concurring in the judgment:
I agree with my colleagues that the SEC’s order cannot stand because Howard did not act, or fail to act, with the requisite scienter of an aider and abettor. I do not agree with the majority’s formulation of the requisite scienter, however, and I therefore concur in the judgment.
In the usual aiding and abetting scenario, we ask three questions: whether “1) another party has committed a securities law violation; 2) the accused aider and abetter had a general awareness that his role was part of an overall activity that was improper; and 3) the accused aider and abetter knowingly and substantially assisted the principal violation.” Investors Research Corp. v. SEC, 628 F.2d 168, 178 (D.C.Cir.1980); see also Dirks v. SEC, 681 F.2d 824, 844 (D.C.Cir.1982), rev’d on other grounds, 463 U.S. 646, 103 S.Ct. 3255, 77 L.Ed.2d 911 (1983). As the majority opinion notes, Maj. Op. at 1143, here the SEC found that Howard’s awawareness that his role was part of improper activity fulfilled the second Investors Research element. I believe the SEC’s finding in this respect is not supported by substantial evidence, see 15 U.S.C. § 78y(a)(4), because Howard was not “recklessly” ignorant and therefore, under our precedent, including Graham v. SEC, 222 F.3d 994 (D.C.Cir.2000), inter alia, we must vacate the SEC’s order. Where I part company with the majority is in its apparent use of two distinct lines of authority regarding recklessness - one applying to a securities violation, whether committed by a primary actor or by an aider and abettor, the other applying to the second Investors Research element of “general awareness of wrong*1151doing” - to announce a new, and I believe, incorrect, scienter level to satisfy the latter.
We have held in the securities area that willfulness can support a primary violation, Wonsover v. SEC, 205 F.3d 408, 416 (D.C.Cir.2000) (concluding that “substantial evidence supports the [SEC]’s determination that Wonsover failed to conduct reasonable inquiry into the sources of the unregistered shares he sold and that his inadequate inquiry in the face of several ‘red flags’ justified a finding of willfulness” (emphasis added)), as can “extreme” recklessness. SEC v. Steadman, 967 F.2d 636, 641 (D.C.Cir.1992). In Steadman, we considered both primary violations of section 10(b) of the Securities Exchange Act and Rule 10b-5, inter alia, allegedly committed by Steadman and others, and Stead-man’s separate liability as an aider and abettor in the defendant corporation’s violations of certain SEC regulations. Relying on Investors Research, we stated that an aider and abettor must “ ‘knowingly and substantially assist[]’ in the commission of a securities violation, with at least ‘a general awareness that his role was part of an overall activity that was improper.’ ” Steadman, 967 F.2d at 647 (quoting Investors Research, 628 F.2d at 178 (alteration in original)).1
More recently, in Graham v. SEC, supra, the court articulated the test for aiding and abetting liability as follows:
Although variously formulated, three principal elements are required to establish liability for aiding and abetting a violation of section 10(b) and Rule 10b — 5: (1) that a principal committed a primary violation; (2) that the aider and abettor provided substantial assistance to the primary violator; and (3) that the aider and abettor had the necessary “scien-ter” — i.e., that she rendered such assistance knowingly or recklessly. See SEC v. Fehn, 97 F.3d 1276, 1287-88 (9th Cir. 1996); Bloor v. Carro, Spanbock, Londin, Rodman & Fass, 754 F.2d 57, 62 (2d Cir.1985); SEC v. Falstaff Brewing Corp., 629 F.2d 62, 72 (D.C.Cir.1980); Investors Research Corp. v. SEC, 628 F.2d 168, 178 (D.C.Cir.1980); see also SEC v. Steadman, 967 F.2d 636, 641 (D.C. Cir.1992).
222 F.3d at 1000. Graham’s third element “that [the aider and abettor] rendered such assistance knowingly or recklessly” can only be a reformulation of the Investors Research “general awareness of wrongdoing” element, both because Graham expressly relies on Investors Research in its articulation and because the other two parts of the Investors Research test are covered in Graham’s first and second elements. Graham later focuses on the “third element” of aiding and abetting liability:
The real question here concerns the third element of aiding and abetting liability: did Graham assist Broumas with *1152the requisite scienter? We have held that knowledge or recklessness is sufficient to satisfy that requirement. See Kowal v. MCI Communications Corp., 16 F.3d 1271, 1276 (D.C.Cir.1994); SEC v. Steadman, 967 F.2d 636, 641 (D.C.Cir. 1992); Zoelsch [v. Arthur Andersen & Co.], 824 F.2d [27,] 36 [(D.C.Cir.1987)]; Dirks v. SEC, 681 F.2d 824, 844-45 (D.C.Cir.1982), rev’d on other grounds, 463 U.S. 646, 103 S.Ct. 3255, 77 L.Ed.2d 911 (1983).... We are satisfied that Graham acted with at least extreme recklessness in aiding Broumas’ stock-kiting scheme.
