December 20, 1999.
1. Statutes — Maxims and Rules of Construction — Generally
Principal objective of statutory construction is to discern legislative intent, and although court may explore a variety of sources to discern that intent, where a statute is unambiguous court will rely on plain and ordinary meaning of words chosen.
2. Marriage — Generally — Definition and Nature
Same-sex couples were not entitled to a marriage license under statutory scheme governing marriages in Vermont, since it was not clear that limiting marriage to opposite-sex couples violated legislature's "intent and spirit"; rather, evidence demonstrated a clear legislative assumption that marriage under statutory scheme consisted of a union between a man and a woman. 15 V.S.A. § 1 — 15 V.S.A. § 7; 18 V.S.A. § 5131 — 18 V.S.A. § 5151.
3. Constitutional Law — Vermont Constitution — Equal Protection
Common benefits clause of Vermont Constitution differs markedly from federal equal protection clause in its language, historical origins, purpose and development; while federal amendment may thus supplement protections afforded by common benefits clause, it does not supplant it as first and primary safeguard of rights and liberties of all Vermonters. U.S. Const. amend. XIV; Vt. Const. ch I, art. 7.
4. Constitutional Law — Vermont Constitution — Equal Protection
Common benefits clause of Vermont Constitution is intended to ensure that benefits and protections conferred by state are for common benefit of community and not for advantage of persons who are only a part of that community. Vt. Const. ch. I. art. 7.
5. Constitutional Law — Vermont Constitution — Equal Protection
In analyzing laws challenged under common benefits clause of Vermont Constitution, court must ultimately ascertain whether omission of a part of community from benefit, protection and security of challenged law bears a reasonable and just relation to the governmental purpose; factors to be considered may include: (1) significance of benefits and protections of challenged law; (2) whether omission of members of community from benefits and protections of challenged law promotes government's stated goals; and (3) whether classification is significantly underinclusive or overinclusive. Vt. Const. ch. I, art. 7.
6. Constitutional Law — Vermont Constitution — Equal Protection
Stated governmental purposes underlying Vermont marriage statutes — furthering the link between procreation and child rearing, and promoting commitment between married couples to promote security of their children and community as a whole — provided no reasonable basis for denying legal benefits and protections of marriage to same-sex couples. Vt. Const. ch. I, art. 7; 15 V.S.A. § 1 — 15 V.S.A. § 7; 18 V.S.A. § 5131 — 18 V.S.A. § 5151.
7. Constitutional Law — Vermont Constitution — Equal Protection
Under common benefits clause of Vermont Constitution, state of Vermont could not permissibly exclude same-sex couples from benefit, protection and security that its laws provided to opposite-sex married couples, since none of the asserted state interests provided a reasonable and just basis for such continued exclusion. Vt. Const. ch. I, art. 7; 15 V.S.A. § 1 — 15 V.S.A. § 7; 18 V.S.A. § 5131 — 18 V.S.A. § 5151.
8. Constitutional Law — Vermont Constitution — Equal Protection
Same-sex couples were entitled, under common benefits clause of Vermont Constitution, to obtain same benefits and protections afforded by Vermont law to married opposite-sex couples; court's judgment was suspended, however, and current statutory scheme governing marriages was to remain in effect for a reasonable period of time, to enable legislature to consider and enact implementing legislation consistent with court's constitutional mandate. Vt. Const. ch. I, art. 7; 15 V.S.A. § 1 — 15 V.S.A. § 7; 18 V.S.A. § 5131 — 18 V.S.A. § 5151.
Appeal from Chittenden Superior Court, Docket No. 1009-97CnC.
In the above-entitled cause, the Clerk will enter:
The judgment of the superior court upholding the constitutionality of the Vermont marriage statutes under Chapter I, Article 7 of the Vermont Constitution is reversed. The effect of the Court's decision is suspended, and jurisdiction is retained in this Court, to permit the Legislature to consider and enact legislation consistent with the constitutional mandate described herein.
Beth Robinson and Susan M. Murray of Langrock Sperry Wool, Middlebury, and Mary Bonauto, Gay Lesbian Advocates Defenders, Boston, Massachusetts, for Plaintiffs-Appellants.
William H. Sorrell, Attorney General, and Eve Jacobs-Carnahan and Timothy Tomasi, Assistant Attorneys General, Montpelier, for Defendant-Appellee State.
Timothy M. Eustace of Stitzel, Page Fletcher, P.C., Burlington, for Defendants-Appellees Town of Shelburne and City of South Burlington.
Gregg H. Wilson of Kolvoord, Overton Wilson, Essex Junction, for Defendant-Appellee Town of Milton.
Harvey Golubock, Montpelier, for Amicus Curiae Vermont Human Rights Commission.
Richard T. Cassidy of Hoff, Curtis, Pacht, Cassidy Frame, P.C., Burlington, and Evan Wolfson, Lambda Legal Defense and Education Fund, Inc., and Lawson M. Vicario and S. Elizabeth Foster of Gibson, Dunn Crutcher LLP, New York, New York, for Amici Curiae Vermont Coalition for Lesbian and Gay Rights, et al.
David Rath of Kohn Rath, Hinesburg, for Amicus Curiae Professors of Legislation and Statutory Interpretation.
Eileen M. Blackwood of Blackwood and Kraynak, P.C., Burlington, David Chambers, White River Junction, and Matthew Coles, American Civil Liberties Union Foundation, New York, New York, for Amici Curiae Parents and Friends of Lesbian and Gay Men, et al.
Peter M. Lawrence of Barr, Sternberg Moss, P.C., Bennington, for Amici Curiae Vermont Organization for Weddings of the Same-Gender, et al.
William M Dorsch of Mickenberg, Dunn, Sirotkin Dorsch, Burlington, for Amici Curiae Vermont NOW, et al.
Philip C. Woodward and Karen McAndrew of Dinse, Knapp McAndrew, P.C., Burlington, for Amici Curiae Vermont Psychiatric Association, et al.
Hal Goldman, Burlington, for Amicus Curiae Take It To the People.
J. Paul Giuliani of McKee, Giuliani Cleveland, Montpelier, and Dwight G. Duncan, North Dartmouth, Massachusetts, for Amici Curiae New Journey, et al.
Robert H. Erdmann, South Burlington, Jay Alan Sekulow and John P. Tuskey, Virginia Beach, Virginia, and Vincent P. McCarthy, New Milford, Connecticut, for Amicus Curiae The American Center for Law and Justice.
Clarke A. Gravel of Gravel Shea, Burlington, and Don Stenberg, Nebraska Attorney General, and L. Steven Grasz, Deputy Attorney General, Lincoln, Nebraska, for Amici Curiae State of Nebraska, et al.
Jon R. Eggleston, Burlington, for Amicus Curiae Professors of Law and Jurisprudence.
Duncan F. Kilmartin, Newport, and David R. Huggins, The National Legal Foundation, Memphis, Tennessee, for Amici Curiae Specialty Research Associates, et al. William M. O'Brien, O'Brien Law Offices, Winooski, Thomas E. McCormick of McCormick Fitzpatrick Kasper Burchard, Burlington, and Von G. Keetch and Alexander Dushku of Kirton McConkie, Salt Lake City, Utah, for Amici Curiae Roman Catholic Diocese of Burlington, Vermont, et al.
John Fitzpatrick, Burlington, and David Zwiebel, New York, New York, for Amicus Curiae Agudath Israel of America.
Duncan F. Kilmartin of Rexford Kilmartin, Newport, and Steven T. McFarland, Kimberlee W. Colby and Samuel B. Casey, Annandale, Virginia, for Amici Curiae Christian Legal Society, et al.
Timothy J. O'Connor, Jr., O'Connor Law Office, Brattleboro, and David Orgon Coolidge, The Catholic University of America, Washington, District of Columbia, for Amici Curiae Hon. Peter Brady, et al.
PRESENT: Amestoy, C.J., Dooley, Morse, Johnson and Skoglund, JJ.
May the State of Vermont exclude same-sex couples from the benefits and protections that its laws provide to opposite-sex married couples? That is the fundamental question we address in this appeal, a question that the Court well knows arouses deeply-felt religious, moral, and political beliefs. Our constitutional responsibility to consider the legal merits of issues properly before us provides no exception for the controversial case. The issue before the Court, moreover, does not turn on the religious or moral debate over intimate same-sex relationships, but rather on the statutory and constitutional basis for the exclusion of same-sex couples from the secular benefits and protections offered married couples.
We conclude that under the Common Benefits Clause of the Vermont Constitution, which, in pertinent part, reads,
That government is, or ought to be, instituted for the common benefit, protection, and security of the people, nation, or community, and not for the particular emolument or advantage of any single person, family, or set of persons, who are a part only of that community. . . .
Vt. Const., ch. I, art 7., plaintiffs may not be deprived of the statutory benefits and protections afforded persons of the opposite sex who choose to marry. We hold that the State is constitutionally required to extend to same-sex couples the common benefits and protections that flow from marriage under Vermont law. Whether this ultimately takes the form of inclusion within the marriage laws themselves or a parallel "domestic partnership" system or some equivalent statutory alternative, rests with the Legislature. Whatever system is chosen, however, must conform with the constitutional imperative to afford all Vermonters the common benefit, protection, and security of the law.
Plaintiffs are three same-sex couples who have lived together in committed relationships for periods ranging from four to twenty-five years. Two of the couples have raised children together. Each couple applied for a marriage license from their respective town clerk, and each was refused a license as ineligible under the applicable state marriage laws. Plaintiffs thereupon filed this lawsuit against defendants — the State of Vermont, the Towns of Milton and Shelburne, and the City of South Burlington — seeking a declaratory judgment that the refusal to issue them a license violated the marriage statutes and the Vermont Constitution.
The State, joined by Shelburne and South Burlington, moved to dismiss the action on the ground that plaintiffs had failed to state a claim for which relief could be granted. The Town of Milton answered the complaint and subsequently moved for judgment on the pleadings. Plaintiffs opposed the motions and cross-moved for judgment on the pleadings. The trial court granted the State's and the Town of Milton's motions, denied plaintiffs' motion, and dismissed the complaint. The court ruled that the marriage statutes could not be construed to permit the issuance of a license to same-sex couples. The court further ruled that the marriage statutes were constitutional because they rationally furthered the State's interest in promoting "the link between procreation and child rearing." This appeal followed.
In their motions, each of the parties presented the trial court with extensive extra-pleading facts and materials, including legislative history, scientific data, and sociological and psychological studies. See V.R.C.P. 12(b) (c) (motion treated as one for summary judgment where "matters outside the pleadings are presented to and not excluded by the court"); Fitzgerald v. Congleton, 155 Vt. 283, 293-94, 583 A.2d 595, 601 (1990) (court effectively converted motion to dismiss into motion for summary judgment where it considered matters outside pleadings and parties had reasonable opportunity to submit extra-pleading materials). The parties have continued to rely on these materials on appeal. In addition, the Court has received numerous amicus curiae briefs, representing a broad array of interests, supportive of each of the parties.
I. The Statutory Claim
Plaintiffs initially contend the trial court erred in concluding that the marriage statutes render them ineligible for a marriage license. It is axiomatic that the principal objective of statutory construction is to discern the legislative intent. See Merkel v. Nationwide Ins. Co., 166 Vt. 311, 314, 693 A.2d 706, 707 (1997). While we may explore a variety of sources to discern that intent, it is also a truism of statutory interpretation that where a statute is unambiguous we rely on the plain and ordinary meaning of the words chosen. See In re P.S., 167 Vt. 63, 70, 702 A.2d 98, 102 (1997). "[W]e rely on the plain meaning of the words because we presume they reflect the Legislature's intent." Braun v. Board of Dental Examiners, 167 Vt. 110, 116, 702 A.2d 124, 127 (1997).
Vermont's marriage statutes are set forth in Chapter 1 of Title 15, entitled "Marriage," which defines the requirements and eligibility for entering into a marriage, and Chapter 105 of Title 18, entitled "Marriage Records and Licenses," which prescribes the forms and procedures for obtaining a license and solemnizing a marriage. Although it is not necessarily the only possible definition, there is no doubt that the plain and ordinary meaning of "marriage" is the union of one man and one woman as husband and wife. See Webster's New International Dictionary 1506 (2d ed. 1955) (marriage consists of state of "being united to a person . . . of the opposite sex as husband or wife"); Black's Law Dictionary 986 (7th ed. 1999) (marriage is "[t]he legal union of a man and woman as husband and wife"). This understanding of the term is well rooted in Vermont common law. See Le Barron v. Le Barron, 35 Vt. 365, 366-71 (1862) (petition by wife to annul marriage for alleged physical impotence of husband); Clark v. Field, 13 Vt. 460, 465 (1841) (suit to declare marriage null and void on ground that husband and wife had not consummated marriage); Overseers of the Poor of the Town of Newbury v. Overseers of the Poor of the Town of Brunswick, 2 Vt. 151, 152 (1829) (dispute between towns over liability for support of family turned, in part, on validity of marriage where justice of peace had not declared parties husband and wife). The legislative understanding is also reflected in the enabling statute governing the issuance of marriage licenses, which provides, in part, that the license "shall be issued by the clerk of the town where either the bride or groom resides." 18 V.S.A. § 5131(a). "Bride" and "groom" are gender-specific terms. See Webster's, supra, at 334 (bride defined as "a woman newly married, or about to be married;" bridegroom defined as "a man newly married, or about to be married").
Further evidence of the legislative assumption that marriage consists of a union of opposite genders may be found in the consanguinity statutes, which expressly prohibit a man from marrying certain female relatives, see 15 V.S.A. § 1, and a woman from marrying certain male relatives, see id. § 2. In addition, the annulment statutes explicitly refer to "husband and wife," see id. § 513, as do other statutes relating to married couples. See, e.g., 12 V.S.A. § 1605 ("husband and wife" may not testify about communications to each other under rule commonly known as "marital privilege," see State v. Wright, 154 Vt. 512, 525, 581 A.2d 720, 728 (1990)); 14 V.S.A. § 461, 465, 470 (referring to interest of "widow" in estate of her "husband"); id. § 10 (requiring three witnesses where "husband or wife" are given beneficial interest in other's will); 15 V.S.A. § 102 (legal protections where "married man . . . deserts, neglects, or abandons his wife").
These statutes, read as a whole, reflect the common understanding that marriage under Vermont law consists of a union between a man and a woman. Plaintiffs essentially concede this fact. They argue, nevertheless, that the underlying purpose of marriage is to protect and encourage the union of committed couples and that, absent an explicit legislative prohibition, the statutes should be interpreted broadly to include committed same-sex couples. Plaintiffs rely principally on our decision in In re B.L.V.B., 160 Vt. 368, 369, 628 A.2d 1271, 1272 (1993). There, we held that a woman who was co-parenting the two children of her same-sex partner could adopt the children without terminating the natural mother's parental rights. Although the statute provided generally that an adoption deprived the natural parents of their legal rights, it contained an exception where the adoption was by the "spouse" of the natural parent. See id. at 370, 628 A.2d at 1273 (citing 12 V.S.A. § 448). Technically, therefore, the exception was inapplicable. We concluded, however, that the purpose of the law was not to restrict the exception to legally married couples, but to safeguard the child, and that to apply the literal language of the statute in these circumstances would defeat the statutory purpose and "reach an absurd result." Id. at 371, 628 A.2d at 1273. Although the Legislature had undoubtedly not even considered same-sex unions when the law was enacted in 1945, our interpretation was consistent with its "general intent and spirit." Id. at 373, 628 A.2d at 1274.
Contrary to plaintiffs' claim, B.L.V.B. does not control our conclusion here. We are not dealing in this case with a narrow statutory exception requiring a broader reading than its literal words would permit in order to avoid a result plainly at odds with the legislative purpose. Unlike B.L.V.B., it is far from clear that limiting marriage to opposite-sex couples violates the Legislature's "intent and spirit." Rather, the evidence demonstrates a clear legislative assumption that marriage under our statutory scheme consists of a union between a man and a woman. Accordingly, we reject plaintiffs' claim that they were entitled to a license under the statutory scheme governing marriage.
II. The Constitutional Claim
Assuming that the marriage statutes preclude their eligibility for a marriage license, plaintiffs contend that the exclusion violates their right to the common benefit and protection of the law guaranteed by Chapter I, Article 7 of the Vermont Constitution. They note that in denying them access to a civil marriage license, the law effectively excludes them from a broad array of legal benefits and protections incident to the marital relation, including access to a spouse's medical, life, and disability insurance, hospital visitation and other medical decision making privileges, spousal support, intestate succession, homestead protections, and many other statutory protections. They claim the trial court erred in upholding the law on the basis that it reasonably served the State's interest in promoting the "link between procreation and child rearing." They argue that the large number of married couples without children, and the increasing incidence of same-sex couples with children, undermines the State's rationale. They note that Vermont law affirmatively guarantees the right to adopt and raise children regardless of the sex of the parents, see 15A V.S.A. § 1-102, and challenge the logic of a legislative scheme that recognizes the rights of same-sex partners as parents, yet denies them — and their children — the same security as spouses.
Although plaintiffs raise a number of additional arguments based on both the United States and the Vermont Constitutions, our resolution of the Common Benefits claim obviates the necessity to address them.
In considering this issue, it is important to emphasize at the outset that it is the Common Benefits Clause of the Vermont Constitution we are construing, rather than its counterpart, the Equal Protection Clause of the Fourteenth Amendment to the United States Constitution. It is altogether fitting and proper that we do so. Vermont's constitutional commitment to equal rights was the product of the successful effort to create an independent republic and a fundamental charter of government, the Constitution of 1777, both of which preceded the adoption of the Fourteenth Amendment by nearly a century. As we explained in State v. Badger, 141 Vt. 430, 448-49, 450 A.2d 336, 347 (1982), "our constitution is not a mere reflection of the federal charter. Historically and textually, it differs from the United States Constitution. It predates the federal counterpart, as it extends back to Vermont's days as an independent republic. It is an independent authority, and Vermont's fundamental law."
As we explain in the discussion that follows, the Common Benefits Clause of the Vermont Constitution differs markedly from the federal Equal Protection Clause in its language, historical origins, purpose, and development. While the federal amendment may thus supplement the protections afforded by the Common Benefits Clause, it does not supplant it as the first and primary safeguard of the rights and liberties of all Vermonters. See id. (Court is free to "provide more generous protection to rights under the Vermont Constitution than afforded by the federal charter"); State v. Jewett, 146 Vt. 221, 224, 500 A.2d 233, 235 (1985) (state constitution may protect Vermonters "however the philosophy of the United States Supreme Court may ebb and flow"); see generally H. Linde, First Things First, Rediscovering the States' Bill of Rights, 9 U. Balt. L. Rev. 379, 381-82 (1980); S. Pollock, State Constitutions as Separate Sources of Fundamental Rights, 35 Rutgers L. Rev. 707, 717-19 (1983).
A. Historical Development
In understanding the import of the Common Benefits Clause, this Court has often referred to principles developed by the federal courts in applying the Equal Protection Clause. See, e.g., Choquette v. Perrault, 153 Vt. 45, 51-52, 569 A.2d 455, 458-59 (1989). At the same time, however, we have recognized that "[a]lthough the provisions have some similarity of purpose, they are not identical." Benning v. State, 161 Vt. 472, 485 n. 7, 641 A.2d 757, 764 n. 7 (1994). Indeed, recent Vermont decisions reflect a very different approach from current federal jurisprudence. That approach may be described as broadly deferential to the legislative prerogative to define and advance governmental ends, while vigorously ensuring that the means chosen bear a just and reasonable relation to the governmental objective.
