(dissenting).
I believe that PF 15 exceeds the rulemak-ing authority of a federal district court. Like my brethren, I shall assume that the United States District Court for the District of Massachusetts intended to incorporate as its own local rule the so-called ethical rule promulgated by the Supreme Judicial Court of Massachusetts. Hence, the primary question is not whether a state rule must yield to the Supremacy Clause but whether a single federal district court is empowered to adopt a rule like this.
I think not. To condition a grand jury subpoena on a court’s prior approval reduces the grand jury’s traditional power to call those witnesses it pleases. The Su*659preme Court in analogous circumstances, and two circuit courts, in directly comparable circumstances, have ruled that to go so far is an improper interference with a grand jury’s historical right to every man’s evidence. See section II, below. Whether or not these precedents actually invalidate PF 15 is not the point. They show at very least that PF 15 operates in a policy area too sensitive, important and controversial to be regulated at a local district court level. Regardless of the merits of the rule, it is a major change in grand jury practice and procedure that should be promulgated only through an amendment to the federal criminal rules, or by Congress through legislation.23
I.
A district court’s power to adopt local rules touching on criminal matters is set out in Fed.R.Crim.P. 57, which provides,
Each district court ... may from time to time ... make and amend rules governing its practice not inconsistent with these rules.
(Emphasis provided.)
Rule 57 and the other federal criminal rules (including Rule 6, pertaining to grand juries,24 and Rule 17, pertaining to subpoenas) were enacted under authority of a statute, 18 U.S.C. §§ 3771-3772 (1984), by which Congress specifically empowered the Supreme Court to prescribe rules of pleading, practice and procedure governing criminal cases for the district courts. Federal and local rulemaking further rest upon authority granted by 28 U.S.C. § 2071 (1982), which delegates to the federal courts the power “from time to time to prescribe rules for the conduct of their business.” Such rules must be “consistent with Acts of Congress and rules of practice and procedure prescribed by the Supreme Court.”
Local rules, therefore, must be consistent both with federal statutory and constitutional law (including judicial interpretations thereof), and with the Federal Rules of Criminal Procedure, which are the primary means for regulating practice and procedure in the district courts. The Advisory Committee’s comments on Fed.R. Crim.P. 57 emphasize that local rulemaking is necessarily meant to be of narrow scope. Describing the Federal Rules of Criminal Procedure as “intended to constitute a comprehensive procedural code for criminal cases in the Federal courts” the Advisory Committee goes on to say that,
Nevertheless it seemed best not to endeavor to prescribe a uniform practice as to some matters of detail, but to leave the individual courts free to regulate them, either by local rules or by usage. Among such matters are the mode of impaneling a jury, the manner and order of interposing challenges to jurors, the manner of selecting the foreman of a trial jury ... and other similar details.
(Emphasis supplied.) This language makes clear that local rules were never conceived to be a means for changing policies properly regulated at a federal level. Rather they are to address “some matters of detail” left open by the criminal rules where uniformity is not required.
For reasons given in section II below, I believe the present novel limitation upon a *660grand jury’s subpoena power is much too controversial and delicate to be the sort of “matter of detail” that fits within a district court’s local rulemaking power. Before turning to these reasons, let me first refer to my colleagues’ argument that PF 15 is not “inconsistent” with the federal criminal rules because it does not literally contravene Fed.R.Civ.P. 17(a) (which unconditionally empowers the clerk to issue subpoenas). My colleagues find no inconsistency because PF 15 does not call for court approval before issuance of a grand jury subpoena — it requires court approval only before the prosecutor serves the subpoena. This is far too fine a distinction to provide a convincing rationale. It is, after all, the Supreme Court, not individual local courts, to which Congress has delegated the primary power to prescribe rules of pleading, practice and procedures for the district courts. See 18 U.S.C. §§ 3771-3772 (1984). Mere absence of a clash between the strict letter of the federal criminal rules and PF 15 does not resolve the question of inconsistency. A local rule may be inconsistent if it is discordant with policies implicit as well as explicit in the federal rules. It may likewise be ultra vires if out of harmony with policies in statutes or in controlling interpretations by the courts. See 28 U.S.C. § 2071 (1982). Thus “inconsistency” will exist where a local rule legislates in an area in which the federal rules are silent and the surrounding circumstances show that such silence manifested an intent by the Supreme Court not to regulate further. This is the situation here. Neither Fed.R.Crim.P. 17 nor any other provision in the federal criminal rules or statutes, provides for judicial approval of a grand jury subpoena before it may be served, while history and the case law, as hereinafter discussed, make it clear that a grand jury’s uninhibited ability to call the witnesses it chooses is a right entitled to the utmost respect. In such circumstances, the rules’ silence is most reasonably interpreted as forbidding further regulation by any body other than the Supreme Court or Congress.
