and that the toll records were likely to advance the criminal investigation. It has also been my experience that very few subpoenas are issued for toll records in situations where criminal acts have been completed. Toll records obtained while certain criminal activity is still ongoing are extremely valuable in that they provide guidance to law enforcement as to the possible identity of co-conspirators. With this information, other investigative techniques can then be used to establish the involvement of co-conspirators. Access to these records in the course of such an ongoing criminal conspiracy often provides law enforcement with a critical opportunity to corroborate and substantiate both the contacts between the conspirators and the illegal purposes furthered by those contacts.
. J.A. 227.
. J.A. 40.
. J.A. 40.
. J.A. 40-41.
. J.A. 106-116; 157-162.
. J.A. 17-21.
. J.A. 20-22.
. J.A. 7-15.
. The figure of 75,000-100,000 is calculated on the basis of the 1974-75 rate of 2,000-3,000 subpoenas each month.
. This subpoena was issued by a grand jury in connection with an investigation into the theft and receipt of stolen United States Government documents, a violation of 18 U.S.C. § 641, which provides that any person who “steals,” “converts” or “sells” any “record” or “thing of value” of the United States or who knowingly “receives” the same with intent to convert it to his own use or gain, shall be imprisoned up to 10 years or fined up to $10,000, or both.
In November 1972 several hundred American Indians occupied the U. S. Bureau of Indian Affairs building. The Washington, D. C., Metropolitan Police Department, Intelligence Section, had an undercover agent inside the building among the Indians. When the Indians eventually vacated the building, they purloined and took with them Government property, including official Government documents. The undercover agent reported to his Section that the Indians had negotiated with Jack Anderson, a journalist, for the purchase of some of these stolen documents and that an Anderson employee was scheduled to receive the documents at the home of one of the Indian leaders. This information was passed on to the FBI. The Indian leader’s apartment building was placed under physical surveillance. At the scheduled time, Mr. Les Whitten, an Anderson employee, arrived at the apartment and was subsequently arrested there in apparent possession of the stolen documents. Mr. Anderson’s and Mr. Whitten’s toll-call records were subpoenaed to obtain further evidence in the case (J.A. 260-66).
. In the first of these espionage cases, the toll-call record subpoena was issued by a grand jury in connection with an investigation into the unauthorized disclosure of a classified, “eyes only” State Department cable, a possible violation of 18 U.S.C. § 793, which provides, inter alia, that any person in lawful possession of any document relating to national defense, knowing that such document could be used to the advantage of any foreign nation, who willfully transmits the document to any person not entitled to receive it, shall be imprisoned up to 10 years or fined up to $10,000, or both.
Jack Anderson wrote and published an article reporting on the indiscretions of an A.I.D. official stationed in Kenya. Anderson claimed in the article to have gleaned the information from a classified State Department cable. Mr. Anderson was not authorized to receive classified information. This circumstance suggested that a Government employee with access to sensitive diplomatic cable traffic was conveying classified information to unauthorized persons. The FBI commenced an investigation to determine the identity of the Government employee. During the investigation the Government subpoenaed Mr. Anderson’s toll records in an effort to determine whether any State Department employees who had access to the cable had contacted Mr. Anderson prior to the article’s appearance. (J.A. 219-26).
The second and third espionage cases related to the investigation of Daniel Ellsberg for the unauthorized disclosure of classified Government documents known as the “Pentagon Papers.” The FBI received independent information indicating that “Knight Newspapers” and employees of the St. Louis Post Dispatch may have obtained information from those classified documents and may have been in telephonic contact with the person who made the unauthorized disclosure. On the basis of this information, the FBI obtained grand jury subpoenas for the toll-call records of “Knight Newspapers” and two employees of the St. Louis Post Dispatch. In his deposition, Richard Dudman, one of the Dispatch’s employees, bears out the accuracy of the Bureau’s lead:
There are cases, too, where an informant may have reason to fear criminal prosecution because of what he tells us. This was the situation when we were trying to find someone who would give us copies of the Pentagon Papers, which were classified top secret. Our eventual source for some of those documents telephoned us anonymously from pay stations, and we never did learn his identity. (J.A. 174).
*1040The EUsberg investigation resulted in an indictment; however, the charges were dismissed during the trial. (J.A. 145).
. This summons was issued by the Internal Revenue Service in an investigation into possible violations of 26 U.S.C. § 7213 and 18 U.S.C. § 1905 by an IRS employee. The former statute makes it a felony for any U. S. employee to disclose income tax return information to any person not authorized to receive it; the latter statute makes it a felony for any U. S. employee to disclose to an unauthorized person financial and other confidential information contained in reports filed with the Government.
In 1973, information came to the attention of the IRS indicating that an IRS employee was divulging to a newspaper reporter sensitive information regarding the criminal tax fraud investigation of a prominent taxpayer then being conducted by the Service’s Intelligence Division. Mr. David Rosenbaum, a reporter for The New York Times, had contacted the taxpayer and told him that he knew that a “full scale” fraud investigation was underway. The taxpayer’s attorney later contacted Mr. Rosenbaum, and Mr. Rosenbaum recited particulars concerning the investigation and cited specific issues raised during the audit. The taxpayer complained to the IRS that his tax affairs were being leaked to the newspaperman. Other evidence surfaced indicating that the investigation was being compromised. The IRS issued a summons for the reporter’s toll records in order to identify the culpable IRS employee. (J.A. 241-248).
. J.A. 151-53. Moreover, even if the Government does subpoena toll records in civil and administrative matters, there is no evidence that the notice procedures provided under AT&T’s new policy do not give a customer sufficient opportunity to challenge government inspection of the records. By its terms, the policy requires immediate telephonic notice upon receipt of a subpoena and written notice within 24 hours of receipt. The notification must include the approximate date on which the company “will” comply with the subpoena. This strongly suggests that notice will usually be prior to or contemporaneous with company release of records, or, if not, at least within sufficient time to give a customer the opportunity effectively to challenge government inspection of the records. This was clearly the intent of AT&T’s policy, and there is no evidence that the policy does not work this way in practice with respect to non-law enforcement cases.