Conventional equal protection analysis under the Fourteenth Amendment employs three "tiers" of judicial review based upon the nature of the right or the class affected. See generally, Cleburne v. Cleburne Living Center, Inc., 473 U.S. 432, 440-41 (1985); 3 R. Rotunda J. Nowak, Treatise on Constitutional Law § 18.3, at 216-10 (3d ed. 1999). The first step in that analysis is to categorize the class affected as more or less similar to race based upon certain judicially-developed criteria. See Personnel Administrator of Mass. v. Feeney, 442 U.S. 256, 272 (1979); see generally, J. Baer, Equality Under the Constitution: Reclaiming the Fourteenth Amendment 253-64 (1983); C. Sunstein, The Anticaste Principle, 92 Mich. L. Rev. 2410, 2441-44 (1994). If a legislative classification implicates a "suspect" class, generally defined in terms of historical discrimination, political powerlessness, or immutable characteristics, the law is subject to strict scrutiny, and the state must demonstrate that it furthers a compelling governmental interest that could not be accomplished by less restrictive means. In addition to race (the original suspect class), alienage and national origin have also been recognized as suspect. See Cleburne, 473 U.S. at 440. The United States Supreme Court has created a "middle-tier" level of review for legislative classifications based on gender or illegitimacy; laws affecting these groups must be substantially related to a sufficiently important governmental interest to withstand constitutional scrutiny. See id. The balance of legislative enactments, including nearly all economic and commercial legislation, are presumptively constitutional and will be upheld if rationally related to any conceivable, legitimate governmental interest. See Minnesota v. Clover Leaf Creamery Co., 449 U.S. 456, 466 (1981); see also Cleburne, 473 U.S. at 440. Thus, as one commentator has explained, rationality review may be "used to uphold laws justified even by hypothesized or ad hoc state interests." J. Wexler, Defending the Middle Way: Intermediate Scrutiny as Judicial Minimalism, 66 Geo. Wash. L. Rev. 298, 300 (1998).
Although our decisions over the last few decades have routinely invoked the rhetoric of suspect class favored by the federal courts, see, e.g., Choquette, 153 Vt. at 51, 569 A.2d at 458, there are notable exceptions. The principal decision in this regard is the landmark case of State v. Ludlow Supermarkets, Inc., 141 Vt. 261, 448 A.2d 791 (1982). There, Chief Justice Albert Barney, writing for the Court, invalidated a Sunday closing law that discriminated among classes of commercial establishments on the basis of their size. After noting that this Court, unlike its federal counterpart, was not constrained by considerations of federalism and the impact of its decision on fifty varying jurisdictions, the Court declared that Article 7 "only allows the statutory classifications . . . if a case of necessity can be established overriding the prohibition of Article 7 by reference to the "`common benefit, protection, and security of the people.'" Id. at 268, 448 A.2d at 795. Applying this test, the Court concluded that the State's justifications for the disparate treatment of large and small businesses failed to withstand constitutional scrutiny. Id. at 269-70, 448 A.2d at 796.
Ludlow, as we later explained, did not alter the traditional requirement under Article 7 that legislative classifications must "reasonably relate to a legitimate public purpose." Choquette, 153 Vt. at 52, 569 A.2d at 459. Nor did it overturn the principle that the justifications demanded of the State may depend upon the nature and importance of the benefits and protections affected by the legislation; indeed, this is implicit in the weighing process. It did establish that Article 7 would require a "more stringent" reasonableness inquiry than was generally associated with rational basis review under the federal constitution. State v. Brunelle, 148 Vt. 347, 351, 534 A.2d 198, 201-202 (1987); see also Hodgeman v. Jard Co., 157 Vt. 461, 464, 599 A.2d 1371, 1373 (1991) (citing Ludlow for principle that Article 7 "may require this Court to examine more closely distinctions drawn by state government than would the Fourteenth Amendment"). Ludlow did not override the traditional deference accorded legislation having any reasonable relation to a legitimate public purpose. It simply signaled that Vermont courts — having "access to specific legislative history and all other proper resources" to evaluate the object and effect of State laws — would engage in a meaningful, case-specific analysis to ensure that any exclusion from the general benefit and protection of the law would bear a just and reasonable relation to the legislative goals. Ludlow, 141 Vt. at 268, 448 A.2d at 795.
In this respect, Ludlow was consistent with an older line of Vermont decisions which, albeit in the Fourteenth Amendment context, routinely subjected laws involving economic classifications to a relatively straightforward reasonableness evaluation, explicitly balancing the rights of the affected class against the State's proffered rationale. See, e.g., State v. Hoyt, 71 Vt. 59, 64, 42 A. 973, 975 (1899) (peddler-licensing classifications must be "based on some reasonable ground, some difference that bears a just and proper relation to the attempted classification, and is not a mere arbitrary selection"); State v. Cadigan, 73 Vt. 245, 252, 50 A. 1079, 1081 (1901) (State must establish "reasonable basis" to support law distinguishing between business partnerships organized in Vermont and those formed in other states); State v. Haskell, 84 Vt. 429, 437, 75 A. 852, 856 (1911) (mill regulation must be "based upon some difference having a reasonable and just relation to the object sought"). These opinions are notable for their detailed examination of the context and purposes of the challenged legislation, the impact on the affected class, and the logical fit between the statutory classification and the public ends to be achieved.
Although it is accurate to point out that since Ludlow our decisions have consistently recited the federal rational-basis/strict-scrutiny tests, it is equally fair to observe that we have been less than consistent in their application. Just as commentators have noted the United States Supreme Court's obvious yet unstated deviations from the rational-basis standard, so have this Court's holdings often departed from the federal test. In Colchester Fire Dist. No. 2 v. Sharrow, 145 Vt. 195, 198-99, 485 A.2d 134, 136-37 (1984), for example, the Court ostensibly applied a rational-basis test to invalidate a payment scheme for revenue-bond assessments. While acknowledging the broad discretion traditionally accorded the Legislature in taxation and other areas of public welfare, the Court nevertheless examined each of the district's rationales in detail and found them to be unpersuasive in light of the record and administrative experience. See id. at 200-201, 485 A.2d at 137 (record established no "plausible relationship between the method of bond assessment and its alleged purposes").
Cass Sunstein, among others, has documented the United States Supreme Court's unacknowledged departures from the deferential rational-basis standard without defining a new kind of scrutiny. See C. Sunstein, Foreword: Leaving Things Undecided, 110 Harv. L. Rev. 6, 59-61 (1996). These cases include Romer v. Evans, 517 U.S. 620, 635 (1996) (holding Colorado statute that banned state or local laws forbidding sexual-orientation discrimination was not rationally related to legitimate governmental objective), City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 450 (1985) (applying rational basis review, Court invalidated zoning discrimination against mentally retarded as based on "irrational prejudice"), and United States Dept. of Agriculture v. Moreno, 413 U.S. 528, 534 (1973) (invalidating regulation that excluded non-family members of household from food stamp program). In each of these decisions, the Court employed a highly contextual, fact-based analysis balancing private rights and public interests even while ostensibly applying minimal rational basis review. Conversely, in Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 237 (1995), the high court itself questioned the notion that strict scrutiny was inevitably "fatal in fact." See G. Gunther, The Supreme Court, 1971 Term — Foreword: In Search of Evolving Doctrine on a Changing Court: A Model for a New Equal Protection, 86 Harv. L. Rev. 1, 8 (1972) (observing that strict scrutiny is generally "`strict' in theory and fatal in fact"). Viewed together, these cases have prompted one commentator to suggest that "[t]he hard edges of the tripartite division have thus softened," and that the Court has moved "toward general balancing of relevant interests." Sunstein, supra, at 77.
In Choquette, 153 Vt. at 51, 569 A.2d at 458, the Court again purported to apply rational-basis review under Article 7 in holding a fence-repair statute to be unconstitutional. Not content to accept arguments derived from a bygone agricultural era, the Court held that the policies underlying the law were outdated and failed to establish a reasonable relation to the public purpose in the light of contemporary circumstances. See id. at 53-54, 569 A.2d at 459-60; see also Oxx v. Department of Taxes, 159 Vt. 371, 376, 618 A.2d 1321, 1324 (1992) (income tax assessment violated Equal Protection and Common Benefits Clauses as applied); Lorrain v. Ryan, 160 Vt. 202, 215 628 A.2d 543, 551 (1993) (statutory scheme denying right of spouse of injured worker to sue third-party tortfeasor for loss of consortium violated Equal Protection and Common Benefits Clauses).
The "more stringent" test was also implicit in our recent decision in MacCallum v. Seymour's Administrator, 165 Vt. 452, 686 A.2d 935 (1996), which involved an Article 7 challenge to an intestacy statute that denied an adopted person's right of inheritance from collateral kin. While employing the rhetoric of minimal scrutiny, our analysis was more rigorous than traditional federal rational-basis review. Indeed, although the State proffered at least a conceivable purpose for the legislative distinction between natural and adopted children, we held that the classification was unreasonable, explaining that "[a]dopted persons have historically been a target of discrimination," id. at 459, 686 A.2d at 939, and that however reasonable the classification when originally enacted, it represented an "outdated" distinction today. Id. at 460, 686 A.2d at 939. Thus, while deferential to the historical purpose underlying the classification, we demanded that it bear a reasonable and just relation to the governmental objective in light of contemporary conditions.
This approach may also be discerned in the Court's recent opinion in Brigham v. State, 166 Vt. 246, 692 A.2d 384 (1997), addressing an Article 7 challenge to the State's educational funding system. Consistent with prior decisions, the Court acknowledged the federal standard, see id. at 265, 692 A.2d at 395, even as it eschewed the federal categories of analysis. Indeed, after weighing the State's justifications for the disparate funding of education against its impact upon public-school students, the Court concluded; "Labels aside, we are simply unable to fathom a legitimate governmental purpose to justify the gross inequities in educational opportunities evident from the record." Id. at 265, 692 A.2d at 396.
Thus, "labels aside," Vermont case law has consistently demanded in practice that statutory exclusions from publicly-conferred benefits and protections must be "premised on an appropriate and overriding public interest." Ludlow, 141 Vt. at 268, 448 A.2d at 795. The rigid categories utilized by the federal courts under the Fourteenth Amendment find no support in our early case law and, while routinely cited, are often effectively ignored in our more recent decisions. As discussed more fully below, these decisions are consistent with the text and history of the Common Benefits Clause which, similarly, yield no rigid categories or formulas of analysis. The balancing approach utilized in Ludlow and implicit in our recent decisions reflects the language, history, and values at the core of the Common Benefits Clause. We turn, accordingly, to a brief examination of constitutional language and history.
We typically look to a variety of sources in construing our Constitution, including the language of the provision in question, historical context, case-law development, the construction of similar provisions in other state constitutions, and sociological materials. See Benning, 161 Vt. at 476, 641 A.2d 759. The Vermont Constitution was adopted with little recorded debate and has undergone remarkably little revision in its 200-year history. Recapturing the meaning of a particular word or phrase as understood by a generation more than two centuries removed from our own requires, in some respects, an immersion in the culture and materials of the past more suited to the work of professional historians than courts and lawyers. See generally, H. Powell, Rules for Originalists, 73 Va. L. Rev. 659, 659-61 (1987); P. Brest, The Misconceived Quest for the Original Understanding, 60 B.U.L. Rev. 204, 204-209 (1980). The responsibility of the Court, however, is distinct from that of the historian, whose interpretation of past thought and actions necessarily informs our analysis of current issues but cannot alone resolve them. See Powell, supra, at 662-68; Brest, supra, at 237. As we observed in State v. Kirchoff, 156 Vt. 1, 6, 587 A.2d 988, 992 (1991), "our duty is to discover the core value that gave life to Article ." (Emphasis added). Out of the shifting and complicated kaleidoscope of events, social forces, and ideas that culminated in the Vermont Constitution of 1777, our task is to distill the essence, the motivating ideal of the framers. The challenge is to remain faithful to that historical ideal, while addressing contemporary issues that the framers undoubtedly could never have imagined.
We first focus on the words of the Constitution themselves, for, as Chief Justice Marshall observed, "although the spirit of an instrument, especially of a constitution, is to be respected not less than its letter, yet the spirit is to be collected chiefly from its words." Sturges v. Crowningshield, 17 U.S. (4 Wheat.) 122, 202 (1819). One of the fundamental rights included in Chapter I of the Vermont Constitution of 1777, entitled "A Declaration of Rights of the Inhabitants of the State of Vermont," the Common Benefits Clause as originally written provided:
That government is, or ought to be, instituted for the common benefit, protection, and security of the people, nation or community; and for the particular emolument or advantage of any single man, family or set of men, who are a part only of that community; and that the community hath an indubitable, unalienable and indefeasible right, to reform, alter or abolish government, in such manner as shall be, by that community, judged most conducive to the public weal.
Vt. Const. of 1777, ch. I, art. VI. The first point to be observed about the text is the affirmative and unequivocal mandate of the first section, providing that government is established for the common benefit of the people and community as a whole. Unlike the Fourteenth Amendment, whose origin and language reflect the solicitude of a dominant white society for an historically-oppressed African-American minority (no state shall "deny" the equal protection of the laws), the Common Benefits Clause mirrors the confidence of a homogeneous, eighteenth-century group of men aggressively laying claim to the same rights as their peers in Great Britain or, for that matter, New York, New Hampshire, or the Upper Connecticut River Valley. See F. Mahady, Toward a Theory of State Constitutional Jurisprudence: A Judge's Thoughts, 13 Vt. L. Rev. 145, 151-52 (1988) (noting distinct eighteenth-century origins of Article 7). The same assumption that all the people should be afforded all the benefits and protections bestowed by government is also reflected in the second section, which prohibits not the denial of rights to the oppressed, but rather the conferral of advantages or emoluments upon the privileged.
The current version differs from the original only in that the gender-neutral terms "person" and "persons" have been substituted for "man" and "men." See Vt. Const., Ch. II § 76. This revision was not intended to "alter the sense, meaning or effect of the" provision. Id.
There is little doubt as to the obligatory nature of the Common Benefits Clause, which provides that "government is, or ought to be, instituted for the common benefit, protection, and security . . . ." (Emphasis added). Indeed the State does not argue that it is merely hortatory or aspirational in effect, an argument that would not be persuasive in any event. See Brigham, 166 Vt. at 261-62, 692 A.2d at 393-94 (1997) (framers "drew no distinction between `ought' and `shall' in defining rights and duties").
The words of the Common Benefits Clause are revealing. While they do not, to be sure, set forth a fully-formed standard of analysis for determining the constitutionality of a given statute, they do express broad principles which usefully inform that analysis. Chief among these is the principle of inclusion. As explained more fully in the discussion that follows, the specific proscription against governmental favoritism toward not only groups or "set[s] of men," but also toward any particular "family" or "single man," underscores the framers' resentment of political preference of any kind. The affirmative right to the "common benefits and protections" of government and the corollary proscription of favoritism in the distribution of public "emoluments and advantages" reflect the framers' overarching objective "not only that everyone enjoy equality before the law or have an equal voice in government but also that everyone have an equal share in the fruits of the common enterprise." W. Adams, The First American Constitutions 188 (1980) (emphasis added). Thus, at its core the Common Benefits Clause expressed a vision of government that afforded every Vermonter its benefit and protection and provided no Vermonter particular advantage.
C. Historical Context
Although historical research yields little direct evidence of the framers' intentions, an examination of the ideological origins of the Common Benefits Clause casts a useful light upon the inclusionary principle at its textual core. Like other provisions of the Vermont Constitution of 1777, the Common Benefits Clause was borrowed verbatim from the Pennsylvania Constitution of 1776, which was based, in turn, upon a similar provision in the Virginia Declaration of Rights of 1776. See J. Shaeffer, A Comparison of the First Constitutions of Vermont and Pennsylvania, 43 Vt. Hist. 33, 33-35 (1975); J. Selsam, The Pennsylvania Constitution of 1776: A Study in Revolutionary Democracy 178 (1936). The original Virginia clause differed from the Pennsylvania and Vermont provisions only in the second section, which was contained in a separate article and provided "[t]hat no man, or set of men, are entitled to exclusive or separate emoluments or privileges from the community, but in consideration of public services." See Virginia Declaration of Rights, art. IV (reprinted in 11 West's Encyclopedia of American Law 82 (1998)).
The use of the word "family" in the Pennsylvania Common Benefits Clause reflects Pennsylvania's history, where elite "proprietors" including the Penns and other established families, had long dominated colonial politics, religion, and economic interests. The revolt against Great Britain presented an opportunity for western Pennsylvania farmers, urban gentry, and dissenting Presbyterians nursing "deep seated and long-felt grievances" to end Eastern domination of the colony, and establish a more democratic form of government. See Selsam, supra, at 1, 255-56.
Although aimed at Great Britain, the American Revolution — as numerous historians have noted — also tapped deep-seated domestic antagonisms. The planter elite in Virginia, the proprietors of Eastern Pennsylvania, and New Yorkers claiming Vermont lands were each the object of long-standing grievances. Selsam, supra, at 255-56; R. Shalhope, Bennington and the Green Mountain Boys: The Emergence of Liberal Democracy in Vermont, 1760-1850 at 70-97 (1996); G. Wood, The Creation of the American Republic, 1776-1787 at 75-82 (1969). Indeed, the revolt against Great Britain unleashed what one historian, speaking of Pennsylvania, has called "a revolution within a revolution." Selsam, supra, at 1. By attempting to claim equal rights for Americans against the English, regardless of birthright or social status, "even the most aristocratic of southern Whig planters . . . were pushed into creating an egalitarian ideology that could be and even as early as 1776 was being turned against themselves." Wood, supra, at 83. While not opposed to the concept of a social elite, the framers of the first state constitutions believed that it should consist of a "natural aristocracy" of talent, rather than an entrenched clique favored by birth or social connections. See id. at 479-80. As the preeminent historian of the ideological origins of the Revolution explained, "while `equality before the law' was a commonplace of the time, `equality without respect to the dignity of the persons concerned' was not; [the Revolution's] emphasis on social equivalence was significant." B. Bailyn, The Ideological Origins of the American Revolution 307 (1967). Thus, while the framers' "egalitarian ideology" conspicuously excluded many oppressed people of the eighteenth century — including African-Americans, Native Americans, and women — it did nevertheless represent a genuine social revolt pitting republican ideals of "virtue," or talent and merit, against a perceived aristocracy of privilege both abroad and at home.
Vermont was not immune to the disruptive forces unleased by the Revolution. One historian has described Vermont on the eve of the Revolution as rife with "factional rivalry [and] regional jealousy." G. Aichele, Making the Vermont Constitution: 1777-1824, 56 Vt. Hist. 166, 177 (1988). Competing factions in the Champlain and Upper Connecticut River Valleys had long vied for political and economic dominance. See id. at 180. Echoing Selsam on Pennsylvania, another historian has spoken of "Vermont's double revolution — a rebellion within a rebellion" to describe the successful revolt against both Great Britain and New York by the yeoman farmers, small-scale proprietors, and moderate land speculators who comprised the bulk of the Green Mountain Boys. D. Smith, Green Mountain Insurgency: Transformation of New York's Forty-Year Land War, 64 Vt. Hist. 197, 197-98, 224 (1996); see also Shalhope, supra, at 169 (egalitarian ideology of American Revolution "resonated powerfully with the visceral feelings" of Green Mountain Boys and others in Vermont).
The powerful movement for "social equivalence" unleashed by the Revolution ultimately found its most complete expression in the first state constitutions adopted in the early years of the rebellion. In Pennsylvania, where social antagonisms were most acute, the result was a fundamental charter that has been described as "the most radical constitution of the Revolution." Wood, supra, at 84-85; see also Shaeffer, supra, at 35-36. Yet the Pennsylvania Constitution's egalitarianism was arguably eclipsed the following year by the Vermont Constitution of 1777. In addition to the commitment to government for the "common benefit, protection, and security," it contained novel provisions abolishing slavery, eliminating property qualifications for voting, and calling for the governor, lieutenant governor, and twelve councilors to be elected by the people rather than appointed by the Legislature. See Shalhope, supra, at 171-72. These and other provisions have led one historian to observe that Vermont's first charter was the "most democratic constitution produced by any of the American states." See id. at 172.