That literal conflict is not the sole ground for “inconsistency” under Fed.R. Civ.P. 57 was made abundantly plain by the Supreme Court in Miner v. Atlass, 363 U.S. 641, 80 S.Ct. 1300, 4 L.Ed.2d 1462 (1960). In that case, despite the absence of a direct conflict with any General Admiralty Rule, the Court struck down a local admiralty rule that allowed the district court to order the taking of oral depositions as part of the discovery process. Like the Federal Rules of Criminal Procedure, the General Admiralty Rules authorized district courts “ ‘to regulate their practice in such a manner as they deem most expedient for the due administration of justice, provided the same are not inconsistent with these rules.’ ” Id. at 647, 80 S.Ct. at 1304, quoting General Admiralty Rule 44 (emphasis in original). Although there was no direct conflict with any General Admiralty Rule, the Court, focusing on the history of and policies implicit in those rules, rejected the local rule as “not consistent” with the General Rules. The Court doubted that “a change so basic as this [should] be effectuated through the local rule-making power,” stating that such fundamental procedural innovations should be introduced “only after mature consideration of informed opinion from all relevant quarters, with all the opportunities for comprehensive and integrated treatment which such consideration affords.” 363 U.S. at 650, 80 S.Ct. at 1306. See In re Grand Jury Proceedings, 558 F.Supp. 532, 535-36 (W.D.Va.1983) (courts should exercise restraint in adopting rules touching on important policy issues) (citing J. Wein-stein, Reform of Federal Court Rulemak-ing Procedures, 76 Colum.L.Rev. 905, 930 (1976)).25
*661In the present case, proponents of PF 15 attempt to minimize its importance by describing it as only a minor procedural device providing a structure for dealing with the special problems of attorney subpoenas. I do not question that the subpoenaing of attorneys may, in some instances, create a serious problem (although I think the extent to which there are serious abuses requiring a special rule can be better ascertained at the level of national rule-making than locally). See note 23, supra. But I do question that a special screening procedure can be properly established other than by an amendment to the federal criminal rules or by legislation. Were we writing on a clean slate, perhaps a different view would be in order. But we are not writing on a clean slate; the Supreme Court and other courts, infra, have made it clear that any prior judicial screening of subpoenas impacts seriously and questionably upon the protected information-gathering powers of the grand jury. PF 15 is at very least in an area so controversial and sensitive as to be well beyond the “matters of detail” which Rule 57 allows a single district court to regulate by local rule.
II.
I turn now to the precedents showing that the silence of the federal rules and of Congress is not an invitation to local rule-making but rather indicates, in this context, that local regulation is foreclosed. These precedents stem from the deep respect and deference accorded to the independence of the grand jury.
The grand jury is a part of neither the executive nor judicial branch. Although closely related to both, it is an entity with independent constitutional status. See Nixon v. Sirica, 487 F.2d 700, 712 n. 54 (D.C.Cir.1973). Its task, of course, is to inquire into the existence of possible criminal conduct. Given this mission, the grand jury has been instilled with broad investigatory powers, foremost of which is its right to every man’s evidence. United States v. Dionisio, 410 U.S. 1, 9, 93 S.Ct. 764, 769, 35 L.Ed.2d 67 (1973); In re Grand Jury Matters, 751 F.2d 13, 16 (1st Cir.1984). In United States v. Mandujano, 425 U.S. 564, 571, 96 S.Ct. 1768, 1774, 48 L.Ed.2d 212 (1976), the Supreme Court stated,
the grand jury’s investigative power must be broad if its public responsibility is adequately to be discharged. Indispensable to the exercise of its power is the authority to compel the attendance and the testimony of witnesses, and to require the production of evidence.