. The Government has taken no position on this issue.
. See 28 U.S.C. § 1331(a) (1976).
. See Powell v. McCormack, 395 U.S. 486, 499, 89 S.Ct. 1944, 1952, 23 L.Ed.2d 491, 504 (1969) (“declaratory judgment can be used as a predicate to further relief, including an injunction”); 10 C. Wright & A. Miller, Federal Practice & Procedure, § 2751 at 866-867 (1973) (declaratory judgment may be “supplemented either by damages or by equitable relief even though coercive relief might have been available at the time of the declaratory action”) (footnotes omitted).
. Emphasis supplied. See generally Chase Nat’l Bank v. City of Norwalk, 291 U.S. 431, 436-437, 54 S.Ct. 475, 477, 78 L.Ed. 894, 898 (1934); Environmental Defense Fund v. EPA, 158 U.S.App.D.C. 1, 5, 485 F.2d 780, 784, aff'd, 412 U.S. 541, 93 S.Ct. 2770, 37 L.Ed.2d 140 (1973); Schwartz v. Galveston Independent School Dist., 309 F.Supp. 1034, 1038 (S.D.Tex. 1970); South Cent. Bell Tel. Co. v. Constant, Inc., 304 F.Supp. 732 (E.D.La.1969), aff'd, 437 F.2d 1207 (5th Cir. 1971); SEC v. Meyers, 285 *1042F.Supp. 743, 750 (D.Md.1968); 7 J. Moore, Federal Practice Digest II 65.13 (2d ed. 1975).
. See authorities cited supra note 24.
. See authorities cited supra note 24, particularly South Cent. Bell Tel. Co. v. Constant. There, prior to litigation, one firm obtained a consent injunction prohibiting another firm from employing the former’s name in advertisements or statements to customers. The issue was whether a telephone company with knowledge of the injunction was obligated to take steps to prevent transmission, by use of its equipment, of statements breaching the terms of the order. The District Court held that it was, stating:
As soon as South Central Bell was apprised of the fact that its subscriber, Constant, was by use of South Central Bell’s equipment, violating the injunction imposed by the Court, it had a duty not to act in any way in concert with Constant to effectuate or perpetuate the violation. South Central Bell had the means to prevent its equipment from being used to violate the injunction, and its failure to do so would, at the very least, have amounted to a passive participation in the violation.
304 F.Supp. at 736. On appeal, the Fifth Circuit affirmed without opinion. See note 24 supra.
. See note 26 supra.
. To repeat, plaintiffs’ cause of action against the Government, of course, is constitutional in nature, but the restraint upon AT&T from Rule 65(d) would merely be incidental to effectuation of appellants’ substantive claim. See text supra following note 27.
. Langston v. Johnson, 156 U.S.App.D.C. 5, 7, 478 F.2d 915, 917 (1973); see, e. g., Ashwander v. TVA, 297 U.S. 288, 346-347, 56 S.Ct. 466, 483, 80 L.Ed. 688, 711 (1936) (Brandeis, J., concurring).
. The District Court made a conclusory finding that there was “state action” in the adoption of AT&T’s policy. We have no occasion to determine whether or not this conclusion was correct.
. This Fourth Amendment claim was pressed in the District Court, but plaintiffs seem to have largely abandoned it on this appeal. However, this case involves the interplay of First and Fourth Amendment protections and, therefore, a thorough discussion of Fourth Amendment aspects of the case is desirable in order to lay the groundwork for analysis of plaintiffs’ First Amendment claims in the ensuing section.
. See, e. g., Terry v. Ohio, 392 U.S. 1, 9, 88 S.Ct. 1868, 1873, 20 L.Ed.2d 889 (1968) (“[Wlherever an individual may harbor a reasonable ‘expectation of privacy’ . . . , he is entitled to be free from unreasonable governmental intrusion”); Katz v. United States, 389 U.S. 347, 351, 88 S.Ct. 507, 511, 19 L.Ed.2d 576 (1967) (“[T]he Fourth Amendment protects people, not places. What a person knowingly exposes to the public, even in his own home or office, is not a subject of Fourth Amendment protection. But what he seeks to preserve as private, even in an area accessible to the public, may be constitutionally protected”); Warden v. Hayden, 387 U.S. 294, 301-304, 87 S.Ct. 1642, 1647, 18 L.Ed.2d 782 (1967) (“[The Fourth Amendment] was intended to protect against invasions of ‘the sanctity of a man’s home and the privacies of life’ from searches under indiscriminate, general authority. . [T]he principal object of the Fourth Amendment is the protection of privacy rather than property. . . .”).
. See, e. g., Katz v. United States, 389 U.S. 347, 88 S.Ct. 507, 19 L.Ed.2d 576 (1967) (Government eavesdropping violated privacy upon which petitioner justifiably relied while using a public telephone booth).
. See, e. g., United States v. United States District Court, 407 U.S. 297, 92 S.Ct. 2125, 32 L.Ed.2d 752 (1972) (Fourth Amendment requires prior judicial issuance of warrant for domestic security electronic surveillance).
. See, e. g., United States v. Dionisio, 410 U.S. 1, 93 S.Ct. 764, 35 L.Ed.2d 67 (1973) (compelled furnishing of voice exemplars did not violate Fourth Amendment since no person has reasonable expectation of privacy in physical characteristics of his own voice); United States v. White, 401 U.S. 745, 91 S.Ct. 1122, 28 L.Ed.2d 453 (1971) (use of informants’ and undercover agents’ testimony does not violate Fourth Amendment protections since “inescapably, one . must realize and risk that his companions may be reporting to the police”).
. “No man is an island, entire of itself; every man is a piece of the Continent, a part of the main.” — John Donne (Devotions XVII).
. See, e. g., United States v. Miller, 425 U.S. 435, 96 S.Ct. 1619, 48 L.Ed.2d 71 (1976); Lewis v. United States, 385 U.S. 206 (1966); Hoffa v. United States, 385 U.S. 293, 87 S.Ct. 408, 17 L.Ed.2d 374 (1966).
