View PDF Version

Nos. 05-380 and 05-1382

In the Supreme Court of the United States









Solicitor General
Counsel of Record
Department of Justice
Washington, D.C. 20530-0001
(202) 514-2217


Respondents' primary submission in these cases is that, in enacting the Partial-Birth Abortion Ban Act of 2003, Con gress sought to overturn this Court's decision in Stenberg v. Carhart, 530 U.S. 914 (2000). The Act itself belies that sub mission. To the contrary, it is clear that, in devising the Act, Congress sought to comply with Stenberg and to remedy both of the constitutional deficiencies that this Court identified in Stenberg. First, Congress made detailed factual findings con cerning the medical necessity of partial-birth abortion, and determined, based on those findings, that a statutory health exception was unnecessary. Second, Congress incorporated a narrower, more precise definition of partial-birth abortion, which was tailored to exclude the more common standard D&E abortion procedure. Congress's considered effort to meet the requirements of Stenberg, while exercising its well- established factfinding capabilities, in no way intrudes on the prerogatives of this Court, but instead represents a healthy exercise of the separation of powers.

For those reasons, and the reasons discussed in peti tioner's opening briefs, Stenberg does not require invalidation of the Act, which denies no woman the ability to obtain a safe abortion and instead takes only the limited step of proscribing a rarely used and inhumane abortion procedure resembling infanticide. A contrary conclusion would have enormous im plications for the ability of Congress to address an issue of grave medical and moral concern to most Americans. It would put courts in the difficult and institutionally problem atic role of second-guessing congressional judgments about medical facts, and mark a return to the days in which abortion regulations-even those not preventing women from obtain ing a safe abortion-were subject to the strictest of scrutiny. The invalidation of the Act would betray the lesson of the joint opinion in Planned Parenthood of Southeastern Penn sylvania v. Casey, 505 U.S. 833 (1992), which rejected strict scrutiny and sanctioned various abortion regulations short of an outright ban. Moreover, acceptance of respondents' broader argument that partial-birth abortion is beyond regu lation would betray Stenberg itself, which acknowledged that, as Justice O'Connor stressed in her concurring opinion, the inhumane and widely condemned practice of partial-birth abortion could constitutionally be prohibited in appropriate circumstances.


A. Stenberg Did Not Adopt A Rule Of Virtual Per Se Invali dation For Abortion Regulations Lacking A Health Ex ception

Respondents in these cases urge the Court to read into Stenberg a rule of virtual per se invalidation for abortion reg ulations lacking a health exception. That proposed rule is inconsistent both with this Court's abortion precedents and with generally applicable principles of law.

Respondents contend that, under this Court's decision in Stenberg, it is sufficient for a plaintiff challenging a statute that lacks a health exception merely to point to "substantial medical authority" that suggests that the statute would create significant health risks (even if it would not prevent any woman from obtaining an abortion). See, e.g., Carhart Br. 21- 23; Planned Parenthood Br. 11-14. Respondents do not deny that, under their view, a plaintiff need not show that the pre ponderance of the evidence supports the proposition that the statute at issue would create significant health risks; instead, they concede only that a plaintiff must come forward with something more than "a lone expert, making bald assertions with no support." Carhart Br. 21.

Stenberg certainly does not purport to impose such a hair trigger for invalidating abortion regulations, and such an in terpretation of Stenberg would effectively put that decision into conflict with the Court's earlier abortion decisions. Most notably, in Casey, the Court held only that it would be uncon stitutional for the government to prohibit or restrict abortion where the regulation at issue would "interfere with a woman's choice to undergo an abortion procedure if continuing her pregnancy would constitute a threat to her health." 505 U.S. at 880. Casey considered the health risks from regulations that might delay an abortion and prolong a pregnancy, not the marginal health risks of two alternative abortion procedures in circumstances in which women would continue to have ac cess to at least one safe procedure. But even in that context, Casey in no way indicated that the relevant constitutional inquiry was whether there was merely a division of medical opinion on the existence of a threat. Cf. ibid. (noting that it was "undisputed" that certain specified conditions could re quire an emergency abortion). Indeed, the joint opinion in Casey criticized the Court's earlier decision in City of Akron v. Akron Center for Reproductive Health, Inc., 462 U.S. 416 (1983) (Akron I), for placing too much weight on the views of an individual physician when they deviated from the best judgment of the legislature. 505 U.S. at 881-887; see Stenberg, 530 U.S. at 969-970 (Kennedy, J., dissenting).

More broadly, respondents' standard-entitling them to judgment based on the testimony of competing experts-is antithetical to normal modes of procedure and ignores both the important countervailing government interests and the dramatic consequences of facial invalidation of a duly enacted statute. Merely identifying a dispute over a material fact is generally enough to survive a motion for summary judgment, not enough to prevail in a facial challenge. And a rule that allows the views of a minority of medical experts to trump the considered views of Congress devalues the important govern ment interests recognized in Casey. As the dissenters in Stenberg noted, such a rule would be tantamount to a require ment that every abortion regulation contain an express health exception, because "there will always be some support for a [banned] procedure and there will always be some doctors who conclude that the procedure is preferable." 530 U.S. at 1012 (opinion of Thomas, J.); see id. at 969 (opinion of Ken nedy, J.) (noting that "[t]he standard of medical practice can not depend on the individual views of [the plaintiff] and his supporters"). In Stenberg, the Court neither explicitly nor implicitly adopted such a per se rule. Under the analysis ac tually applied in Stenberg, the Act is clearly constitutional.1

B. When Analyzed Under The Proper Standard, The Record Overwhelmingly Supports Congress's Judgment That No Health Exception Was Required

1. Congress's Findings On The Medical Necessity Of Partial-Birth Abortion Are Entitled To Deference

Respondents do not seriously dispute the general proposi tion that courts should afford a high degree of deference to congressional factual findings that inform the constitutional ity of federal statutes. See, e.g., Turner Broad. Sys., Inc. v. FCC, 520 U.S. 180, 195 (1997) (Turner II). Instead, respon dents offer a variety of arguments as to why Congress's find ings here on the medical necessity of partial-birth abortion are not entitled to deference. Those arguments lack merit.