(Citations omitted.)
The grand jury’s unencumbered power to compel people to appear before it has deep historical roots. See Blair v. United States, 250 U.S. 273, 280, 39 S.Ct. 468, 470, 63 L.Ed. 979 (1919). “[T]he grand jury’s right to every man’s evidence is substantively limited only by express ‘constitutional, common-law or statutory privileges.’ ” In re Grand Jury Matters,. 751 F.2d 13, 17 (1st Cir.1984), quoting Branzburg v. Hayes, 408 U.S. 665, 688, 92 S.Ct. 2646, 2660, 33 L.Ed.2d 626 (1972).
PF 15’s requirement that a prosecutor not subpoena certain lawyers without prior judicial approval is a significant limitation upon the grand jury’s right to every man’s evidence. It both imposes a procedural block and raises the likelihood that, in deciding whether to approve or disapprove a subpoena, an individual judge will impose substantive limitations of his own devising going beyond the “express ‘constitutional, common-law or statutory privileges’ ” currently recognized.26 Id. It forces the *662grand jury to conform its investigatory powers in advance to the predilections of a judge.
The Supreme Court has refused in the past to compel the grand jury to submit to prior judicial screening of its subpoena power. In Branzburg v. Hayes, the Court rejected two journalists' contention that compelling them to appear and testify before a grand jury would violate the First Amendment’s freedom of speech and press guarantees. The newsmen, it will be noted, did not claim an absolute privilege. As here, they argued only that a reporter should not be forced to appear and testify until the government shows that his testimony is relevant and unavailable from other sources. 408 U.S. at 680. The Court found this argument unpersuasive: “ ‘[The grand jury] is a grand inquest, a body with powers of investigation and inquisition, the scope of whose inquiries is not to be limited narrowly by questions of propriety or forecasts of the probable result of the investi-gation_Id. at 688, 92 S.Ct. at 2660, quoting Blair v. United States, 250 U.S. at 282, 39 S.Ct. at 471; see also United States v. Dionisio, 410 U.S. 1, 93 S.Ct. 764, 35 L.Ed.2d 67 (1973) (no showing of need required before grand jury may subpoena witnesses for purposes of obtaining voice exemplar).
None of the provisions in the Federal Rules of Criminal Procedure (nor in any statute) which deal with grand juries and subpoenas in any way hint that the district court is entitled to exercise a power to screen subpoenas, as PF 15 prescribes, nor is there historical precedent for this being done. Rather, the federal rules assign to the district courts a largely ministerial role in the subpoena process.27 See Fed.R. Crim.P. 6 & 17. PF 15 intrudes upon this open landscape by assigning to the district court a new role in the subpoena process. Whether a good or bad concept, given the case law discussed, PF 15 is clearly inconsistent with the degree of independence currently granted to grand juries under the Federal Rules of Criminal Procedure.
As indicated, the grand jury’s power to compel every man’s evidence has traditionally been limited only by “express constitutional, common-law or statutory privileges.” In re Grand Jury Matters, 751 F.2d at 17. Here, the privilege most freely invoked as supporting PF 15 is the Sixth Amendment right to counsel. The Supreme Court, however, has determined that Sixth Amendment rights do not attach until “the time that adversary judicial proceedings have been initiated.” Kirby v. Illinois, 406 U.S. 682, 688, 92 S.Ct. 1877, 1881, 32 L.Ed.2d 411 (1972).28 PF 15 thus cannot be viewed as a procedural device intended merely to implement existing privileges. It is clearly something new and different— with substantive as well as procedural implications of its own, since in determining whether to permit the subpoenaing of an attorney, each judge must formulate new criteria outside the current privileges. See note 26, supra.