The historical origins of the Vermont Constitution thus reveal that the framers, although enlightened for their day, were not principally concerned with civil rights for African-Americans and other minorities, but with equal access to public benefits and protections for the community as a whole. The concept of equality at the core of the Common Benefits Clause was not the eradication of racial or class distinctions, but rather the elimination of artificial governmental preferments and advantages. The Vermont Constitution would ensure that the law uniformly afforded every Vermonter its benefit, protection, and security so that social and political preeminence would reflect differences of capacity, disposition, and virtue, rather than governmental favor and privilege. D. Analysis under Article 7
This Court has noted that interpretations of similar constitutional provisions from other states may be instructive in understanding our own. See Benning, 161 Vt. at 476, 641 A.2d at 759. "Common Benefits" decisions from other states, however, are scarce. Pennsylvania eliminated the Common Benefits Clause when it replaced its constitution in 1790, and Virginia courts have not explored in any depth the meaning of its clause. The New Hampshire Constitution of 1783 also included a common benefits section substantially similar to Vermont's. See N.H. Const., Pt. 1, art. 10. Although New Hampshire courts have not developed an independent Common Benefits jurisprudence, several early New Hampshire decisions noted the provision's significance. See State v. Pennoyer, 18 A.2d 878, 881 (1889) (relying on Common Benefits Clause to strike down physician-licensing statute that exempted physicians who had resided in one place for four years); Rosenblum v. Griffin, 197 A. 701, 706 (1938) (noting that under Common Benefits Clause, "[e]quality of benefit is no less required than equality of burden. Otherwise equal protection is denied"). Massachusetts included a variation on Vermont's Common Benefits Clause in its Constitution of 1780, as well as a separate "emoluments" provision. See Mass. Const., Pt. 1, arts. VI VII (adopted 1780). Massachusetts has not relied on the Common Benefits provision as a separate source of equal protections rights. See Town of Brookline v. Secretary of Com., 631 N.E.2d 968, 978 n. 19 (Mass. 1994).
In the nineteenth century, a number of additional states adopted variations on the Common Benefits Clause. See, e.g., Conn. Const. of 1818, art. 1, § 2 ("[A]ll political power is inherent in the people, and all free governments are founded on their authority, and instituted for their benefit."); Ohio Const. of 1851, art. 1, § 2 ("All political power is inherent in the people. Government is instituted for their equal protection and benefit."); W. Va. Const. Const., art. III, § 3 (adopted 1872) ("Government is instituted for the common benefit, protection and security of the people, nation or community."). Even assuming that provisions enacted in the nineteenth century have some bearing on the meaning of a Revolutionary-era document, these sister-state constitutions provide little guidance. Ohio has held that the state clause is the "functional equivalent" of the Equal Protection Clause with similar standards. See American Ass'n of Univ. Professors v. Central State Univ., 699 N.E.2d 463, 467 (Ohio 1998). The West Virginia Supreme Court, in contrast, has relied on the Common Benefits Clause to hold that the State constitution provides greater individual protection than the United States Constitution. See United Mine Workers of Am. Inter. Union v. Parsons, 305 S.E.2d 343, 353-54 (W.Va. 1983). Apart from noting the absence of an equivalent provision in the federal constitution, however, the West Virginia court has not engaged in any extensive textual or historical analysis.
A number of states during the Revolutionary and early National periods also adopted separate provisions, apparently modeled on the Pennsylvania and Virginia clauses, declaring that no men, or set of men, are entitled to exclusive or separate emoluments or privileges from the community, but in consideration of public services. See, e.g., N.C. Const. of 1776, Decl. of Rights, § 3; Mass. Const., Pt. 1, art. VI; Conn. Const. of 1818, art. I, § 1; Miss. Const. of 1832, art. I, § 1; Ky. Const. of 1792, art. XII, § 1. These "emoluments and privileges" clauses have been extensively cited and applied, often in the context of taxpayer suits challenging public expenditures as unconstitutional "gifts" of public funds without consideration of public service, or suits challenging legislative acts granting special credits, payments, or exemptions to a specific class. see, e.g., Commissioner of Pub. Works v. City of Middletown. 731 A.2d 749. 757 (Conn. 1999) (challenge to tax exemption); Driscoll v. City of New Haven, 52 A. 618, 622 (Conn. 1902) (taxpayer suit to enjoin municipal grant of land to private company); Kentucky Union R.R. Co. v. Bourbon County, 2. S.W. 687, 690 (Ky. 1887) (taxpayer suit to enjoin subscription of bonds for railroad purposes); Brumley v. Baxter, 36 S.E.2d 281, 286 (N.C. 1945) (taxpayer suit to enjoin municipal grant of real property for use by military veterans); see also Gross v. Auditor of Accounts, 109 Vt. 156, 159, 194 A. 465, 467 (1937) (Article 7 challenge to payment to sheriff's widow as "emolument" without consideration of public service). These cases generally turned on whether the challenged action promoted a public purpose or was made without some consideration of public service. They represent, in effect, the reverse of the Common Benefits Clause, prohibiting the grant of special privileges to a select class of persons over and above those granted to the general community, as the Common Benefits Clause requires the equal enjoyment of general benefits and protections by the whole community.
The language and history of the Common Benefits Clause thus reinforce the conclusion that a relatively uniform standard, reflective of the inclusionary principle at its core, must govern our analysis of laws challenged under the Clause. Accordingly, we conclude that this approach, rather than the rigid, multi-tiered analysis evolved by the federal courts under the Fourteenth Amendment, shall direct our inquiry under Article 7. As noted, Article 7 is intended to ensure that the benefits and protections conferred by the State are for the common benefit of the community and are not for the advantage of persons "who are a part only of that community." When a statute is challenged under Article 7, we first define that "part of the community" disadvantaged by the law. We examine the statutory basis that distinguishes those protected by the law from those excluded from the State's protection. Our concern here is with delineating, not with labelling the excluded class as "suspect," "quasi-suspect," or "non-suspect" for purposes of determining different levels of judicial scrutiny.
The concurring opinion would tie its analysis to the presumably "objective" test of suspect class. But suspect class analysis has never provided a stable mooring for constitutional application of Vermont's Common Benefits Clause. Although the concurrence identifies precedents of this Court holding that a more searching scrutiny is required when a statutory scheme involves suspect classes, we have never established the criteria for determining what constitutes a suspect class under the Vermont Constitution nor have we ever identified a suspect class under Article 7. Moreover, the concurrence applies strict scrutiny predicated on a finding that lesbians and gay men are a suspect class, although the overwhelming majority of decisions have rejected such claims. See Ben-Shalom v. Marsh, 881 F.2d 454, 464-66 (7th Cir. 1989), cert denied, 494 U.S. 1004 (1990); Equality Found'n of Greater Cincinnati, Inc. v. City of Cincinnati, 128 F.3d 289, 292-93 (6th Cir. 1997); Thomasson v. Perry, 80 F.3d 915, 927 (4th Cir.), cert. denied, 519 U.S. 948 (1996); Richenberg v. Perry, 97 F.3d 256, 260-61 (8th Cir. 1996), cert. denied, 522 U.S. 807) (1997); High Tech Gays v. Defense Indus. Sec. Clearance Office, 895 F.2d 563, 571-72 (9th Cir. 1990); Woodward v. United States, 871 F.2d 1068, 1076 (Fed. Cir. 1989), cert. denied, 494 U.S. 1002 (1990); Padula v. Webster, 822 F.2d 97, 103 (D.C. Cir. 1987); Baker v. Wade, 769 F.2d 289, 292 (5th Cir. 1985) (en banc), cert. denied, 478 U.S. 1035 (1986); National Gay Task Force v. Board of Educ., 729 F.2d 1270, 1273 (10th Cir. 1984), aff'd 470 U.S. 903 (1985); Opinion of the Justices, 530 A.2d 21, 24 (N.H. 1987).
The Court — no less than the concurrence — seeks a rationale faithful to our Constitution and careful in the exercise of this Court's limited powers. The concurrence suggests that the Oregon Supreme Court's decision in Hewitt v. State Accident Ins. Fund Corp., 653 P.2d 970, 977-78 (Or. 1982) should be relied upon to supply the missing Vermont jurisprudence of suspect class criteria. Yet, the Oregon Court of Appeals found it necessary to abandon the immutable personal-characteristic criterion of Hewitt in order to find that homosexuals were a suspect class entitled to heightened scrutiny. See Tanner v. Oregon Health Sciences Univ., 971 P.2d 435, 446 (Or.Ct.App. 1998). The "adverse stereotyping" analysis used in its place, see id., may provide one intermediate appellate court's answer to the question of whether homosexuals are a suspect class, but it is far from an "exacting standard" by which to measure the prudence of a court's exercise of its powers. It is difficult to imagine a legal framework that could provide less predictability in the outcome of future cases than one which gives a court free reign to decide which groups have been the subject of "adverse social or political stereotyping." Id. The artificiality of suspect-class labeling should be avoided where, as here, the plaintiffs are afforded the common benefits and protections of Article 7, not because they are part of a "suspect class," but because they are part of the Vermont community.
We look next to the government's purpose in drawing a classification that includes some members of the community within the scope of the challenged law but excludes others. Consistent with Article 7's guiding principle of affording the protection and benefit of the law to all members of the Vermont community, we examine the nature of the classification to determine whether it is reasonably necessary to accomplish the State's claimed objectives.
We must ultimately ascertain whether the omission of a part of the community from the benefit, protection and security of the challenged law bears a reasonable and just relation to the governmental purpose. Consistent with the core presumption of inclusion, factors to be considered in this determination may include: (1) the significance of the benefits and protections of the challenged law; (2) whether the omission of members of the community from the benefits and protections of the challenged law promotes the government's stated goals; and (3) whether the classification is significantly underinclusive or overinclusive. As Justice Souter has observed in a different context, this approach necessarily "calls for a court to assess the relative `weights' or dignities of the contending interests." Washington v. Glucksberg, 521 U.S. 702, 767 (1997) (Souter, J., concurring). What keeps that assessment grounded and objective, and not based upon the private sensitivities or values of individual judges, is that in assessing the relative weights of competing interests courts must look to the history and "`traditions from which [the State] developed'" as well as those "`from which it broke,'" id. at 767 (quoting Poe v. Ullman, 367 U.S. 497, 542 (1961) (Harlan, J., dissenting)), and not to merely personal notions. Moreover, the process of review is necessarily "one of close criticism going to the details of the opposing interests and their relationships with the historically recognized principles that lend them weight or value." Id. at 769 (emphasis added). Ultimately, the answers to these questions, however useful, cannot substitute for "`[t]he inescapable fact . . . that adjudication of . . . claims may call upon the Court in interpreting the Constitution to exercise that same capacity which by tradition courts always have exercised: reasoned judgment.'" Id. (quoting Planned Parenthood of Southeastern Pa. v. Casey, 505 U.S. 833, 849 (1992)). The balance between individual liberty and organized society which courts are continually called upon to weigh does not lend itself to the precision of a scale. It is, indeed, a recognition of the imprecision of "reasoned judgment" that compels both judicial restraint and respect for tradition in constitutional interpretation.
The concurring and concurring and dissenting opinions are mistaken in suggesting that this standard places identical burdens upon the State regardless of the nature of the rights affected. As explained above, the significance of the benefits and protections at issue may well affect the justifications required of the State to support a statutory classification. This is plainly demonstrated in the discussion of marriage benefits and protections which follows. Nor is there any merit to the assertion that this standard invites a more "activist" review of economic and social welfare legislation. See post, at 15 (Dooley, J., concurring). Characterizing a case as affecting "economic" interests, "civil rights," "fundamental" rights, or "suspect classes" — as our colleagues apparently prefer — is no less an exercise in judgment. Indeed, it may disguise the court's value judgments with a label, rather than explain its reasoning in terms that the public and the litigants are entitled to understand. "It is a comparison of the relative strengths of opposing claims that informs the judicial task, not a deduction from some first premise." Glucksberg, 521 U.S. at 764 (Souter, J., concurring). That is a task we trust will continue to be undertaken in a legal climate that recognizes that "constitutional review, not judicial lawmaking, is a court's business here." Id. at 768.
Justice Harlan has described the process of constitutional interpretation as follows:
If the supplying of content to this Constitutional concept has of necessity been a rational process, it certainly has not been one where judges have felt free to roam where unguided speculation might take them. The balance of which I speak is the balance struck by this country, having regard to what history teaches are the traditions from which it developed as well as the traditions from which it broke. That tradition is a living thing. A decision of this Court which radically departs from it could not long survive, while a decision which builds on what has survived is likely to be sound. No formula could serve as a substitute, in this area, for judgment and restraint. Poe, 367 U.S. at 542 (Harlan, J. dissenting).
E. The Standard Applied
With these general precepts in mind, we turn to the question of whether the exclusion of same-sex couples from the benefits and protections incident to marriage under Vermont law contravenes Article 7. The first step in our analysis is to identify the nature of the statutory classification. As noted, the marriage statutes apply expressly to opposite-sex couples. Thus, the statutes exclude anyone who wishes to marry someone of the same sex. Next, we must identify the governmental purpose or purposes to be served by the statutory classification. The principal purpose the State advances in support of the excluding same-sex couples from the legal benefits of marriage is the government's interest in "furthering the link between procreation and child rearing." The State has a strong interest, it argues, in promoting a permanent commitment between couples who have children to ensure that their offspring are considered legitimate and receive ongoing parental support. The State contends, further, that the Legislature could reasonably believe that sanctioning same-sex unions "would diminish society's perception of the link between procreation and child rearing . . . [and] advance the notion that fathers or mothers . . . are mere surplusage to the functions of procreation and child rearing." The State argues that since same-sex couples cannot conceive a child on their own, state-sanctioned same-sex unions "could be seen by the Legislature to separate further the connection between procreation and parental responsibilities for raising children." Hence, the Legislature is justified, the State concludes, "in using the marriage statutes to send a public message that procreation and child rearing are intertwined."
Relying largely on federal precedents, our colleague in her concurring and dissenting opinion suggests that the statutory exclusion of same-sex couples from the benefits and protections of marriage should be subject to heightened scrutiny as a "suspect" or "quasi-suspect" classification based on sex. All of the seminal sex-discrimination decisions, however, have invalidated statutes that single-out men or women as a discrete class for unequal treatment. See, e.g., United States v. Virginia, 518 U.S. 515, 555-56 (1996) (repudiating statute that precluded women from attending Virginia Military Institute); Mississippi Univ. for Women v. Hogan, 458 U.S. 718, 731 (1982) (invalidating admission policy that excluded males from attending state-supported nursing school); Craig v. Boren, 429 U.S. 190, 204 (1976) (invalidating statute that allowed women to purchase non-intoxicating beer at younger age than men); Frontiero v. Richardson, 411 U.S. 677, 690 (1973) (striking statute that imposed more onerous requirements upon female members of armed services to claim spouses as dependents).
Although this Court has not addressed the issue, see State v. George, 157 Vt. 580, 588, 602 A.2d 953, 957 (1991), we do not doubt that a statute that discriminated on the basis of sex would bear a heavy burden under the Article 7 analysis set forth above. The difficulty here is that the marriage laws are facially neutral; they do not single-out men or women as a class for disparate treatment, but rather prohibit men and women equally from marrying a person of the same sex. As we observed in George, 157 Vt. at 585, 602 A.2d at 956, "[i]n order to trigger equal protection analysis at all . . . a defendant must show that he was treated differently as a member of one class from treatment of members of another class similarly situated." (Emphasis added). Here, there is no discrete class subject to differential treatment solely on the basis of sex; each sex is equally prohibited from precisely the same conduct.
Indeed, most appellate courts that have addressed the issue have rejected the claim that defining marriage as the union of one man and one woman discriminates on the basis of sex. See, e.g. Baker v. Nelson, 191 N.W.2d 185, 186-87 (Minn. 1971); Singer v. Hara, 522 P.2d 1187, 1191-92 (Wash.Ct.App. 1974); see also Phillips v. Wisconsin Personnel Comm'n, 482 N.W.2d 121, 129 (Wis.Ct.App. 1992) (holding that health insurance regulation limiting state employee's dependent coverage to spouse did not constitute sex discrimination because coverage was "unavailable to unmarried companions of both male and female employees"); State v. Walsh, 713 S.W.2d 508, 510 (Mo. 1986) (rejecting claim that sodomy statute imposed sex-based classification because it "applie[d] equally to men and women [in] prohibit[ing] both classes from engaging in sexual activity with members of their own sex"). But see Baehr v. Lewin, 852 P.2d 44, 64 (Haw. 1993) (plurality opinion holding that state's marriage laws discriminated on basis of sex).
Although the concurring and dissenting opinion invokes the United States Supreme Court decision in Loving v. Virginia, 388 U.S. 1 (1967), the reliance is misplaced. There the high court had little difficulty in looking behind the superficial neutrality of Virginia's anti-miscegenation statute to hold that its real purpose was to maintain the pernicious doctrine of white supremacy. Id. at 11. Our colleague argues, by analogy, that the effect, if not the purpose, of the exclusion of same-sex partners from the marriage laws is to maintain certain male and female stereotypes to the detriment of both. To support the claim, she cites a number of antiquated statutes that denied married women a variety of freedoms, including the right to enter into contracts and hold property.
The test to evaluate whether a facially gender-neutral statute discriminates on the basis of sex is whether the law "can be traced to a discriminatory purpose." Personnel Administrator v. Feeney, 442 U.S. 256, 272 (1979). The evidence does not demonstrate such a purpose. It is one thing to show that long-repealed marriage statutes subordinated women to men within the marital relation. It is quite another to demonstrate that the authors of the marriage laws excluded same-sex couples because of incorrect and discriminatory assumptions about gender roles or anxiety about gender-role confusion. That evidence is not before us. Accordingly, we are not persuaded that sex discrimination offers a useful analytic framework for determining plaintiffs' rights under the Common Benefits Clause.
Do these concerns represent valid public interests that are reasonably furthered by the exclusion of same-sex couples from the benefits and protections that flow from the marital relation? It is beyond dispute that the State has a legitimate and long-standing interest in promoting a permanent commitment between couples for the security of their children. It is equally undeniable that the State's interest has been advanced by extending formal public sanction and protection to the union, or marriage, of those couples considered capable of having children, i.e., men and women. And there is no doubt that the overwhelming majority of births today continue to result from natural conception between one man and one woman. See J. Robertson, Assisted Reproductive Technology and the Family, 47 Hast. L. J. 911, 911-12 (1996) (noting the number of births resulting from assisted-reproductive technology, which remain small compared to overall number of births).
It is equally undisputed that many opposite-sex couples marry for reasons unrelated to procreation, that some of these couples never intend to have children, and that others are incapable of having children. Therefore, if the purpose of the statutory exclusion of same-sex couples is to "further the link between procreation and child rearing," it is significantly under-inclusive. The law extends the benefits and protections of marriage to many persons with no logical connection to the stated governmental goal.
Furthermore, while accurate statistics are difficult to obtain, there is no dispute that a significant number of children today are actually being raised by same-sex parents, and that increasing numbers of children are being conceived by such parents through a variety of assisted-reproductive techniques. See D. Flaks, et al., Lesbians Choosing Motherhood: A Comparative Study of Lesbian and Heterosexual Parents and Their Children, 31 Dev. Psychol. 105, 105 (1995) (citing estimates that between 1.5 and 5 million lesbian mothers resided with their children in United States between 1989 and 1990, and that thousands of lesbian mothers have chosen motherhood through donor insemination or adoption); G. Green and F. Bozett, Lesbian Mothers and Gay Fathers, in Homosexuality: Research Implications for Public Policy 197, 198 (J. Gonsiorek et al. eds., 1991) (estimating that numbers of children of either gay fathers or lesbian mothers range between six and fourteen million); C. Patterson, Children of the Lesbian Baby Boom: Behavioral Adjustment, Self-Concepts, and Sex Role Identity, in Lesbian and Gay Psychology (B. Greene et al. eds., 1994) (observing that although precise estimates are difficult, number of families with lesbian mothers is growing); E. Shapiro L. Schultz, Single-Sex Families: The Impact of Birth Innovations Upon Traditional Family Notions, 24 J. Fam. L. 271, 281 (1985) ("[I]t is a fact that children are being born to single-sex families on a biological basis, and that they are being so born in considerable numbers").
Thus, with or without the marriage sanction, the reality today is that increasing numbers of same-sex couples are employing increasingly efficient assisted-reproductive techniques to conceive and raise children. See L. Ikemoto, The In/Fertile, the Too Fertile, and the Dysfertile, 47 Hast. L. J. 1007, 1056 n. 170 (1996). The Vermont Legislature has not only recognized this reality, but has acted affirmatively to remove legal barriers so that same-sex couples may legally adopt and rear the children conceived through such efforts. See 15A V.S.A. § 1-102(b) (allowing partner of biological parent to adopt if in child's best interest without reference to sex). The State has also acted to expand the domestic relations laws to safeguard the interests of same-sex parents and their children when such couples terminate their domestic relationship. See 15A V.S.A. § 1-112 (vesting family court with jurisdiction over parental rights and responsibilities, parent-child contact, and child support when unmarried persons who have adopted minor child "terminate their domestic relationship").
