dissenting.
The main themes of the Court’s opinion are that the text of § 10(b) of the Securities Exchange Act of 1934 (Exchange Act), 15 U. S. C. § 78j(b), does not expressly mention aiding and abetting liability, and that Congress knows how to legislate. Both propositions are unexceptionable, but neither is reason to eliminate the private right of action against aiders and abettors of violations of § 10(b) and the Securities and Exchange Commission’s (SEC’s) Rule 10b-5. Because the majority gives short shrift to a long history of aider and abettor liability under § 10(b) and Rule 10b-5, and because its rationale imperils other well-established forms of secondary liability not expressly addressed in the securities laws, I respectfully dissent.
In hundreds of judicial and administrative proceedings in every Circuit in the federal system, the courts and the SEC have concluded that aiders and abettors are subject to liability under § 10(b) and Rule 10b-5. See 5B A. Jacobs, Litigation and Practice Under Rule 10b-5 §40.02 (rev. ed. 1993) (citing cases). While we have reserved decision on the legitimacy of the theory in two cases that did not present it, all 11 Courts of Appeals to have considered the question have recognized a private cause of action against aiders and abettors under § 10(b) and Rule lOb-5.1 The early aiding *193and abetting decisions relied upon principles borrowed from tort law; in those cases, judges closer to the times and climate of the 73d Congress than we concluded that holding aiders and abettors liable was consonant with the Exchange Act’s purpose to strengthen the antifraud remedies of the common law.2 One described the aiding and abetting theory, grounded in “general principles of tort law,” as a “logical and natural complement” to the private § 10(b) action that furthered the Exchange Act’s purpose of “creation and maintenance of a post-issuance securities market that is free from fraudulent practices.” Brennan v. Midwestern United Life Ins. Co., 259 F. Supp. 673, 680 (ND Ind. 1966) (borrowing
*194formulation from the Restatement of Torts § 876(b) (1939)), later opinion, 286 F. Supp. 702 (1968), aff’d, 417 F. 2d 147 (CA7 1969), cert. denied, 397 U. S. 989 (1970). See also Pettit v. American Stock Exchange, 217 F Supp. 21, 28 (SDNY 1963).
The Courts of Appeals have usually applied a familiar three-part test for aider and abettor liability, patterned on the Restatement of Torts formulation, that requires (i) the existence of a primary violation of § 10(b) or Rule 10b-5, (ii) the defendant’s knowledge of (or recklessness as to) that primary violation, and (iii) “substantial assistance” of the violation by the defendant. See, e. g., Cleary v. Perfectune, Inc., 700 F. 2d 774, 776-777 (CA1 1983); I IT, An Int’l Investment Trust v. Cornfeld, 619 F 2d 909, 922 (CA2 1980). If indeed there has been “continuing confusion” concerning the private right of action against aiders and abettors, that confusion has not concerned its basic structure, still less its “existence.” See ante, at 170. Indeed, in this case, petitioner assumed the existence of a right of action against aiders and abettors, and sought review only of the subsidiary questions whether an indenture trustee could be found liable as an aider and abettor absent a breach of an indenture agreement or other duty under state law, and whether it could be liable as an aider and abettor based only on a showing of recklessness. These questions, it is true, have engendered genuine disagreement in the Courts of Appeals.3 But instead of simply addressing the questions presented by the parties, on which the law really was unsettled, the Court sua sponte directed *195the parties to address a question on which even the petitioner justifiably thought the law was settled, and reaches out to overturn a most considerable body of precedent.4
Many of the observations in the majority’s opinion would be persuasive if we were considering whether to recognize a private right of action based upon a securities statute enacted recently. Our approach to implied causes of action, as to other matters of statutory construction, has changed markedly since the Exchange Act’s passage in 1934. At that time, and indeed until quite recently, courts regularly assumed, in accord with the traditional common-law presumption, that a statute enacted for the benefit of a particular class conferred on members of that class the right to sue violators of that statute.5 Moreover, shortly before the Exchange Act was passed, this Court instructed that such “remedial” legislation should receive “a broader and more liberal interpretation than that to be drawn from mere dictionary definitions of the words employed by Congress.” Piedmont & Northern R. Co. v. ICC, 286 U. S. 299, 311 (1932). There is a risk of anachronistic error in applying our current approach to implied causes of action, ante, at 176-177, to a statute enacted when courts commonly read stat*196utes of this kind broadly to accord with their remedial purposes and regularly approved rights to sue despite statutory silence.
