LEAVITT, GOVERNOR OF UTAH, ET AL. v. JANE L. ET AL.
No. 95-1242
Supreme Court of the United States
June 17, 1996
518 U.S. 137
The State of Utah seeks review of a ruling by the Court of Appeals for the Tenth Circuit which declared invalid a provision of Utah law regulating abortions “[a]fter 20 weeks gestational age.”
Utah law, as amended by legislation enacted in 1991, establishes two regimes of regulation for abortion, based on the term of the pregnancy. With respect to pregnancies 20 weeks old or less,
Severability is of course a matter of state law. In Utah, as the Court of Appeals acknowledged, the matter “is determined first and foremost by answering the following question: Would the legislature have passed the statute without the unconstitutional section?” Id., at 1497 (citing Stewart v. Utah Public Service Comm‘n, 885 P.2d 759, 779 (Utah 1994)). A provision of the abortion part of the Utah Code, to which these two sections were added, answers that question. Section 317 provides:
“If any one or more provision, section, subsection, sentence, clause, phrase or word of this part or the application thereof to any person or circumstance is found to be unconstitutional, the same is hereby declared to be
severable and the balance of this part shall remain effective notwithstanding such unconstitutionality. The legislature hereby declares that it would have passed this part, and each provision, section, subsection, sentence, clause, phrase or word thereof, irrespective of the fact that any one or more provision, section, subsection, sentence, clause, phrase, or word be declared unconstitutional.” Utah Code Ann. § 76-7-317 (1995) (emphasis added).
In the face of this statement by the Utah Legislature of its own intent in enacting regulations of abortion, the Court of Appeals nonetheless concluded that
The dichotomy between “structural” and “substantive” intents is nowhere to be found in the Utah cases cited as authority by the Court of Appeals. Indeed, none of those cases even speaks in terms of “conflicts among legislative intentions,” id., at 1498. The cases do support the proposition that, “even where a savings clause exist[s], where the provisions of the statute are interrelated, it is not within the scope of th[e] court‘s function to select the valid portions of
But even if the Court of Appeals were correct in treating
This mode of analysis, if carried out in every case, would operate to defeat every claim of severability. Every legislature that adopts, in a single enactment, provision A plus provision B intends (A+B); and that enactment, which reads (A+B), is invariably a “unified expression of that intent,” so that taking away A from (A+B), leaving only B, will invariably “clearly undermine the legislative purpose” to enact (A+B). But the fallacy in applying this reasoning to the severability question is that it is not the severing that will take away A from (A+B) and thus foil the legislature‘s intent; it is the invalidation of A (in this case, because of its unconstitutionality) which does so—an invalidation that occurs whether or not the two provisions are severed. The relevant question, in other words, is not whether the legislature would prefer (A+B) to B, because by reason of the invalidation of A that choice is no longer available. The relevant question is whether the legislature would prefer not to have B if it could not have A as well. Here, the Court of Appeals in effect said yes. It determined that a legislature bent
*
*
*
We have summarily set aside unsupportable judgments in cases involving only individual claims, see, e. g., Board of Ed. of Rogers v. McCluskey, 458 U. S. 966, 969-971 (1982); National Bank of North America v. Associates of Obstetrics & Female Surgery, Inc., 425 U. S. 460, 460-461 (1976). Much more is that appropriate when what is at issue is the total invalidation of a statewide law, see, e. g., Idaho Dept. of Employment v. Smith, 434 U. S. 100, 100-102 (1977). To be sure, we do not normally grant petitions for certiorari solely to review what purports to be an application of state law; but we have done so, see Steele v. General Mills, Inc., 329 U. S. 433, 438, 440-441 (1947); Wichita Royalty Co. v. City Nat. Bank of Wichita Falls, 306 U. S. 103, 107 (1939),5 and undoubtedly should do so where the alternative is allowing
Finally, the dissent‘s appeal to the supposed greater expertise of courts of appeals regarding state law is particularly weak (if not indeed counterindicative) where a Court of Appeals panel consisting of judges from Oklahoma, Colorado, and Kansas has reversed the District Court of Utah on a point of Utah law. If, as we have said, the courts of appeals owe no deference to district court adjudications of state law, see Salve Regina College v. Russell, 499 U. S. 225, 239-240 (1991), surely there is no basis for regarding panels of circuit judges as “better qualified” than we to pass on such questions, see post, at 146. Our general presumption that courts of appeals correctly decide questions of state law reflects a judgment as to the utility of reviewing them in most cases, see Salve Regina College, supra, at 235, n. 3, not a belief that the courts of appeals have some natural advantage in this domain, cf. Brockett v. Spokane Arcades, Inc., 472 U. S. 491, 500 (1985) (“[W]e surely have the authority to differ with the lower federal courts as to the meaning of a state statute“); Cole v. Richardson, 405 U. S. 676, 683-685 (1972). That general presumption is obviously inapplicable where the court of appeals’ state-law ruling is plainly wrong, a conclusion that the dissent does not even contest in this case.
It is so ordered.
JUSTICE STEVENS, with whom JUSTICE SOUTER, JUSTICE GINSBURG, and JUSTICE BREYER join, dissenting.