Therefore, to the extent that the State's purpose in licensing civil marriage was, and is, to legitimize children and provide for their security, the statutes plainly exclude many same-sex couples who are no different from opposite-sex couples with respect to these objectives. If anything, the exclusion of same-sex couples from the legal protections incident to marriage exposes their children to the precise risks that the State argues the marriage laws are designed to secure against. In short, the marital exclusion treats persons who are similarly situated for purposes of the law, differently.
The State also argues that because same-sex couples cannot conceive a child on their own, their exclusion promotes a "perception of the link between procreation and child rearing," and that to discard it would "advance the notion that mothers and fathers . . . are mere surplusage to the functions of procreation and child rearing" Apart from the bare assertion, the State offers no persuasive reasoning to support these claims. Indeed, it is undisputed that most of those who utilize non-traditional means of conception are infertile married couples, see Shapior and Schultz, supra, at 275, and that many assisted-reproductive techniques involve only one of the married partner's genetic material, the other being supplied by a third party through sperm, egg, or embryo donation. See E. May, Barren in the Promised Land: Childless Americans and the Pursuit of Happiness, 217, 242 (1995); Robertson, supra, at 911-12, 922-27. The State does not suggest that the use of these technologies undermines a married couple's sense of parental responsibility, or fosters the perception that they are "mere surplusage" to the conception and parenting of the child so conceived. Nor does it even remotely suggest that access to such techniques ought to be restricted as a matter of public policy to "send a public message that procreation and child rearing are intertwined." Accordingly, there is no reasonable basis to conclude that a same-sex couple's use of the same technologies would undermine the bonds of parenthood, or society's perception of parenthood.
The question thus becomes whether the exclusion of a relatively small but significant number of otherwise qualified same-sex couples from the same legal benefits and protections afforded their opposite-sex counterparts contravenes the mandates of Article 7. It is, of course, well settled that statutes are not necessarily unconstitutional because they fail to extend legal protection to all who are similarly situated. See Benning, 161 Vt. at 486, 641 A.2d at 764 ("A statute need not regulate the whole of a field to pass constitutional muster."). Courts have upheld underinclusive statutes out of a recognition that, for reasons of pragmatism or administrative convenience, the legislature may choose to address problems incrementally. See, e.g., City of New Orleans v. Dukes, 427 U.S. 297, 303 (1976) (legislature may adopt regulations "that only partially ameliorate a perceived evil"); Williamson v. Lee Optical of Okla., Inc., 348 U.S. 483, 489 (1955) ("The legislature may select one phase of one field and apply a remedy there, neglecting the others."). The State does not contend, however, that the same-sex exclusion is necessary as a matter of pragmatism or administrative convenience. We turn, accordingly, from the principal justifications advanced by the State to the interests asserted by plaintiffs.
As noted, in determining whether a statutory exclusion reasonably relates to the governmental purpose it is appropriate to consider the history and significance of the benefits denied. See Glucksberg, 521 U.S. at 710 (to assess importance of rights and interests affected by statutory classifications, courts must look to "history, legal traditions and practices"). What do these considerations reveal about the benefits and protections at issue here? In Loving v. Virginia, 388 U.S. 1, 12 (1967), the United States Supreme Court, striking down Virginia's anti-miscegenation law, observed that "[t]he freedom to marry has long been recognized as one of the vital personal rights." The Court's point was clear; access to a civil marriage license and the multitude of legal benefits, protections, and obligations that flow from it significantly enhance the quality of life in our society.
The Supreme Court's observations in Loving merely acknowledged what many states, including Vermont, had long recognized. One hundred thirty-seven years before Loving, this Court characterized the reciprocal rights and responsibilities flowing from the marriage laws as "the natural rights of human nature." See Overseers of the Poor, 2 Vt. at 159. Decisions in other New England states noted the unique legal and economic ramifications flowing from the marriage relation. See, e.g., Adams v. Palmer, 51 Maine 481, 485 (Me. 1863) ("it establishes fundamental and most important domestic relations"). Early decisions recognized that a marriage contract, although similar to other civil agreements, represents much more because once formed, the law imposes a variety of obligations, protections, and benefits. As the Maine Supreme Judicial Court observed, the rights and obligations of marriage rest not upon contract, "but upon the general law of the State, statutory or common, which defines and prescribes those rights duties and obligations. They are of law, not contract." See id. at 483; see also Ditson v. Ditson, 4 R.I. 87, 105 (1856) (marriage transcends contract because "it gives rights, and imposes duties and restrictions upon the parties to it"). In short, the marriage laws transform a private agreement into a source of significant public benefits and protections. While the laws relating to marriage have undergone many changes during the last century, largely toward the goal of equalizing the status of husbands and wives, the benefits of marriage have not diminished in value. On the contrary, the benefits and protections incident to a marriage license under Vermont law have never been greater. They include, for example, the right to receive a portion of the estate of a spouse who dies intestate and protection against disinheritance through elective share provisions, under 14 V.S.A. § 401-404, 551; preference in being appointed as the personal representative of a spouse who dies intestate, under 14 V.S.A. § 903; the right to bring a lawsuit for the wrongful death of a spouse, under 14 V.S.A. § 1492; the right to bring an action for loss of consortium, under 12 V.S.A. § 5431; the right to workers' compensation survivor benefits under 21 V.S.A. § 632; the right to spousal benefits statutorily guaranteed to public employees, including health, life, disability, and accident insurance, under 3 V.S.A. § 631; the opportunity to be covered as a spouse under group life insurance policies issued to an employee, under 8 V.S.A. § 3811; the opportunity to be covered as the insured's spouse under an individual health insurance policy, under 8 V.S.A. § 4063; the right to claim an evidentiary privilege for marital communications, under V.R.E. 504; homestead rights and protections, under 27 V.S.A. § 105-108, 141-142; the presumption of joint ownership of property and the concomitant right of survivorship, under 27 V.S.A. § 2; hospital visitation and other rights incident to the medical treatment of a family member, under 18 V.S.A. § 1852; and the right to receive, and the obligation to provide, spousal support, maintenance, and property division in the event of separation or divorce, under 15 V.S.A. § 751-752. Other courts and commentators have noted the collection of rights, powers, privileges, and responsibilities triggered by marriage. See generally Baehr v. Lewin, 852 P.2d 44, 59 (Haw. 1993); D. Chambers, What If? The Legal Consequences of Marriage and the Legal Needs of Lesbian and Gay Male Couples, 95 Mich. L. Rev. 447, passim; J. Robbenolt M. Johnson, Legal Planning for Unmarried Committed Parties: Empirical Lessons for a Preventive and Therapeutic Approach, 41 Ariz. L. Rev. 417, passim (1999); J. Trosino, American Wedding: Same-Sex Marriage and the Miscegenation Analogy, 73 B.U.L. Rev. 93, 96 (1993).
While other statutes could be added to this list, the point is clear. The legal benefits and protections flowing from a marriage license are of such significance that any statutory exclusion must necessarily be grounded on public concerns of sufficient weight, cogency, and authority that the justice of the deprivation cannot seriously be questioned. Considered in light of the extreme logical disjunction between the classification and the stated purposes of the law — protecting children and "furthering the link between procreation and child rearing" — the exclusion falls substantially short of this standard. The laudable governmental goal of promoting a commitment between married couples to promote the security of their children and the community as a whole provides no reasonable basis for denying the legal benefits and protections of marriage to same-sex couples, who are no differently situated with respect to this goal than their opposite-sex counterparts. Promoting a link between procreation and childrearing similarly fails to support the exclusion. We turn, accordingly, to the remaining interests identified by the State in support of the statutory exclusion.
The State asserts that a number of additional rationales could support a legislative decision to exclude same-sex partners from the statutory benefits and protections of marriage. Among these are the State's purported interests in "promoting child rearing in a setting that provides both male and female role models," minimizing the legal complications of surrogacy contracts and sperm donors, "bridging differences" between the sexes, discouraging marriages of convenience for tax, housing or other benefits, maintaining uniformity with marriage laws in other states, and generally protecting marriage from "destabilizing changes." The most substantive of the State's remaining claims relates to the issue of childrearing. It is conceivable that the Legislature could conclude that opposite-sex partners offer advantages in this area, although we note that child-development experts disagree and the answer is decidedly uncertain. The argument, however, contains a more fundamental flaw, and that is the Legislature's endorsement of a policy diametrically at odds with the State's claim. In 1996, the Vermont General Assembly enacted, and the Governor signed, a law removing all prior legal barriers to the adoption of children by same-sex couples. See 15A V.S.A. § 1-102. At the same time, the Legislature provided additional legal protections in the form of court-ordered child support and parent-child contact in the event that same-sex parents dissolved their "domestic relationship." Id. § 1-112. In light of these express policy choices, the State's arguments that Vermont public policy favors opposite-sex over same-sex parents or disfavors the use of artificial reproductive technologies, are patently without substance.
Similarly, the State's argument that Vermont's marriage laws serve a substantial governmental interest in maintaining uniformity with other jurisdictions cannot be reconciled with Vermont's recognition of unions, such as first-cousin marriages, not uniformly sanctioned in other states. See 15 V.S.A. § 1-2 (consanguinity statutes do not exclude first cousins); 1 H. Clark, The Law of Domestic Relations in the United States § 2.9, at 153-54 (2d ed. 1987) (noting states that prohibit first-cousin marriage). In an analogous context, Vermont has sanctioned adoptions by same-sex partners, see 15A V.S.A. § 1-102, notwithstanding the fact that many states have not. See generally, Annotation, Adoption of Child By Same-Sex Partners, 27 A.L.R. 5th 54, 68-72 (1995). Thus, the State's claim that Vermont's marriage laws were adopted because the Legislature sought to conform to those of the other forty-nine states is not only speculative, but refuted by two relevant legislative choices which demonstrate that uniformity with other jurisdictions has not been a governmental purpose.
The State's remaining claims (e.g., recognition of same-sex unions might foster marriages of convenience or otherwise affect the institution in "unpredictable" ways) may be plausible forecasts as to what the future may hold, but cannot reasonably be construed to provide a reasonable and just basis for the statutory exclusion. The State's conjectures are not, in any event, susceptible to empirical proof before they occur.
It would, for example, serve no useful purpose to remand this matter for hearings on whether marriages of convenience (i.e., unions for the purpose of obtaining certain statutory benefits) would result from providing same-sex couples with the statutory benefits and protections accorded opposite-sex couples under marriage laws. For the reasons we have stated in this opinion, it is not a failure of proof that is fatal to the State's arguments, it is a failure of logic.
Finally, it is suggested that the long history of official intolerance of intimate same-sex relationships cannot be reconciled with an interpretation of Article 7 that would give state-sanctioned benefits and protection to individuals of the same sex who commit to a permanent domestic relationship. We find the argument to be unpersuasive for several reasons. First, to the extent that state action historically has been motivated by an animus against a class, that history cannot provide a legitimate basis for continued unequal application of the law. See MacCallum, 165 Vt. at 459-60, 686 A.2d at 939 (holding that although adopted persons had "historically been a target of discrimination," social prejudices failed to support their continued exclusion from intestacy law). As we observed recently in Brigham, 166 Vt. at 267, 692 A.2d at 396, "equal protection of the laws cannot be limited by eighteenth-century standards." Second, whatever claim may be made in light of the undeniable fact that federal and state statutes — including those in Vermont — have historically disfavored same-sex relationships, more recent legislation plainly undermines the contention. See, e.g., Laws of Vermont, 1977, No. 51, § 2, 3 (repealing former § 2603 of Title 13, which criminalized fellatio). In 1991, Vermont was one of the first states to enact statewide legislation prohibiting discrimination in employment, housing, and other services based on sexual orientation. See 21 V.S.A. § 495 (employment); 9 V.S.A. § 4503 (housing); 8 V.S.A. § 4724 (insurance); 9 V.S.A. § 4502 (public accommodations). Sexual orientation is among the categories specifically protected against hate-motivated crimes in Vermont. See 13 V.S.A. § 1455. Furthermore, as noted earlier, recent enactments of the General Assembly have removed barriers to adoption by same-sex couples, and have extended legal rights and protections to such couples who dissolve their "domestic relationship." See 15A V.S.A. § 1-102, 1-112.
Thus, viewed in the light of history, logic, and experience, we conclude that none of the interests asserted by the State provides a reasonable and just basis for the continued exclusion of same-sex couples from the benefits incident to a civil marriage license under Vermont law. Accordingly, in the faith that a case beyond the imagining of the framers of our Constitution may, nevertheless, be safely anchored in the values that infused it, we find a constitutional obligation to extend to plaintiffs the common benefit, protection, and security that Vermont law provides opposite-sex married couples. It remains only to determine the appropriate means and scope of relief compelled by this constitutional mandate.
It is important to state clearly the parameters of today's ruling. Although plaintiffs sought injunctive and declaratory relief designed to secure a marriage license, their claims and arguments here have focused primarily upon the consequences of official exclusion from the statutory benefits, protections, and security incident to marriage under Vermont law. While some future case may attempt to establish that — notwithstanding equal benefits and protections under Vermont law — the denial of a marriage license operates per se to deny constitutionally-protected rights, that is not the claim we address today.
We hold only that plaintiffs are entitled under Chapter I, Article 7, of the Vermont Constitution to obtain the same benefits and protections afforded by Vermont law to married opposite-sex couples. We do not purport to infringe upon the prerogatives of the Legislature to craft an appropriate means of addressing this constitutional mandate, other than to note that the record here refers to a number of potentially constitutional statutory schemes from other jurisdictions. These include what are typically referred to as "domestic partnership" or "registered partnership" acts, which generally establish an alternative legal status to marriage for same-sex couples, impose similar formal requirements and limitations, create a parallel licensing or registration scheme, and extend all or most of the same rights and obligations provided by the law to married partners. See Report, Hawaii Commission on Sexual Orientation and the Law (Appendix D-1B) (1995) (recommending enactment of "Universal Comprehensive Domestic Partnership Act" to establish equivalent licensing and eligibility scheme and confer upon domestic partners "the same rights and obligations under the law that are conferred on spouses in a marriage relationship") (emphasis added); C. Christensen, If Not Marriage? On Securing Gay and Lesbian Family Values by a "Simulacrum of Marriage", 66 Fordham L. Rev. 1699, 1734-45 (1998) (discussing various domestic and foreign domestic partnership acts); A. Friedman, Same-Sex Marriage and the Right to Privacy: Abandoning Scriptural, Canonical, and Natural Law Based Definitions of Marriage, 35 How. L. J. 173, 217-220 n. 237 (reprinting Denmark's "Registered Partnership Act"); see generally, Note, A More Perfect Union: A Legal and Social Analysis of Domestic Partnership Ordinances, 92 Colum. L. Rev. 1164 (1992) (discussing local domestic partnership laws); M. Pedersen, Denmark: Homosexual Marriage and New Rules Regarding Separation and Divorce, 30 J. Fam. L. 289 (1992) (discussing amendments to Denmark's Registered Partnership Act); M. Roth, The Norwegian Act on Registered Partnership for Homosexual Couples, 35 J. Fam. L. 467 (1997) (discussing Norway's Act on Registered Partnership for Homosexual Couples). We do not intend specifically to endorse any one or all of the referenced acts, particularly in view of the significant benefits omitted from several of the laws.
Further, while the State's prediction of "destabilization" cannot be a ground for denying relief, it is not altogether irrelevant. A sudden change in the marriage laws or the statutory benefits traditionally incidental to marriage may have disruptive and unforeseen consequences. Absent legislative guidelines defining the status and rights of same-sex couples, consistent with constitutional requirements, uncertainty and confusion could result. Therefore, we hold that the current statutory scheme shall remain in effect for a reasonable period of time to enable the Legislature to consider and enact implementing legislation in an orderly and expeditious fashion. See Linkletter v. Walker, 381 U.S. 618, 628 (1965) (no constitutional rule impedes court's discretion to postpone operative date of ruling where exigencies require); Smith v. State, 473 P.2d 937, 950 (Idaho 1970) (staying operative effect of decision abrogating rule of sovereign immunity until adjournment of next legislative session); Spanel v. Mounds View School Dist. No. 621, 118 N.W.2d 795, 803-04 (Minn. 1962) (same). In the event that the benefits and protections in question are not statutorily granted, plaintiffs may petition this Court to order the remedy they originally sought.
Contrary to the characterization in the concurring and dissenting opinion, we do not "decline to provide plaintiffs with a marriage license" because of uncertainty and confusion that change may bring. Post, at 11. Rather, it is to avoid the uncertainty that might result during the period when the Legislature is considering potential constitutional remedies that we consider it prudent to suspend the Court's judgment for a reasonable period.
Our colleague asserts that granting the relief requested by plaintiffs — an injunction prohibiting defendants from withholding a marriage license — is our "constitutional duty." Post, at 3. (Johnson, J., concurring in part and dissenting in part). We believe the argument is predicated upon a fundamental misinterpretation of our opinion. It appears to assume that we hold plaintiffs are entitled to a marriage license. We do not. We hold that the State is constitutionally required to extend to same-sex couples the common benefits and protections that flow from marriage under Vermont law. That the State could do so through a marriage license is obvious. But it is not required to do so, and the mandate proposed by our colleague is inconsistent with the Court's holding.
The dissenting and concurring opinion also invokes the United States Supreme Court's desegregation decision in Watson v. City of Memphis, 373 U.S. 526 (1963), suggesting that the circumstances here are comparable, and demand a comparable judicial response. The analogy is flawed. We do not confront in this case the evil that was institutionalized racism, an evil that was widely recognized well before the Court's decision in Watson and its more famous predecessor, Brown v. Board of Education, 347 U.S. 483 (1954). Plaintiffs have not demonstrated that the exclusion of same-sex couples from the definition of marriage was intended to discriminate against women or lesbians and gay men, as racial segregation was designed to maintain the pernicious doctrine of white supremacy. See Loving, 388 U.S. at 11 (holding anti-miscegenation statutes violated Equal Protection Clause as invidious effort to maintain white supremacy). The concurring and dissenting opinion also overlooks the fact that the Supreme Court's urgency in Watson was impelled by the City's eight year delay in implementing its decision extending Brown to public recreational facilities, and "the significant fact that the governing constitutional principles no longer bear the imprint of newly enunciated doctrine." See Watson, 373 U.S. at 529; Dawson v. Mayor and City Council of Baltimore, 220 F.2d 386, aff'd, 350 U.S. 877 (1955). Unlike Watson, our decision declares decidedly new doctrine.
The concurring and dissenting opinion further claims that our mandate represents an "abdicat[ion]" of the constitutional duty to decide, and an inexplicable failure to implement "the most straightforward and effective remedy." Post, at 3, 10. Our colleague greatly underestimates what we decide today and greatly overestimates the simplicity and effectiveness of her proposed mandate. First, our opinion provides greater recognition of — and protection for — same sex relationships than has been recognized by any court of final jurisdiction in this country with the instructive exception of the Hawaii Supreme Court in Baehr, 825 P.2d 44. See Hawaii Const., art. I, § 23 (state constitutional amendment overturned same-sex marriage decision in Baehr by returning power to Legislature "to reserve marriage to opposite-sex couples"). Second, the dissent's suggestion that her mandate would avoid the "political caldron" (post, at 4) of public debate is — even allowing for the welcome lack of political sophistication of the judiciary — significantly insulated from reality. See Hawaii Const., art. I, § 23; see also Alaska Const., art. I, § 25 (state constitutional amendment reversed trial court decision in favor of same-sex marriage, Brause v. Bureau of Vital Statistics, No. 3AN-95-6562 CI, 1998 WL 88743 (Alaska Super. Ct. Feb. 27, 1998), by providing that "a marriage may exist only between one man and one woman").