. United States v. Miller, 425 U.S. 435, 96 S.Ct. 1619, 48 L.Ed.2d 71 (1976).
. United States v. Miller, 425 U.S. 435, 96 S.Ct. 1619, 48 L.Ed.2d 71 (1976) (bank depositor has no Fourth Amendment interest in bank records relating to his transactions); Donaldson v. United States, 400 U.S. 517, 91 S.Ct. 534, 27 L.Ed.2d 580 (1971) (taxpayer has no standing to challenge IRS summons directed at employer’s records relating to taxpayer); id. at 537, 91 S.Ct. at 545 (Douglas, J., concurring) (“[I]t is difficult to see how the summoning of a third party, and the records of a third party, can violate the rights of a taxpayer, even if a criminal prosecution is contemplated or in progress. There is no right to be free from incrimination by the records or testimony of others.”); First National Bank v. United States, 267 U.S. 576, 45 S.Ct. 231, 69 L.Ed. 796 (1925), aff'g without opinion 295 F. 142 (S.D.Ala.1924) (taxpayer has no Fourth Amendment basis for challenging internal revenue summons directed at third-party bank records relating to taxpayer). Cf. California Bankers Ass’n v. Shultz, 416 U.S. 21, 94 S.Ct. 1494, 39 L.Ed.2d 812 (1974); Couch v. United States, 409 U.S. 322, 93 S.Ct. 611, 34 L.Ed.2d 548 (1973) (taxpayer has no Fourth or Fifth Amendment right to contest a third party summons directed at taxpayer’s accountant, even when it required production of taxpayer’s own records); Wilson v. United States, 221 U.S. 361, 31 S.Ct. 538, 55 L.Ed. 771 (1911) (corporation president has no Fourth Amendment interest in corporation’s business records).
. United States v. House, 524 F.2d 1035, 1044 (3rd Cir. 1975) (use of evidence obtained through seizure of bank records without summons did not violate bank depositors’ Fourth or Fifth Amendment rights: “Certainly it would be difficult ... to imagine on what basis we could hold that the taxpayers, when they made their bank deposits and invoked the bank’s assistance in collecting third party negotiable instruments, entered a constitutionally protected zone of privacy.”); United States v. Continental Bank & Trust Co., 503 F.2d 45 (10th Cir. 1974) (taxpayers not entitled to notice of IRS summons directed at bank records relating to them, since they had no interest in such records that could be vindicated through intervention); Scarafiotti v. Shea, 456 F.2d 1052 (10th Cir. 1972) (mandamus unavailable to compel revenue agent to provide notice of investigative activities involving third parties); Harris v. United States, 413 F.2d 316 (9th Cir. 1969) (bank depositor had no constitutional basis for challenging summons directed at bank records relating to him); Galbraith v. United States, 387 F.2d 617, 618 (10th Cir.1968) (bank customer had no constitutional standing to challenge SEC administrative summons of bank records relating to him: “The guarantees of the Fourth and Fifth Amendment are essentially personal privileges that cannot be projected to the seizure of papers and effects of another or to suppress incriminating aspects of such records”); Application of Cole, 342 F.2d 5, 7-8 (2d Cir.), cert. denied, 381 U.S. 950, 85 S.Ct. 1803, 14 L.Ed.2d 723 (1965) (IRS was not obligated to notify taxpayers of an administrative summons which called for the production of bank records relating to them: “[A]ll of the records, documents and papers which were the subject matter of the summons were the property of the Bank on whom the summons was served. None of the material sought belonged to the taxpayers or involved the work product of their attorneys. They have no interest in any of them in the sense that they had a right to any of them. Under these circumstances the Commissioner had no duty to give advance notice to the taxpayers or their counsel of his intention to examine a third party and the third party’s own records and papers.”) (emphasis supplied); McMann v. SEC, 87 F.2d 377 (2d Cir.) (L. Hand, J.), cert. denied, 301 U.S. 684, 57 S.Ct. 785, 81 L.Ed. 1342 (1937) (customer held not entitled to prevent brokers from complying with SEC subpoenas for copy of customer’s account on grounds that subpoenas infringed constitutional rights of customer).
. 425 U.S. 435, 96 S.Ct. 1619, 48 L.Ed.2d 71 (1976).
. Id. at 442-45, 96 S.Ct. 1619.
. Id. at 443 n.5; 445, 96 S.Ct. 1619.
. Katz v. United States, 389 U.S. at 351, 88 S.Ct. at 511.
. United States v. Miller, 425 U.S. at 442—44, 96 S.Ct. at 1623-24 (citations and footnotes omitted) (emphasis added).
. See cases cited at note 49, infra.
. 452 F.2d 505 (9th Cir. 1971).
. Id. at 506.