a. Respondents repeatedly contend that Congress was seeking to "circumvent the constitutional rule established in Stenberg," Planned Parenthood Br. 24, or to "alter the mean ing and scope of substantive constitutional rights," Carhart Br. 24. Congress's findings in the Act, however, do not take issue with the constitutional rule identified by this Court in Stenberg. Rather, Congress took that rule as its starting point and merely made findings relevant to that rule. See, e.g., Act § 2(3), 117 Stat. 1201. Congress's findings-including its ultimate and most relevant finding that it is never medi cally necessary for a mother to undergo a partial-birth abor tion-certainly bear on the application of the constitutional test articulated by this Court in Stenberg. But Congress's unobjectionable effort to comply with this Court's precedents is a far cry from an effort to defy those precedents. These cases are thus readily distinguishable from cases in which Congress sought to alter the constitutional rule of decision laid down by this Court, see Dickerson v. United States, 530 U.S. 428 (2000); sought to replace a constitutional ruling of this Court with a different substantive rule of decision under Section 5 of the Fourteenth Amendment, see, e.g., City of Boerne v. Flores, 521 U.S. 507 (1997); or made findings that were insufficient as a matter of law to sustain Congress's ex ercise of a constitutional power, see United States v. Morri son, 529 U.S. 598 (2000).2

Respondents suggest that "[e]xtension of * * * deference to this case would effectively provide Congress with carte blanche to violate the Constitution simply by making carefully chosen 'findings.'" Carhart Br. 24. Respondents, however, simply overlook the fact that, while congressional factual find ings are entitled to substantial deference, see Turner II, 520 U.S. at 195, that deference does not amount to a blank check. Where Congress's findings are plainly not supported by sub stantial evidence, or where Congress's findings amount to legal conclusions, courts remain free to disregard those find ings. See ibid.; Lamprecht v. FCC, 958 F.2d 382, 392 n.2 (D.C. Cir. 1992) (Thomas, Circuit Justice) (rejecting proposition that "a legislature could make a statute constitutional simply by 'finding' that black is white or freedom, slavery").

b. In a similar vein, respondents suggest that Congress was foreclosed from "making contradictory findings" to those made in the Stenberg litigation. Carhart Br. 31. But it is un clear exactly which argument respondents embrace. Respon dents appear reluctant to defend the Eighth Circuit's conclu sion that, because the medical necessity of partial-birth abor tion was a "legislative" fact, Stenberg foreclosed Congress from making findings on that question (except on the basis of new evidence). See 05-380 Pet. App. 16a-20a. That reluctance is understandable. Deference to Congress's superior fact- finding function is appropriate precisely for such "legislative" facts, and, in any event, there is little evidence to support the premise that, in Stenberg, this Court was making "legislative" factual findings of its own. To the contrary, the Court repeat edly relied on the evidence presented to, and the factual find ings made by, the district court. See, e.g., Stenberg, 530 U.S. at 931-932, 934, 936-937.

It cannot be that the case-specific factual findings of a sin gle district judge, even when reviewed under an appropriately deferential standard by this Court, somehow foreclose Con gress from making its own, truly "legislative" factual findings, based both on its superior institutional capacity to make such findings and, in this instance, on a fuller and updated eviden tiary record. See Act § 2(5), 117 Stat. 1202.3 Such a rule would effectively convert the Judiciary's power "to say what the law is," Marbury v. Madison, 5 U.S. (1 Cranch) 137, 177 (1803), into a power to say what the law and facts are, even when those facts pertain to an appropriate subject of legisla tion. Nothing in our Constitution or this Court's precedents supports that view. Instead, the Legislative Branch is plainly authorized to engage in its own factfinding, and is the branch best situated to collect the full range of relevant data and evaluate it from a broad spectrum of perspectives. Congress does not lose that authority just because a court makes factual findings first in a particular case. Just as an agency does not lose its ability to resolve statutory ambiguities because a court construed the statute before an authoritative agency construction, see National Cable & Telecomms. Ass'n v. Brand X Internet Servs., 125 S. Ct. 2688, 2700-2702 (2005), Congress does not lose its ability to find facts and legislate because of prior litigation.

c. Respondents contend that, even if courts should gener ally afford a high degree of deference to congressional factual findings, they should not do so "where the case requires the application of heightened scrutiny to violations of fundamen tal substantive rights." Carhart Br. 29. In Turner II, how ever, this Court deferred to Congress's express finding that statutory provisions requiring cable-television systems to carry local stations were necessary to preserve those stations in upholding those provisions under the First Amendment, applying the heightened (intermediate) scrutiny afforded to content-neutral regulations. See 520 U.S. at 196. Similarly, in Board of Education v. Mergens, 496 U.S. 226 (1990), a plu rality of the Court (in reasoning with which none of the other Justices directly disagreed) deferred to Congress's finding that high school students were unlikely to confuse an equal- access policy with state sponsorship of religion in concluding that the Equal Access Act did not have the primary effect of advancing religion (and was thus valid under the Establish ment Clause). See id. at 251. Those cases confirm that the degree of deference owed to congressional findings does not turn on the right at issue or the applicable level of scrutiny.4

d. Finally, respondents contend that Congress's findings concerning the medical necessity of partial-birth abortion are not entitled to deference because Congress has no particular institutional expertise in that area. See, e.g., Carhart Br. 31- 32. Congress, however, has long exercised its authority under the Commerce Clause to regulate various aspects of medical practice-whether directly through its regulation of the sale, composition, and labeling of drugs (and the use of medical devices), see Federal Food, Drug & Cosmetic Act, 21 U.S.C. 301 et seq.; Controlled Substances Act, 21 U.S.C. 801 et seq., or indirectly through its provision of funding for various types of medical expenses, see Social Security Act, 42 U.S.C. 301 et seq. And this Court has repeatedly deferred to Congress's findings on legislation involving medical matters. See, e.g., Jones v. United States, 463 U.S. 354, 363-366 & n.13 (1983); Usery v. Turner Elkhorn Mining Co., 428 U.S. 1, 33-34 (1976); Lambert v. Yellowley, 272 U.S. 581, 588-597 (1926). In any event, the principle of deference to congressional factual findings does not turn on a court's assessment of Congress's true level of expertise regarding a particular subject, but is instead based more generally on Congress's superior institu tional ability to "amass and evaluate * * * data bearing upon legislative questions," Turner II, 520 U.S. at 195 (inter nal quotation marks omitted), as well as its constitutional responsibility to "weigh[] conflicting evidence in the legisla tive process," id. at 199. That capability surely exceeds the factfinding powers of a single district judge.