That PF 15 is no mere routine procedural device, but is in fact a novel and controversial inroad upon present federal practice and procedure, is further demonstrated by *663the fact that two other circuits, perceiving a fundamental inconsistency between the grand jury’s mission and a court-approved showing of need, have both rejected such approval in the specific context before us, the attorney-witness. In In re Grand Jury Subpoena Served upon Doe, 781 F.2d 238, 248 (2d Cir.1985) (en banc), cert. denied, 475 U.S. 1108, 106 S.Ct. 1515, 89 L.Ed.2d 914 (1986), the Second Circuit said,
To impose additional requirements that the government show its need for the information sought and that the attorney is the only source for that information would hamper severely the investigative function of the grand jury, if not stop the grand jury ‘dead in its tracks.’
The Seventh Circuit has stated that to require a showing of need before subpoenaing attorneys would run contrary to the method by which the grand jury operates.
A grand jury tracks down leads, and even innocent-looking information may be useful. If the grand jury has some information on a subject, it may seek more to confirm or contradict what it has. How much information is “enough” is a matter for the judgment of the grand jury and the prosecutors rather than the courts.
In re Klein, 776 F.2d 628, 632 (7th Cir.1985). As a grand jury may be pursuing a number of interrelated criminal activities, the relevance of any one piece of testimony or evidence may not be apparent until the end of the investigation. See United States v. Dionisio, 410 U.S. at 16, 93 S.Ct. at 772. It may, therefore, be difficult or impossible to articulate the need for certain testimony at a midinvestigation hearing.
The Supreme Court in Dionisio expressed the similar view that, “Any holding that would saddle a grand jury with mini-trials and preliminary showings would assuredly impede its investigation and frustrate the public’s interest in the fair and expeditious administration of the criminal laws.” 410 U.S. at 17, 93 S.Ct. at 773. Given the Court’s view in Dionisio and Branzburg, supra, and the above circuit opinions, it is obvious at very least that PF 15 ventures into difficult and controversial territory. This is not to' say that a rule of the type set out in PF 15 may never be adopted by the Congress or the Supreme Court. Compare Burlington Northern Railroad Company v. Woods, — U.S. -, 107 S.Ct. 967, 94 L.Ed.2d 1 (1987). Perhaps when confronted with the evidence which proponents of PF 15 submit, the Court would modify the views projected in Branzburg and Dionisio, so as to accommodate such a rule. See note 23, supra. My point is simply that the policy implications of the rule far exceed what a single district court may properly adopt on its own in the guise of a local rule. What is involved is no mere local procedural supplement but a controversial change in federal grand jury practice and procedure, which should be adopted only at the national level after “mature consideration of informed opinion from all relevant quarters.” Miner v. Atlass, 363 U.S. at 650, 80 S.Ct. at 1306.29
Whatever the merits of PF 15 — upon which I do not pass judgment — it legislates in an area far beyond Congress’s and the Supreme Court’s limited delegation of rule-making authority to local tribunals. PF 15 is not simply some minor procedural or “ethical” variation. It marks a dramatic and controversial departure from the historically limited power of the courts to supervise the grand jury, interposing a step that, in very similar contexts, both the Supreme Court and circuit courts have, so far, disallowed.
If a rule like PF 15 is required, Congress or, at least, the Supreme Court under its rulemaking authority, acting at the national level, should promulgate it. A district *664court may not effect such a fundamental change through local rules.
Believing as I do that the district court exceeded its rulemaking power, I also believe for like reasons that the Supremacy Clause prevents Massachusetts from ever enforcing its PF 15 against a federal prosecutor.
For these reasons, I dissent.
