concurring.
Stenberg v. Carhart, 530 U.S. 914, 120 S.Ct. 2597, 147 L.Ed.2d 743 (2000), effectively held that the deeply disturbing — and morally offensive — destruction of the life of a partially born child cannot be banned by a legislature without an exception for the mother’s health (as determined by her doctor). It did so, despite the existence of other established methods of terminating such late-term pregnancies, upon the basis that some medical opinion exists to the effect that partial-birth abortion is safer for some women in some circumstances.
The underlying facts before us, consisting principally of a range of medical opinions, are not materially different from those before the Court in Stenberg; thus it is my duty to follow that precedent no matter how personally distasteful the fulfillment of that duty may be. I join Judge Newman’s carefully-crafted opinion accordingly. I write separately, however, to express certain concerns with the Supreme Court’s abortion jurisprudence generally and with Stenberg in particular.
I can think of no other field of law that has been subject to such sweeping constitutionalization as the field of abortion. Under the Supreme Court’s current jurisprudence, the legislature is all but foreclosed from setting policy regulating the practice; instead, federal courts must give their constitutional blessing to nearly every increment of social regulation that touches upon abortion — from gathering statistics about its frequency to establishing informed-consent standards that govern its use. In the process, the Supreme Court has sanctioned a mode of constitutional analysis in abortion cases that has removed the lower courts from their traditional role as arbiter of specific factual disputes and instead asked them to exercise their “gravest and most delicate duty,” the invalidation of a statute, Blodgett v. Holden, 275 U.S. 142, 148, 276 U.S. 594, 48 S.Ct. 105, 72 L.Ed. 206 (1927) (Holmes, J.), based upon a speculative showing that a statute might, in some yet-to-be-presented circumstance, have an unconstitutional application.
The Supreme Court’s decision in Stenberg exemplifies these larger problems. Faced with a statute that sought to ban a single method of abortion that many Americans — probably most Americans— find exceedingly offensive on moral *291grounds, the Court determined that, even though other methods of abortion are safe, a state cannot ban the procedure as long as it might be significantly safer for some unproven number of women. See Stenberg, 530 U.S. at 934-38, 120 S.Ct. 2597. The Stenberg Court’s holding is flawed in at least three respects: (1) it equates the denial of a potential health benefit (in the eyes of some doctors) with the imposition of a health risk and, in the process, promotes marginal safety above all other values, however worthy they may be; (2) it endorses a rule that permits the lower courts to hold a statute facially invalid upon a speculative showing of harm, even if, in the vast majority of cases, the statute’s application would not lead to an unconstitutional result; and (3) it establishes an evidentiary standard that all but removes the legislature from the field of abortion policy. I address each point in turn.
1. The Stenberg Court, By Equating The Denial Of A Potential Health Benefit With The Imposition Of A Health Risk, Has Promoted Marginal Safety Above All Other Values.
In Stenberg, the Court held that under the “governing standard” articulated in Casey, a statute must include “an exception ‘where it is necessary, in appropriate medical judgment for the preservation of the life or health of the mother.’ ” Id. at 931, 112 S.Ct. 2791 (quoting Planned Parenthood v. Casey, 505 U.S. 833, 879, 112 S.Ct. 2791, 120 L.Ed.2d 674 (1992)). The Court went on to explain that the requirement of a health exception is not limited to situations where the pregnancy itself endangers women’s health, but also includes situations where a statute that operates to preclude a particular method of abortion would pose a “significant health risk.” Id. As the Court explained, “Our cases have repeatedly invalidated statutes that in the process of regulating the methods of abortion, imposed significant health risks.” Id.; see also- id. (explaining that “a State cannot subject women’s health to significant risks ... where state regulations force women to use riskier methods of abortion” (emphasis added)). Based on these standards, the majority had little trouble concluding that “a statute that altogether forbids D & X creates a significant health risk.” Id. at 938, 112 S.Ct. 2791.
In reaching this conclusion, however, the Court never identified why a statute that altogether forbids D & X creates a significant health risk; it simply noted that, while other methods of abortion are “ ‘safe,’ ” some doctors believe that “the D & X method [is] significantly safer in certain circumstances.” Id. at 934, 112 S.Ct. 2791 (second emphasis in the original). Of course, this only establishes that a statute that altogether forbids D & X would deny some women a potential health benefit over an objectively “safe” baseline; it does not establish that such a statute would pose a constitutionally significant health risk.