2. Congress's Findings Are Supported By Substantial Evidence

Respondents prefer to focus on subsidiary findings and mistaken arguments that Congress's findings here do not enjoy the normal deference, rather than making a serious effort to challenge Congress's ultimate and most relevant finding that "partial-birth abortion is never medically indi cated to preserve the health of the mother," Act § 2(14)(O), 117 Stat. 1206, under the deferential standard set out in Turner II. To the extent that respondents do take issue with Congress's ultimate finding, their effort to undermine it is unavailing.

a. Respondents primarily rely on the district courts' find ings in their respective cases that substantial medical author ity supported the proposition that partial-birth abortion was sometimes medically necessary, contending that those find ings are entitled to deference (and indeed are reviewable only for clear error). See, e.g., Carhart Br. 2-3, 5, 21-23; Planned Parenthood Br. 14-24. That contention is erroneous, however, both because it misapprehends the applicable legal standard, see pp. 2-4, supra, and because it turns well-established prin ciples concerning Congress's superior factfinding capacity on their head. The constitutionality of nationwide legislation properly depends on the credibility judgments of Congress, not those of individual district court judges, which of course can vary. And deference to congressional factfinding does not become inapplicable simply because a district court hears witnesses who did not testify before Congress.5

b. To the extent that respondents attempt to come to terms with Congress's factual findings, they primarily attack Congress's subsidiary findings. See, e.g., Planned Parent hood Br. 26-29. The relevant legal question before this Court, however, is not whether all of Congress's findings, including those that do not bear directly on the underlying legal ques tion, are supported by substantial evidence. Instead, it is whether substantial evidence supports Congress's ultimate, and most relevant, finding that "partial-birth abortion is never medically indicated to preserve the health of the mother." Act § 2(14)(O), 117 Stat. 1206. By focusing on Con gress's other findings, respondents commit the same error as the courts below, which (based on their misreading of Stenberg) asked only whether substantial evidence supported the proposition that there is no division of opinion as to whether partial-birth abortion is medically necessary in some circumstances. See, e.g., 05-380 Pet. App. 460a-461a; 05-1382 Pet. App. 15a.

c. Substantial evidence supports Congress's key findings, including its ultimate finding concerning the medical neces sity of partial-birth abortion. The starting point for the "sub stantial evidence" inquiry is the evidence that was actually before Congress when it made its findings. As petitioner's opening briefs in these cases explain, numerous physicians who appeared before Congress testified that there were no circumstances in which partial-birth abortion was the only appropriate type of abortion; that partial-birth abortion offers no safety advantages over other types of abortion; and that partial-birth abortion presents various safety risks that other abortion procedures do not. 05-380 Br. 31-33; 05-1382 Br. 22. Respondents' only answer to that testimony is the unela borated assertion that "[n]one of th[o]se physicians are ex perts in second-trimester surgical abortion, and several, if not all, oppose abortion by any method." Planned Parenthood Br. 29 n.24. That does not suffice.

While it is true (and unsurprising) that none of the physi cians who believed that the partial-birth abortion procedure was unnecessary and posed health risks had ever personally performed such an abortion, all of those physicians were expe rienced obstetricians, and some were maternal-fetal experts who specialized in treating women with high-risk pregnancies (and therefore could readily assess the risks that would at tend particular types of abortion procedures). And while at least some of those physicians may be opposed to abortion more generally, it does not follow that those physicians would be biased in their assessment of the comparative risks pre sented by different abortion methods. Because Congress could readily have concluded that those physicians were credi ble witnesses, there is no basis for discounting their testimony in assessing the validity of Congress's findings.

Congress's findings concerning the medical necessity of partial-birth abortion, moreover, were supported not only by the testimony of those physicians who testified before Con gress, but also by (1) other evidence in the legislative record, including statements from leading physician groups, articles in medical journals, and written statements from other physi cians, and (2) the testimony of still other physicians in the trials in these cases. See 05-380 Pet. Br. 33-37; 05-1382 Pet. Br. 22-24. Respondents merely challenge selected pieces of that evidence, not the substantiality of the record as a whole, and their specific challenges in any event lack merit.

With regard to the other evidence in the legislative record, respondents suggest that the American Medical Association (AMA) and the American College of Obstetricians and Gyne cologists (ACOG) actually held the view that partial-birth abortion was sometimes medically necessary. See, e.g., Carhart Br. 34-35. Notwithstanding its ultimate decision to oppose the Act, however, the AMA repeatedly expressed the view that partial-birth abortion is never the only appropriate procedure to protect a woman's health. See 05-380 Pet. C.A. App. 758, 775, 1002, 1004. And while a select panel convened by ACOG did suggest that partial-birth abortion could in some circumstances be the "best" choice, it ultimately con cluded that it "could identify no circumstances under which [partial-birth abortion] would be the only option to save the life or preserve the health of the woman." See 05-1382 J.A. 856-857, 860; 05-380 Pet. C.A. App. 1056, 1297. Consistent with those views, no one suggests that a doctor who refuses to perform a D&X abortion, but instead employs only the stan dard D&E procedure, is committing malpractice or operating outside the accepted standards of medical practice. There is therefore no bar to the legislature making the same judgment that any doctor can safely and responsibly make-to rule out the D&X procedure.

With regard to the trial testimony in these cases, respon dents note that the district courts credited the testimony of their experts over the testimony of the government's. See, e.g., Planned Parenthood Br. 7-8, 15-16. Even assuming the validity of those credibility determinations-and there is good reason to question them, see 05-1382 Pet. Br. 25-27-the cred ibility judgments of individual district court judges cannot trump the credibility judgments of Congress.6

d. Respondents heavily rely on the evidence they pre sented at trial, which in their view shows that partial-birth abortion is safer than other types of abortion either in specific circumstances or as a more categorical matter. The whole point of "substantial evidence" review, however, is that a re viewing court is required to defer to congressional findings even if the evidence is in conflict (and more than one conclu sion could thus be drawn from that evidence). See, e.g., Turner II, 520 U.S. at 208, 210. To the extent that the lower courts in these cases found that there was a division of opinion on the medical necessity of partial-birth abortion, see, e.g., 05- 380 Pet. App. 463a; 05-1382 Pet. App. 22a, 215a, the necessary implication is that there was, at least, substantial evidence to support Congress's finding that partial-birth abortion is never medically indicated.