The concurring and dissenting opinion confuses decisiveness with wisdom and judicial authority with finality. Our mandate is predicated upon a fundamental respect for the ultimate source of constitutional authority, not a fear of decisiveness. No court was ever more decisive than the United States Supreme Court in Dred Scott, 60 U.S. (19 How.) 393 (1857). Nor more wrong. Ironically it was a Vermonter, Stephen Douglas, who in defending the decision said — as the dissent in essence does here — "I never heard before of an appeal being taken from the Supreme Court." See A. Bickel, The Morality of Consent 101 (1975). But it was a profound understanding of the law and the "unruliness of the human condition," id. at 11, that prompted Abraham Lincoln to respond that the Court does not issue Holy Writ. See id. at 101. Our colleague may be correct that a mandate intended to provide the Legislature with the opportunity to implement the holding of this Court in an orderly and expeditious fashion will have precisely the opposite effect. Yet it cannot be doubted that judicial authority is not ultimate authority. It is certainly not the only repository of wisdom.
When a democracy is in moral flux, courts may not have the best or the final answers. Judicial answers may be wrong. They may be counterproductive even if they are right. Courts do best by proceeding in a way that is catalytic rather than preclusive and that is closely attuned to the fact that courts are participants in the system of democratic deliberation.
C. Sunstein, Foreword: Leaving Things Undecided, 110 Harv. L. Rev. 6, 101 (1996).
The implementation by the Vermont Legislature of a constitutional right expounded by this Court pursuant to the Constitution for the common benefit and protection of the Vermont community is not an abdication of judicial duty, it is the fulfillment of constitutional responsibility.
While many have noted the symbolic or spiritual significance of the marital relation, it is plaintiffs' claim to the secular benefits and protections of a singularly human relationship that, in our view, characterizes this case. The State's interest in extending official recognition and legal protection to the professed commitment of two individuals to a lasting relationship of mutual affection is predicated on the belief that legal support of a couple's commitment provides stability for the individuals, their family, and the broader community. Although plaintiffs' interest in seeking state recognition and protection of their mutual commitment may — in view of divorce statistics — represent "the triumph of hope over experience," the essential aspect of their claim is simply and fundamentally for inclusion in the family of State-sanctioned human relations.
J. Boswell, Life of Johnson (1791) (reprinted in Bartlett's Familiar Quotations 54 (15th ed. 1980).
The past provides many instances where the law refused to see a human being when it should have. See, e.g., Dred Scott, 60 U.S. at 407 (concluding that African slaves and their descendants had "no rights which the white man was bound to respect"). The future may provide instances where the law will be asked to see a human when it should not. See, e.g., G. Smith, Judicial Decision making in the Age of Biotechnology, 13 Notre Dame J. Ethics Pub. Policy 93, 114 (1999) (noting concerns that genetically engineering humans may threaten very nature of human individuality and identity). The challenge for future generations will be to define what is most essentially human. The extension of the Common Benefits Clause to acknowledge plaintiffs as Vermonters who seek nothing more, nor less, than legal protection and security for their avowed commitment to an intimate and lasting human relationship is simply, when all is said and done, a recognition of our common humanity.
The judgment of the superior court upholding the constitutionality of the Vermont marriage statutes under Chapter I, Article 7 of the Vermont Constitution is reversed. The effect of the Court's decision is suspended, and jurisdiction is retained in this Court, to permit the Legislature to consider and enact legislation consistent with the constitutional mandate described herein.
I concur in Part I of the majority opinion, the holding of Part II, and the mandate. I do not, however, concur in the reasoning of Part II. I recognize that to most observers the significance of this decision lies in its result and remedy. In the cases that come before us in the future, however, the significance of this case will lie in its rationale — that is, how we interpret and apply Chapter I, Article 7 of the Vermont Constitution. Moreover, in this, the most closely-watched opinion in this Court's history, its acceptability will be based on whether its reasoning and result are clearly commanded by the Constitution and our precedents, and whether it is a careful and necessary exercise of the Court's limited powers. I do not believe that the majority's rationale meets this exacting standard, and I fear how it may be applied — or ignored — in the future.
This is a concurrence and not a dissent. I agree with the majority that the consequence of limiting marriage to a man and woman is the exclusion of these plaintiffs, and many persons similarly situated, from numerous rights, benefits, and duties that government and society provide to — and impose on — married persons. However we might have described marriage in relation to the very limited government that was created by our Constitution, the complexity of the current system of government-created benefits and burdens has made civil marriage a modern-day emolument, a government recognized and supported special status for which these plaintiffs are not eligible.
This is a civil rights case, very different from a claim of discrimination with respect to, for example, a peddler's fee, see State v. Hoyt, 71 Vt. 59, 42 A. 973 (1899), operation of partnerships, see State v. Cadigan, 73 Vt. 245, 50 A. 1079 (1901), or regulation of river pollution, see State v. Haskell, 84 Vt. 429, 75 A. 852 (1911). It is also very different from a claim that exemptions to a Sunday closing law unconstitutionally discriminated against large stores, the issue in State v. Ludlow Supermarkets, Inc., 141 Vt. 261, 448 A.2d 791 (1982). The United States Supreme Court has recognized that discrimination based on race, alienage, national origin, or sex requires greater justification than economic discrimination, such as discrimination in the fees charged certain peddlers based on the type of goods they are selling. See Cleburne v. Cleburne Living Ctr., 473 U.S. 432, 440-41 (1985) (discussing the standards for scrutinizing various classifications). Compare United States v. Virginia, 518 U.S. 515, 532 (1996) (sex), and Loving v. Virginia, 388 U.S. 1, 11 (1967) (race), with Williamson v. Lee Optical, Inc., 348 U.S. 483, 486-88 (1955) (economic regulation). Until this decision, we also recognized this distinction. As we stated in Brigham v. State, 166 Vt. 246, 265, 692 A.2d 384, 396 (1997): "Where a statutory scheme affects fundamental constitutional rights or involves suspect classifications, both federal and state decisions have recognized that proper equal protection analysis necessitates a more searching scrutiny."
The marriage statutes do not facially discriminate on the basis of sexual orientation. There is, however, no doubt that the requirement that civil marriage be a union of one man and one woman has the effect of discriminating against lesbian and gay couples, like the plaintiffs in this case, who are unable to marry the life partners of their choice. The majority proclaims that most decisions have concluded that lesbians and gay men are not a suspect classification, inferring that any conclusion to the contrary is wrong. See ante, at 24 n. 10. On this point, however, I believe the central analysis of Ludlow is critical:
[A] state court reviewing state legislation is in a very different posture the United States Supreme Court when it undertakes the parallel task. Rather than disposing of a case on the premise that its impact will presumably affect more than fifty varying jurisdictions, a state court reaches its result in the legal climate of the single jurisdiction with which it is associated, if federal proscriptions are not transgressed.
141 Vt. at 268, 448 A.2d at 795. Although our precedents mandate use of at least a close cousin of the federal equal protection test, we must, as we said in Ludlow, apply that test in our own "legal climate."
Vermont's legal climate differs considerably from that in other jurisdictions where courts have held that lesbians and gay men are not a suspect classification. Indeed, the federal analysis of the rights of lesbians and gay men almost always starts with Bowers v. Hardwick, 478 U.S. 186 (1986), a decision that reflects a legal climate quite hostile to those rights. Bowers upheld a Georgia conviction for sodomy based on a sex act committed by two males in the bedroom of defendant's home. See id. at 196. It held that, for due process purposes, individuals do not have "a fundamental right to engage in homosexual sodomy." Id. at 191.
Federal courts considering equal-protection challenges have relied on Bowers to conclude that lesbians and gay men are not a suspect classification. They rationalize that if homosexual conduct can constitutionally be criminalized, homosexuals cannot constitute a suspect class. See, e.g., Equality Found. of Greater Cincinnati, Inc. v. City of Cincinnati, 128 F.3d 289, 292-93 (6th Cir. 1997) (holding that under Bowers and its progeny, homosexuals do not constitute suspect class because conduct which defined them as homosexuals could constitutionally be proscribed); Ben-Shalom v. Marsh, 881 F.2d 454, 464-65 (7th Cir. 1989) (citing Bowers and holding that because homosexual conduct may constitutionally be criminalized, homosexuals do not constitute a suspect class); High Tech Gays v. Defense Indus. Sec. Clearance Office, 895 F.2d 563, 571 (9th Cir. 1990) (same);Woodward v. United States, 871 F.2d 1068, 1074-76 (Fed. Cir. 1989) (same); Padula v. Webster, 822 F.2d 97, 102-03 (D.C. Cir. 1987) (same); see also Opinion of the Justices, 530 A.2d 21, 24 (N.H. 1987) (stating that for federal equal-protection analysis homosexuals do not constitute a suspect class, nor is there a fundamental right to engage in sodomy according to Bowers). The majority errs in relying on these cases because the Bowers rationale applied in all of them is not applicable in Vermont today. Although Vermont, like all states, once criminalized sodomy, and had a "fellation" law, see State v. LaForrest, 71 Vt. 311, 312, 45 A.2d 225, 226 (1899) (holding sodomy a crime by virtue of 1 V.S.A. § 271 — formerly V.S. 898 — and adopting common law so far as applicable in Vermont); 13 V.S.A. § 2603 (repealed 1977, No. 51, § 2), it repealed this law in 1977 and does not now prohibit, or otherwise restrict, homosexual conduct between adults, except on the same terms that it restricts heterosexual conduct. See, e.g., 13 V.S.A. § 3252 (sexual assault); 13 V.S.A. § 3253 (aggravated assault); 13 V.S.A. § 2601 (lewd and lascivious conduct).
Since 1992, it has generally been the policy of Vermont to prohibit discrimination based on sexual orientation. See 1991, No. 135 (Adj. Sess.). This includes discrimination based on "male or female homosexuality." 1 V.S.A. § 143. Thus, I believe our "legal climate" is vastly different from that in Bowers, where, after considering that twenty-four states had criminalized sodomy between consenting adults, the United States Supreme Court concluded that there was no fundamental right, deeply rooted in the Nation's history, to engage in such conduct. My point here is simply that the rationale in federal decisions for withholding a more searching scrutiny does not apply in Vermont. The majority errs in relying on these decisions and the state court decisions applying the same federal analysis.
Chapter I, Article 7 of the Vermont Constitution actually contains three clauses, the most important of which is the second, which contains the prohibition on governmental actions "for the particular emolument or advantage of any single person, family, or set of persons, who are a part only of that community." This anti-privilege language, and variations on it, is contained in the vast majority of pre-civil war state constitutions. See, e.g., Conn. Const. of 1818, art. I, § 1; Ky. Const. of 1792, art. XII, § 1; Mass. Const., art. VI (adopted in 1780); N.H. Const., art. X (adopted in 1784); N.C. Const. of 1776, art. III; Ohio Const. of 1851, art. I, § 2; Va. Const. of 1776, Bill of Rights, § 4; Tx. Const. of 1845, art. I, § 2. At least in this century, the jurisprudence in Vermont is similar to that in most states. See, e.g., Town of Emerald Isle v. State, 360 S.E.2d 756, 764 (N.C. 1987) (classification is not exclusive emolument if intended to promote general welfare and reasonable basis exists to conclude it serves public interest); Primes v. Tyler, 331 N.E.2d 723, 728-29 (Ohio 1975) (statute violates constitution because no governmental interest justifies grant of special privilege and immunity); Rosenblum v. Griffin, 197 A. 701, 706 (N.H. 1938) (classification is constitutional under New Hampshire or federal law if based on some reasonable ground); City of Corbin v. Louisville Nashville R.R. Co., 26 S.W.2d 539, 540 (Ky. 1930) (purpose of emoluments and privileges clause is to place all similarly situated citizens on plane of equality under law).
Oregon, like Vermont, has developed an independent state constitutional jurisprudence. Article I, Section 20 of the Oregon Constitution, adopted in 1859, provides that no law shall "grant to any citizen or class of citizens privileges, or immunities, which, upon the same terms, shall not equally belong to all citizens." This provision is similar in purpose and effect to our Common Benefits Clause. See D. Schuman, The Right to "Equal Privileges and Immunities": A State Version of "Equal Protection," 13 Vt. L. Rev. 221, 222-25 (1988). The Oregon Supreme Court has described that provision precisely how we today have described Chapter I, Article 7: "Antedating the Civil War and the equal protection clause of the fourteenth amendment, its language reflects early egalitarian objections to favoritism and special privileges for a few rather than the concern of the Reconstruction Congress about discrimination against disfavored individuals or groups." State v. Clark, 630 P.2d 810, 814 (Or. 1981). Just as this Court has acknowledged in developing its Article 7 jurisprudence, the Oregon court has recognized that a privilege for a person or group of persons means discrimination against others. See id. at 814 (Article I, Section 20 of Oregon Constitution protects against adverse discrimination as well as against favoritism). Thus, while developing an independent state constitutional jurisprudence, the Oregon Supreme Court has looked to the decisions of United States Supreme Court, but has adopted the federal analysis only where the court finds it persuasive. See State v. Kennedy, 666 P.2d 1316, 1321 (Or. 1983). See, e.g., Hewitt v. State Accident Ins. Fund Corp., 653 P.2d 970, 976 (Or. 1982) (declining to adopt federal standard of intermediate scrutiny for sex-based classifications).
The Oregon Supreme Court, like this Court, has adopted the federal, tiered framework for analyzing equal-protection type constitutional challenges. See Hewitt, 653 P.2d at 976 (following United States Supreme Court analysis that asks whether classification is made on basis of suspect classification, and if so, whether such classification is subject to strict scrutiny). Moreover, it has held, as we have held, that its state constitution "prohibits disparate treatment of groups or individuals by virtue of `invidious' social categories" and that discrimination against a suspect class is subject to strict scrutiny. Id.; see MacCallum v. Seymour's Adm'r, 165 Vt. 452, 460, 686 A.2d 935, 939 (1996) (Article 7 protects against invidious discrimination). I point out the similarities between our Article 7 jurisprudence and Oregon's Section 20 jurisprudence because this Court has not established the criteria for identifying suspect classifications, while the Oregon courts have. Because of the historical similarity, I find it useful to look to Oregon case law, and the United States Supreme Court decisions upon which it relies, in considering whether lesbians and gay men are a suspect classification under Article 7. In Hewitt, the Oregon Supreme Court determined that sex-based classifications are suspect because (1) they focus on an immutable personal characteristic and thus "can be suspected of reflecting `invidious' social or political premises, that is to say, prejudice or stereotyped prejudgments," and (2) "[t]he purposeful historical, legal, economic and political unequal treatment of women is well known." 653 P.2d at 977. Accordingly, the court held that sex-based classifications are inherently suspect, like the United States Supreme Court found classifications based on race, alienage, and nationality. See id. at 977-78 (citing Loving v. Virginia, 388 U.S. 1, 11 (1967) (race); Graham v. Richardson, 403 U.S. 365, 372 (1971) (alienage); Oyama v. California, 332 U.S. 633, 646 (1948) (nationality)).
Although the Oregon Supreme Court has not addressed whether lesbians and gay men are a suspect classification, the Oregon Court of Appeals has recently done so. See Tanner v. Oregon Health Sciences Univ., 971 P.2d 435 (Or.Ct.App. 1998). In Tanner, the court held that Article I, Section 20 of the Oregon Constitution requires the Oregon Health Sciences University to extend health and life insurance benefits to the unmarried domestic partners of its homosexual employees. See id. at 448. The Tanner court examined the Hewitt two-part test for defining suspect classes and determined that "immutability — in the sense of inability to alter or change — is not necessary" because alienage and religious affiliation — which may be changed — have been held to be suspect classifications. Thus, it held that defining a suspect class depends not on the immutability of a class-defining characteristic, but upon (1) whether the characteristic has historically been regarded as defining a distinct socially-recognized group, and if so (2) whether that group has been the subject of adverse social or political stereotyping. See id. at 446. Applying this test, the court concluded that the class of homosexual couples is clearly defined in terms of stereotyped personal and social characteristics; is widely regarded as a distinct, socially recognized group; and indisputably has "been and continues to be the subject of adverse social and political stereotyping and prejudice." Id. at 447. Thus, the court found that the plaintiffs, three lesbian couples, were members of a suspect class.
In this concurrence, I do not detail a suspect-classification analysis, but I can summarize my opinion by saying that I agree with the general framework adopted by the Oregon courts in Hewitt and Tanner. These decisions concerning Article I, Section 20 of that state's constitution are entirely consistent with the law we have developed under Chapter I, Article 7 of the Vermont Constitution, at least prior to this decision. I find Hewitt and Tanner far more persuasive than the majority's decision, which backtracks from the established legal framework under Article 7 and fails to provide any guidelines whatsoever for the Legislature, the trial courts, or Vermonters in general to predict the outcome of future cases.
I agree with the majority that the State cannot justify the denial of legal benefits and responsibilities of civil marriage to gay and lesbian couples. And I agree that the appropriate remedy is either to require the State to extend the option of receiving these benefits and associated responsibilities to these couples, or to require that it offer the opportunity for civil marriage on equal terms. I will briefly explain my disagreement with the majority's rationale for reaching the same result.
The majority's analysis under Chapter I, Article 7 proceeds in three steps: (1) there is one equality standard imposed by Article 7, and it applies to claims of civil rights discrimination and economic discrimination alike; (2) the equality standard is higher, that is, more active, than the standard imposed by the Equal Protection Clause of the Fourteenth Amendment for analyzing claims of economic discrimination; and (3) under the new standard, the denial of the benefits of marriage to lesbians and gay men violates Chapter I, Article 7. In the first two steps, the majority makes statements entirely contrary to our existing Article 7 jurisprudence. As to the third step, I find no standard in the Court's decision — it is entirely a matter of "judgment."
The first step in the Court's analysis requires overruling a long series of precedents holding that where a statutory scheme affects fundamental constitutional rights or involves suspect classifications, Article 7 requires "a more searching scrutiny." Brigham, 166 Vt. at 265, 692 A.2d at 396. Among the decisions that have stated this standard are L'Esperance v. Town of Charlotte, 167 Vt. 162, 165, 704 A.2d 760, 762 (1997); McCallum, 165 Vt. at 457, 686 A.2d at 936-37; Benning v. State, 161 Vt. 472, 486, 641 A.2d 757, 764 (1994); In re Sherman Hollow, Inc., 160 Vt. 627, 628, 641 A.2d 753, 755 (1993) (mem.); Oxx v. Department of Taxes, 159 Vt. 371, 376, 618 A.2d 1321, 1324 (1992); Hodgeman v. Jard Co., 157 Vt. 461, 464, 599 A.2d 1371, 1373 (1991); State v. George, 157 Vt. 580, 588, 602 A.2d 953, 957 (1991); Town of Sandgate v. Colehamer, 156 Vt. 77, 88, 589 A.2d 1205, 1211 (1990); and Choquette v. Perrault, 153 Vt. 45, 51-52, 569 A.2d 455, 459 (1989). The majority barely acknowledges the multi-tiered standard stated in those cases, and dismisses it as a "rigid" analysis. See ante, at 23. It is ironic that in a civil rights case we overrule our precedent requiring the State to meet a higher burden in civil rights cases, but still conclude, under the lower standard, that the State has not met its burden.
The majority's characterization of Brigham is neither fair nor accurate. The majority states that Brigham "acknowledged the federal standard," but "eschewed the federal categories of analysis." Ante, at 14. Far beyond "acknowledging" the federal standards, Brigham held explicitly that they applied under Article 7 — a holding now implicitly overruled by the majority decision. Rather than eschewing the federal standards, we held that the educational financing system advanced no "legitimate governmental purpose" under any standard. See Brigham, 166 Vt. at 265, 692 A.2d at 396.
The majority's statement that suspect class analysis is "often effectively ignored in our more recent decisions" is inaccurate, unless our statements that we need not reach the issue in a case somehow "ignores" suspect-class analysis. Ante, at 15. See, e.g., MacCallum, 165 Vt. at 457 n. 1, 686 A.2d at 938 n. 1 (in view of our disposition, we need not reach plaintiff's claim that adopted persons are suspect class).