By conceding that in all circumstances there are objectively “safe” alternatives to the D & X procedure, the Stenberg Court announced a rule that places relative safety above all other values. See Stenberg, 530 U.S. at 934-35, 120 S.Ct. 2597. Under the Court’s Logic, other available procedures might be deemed safe — even safer than natural childbirth — but if there is a marginally safer alternative in the opinion of some credible professionals, the state must make it available, no matter how morally repugnant society deems that method. This fundamental flaw — holding that the denial of a marginal health benefit constitutes the imposition of a significant health risk — permeates the Stenberg decision and renders it at odds with Casey.
*292I cannot square the undue burden standard of Casey with a holding that, while conceding the existence of alternative safe procedures, denies legislatures the ability to promote important interests above the conferral upon some citizens of a marginal health benefit. Certainly, nothing in Casey, which Stenberg purports to apply, compels such a result. Indeed, Casey expressly held that even though Roe v. Wade spoke of the state’s “important and legitimate interest in potential life,” 410 U.S. 113, 163, 93 S.Ct. 705, 35 L.Ed.2d 147 (1973), “[t]hat portion of the decision in Roe [had] been given too little acknowledgment and implementation by the Court in its subsequent cases,” Casey, 505 U.S. at 871, 112 S.Ct. 2791. And the decision went further still. Casey also established that the government has a legitimate interest in informing a woman about the “consequences to the fetus” during an abortion, “even when those consequences have no direct relation to [a woman’s] health,” id. at 882, 112 S.Ct. 2791, and even though the 24-hour waiting-period associated with those informed-consent provisions could at least arguably impose an increased health risk in some situations, see id. at 886-87, 112 S.Ct. 2791. Simply put, Casey permitted the state to advance important governmental interests, provided that in promoting those interests, it does not deprive women of their right to terminate a pregnancy by way of an objectively safe abortion procedure.8 Cf. id. at 875, 112 S.Ct. 2791 (stating that, prior to Casey, the Court erred in “striking down ... abortion regulations which in no real sense deprived women of [their] ultimate decision”). After Stenberg, lower courts can only guess as to how the Casey standard should be applied.9
*2932. The Stenberg Court Endorsed A Rule That Permits The Lower Courts To Hold A Statute Facially Invalid Upon A Speculative Showing Of Harm, Even If, In The Vast Majority Of Cases, The Statute’s Application Would Not Lead To An Unconstitutional Result.
At least since the landmark decision of Marburg v. Madison, 5 U.S. (1 Cranch) 137, 2 L.Ed. 60 (1803), federal courts have traditionally reviewed the constitutionality of legislation as applied to particular facts on a case-by-case basis. Under this mode of judicial review, a litigant is required to make a showing that a statute will work an unconstitutional result as applied to the facts and circumstances associated with that litigant’s conduct. Once a court determines that a statute will work an unconstitutional result, it will typically prevent the enforcement of that statute against the challenger.
In certain circumstances, courts are also permitted to declare a statute facially unconstitutional and invalidate the challenged legislation in all of its applications. The standard for a “facial challenge,” however, is typically much more onerous: “the challenger must establish that no set of circumstances exists under which the Act would be valid.” United States v. Salerno, 481 U.S. 739, 745, 107 S.Ct. 2095, 95 L.Ed.2d 697 (1987). Even under this “difficult” standard, id., a facial challenge remains a departure from a court’s traditional role as arbiter of a specific factual dispute. Because a facial challenge requires a litigant to show that a statute is incapable of a single valid application, it necessarily requires a court to speculate about potential applications of the challenged statute to hypothetical situations. This, in turn, requires a relaxing of many of the familiar requirements associated with Article III of the Constitution, which limits the federal judiciary’s authority to specific “Cases or Controversies.” As the Supreme Court has long recognized, a federal court “has no jurisdiction to pronounce any statute ... void, because irreconcilable with the constitution, except as it is called upon to adjudge the legal rights of litigants in actual controversies.” Liverpool, N.Y. & Philadelphia S.S. Co. v. Comm’rs of Emigration, 113 U.S. 33, 39, 5 S.Ct. 352, 28 L.Ed. 899 (1885).