In any event, the better view, based on the evidence in all three of the cases challenging the constitutionality of the Act, is that partial-birth abortion is not safer than other types of abortion either generally or in any specific circumstance. With regard to safety generally, one of the district courts to have considered the issue found that "the Government's ex pert witnesses reasonably and effectively refuted Plaintiffs' proffered bases for the opinion that D&X has safety advan tages over other second-trimester abortion procedures," and that "many of Plaintiffs' purported reasons for why D&X is medically necessary" are either only "theoretical" or simply "false." National Abortion Fed'n v. Ashcroft, 330 F. Supp. 2d 436, 479, 480 (S.D.N.Y. 2004), aff'd, 437 F.3d 278 (2d Cir. 2006) (NAF). While the plaintiffs in these cases have claimed that partial-birth abortions are generally safer than standard D&E abortions because partial-birth abortions present a lower risk of (1) cervical or uterine trauma, (2) retained fetal parts, (3) laceration from bony fragments, and (4) hemorrhage or infection, evidence presented at the trials suggested that each of those asserted safety benefits was hypothetical at best.7 In response, respondents presented only anecdotal evidence and intuition, and no scientific evidence, to support their claims. While respondents relied on a peer-reviewed study led by a plaintiff in another of the three cases, Dr. Ste phen Chasen, and based on abortions performed by Dr. Chasen and one of his partners, that study concludes that partial-birth abortion and standard D&E abortion had no significant differences in short-term complication rates, blood loss, or procedure time8-and actually documents (albeit be low a level deemed statistically robust) that the D&X proce dure resulted in a higher rate of subsequent premature birth. 05-380 J.A. 479-494, 615-617; 05-380 Pet. C.A. App. 2114.9

With regard to safety in specific circumstances, the district court in NAF found that "[i]n no case * * * could Plaintiffs point to a specific patient or actual circumstance in which D&X was necessary to protect a woman's health." 330 F. Supp. 2d at 480. Numerous experts, including several of re spondents' own experts, testified that there was no particular circumstance in which partial-birth abortion was medically necessary. See, e.g., 05-1382 J.A. 536, 938-939; 05-380 C.A. App. 1377. It should be emphasized, as well, that the rela tively long lead time necessary for sufficient dilation makes the D&X procedure particularly ill-suited for dealing with emergency health issues when time is of the essence. While respondents list various medical conditions for which they believe partial-birth abortion would be the safest abortion method, see, e.g., Carhart Br. 6-7; Planned Parenthood Br. 20- 22, the district court in Carhart was the only one that agreed with respondents, and it did so only with respect to two condi tions: placental cancer (in combination with preeclampsia) and placenta previa. See 05-380 Pet. App. 481a-482a. The underlying evidence on which that court relied does not sup port respondents' claims with regard to either of those condi tions. As to placental cancer, the court cited the testimony of only one physician, Dr. Joanna Cain-but she based her con clusion that partial-birth abortion was the safest abortion method for such cases solely on statements made in the delib erations of ACOG's task force, see 05-380 J.A. 500, which ulti mately concluded that it "could identify no circumstances under which [partial-birth abortion] would be the only option to save the life or preserve the health of the woman." 05-1382 J.A. 856. As to placenta previa, the court primarily relied on the testimony of Dr. Cassing Hammond-but he testified only that a D&E abortion would be safer than a hysterotomy in cases involving that condition, not that the D&X procedure would be safer than a standard D&E procedure. See 05-380 J.A. 749-750.

At a minimum, respondents' trial evidence concerning the medical necessity of partial-birth abortion was not so over whelming as to render Congress's factual findings unsup ported by substantial evidence. Those findings are therefore valid and entitled to deference.

C. Even Assuming That Partial-Birth Abortion Has Mar ginal Health Advantages In Some Cases, A Statute That Prohibits Partial-Birth Abortion Does Not Impose An Undue Burden On A Woman's Access To An Abortion

Even if the Court refused to defer to Congress's factual findings, it should conclude that the Act would not impose an undue burden on a woman's access to an abortion. Respon dents do not appear to contend that partial-birth abortion is necessary for the preservation of the mother's health in the sense that, when a mother requires an abortion for a non-life- threatening health condition, partial-birth abortion will be her only safe or practical option. That is unsurprising, because two of the three district courts to have considered the issue have squarely found that the plaintiffs in those cases failed to identify any such situation. See 05-1382 Pet. App. 147a; NAF, 330 F. Supp. 2d at 480. Moreover, respondents appear to con cede that standard D&E abortions are generally safe-and respondents' amici expressly make that concession. See, e.g., AMWA Br. 14 (stating that D&E by dismemberment is "ex tremely safe"); Chasen Br. 20 (stating that complications from D&E are "fortunately[] rare"). That is also unsurprising, because respondents' experts repeatedly testified that they considered standard D&E abortions to be "very" or "ex tremely" safe. See, e.g., 05-1382 J.A. 132, 237, 265, 409. And there is no suggestion that doctors who perform only D&E abortions are endangering their patients or deviating from the appropriate standard of care.

The relevant constitutional question in these cases is there fore whether it would impose an undue burden on any given woman's right to an abortion to prohibit an inhumane and rarely used type of partial-birth abortion procedure when that prohibition does not deny the woman a safe abortion, includ ing a safe late-term abortion, by a more common procedure. The answer to that question is clearly no. The Act implicates not only the government's compelling interest in protecting human life, but also the government's specific (and no less compelling) interest in prohibiting a particular type of abor tion procedure that closely resembles infanticide. See, e.g., Stenberg, 530 U.S. at 960, 962 (Kennedy, J., dissenting) (con cluding that "Nebraska's ban on partial birth abortion fur thers purposes States are entitled to pursue" and that "Ne braska was entitled to find the existence of a consequential moral difference between the [D&X and D&E] procedures").

Respondents and their amici contend that the latter inter est is not compelling, or even legitimate, because it consti tutes merely a "moral" interest. Planned Parenthood Br. 31- 32; AMWA Br. 27-29; NWLC Br. 2-3. That is both mistaken and beside the point. The line between being in utero and being outside the womb has long been the line that separates abortion from infanticide. A law that reinforces that line in the class of abortions that take place just before viability clearly promotes the government's interest in protecting life. But even if the government's interest in avoiding the slippery slope to infanticide is somehow categorically different from the interest in protecting life, but cf. Washington v. Glucks- berg, 521 U.S. 702, 728-735 (1997), the interest in promoting life could similarly be described as a moral interest. That interest, however, was described in the joint opinion in Casey as "substantial," "profound," and "important," see, e.g., 505 U.S. at 871, 876, 878, and is the interest that underpins all of the government's valid abortion regulations. If the govern ment was foreclosed from relying on "moral" interests in reg ulating abortion, it would be largely disabled not just from regulating late-term abortion procedures such as partial-birth abortion, but from regulating abortion altogether. That posi tion is sharply at odds with the joint opinion in Casey, which emphasized that the "the State's important and legitimate interest in potential life * * * has been given too little ac knowledgment." Id. at 871 (citation omitted).