222 F.3d at 1004. Graham’s use of “extreme” recklessness here and elsewhere, see id. at 1006 (“Given the abundance of red flags here, it would be very hard to characterize Graham’s conduct as anything but extremely reckless, regardless of the approvals she received.... ”), describes the extent of Graham’s recklessness; it does not impose a requirement of extreme recklessness to support the “third element” (Investors Research’s second element) of aiding and abetting liability. This reading is plain, most notably from Graham’s own recognition that “recklessness is sufficient” as well as its express reliance on Dirks. See supra n. 1.
The majority opinion, however, misreads both Steadman and Graham to “hold that ‘extreme recklessness’ may support [the second Investors Research element of] aiding and aiding liability.” Maj. Op. at 1143 (emphasis added). That “may” means “must” in the majority’s view - and that the majority is in fact addressing the second Investors Research element - is apparent from its subsequent discussion, particularly the following passage: “We are willing to assume that the SEC thought - incorrectly - that reckless conduct amounted to a form of awareness of wrongdoing. But we are unwilling to assume that it properly evaluated Howard’s under an extreme recklessness standard.” Maj. Op. at 1143. It ultimately concludes that the SEC improperly evaluated Howard’s conduct because it used “recklessness” rather than “extreme recklessness” as the requisite level of scienter. Maj. Op. at 1147. I believe its application of an “extreme” recklessness standard is wrong.
While I characterized the majority’s error as using two “distinct” lines of authority regarding recklessness, supra at 1150, one line is, at least to me, not altogether clear. Although we said in Steadman that “extreme” recklessness satisfies the intent requirement, we relied on Circuit precedent that held that recklessness suffices. Supra n. I.2 Is there a difference? The majority plainly thinks so. Whether the two terms in fact impose different standards in satisfying the “[Hochfelder] intent requirement,” Steadman, 967 F.2d at 641, or are merely descriptive of the degree of recklessness exhibited,3 we need not decide in this case because the separate standard - applicable to an element of aiding and abetting, but not primary, liability - for determining whether Howard’s lack of awareness of the primary violations of Rule 10b-9 is sanctionable is recklessness. This line of authority is distinct and clear. Graham, 222 F.3d at 1000, 1004; Dirks, 681 F.2d at 844; Steadman, 967 F.2d at 647; Investors Research, 628 F.2d at 178; see also Zoelsch, 824 F.2d at 36. The SEC, relying on Graham, J.A. 517, 522 *1153n.17, correctly applied the recklessness standard to Howard’s unawareness of the improper activity; its error lies in its conclusion that Howard’s ignorance was in fact reckless. For the foregoing reasons, I concur in the vacatur of the SEC’s order as well as the partial remand.
. Earlier in Steadman, we observed that "we have determined, along with a number of other circuits, that extreme recklessness may also satisfy this [Hochfelder] intent requirement.” 967 F.2d at 641 (emphasis added). The D.C. circuit precedent it then cited, however, never spoke of "extreme” recklessness. On the contrary, it stated repeatedly that "recklessness satisfies the 10b-5 scienter requirement.” Dirks v. SEC, 681 F.2d at 844; see also id. at 845 n. 28. Perhaps the Steadman court equated the Seventh Circuit’s "extreme departure from the standards of ordinary care” language with extreme recklessness. Sundstrand Corp. v. Sun Chem. Corp., 553 F.2d 1033, 1045 (7th Cir.1977) (internal quotation marks omitted and emphasis added). In any event, to the extent (if any) it intended to impose a higher degree of recklessness, it did not do so with respect to the "general awareness of wrongdoing” element of Steadman's aiding and abetting liability. That discussion, as noted above, adopted the Investors Research test.
. See also Kowal, 16 F.3d at 1276 ("To state a claim for securities fraud under Rule 10b-5, a plaintiff must allege that the defendant knowingly or recklessly made a false or misleading statement of material fact in connection with the purchase or sale of a security, upon which plaintiff reasonably relied, proximately causing his injury.”)
. See generally Marrie v. SEC, 374 F.3d 1196 (D.C.Cir.2004).