Facial challenges are tolerated, in part, because they save judicial resources; they permit a single injured party to assert the claims of all future litigants by making a showing that each time that a statute is *294enforced, it will necessarily yield an unconstitutional result. And because a litigant is required to show that a statute has no valid applications before a court can declare that statute unconstitutional in its entirety, a facial challenge provides no more relief than would be obtained over an exhaustive series of as-applied challenges.
When courts depart from the Salerno standard, however, they find themselves even further removed from their core function — resolving specific factual disputes between parties. If, for example, a court were permitted to hold a statute facially invalid based upon a finding that a “large fraction” of its applications would yield an unconstitutional result, that court would face an extraordinarily difficult inquiry: first, it would have to “consider every conceivable situation which might possibly arise in the application of complex and comprehensive legislation,” Barrows v. Jackson, 346 U.S. 249, 256, 73 S.Ct. 1031, 97 L.Ed. 1586 (1953), and then it would have to determine whether the statute worked an unconstitutional result in a “large fraction” of these hypothetical situations. Moreover, if a court were permitted to declare a statute facially invalid based on a showing less than Salerno’s “no set of circumstances” standard, its relief might extend to situations where the statute’s application would not raise constitutional concerns.
As it stands now, however, the Supreme Court appears to have adopted the “large fraction” standard (perhaps modified by Stenberg to mean a “not-so-large fraction” standard) for those who seek to challenge an abortion regulation as facially unconstitutional. In Casey, the challengers to a spousal-notification provision did not have to show that there were “no set of circumstances” under which the provision would be valid; they simply had to show that, in the situations where the provision was “relevant,” it worked an unconstitutional result in a “large fraction of cases.” See Casey, 505 U.S. at 895, -112 S.Ct. 2791 (stating that “in a large fraction of cases in which [the challenged provision] is relevant, it will operate as a substantial obstacle to a woman’s choice to undergo an abortion”); cf. id. at 893, 894 (stating that the challenged provision was likely to prevent a “significant number of women” from seeking abortions). And “in Stenberg v. Carhart, without so much as a mention of Salerno, the Court held invalid, in a preenforcement challenge, an abortion statute that might have been construed ... to have at least some proper applications.” A Woman’s Choice — East Side Women’s Clinic v. Newman, 305 F.3d 684, 687 (7th Cir.2002) (citation omitted); see also Stenberg, 530 U.S. at 934, 120 S.Ct. 2597 (stating that, in a challenge to a ban on an “infrequently used abortion procedure,” the “procedure’s relative rarity ... is not highly relevant”). As a result of these decisions, at least “seven circuits have concluded that Salerno does not govern facial challenges to abortion regulations.” Richmond Med. Ctr. for Women v. Hicks, 409 F.3d 619, 627 (4th Cir.2005) (collecting cases); see also Sabri v. United States, 541 U.S. 600, 609-10, 124 S.Ct. 1941, 158 L.Ed.2d 891 (2004) (stating in dicta that the Supreme Court has recognized “the validity of facial attacks alleging over-breadth (though not necessarily using that term) in” the field of “abortion”) (citing Stenberg, 530 U.S. at 938-46, 120 S.Ct. 2597).
There may be adequate reasons for suspending Salerno’s “no set of circumstances” test in the field of abortion — • including possibly a concern that, under the Salémo standard, it would be difficult for a woman to bring an as-applied challenge under exigent circumstances. But the Supreme Court has never told us what has happened to the Salerno doctrine in *295the abortion context;10 it has never balanced the jurisprudential and administrative considerations associated with jettisoning Salerno against whatever medical concerns might militate in favor of a modified standard of proof. More importantly, the Court has never considered whether Casey or Stenberg struck the appropriate balance. Perhaps a better standard can be articulated — one that requires a regulation’s challenger to make an affirmative showing of proof regarding the way in which women will be adversely affected by the challenged abortion regulation. Instead, the Court has sanctioned a mode of constitutional analysis that permits the lower courts to invalidate an abortion regulation based upon a speculative showing that the challenged provision might work an unconstitutional result. See Stenberg, 530 U.S. at 937, 120 S.Ct. 2597 (stating that “[wjhere a significant body of medical opinion believes a procedure may bring with it greater safety for some patients and explains that view, we cannot say that the presence of a different view by itself proves the contrary” (emphasis added)); but cf. Broadrick v. Oklahoma, 413 U.S. 601, 615, 93 S.Ct. 2908, 37 L.Ed.2d 830 (1973) (explaining that, under the First Amendment doctrine of overbreadth, unless the unconstitutional reach of a statute is “not only real, but substantial,” the fact that a statute “may deter protected speech to some unknown extent” cannot “justify invalidating a statute on its face and so prohibiting a State from enforcing the statute against conduct that is admittedly within its power to proscribe” (emphasis added)).