D. To The Extent That The Court Believes That Stenberg Compels A Different Result, It Should Be Overruled

For the reasons explained above and in petitioner's open ing briefs, the Court's decision in Stenberg is not controlling. If, however, the Court were to conclude that its decision in Stenberg compels the conclusion that the Act is unconstitu tional, then Stenberg should be overruled.

As this Court has explained, "[s]tare decisis is the pre ferred course because it promotes the evenhanded, predict able, and consistent development of legal principles, fosters reliance on judicial decisions, and contributes to the actual and perceived integrity of the judicial process." Payne v. Tennessee, 501 U.S. 808, 827 (1991). Particularly in constitu tional cases, however, it is "common wisdom" that stare decisis is not an "inexorable command." Casey, 505 U.S. at 854. Contrary to respondents' arguments (Carhart Br. 35-37; Planned Parenthood Br. 29-31), several considerations weigh strongly against affording stare decisis effect to Stenberg.

As a preliminary matter, Stenberg is a recent and deeply divided decision that this Court has not followed in any other cases. Stenberg was grounded on a reading of another 5-4 decision, Casey, and was sharply criticized by one of the Jus tices in the Casey majority as an unjustified departure from that decision. See Stenberg, 530 U.S. at 956-957, 979 (Ken nedy, J., dissenting). Even in the short time since its issu ance, moreover, Stenberg has already proven to be unwork able in practice, as reflected in the different analytical ap proaches taken in the lower-court opinions that have at tempted to apply it. This is true not only of courts consider ing the constitutionality of the Act, but also of courts consid ering the validity of other abortion regulations, which have reached conflicting conclusions as to, inter alia, whether Stenberg requires every abortion regulation to contain an express health exception and, if not, what evidentiary showing must be made before such an exception is required. See, e.g., Richmond Med. Ctr. for Women v. Hicks, 409 F.3d 619, 622, 625-626 (4th Cir. 2005), petition for cert. pending, No. 05-730 (filed Dec. 1, 2005); Women's Med. Prof. Corp. v. Taft, 353 F.3d 436, 448 (6th Cir. 2003); A Woman's Choice-East Side Women's Clinic v. Newman, 305 F.3d 684, 688 (7th Cir. 2002), cert. denied, 537 U.S. 1192 (2003).

Moreover, if Stenberg really requires courts to engage in extensive evidentiary proceedings to second-guess congres sional judgments about medicine, then it has proven unwork able by thrusting courts into a role for which they are institu tionally ill-suited. It is one thing to require courts to ensure that Congress had substantial evidence, but it is quite another matter to force courts to engage in extensive factfinding to determine whether there is a sufficient consensus of medical opinion. This Court observed over a century ago that "[i]t is no part of the function of a court or a jury to determine," when medical opinion is divided, "which one of two modes was likely to be most effective for the protection of the public against disease." Jacobson v. Massachusetts, 197 U.S. 11, 30 (1905). Federal courts "are ill-equipped to evaluate the rela tive worth of particular surgical procedures," Stenberg, 530 U.S. at 968 (Kennedy, J., dissenting), and cannot sit as "the Nation's ex officio medical board[s]," ibid. (quoting Akron I, 462 U.S. at 456 (O'Connor, J., dissenting)).

Overruling Stenberg would not affect any relevant reliance interests, because it could hardly be said that "people have organized intimate relationships and made choices that define their views of themselves and their places in society[] in reli ance on the availability" of the gruesome and rarely used pro cedure of partial-birth abortion. Casey, 505 U.S. at 856. Moreover, there could be no reasonable reliance on Stenberg here, because the extension of Stenberg that respondents seek would put it in conflict with numerous other decisions of this Court: most notably, but by no means limited to, the Court's decision in Casey. See Stenberg, 530 U.S. at 957, 960-963, 979 (Kennedy, J., dissenting); id. at 1005-1020 (Thomas, J., dis senting). Such a reading would mark a return to the strict scrutiny expressly abandoned in Casey; would extend that strict scrutiny to legislative judgments concerning health, despite a century of precedents employing a much more def erential analysis, see, e.g., Kansas v. Hendricks, 521 U.S. 346, 360 n.3 (1997); Jacobson, 197 U.S. at 30; and would resurrect the Akron I requirement of near-absolute deference to the treating physician, which was overruled by the joint opinion in Casey, see 505 U.S. at 881-887. If Stenberg must be read to require a choice between Stenberg and all these other prece dents of the Court, the Court should jettison Stenberg and preserve the latter.


A. The Act Is Not Unconstitutionally Overbroad

1. Unlike the statute in Stenberg, the Act defines "partial- birth abortion" in a way that unambiguously does not reach standard D&E abortions. The Act applies only where a physi cian "deliberately and intentionally vaginally delivers a living fetus" past a specified anatomical "landmark" and then "per forms [an] overt act, other than completion of delivery, that kills the partially delivered living fetus"-and delivers the fetus with the purpose of performing that overt act. The Act thus excludes standard D&E abortions, in which the dismem berment of the fetus is not a discrete act from the delivery of a portion of the fetus and in which a major portion of the fetus is usually not delivered while the fetus is still living.

2. Respondents contend that, in a standard D&E abortion, a physician would like to deliver the fetus as intact as possi ble, and that, where the physician is able to remove most of the fetus but is unable to remove the fetus intact (e.g., because the head becomes stuck), the physician must sometimes per form a partial-birth abortion instead (e.g., by puncturing the fetus's skull and vacuuming out its brain). Carhart Br. 42-43; Planned Parenthood Br. 34-37. Indeed, the Planned Parent hood respondents go even further and make the novel sugges tion that partial-birth abortion does not even exist as a dis crete procedure, because physicians always set out to remove as much of the fetus as possible. Br. 41. Under that view, D&X is not a distinct procedure, only a particularly "success ful" variant of the D&E procedure. That suggestion, how ever, cannot be squared with the history of the partial-birth abortion procedure, which makes clear that partial-birth abor tion was developed as a distinct type of procedure in which the physician intentionally seeks to achieve a greater degree of dilation than in a standard D&E abortion, for the specific purpose of removing the fetus intact. See Stenberg, 530 U.S. at 927-928; id. at 959-960 (Kennedy, J., dissenting); id. at 985- 989 (Thomas, J., dissenting); cf. AMWA Br. 15 (noting that, in initially explaining the procedure of partial-birth abortion, the procedure's inventor "described both dilation techniques and intra-operative techniques used to maximize the possibility of intact removal"). That suggestion also defies the analysis of Stenberg, which faulted the Nebraska statute for failing to distinguish clearly between the D&X and D&E procedures (which would hardly be a failing if no meaningful difference existed). It likewise defies the innumerable references in the congressional and lower-court records to D&X and D&E as distinct procedures. Finally, that suggestion squarely con flicts with this Court's promise in Stenberg that a properly drawn statute could distinguish between D&X and D&E and validly regulate the former.10