. United States v. Lustig, 555 F.2d 737, 747 n.10 (9th Cir. 1977), cert. denied, 434 U.S. 1043, 98 S.Ct. 889, 54 L.Ed.2d 796 (1978) (“It is well established that the ‘expectation of privacy’ only extends to the content of telephone conversations, not to records that conversations took place.”); United States v. Baxter, 492 F.2d 150, 167 (9th Cir. 1973), cert. denied, 416 U.S. 940, 94 S.Ct. 1945, 40 L.Ed.2d 292 (1974) (“[Tjhe test of whether a seizure prior to adherence to judicial processes violates the Fourth Amendment is whether there is an invasion of the defendant’s ‘constitutionally justifiable expectations of privacy’. . . . Telephone subscribers are fully aware that records will be made of their toll calls. This Court has held that the expectation of privacy protected by the Fourth Amendment attaches to the content of a telephone conversation not to the fact that a conversation took place. The defendants have failed to show a violation of their Fourth Amendment rights.”); United States v. Fithian, 452 F.2d 505 (9th Cir. 1971); Nolan v. United States, 423 F.2d 1031 (10th Cir. 1969), cert. denied, 400 U.S. 848, 91 S.Ct. 47, 27 L.Ed.2d 85 (1970) (use of telephone company toll records at trial and before grand jury to show interstate calls between defendant’s telephone and that of an alleged co-conspirator did not constitute a violation of the Fourth Amendment); United States v. Covello, 410 F.2d 536, 542 (2d Cir.), cert. denied, 396 U.S. 879, 90 S.Ct. 150, 24 L.Ed.2d 136 (1969) (“[Tjhe keeping of toll records is a necessary part of the ordinary course of the telephone company’s business and is necessary in order that the company may substantiate its charges to its customers. Toll records are kept for all telephone subscribers and are not kept just for subscribers being investigated by officers of the law, or ones suspected of criminal proclivities. The subscriber is fully aware that such records will be made . and the records of the telephone company so kept in the ordinary course of the company’s business are entitled to the same evidentiary treatment as the records of other businesses.”); United States v. Gallo, 123 F.2d 229 (2nd Cir. 1941) (L. Hand, Swan, A. Hand, JJ.) (“When a person takes up a telephone he knows that the company will make, or may make, some kind of a record of the event, and he must be deemed to consent to whatever record the business convenience of the company requires.”); United States v. Kohne, 347 F.Supp. 1178 (W.D.Pa.1972). See also the following cases upholding admissibility of toll-call-billing records in criminal trials: Wood v. United States, 84 F.2d 749 (5th Cir.), cert. denied, 299 U.S. 589, 57 S.Ct. 116, 81 L.Ed. 434 (1936); Brink v. United States, 60 F.2d 231 (6th Cir.), cert. denied, 287 U.S. 667, 53 S.Ct. 291, 77 L.Ed. 575 (1932); United States v. Radov, 44 F.2d 155 (3rd Cir. 1930); Blakeslee v. United States, 32 F.2d 15 (1st Cir. 1929).
It has also been held that a telephone subscriber has no Fourth Amendment interest in local call records obtained by means of a pen register installed without his knowledge. Hodge v. Mountain States Tel. & Tel. Co., 555 F.2d 254 (9th Cir. 1977); United States v. Clegg, 509 F.2d 605 (5th Cir. 1975).
. Judge Robinson concurs in all of Part IV, with the exception of my analysis of the relationship between the First, Fourth, and Fifth Amendments (subsection IV.A. Lb.) which he finds unnecessary to our decision. While there may be slight differences in emphasis on the First Amendment issues between my opinion for the Court and Judge Robinson’s concurring opinion, we believe our views are basically consistent. I am in fundamental agreement with Judge Robinson’s discussion of the remedial issue (Part II of the concurring opinion) which amplifies my own (Part IV.B of this opinion).
. Brief for Appellants at 21-28.
. By use of the term “bad faith” we mean to include not only investigations actuated by bad intentions but also investigations that probe at will without relation to law enforcement needs and that expose for the sake of exposure. See Branzburg v. Hayes, 408 U.S. 665, 700, 92 S.Ct. 2646, 23 L.Ed.2d 646 (1972).
. 408 U.S. 665, 92 S.Ct. 2646, 23 L.Ed.2d 646 (1972).
. Id at 707, 92 S.Ct. at 267.
. Brief for Appellants at 14: “The Supreme Court in Branzburg explicitly recognized that reporters have a First Amendment interest, personal to themselves, in maintaining the confidentiality of their news sources.”
. Brief for Appellants at 11: “The Court [in Branzburg ] recognized that valid First Amendment interests are involved whenever the confidentiality of a journalist’s sources are threatened. . . . ” (emphasis added).
. See Part III of this opinion.
. See 26 U.S.C.A. § 7213(a)(1) (West Supp. 1977).
. See id. § 7213(a)(3) (West Supp.1977).
. 408 U.S. 665, 92 S.Ct. 2646, 23 L.Ed.2d 646 (1972).
. 408 U.S. at 690, 92 S.Ct. at 2661.
. Id.
. Id. at 682, 92 S.Ct. 2646, 2657.
. Id.
. Id. at 682, 698-99, 92 S.Ct. 2646.
. Id. at 695, 92 S.Ct. at 2664.
. Plaintiffs and the dissent characterize the instant case as involving the integrity of “confidential relationships” between journalists and their sources and contend that the journalists’ right to notice of third-party subpoenas derives from a right to preserve the “confidentiality” of their sources. For example, plaintiffs assert:
It is the confidentiality of their news sources that plaintiffs are entitled to protect. Plaintiffs have been injured by the invasion of confidentiality that resulted from previous disclosures of toll records. (Reply Brief of Appellants at 16.)
In the same vein, the dissent states:
[Tjhere [can be no] doubt that newsgathering encompasses contacts with confidential sources of information . And the circumstances of this case, more strikingly than testimony before a grand jury, betray the potential for substantial infringement of this confidentiality . . . . (Dissent at-of 192 U.S.App.D.C., at 1085 of 593 F.2d.)
These characterizations are disingenuous and seriously misleading. In point of fact, this *1051case, unlike Branzburg v. Hayes, does not involve' “confidential relationships”; it involves non-confidential third-party information.
A “confidential” relationship exists in the law only where parties undertake to keep some matter secret from the rest of the world, and “confidentiality” exists only between those parties. Ordinarily mutual promises of secrecy have no legal significance; the state may compel each party to betray the confidence of the other in order to obtain evidence in criminal proceedings. In a very narrow class of cases, however, mutual promises of secrecy are respected in order to foster certain socially important relationships. In these cases, the law will protect one party to the relationship from voluntary or compelled betrayal by the other; that is, it recognizes only that one party has the “privilege” of preventing the other party from testifying about the secret matter. This is a “testimonial privilege,” and this is the privilege that journalists unsuccessfully claimed in Branzburg v. Hayes. That case involved Government efforts to compel journalists themselves to betray their sources’ confidences. Here, there is no such compulsion; neither journalists nor their sources are being compelled to reveal any confidence. Rather, the Government is collecting third-party information. The law does not protect parties to a “confidential” relationship from compromise by neutral third-parties to whom they have knowingly imparted their secret. This risk of compromise is entirely within the parties’ own control. It is an elementary principle that when parties to a “confidential” relationship reveal their secret to third parties, the confidentiality is destroyed, and the Government is free to exact the information from third parties without violating any of the policies which may have initially given rise to a testimonial privilege. In short, the law will only go so far as to protect each party from betrayal by the other, but it will not extend the web of secrecy to third parties.