The Supreme Court recently had an opportunity to address the continued vitality of Salerno in the abortion context. See Planned Parenthood v. Heed, 390 F.3d 53, 57-58 (1st Cir.2004), vacated, Ayotte v. Planned Parenthood, — U.S. -, 126 S.Ct. 961, — L.Ed.2d- (2006). The Court, however, chose not to address the most controversial issue before it — the quantum of proof necessary to bring a facial challenge to an abortion regulation. Instead, the Court simply held that, even where a challenger has brought forth evidence sufficient to satisfy the undefined standard of proof for facial challenges to abortion regulations, the lower courts should refrain from granting “the most blunt remedy” — permanently enjoining a regulation in its entirety' — -when a more limited remedy might suffice. Ayotte, 126 S.Ct. at 969. Still an open question is the appropriate standard of proof necessary to sustain a facial challenge to an abortion regulation.
Without the guidance of the Supreme Court on the question of proof, today we declare, under the binding precedent of Stenberg, that a statute is unconstitutional because there is evidence that, in some circumstances, its application might deny some women access to a marginally safer procedure. The Supreme Court needs to inform us how much evidence is required to sustain such challenges. Until it does, the lower courts will continue to labor under a standard that is both unclear and difficult to apply with any certainty, while the legislatures will lack sufficient eonsti-*296tutional guidance on the standard that will be used to challenge their enactments.
3. The Stenberg Court Established An Impossible Evidentiary Standard In An Area Where Legislatures Should Be Permitted To Set Policy.
Based upon my review of the record both before Congress and the district court, Congress had before it substantial evidence that, even if the D & X procedure is wholly prohibited, a woman can obtain a safe abortion in almost every conceivable situation. But that is not the standard under Stenberg. Under Stenberg, the government must establish that the D & X procedure is never necessary, in appropriate medical judgment, to preserve the health of the mother. Stenberg, 530 U.S. at 932, 936-38, 120 S.Ct. 2597. Not only must the government establish a negative, but it also must establish that proponents of the D & X procedure are not exercising responsible medical judgment when they conclude that D & X might be necessary, in some conceivable situation, based on the estimated health risks and benefits. See id. at 937, 120 S.Ct. 2597. This is a virtually insurmountable evidentiary hurdle.
The standard announced in Stenberg is rendered all the more questionable when one considers that the constitutional provision that the Court invoked to strike down Nebraska’s statute — the Due Process Clause of the Fourteenth Amendment— has generally been interpreted as a restraint on arbitrary government action. The Supreme Court should tell us what it is about abortion cases that necessitates an exception to this rule.
In my opinion, where the government still makes available to women an objectively safe and convenient means to terminate a pregnancy, it is inappropriate for a court to preclude the legislature from making a reasonable policy judgment about a particular procedure. As Justice Kennedy explained in his dissenting opinion in Sten-berg, the legislature has always been empowered to-make difficult policy choices in the field of medicine, even in the face of substantial disagreement over the virtues of a particular medical procedure. Id. at 970, 120 S.Ct. 2597 (Kennedy, J., dissenting). Indeed, “it is precisely where such disagreement exists that legislatures have been afforded the widest latitude.” Id. (quoting Kansas v. Hendricks, 521 U.S. 346, 360 n. 3, 117 S.Ct. 2072, 138 L.Ed.2d 501 (1997)).
sH «Is «¡» •}*
In the end, I cannot escape the conclusion that, in these abortion cases, the federal courts have been transformed into a sort of super regulatory agency — a role for which courts are institutionally ill-suited and one that is divorced from accepted norms of constitutional adjudication. In today’s case, we are compelled by a precedent to invalidate a statute that bans a morally repugnant practice, not because it poses a significant health risk, but because its application might deny some unproven number of women a marginal health benefit. Is it too much to hope for a better approach to the law of abortion — one that accommodates the reasonable policy judgments of Congress and the state legislatures without departing from established, generally applicable, tenets of constitutional law?