Even had § 10(b) not been enacted against a backdrop of liberal construction of remedial statutes and judicial favor toward implied rights of action, I would still disagree with the majority for the simple reason that a “settled construction of an important federal statute should not be disturbed unless and until Congress so decides.” Reves v. Ernst & Young, 494 U. S. 56, 74 (1990) (Stevens, J., concurring). See Blue Chip Stamps v. Manor Drug Stores, 421 U. S. 723, 733 (1975) (the “longstanding acceptance by the courts” and “Congress’ failure to reject” rule announced in landmark Court of Appeals decision favored retention of the rule).6 A policy of respect for consistent judicial and administrative interpretations leaves it to elected representatives to assess settled law and to evaluate the merits and demerits of changing it.7 Even when there is no affirmative evidence of rati*197fication, the Legislature’s failure to reject a consistent judicial or administrative construction counsels hesitation from a court asked to invalidate it. Cf. Burnet v. Coronado Oil & Gas Co., 285 U. S. 393, 406 (1932) (Brandeis, J., dissenting). Here, however, the available evidence suggests congressional approval of aider and abettor liability in private § 10(b) actions. In its comprehensive revision of the Exchange Act in 1975, Congress left untouched the sizable body of case law approving aiding and abetting liability in private actions under § 10(b) and Rule 10b-5.8 The case for leaving aiding *198and abetting liability intact draws further strength from the fact that the SEC itself has consistently understood § 10(b) to impose aider and abettor liability since shortly after the rule’s promulgation. See Ernst & Young, 494 U. S., at 75 (Stevens, J., concurring). In short, one need not agree as an original matter with the many decisions recognizing the private right against aiders and abettors to concede that the right fits comfortably within the statutory scheme, and that it has become a part of the established system of private enforcement. We should leave it to Congress to alter that scheme.
The Court would be on firmer footing if it had been shown that aider and abettor liability “detracts from the effectiveness of the 10b-5 implied action or interferes with the effective operation of the securities laws.” See Musick, Peeler & Garrett v. Employers Ins. of Wausau, 508 U. S. 286, 298 (1993). However, the line of decisions recognizing aider and abettor liability suffers from no such infirmities. The language of both § 10(b) and Rule 10b-5 encompasses “any person” who violates the Commission’s antifraud rules, whether “directly or indirectly”; we have read this “broad” language “not technically and restrictively, but flexibly to effectuate its remedial purposes.” Affiliated Ute Citizens of Utah v. United States, 406 U. S. 128, 151 (1972). In light of the encompassing language of § 10(b), and its acknowledged purpose to strengthen the antifraud remedies of the common law, it was certainly no wild extrapolation for courts to conclude that aiders and abettors should be subject to the *199private action under § 10(b).9 Allowing aider and abettor claims in private § 10(b) actions can hardly be said to impose unfair legal duties on those whom Congress has opted to leave unregulated: Aiders and abettors of § 10(b) and Rule 10b-5 violations have always been subject to criminal liability under 18 U. S. C. §2. See 15 U. S. C. § 78ff (criminal liability for willful violations of securities statutes and rules promulgated under them). Although the Court canvasses policy arguments against aider and abettor liability, ante, at 188-190, it does not suggest that the aiding and abetting theory has had such deleterious consequences that we should dispense with it on those grounds.10 The agency charged with primary responsibility for enforcing the securities laws does not perceive such drawbacks, and urges retention of the private right to sue aiders and abettors. See Brief for SEC as Amicus Curiae 5-17.
As framed by the Court’s order redrafting the questions presented, this case concerns only the existence and scope of aiding and abetting liability in suits brought by private parties under § 10(b) and Rule 10b-5. The majority’s rationale, *200however, sweeps far beyond even those important issues. The majority leaves little doubt that the Exchange Act does not even permit the SEC to pursue aiders and abettors in civil enforcement actions under § 10(b) and Rule 10b-5. See ante, at 177 (finding it dispositive that “the text of the 1934 Act does not itself reach those who aid and abet a § 10(b) violation”). Aiding and abetting liability has a long pedigree in civil proceedings brought by the SEC under § 10(b) and Rule 10b-5, and has become an important part of the SEC’s enforcement arsenal.11 Moreover, the majority’s approach to aiding and abetting at the very least casts serious doubt, both for private and SEC actions, on other forms of secondary liability that, like the aiding and abetting theory, have long been recognized by the SEC and the courts but are not expressly spelled out in the securities statutes.12 *201The principle the Court espouses .today — that liability may not be imposed on parties who are not within the scope of § 10(b)’s plain language — is inconsistent with long-established SEC and judicial precedent.
