“[T]he ‘absolute equality of all’ persons before the law [is] ‘the very foundation principle of our government.’ ” (Varnum v. Brien (Iowa 2009) 763 N.W.2d 862, 877.)
In In re Marriage Cases (2008) 43 Cal.4th 757, 855-856 [76 Cal.Rptr.3d 683, 183 P.3d 384] (Marriage Cases), we held that denying same-sex couples the right to marry denies them equal protection of the law. Proposition 8 partially abrogated that decision by amending the California Constitution to deny same-sex couples fully equal treatment by adding the words: “Only marriage between a man and a woman is valid or recognized in California.”
The question before us is not whether the language inserted into the California Constitution by Proposition 8 discriminates against same-sex couples and denies them equal protection of the law; we already decided in the Marriage Cases that it does. The question before us today is whether such a change to one of the core values upon which our state Constitution is founded can be accomplished by amending the Constitution through an initiative measure placed upon the ballot by the signatures of 8 percent of the number of persons who voted in the last gubernatorial election and passed by a simple majority of the voters. (Cal. Const., art. II, § 8.) Or is this limitation on the scope of the equal protection clause to deny the full protection of the law to a minority group based upon a suspect classification such a fundamental change that it can only be accomplished by revising the California Constitution, either through a constitutional convention or by a measure passed by a two-thirds vote of both houses of the Legislature and approved by the voters? (Cal. Const., art. XVIII.)
For reasons elaborated below, I conclude that requiring discrimination against a minority group on the basis of a suspect classification strikes at the *484core of the promise of equality that underlies our California Constitution and thus “represents such a drastic and far-reaching change in the nature and operation of our governmental structure that it must be considered a ‘revision’ of the state Constitution rather than a mere ‘amendment’ thereof.” (Amador Valley Joint Union High Sch. Dist. v. State Bd. of Equalization (1978) 22 Cal.3d 208, 221 [149 Cal.Rptr. 239, 583 P.2d 1281] (Amador Valley).) .The rule the majority crafts today not only allows same-sex couples to be stripped of the right to marry that this court recognized in the Marriage Cases, it places at risk the state constitutional rights of all disfavored minorities. It weakens the status of our state Constitution as a bulwark of fundamental rights for minorities protected from the will of the majority. I therefore dissent.1
Equal protection principles lie at the core of the California Constitution and have been embodied in that document from its inception. (Grodin et al., The California State Constitution: A Reference Guide (1993) p. 47.) Former section 11 of article I of the original 1849 Constitution stated, “All laws of a general nature shall have a uniform operation” and section 21 of article I of the 1879 Constitution added, “nor shall any citizen, or class of citizens, be granted privileges or immunities which, upon the same terms, shall not be granted to all citizens.” These provisions were “substantially the equivalent of the equal protection clause of the Fourteenth Amendment to the United States Constitution.” (Department of Mental Hygiene v. Kirchner (1965) 62 Cal.2d 586, 588 [43 Cal.Rptr. 329, 400 P.2d 321]; see Sail’er Inn, Inc. v. Kirby (1971) 5 Cal.3d 1, 15, fn. 13 [95 Cal.Rptr. 329, 485 P.2d 529].) In 1974, an express equal protection clause was added to the California Constitution that mirrors the language of the Fourteenth Amendment to the United States Constitution.2
Ensuring equal protection prevents “governmental decisionmakers from treating differently persons who are in all relevant respects alike. [Citation.]” (Nordlinger v. Hahn (1992) 505 U.S. 1, 10 [120 L.Ed.2d 1, 112 S.Ct. 2326].) The doctrine’s purpose is to protect “against intentional and arbitrary discrimination.” (Sunday Lake Iron Co. v. Wakefield (1918) 247 U.S. 350, 352 [62 *485L.Ed. 1154, 38 S.Ct. 495].) As such, it is a shield against arbitrary government power, because equal protection “requires the democratic majority to accept for themselves and their loved ones what they impose on you and me.” (Cruzan v. Director, Missouri Dept. of Health (1990) 497 U.S. 261, 300 [111 L.Ed.2d 224, 110 S.Ct. 2841] (conc. opn. of Scalia, J.).) Thus, it is not so much a discrete constitutional right as it is a basic constitutional principle that guides all legislation and compels the will of the majority to be tempered by justice. The Iowa Supreme Court, in affirming the constitutional right of gays and lesbians to marry, recently recognized the importance of this promise of equality, stating: “If gay and lesbian people must submit to different treatment without an exceedingly persuasive justification, they are deprived of the benefits of the principle of equal protection upon which the rule of law is founded.” (Varnum v. Brien, supra, 163 N.W.2d 862, 906, italics added.)