3. Respondents contend that the Act is overbroad because it in fact reaches standard D&E abortions in which a physi cian does not set out to perform a partial-birth abortion, but ends up having to perform one anyway (and that the Act would therefore chill physicians from ever performing stan dard D&E abortions). Carhart Br. 44-45; Planned Parent hood Br. 42-44. In such cases, however, the physician would lack the required intent to be covered by the Act: i.e., the specific intent, at the outset of the procedure, to deliver the requisite portion of the fetus for the purpose of performing the ultimate lethal act. Contrary to respondents' contentions, that interpretation is compelled by the text of the Act itself, which applies only where the person performing the abortion "deliberately and intentionally" delivers the requisite portion of the fetus "for the purpose" of performing the ultimate le thal act. It is clear from that compound mens rea require ment that the physician must have the "purpose" of perform ing the ultimate lethal act at the outset of the procedure-not that it is sufficient for the physician to develop that "purpose" once the procedure is underway. See 05-380 Pet. Br. 47-48; 05-1382 Pet. Br. 32-33. Where a physician originally intends to perform a standard D&E abortion, therefore, the physician will not be subject to the Act.

4. The Carhart respondents repeatedly contend that the Act "ban[s] some non-intact D&Es" because it would reach abortions in which a physician delivers the requisite portion of the fetus and only then performs a discrete act of dismem berment (and delivers the fetus with the purpose of perform ing that act). Br. 38; see Br. 2, 17, 41-42. The Planned Par enthood respondents, however, concede that there is no evi dence that physicians ever perform such delivery-followed-by- dismemberment abortions. See Br. 41 n.34. And even if they did, there is no sense in which such abortions could be classi fied as standard D&E abortions; instead, such abortions are simply partial-birth abortions that are effectuated in a partic ularly gruesome manner. Because the Act does not reach any standard D&E abortions, and because the Act does not reach partial-birth abortions carried out by physicians who had in tended to perform standard D&E abortions instead, it is not unconstitutionally overbroad.

B. The Act Is Not Unconstitutionally Vague

1. Respondents renew their contention that the operative provisions of the Act more generally are unconstitutionally vague. Carhart Br. 45-47; Planned Parenthood Br. 44-47. As has already been shown, however, that contention fails be cause the Act unambiguously excludes standard D&E abor tions (and, at a minimum, could reasonably be so construed), and otherwise precisely defines the conduct that it prohibits. The Carhart respondents do not contend that any specific phrases within the Act are impermissibly vague. While the Planned Parenthood respondents challenge the Act's use of the phrase "overt act," Br. 45, their real complaint is with the breadth of that phrase, rather than with its clarity. The phrase "overt act" plainly refers to any act distinct from the act of delivery, other than the "completion of delivery"-a statutory exception that is designed to exclude induction abor tions. The phrase "overt act" is broad in reach precisely in order to ensure that a physician cannot evade the Act simply by performing an atypical lethal act: e.g., by dismembering the partially delivered fetus, rather than crushing its skull. There is nothing vague about the phrase "overt act," or, for that matter, any other term or phrase in the statute.11

2. With regard both to the Act's asserted overbreadth and to its asserted vagueness, respondents fail to suggest any way in which Congress could have drafted the Act to remedy the alleged constitutional deficiencies. Indeed, respondents virtu ally acknowledge that, in their view, Congress could not draft any statutory language-including a ban on the D&X proce dure, in terms-that would effectively regulate partial-birth abortion while navigating the Scylla and Charybdis of overbreadth and vagueness. See, e.g., Carhart Br. 47 n.29. If, for example, Congress sought to address the Act's asserted overbreadth by passing a less specific statute that simply prohibited "partial-birth" abortions (or "D&X" or "intact D&E" abortions), respondents would undoubtedly argue that the statute was unconstitutionally vague-and indeed, insofar as they believe that partial-birth abortions and standard D&E abortions are not meaningfully distinct, would likely argue that the statute is still overbroad. If, on the other hand, Con gress sought to address the Act's asserted vagueness by pass ing a statute that contained a different definition of "partial- birth abortion" (e.g., by prohibiting only the "crushing of the fetus's skull," rather than an "overt act[] other than comple tion of delivery"), the resulting statute could readily be cir cumvented. At bottom, respondents' approach is irreconcil able with the position of a majority of this Court in Stenberg: namely, that a valid statute could be drafted to prohibit partial-birth abortion. See, e.g., 530 U.S. at 951 (O'Connor, J., concurring).

C. The Court May Construe The Act To Avoid Any Conceiv able Overbreadth Or Vagueness Deficiency

When it comes to respondents' overbreadth and vagueness arguments, these cases differ from Stenberg in another criti cal respect. Because these cases involve the interpretation of a federal statute, rather than a state statute, the Court may and, if it deems it necessary to do so, should construe the Act to avoid any constitutional deficiencies it might identify. For example, respondents' own arguments as to how a delivery- and-dismemberment procedure would not constitute a D&X amply justify Congress's decision to avoid that medical term in its prohibition. Cf. Hendricks, 521 U.S. at 359 (noting that "[l]egal definitions * * * need not mirror those advanced by the medical profession"). But if the Constitution requires the statute to prohibit the D&X procedure as such, the Act can readily be so construed. More broadly, as long as the Court applies the normal canons of constitutional avoidance, rather than the "canon of construction under which in cases involving abortion, a permissible reading of a statute is to be avoided at all costs," Thornburgh v. American Coll. of Obstetricians & Gynecologists, 476 U.S. 747, 829 (1986) (O'Connor, J., dissent ing) (internal quotation marks and citation omitted), there is no obstacle to construing the statute to avoid constitutional difficulty.