. As Justice O'Connor recognized in her dissenting opinion in Thornburgh v. American College of Obstetricians and Gynecologists, 476 U.S. 747, 106 S.Ct. 2169, 90 L.Ed.2d 779 (1986), overruled in part and superseded by Planned Parenthood v. Casey, 505 U.S. 833, 112 S.Ct. 2791, 120 L.Ed.2d 674 (1992), an abortion regulation does not necessarily impose an undue burden where it requires a "trade-off” between the relative health benefits associated with a desired abortion procedure and the important governmental interests associated with an alternative procedure. See Thornburgh, 476 U.S. at 832, 106 S.Ct. 2169 (O'Connor, J., dissenting) (stating that, absent the imposition of a health risk rising to the level of a "real and identifiable one,” where a statute promotes a "trade-off” between alternative abortions procedures, "there is little possibility that a woman's abortion decision will be unduly burdened by risks falling below that threshold”).
. If it were not for Stenberg, I believe this case would be governed by the thoughtful analysis that Judge Straub articulates in his dissenting opinion. See infra. In my opinion, Judge Straub carefully demonstrates why banning partial-birth abortion would not impose an undue burden on a woman's right to choose to terminate an unwanted pregnancy — at least, as that standard was defined in Casey. For at least three principal reasons, however, I believe his analysis is foreclosed by Stenberg.
First, I believe that Judge Straub overstates the Supreme Court’s reliance on the district court's findings in Stenberg. The Stenberg Court did not limit itself to the record on appeal; it also considered the conclusions of other federal district courts, Stenberg, 530 U.S. at 932-33, 120 S.Ct. 2597 (noting that, with one exception, the federal trial courts that heard evidence on partial-birth abortion concluded that D&X may be the best or most appropriate procedure in some circumstances); statements in medical textbooks, id. at 927, 120 S.Ct. 2597, medical journals, id. at 929, 120 S.Ct. 2597, and amicus curiae briefs, id. at 933, 120 S.Ct. 2597; and evidence before Congress, id. at 929-30, 120 S.Ct. 2597, 935; see also id. at 960, 966, 120 S.Ct. 2597 (Kennedy, J., dissenting); id. at 987 & n. 5, 993-96 & nn.ll & 13, 1016-17 & n. 23, 120 S.Ct. 2597 (Thomas, J., dissenting).
Second, the evidence has not changed since the Supreme Court decided Stenberg — only the conclusions that Congress decided to draw from that evidence. After Stenberg was decided, but before Congress held any new hearings on partial-birth abortion, Representative Chabot introduced House Bill 4965, the *293legislative precursor to the Partial-Birth Abortion Ban Act of 2003, complete with the same detailed factual findings that were ultimately enacted into law. Compare The Partial-Birth Abortion Ban Act of 2003, Pub.L. No. 108-105, § 2, 117 Stat. 1201, 1201-06 (2003), with Partial-Birth Abortion Ban Act of 2002, H.R. 4965, 107th Cong. § 2 (2002) (introduced June 19, 2002). When House Bill 4965 was introduced, with its findings already in place, Congress had not considered any new evidence. When it did hold hearings, the same divisions of medical opinion over whether D & X is safer than other procedures were presented, which is all that Stenberg required to ban the procedure.
Third, I do not believe that this case can be distinguished from Stenberg based upon the government’s interest in holding the line between infanticide and abortion. As the government was required to concede at oral argument, the Act applies both before and after a fetus becomes viable such that it would survive outside the womb. Absent a provision that limits the Act’s reach to viable fetuses, it is difficult to rest the Act's constitutionality upon an interest in maintaining a distinction between infanticide and abortion. In any event, the Stenberg Court was aware of the government's interest in maintaining such a distinction, see, e.g., Stenberg, 530 U.S. at 963, 120 S.Ct. 2597 (Kennedy, J., dissenting), but apparently the Court did not find this interest sufficient to justify a complete ban on partial-birth abortions.
. As Judge Easterbrook explained:
When the Justices themselves disregard rather than overrule a decision — as the majority did in Stenberg, and the plurality did in Casey — they put courts of appeals in a pickle. We cannot follow Salerno without departing from the approach taken in both Stenberg and Casey; yet we cannot disregard Salerno without departing from the principle that only an express overruling relieves an inferior court of the duty to follow decisions on the books.
Choice-East Side Women’s Clinic, 305 F.3d at 687.