Of particular importance for this case is that discrimination against disfavored minorities is presumptively suspect under the equal protection clause. As we affirmed in the Marriage Cases, supra, 43 Cal.4th at page 842, and as the majority reaffirms today (maj. opn., ante, at p. 411), sexual orientation is such a suspect classification. Under our state equal protection jurisprudence, as in federal law, laws that involve suspect classifications or touch upon fundamental interests are subject to strict scrutiny, meaning that “ ‘ “ ‘ “the state bears the burden of establishing not only that it has a compelling interest which justifies the law but that the distinctions drawn by the law are necessary to further its purpose.” [Citation.]’ ” ’ ” (Marriage Cases, supra, 43 Cal.4th at p. 832, italics omitted.)
The equal protection clause is therefore, by its nature, inherently countermajoritarian. As a logical matter, it cannot depend on the will of the majority for its enforcement, for it is the will of the majority against which the equal protection clause is designed to protect. Rather, the enforcement of the equal protection clause is especially dependent on “the power of the courts to test legislative and executive acts by the light of constitutional mandate and in particular to preserve constitutional rights, whether of individual or minority, from obliteration by the majority.” (Bixby v. Pierno (1971) 4 Cal.3d 130, 141 [93 Cal.Rptr. 234, 481 P.2d 242].)
California’s equal protection doctrine has not been confined to that of federal Fourteenth Amendment jurisprudence: “[0]ur state equal protection provisions ... are possessed of an independent vitality which, in a given case, may demand an analysis different from that which would obtain if only the federal standard were applicable.” (Serrano v. Priest (1976) 18 Cal.3d 728, 764 [135 Cal.Rptr. 345, 557 P.2d 929].) The equal protection clause of our state Constitution is important as a provision of independent force and effect *486only when this court extends greater protection under that provision than the high court has extended under the equal protection clause of the federal Constitution.
The majority upholds Proposition 8 by reasoning that it does not “fundamentally alter the meaning and substance of state constitutional equal protection principles as articulated” in the Marriage Cases, because it merely “carves out a narrow and limited exception to these state constitutional rights, reserving the official designation of the term ‘marriage’ for the union of opposite-sex couples . . . .” (Maj. opn., ante, at p. 388.) The majority protests that it does not mean to “diminish or minimize the significance that the official designation of ‘marriage’ holds” (ibid.), but that is exactly the effect of its decision.
Denying the designation of marriage to same-sex couples cannot fairly be described as a “narrow” or “limited” exception to the requirement of equal protection; the passionate public debate over whether same-sex couples should be allowed to marry, even in a state that offers largely equivalent substantive rights through the alternative of domestic partnership, belies such a description. “[T]he constitutional right to marry . . . has been recognized as one of the basic, inalienable civil rights guaranteed to an individual by the California Constitution . . . .” (Marriage Cases, supra, 43 Cal.4th at p. 781.) Prior to the enactment of Proposition 8, the California Constitution guaranteed “this basic civil right to all Californians, whether gay or heterosexual, and to same-sex couples as well as to opposite-sex couples.” (43 Cal.4th at p. 782.) “In light of the fundamental nature of the substantive rights embodied in the right to marry—and their central importance to an individual’s opportunity to live a happy, meaningful, and satisfying life as a full member of society—the California Constitution properly must be interpreted to guarantee this basic civil right to all individuals and couples, without regard to their sexual orientation.” (Id. at p. 820, fn. omitted.)
We recognized in the Marriage Cases that “drawing] a distinction between the name for the official family relationship of opposite-sex couples (marriage) and that for same-sex couples (domestic partnership)” (Marriage Cases, supra, 43 Cal.4th at p. 782) “impinges upon a same-sex couple’s fundamental interest in having their family relationship accorded the same respect and dignity enjoyed by an opposite-sex couple” (id. at p. 784). Denying same-sex couples the right to call their relationships marriages treats them as “ ‘second-class citizens.’ ” (Id. at p. 785.) As we observed in the Marriage Cases, “there exists a substantial risk that a judicial decision upholding the differential treatment of opposite-sex and same-sex couples would be understood as validating a more general proposition that our state by now has repudiated; that it is permissible, under the law, for society to *487treat gay individuals and same-sex couples differently from, and less favorably than, heterosexual individuals and opposite-sex couples.” (43 Cal.4th at p. 855.)