As a general principle, I agree, “the creation of new rights ought to be left to legislatures, not courts.” Musick, Peeler, 508 U. S., at 291. But judicial restraint does not always favor the narrowest possible interpretation of rights derived from federal statutes. While we are now properly reluctant to recognize private rights of action without an instruction from Congress, we should also be reluctant to lop off rights of action that have been recognized for decades, even if the judicial methodology that gave them birth is now out of favor. Caution is particularly appropriate here, because the judicially recognized right in question accords with the longstanding construction of the agency Congress has assigned to enforce the securities laws. Once again the Court has refused to build upon a “‘secure foundation . . . laid by others,’ ” Patterson v. McLean Credit Union, 491 U. S. 164, 222 (1989) (Stevens, J., dissenting) (quoting B. Cardozo, The Nature of the Judicial Process 149 (1921)).
I respectfully dissent.
See, e. g., Cleary v. Perfectune, Inc., 700 F. 2d 774, 777 (CA1 1983); IIT v. Cornfeld, 619 F. 2d 909, 922 (CA2 1980); Monsen v. Consolidated Dressed Beef Co., 579 F. 2d 793, 799-800 (CA3 1978); Schatz v. Rosenberg, 943 F. 2d 485, 496-497 (CA4 1991); Fine v. American Solar King Corp., 919 F. 2d 290, 300 (CA5 1990); Moore v. Fenex, Inc., 809 F. 2d 297, 303 (CA6), cert. denied sub nom. Moore v. Frost, 483 U. S. 1006 (1987); Schlifke v. Seafirst Corp., 866 F. 2d 935, 947 (CA7 1989); K & S Partnership v. Continental Bank, N. A, 952 F. 2d 971, 977 (CA8 1991); Levine v. *193Diamanthuset, Inc., 950 F. 2d 1478, 1483 (CA9 1991); Farlow v. Peat, Marwick, Mitchell & Co., 956 F. 2d 982, 986 (CA10 1992); Schneberger v. Wheeler, 859 F. 2d 1477, 1480 (CA11 1988). The only court not to have squarely recognized aiding and abetting in private § 10(b) actions has done so in an action brought by the SEC, see Dirks v. SEC, 681 F. 2d 824, 844 (CADC), rev’d on other grounds, 463 U. S. 646 (1983), and has suggested that such a claim was available in private actions, see Zoelsch v. Arthur Andersen & Co., 824 F. 2d 27, 35-36 (CADC 1987). The Seventh Circuit’s test differs markedly from the other Circuits’ in that it requires that the aider and abettor “commit one of the ‘manipulative or deceptive’ acts prohibited under section 10(b) and rule 10b-5.” Robin v. Arthur Young & Co., 915 F. 2d 1120, 1123 (CA7 1990).
When § 10(b) was enacted, aiding and abetting liability was widely, albeit not universally, recognized in the law of torts and in state legislation prohibiting misrepresentation in the marketing of securities. See, e. g., 1 T. Cooley, Law of Torts 244 (3d ed. 1906) (“All who actively participate in any manner in the commission of a tort, or who command, direct, advise, encourage, aid or abet its commission, are jointly and severally liable therefor”). Section 16(1) of the Uniform Sale of Securities Act, 9 U. L. A. 385 (1932), conferred a right to sue aiders and abettors of securities fraud, as did the blue sky laws of 11 States. See Abrams, The Scope of Liability Under Section 12 of the Securities Act of 1933: “Participation” and the Pertinent Legislative Materials, 15 Ford. Urb. L. J. 877, 945 (1987). The courts’ reliance on common-law tort principles in defining the scope of liability under § 10(b) was by no means an anomaly. See, e. g., American Soc. of Mechanical Engineers, Inc. v. Hydrolevel Corp., 456 U. S. 556, 565-574 (1982).
Compare, for example, the discussion in the opinion below of scienter in cases in which defendant has no disclosure duty, 969 F. 2d 891, 902-903 (CA10 1993), with that in Schatz v. Rosenberg, 943 F. 2d 485, 496 (CA4 1991), and Ross v. Bolton, 904 F. 2d 819, 824 (CA2 1990). See also Kuehnle, Secondary Liability Under The Federal Securities Laws — Aiding and Abetting, Conspiracy, Controlling Person, and Agency: Common-Law Principles and The Statutory Scheme, 14 J. Corp. L. 313, 323-324, and n. 53 (1988).
