dissenting from the denial of rehearing en banc, with whom PREGERSON and TASHIMA, Circuit Judges, join, and with whom MICHAEL DALY HAWKINS and GRABER, Circuit Judges, join with respect to Part I:

By failing to take this case en banc, this court appears to endorse a decision of a divided panel that ignores the plain directives of Congress, casts aside the prior decisions of this court, and creates a striking conflict with our fellow circuits. Because I believe that we have a duty to correct an opinion that ignores the authority of Congress’s words, the dictates of stare decisis, and the uniform conclusion of other circuit courts, I dissent from the denial of rehearing en banc.

I.

The excellent dissent from the panel opinion by Judge Browning more than adequately describes the myriad errors made by the panel majority. I write separately only to emphasize a few additional points. First, contrary to the majority’s claim that at the time it wrote “there [was] widespread disagreement among courts as to the proper interpretation of the PSLRA’s pleading requirement,” circuit courts were in fact unanimous about what the statute did not do. As even the majority’s own description of the cases from other circuits makes clear, see In re Silicon Graphics Inc. Sec. Litig., 183 F.3d 970, 974 (9th Cir.1999), every other circuit to have interpreted the PSLRA’s pleading requirement had concluded that the statute did not overturn the well-established, preexisting law governing the mental state required to establish securities fraud. 1 In fact, until this court decided Silicon Graphics, every circuit court to consider the issue was in agreement that under the PSLRA, as before, a showing of “recklessness” sufficed to establish § 10(b) Lability. See Hoffman v. Comshare, Inc. (In re Comshare, Inc. Sec. Litig.), 183 F.3d 542, 550 (6th Cir.1999) (holding that evidence of “recklessness” still suffices to prove scienter); In re Advanta Corp. Sec. Litig., 180 F.3d 525, 535 (3rd Cir.1999) (same); Press v. Chemical Inv. Serv. Corp., 166 F.3d 529 , 538 (2nd *523 Cir.1999) (same). Thus, our circuit was the first to arrive at the remarkable conclusion that proving recklessness is no longer sufficient.

We did not start a trend, however. Since the Silicon Graphics panel issued its opinion, the Eleventh Circuit has explicitly rejected the panel’s conclusion and followed the well-reasoned analyses of other circuits. In so doing, it noted our outlier status: “The opinion of the Ninth Circuit in Silicon Graphics would seem to indicate that the Reform Act substantively raised the required level of scienter, while the Second, Third, and Sixth Circuits hold that fact-specific allegations of recklessness are still sufficient.” Bryant v. Avado Brands, Inc., 187 F.3d 1271, 1283 (11th Cir.1999). After noting that “[ejvery circuit to address the question before the passage of the Reform Act held that a showing of recklessness was sufficient to allege scien-ter,” and discussing some of the compelling evidence that demonstrates that Congress did not intend the PSLRA to alter the scienter requirement, the Eleventh Circuit concluded: “We are persuaded that the plain text of the statute makes clear that recklessness was not eliminated as a basis for liability under the Reform Act.” Bryant, 187 F.3d at 1284 . By siding with the Second, Third, and Sixth Circuits, the Eleventh Circuit left this court out in the cold, as the only circuit to conclude in the face of compelling evidence to the contrary that “recklessness” is no longer sufficient to establish liability under § 10(b).

It is difficult to know why the majority, in interpreting a statutory enactment that so clearly is limited to dealing with the particularity of pleadings, decided to change the substantive standard governing securities fraud liability. It is similarly difficult to understand why the majority would create a new term — and with it a wholly new and untested standard — to replace a term commonly accepted and understood by all. Prior to the enactment of the PSLRA, circuits around the country had all concluded that “recklessness” described the appropriate standard for liability under § 10(b), and in enacting the new statute Congress exhibited no intention of changing that rule. To the contrary, the language and legislative history of the Act clearly reflect Congress’s intention to retain the existing standard. In spite of this, a panel of this court has now decided that it prefers a new standard — “deliberate recklessness” — and we are left to speculate about how that somewhat anomalous term will be construed, and misconstrued, in the future. For all its ambiguity and peculiarity, however, two things are clear: (1) the “deliberate recklessness” standard is deliberately designed to make it more difficult for innocent persons injured by the reckless conduct of the issuers of securities to obtain recoveries, and (2) the substantive change in the law was made not by Congress but by a panel of this court that substituted its own policy views for those of the legislative branch.

II.

This court’s decision not to convene an en banc court to correct the panel’s erroneous holding in this highly controversial and exceptionally important case is equally difficult to understand. In light of our internal rules, however, the only fact the public will be able to glean from the minimalist order we issue is that affirmative votes to go en banc were not cast by a majority of the active nonreeused members of the court. We are not permitted to reveal the identities of judges who support and oppose particular en banc calls, and we are even prohibited from telling the public whether a majority of the eligible judges were recorded as being opposed to rehearing. 2 Information of this nature *524 should not be withheld: it is important and useful to those interested in our decisions, whether as actual or prospective litigants, or as biased or unbiased commentators on our court’s operations. The public is entitled to know whether an en banc hearing was denied because the court was equally divided, because a bare majority opted not to invoke the en banc process, or because the court voted overwhelmingly not to go en banc. 3 Unfortunately, the number of signatories to a dissent from the denial of en banc review ordinarily provides little significant information. A number of judges, as a matter of principle, do not join in en banc dissents even when they have voted unsuccessfully to go en banc and agree with the content of the dissent.

In short, although I will not rehash here the arguments I have made on a number of occasions, see United States v. Koon, 45 F.3d 1303, 1308-10 (9th Cir.1995) (Reinhardt, J., dissenting from the denial of en banc review); Elder v. Holloway, 984 F.2d 991, 1001-02 (9th Cir.1993) (same); Brewer v. Lewis, 997 F.2d 550, 556 (9th Cir. 1993) (same); Harris v. Vasquez, 949 F.2d 1497, 1539-41 (9th Cir.1990) (same), I feel compelled to reiterate that “the public has a right to know how all public officials, including judges, conduct the public’s business.” Koon, 45 F.3d at 1310 . When we refuse to disclose our votes in important cases, we wrongfully deprive the public of information to which it is entitled, and improperly shield our individual actions from public scrutiny.

. As the Silicon Graphics majority was forced to acknowledge, the preexisting law of this court included an en banc opinion that squarely held that a showing of “recklessness” was sufficient to establish liability under § 10(b). See Hollinger v. Titan Capital Corp., 914 F.2d 1564 (9th Cir.1990). In Hol-linger, we included a detailed description of what constitutes reckless conduct, and nowhere does the description hint that the proper standard is actually something called "deliberate recklessness.” See id. at 1569 . To the contrary, the new “deliberate recklessness” standard was simply invented by the Silicon Graphics panel.

. A negative ballot and a failure to vote are both recorded as "no” votes. See General Order § 5.5.b ("A judge's failure to vote shall be considered a no’ vote.”). A vote fails unless more than half of the nonreeused active judges cast affirmative votes. See General Order § 5.5.c.

. It may also be important in some instances to know whether the votes of one or more judges who were recused in a particular case, but not in similar cases, might have made the difference, at least from a mathematical standpoint.