concurring specially.

While I concur in the decision of the majority, I do not join in that section of the *383 opinion which addresses the issue of duty to disclose. This case provides the circuit with an opportunity to clarify its position on the effect that Ernst & Ernst v. Hochfelder, 425 U.S. 185 , 96 S.Ct. 1375 , 47 L.Ed.2d 668 (1976), and Nelson v. Serwold, 576 F.2d 1332 (9th Cir.), cert. denied, 439 U.S. 970 , 99 S.Ct. 464 , 58 L.Ed.2d 431 (1978), had on our 1974 decision in White v. Abrams, 495 F.2d 72 (9th Cir. 1974).

In the three years since the Supreme Court decided Hochfelder , overruling the “flexible duty” negligence standard of White v. Abrams in favor of a form of scienter involving more than negligence, this circuit has addressed the question of White v. Abrams ’ continuing vitality several times — with varying results.

Initially, this circuit seemed to take the position that the Supreme Court had dealt a death blow to White v. Abrams in footnote 12 of Hochfelder , where the Court listed the decision as an example of the negligence standard it was repudiating. In Robinson v. Heilman, this court held that the scienter standard encompassed some of the factors from the duty analysis, but that there should not be a separate duty inquiry. Robinson v. Heilman, 563 F.2d 1304 (9th Cir. 1977). The court stated: “Of course, the closer the relationship of the person charged to the corporation and the greater his participation in the transactions attacked, the easier it will be to prove the requisite scienter. That state of affairs, however, provides no justification for any inference that direct participants have any greater or different duty to buyers or sellers of securities than those whose connections with the transactions may be more remote.” Id. at 1308 .

Then, the circuit resurrected White in Crocker Citizens National Bank v. Control Metals Corp., 566 F.2d 631 (9th Cir. 1977). The court held that “[ajbsent the possibility of liability based upon negligence, Hochfelder is not inconsistent, on its facts, with the flexible duty standard established in White v. Abrams.” Id. at 636 n. 2. The circuit reiterated the Crocker view in Zweig v. Hearst, 594 F.2d 1261 , 1268 n. 13 (9th Cir. 1979), and again in Kidwell ex rel. Penfold v. Meikle, 597 F.2d 1273, 1294 (9th Cir. 1979); yet .it took a different approach in Ritzau v. Warm Springs West and treated the standard as “rejected”:

Ritzau v. Warm Springs West, 589 F.2d 1370, 1374 (9th Cir. 1979). In addition, when the circuit decided that recklessness satisfies Hochfelder’s scienter requirements, it made no mention of a separate duty analysis. Nelson v. Serwold, supra, 576 F.2d 1332 .

Two district courts in the Ninth Circuit have faced the Hochfelder/White v. Abrams issue. In the first case, which predated the opinions listed above, the court held that the flexible duty test and the scienter requirements of Hochfelder were each prerequisites to 10b-5 liability. Carr v. New York Stock Exchange, 414 F.Supp. 1292, 1300 (N.D.Cal.1976). More recently, a district court observed that “[b]ecause [the] flexible duty standard permitted liability in certain circumstances based upon negligence alone, its precedential value is in question after Hochfelder.” In re Gap Stores Securities Litigation, 457 F.Supp. 1135, 1140-41 (N.D.Cal.1978). After comparing Crocker-Citizens National Bank v. Control Metals Corp., supra, with Robinson v. Heilman, supra, the court concluded: “Until the Ninth Circuit explicitly re-examines White v. Abrams , however, it should be followed to the extent it is not directly inconsistent with Hochfelder.” In re Gap Stores Securities Litigation, supra, at 1141.

I believe that a re-examination of the flexible duty standard in light of Hochfelder shows that the time has come to lay White v. Abrams to rest. The five-factor duty analysis helps a court evaluate the foreseeability of harm to a plaintiff from the defendant’s actions and the level of *384 culpability of the defendant’s conduct. These evaluations are relevant, and even necessary, to a finding of liability based on negligence. After Hoehfelder, supra, and Nelson v. Serwold, supra, however, 10b-5 liability requires proof of recklessness, and the duty analysis is no longer appropriate.

This court has not defined recklessness in the context of securities violations, 1 but the Second and Seventh Circuits have described “[r]eckless conduct [as], at the least, conduct which is ‘highly unreasonable’ and which represents ‘an extreme departure from the standards of ordinary care . to the extent that the danger was either known to the defendant or so obvious that the defendant must have been aware of it.’ ” 2 Rolf v. Blyth, Eastman Dillon & Co., 570 F.2d 38, 47 (2d Cir. 1978), quoting Sanders v. John Nuveen & Co., 554 F.2d 790, 793 (7th Cir. 1977).

If this circuit adopts a similar definition of recklessness, the scienter requirements of Hoehfelder and Nelson will be satisfied when a court concludes that a defendant’s conduct was highly unreasonable and presented a known or highly obvious danger to the plaintiff. If the circuit should adopt a different definition, the scienter requirements will be satisfied when the test set forth in that definition is met. There is no reason to require the court then to engage in a complex, multi-facted duty analysis which was designed to test negligence.