“As I have said before, ‘the adversary process functions most effectively when we rely on the initiative of lawyers, rather than the activism of judges, to fashion the questions for review.’ New Jersey v. T. L. O., 468 U. S. 1214, 1216 (1984) (dissenting from order directing reargument).” Patterson v. McLean Credit Union, 485 U. S. 617, 623 (1988) (Stevens, J., dissenting from order directing reargument).
See Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Curran, 456 U. S. 353, 374-378 (1982); Middlesex County Sewerage Authority v. National Sea Clammers Assn., 453 U. S. 1, 22-25 (1981) (Stevens, J., concurring in judgment in part and dissenting in part); California v. Sierra Club, 451 U. S. 287, 298-301 (1981) (Stevens, J., concurring). A discussion of the common-law presumption is found in Justice Pitney’s opinion for the Court in Texas & Pacific R. Co. v. Rigsby, 241 U. S. 33, 39-40 (1916). See also, e. g., Texas & New Orleans R. Co. v. Railway Clerks, 281 U. S. 548, 568-570 (1930).
None of the cases the majority relies upon to support its strict construction of § 10(b), ante, at 173-175, even arguably involved a settled course of lower court decisions. See Mertens v. Hewitt Associates, 508 U. S. 248 (1993); Pinter v. Dahl, 486 U. S. 622, 635, n. 12 (1988); Chiarella v. United States, 445 U. S. 222, 229, n. 11 (1980); Sante Fe Industries, Inc. v. Green, 430 U. S. 462, 475-476, n. 15 (1977); Ernst & Ernst v. Hochfelder, 425 U. S. 185, 191-192, n. 7 (1976).
Of course, when a decision of this Court upsets settled law, Congress may step in to reinstate the old law, cf. Securities Exchange Act § 27A, as added by Pub. L. 102-242, §476,105 Stat. 2236,2387, codified at 15 U. S. C. §78aa-l (1988 ed., Supp. IV) (providing that relevant state limitations period should govern actions pending when Lampf Pleva, Lipkind, Prupis & Petigrow v. Gilbertson, 501 U. S. 350 (1991), came down). However, we should not lightly heap new tasks on the Legislature’s already full plate. Moreover, congressional efforts to address the problems posed by judicial decisions that disrupt settled law frequently create special difficulties of their own. See, e. g., Plant v. Spendthrift Farm, Inc., 1 F. 3d 1487 (CA6 1993) (holding §27A unconstitutional), cert, pending, No. 93-1121; Pacific Mut. Life Ins. Co. v. First RepublicBank Corp., 997 F 2d 39 (CA5 1993) (upholding it), cert. granted, 510 U. S. 1039 (1994). See also Rivers v. Roadway Express, Inc., post, at 304-313.
By 1975, the renowned decision in Brennan v. Midwestern United Life Ins. Co., 259 F. Supp. 673, 680 (ND Ind. 1966), had been on the books almost a decade and several Courts of Appeals had recognized aider and abettor liability in private actions brought under § 10(b) and Rule 10b-5. See Kerbs v. Fall River Industries, Inc., 502 F. 2d 731, 739-740 (CA10 1974); Landy v. FDIC, 486 F. 2d 139, 162-163 (CA3 1973), cert. denied, 416 U. S. 960 (1974); Strong v. France, 474 F. 2d 747, 752 (CA9 1973); Buttrey v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 410 F. 2d 135, 144 (CA7), cert. denied, 396 U. S. 838 (1969). See also Lanza v. Drexel & Co., 479 F. 2d 1277, 1301, 1303-1304 (CA2 1973) (en banc); Ruder, Multiple Defendants in Securities Law Fraud Cases: Aiding and Abetting, Conspiracy, In Pari Delicto, Indemnification, and Contribution, 120 U. Pa. L. Rev. 597, 620-638 (1972). We have noted the significance of the 1975 amendments in another case involving a “consistent line of judicial decisions” on the implied right of action under § 10(b) and Rule 10b-5. See Herman & MacLean v. Huddleston, 459 U. S. 375, 384-386 (1983). Those amendments emerged from “ ‘the most searching reexamination of the competitive, statutory, and economic issues facing the securities markets, the securities industry, and, of course, public investors, since the 1930’s.’” Id., at 385, n. 20 (quoting H.R. Conf. Rep. No. 94-229, p. 91 (1975)).