In this case, plaintiffs claim that the mere fact of their relationship with certain sources is “confidential”. If this is true, then plaintiffs must keep this fact secret in order to preserve “confidentiality”. However, plaintiffs and their sources have contacted each other through the facilities of third parties, without taking available precautions to protect their identities. In so doing, they have knowingly assumed the risk of disclosure by placing in the hands of those third parties documentary evidence that their relationship exists. Under these circumstances, plaintiffs cannot appeal to the “confidentiality” of their relationships to bar Government access to this third-party evidence. See generally 8 Wigmore, Evidence §§ 2285-2241 (McNaughten rev.ed. 1961).
. See Zurcher v. Stanford Daily, 436 U.S. 547, 98 S.Ct. 1970, 56 L.Ed.2d 525 (1978) (rejecting journalists’ assertion that search of newspaper office for mere evidence pursuant to a warrant violated First Amendment); Pell v. Procunier, 417 U.S. 817, 833, 94 S.Ct. 2800, 2810, 41 L.Ed.2d 495 (1974) (rejecting media plaintiffs’ assertion that face-to-face interviews with designated prison inmates were such an effective and superior method of news-gathering that its curtailment amounted to unconstitutional state interference with a free press: “It has generally been held that the First Amendment does not guarantee the press a constitutional right of special access to information not available to the public generally,” quoting Branzburg); Saxbe v. Washington Post Co., 417 U.S. 843, 94 S.Ct. 2811, 41 L.Ed.2d 514 (1974); Zemel v. Rusk, 381 U.S. 1, 16-17, 85 S.Ct. 1271, 1280, 14 L.Ed.2d 179 (1965) (sustaining Government refusal to validate passports to Cuba even though that restriction “rendered less than wholly free the flow of information concerning that country”) (“[T]he right to speak and publish does not carry with it the unrestrained right to gather information.”); Estes v. Texas, 381 U.S. 532, 539-40, 85 S.Ct. 1628, 14 L.Ed.2d 543 (1965); Associated Press v. NLRB, 301 U.S. 103, 132-33, 57 S.Ct. 650, 81 L.Ed. 953 (1937) (“The publisher of a newspaper has no special immunity from the application of general laws. He has no special privilege to invade the rights and liberties of others. He must answer for libel. He may be punished for contempt of court. He is subject to the antitrust laws. Like others he must pay equitable and non-discriminatory taxes on his business.”)
. See Lovell v. Griffin, 303 U.S. 444, 58 S.Ct. 666, 82 L.Ed. 949 (1938); Near v. Minnesota, 283 U.S. 697, 51 S.Ct. 625, 75 L.Ed. 1357 (1931); Schenk v. United States, 249 U.S. 47, 39 S.Ct. 247, 63 L.Ed. 470 (1919).
. Chaplinsky v. New Hampshire, 315 U.S. 568, 572 n.3, 62 S.Ct. 766, 86 L.Ed. 1031 (1942).
. E. g., Baird v. State Bar of Arizona, 401 U.S. 1, 91 S.Ct. 702, 27 L.Ed.2d 639 (1971); Keyishian v. Board of Regents, 385 U.S. 589, 87 S.Ct. 675, 17 L.Ed.2d 629 (1967); Lamont v. Postmaster General, 381 U.S. 301, 85 S.Ct. 1493, 14 L.Ed.2d 398 (1965); Baggett v. Bullitt, 377 U.S. 360, 84 S.Ct. 1316, 12 L.Ed.2d 377 (1964).
. E. g., N.A.A.C.P. v. Alabama, 357 U.S. 449, 78 S.Ct. 1163, 2 L.Ed.2d 1488 (1958) (contempt citation); Dombrowski v. Pfister, 380 U.S. 479, 85 S.Ct. 1116, 14 L.Ed.2d 22 (1965) (criminal prosecution); Baird v. State Bar of Arizona, 401 U.S. 1, 91 S.Ct. 702, 27 L.Ed.2d 639 (1971) (exclusion from profession).
. 408 U.S. 1, 11, 92 S.Ct. 2318, 2324, 33 L.Ed.2d 154 (1972).
. 408 U.S. at 680-81, 699-700, 92 S.Ct. 2318. See Zurcher v. Stanford Daily, 436 U.S. at 567, 98 S.Ct. at 1982 (1978) (“Not every such seizure, and not even most, will impose a prior restraint. And surely a warrant to search newspaper premises for criminal evidence . carries no realistic threat of prior restraint or of any direct restraint whatsoever”).
. The dissent’s reasoning proceeds in three steps. First, it contends that plaintiffs have a First Amendment right to gather information; second, it asserts that this information-gathering right includes a right to collect information from secret sources; and, third, it reasons that any government action which threatens to compromise the secrecy of a source necessarily invades the right to collect information from secret sources. The dissent discusses these points in terms of “interests.” Thus, it concludes that plaintiffs have First Amendment “interests” in their toll-call records and that these “interests” are implicated whenever the Government obtains access to these records. Further, the dissent suggests that we in the majority must agree that plaintiffs’ First Amendment “interests” are threatened by the Government’s toll-record subpoena practice. Dissent at---, - of 192 U.S.App.D.C., at 1086-1087, 1093 of 593 F.2d. In fact, we do not agree and, in our view, the dissent’s conclusions gloss over the truly difficult issues posed.
When the dissent speaks in terms of First Amendment “interests,” we assume it is referring to First Amendment “rights.” We agree with the dissent that the First Amendment extends some protection to news-gathering. To this extent, it may be said that plaintiffs have First Amendment “rights” or “interests” in news-gathering. But the existence of this right does not necessarily mean that plaintiffs have a First Amendment “right” or “interest” in using their telephones immune from the prospect of good faith toll-record subpoenas. The issue in this case is precisely whether or not toll-record subpoenas abridge the right to gather information that is guaranteed by the First Amendment. If these subpoenas abridge that right, then plaintiffs would have a First Amendment “interest” in their toll-call records, and the procedural protections called for by the dissent would be appropriate. However, if these subpoenas do not abridge the right to gather information, then plaintiffs would not have a First Amendment “interest” in their toll-call records. It is our position that the subpoena practices reflected in this record do not abridge the right to gather information that is guaranteed by the First Amendment. Hence, in the factual circumstances of this case, we do not agree with the dissent that plaintiffs have First Amendment “interests” in their toll-call records. We have reached this conclusion through a two-part analysis by which we have focused, first, on the scope of the right to gather information itself and, second, on the nature of the impact that the challenged government practice has on that right.