Describing the effect of Proposition 8 as narrow and limited fails to acknowledge the significance of the discrimination it requires. But even a narrow and limited exception to the promise of full equality strikes at the core of, and thus fundamentally alters, the guarantee of equal treatment that has pervaded the California Constitution since 1849. Promising equal treatment to some is fundamentally different from promising equal treatment to all. Promising treatment that is almost equal is fundamentally different from ensuring truly equal treatment. Granting a disfavored minority only some of the rights enjoyed by the majority is fundamentally different from recognizing, as a constitutional imperative, that they must be granted all of those rights. Granting same-sex couples all of the rights enjoyed by opposite-sex couples, except the right to call their “ ‘officially recognized, and protected family relationship’ ” (maj. opn., ante, at p. 388) a marriage, still denies them equal treatment.
There is no doubt that the ultimate authority over the content of the California Constitution lies with the people. “All political power is inherent in the people. Government is instituted for their protection, security, and benefit, and they have the right to alter or reform it when the public good may require.” (Cal. Const., art. II, § 1.) But there are two methods for the people to alter the California Constitution: by revising it or by amending it. A revision to the Constitution must be initiated by the Legislature in one of two ways: the Legislature, by a two-thirds vote, “may submit at a general election the question whether to call a convention to revise the Constitution” (Cal. Const., art. XVIII, § 2), or the Legislature, by a two-thirds vote, may propose a revision of the Constitution to be submitted to the voters (Cal. Const., art. XVIII, § 1). This is in contrast to a constitutional amendment, which can be accomplished by a majority of the electorate after the signatures of 8 percent of the number of persons who voted in the last gubernatorial election have qualified it for the ballot. (Cal. Const., art. II, § 8, subd. (b).)
We have long recognized the importance of this distinction between revising and amending the Constitution. In Livermore v. Waite (1894) 102 Cal. 113 [36 P. 424], which was decided before the initiative process was created in 1911, we observed that, at that time, there were “two methods by which changes may be effected in [the California Constitution], one by a convention of delegates chosen by the people for the express purpose of revising the entire instrument, and the other through the adoption by the people of propositions for specific amendments that have been previously submitted to it by two-thirds of the members of each branch of the legislature.” (Id. at *488p. 117.) We noted that there was a basic difference between the process of revising the Constitution by means of the constitutional convention and amending the Constitution. “Under the first of these methods [revision] the entire sovereignty of the people is represented in the convention. The character and extent of a constitution that may be framed by that body is freed from any limitations other than those contained in the constitution of the United States.” (Ibid.) The power of amendment, however, was much more limited: “The power of the legislature to initiate any change in the existing organic law is, however, of greatly less extent, and, being a delegated power, is to be strictly construed under the limitations by which it has been conferred. . . . The legislature is not authorized to assume the function of a constitutional convention, and propose for adoption by the people a revision of the entire constitution under the form of an amendment . . . .” (Id. at pp. 117-118.)
We took care in Livermore to explain the reason for this difference between the broad power of revision and the greatly limited power of amendment: “The very term ‘constitution’ implies an instrument of a permanent and abiding nature, and the provisions contained therein for its revision indicate the will of the people that the underlying principles upon which it rests, as well as the substantial entirety of the instrument, shall be of a like permanent and abiding nature. On the other hand, the significance of the term ‘amendment’ implies such an addition or change within the lines of the original instrument as will effect an improvement, or better carry out the purpose for which it was framed.” (Livermore v. Waite, supra, 102 Cal. at pp. 118-119.)3
The emergence of the initiative process did nothing to alter the distinction between amending and revising the Constitution. The initiative process was *489created in 1911 to permit the people to directly enact statutes and amend, but not revise, the Constitution. As has been well documented and often recounted, the introduction of direct democracy in California in the form of the initiative, referendum, and recall processes, was in response to government corruption prevalent at the beginning of the last century. (See Cal. Com. on Campaign Financing, Democracy by Initiative: Shaping California’s Fourth Branch of Government (1992) pp. 36-40.) Corporate power, principally that of the Southern Pacific Railroad, dominated state government and had undermined both the independence of the judiciary and the Legislature’s role as a servant of the popular will. (Id. at pp. 36-38.) Also of concern were corrupt political bosses and big-city machines. (Id. at pp. 39-40.) Hiram Johnson and his allies in the Progressive movement sought to restore the connection between government and the majority will by allowing the people to bypass an unresponsive Legislature and enact their own legislation. (Id. at pp. 40-42.)