The severability issue discussed in the Court‘s per curiam opinion is purely a question of Utah law. It is contrary to our settled practice to grant a petition for certiorari for the sole purpose of deciding a state-law question ruled upon by a federal court of appeals. The justifications for that practice are well established: The courts of appeals are more familiar with and thus better qualified than we to interpret the laws of the States within their Circuits; the decision of a federal court (even this Court) on a question of state law is not binding on state tribunals; and a decision of a state-law issue by a court of appeals, whether right or wrong, does not have the kind of national significance that is the typical predicate for the exercise of our certiorari jurisdiction.*
The underlying substantive issue in this case generates what Justice Holmes once described as a kind of “hydraulic pressure” that motivates ad hoc decisionmaking. Northern Securities Co. v. United States, 193 U. S. 197, 401 (1904) (dissenting opinion). Even if the court of appeals has rendered an incorrect decision, that is no reason for us to jettison the traditional guides to our practice of certiorari review. The doctrine of judicial restraint counsels the opposite course.
That leaves the single example of Steele v. General Mills, Inc., 329 U. S. 433 (1947), in which this Court granted certiorari because the lower court‘s judgment “undermine[d] the transportation policy of Texas.” Id., at 438. Decided nearly 50 years ago and without successor, Steele is the exception that proves the rule.
However irregular such grants were in the past, they are now virtually unheard of. Indeed, in 1980 we codified our already longstanding practice by eliminating as a consideration for deciding whether to review a case the fact that “a court of appeals has . . . decided an important state or territorial question in a way in conflict with applicable state or territorial law.” Compare this Court‘s Rule 19(1)(b) (1970) with this Court‘s Rule 17.1 (1980). That deletion—the only deletion of an entire category of cases—was intended to communicate our view that errors in the application of state law are not a sound reason for granting certiorari, except in the most extraordinary cases. Tellingly, the majority does not cite a single example during the past 16 years in which we
Accordingly, I respectfully dissent from the decision to grant the petition.
Notes
The two subsections state:
“(2) An abortion may be performed in this state only under the following circumstances:
“(a) in the professional judgment of the pregnant woman‘s attending physician, the abortion is necessary to save the pregnant woman‘s life;
“(b) the pregnancy is the result of rape or rape of a child . . . that was reported to a law enforcement agency prior to the abortion;
“(c) the pregnancy is the result of incest . . . and the incident was reported to a law enforcement agency prior to the abortion;
“(d) in the professional judgment of the pregnant woman‘s attending physician, to prevent grave damage to the pregnant woman‘s medical health; or
“(e) in the professional judgment of the pregnant woman‘s attending physician, to prevent the birth of a child that would be born with grave defects.
“(3) After 20 weeks gestational age, measured from the date of conception, an abortion may be performed only for those purposes and circumstances described in Subsections (2)(a), (d), and (e).”
In none of the Utah cases relied upon by the Court of Appeals was there a legislative statement of this sort. In both Salt Lake City v. International Assn. of Firefighters, 563 P.2d 786 (1977), and Carter v. Beaver County Service Area No. One, 399 P.2d 440 (1965), the saving clauses at issue simply declared: “If any provision of this act, or the application of any provision to any person or circumstance, is held invalid, the remainder of this act shall not be affected thereby.” See 1975 Utah Laws, ch. 102, § 10; 1961 Utah Laws, ch. 34, § 3. And in State v. Salt Lake City, 445 P.2d 691, 696 (1968), the court treated the saving clause in the municipal ordinance under review as no different from the one discussed in Carter, upon which the court relied.
Compare International Assn. of Firefighters, supra, at 791 (“The [invalidated] provisions . . . are an integral part of the act. . . . The concept of binding arbitration is wholly interdependent with the other provisions of the act“); Carter, supra, at 441-442 (“[T]he separability clause . . . is ineffective, because of the dependency of the remaining sections upon the provisions declared inoperative“) (emphases added).
The Court of Appeals also adverted to
The dissent says that our review in Wichita Royalty Co. “was plainly motivated by a concern to give effect to [the] new mandate” of Erie R. Co. v. Tompkins, 304 U. S. 64 (1938), that federal courts apply state substantive law in diversity cases. Post, at 147. It remains the case, however, that “the only question for our decision” was whether the Court of Appeals was correct in its interpretation of state law. 306 U. S., at 107 (emphasis added). As for Steele v. General Mills, Inc., 329 U. S. 433 (1947), there our review was prompted by concern that the judgment below “undermine[d] the transportation policy of Texas,” id., at 438. But unless we were wrong in Steele to regard this as “a question of such importance” as to justify review, ibid., the Tenth Circuit‘s “undermin[ing] [of] the [abortion] policy of [Utah]” presents an issue equally worth our attention. If the dissent is correct that Steele was our last case of this sort, it indicates only that we have not since been faced with a federal court‘s equivalently clear misinterpretation of a state law of equivalent significance.
The majority finds deference to the Court of Appeals “counterindicative” because it reversed the District Court for the District of Utah on a point of Utah law. Ante, at 145. But courts of appeals owe district courts no deference on state-law questions; they review such matters de novo. See Salve Regina College v. Russell, 499 U. S. 225, 235-240 (1991) (rejecting reliance on the “local expertise” of the District Court). The geography of the Circuit, see ante, at 145, is utterly irrelevant.