It may be useful at this point to consider a bit of the history of White v. Abrams. Before Hoehfelder established that negligence could not be the basis for 10b-5 liability, the circuit courts wrestled for years with different state of mind requirements. When this circuit decided White v. Abrams , Professor Bromberg hailed the five-factor flexible duty standard as “an important break from traditional verbalisms, . promising] emphasis on more significant aspects of cases.” A. Bromberg, 4 Securities Law, Fraud § 8.4(500) (Supp.1975).

The verbalisms from which White departed were those of common law fraud which had dominated 10b-5 law up to that time. The White court’s departure was based, in part, on Affiliated Ute Citizens v. United States, 406 U.S. 128 , 92 S.Ct. 1456 , 31 L.Ed.2d 741 (1972), where the Supreme Court had held that causation was established without proof of reliance, where the defendants had a duty to disclose but nonetheless withheld material facts. Id. at 154 , 92 S.Ct. 1472 . This court saw Ute as disregarding common law fraud elements for an analysis of “what kind of a duty rule 10b-5 imposes.” White v. Abrams, supra, 495 F.2d at 731 . The court stated:

Id. at 731-32.

Unfortunately, the duty standard in application sometimes does mask complex determinations. The five factors that the White court set out for consideration are:

Id. at 735-36 (footnotes omitted).

Not all factors are necessary to a finding of liability in a given case. See, e. g., Kidwell ex rel. Penfold v. Meikle, supra, 597 F.2d at 1296-97 ; Zweig v. Hearst, supra, 594 F.2d at 1269 . Instead, a court applying the test must “[w]eigh[] all these facts in light of the principles of full disclosure that Rule 10b-5 was meant to foster.” Kidwell ex rel. Penfold v. Meikle, supra, 597 F.2d at 1297 .

A court examining elements such as the relationship of defendant to plaintiff and plaintiff’s and defendant’s relative access to information must weigh many of the same elements that figure into the traditional concepts of materiality and causation. 3 For example, whether a statement could reasonably be expected to influence the decision to sell and was therefore material will be affected by the two factors just listed as well as by defendant’s awareness of plaintiff’s reliance. Factors (3) and (5), which go to the defendant’s culpability, may not be subsumed as clearly within the common law standards, but the two considerations may figure somewhat in the scienter analysis. To the extent that they are not thus incorporated, the factors only serve to confuse the liability determination. There is nothing in Hochfelder to suggest that a defendant who intentionally causes another to suffer damages through the purchase or sale of a security should be exonerated because he does not benefit financially from the relationship.

Application of the factors sometimes produces awkward results in other ways, as well. This case demonstrates the problem. Factor one looks to the relationship between defendant and plaintiff, properly contemplating a greater likelihood of liability where plaintiff and defendant are in a fiduciary relationship. See Rolf v. Blyth, Eastman Dillon & Co., Inc., supra, 570 F.2d at 44 . There is no real relationship between plaintiff and defendant here, and there seldom would be one unless defendant were the buyer, seller or broker. Thus the court is confronted with the necessity of manufacturing the relationship from a tenuously defined benefit that the defendant receives from the transaction, as here, or dropping the requirement altogether, as it did in Zweig v. Hearst Corp., 594 F.2d 1261, 1267 (9th Cir. 1979).

We may avoid this problem quite easily by recognizing that the flexible duty negligence standard is just that — a negligence standard. We have followed the Supreme Court’s dictates in Hochfelder and required scienter instead of negligence. We should recognize that it is, therefore, no longer appropriate to apply a duty standard that was designed to compartmentalize and simplify a negligence inquiry. As does the case before us, Hochfelder addressed the liability of an accounting firm for omissions made in connection with the sale of a security. The Supreme Court inquired into scienter and made no separate duty inquiry, and we should do the same.

. In Nelson v. Serwold, supra, 576 F.2d at 1338 , we held only that “[t]he evidence supports a finding of recklessness, or some degree of intent not sufficiently aggravated to be characterized as ‘deliberate and cold-blooded.’ ”

. The Second Circuit left open the question whether it would find 10b-5 liability when recklessness was determined under the less strict test established in Stern v. American Bankshares Corp., 429 F.Supp. 818, 826 (E.D.Wis.1977): “plaintiff must allege that the defendants knew or should have known of the facts and circumstances concerning the fraud.” See Rolf v. Blyth, Eastman Dillon & Co., Inc., 570 F.2d at 47 n. 16.

. These concepts remain central to 10b-5 liability determinations. The Supreme Court framed the duty approach in Ute in terms of materiality and causation, Affiliated Ute Citizens v. United States, supra, 406 U.S. at 154 , 92 S.Ct. 1456 ; and the lower courts continue to require these elements. See, e. g., Zweig v. Hearst Corp., 594 F.2d 1261, 1266 (9th Cir. 1979); Nelson v. Serwold, supra, 576 F.2d at 1335 ; St. Louis Union Trust Co. v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 562 F.2d 1040, 1048 (8th Cir. 1977); Valente v. Pepsico, Inc., 454 F.Supp. 1228, 1236 (D.Del.1978).