Respondents also err in arguing that the Act must be inval idated in its entirety if the Court concludes it may be uncon stitutional in any application.

1. As a preliminary matter, the Planned Parenthood re spondents suggest that, in determining whether narrower injunctive relief is appropriate, a reviewing court should em ploy a presumption in favor of facial invalidation (and against narrower injunctive relief), on the ground that "Congress can always follow the Court's guidance and re-enact a law that conforms to the Constitution." Br. 48 n.38. But that sugges tion is flatly inconsistent with this Court's decision in Ayotte v. Planned Parenthood of Northern New England, 126 S. Ct. 961 (2006), which made clear that the "normal rule" is to en join a statute only as to its unconstitutional applications, rather than to invalidate the statute facially as to all of its applications. Id. at 968. Such a presumption in favor of nar rower injunctive relief makes eminent sense, because it en sures that a court will not "frustrate the expressed will of Congress or that of the state legislatures" by enjoining the enforcement of a law even in situations in which such enforce ment is, or would be, constitutional. Barrows v. Jackson, 346 U.S. 249, 256-257 (1953).

2. Respondents contend that, to the extent that the Act is invalid because it lacks a health exception, the presumption against facial invalidation has been overcome because it is clear that Congress purposefully adopted a statute without a health exception. Carhart Br. 47-49; Planned Parenthood Br. 47-48. That contention too, however, would invert the Ayotte presumption. Anytime a constitutional defect results from the omission of a necessary exception, that omission will al ways be, at least to some extent, intentional. If that were enough to invalidate a law, the Court in Ayotte would have affirmed the facial invalidation of the statute at issue. It did not, because, under Ayotte, the relevant question is whether Congress would have preferred no statute at all to a statute with a (judicially crafted) health exception. See 126 S. Ct. at 968. Respondents' only answer is the non sequitur that Con gress deliberately delayed passage of the Act for political reasons. See Planned Parenthood Br. 48. But even assuming that were true, it would say nothing about the substantive scope of the Act. The better view, based both on Congress's statutory findings and the statements of various members, is that the proponents of the Act would have preferred to ban partial-birth abortion in at least some circumstances, even if they could not have banned it altogether. See 05-1382 Pet. Br. 43. That would be particularly true to the extent that a court could craft a narrow health exception applicable only to par ticular verifiable medical conditions, which would pose few, if any, of the concerns evident in the debate over a blanket health exception.

3. Respondents and their amici contend that narrower injunctive relief would nevertheless be inappropriate because (1) it would be difficult to predict what sort of health excep tion Congress would have wanted and (2) it would require the Court to anticipate all of the specific circumstances in which the Act could not constitutionally be applied. Carhart Br. 49; NARAL Br. 10-12. As to the first concern, however, given the abundant evidence that Congress would have preferred to prohibit partial-birth abortion to the fullest extent possible, it is safe to assume that Congress would have opted for the narrowest exception that meets constitutional requirements. And as to the second concern, it is the plaintiffs' obligation in a facial challenge to identify the specific circumstances in which the statute being challenged cannot constitutionally be applied (and in which the statute's application should there fore be enjoined). Any difficulty in identifying those circum stances here is attributable to the deficiencies in respondents' evidence. And to the extent that this Court were to conclude that any such circumstances existed, it could readily enjoin the Act's application in those circumstances-without fore closing later plaintiffs from showing that the Act cannot con stitutionally be applied in other circumstances (and that the Act thus should be enjoined in those circumstances as well).

4. Finally, respondents argue that any overbreadth or vagueness in the Act cannot be cured by a narrower injunc tion. Carhart Br. 49-50; Planned Parenthood Br. 49-50. As respondents acknowledge, however, insofar as their argument is that it would be impossible to craft an injunction that would clarify that the Act reaches only standard D&E abortions, that is just another way of saying that Congress cannot devise any statute that would permissibly prohibit the inhumane, and, in Congress's informed judgment, medically unneces sary, practice of partial-birth abortion. See, e.g., id. at 50. As explained above, that position is at odds with this Court's prior decisions and should be rejected here. The promise of Stenberg was that the D&X procedure could permissibly be regulated. The Act as written is an example of such a permis sible regulation, and the Court should resist respondents' invitation to revoke that promise through a misguided appli cation of Ayotte.


For the foregoing reasons and those stated in petitioner's opening briefs, the judgments of the courts of appeals should be reversed.

Respectfully submitted.

Solicitor General


1 Respondents do not attempt to defend the Ninth Circuit's holding that, under Stenberg, the appropriate constitutional inquiry is whether "there is a medical consensus that no circumstance exists in which the procedure would be necessary to preserve a woman's health." 05-1382 Pet. App. 15a (emphasis added). Even respondents seem to recognize the implausibility of such a rule, which in the facial challenge context would turn the applicable standard entirely on its head. Moreover, while respondents describe application of the "large fraction" standard for facial challenges applied to the spousal-notifica tion provision at issue in Casey as "chillingly callous," Planned Parenthood Br. 13, nothing in Stenberg suggested that the Court was adopting a new, substantially more permissive rule for facial challenges, and there is nothing callous about requiring a plaintiff who seeks invalidation of a statute in all its applications to show that the statute is unconstitutional in all, or at least a large fraction, of those applications.

2 Respondents and their amici repeatedly note that some supporters of the Act made statements critical of the Court's decision in Stenberg. See, e.g., Carhart Br. 1; Fifty-Two Members of Congress Br. 2, 4. But the views of individual legislators concerning the relative merits of this Court's precedents hardly taint their votes or undermine legislative efforts to comply with decisions with which some members may disagree. Moreover, the relevant focus is on the findings of Congress, not the views of individual legislators.

3 Respondents repeatedly contend that Congress's factual findings should not receive deference because Congress made those findings based on "virtually the same factual record that informed this Court's decision in Stenberg." Carhart Br. 15; see id. at 19 & n.12; Planned Parenthood Br. 4-5 & n.6. Even if it were true that Congress did not consider any new evidence, Congress would not be foreclosed from revisiting the factual findings made by the district judge in Stenberg and making contrary findings of its own. Deference to congressional factual findings is based not just on Congress's superior capacity to root out raw data, but also on its ability to analyze that data more thoroughly and through the perspectives of elected officials representing diverse views and constituents across the Nation. See p. 10, infra. In any event, Congress did consider additional evidence in making its findings. Specifically, Congress held two post-Stenberg hearings in which it heard testimony from two physicians, Dr. Kathi Aultman and Dr. Mark Neerhof, who had not previously testified, see 05-1382 C.A. E.R. 578-586, 604, 862-866; heard testimony from a third physician, Dr. Curtis Cook, who had previously testified but on different topics, compare id. at 597-618 (2002 testimony) with id. at 544- 546, 550-551 (1997 testimony); and received new documentary evidence.