The effect of the majority decision is that the State now bears no higher burden to justify discrimination against African-Americans or women than it does to justify discrimination against large retail stores as in Ludlow. I doubt that the framers of our Constitution, concerned with preventing the equivalent of British royalty, would believe that the inevitable line-drawing that must occur in economic regulation should be equated with the denial of civil and human rights. I do not believe that the new standard is required by, or even consistent with, the history on which the majority bases it.
The second step is also at variance with our Article 7 law, even as it seeks to rely upon it. The majority holds that Article 7 requires a more active standard of constitutional review than the Fourteenth Amendment, as interpreted by the United States Supreme Court, in the absence of a fundamental right or suspect classification. See ante, at 11-12. This means that in the future this Court is less likely to defer to the Legislature and more likely to find its acts unconstitutional than would the United States Supreme Court. Again, I find great irony in the fact that we are doing this unnecessarily in a case where the main theme of the State and many amici is that we must defer to the Legislature on the issue before us.
I agree that Ludlow, Choquette, and MacCallum contain important holdings about how equality challenges are addressed by a state court. Ludlow holds that we must look at justifications for distinctions that are realistic in view of Vermont's unique legal culture. See Ludlow, 141 Vt. at 268, 448 A.2d at 795. Choquette and MacCallum hold that such justifications must be relevant to contemporary circumstances and not be wholly archaic. See Choquette, 153 Vt. at 53-54, 569 A.2d at 460; MacCallum, 165 Vt. at 461, 686 A.2d at 940. None of these decisions demonstrate that "Vermont decisions reflect a very different approach from current federal jurisprudence," which is how the majority characterizes them. Ante, at 10. Indeed, we have said over and over that the test, where no fundamental right or suspect class is involved, "is the same under the Equal Protection Clause of the Fourteenth Amendment to the United States Constitution" as under Article 7. Lorrain v. Ryan, 160 Vt. 202, 212, 628 A.2d 543, 550 (1993); see Brigham, 166 Vt. at 265, 692 A.2d at 395; L'Esperance, 167 Vt. at 165, 704 A.2d at 762. Although the majority seeks to rely on isolated statements from Ludlow, in fact, we are by this decision creating a new, more active standard of review in Article 7 challenges. We have wisely, in the past, avoided the path the majority now chooses, a path worn and abandoned in many other states. When Justice Hayes decried the failure of litigants to raise state constitutional issues, see State v. Jewett, 146 Vt. 221, 229, 500 A.2d 233, 238 (1985), he could not have been referring to challenges under state anti-emolument and equality provisions. In state after state, throughout the nineteenth and early twentieth centuries, state supreme courts routinely struck down economic and social welfare statutes under these provisions using an analysis similar to that employed by the majority in this case. See H. Gillman, The Constitution Besieged: The Rise and Demise of Lochner Era Police Powers Jurisprudence 9 (1993). For example, in Auditor of Lucas County v. State, 78 N.E. 955, 957 (Ohio 1906), the Ohio Supreme Court struck down an Ohio law that provided a stipend of $25 each quarter to adult blind persons because it was over-inclusive — including rich and poor — and under-inclusive — including only some disabled persons. See also Cincinnati v. Cook, 140 N.E. 655, 656 (Ohio 1923) (striking down ordinance that allowed parking in front of train station only with consent of supervisor of station, in part because it created "privilege or immunity" in those who were allowed to park); Low v. Rees Printing Co., 59 N.W. 362, 368 (Neb. 1894) (striking down eight-hour-day law because it exempted farm or domestic labor); State v. Pennoyer, 18 A. 878, 881 (N.H. 1889) (striking down statute requiring licensing of all physicians, except those who resided in only one town between 1875 and 1879, because it imposed unequal burden on members of same class); Millett v. People, 7 N.E. 631, 636 (Ill. 1886) (striking down statute requiring mine operators who tied wages to amount of coal extracted to keep scale at mine so coal could be weighed before managers had chance to separate unusable material); In re Jacobs, 98 N.Y. 98, 112-14 (N.Y. 1885) (striking down act addressing deplorable working conditions under which cigar makers labored in tenements by banning the manufacturing of cigars in those dwellings); Ex parte Westerfield, 55 Cal. 550, 551 (Cal.Sup.Ct. 1880) (striking down law making it misdemeanor for bakers to force employees to work between six o'clock Saturday evening and six o'clock Sunday evening).
My concern about the effect of this decision as a precedent is heightened by the majority's treatment of the Ludlow decision. It is fair to say that for some purposes, there have been two versions of the Ludlow decision. First, there is the one we have described in dicta, usually as a historical event. See State v. Brunelle, 148 Vt. 347, 351, 534 A.2d 198, 201-02 (1987); Hodgeman, 157 Vt. at 464, 599 A.2d at 1373. This one holds that Article 7 is "more stringent than the federal constitutional standard which requires only a rational justification." Brunelle, 148 Vt. at 351, 534 A.2d at 201-02. Second, there is the Ludlow decision that we have actually used in deciding cases. See, e.g., Choquette, 153 Vt. at 52, 569 A.2d at 459; In re Property of One Church Street, 152 Vt. 260, 263-65, 565 A.2d 1349, 1350-51 (1989). This version of Ludlow holds that the Article 7 standard is the reasonable-relationship test applicable under the Fourteenth Amendment to the United States Constitution. See Choquette, 153 Vt. at 52, 569 A.2d at 459; see also Lorrain, 160 Vt. at 212, 628 A.2d at 550 (test under Article 7 is same as that under federal Equal Protection Clause).
Obviously, these versions of Ludlow are irreconcilable, and only one can be accurate. In case after case, advocates pursuing Article 7 challenges have tried, and failed, to get us to adopt the first version of Ludlow as the basis for a favorable decision. The first version has appeared only in dicta in two isolated cases. Today, seventeen years after the Ludlow decision, the advocates have finally succeeded, with a begrudging acknowledgment from the majority that our decisions "have consistently recited" the federal test and are now wholesale overruled.
In view of this history of treatment of Ludlow, I find incredible the majority's statement that "Vermont case law has consistently demanded in practice that statutory exclusions from publicly-conferred benefits and protections must be `premised on an appropriate and overriding public interest,'" ante, at 15, quoting Ludlow as if all of our decisions after Ludlow disingenuously mouthed one deferential constitutional standard but silently employed a more activist standard. If one general statement could be made, it would be that we have never actually employed the standard quoted by the majority in any case, until this one.
My fear is that once we get beyond this controversial decision, we will end up with two versions of it. Will we go back to minimalist review when we get a claim of discrimination, for example, between large stores and small ones, or will the more activist review promised by this decision prevail? Our history in applying Ludlow says that we will do the former, which I find to be the more desirable, but a serious blow will have been dealt to our ability to develop neutral constitutional doctrine.
Most of these decisions reflect judicial attitudes prevalent in the era of Lochner v. New York, 198 U.S. 45 (1905), when the United States Supreme Court was routinely striking down economic and social welfare legislation. As the United States Supreme Court modified its jurisprudence to give primacy to the federal and state legislative role in economic and social welfare legislation, state courts did likewise, often on the basis that Fourteenth Amendment jurisprudence was equally applicable under state due process and equality provisions. See Gillman, supra, at 62. See, e.g., Department of Mental Hygiene v. Kirchner, 400 P.2d 321, 322 (Cal. 1965) ( Fourteenth Amendment to federal constitution and Sections 11 and 21 of Article I of California Constitution provide generally equivalent but independent protections in their respective jurisdictions); People v. Willi, 179 N.Y.S. 542, 547 (Del. Cty. Ct. 1919) (methods of analysis under Fourteenth Amendment and state constitution are identical); City of Chicago v. Rhine, 2 N.E.2d 905, 908 (Ill. 1936) (simultaneously analyzing federal and state equal protection claims); Ex Parte Caldwell, 118 N.W. 133, 134 (Neb. 1908) (upholding under state and federal constitutions statute prohibiting common labor on Sunday).
The Vermont Supreme Court never adopted an activist stance in reviewing economic and social welfare legislation, and history shows we chose the right course. We could have relied upon the looser and more activist language that prevailed in the federal cases in the early twentieth century — the same language that the majority relies upon today, ante, at 12 n. 4 — to substitute our judgment for the Legislature, but wisely we did not. Unfortunately, we have now resurrected that approach. I can find no justification for the holding that Article 7 requires a more activist approach than the Fourteenth Amendment for reviewing social welfare and economic legislation. We were right in Lorrain, Brigham, and L'Esperance on this point and should adhere to those precedents. Finally, concerning the third step of the majority's analysis, I question whether the majority's new standard is ascertainable, is consistent with our limited role in constitutional review, and contains appropriate judicial discretion. As Justice Johnson explains in her dissent, see post, at 21 n. 13, the strength of the federal approach is that it disciplines judicial discretion and promotes predictability. See C. Sunstein, Foreword: Leaving Things Undecided, 110 Harv. L. Rev. 4, 78 (1996). Indeed, the Oregon courts have followed the federal approach in this area to avoid a balancing process "of pragmatic considerations about which reasonable people may differ over time," Kennedy, 666 P.2d at 1321, and "policy choices disguised as ad hoc evaluations based on comparison of incommensurable," Schuman, supra, at 227. The majority calls the federal approach "rigid" at one point, ante, at 23, but then describes it, as applied in Tanner, as an invitation to subjective judicial decision-making. Ante, at 24 n. 10. The two criticisms are as inconsistent as any criticisms could be. I accept the former — rigid — as accurate, at least in comparison with the wide judicial discretion the majority claims here as an alternative. The latter — subjective judicial decision-making — is, however, the least accurate criticism the majority could level.
Two points about the new standard are particularly troublesome for me. The majority now requires that legislative classifications be "reasonably necessary to accomplish the State's claimed objectives." Ante, at 24. In our imperfect world, few legislative classifications are "necessary," and most legislation could be more narrowly tailored to the state's objective. I cannot square this standard with our limited role in constitutional adjudication. As I noted earlier, while language to this effect appears in Ludlow, it has never been used as the basis of one of our decisions until today.
More importantly, I cannot endorse, in this vitally important area of constitutional review, a standard that relies wholly on factors and balancing, with no mooring in any criteria or guidelines, however imperfect they may be. On this point, I agree with Justice Johnson. See post, at 21 n. 13. I accept the majority's assertion that it has attempted to avoid a standard based on "personal notions," and that all constitutional adjudication requires reasoned judgment, but I do not believe that it has succeeded in properly applying the critical considerations it has identified. Ante, at 25. Instead of mooring its analysis within the framework of fundamental rights and suspect classifications, the majority professes to make its new Article 7 standard "objective and grounded" by requiring courts, in balancing the competing interests, to "look to the history and `traditions from which [the State] developed' as well as those `from which it broke.'" Ante, at 25. It is difficult to conceive that any persons sitting on this Court, whatever their philosophical persuasions, would be insensitive to the history and traditions from which Vermont developed, and those from which it broke, but how this standard will be applied to Article 7 challenges is not at all predictable. In the end, the approach the majority has developed relies too much on the identities and personal philosophies of the men and women who fill the chairs at the Supreme Court, too little on ascertainable standards that judges of different backgrounds and philosophies can apply equally, and very little, if any, on deference to the legislative branch.
The final irony in this decision for me is that the balancing and weighing process set forth in the Court's opinion describes exactly the process we would expect legislators to go through if they were facing the question before us. We are judges, not legislators.
For the above reasons, I concur in the mandate, but respectfully disagree with Part II of the Court's decision, the majority's rationale for reaching this mandate.
Concurring and Dissenting
Forty years ago, in reversing a decision that had denied injunctive relief for the immediate desegregation of publicly owned parks and recreational facilities in Memphis, Tennessee, a unanimous United States Supreme Court stated:
The basic guarantees of our Constitution are warrants for the here and now and, unless there is an overwhelming compelling reason, they are to be promptly fulfilled.
Watson v. City of Memphis, 373 U.S. 526, 533 (1963).
Plaintiffs come before this Court claiming that the State has unconstitutionally deprived them of the benefits of marriage based solely upon a discriminatory classification that violates their civil rights. They ask the Court to remedy the unlawful discrimination by enjoining the State and its municipalities from denying them the license that serves to identify the persons entitled to those benefits. The majority agrees that the Common Benefits Clause of the Vermont Constitution entitles plaintiffs to obtain the same benefits and protections as those bestowed upon married opposite-sex couples, yet it declines to give them any relief other than an exhortation to the Legislature to deal with the problem. I concur with the majority's holding, but I respectfully dissent from its novel and truncated remedy, which in my view abdicates this Court's constitutional duty to redress violations of constitutional rights. I would grant the requested relief and enjoin defendants from denying plaintiffs a marriage license based solely on the sex of the applicants.
The majority declares that the issue before this Court does not turn on the heated moral debate over intimate same-sex relationships, and further, that this Court has a constitutional responsibility to consider the legal merits of even controversial cases. See ante, at 3. Yet, notwithstanding these pronouncements, the majority elects to send plaintiffs to an uncertain fate in the political caldron of that very same moral debate. And to what end? Passing this case on to the Legislature will not alleviate the instability and uncertainty that the majority seeks to avoid, and will unnecessarily entangle this Court in the Legislature's efforts to accommodate the majority's mandate within a "reasonable period of time." Ante, at 41.
In the 1999 legislative session, while the instant case was pending before this Court, fifty-seven representatives signed H. 479, which sought to amend the marriage statutes by providing that a man shall not marry another man, and a woman shall not marry another woman.
In 1948, when the California Supreme Court struck down a state law prohibiting the issuance of a license authorizing interracial marriages, the Court did not suspend its judgment to allow the Legislature an opportunity to enact a separate licensing scheme for interracial marriages. See Perez v. Lippold, 198 P.2d 17, 29 (Cal. 1948) (granting writ of mandamus compelling county clerk to issue certificate of registry). Indeed, such a mandate in that context would be unfathomable to us today. Here, as in Perez, we have held that the State has unconstitutionally discriminated against plaintiffs, thereby depriving them of civil rights to which they are entitled. Like the Hawaii Circuit Court in Baehr v. Mike, No. Civ. 91-1394, 1996 WL 694235, at *22 (Haw. Cir. Ct., Dec. 3, 1996), which rejected the State's reasons for excluding same-sex couples from marriage, we should simply enjoin the State from denying marriage licenses to plaintiffs based on sex or sexual orientation. That remedy would provide prompt and complete relief to plaintiffs and create reliable expectations that would stabilize the legal rights and duties of all couples.
I.My dissent from the majority's mandate is grounded on the government's limited interest in dictating public morals outside the scope of its police power, and the differing roles of the judicial and legislative branches in our tripartite system of government. I first examine the State's narrow interest in licensing marriages, then contrast that interest with the judiciary's fundamental duty to remedy civil rights violations, and lastly emphasize the majority's failure to adequately explain why it is taking the unusual step of suspending its judgment to allow the Legislature an opportunity to redress the unconstitutional discrimination that we have found.
This case concerns the secular licensing of marriage. The State's interest in licensing marriages is regulatory in nature. See Southview Coop. Housing v. Rent Control Bd., 486 N.E.2d 700, 704 (Mass. 1985) ("Licensing is simply a means of regulating."). The regulatory purpose of the licensing scheme is to create public records for the orderly allocation of benefits, imposition of obligations, and distribution of property through inheritance. Thus, a marriage license merely acts as a trigger for state-conferred benefits. See Priddy v. City of Tulsa, 882 P.2d 81, 83 (Okla.Crim.App. 1993) (license gives to licensee special privilege not accorded to others, which licensee otherwise would not enjoy). In granting a marriage license, the State is not espousing certain morals, lifestyles, or relationships, but only identifying those persons entitled to the benefits of the marital status. See People v. County of Mendocino, 683 P.2d 1150, 1155 (Cal. 1984) (licensing regulates activity based on determination of qualification of licensee).
Although the State's licensing procedures do not signal official approval or recognition of any particular lifestyles or relationships, commentators have noted that denying same-sex couples a marriage license is viewed by many as indicating that same-sex relationships are not entitled to the same status as opposite-sex relationships. See, e.g., C. Christensen, If Not Marriage? On Securing Gay and Lesbian Family Values by a "Similacrum of Marriage", 66 Fordham L. Rev. 1699, 1783-84 (1998) (most far reaching consequence of legalizing same-sex marriage would be symbolic shedding of sexual outlaw image and civil recognition of shared humanity); D. Chambers, What If? The Legal Consequences of Marriage and the Legal Needs of Lesbian and Gay Male Couples, 95 Mich. L. Rev. 447, 450 (1996) (allowing same-sex couples to marry would signify acknowledgement of same-sex couples as equal citizens). This Court has recognized that singling out a particular group for special treatment may have a stigmatizing effect more significant than any economic consequences. See MacCallum v. Seymour's Administrator, 165 Vt. 452, 460, 686 A.2d 935, 939 (1996) (noting that symbolic and psychological damage resulting from unconstitutional classification depriving adopted children of right to inherit from collateral kin may be more significant than any concern over material values). The United States Supreme Court has also recognized this phenomenon. See Romer v. Evans, 517 U.S. 620, 634 (1996) (laws singling out gays and lesbians for special treatment "raise the inevitable inference that the disadvantage imposed is born of animosity toward the class of persons affected"); Heckler v. Matthews, 465 U.S. 728, 739-40 (1984) (stigmatizing members of disfavored group as less worthy participants in community "can cause serious noneconomic injuries . . . solely because of their membership in a disfavored group"). Because enjoining defendants from denying plaintiffs a marriage license is the most effective and complete way to remedy the constitutional violation we have found, it is not necessary to reach the issue of whether depriving plaintiffs of the "status" of being able to obtain the same state-conferred marriage license provided to opposite-sex couples violates their civil rights.
Apart from establishing restrictions on age and consanguinity related to public health and safety, see 18 V.S.A. § 5142 (minors and incompetent persons); 15 V.S.A. § 1, 2 (consanguinity), the statutory scheme at issue here makes no qualitative judgment about which persons may obtain a marriage license. See Leduc v. Commonwealth, 657 N.E.2d 755, 756-57 (Mass. 1995) (historical aim of licensure is generally to preserve public health, safety and welfare). Hence, the State's interest concerning the challenged licensing statute is a narrow one, and plaintiffs have prevailed on their constitutional claim because the State has failed to raise any legitimate reasons related to public health or safety for denying marital benefits to same-sex couples. See Commonwealth v. Bonadio, 415 A.2d 47, 50 (Pa. 1980) ("With respect to regulation of morals, the police power should properly be exercised to protect each individual's right to be free from interference in defining and pursuing his own morality but not to enforce a majority morality on persons whose conduct does not harm others."). In my view, the State's interest in licensing marriages would be undisturbed by this Court enjoining defendants from denying plaintiffs a license.
While the State's interest in licensing marriages is narrow, the judiciary's obligation to remedy constitutional violations is central to our form of government. Indeed, one of the fundamental principles of our tripartite system of government is that the judiciary interprets and gives effect to the constitution in cases and controversies concerning individual rights. See Marbury v. Madison, 5 U.S. (1 Cranch) 137, 163, 177-78 (1803); see also Shields v. Gerhart, 163 Vt. 219, 223, 658 A.2d 924, 927-28 (1995) (emphasizing "the preeminence of the Vermont Constitution in our governmental scheme," which includes right of citizens under Chapter I, Article 4 to find a certain remedy promptly and without delay).
Unlike the Vermont Constitution, see Vt. Const. ch. II, § 5 ("The Legislative, Executive, and Judiciary departments, shall be separate and distinct, so that neither exercises the powers properly belonging to the others."), the United States Constitution does not contain an explicit separation-of-powers provision; however, the United States Supreme Court has derived a separation-of-powers requirement from the federal constitution's statement of the powers of each of the branches of government. See, e.g., Bowsher v. Synar, 478 U.S. 714, 721-22 (1986). Because we have relied upon federal separation-of-powers jurisprudence in interpreting Chapter II, Section 5, see Trybulski v. Bellows Falls Hydro-Elec. Corp., 112 Vt. 1, 7, 20 A.2d 117, 120 (1941), I draw upon federal case law for analysis and support in discussing separation-of-powers principles. See In re D.L., 164 Vt. 223, 228 n. 3, 669 A.2d 1172, 1176 n. 3 (1995); see also In re Constitutionality of House Bill 88, 115 Vt. 524, 529, 64 A.2d 169, 172 (1949) (noting that judicial power of Vermont Supreme Court and United States Supreme Court is same).