Congress’ more recent visits to the securities laws also suggest approval of the aiding and abetting theory in private § 10(b) actions. The House Report accompanying an aiding and abetting provision of the 1983 Insider Trading Sanctions Act, see 15 U. S. C. § 78u(d)(2)(A) (1982 ed., Supp. V), contains an approving reference to “judicial application of the concept of aiding and abetting liability to achieve the remedial purposes of the securities laws,” H. R. Rep. No. 98-355, p. 10 (1983), and notes with favor Rolfv. Blyth, Eastman Dillon & Co., 570 F. 2d 38 (CA2), cert, denied, 439 U. S. 1039 (1978), which affirmed a judgment against an aider and abettor in a private action under § 10(b) and Rule 10b-5. Moreover, § 5 of the *198Insider Trading and Securities Fraud Enforcement Act of 1988, Pub. L. 100-704, 102 Stat. 4681, contains an express “acknowledgment,” Musick, Peeler & Garrett v. Employers Ins. of Wausau, 508 U. S. 286, 294 (1993), of causes of action “implied from a provision of this title,” 15 U. S. C. § 78t-l(d).
In a similar context we recognized a private right of action against secondary violators of a statutory duty despite the absence of a provision explicitly covering them. See Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Curran, 456 U. S., at 394 (“Having concluded that exchanges can be held accountable for breaching their statutory duties to-enforce their own rules prohibiting price manipulation, it necessarily follows that those persons who are participants in a conspiracy to manipulate the market in violation of those rules are also subject to suit by futures traders who can prove injury from these violations”).
Indeed, the Court anticipates, ante, at 191, that many aiders and abettors will be subject to liability as primary violators. For example, an accountant, lawyer, or other person making oral or written misrepresentations (or omissions, if the person owes a duty to the injured purchaser or seller, cf. Dirks v. SEC, 463 U. S. 646, 654-655 (1983)) in connection with the purchase or sale of securities may be liable for a primary violation of § 10(b) and Rule 10b-5. See, e. g., Akin v. Q-L Investments, Inc., 959 F. 2d 521, 525-526 (CA5 1992).
See, e. g., SEC v. Coffey, 493 F. 2d 1304, 1316 (CA6 1974); Ruder, 120 U. Pa. L. Rev., at 625-626, nn. 124 and 125. The SEC reports that it asserted aiding and abetting claims in 15 percent of its civil enforcement proceedings in fiscal year 1992, and that elimination of aiding and abetting liability would “sharply diminish the effectiveness of Commission actions.” Brief for SEC as Amicus Curiae 18, n. 15.
The Court’s rationale would sweep away the decisions recognizing that a defendant may be found liable in a private action for conspiring to violate § 10(b) and Rule 10b-5. See, e. g., U. S. Industries, Inc. v. Touche Ross & Co., 854 F. 2d 1223, 1231 (CA10 1988); SEC v. Coffey, 493 F. 2d 1304, 1316 (CA6 1974); Ferguson v. Omnimedia, Inc., 469 F. 2d 194, 197-198 (CA1 1972); Shell v. Hensley, 430 F. 2d 819, 827, n. 13 (CA5 1970); Dasho v. Susquehanna Corp., 380 F. 2d 262, 267, n. 2 (CA7), cert. denied sub nom. Bard v. Dasho, 389 U. S. 977 (1967). See generally Kuehnle, 14 J. Corp. L., at 343-348. Secondary liability is as old as the implied right of action under § 10(b) itself; the very first decision to recognize a private cause of action under the section and rule, Kardon v. National Gypsum Co., 69 F. Supp. 512 (ED Pa. 1946), involved an alleged conspiracy. See also Fry v. Schumaker, 83 F. Supp. 476, 478 (ED Pa. 1947) (Kirkpatrick, C. J.). In addition, many courts, concluding that §20(a)’s “controlling person” provisions, 15 U. S. C. § 78t, are not the exclusive source of secondary liability under the Exchange Act, have imposed liability in § 10(b) actions based upon respondeat superior and other common-law agency principles. See, e. g., Hollinger v. Titan Capital Corp., 914 F. 2d 1564, 1576-1577, and *201n. 27 (CA9 1990) (en banc) (citing and following decisions to this effect from six other Circuits). See generally Kuehnle, 14 J. Corp. L., at 350-376. These decisions likewise appear unlikely to survive the Court’s decision. See ante, at 184.