4 In contending that congressional findings are not entitled to deference where heightened scrutiny is involved, respondents primarily rely on this Court's decisions in four First Amendment cases: Landmark Communica tions, Inc. v. Virginia, 435 U.S. 829 (1978); Sable Communications of California, Inc. v. FCC, 492 U.S. 115 (1989); Ashcroft v. Free Speech Coalition, 535 U.S. 234 (2002); and Randall v. Sorrell, 126 S. Ct. 2479 (2006). In Landmark, however, this Court merely refused to defer to a state legislature's essentially legal conclusion that the divulgence of confidential information would pose a clear and present danger to the orderly administration of justice. See 435 U.S. at 843. In Sable, the Court rejected a request for deference on the ground that "the congressional record contain[ed] no legislative findings" pertinent to the constitutional question presented. See 492 U.S. at 129. Landmark and Sable, moreover, have since been cited only for the modest proposition that "[the fact] [t]hat Congress' predictive judgments are entitled to substantial deference does not mean * * * that they are insulated from meaningful judicial review altogether." Turner Broad. Sys., Inc. v. FCC, 512 U.S. 622, 666 (1994) (Turner I) (plurality opinion). In Free Speech Coalition, the Court did not explicitly refuse to defer to congressional findings at all (and, to the extent that it did so implicitly, did so only on the ground that any congressional findings were legally irrelevant). See, e.g., 535 U.S. at 253, 254. Finally, in Randall, the plurality opinion concluded only that, while legislatures had considerable leeway to make "empirical judgments" with regard to campaign-finance regulations, courts were required to exercise "independent judicial judgment" when "assessing [a] statute's 'tailoring,' that is, * * * assessing the proportionality of the [statute's] restrictions." 126 S. Ct. at 2492. In the end, the Court's decision, on occasion, to reject a statute even under a deferential review of legislative factfinding does not amount to a determination that deference is inappropriate in the first place, any more than a conclusion that an agency's interpretation of an ambiguous statute is unreasonable would amount to a determination that deference to the agency is inappropriate.

5 Where a reviewing court has before it evidence that was not before Congress, it can consider that evidence in engaging in "substantial evidence" review. See Turner II, 520 U.S. at 195. The relevant question, however, is not whether the reviewing court would reach the same determination as Congress based on the record as supplemented; instead, it is whether there is sufficient evidence to suggest that Congress's determination was reasonable in the first place. See, e.g., id. at 210-211; Rostker v. Goldberg, 453 U.S. 57, 82-83 (1981).

6 Respondents contend that the government's own experts recognized that "the hypothesized risks of [partial-birth abortion] are overstated or specula tive." Planned Parenthood Br. 19. Notwithstanding some isolated statements to the contrary, however, the trial and congressional records are replete with statements that partial-birth abortion carries greater risks of cervical incompe tence, see, e.g., 05-1382 J.A. 653-655, 776-779, 881; 05-1382 C.A. E.R. 145, 545, 579-580, 970, and hemorrhaging and infection, see, e.g., 05-1382 J.A. 660-662, 938, 1013-1016; 05-1382 C.A. E.R. 145-146, 863, 971, and poses additional risks from the internal rotation of the fetus to a feet-first position, see, e.g., 05-1382 J.A. 657-660, 937-938; 05-1382 C.A. E.R. 146, 545, 863.

7 See, e.g., 05-1382 J.A. 148, 322, 880, 957-958 (risk of cervical or uterine trauma in standard D&E abortions was minimized by the use of smooth, rounded forceps and standard ultrasound techniques); id. at 148, 319-320 (risk of retained fetal parts was minimized by the use of ultrasound and by inspection of the uterus and inventory of removed parts); id. at 374 (risk of bony fragments was minimized by proper use of forceps); 05-380 J.A. 480, 484 (risk of hemorrhage or infection was the same in partial-birth abortion and standard D&E abortion because both result in similar blood loss).

8 Although respondents contend that the Chasen study shows that partial- birth abortion is in fact safer because the median gestational age for women undergoing partial-birth abortions was higher than for women undergoing standard D&E abortions, see Carhart Br. 10; Planned Parenthood Br. 19, the Chasen study merely concluded that the complication rates from the two types of procedure appeared to be similar, without making any claim concerning the implications of any difference in gestational age. See 05-1382 J.A. 534. Moreover, other factors, such as the higher average age of the women undergoing the standard D&E procedure, would seem to point in the opposite direction.

9 In an amicus brief in these cases, Dr. Chasen (together with the other plaintiffs in NAF) cites a later article in which he and his co-authors attempt to explain the evidence concerning the higher rate of subsequent premature birth. Br. 13. That article, however, was not written until 2005 and is not contained in the trial record in any of the three cases.

10 It is undoubtedly true that partial-birth abortions are much rarer than standard D&E abortions. See, e.g., Stenberg, 530 U.S. at 924-927, 929. But that does not mean they cannot be regulated. See id. at 964 (Kennedy, J., dis senting) ("It ill-serves the Court, its institutional position, and the constitu tional sources it seeks to invoke to refuse to issue a forthright affirmation of Nebraska's right to declare that critical moral differences exist between the two procedures.").

11 Amicus California Medical Association contends that the phrase "in or affecting * * * commerce" is impermissibly vague. Br. 4-18. As a prelimi nary matter, the plaintiffs in all three cases conspicuously failed to argue that the Act constituted an impermissible exercise of Congress's Commerce Clause authority-nor would such a challenge be likely to succeed, as CMA appears to recognize. See, e.g., Br. 6-7 (discussing Freedom of Access to Clinic Entrances Act of 1994, 18 U.S.C. 248). CMA's derivative vagueness argument is meritless, because the phrase "in or affecting * * * commerce" unambigu ously modifies the phrase "performs a partial-birth abortion," and the Act thus regulates partial-birth abortions to the full extent of Congress's Commerce Clause authority. See, e.g., United States v. Carter, 981 F.2d 645, 646-647 (2d Cir. 1992) (rejecting similar vagueness challenge to 18 U.S.C. 922(g)).