The dissent has attempted to magnify the impact that the Government’s subpoena practice has on plaintiffs news-gathering, claiming that plaintiffs have been “restrained” and “foreclosed” from using the telephone. Dissent at - of 192 U.S.App.D.C., at 1090 of 593 F.2d. This claim is simply not true. Plaintiffs are no more foreclosed from using the telephone by the prospect of toll-record subpoenas than they are foreclosed from using the streets by the prospect of physical surveillance. See footnote 54 of Judge Robinson’s concurring opinion.
. Judge Robinson does not join in the analysis in this particular subsection IV.A.l.b., finding it unnecessary to our decision.
. See Part III of this Opinion.
. See Zurcher v. Stanford Daily, 436 U.S. 547, 98 S.Ct. 1970, 56 L.Ed.2d 525 (1978); United States v. United States District Court, 407 U.S. 297, 92 S.Ct. 2125, 32 L.Ed.2d 752 (1972); Stanford v. Texas, 379 U.S. 476, 85 S.Ct. 506, 13 L.Ed.2d 431 (1965); Marcus v. Search Warrant, 367 U.S. 717, 81 S.Ct. 1708, 6 L.Ed.2d 1127 (1961).
. See generally 8 Wigmore, Evidence §§ 2214, 2251 (McNaughton rev. ed. 1961).
. E. g., Zurcher v. Stanford Daily, 436 U.S. 547, 98 S.Ct. 1970, 56 L.Ed.2d 525 (1978); United States v. United States District Court, 407 U.S. 297, 92 S.Ct. 2125, 32 L.Ed.2d 752 (1972); Stanford v. Texas, 379 U.S. 476, 85 S.Ct. 506, 13 L.Ed.2d 431 (1965). See also United States v. Ramsey, 431 U.S. 606, 97 S.Ct. 1972, 52 L.Ed.2d 612 (1977); Roaden v. Kentucky, 413 U.S. 496, 93 S.Ct. 2796, 37 L.Ed.2d 757 (1973).
. See cases cited at note 78, supra. See also In re Possible Violations of 18 U.S.C. 371, 641, 1503, 184 U.S.App.D.C. 82, 564 F.2d 567 (1977).
. To be distinguished are two lines of cases: One line of cases deals with the use of warrants, not as means of investigation, but rather as direct means of prior restraint. In these cases the seizures were not to obtain evidence, *1055but directly to suppress the circulation of printed matter. See, e. g., Marcus v. Search Warrant, 367 U.S. 717, 81 S.Ct. 1708, 6 L.Ed.2d 1127 (1961) (seizure of 11,000 copies of 280 allegedly “obscene” publications); A Quantity of Copies of Books v. Kansas, 378 U.S. 205, 84 S.Ct. 268, 11 L.Ed.2d 163 (1964) (seizure of 1,715 copies of 31 allegedly obscene publications). These cases, therefore, do not apply to investigative action that effects no prior restraint on First Amendment activity.
The second line of cases deals with associational privacy. The First Amendment prohibits the use of compulsion to exact from individuals (or groups) the wholesale disclosure of their associational ties where such inquiry is not germane to the determination of whether a crime has been committed. See, e. g., DeGregory v. Attorney General of New Hampshire, 383 U.S. 825, 86 S.Ct. 1148, 16 L.Ed.2d 292 (1966); Bates v. Little Rock, 361 U.S. 516, 80 S.Ct. 412, 4 L.Ed.2d 480 (1960); N. A. A. C. P. v. Alabama, 357 U.S. 449, 78 S.Ct. 1163, 2 L.Ed.2d 1488 (1958); Watkins v. United States, 354 U.S. 178, 77 S.Ct. 1173, 1 L.Ed.2d 1273 (1957). In such cases, the challenged governmental action is not strictly investigative; the probing is not germane to the detection of specific criminal conduct and is frequently conducted in public so that the information obtained by the government is simultaneously and indiscriminately broadcast to the public at large. In these cases, the First Amendment gives rise to a personal testimonial privilege, similar to the privilege against self-incrimination, by which the individual or group can fend off such interrogation. See 8 Wigmore, Evidence §§ 2214, 2251. These cases have no applicability to good faith criminal investigative efforts. Branzburg v. Hayes, 408 U.S. 665, 700, 92 S.Ct. 2646, 23 L.Ed.2d 626 (1972). Cf. Wilkinson v. United States, 365 U.S. 399, 81 S.Ct. 567, 5 L.Ed.2d 633 (1961); Barenblatt v. United States, 360 U.S. 109, 79 S.Ct. 1081, 3 L.Ed.2d 1115 (1959). Moreover, these cases recognize only a personal testimonial privilege to resist compelled self-disclosure. They do not apply to the good faith collection of information from third parties. If they were so construed, it would be impossible for law enforcement officials to investigate organizations suspected of criminal activity or individuals who could be linked to such organizations. Closely related to this line of cases are cases involving sweeping legislative investigations of First Amendment activity. See United States v. Rumely, 345 U.S. 41, 73 S.Ct. 543, 97 L.Ed. 770 (1953).
. 436 U.S. 547, 98 S.Ct. 1970, 56 L.Ed.2d 525 (1978).
. See id., 436 U.S. at 564, 98 S.Ct. at 1981.
. See id., 436 U.S. at 565, 98 S.Ct. at 1981 1982.
. See id., 436 U.S. at 569-70, 98 S.Ct. at 1984 (concurring opinion).