Although this initiative process was thereby instituted as a remedy for government corruption, and to free legislation from the influence of powerful special interests and the Legislature’s own self-serving inertia, there is no indication that this process was intended to prevent courts from performing their traditional constitutional function of protecting persecuted minorities from the majority will. There is a fundamental difference between preventing politically powerful minorities from unduly influencing legislative and judicial decisions on the one hand, and preventing courts from protecting the rights of disfavored minorities unable to obtain equal rights through the usual maj oritarian processes on the other. There is no indication that the Progressives who framed the initiative process were insensible to that distinction, or that they sought to abolish the judiciary’s role as the guardian of minorities’ fundamental rights.
The initiative process was itself initiated by a 1911 ballot proposition that amended article IV, section 1 of the Constitution to provide in relevant part that “the people reserve to themselves the power to propose laws and amendments to the constitution, and to adopt or reject the same, at the polls independent of the legislature . . . .” There is no evidence that those enacting the initiative process intended to alter the distinction between amending and revising the Constitution that this court had recognized in Livermore v. Waite, supra, 102 Cal. 113, some 17 years earlier, and the language of that decision remains valid today. Nor did the subsequent 1962 constitutional amendment, Proposition 7, which permitted the Legislature by a two-thirds vote to propose constitutional revisions to the electorate short of a constitutional convention (see maj. opn., ante, at pp. 425-426) change the meaning of a *490revision.4 “[T]he underlying principles upon which [the Constitution] rests . . . shall be of a . . . permanent and abiding nature” and may only be altered by revising, rather than amending, the Constitution. (Livermore v. Waite, supra, 102 Cal. at pp. 118-119.)
As discussed, there is no “underlying” principle more basic to our Constitution than that the equal protection clause protects the fundamental rights of minorities from the will of the majority. Accordingly, Proposition 8’s withdrawal of any of those rights from gays and lesbians cannot be accomplished through constitutional amendment.
The majority concludes that in order to constitute a revision, a change in the Constitution must effect a “fundamental change in the basic governmental plan or framework established by the preexisting provisions of the California Constitution—that is ‘in [the government’s] fundamental structure or the foundational powers of its branches.’ [Citation.]” (Maj. opn., ante, at p. 441.) The cases cited by the majority do indeed hold that a change to the Constitution that alters the structure or framework of government is a revision, but these cases do not, as the majority erroneously concludes, also stand for the inverse of this proposition: that a change to the Constitution that does not alter the structure or framework of the Constitution cannot constitute a revision and, thus, necessarily must be an amendment. The reason is simple. None of the cases cited by the majority considered this issue, because it was not raised.
*491We recognized in Amador Valley that whether a proposed amendment constitutes a revision could turn on either the scope or the substance of the proposed change: “[0]ur analysis in determining whether a particular constitutional enactment is a revision or an amendment must be both quantitative and qualitative in nature. For example, an enactment which is so extensive in its provisions as to change directly the ‘substantial entirety’ of the Constitution by the deletion or alteration of numerous existing provisions may well constitute a revision thereof. However, even a relatively simple enactment may accomplish such far reaching changes in the nature of our basic governmental plan as to amount to a revision also. In illustration, the parties herein appear to agree that an enactment which purported to vest all judicial power in the Legislature would amount to a revision without regard either to the length or complexity of the measure or the number of existing articles or sections affected by such change.” (Amador Valley, supra, 22 Cal.3d at p. 223, italics added.) We also rejected as hyperbolic the arguments that Proposition 13 constituted a major change in governmental structure involving loss of home rule or of a republican form of government. (22 Cal.3d at pp. 224-228.)