. I am aware that PF 15 is not designed to undercut the grand jury but to make sure that a defendant’s right to counsel is properly respected. It seeks to strike a new balance between the grand jury’s and the prosecutor’s right of access to evidence, and a defendant’s right to the full protection of his counsel. It may be time for such a realignment; I do not say it is not. My view is merely that this sort of rebalancing of fundamental rights is work either for the Supreme Court, under its rulemaking powers, or for the Congress. It is not the sort of task for which individual district courts were granted their limited rulemaking powers. If district courts can do this, why should not each district court adopt different procedural rules — additional discovery practices, for example, or additional procedural devices, such as demurrers? Clearly the Supreme Court was granted the power to prescribe uniform rules of pleading, practice and procedure for the district courts in order to avoid just this sort of variance in practice and procedure among the different courts. The present issue is one where uniformity, as well as the protections of a national rulemaking procedure, are called for.
. Grand jury practice and procedure is regulated by Fed.R.Crim.P. 6 and also by a number of statutes. See, e.g., 18 U.S.C. §§ 3321, 1331-34, etc.
. It has been repeatedly emphasized that court rules are adopted under a limited congressional grant of legislative power. See, e.g., Sibbach v. Wilson & Co., 312 U.S. 1, 9-10, 61 S.Ct. 422, 424, 85 L.Ed. 479 (1941); Wayman v. Southard, 23 U.S. (10 Wheat.) 1, 41-42, 6 L.Ed. 253 (1825); J. Weinstein, Reform of Federal Court Rulemaking Procedures, 76 Colum.L.Rev. 905, 927-31 (1976) ("the federal courts have recognized that rule-making is ultimately a legislative power residing in Congress, although delegated in large measure to the courts"). Courts must construe this delegation of rulemaking power strictly, including the requirement that local rules encroach neither upon federal statutory law nor upon the rules of civil and criminal procedure promul*661gated by the Supreme Court as part of a national rulemaking process. See, e.g., Carter v. Clark, 616 F.2d 228 (5th Cir.1980) (local rule requiring verification of allegations in prisoner complaints held to be inconsistent with policy reflected in federal statute providing that written declarations made under penalty of perjury were permissible in lieu of sworn affidavits); Hawes v. Club Ecuestre el Comandante, 535 F.2d 140, 144 (1st Cir.1976) (local rule may not "subvert the overall purpose of the [federal] rules").
. PF 15 sets out no standard by which a judge is to determine whether or not to allow an attorney to be subpoenaed. As pointed out below, the screening provision in PF' 15 does not necessarily dovetail with any particular privilege.
. The federal rules expressly allow the district court to quash an oppressive subpoena duces tecum, but make no provision for screening in advance of service. Fed.R.Crim.P. 17(c). There is no corresponding grant of express authority to quash a subpoena ad testificandum, although, as the majority notes, some courts have assumed the existence of such authority to quash. Other courts, in what Professor Wright has called the "better approach" and in recognition of the limited power of courts to control who testifies before the grand jury, have found that they lacked the power to quash subpoenas ad testificandum. See 2 C. Wright, Federal Practice & Procedure, § 273 at 149 & n. 12, § 275 at Supp. 16 & n. 7.1 (1982 & Supp.1986) (compiling cases).
. Although his client’s Sixth Amendment rights have not attached, an attorney testifying before a grand jury may assert an attorney-client privilege to avoid revealing protected confidences. See, e.g., In re Grand Jury Proceedings in the Matter of Freeman, 708 F.2d 1571, 1574-75 (11th Cir.1983). It is well established, however, that the privilege does not excuse a witness from the duty first to appear and claim the privilege in response to particular questions. In re Certain Complaints under Investigation, 783 F.2d 1488, 1518 (11th Cir.), cert.denied, — U.S. -, 106 S.Ct. 3273, 91 L.Ed.2d 563 (1986).
. I recognize, as my colleagues argue, that the Supreme Court allowed district courts to regulate by local rule the size of civil petit juries — no small matter, to be sure. Colgrove v. Battin, 413 U.S. 149, 93 S.Ct. 2448, 37 L.Ed.2d 522 (1973). But there was a tradition, lacking here, of local control over jury arrangements. More important, there was no such history of questioning by the Court as has been noted here to the innovation in question. This is not an area where it can be reasonably presumed that the Court would be satisfied to leave the matter to local rulemaking.