. Lustiger v. United States, 386 F.2d 132, 139 (9th Cir. 1967) (“[T]he Fourth Amendment does not preclude postal inspectors from copying information contained on the outside of sealed envelopes in the mail”); Cohen v. United States, 378 F.2d 751, 760 (9th Cir. 1967); Cana*1057day v. United States, 354 F.2d 849, 856-57 (8th Cir. 1966); United States v. Shwartz, 283 F.2d 107 (3rd Cir. 1960); United States v. Costello, 255 F.2d 876 (2d Cir. 1958).
. See Lamont v. Postmaster General, 381 U.S. 301, 85 S.Ct. 1493, 14 L.Ed.2d 398 (1965).
. Milwaukee Social Democratic Club Pub. Co. v. Burleson, 255 U.S. 407, 437, 49 S.Ct. 352, 65 L.Ed. 704 (1921) (Holmes, J., dissenting) quoted with approval in Blount v. Rizzi, 400 U.S. 410, 416, 91 S.Ct. 423, 27 L.Ed.2d 498 (1971).
. Cohen v. United States, 378 F.2d 751, 760 (9th Cir. 1967).
. See United States v. White, 401 U.S. 745, 91 S.Ct. 1122, 28 L.Ed.2d 453 (1971); Lewis v. United States, 385 U.S. 206, 87 S.Ct. 424, 17 L.Ed.2d 312 (1966); Hoffa v. United States, 385 U.S. 293, 87 S.Ct. 408, 17 L.Ed.2d 374 (1966); Lopez v. United States, 373 U.S. 427, 83 S.Ct. 1381, 10 L.Ed.2d 462 (1963).
. For example, informants are the primary source of information in approximately 85 percent of intelligence investigations conducted'by the Federal Bureau of Investigation, as compared with 5 percent in which the primary source is electronic surveillance. Senate Select Comm, to Study Governmental Operations with Respect to Intelligence Activities, Supplementary Detailed Staff Reports on Intelligence Activities and the Rights of Americans, S.Rep.No. 755, 94th Cong., 2d Sess. 228 (1976).
. 373 U.S. 427, 83 S.Ct. 1381, 10 L.Ed.2d 462 (1963).
. See id., 373 U.S. at 452, 469-71, 83 S.Ct. at 1395.
. 401 U.S. 745, 91 S.Ct. 1122, 28 L.Ed.2d 453 (1971).
. See id., 401 U.S. at 762-65, 91 S.Ct. at 1131 1132. (Douglas, J., dissenting).
. See id., 401 U.S. at 752-53, 91 S.Ct. 1122.
. See id., 401 U.S. at 752, 91 S.Ct. at 1126.
. See, e. g., Oaks v. United States, 527 F.2d 937, 941 (9th Cir. 1975), cert. denied, 426 U.S. 952, 96 S.Ct. 3177, 49 L.Ed.2d 1191 (1976) (surveillance by undercover agent of public meetings of tax rebellion group). Cf. Fifth Avenue Peace Parade Comm. v. Gray, 480 F.2d 326 (2d Cir. 1973), cert. denied, 415 U.S. 948, 94 S.Ct. 1469, 39 L.Ed.2d 563 (1974) (no justiciable controversy where chilling effect claimed to stem from examination of bank records of antiwar organizer, ascertainment of number of buses obtained to transport demonstrators, and observation of bus departures); Philadelphia Yearly Meeting of the Religious Society of Friends v. Tate, 519 F.2d 1335 (3d Cir. 1975) (no judicially cognizable chilling effect where surveillance activities were limited to police photographing and gathering data at public meetings); Donahue v. Duling, 465 F.2d 196 (4th Cir. 1972) (no judicially cognizable chilling effect where uniformed police photographed those present at a public demonstration).
. 527 F.2d 937 (9th Cir. 1975).
. Id. at 941.
. 389 U.S. 347, 350 n.5, 88 S.Ct. 507, 511, 19 L.Ed.2d 576 (1967) (“Virtually every governmental action interferes with personal privacy to some degree. The question in each case is whether that interference violates a command of the United States Constitution.”)
. 408 U.S. at 704-705, 92 S.Ct. at 2668.
. Brief for Appellants at 28-33.
. The dissent stresses the need for “judicial superintendence” of actions which may impinge on First Amendment rights. (Dissent at -of 192 U.S.App.D.C.. at 1088 of 593 F.2d). It cites, inter alia, the following cases: Bantam Books, Inc. v. Sullivan, 372 U.S. 58, 83 S.Ct. 631, 9 L.Ed.2d 584 (1963); Carroll v. President and Comm’rs of Princess Anne, 393 U.S. 175, 89 S.Ct. 347, 21 L.Ed.2d 325 (1968); Freedman v. Maryland, 380 U.S. 51, 185 S.Ct. 734, 13 L.Ed.2d 649 (1965). All these cases, however, involved systems of State regulation which imposed prior restraints on expression predicated on a factual determination that the suppressed expression was not protected by the First Amendment. These cases stand for the proposition that such systems must assure an almost immediate judicial determination of the validity of the restraint. This determination is a factual determination as to whether a particular form of expression is protected or not under the First Amendment; it does not involve a case-by-case balancing of the quantum of “interests” in a protected form of expression against the quantum of governmental interests in infringing on that expression. Thus, in Bantam Books, Inc., the Court did not suggest that the merits of a particular obscene publication be balanced in each case against the Government’s interest in suppressing that publication. This was not why the Court stated that “judicial superintendence” was necessary. Rather, it was declared necessary to ensure that either before or soon after a prior restraint was placed on an allegedly obscene publication, there be a procedurally adequate hearing at which the fact of obscenity could be determined.
. 408 U.S. 707-708, 92 S.Ct. at 2670 (footnote omitted).
. Justice Powell’s concurring opinion in Branzburg is fully consistent with this analysis. It focuses on the concluding portion of the Court’s opinion and reiterates that there is a remedy available to reporters in the event of bad faith harassment:
As indicated in the concluding portion of the opinion, the Court states that no harassment of newsmen will be tolerated. If the newsman believes that the grand jury investigation is not being conducted in good faith he is not without remedy. Indeed, if the newsman is called upon to . . . implicate [ ] confidential source relationships without a legitimate need of law enforcement, he will have access to the court on a motion to quash . . .. 408 U.S. at 709-10, 92 S.Ct. at 2671
(emphasis added).