This power is "not merely to rule on cases, but to decide them." Plaut v. Spendthrift Farm, Inc., 514 U.S. 211, 218-19 (1995) (emphasis in original); see Records of the Council of Censors of the State of Vermont 431 (P. Gillies and D. Sanford eds., 1991) (supreme judicial tribunals are to regard constitution as fundamental law superior to legislative enactment; consequently, if enactment is repugnant to constitution, judges are bound to pronounce it inoperative and void). As this Court has stated on numerous occasions, when measures enacted pursuant to the State's police powers have no real or substantial relation to any legitimate purpose of those powers and invade individual "`rights secured by the fundamental law, it is the duty of the courts to so adjudge, and thereby give effect to the Constitution.'" State v. Morse, 84 Vt. 387, 394, 80 A. 189, 191-92 (1911) (quoting Mugler v. Kansas, 123 U.S. 623 (1887)); see Beecham v. Leahy, 130 Vt. 164, 172, 287 A.2d 836, 841 (1972) ("It is the function of the judicial branch to pass upon the appropriateness and reasonableness of the legislative exercise of police power."). This Court emphasized in Morse that "in its last analysis, the question of the validity of such measures [enacted under the police powers] is one for the court." 84 Vt. at 394, 80 A. at 191.
The power of courts to fashion remedies for constitutional violations is well established in both this Court's and the United States Supreme Court's jurisprudence concerning individual rights and equal protection. See MacCallum v. Seymour's Adm'r, 165 Vt. 452, 462, 686 A.2d 935, 941 (1996) (holding that statute denying adopted children right to inherit from collateral heirs violated Common Benefits Clause, and declaring plaintiff to be lawful heir of estate of collateral relative); Medical Ctr. Hosp. v. Lorrain, 165 Vt. 12, 14-15, 675 A.2d 1326, 1329 (1996) (determining that doctrine making husbands liable to creditors for necessary items provided to wives violated principle of equal protection when applied only to men, and choosing to abolish doctrine rather than to extend it to both men and women); see also Heckler v. Mathews, 465 U.S. 728, 740 (1984) (when right invoked is that to equal treatment, "the appropriate remedy is a mandate of equal treatment"); Davis v. Passman, 442 U.S. 228, 241-42 (1979) (within "great outlines" of Constitution, "judiciary is clearly discernible as the primary means through which rights may be enforced"; unless Constitution commits issue to coordinate branch, "we presume that justiciable constitutional rights are to be enforced through the courts"). Particularly in civil rights cases involving discrimination against a disfavored group, "courts do not need specific [legislative] authorization to employ a remedy, at law or in equity, that is tailored to correct a constitutional wrong." Aguayo v. Christopher, 865 F. Supp. 479, 487-88 (N.D.Ill. 1994) (finding unconstitutional on its face statute making citizenship available to foreign-born children of citizen fathers, but not citizen mothers, and issuing judgment declaring plaintiff to be citizen).
Accordingly, absent "compelling" reasons that dictate otherwise, it is not only the prerogative but the duty of courts to provide prompt relief for violations of individual civil rights. See Watson, 373 U.S. at 532-33 (defendants have heavy burden of showing that delay in desegregating public parks and recreational facilities is "manifestly compelled by constitutionally cognizable circumstances"). This basic principle is designed to assure that laws enacted through the will of the majority do not unconstitutionally infringe upon the rights of a disfavored minority.
There may be situations, of course, when legislative action is required before a court-ordered remedy can be fulfilled. For example, in Brigham v. State, 166 Vt. 246, 249, 269, 692 A.2d 384, 386, 398 (1997), this Court declared that Vermont's system for funding public education unconstitutionally deprived Vermont schoolchildren of a right to an equal educational opportunity, and then retained jurisdiction until the Legislature enacted legislation that satisfied the Court's holding. Plainly, it was not within the province of this Court to create a new funding system to replace the one that we had declared unconstitutional. The Legislature needed to enact legislation that addressed issues such as the level of state funding for public schools, the sources of additional revenue, and the framework for distributing state funds. See Act 60, 16 V.S.A. § 4000-4029. In finding a funding source, the Legislature had to consider whether to apply a flat or progressive tax on persons, property, entities, activities or income. These considerations, in turn, required the Legislature to consider what state programs would have to be curtailed to make up for the projected additional school funding. All of these complex political decisions entailed core legislative functions that were a necessary predicate to fulfillment of our holding. See Brigham, 166 Vt. at 249, 692 A.2d at 386 (devising system for funding public education lies within prerogative of Legislature).
A completely different situation exists here. We have held that the Vermont Constitution entitles plaintiffs "to obtain the same benefits and protections afforded by Vermont law to married opposite-sex couples." Ante, at 39. Given this holding, the most straightforward and effective remedy is simply to enjoin the State from denying plaintiffs a marriage license, which would designate them as persons entitled to those benefits and protections. No legislation is required to redress the constitutional violation that the Court has found. Cf. Watson, 373 U.S. at 532 (desegregation of recreational facilities does not present same kind of cognizable difficulties inherent in desegregating schools). Nor does our paramount interest in vindicating plaintiffs' constitutional rights interfere in any way with the State's interest in licensing marriages. Far from intruding upon the State's narrow interest in its licensing statute, allowing plaintiffs to obtain a license would further the overall goals of marriage, as defined by the majority — to provide stability to individuals, their families, and the broader community by clarifying and protecting the rights of married persons, see ante, at 35. Cf. In re B.L.V.B., 160 Vt. 368, 372, 375, 628 A.2d 1271, 1274-75 (1993) (purpose of adoption statute read in its entirety is to clarify and protect legal rights of adopted persons, not to proscribe adoptions by certain combinations of individuals; denying children of same-sex partners security of legally recognized relationship with second parent serves no legitimate state interest).
I do not misinterpret the majority's holding. See ante, at 41. I am aware that the Legislature is not obligated to give plaintiffs a marriage license, or any other remedy for that matter. It is this Court, not the Legislature, that has the duty to remedy the constitutional violation we have found. We are left to speculate why the majority is not enjoining defendants from denying plaintiffs the regulatory license that they seek and that would entitle them to the same benefits and protections to which they are entitled under the majority's holding.
The majority declines to provide plaintiffs with a marriage license, however, because a sudden change in the marriage laws "may have disruptive and unforeseen consequences," and "uncertainty and confusion could result." Ante, at 40. Thus, within a few pages of rejecting the State's doomsday speculations as a basis for upholding the unconstitutionally discriminatory classification, the majority relies upon those same speculations to deny plaintiffs the relief to which they are entitled as the result of the discrimination. See ante, at 37, 39.
During the civil rights movement of the 1960's, state and local governments defended segregation or gradual desegregation on the grounds that mixing the races would lead to interracial disturbances. The Supreme Court's "compelling answer" to that contention was "that constitutional rights may not be denied simply because of hostility to their assertion or exercise." See Watson, 373 U.S. at 535. Here, too, we should not relinquish our duty to redress the unconstitutional discrimination that we have found merely because of "personal speculations" or "vague disquietudes." Id. at 536. While the laudatory goals of preserving institutional credibility and public confidence in our government may require elected bodies to wait for changing attitudes concerning public morals, those same goals require courts to act independently and decisively to protect civil rights guaranteed by our Constitution. None of the cases cited by the majority support its mandate suspending the Court's judgment to allow the Legislature to provide a remedy. In Linkletter v. Walker, 381 U.S. 618, 622 (1965), the issue was whether the decision in Mapp v. Ohio, 367 U.S. 643 (1961) extending the exclusionary rule to the states through the federal due process clause applied to all state court convictions that had become final before Mapp. The Court declined to apply Mapp retroactively, stating that both defendants and the states had relied upon the decision that Mapp had overruled, that the fairness of the underlying trials had not been placed at issue, and that applying Mapp retroactively would severely tax the administration of justice in state courts. See Linkletter, 381 U.S. at 637-39. After noting that it was not concerned with "pure" prospectivity because the exclusionary rule had been applied in Mapp itself, the Court held that new rules may be applied prospectively "where the exigencies of the situation require such an application." See id. at 622, 628.
The majority states that my analogy to the circumstances in Watson is "flawed" because (1) we are not confronting the evil of institutionalized racism; and (2) our ruling today is "decidedly new doctrine." Ante, at 42. The majority's first point implies that our duty to remedy unconstitutional discrimination is somehow limited when that discrimination is based on sex or sexual orientation rather than race. I would not prioritize among types of civil rights violations; our duty to remedy them is the same, once a constitutional violation is found.
Regarding the second point, the Court in Watson enunciated "the usual principle that any deprivation of constitutional rights calls for prompt rectification," stating further that the unavoidable delay in implementing the desegregation of schools ordered in Brown v. Board of Educ., 347 U.S. 483 (1954) was "a narrowly drawn, and carefully limited, qualification upon usual precepts of constitutional adjudication and is not to be unnecessarily expanded in application." 373 U.S. at 532-33. The majority has not explained why it is diverging from that basic principle in this case. Further, as both the majority and concurrence acknowledge, see ante, at 36-38; ante, at 6 (Dooley, J., concurring), allowing same-sex couples to obtain the benefits and protections of marriage is a logical extension of Vermont's legislatively enacted public policy prohibiting discrimination on the basis of sex and sexual orientation, see 1991, No. 135 (Adj. Sess.), decriminalizing consensual homosexual conduct between adults, see 1977, No. 51, § 22, and permitting same-sex partners to adopt children, see 15A V.S.A. § 1-102(b) (codifying holding in B.L.V.B., 160 Vt. at 369, 628 A.2d at 1272, which allowed same-sex partner of natural parent to adopt parent's child without terminating parent's rights); 15A V.S.A. § 1-112 (giving family court jurisdiction to adjudicate issues pertaining to parental rights and responsibilities and child support with respect to adopted children of domestic partners). Yet, the majority suggests that there is "wisdom" in delaying relief for plaintiffs until the Legislature has had a chance to act, ante, at 43, much as the City of Memphis urged the "wisdom of proceeding slowly and gradually in its desegregation efforts." Watson, 373 U.S. at 528.
This rule requires the exclusion of evidence obtained as the result of unconstitutional searches and seizures.
Unlike Linkletter, the issue here is not whether the majority's holding should be applied retroactively or prospectively, but rather whether the relief it has promised should be provided promptly by this Court or at some uncertain future time by the Legislature. Neither these plaintiffs, nor any same-sex couples seeking the benefits and protections of marriage, obtain any relief until the Legislature acts, or failing that, this Court acts again. Thus, the majority is not applying its holding on even a purely prospective basis. In any event, assuming that Linkletter continues to have vitality in cases involving civil rights violations, see Fairfax Covenant Church v. Fairfax County Sch. Bd., 17 F.3d 703, 709, 710 (4th Cir. 1994) (stating that Supreme Court has recently cast serious doubt upon practice of departing from traditional rule of retroactive application, which is "the rule inherent in the judicial function" of applying and interpreting law in real controversies), the "unforeseen consequences" alluded to by the majority cannot be considered "exigencies" warranting relief only at some unspecified future time.
The other two cases cited by the majority also concern whether court rulings should be applied prospectively or retroactively. In those cases, the courts weighed the potential consequences of a decision to abrogate common-law sovereign immunity — the doctrine declaring that the government is immune from lawsuits. See Smith v. State, 473 P.2d 937, 950 (Idaho 1970) (applying decision to abrogate doctrine of sovereign immunity to cases before court but otherwise staying decision until adjournment of following legislative session to prevent undue hardship to government agencies that relied on doctrine); Spanel v. Mounds View Sch. Dist. No. 621, 118 N.W.2d 795, 803-04 (Minn. 1962) (staying decision to abrogate sovereign immunity until following legislative session to prevent hardship to government agencies that relied on doctrine); cf. Presley v. Miss. State Highway Comm'n, 608 So.2d 1288, 1298 (Miss. 1992) (giving retroactive application to decision finding sovereign immunity act unconstitutional would pose fiscally disastrous consequences to state agencies). These courts simply acknowledged that retroactively applying their holding abrogating sovereign immunity, without affording the Legislature an opportunity either to alter insurance coverage or enact an immunity statute, would have potentially disastrous fiscal consequences for the state. See Hillerby v. Town of Colchester, 167 Vt. 270, 293, 706 A.2d 446, 459 (1997) (Johnson, J., dissenting) (favoring quasi-prospective approach that would afford Legislature time to react to holding abrogating general municipal immunity). That is not the situation here, where no disastrous consequences, fiscal or otherwise, have been identified.
I recognize that the Legislature is, and has been, free to pass legislation that would provide same-sex couples with marital benefits. But the majority does not explain why it is necessary for the Legislature to act before we remedy the constitutional violation that we have found. In our system of government, civil rights violations are remedied by courts, not because we issue "Holy Writ" or because we are "the only repository of wisdom." Ante, at 43-44. It is because the courts "must ultimately define and defend individual rights against government in terms independent of consensus or majority will." L. Tribe, American Constitutional Law § 15.3, at 896 (1978). "`[G]roups that have historically been the target of discrimination cannot be expected to wait patiently for the protection of their human dignity and equal rights while governments move toward reform one step at a time.'" Rosenberg v. Canada, Docket No. C22807 (Ontario Court of Appeals, April 23, 1998, at 17-18 (quoting Vriend v. Alberta,  S.C.J. No. 29 (Q.L.), at para. 122). Once a court has determined that a discriminatory classification has deprived plaintiffs of a constitutionally ripe entitlement, the court must decide if the classification "is demonstrably justifiable in a free and democratic society, not whether there might be a more propitious time to remedy it." Id. at 18.
Judicial authority is not, however, the ultimate source of constitutional authority. Within our constitutional framework, the people are the final arbiters of what law governs us; they retain the power to amend our fundamental law. If the people of Vermont wish to overturn a constitutionally based decision, as happened in Alaska and Hawaii, they may do so. The possibility that they may do so, however, should not, in my view, deprive these plaintiffs of the remedy to which they are entitled.
Today's decision, which is little more than a declaration of rights, abdicates that responsibility. The majority declares that plaintiffs have been unconstitutionally deprived of the benefits of marriage, but does not hold that the marriage laws are unconstitutional, does not hold that plaintiffs are entitled to the license that triggers those benefits, and does not provide plaintiffs with any other specific or direct remedy for the constitutional violation that the Court has found to exist. By suspending its judgment and allowing the Legislature to choose a remedy, the majority, in effect, issues an advisory opinion that leaves plaintiffs without redress and sends the matter to an uncertain fate in the Legislature. Cf. In re Williams, 154 Vt. 318, 318-19, 321, 577 A.2d 686, 686-87 (1990) (statute requiring district court to hold hearings, issue findings, and advise local legislative bodies concerning alleged police misconduct violated separation of powers between judicial and legislative branches by requiring courts to give advisory opinions, upon which municipalities might or might not act). Ironically, today's mandate will only increase "the uncertainty and confusion" that the majority states it is designed to avoid. Ante, at 40.
No decision of this Court will abate the moral and political debate over same-sex marriage. My view as to the appropriateness of granting plaintiffs the license they seek is not based on any overestimate (or any estimate) of its effectiveness, nor on a miscalculation (or any calculation) as to its likely permanence, were it to have received the support of a majority of this Court. Rather, it is based on what I believe are the commands of our Constitution.
Although I concur with the majority's conclusion that Vermont law unconstitutionally excludes same-sex couples from the benefits of marriage, I write separately to state my belief that this is a straightforward case of sex discrimination.
As I argue below, the marriage statutes establish a classification based on sex. Whether such classification is legally justifiable should be analyzed under our common-benefits jurisprudence, which until today, has been closely akin to the federal equal-protection analysis under the Fourteenth Amendment. Therefore, the State must show that the classification is narrowly tailored to further important, if not compelling, interests. Not only do the rationalizations advanced by the State fail to pass constitutional muster under this or any other form of heightened scrutiny, they fail to satisfy the rational-basis test as articulated under the Common Benefits Clause.
The majority misconstrues my opinion. See ante, at 27 n. 13. I do not reach the issue of whether heightened scrutiny is appropriate for sex-based classifications under the Common Benefits Clause. See Ashwander v. Tennessee Valley Auth., 297 U.S. 288, 347 (1936) (Brandeis, J., concurring) (courts should not formulate rules of constitutional law broader than is required by precise facts to which they are to be applied). I mention federal law and that of other states merely to acknowledge the approach of other jurisdictions on an issue that we have not yet decided. I analyze the sex-based classification under our current test for rational-basis review.
In its brief, the State notes that if the Court declares that heightened scrutiny is applicable, it might offer additional arguments and justifications to demonstrate a compelling State interest in the marriage statutes. Obviously, in its extensive filings both in the trial court and here, which included a one-hundred-page appellate brief, the State made every conceivable argument in support of the marriage laws, including what it perceived to be its best arguments. For the reasons stated by the majority, see ante, at 4 n. 1, 37 n. 14, I agree that it would be pointless to remand this matter for further proceedings in the trial court.
"We have held that the Common Benefits Clause in the Vermont Constitution, see ch. I, art. 7, is generally coextensive with the equivalent guarantee in the United States Constitution, and imports similar methods of analysis." Brigham, 166 Vt. at 265, 692 A.2d at 395; see also Lorrain, 160 Vt. at 212, 628 A.2d at 550 (test under Common Benefits Clause is same as test under federal Equal Protection Clause). Where the statutory scheme affects a fundamental constitutional right or involves a suspect classification, "the State must demonstrate that any discrimination occasioned by the law serves a compelling governmental interest, and is narrowly tailored to serve that objective." Brigham, 166 Vt. at 265, 692 A.2d at 396. Otherwise, classifications are constitutional if they are "reasonably related to the promotion of a valid public purpose." MacCallum, 165 Vt. at 457, 686 A.2d at 937-38. As the majority states, the marriage "statutes, read as a whole, reflect the common understanding that marriage under Vermont law consists of a union between a man and a woman." Ante, at 6. Thus, the statutes impose a sex-based classification. See, e.g., Brause v. Bureau of Vital Statistics, No. 3AN-95-6562 CI, *6, 1998 WL 88743 (Alaska Super. Feb. 27, 1998) (prohibition on same-sex marriage is sex-based classification); Baehr v. Lewin, 852 P.2d 44, 64 (Haw. 1993) (Levinson, J., plurality opinion) (same). A woman is denied the right to marry another woman because her would-be partner is a woman, not because one or both are lesbians. Similarly, a man is denied the right to marry another man because his would-be partner is a man, not because one or both are gay. Thus, an individual's right to marry a person of the same sex is prohibited solely on the basis of sex, not on the basis of sexual orientation. Indeed, sexual orientation does not appear as a qualification for marriage under the marriage statutes. The State makes no inquiry into the sexual practices or identities of a couple seeking a license.
The State advances two arguments in support of its position that Vermont's marriage laws do not establish a sex-based classification. The State first contends that the marriage statutes merely acknowledge that marriage, by its very nature, cannot be comprised of two persons of the same sex. Thus, in the State's view, it is the definition of marriage, not the statutes, that restricts marriage to two people of the opposite sex. This argument is circular. It is the State that defines civil marriage under its statute. The issue before us today is whether the State may continue to deprive same-sex couples of the benefits of marriage. This question is not resolved by resorting to a historical definition of marriage; it is that very definition that is being challenged in this case.
The State's second argument, also propounded by the majority, see ante, at 27 n. 13, is that the marriage statutes do not discriminate on the basis of sex because they treat similarly situated males the same as similarly situated females. Under this argument, there can be no sex discrimination here because "[i]f a man wants to marry a man, he is barred; a woman seeking to marry a woman is barred in precisely the same way. For this reason, women and men are not treated differently." C. Sunstein, Homosexuality and the Constitution, 70 Ind. L.J. 1, 19 (1994). But consider the following example. Dr. A and Dr. B both want to marry Ms. C, an X-ray technician. Dr. A may do so because Dr. A is a man. Dr. B may not because Dr. B is a woman. Dr. A and Dr. B are people of opposite sexes who are similarly situated in the sense that they both want to marry a person of their choice. The statute disqualifies Dr. B from marriage solely on the basis of her sex and treats her differently from Dr. A, a man. This is sex discrimination.