Although Justice Powell refers to case-by-case “balancing,” it is clear that he is actually referring to the availability of judicial case-by-case screening out of bad faith “improper and prejudicial” interrogation. Indeed, this court has *1062already so interpreted Justice Powell’s opinion in In re Possible Violations of 371, 641, 1503, 184 U.S.App.D.C. 82, 564 F.2d 567 (1977):
The Branzburg decision did not leave newsmen completely without protection from indiscriminate probing for news sources. In particular, the Court observed that official harassment of the press undertaken solely to disrupt a reporter’s relationship with news sources would clearly be subject to judicial control. In a separate opinion, Mr. Justice Powell, who also concurred in the opinion of the Court, emphasized and elaborated upon this aspect of the majority opinion .
We conclude that Branzburg squarely rejected the very privilege appellant asserts that it established. A newsman can claim no general immunity, qualified or otherwise, from grand jury questioning. On the contrary, like all other witnesses, he must appear and normally must answer. If the grand jury’s questions are put in bad faith for the purpose of harassment, he can call on the courts for protection.
. See discussion at pp.---of 192 U.S.App.D.C., at pp. 1067-1068 of 593 F.2d, infra.
. 408 U.S. at 698-699, 92 S.Ct. 2646, 2665.
. Id.
. Id.
. Id.
. See id, 408 U.S. at 680, 92 S.Ct. 2646.
. See id., 408 U.S. at 703-04, 92 S.Ct. at 2668.
. Id.
. See id., 408 U.S. at 705, 92 S.Ct. at 2669.
. See id., 408 U.S. at 705 n.40, 92 S.Ct. 2646.
. See Allee v. Medrano, 416 U.S. 802, 92 S.Ct. 2191, 40 L.Ed.2d 566 (1974); Hague v. CIO, 307 U.S. 496, 59 S.Ct. 954, 83 L.Ed. 1483 (1939).
. Branzburg v. Hayes, 408 U.S. 665, 707-08, 92 S.Ct. 2646, 2670, 23 L.Ed.2d 646 (1972).
. Id. (footnote omitted).
. O. Fiss, Injunctions 9 (1972) (emphasis added). See also Younger v. Harris, 401 U.S. 37, 91 S.Ct. 746, 27 L.Ed.2d 669 (1971); Dombrowski v. Pfister, 380 U.S. 479, 85 S.Ct. 1116, 14 L.Ed.2d 22 (1965).
. 408 U.S. 1, 15, 92 S.Ct. 2318, 2326, 33 L.Ed.2d 154 (1972). See also Socialist Workers Party v. Attorney General, 510 F.2d 253 (2d Cir. 1974) (injunction against FBI surveillance of SWP convention held abuse of discretion), application for stay denied, 419 U.S. 1314, 95 S.Ct. 425, 428, 42 L.Ed.2d 627 (1974) (Justice Marshall, as circuit justice) (“[Ojur abhorrence for abuse of governmental investigative authority cannot be permitted to lead to an indiscriminate willingness to enjoin undercover investigation of any nature, whenever a countervailing First Amendment claim is raised.”).
The broad power in the police and the grand jury to probe widely for evidence of crime is vital because the ability of the police to go directly to the suspect or his property for evidence of guilt is significantly limited by Fourth and Fifth Amendment protections. The Government is therefore remitted to more extensive surveillance and circumstantial investigation to detect criminality. A court should not interdict these investigations in the absence of clear proof of bad faith or arbitrariness.
. National Association of Government Employees v. Campbell, 192 U.S.App.D.C. -, ---, 593 F.2d 1023, 1027 (1978) (footnotes omitted).
. Brief for Appellants at 8.
. Brief for Appellants at 22.
. J.A. 167-198.
. J.A. 188.
. J.A. 267.
. J.A. 267.
. Cf. Branzburg v. Hayes, 408 U.S. at 693-94, 92 S.Ct. 2646.
. Rizzo v. Goode, 423 U.S. 362, 373-75, 96 S.Ct. 598, 46 L.Ed.2d 561 (1976); Allee v. Medrano, 416 U.S. 802, 815, 92 S.Ct. 2191, 40 L.Ed.2d 566 (1974); Lewis v. Kugler, 446 F.2d 1343 (3d Cir. 1971); Lankford v. Gelston, 364 F.2d 197 (4th Cir. 1966).
. Rizzo v. Goode, 423 U.S. 362, 96 S.Ct. 598, 46 L.Ed.2d 561 (1976); Long v. District of Columbia, 152 U.S.App.D.C. 189, 469 F.2d 927 (1972); Wilson v. Webster, 315 F.Supp. 1104 (C.D.Cal.1970), vacated and remanded, 467 F.2d 1282 (9th Cir. 1972) (denial of injunction held proper); Hughes v. Rizzo, 282 F.Supp. 881 (E.D.Pa.1968).
. 152 U.S.App.D.C. 187, 469 F.2d 927 (1972).
. Id. 152 U.S.App.D.C. at 189, 469 F.2d at 929.
. Id. 152 U.S.App.D.C. at 192, 469 F.2d at 932 (emphasis added).
. See part II of Judge Robinson’s opinion.
. Brief for Appellants at 7.
. J.A. 165.
. See Zurcher v. Stanford Daily, 436 U.S. 547, 98 S.Ct. 1970, 56 L.Ed.2d 525 (1978) (“The fact is that respondents . have pointed to only a very few instances in the entire United States since 1971 involving the issuance of warrants for searching newspaper premises. This reality hardly suggests abuse; and if abuse occurs, there will be time enough to deal with it.”).
. 423 U.S. 362, 96 S.Ct. 598, 46 L.Ed.2d 561 (1976).
. 416 U.S. 802, 92 S.Ct. 2191, 40 L.Ed.2d 566 (1974).
. 408 U.S. 1, 92 S.Ct. 2318, 33 L.Ed.2d 154 (1972).