Under the State's analysis, a statute that required courts to give custody of male children to fathers and female children to mothers would not be sex discrimination. Although such a law would not treat men and women differently, I believe it would discriminate on the basis of sex. Apparently, the Legislature agrees. By prohibiting consideration of the sex of the child or parent in custody decisions, see 15 V.S.A. § 665(c), the Legislature undoubtedly intended to prohibit sex discrimination, even if the rules applied equally to men and women. See Harris v. Harris, 162 Vt. 174, 182, 647 A.2d 309, 314 (1994) (stating the family court's custody decision would have to be reversed if it had been based on preference that child remain with his father because of his gender).
I recognize, of course, that although the classification here is sex-based on its face, its most direct impact is on lesbians and gay men, the class of individuals most likely to seek same-sex marriage. Viewing the discrimination as sex-based, however, is important. Although the original purpose of the marriage statutes was not to exclude same-sex couples, for the simple reason that same-sex marriage was very likely not on the minds of the Legislature when it passed the licensing statute, the preservation of the sex-based classification deprives lesbians and gay men of the right to marry the life partner of their choice. If, as I argue below, the sex-based classification contained in the marriage laws is unrelated to any valid purpose, but rather is a vestige of sex-role stereotyping that applies to both men and women, the classification is still unlawful sex discrimination even if it applies equally to men and women. See McCallum, 165 Vt. at 459, 686 A.2d at 939 (Constitution does not permit law to give effect, either directly or indirectly, to private biases; when government itself makes the classification, it is obliged to afford all persons equal protection of the law); Loving v. Virginia, 388 U.S. 1, 8-9, 11 (1967) (statute prohibiting racial intermarriage violates Equal Protection Clause although it applies equally to Whites and Blacks because classification was designed to maintain White Supremacy.) Although Vermont has not had occasion to consider the question, most, if not all, courts have held that the denial of rights or benefits on the basis of sex subject the state's action to some level of heightened scrutiny. This is so because the sex of an individual "frequently bears no relation to ability to perform or contribute to society." Frontiero v. Richardson, 411 U.S. 677, 686 (1973) (plurality opinion). Moreover, in some cases, such as here, sex-based classifications "very likely reflect outmoded notions of the relative capabilities of men and women." Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 441 (1985).
I do not contend, as the majority suggests, that the real purpose of the exclusion of same-sex partners from the marriage laws was to maintain certain male and female stereotypes. See ante, at 28 n. 13. As noted above, I agree that the original purpose was very likely not intentionally discriminatory toward same-sex couples. The question is whether the State may maintain a classification today only by giving credence to generally discredited sex-role stereotyping. I believe our decision in MacCallum says no. See Sunstein, supra, at 23, 27 (exclusion of same-sex couples from marriage is, in reality, impermissible sex-role stereotyping, and therefore, is discrimination on basis of sex); J. Culhane, Uprooting the Arguments Against Same-Sex Marriage, 20 Cardozo L.Rev. 1119, 1171-75 (1999) (accord).
See, e.g., United States v. Virginia, 518 U.S. 515, 533 (1996) (concluding that sex-based classifications are subject to heightened standard of review less rigorous than that imposed for race or national origin classifications); Frontiero v. Richardson, 411 U.S. 677, 684, 686 (1973) (plurality opinion) (concluding that sex is suspect classification under two-part test inquiring whether class is defined by immutable characteristic and whether there is history of invidious discrimination against class); Sailer Inn, Inc. v. Kirby, 485 P.2d 529, 540 (Cal. 1971) (applying federal two-part test and concluding that sex is immutable trait and women have historically labored under severe legal and social disabilities); Hewitt v. State Accident Ins. Fund Corp., 653 P.2d 970, 977 (Or. 1982) (applying federal two-part test and concluding that sex is immutable personal characteristic and purposeful unequal treatment of women is well known).
I do not believe that it is necessary to reach the question in this case, however, because in my view, the justifications asserted by the State do not satisfy even our rational-basis standard under the Common Benefits Clause, which requires that the classification be "reasonably related to the promotion of a valid public purpose." MacCallum, 165 Vt. at 457 n. 1, 686 A.2d at 938 n. 1 (because statute failed to pass constitutional muster under rational-basis test, no need to determine whether adopted persons are suspect class). In MacCallum, we invalidated, under the Common Benefits Clause, a statute denying an adopted person's right of inheritance from collateral kin, stating that the statute was grounded on outdated prejudices instead of a valid public purpose. See id. at 460-62, 686 A.2d at 939-41. Rather than blindly accept any conceivable justification proffered by the State in that case, we carefully considered the State's rationales to determine whether the discriminatory classification rested upon a reasonable consideration of legislative policy. See id. at 457, 459-61, 696 A.2d at 938, 939-40; see also Romer v. Evans, 517 U.S. 620, 635-36 (state constitutional amendment prohibiting all legislative, executive, or judicial action designed to protect homosexuals from discrimination violated Equal Protection Clause under rational-basis test because it was discriminatory and had no proper legislative end); Cleburne, 473 U.S. at 450 (ordinance requiring special use permit for operation of home for mentally retarded violated Equal Protection Clause under rational basis test because it rested on irrational prejudice rather than legitimate government purpose).
The question remains why I feel it is necessary to identify the class of persons being discriminated against in this case if the majority and I reach the same conclusion. It is important because I have concerns about the test that the majority devises to review equal-protection challenges under the Common Benefits Clause. The majority rejects the notion that the Court should accord some measure of heightened scrutiny for classifications denying benefits to historically disadvantaged groups. It argues that the history of the Common Benefits Clause supports the Court's adoption of a uniform standard that is reflective of the broad inclusionary principle at its core. Therefore, rather than accord any particular group heightened scrutiny, it will balance all the factors in the case and reach a just result. While this notion is superficially attractive in its attempt to achieve fundamental fairness for all Vermonters, it is flawed with respect to an equal-protection analysis. The guarantee of equal protection is about fundamental fairness in a large sense, but its most important purpose is to secure the rights of historically disadvantaged groups whose exclusion from full participation in all facets of society has resulted from hatred and prejudice.
I share Justice Dooley's concern that the new standard enunciated by the majority may not give sufficient deference to the Legislature's judgment in economic and commercial legislation. See ante, at 15-16 (Dooley, J., concurring). It is the Legislature's prerogative to decide whether, for example, to give "optometrists" more protection than "opticians." See Cleburne, 473 U.S. at 471 (Marshall, J., concurring in part and dissenting in part). Such classifications ought not to become a matter of serious constitutional review, even though optometrists and opticians comprise "a part of the community" and may have vital economic interests in the manner in which they are regulated. I am certain the majority would agree with that proposition and argue that its balancing of all the relevant factors in that kind of a case would not result in striking down a classification that treated those two groups differently. But therein lies my concern with the majority's approach. Although we might agree on the optometrists/opticians classification, a balancing of all relevant factors in all equal-protection cases puts the rule of law at "excessive risk." C. Sunstein, Foreward: Leaving Things Undecided, 110 Harvard L. Rev. 4, 78 (1996). As Professor Sunstein explains:
The use of `tiers' has two important goals. The first is to ensure that courts are most skeptical in cases in which it is highly predictable that illegitimate motives are at work. . . . The second goal of a tiered system is to discipline judicial discretion while promoting planning and predictability for future cases. Without tiers, it would be difficult to predict judicial judgments under the Equal Protection Clause, and judges would make decisions based on ad hoc assessments of the equities. The Chancellor's foot[*] is not a promising basis for anti-discrimination law.
Id. The majority argues that subjective judgment is required to make choices about classes who are entitled to heightened review and, therefore, that a tiered approach is not more precise than the balancing-of-factors approach. See ante, at 24 n. 10. But, in choosing the suspect class, it would be incumbent upon the Court to articulate its rationale, thereby providing predictive value in future cases of discrimination rather than depending on the "perspicacity of judges to see it." Cleburne, 473 U.S. at 466 (Marshall, J., concurring in part and dissenting in part).
[*The reference to the Chancellor's foot in the Sunstein quote is from John Seldon's (1584-1654) critique of equity, which is relevant here:]
Equity is a roguish thing. For Law we have a measure, know what to trust to; Equity is according to the conscience of him that is Chancellor, and as that is larger or narrower, so is Equity. `Tis all one as if they should make the standard for the measure we call a "foot" a Chancellor's foot; what an uncertain measure would this be! One Chancellor has a long foot, another a short foot, a third an indifferent foot. `Tis the same thing in the Chancellor's conscience.
J. Bartlett, Familiar Quotations, 263 (15th ed. 1980).
Before applying the rational-basis standard to the State's justifications, it is helpful to examine the history of the marriage laws in Vermont. There is no doubt that, historically, the marriage laws imposed sex-based roles for the partners to a marriage — male provider and female dependent — that bore no relation to their inherent abilities to contribute to society. Under the common law, husband and wife were one person. See R. E. Builders, Inc. v. Chandler, 144 Vt. 302, 303-04, 476 A.2d 540, 541 (1984). The legal existence of a woman was suspended by marriage; she merged with her husband and held no separate rights to enter into a contract or execute a deed. See id. She could not sue without her husband's consent or be sued without joining her husband as a defendant. See id. Moreover, if a woman did not hold property for her `sole and separate use' prior to marriage, the husband received a freehold interest in all her property, entitling him to all the rents and profits from the property. See id.
Starting in the late nineteenth century, Vermont, like other states, began to enact statutes, such as the Rights of Married Women Act, see 15 V.S.A. § 61-69, to grant married women property and contractual rights independent of their husbands. See Medical Ctr. Hosp. v. Lorrain, 165 Vt. 12, 14, 675 A.2d 1326, 1328 (1996). The Legislature's intent in enacting the Rights of Married Women Act was to "reject the archaic principle that husband and wife are `one person,'" and "to set a married woman free `from the thraldom of the common law.'" Richard v. Richard, 131 Vt. 98, 102, 106, 300 A.2d 637, 639, 641 (1973). Thus, we recognized that the legal existence of married women was no longer merged into that of their husbands, see Lorrain, 165 Vt. at 15, 676 A.2d at 1329, and that "a married woman is a `person' under the Constitution of Vermont." Richard, 131 Vt. at 106, 300 A.2d at 641.
Today, the partners to a marriage are equal before the law. See R E Builders, 144 Vt. at 304, 476 A.2d at 541 (modern statutes attempt to accord wives legal rights equal to husbands). A married woman may now enter contracts, sue and be sued without joining her husband, purchase and convey property separate from her husband, own property, and collect rents and profits from it. See Lorrain, 165 Vt. at 15, 675 A.2d at 1329 (women have property and contractual rights equal to men regardless of their marital status). As the Legislature enacted statutes to confer rights upon married women, this Court abolished common-law doctrines arising from the common law theory that husband and wife were one person and that the wife had no independent legal existence. See, e.g., Richard, 131 Vt. at 106, 300 A.2d at 641 (abolishing interspousal immunity, which was based on "archaic principle" that husband and wife are one person, to allow passenger wife to sue husband for personal injuries arising from husband's negligence in operating automobile).
The question now is whether the sex-based classification in the marriage law is simply a vestige of the common-law unequal marriage relationship or whether there is some valid governmental purpose for the classification today. See MacCallum, 165 Vt. at 460-62, 686 A.2d at 939-41 (State's rationales proffered to validate statutory classification cannot rest on outdated presumptions not reasonable today when vast cultural and social changes have occurred). In support of the marriage statutes, the State advances public purposes that fall into three general categories.
In the first category, the State asserts public purposes — uniting men and women to celebrate the "complementarity" (sic) of the sexes and providing male and female role models for children — based on broad and vague generalizations about the roles of men and women that reflect outdated sex-role stereotyping. The State contends that (1) marriage unites the rich physical and psychological differences between the sexes; (2) sex differences strengthen and stabilize a marriage; (3) each sex contributes differently to a family unit and to society; and (4) uniting the different male and female qualities and contributions in the same institution instructs the young of the value of such a union. The State relies on social science literature, such as Carol Gilligan's In a Different Voice: Psychological Theory and Women's Development (1982), to support its contention that there are sex differences that justify the State requiring two people to be of opposite sex to marry.
The State attempts to analogize this case to the changes in law brought about by women's participation in the legal profession starting in the 1970s, arguing that women have brought a different voice to legal theory and practice. The State also points to United States v. Virginia, 518 U.S. 515, 533 (1996) (hereinafter VMI), arguing that an institution or community made up exclusively of one sex is different from a community composed of both. The goal of diversity has been recognized to justify affirmative action programs in public broadcasting and education. See, e.g., Metro v. Broadcasting, Inc. FCC, 497 U.S. 547, 567-68 (1990) (holding that state interest in racial diversity in broadcasting justified affirmative-action racial classification); Regents of Univ. of Calif. v. Bakke, 438 U.S. 265, 311-319 (1978) (opinion of Powell, J.) (endorsing race classification in university admission as legitimate means of achieving diversity). Similarly, the recognition that women may contribute differently from men is a valid argument for women's full participation in all aspects of public life. The goal of community diversity has no place, however, as a requirement of marriage.
To begin with, carried to its logical conclusion, the State's rationale could require all marriages to be between people, not just of the opposite sex, but of different races, religions, national origins, and so forth, to promote diversity. Moreover, while it may be true that the female voice or point of view is sometimes different from the male, such differences are not necessarily found in comparing any given man and any given woman. The State's implicit assertion otherwise is sex stereotyping of the most retrograde sort. Nor could the State show that the undoubted differences between any given man and woman who wish to marry are more related to their sex than to other characteristics and life experiences. In short, the "diversity" argument is based on illogical conclusions from stereotypical imaginings that would be condemned by the very case cited for its support. See VMI, 518 U.S. at 533 (justifications for sex-based classifications "must not rely on overbroad generalizations about different talents, capabilities, or preferences of males and females.").
In the second category, the State asserts, under several different guises, the public purpose of maintaining the sex-based classification. First, the State claims an interest in "preserving the existing marital structure." Second, the State claims an interest in "instructing the young of the value of uniting male and female qualities." This is mere tautology. The State's objective is to preserve the status quo, but that does not address the question of whether the classification can be justified. Perpetuating the classification, in and of itself, is not a valid purpose for the classification. See id. at 545 (rejecting as circular governmental justification that sex-based classification is essential to governmental objective of single-sex education). Many of the State's remaining justifications, which I place into a third category, assume highly questionable public purposes. But because none of these justifications are even remotely, much less reasonably, related to the challenged classification, I accept, for the sake of argument, the premise that each of them concerns a legitimate state interest.
The State contends, for example, that prohibiting individuals from marrying a person of the same sex promotes the public purpose of minimizing custody and visitation disputes arising from surrogacy contracts because the prohibition may deter use of technologically assisted reproduction by same-sex couples. Further, the State argues that increased use of technologically assisted reproduction "may lead men who conceive children by sexual union to perceive themselves as sperm donors, without any responsibility for their offspring." Both of these reasons suffer from the same constitutional deficiency. If the state purpose is to discourage technologically assisted reproduction, I agree with the majority that the classification is significantly underinclusive. The State does nothing to discourage technologically assisted reproduction by individuals or opposite-sex couples. Moreover, opposite-sex couples may obtain marriage licenses without regard to whether or not they will use technologically assisted reproduction. The public purpose provides no rationale for the different treatment.
The State does not address the apparent conflict between the public purposes it asserts and the legislative policy of this State. Vermont does not prohibit the donation of sperm or the use of technologically assisted methods of reproduction. Thus, same-sex partners and single individuals may use technologically assisted reproduction, all without the benefit of marriage. It is impossible to accept that the classification in the marriage statutes serves as a reasonable deterrent to such methods.
The State also asserts that it has an interest in furthering the link between procreation and child rearing "to ensure that couples who engage in sexual intercourse accept responsibility for the potential children they might create." But the State cannot explain how the failure of opposite-sex couples to accept responsibility for the children they create relates at all to the exclusion of same-sex couples from the benefits of marriage. To the extent that couples, same-sex or opposite-sex, will fail to take responsibility for the children they create, the risk is greater where the couples are not married. Therefore, denying same-sex couples the benefits of marriage on this ground is not only arbitrary but completely at odds with the stated government purpose.
The State further contends that prohibiting individuals from marrying same-sex partners will deter marriages of convenience entered into solely to obtain tax benefits or government assistance. Two persons of the opposite sex are completely free to enter into a marriage of convenience, however, without the State examining their motives. Indeed, the pool of opposite-sex couples who may choose to enter into such marriages is much greater than the pool of same-sex couples. Once again, the public purpose provides no rationale for treating individuals who choose same-sex partners differently from those who choose opposite-sex partners.
Although "[a] statute need not regulate the whole of a field to pass constitutional muster," Benning v. State, 161 Vt. 472, 486, 641 A.2d 757, 764 (1994), there still must be some rational basis for an underinclusive classification. Here, none of the alleged governmental purposes within the third category of State justifications provides a rational basis for treating similarly situated people differently, or for applying the classification in an underinclusive manner. See Cleburne, 473 U.S. at 446 (State may not impose classification where relationship to asserted goal is so attenuated as to render distinction arbitrary or irrational). The State's justifications are nothing more than post-hoc rationalizations completely unrelated to any rational reason for excluding same-sex couples from obtaining the benefits of marriage.
Finally, the State claims a valid public purpose in adopting a classification to align itself with the other states. The Vermont Constitution is freestanding authority, however, and may protect rights not protected under the federal constitution or other state constitutions. Brigham, 166 Vt. at 257, 268, 692 A.2d at 391, 397 (recognizing right to equal education under Vermont Constitution, while acknowledging that this right is not recognized under federal constitution and is recognized under only some state constitutions). This Court does not limit the protections the Vermont Constitution confers on Vermonters solely to make Vermont law consistent with that of other states. See id. at 257, 692 A.2d at 391 (decisions in other states are of limited precedential value because each state's constitutional evolution is unique). Indeed, as the majority notes, Vermont's marriage laws are already distinct in several ways from the laws of other states.
In sum, the State treats similarly situated people — those who wish to marry — differently, on the basis of the sex of the person they wish to marry. The State provides no legally valid rationale for the different treatment. The justifications asserted by the State for the classification are tautological, wholly arbitrary, or based on impermissible assumptions about the roles of men and women. None of the State's justifications meets the rational-basis test under the Common Benefits Clause. Finding no legally valid justification for the sex-based classification, I conclude that the classification is a vestige of the historical unequal marriage relationship that more recent legislative enactments and our own jurisprudence have unequivocally rejected. The protections conferred on Vermonters by the Common Benefits Clause cannot be restricted by the outmoded conception that marriage requires one man and one woman, creating one person — the husband. As this Court recently stated, "equal protection of the laws cannot be limited by eighteenth-century standards." See Brigham, 166 Vt. at 267, 692 A.2d at 396.
This case is undoubtedly one of the most controversial ever to come before this Court. Newspaper, radio and television media have disclosed widespread public interest in its outcome, as well as the full spectrum of opinion as to what that outcome should be and what its ramifications may be for our society as a whole. One line of opinion contends that this is an issue that ought to be decided only by the most broadly democratic of our governmental institutions, the Legislature, and that the small group of men and women comprising this Court has no business deciding an issue of such enormous moment. For better or for worse, however, this is simply not so. This case came before us because citizens of the state invoked their constitutional right to seek redress through the judicial process of a perceived deprivation under state law. The Vermont Constitution does not permit the courts to decline to adjudicate a matter because its subject is controversial, or because the outcome may be deeply offensive to the strongly held beliefs of many of our citizens. We do not have, as does the Supreme Court of the United States, certiorari jurisdiction, which allows that Court, in its sole discretion, to decline to hear almost any case. To the contrary, if a case has been brought before us, and if the established procedures have been followed, as they were here, we must hear and decide it.
Moreover, we must decide the case on legal grounds. However much history, sociology, religious belief, personal experience or other considerations may inform our individual or collective deliberations, we must decide this case, and all cases, on the basis of our understanding of the law, and the law alone. This must be the true and constant effort of every member of the judiciary. That effort, needless to say, is not a guarantee of infallibility, nor even an assurance of wisdom. It is, however, the fulfillment of our pledge of office.