In Brosnahan v. Brown (1982) 32 Cal.3d 236, 243 [186 Cal.Rptr. 30, 651 P.2d 274] (Brosnahan), we considered the validity of the 1982 Proposition 8 which, among other things, amended the Constitution by adding article I, section 28, subdivision (d) (section 28(d)) to the California Constitution—the so-called “truth-in-evidence provision,” which provides that “relevant evidence shall not be excluded in any criminal proceeding.” This court quickly rejected the argument that the initiative was “such a ‘drastic and far-reaching’ measure” that it constituted a revision rather than an amendment to the Constitution. (Brosnahan, supra, 32 Cal.3d at p. 260.) Citing our decision in Amador Valley, the court employed both a quantitative and qualitative analysis. The court concluded: “From a qualitative point of view, while [the 1982] Proposition 8 does accomplish substantial changes in our criminal justice system, even in combination these changes fall considerably short of constituting ‘such far reaching changes in the nature of our basic governmental plan as to amount to a revision . . . .’ [Citations.]” (Brosnahan, supra, 32 Cal.3d at p. 260.) We further rejected the contentions that the 1982 Proposition 8 would lead to significant changes in the structure of government because it would result in “(1) the inability of the judiciary to perform its constitutional duty to decide cases, particularly civil cases; and (2) the abridgement of the constitutional right to public education,” comparing this dire forecast to the predictions of loss of home rule and republican government we found baseless in Amador Valley. (Brosnahan, supra, 32 Cal.3d at p. 261.)
In its concluding statement, the Brosnahan court substituted the word “framework” for the word “plan” in restating the rule in Amador Valley that a *492revision must alter “our basic governmental plan” (Amador Valley, supra, 22 Cal.3d at p. 223), stating: “For the above reasons, nothing contained in [the 1982] Proposition 8 necessarily or inevitably will alter the basic governmental framework set forth in our Constitution. It follows that Proposition 8 did not accomplish a ‘revision’ of the Constitution . . . .” (Brosnahan, supra, 32 Cal.3d at p. 261.) The court in Brosnahan did not discuss or explain why it substituted the word “framework” for the word “plan.” Nothing in the opinion in Brosnahan indicates that the court attached any significance to this single use of the word “framework.” There is nothing to indicate that in substituting the word “framework” for the word “plan” in this one instance, the court meant to signal a departure from its holding in Amador Valley or to restrict its analysis to whether a proposed amendment would affect the structure of the government. The decision in Brosnahan never addressed whether the 1982 Proposition 8 revised the Constitution because it altered fundamental rights. Rather, it simply applied the rule stated in Amador Valley that the amendment was proper because it did not make “far reaching changes in the nature of our basic governmental plan.”
The idea that the electorate may, by amendment, significantly curtail the constitutional rights of minorities is not, contrary to the majority, squarely supported by case law. Even in the area of criminal law and procedure, in which the initiative process has perhaps made its boldest forays into the field of constitutional rights, this court has stopped short of approving the kind of basic constitutional change at issue in the present case. In In re Lance W. (1985) 37 Cal.3d 873, 885 [210 Cal.Rptr. 631, 694 P.2d 744], this court considered the 1982 Proposition 8 and rejected the argument that the addition of section 28(d) to the California Constitution—the “truth-in-evidence provision”—constituted “an impermissible constitutional revision, rather than amendment, because it abrogates the judicial function of fashioning appropriate remedies for violation of constitutional rights.”
In upholding section 28(d), we equated the power to amend the Constitution to legislative power: “The Legislature and, a fortiori, the people acting through either the reserved power of statutory initiative or the power to initiate and adopt constitutional amendments (art. II, § 8) may prescribe rules of procedure and of evidence to be followed in the courts of this state.” (In re Lance W., supra, 37 Cal.3d at p. 891.) We thus concluded that restricting the judicially created exclusionary rule “cannot be considered such a sweeping change either in the distribution of powers made in the organic document or in the powers which it vests in the judicial branch as to constitute a revision of the Constitution . . . .” (Id. at p. 892.)
Our decision in Lance W. did state, in dicta and without explanation or citation to authority: “The people could by amendment of the Constitution *493repeal section 13 of article I in its entirety.” (In re Lance W., supra, 37 Cal.3d at p. 892.)5 This passing observation was unnecessary to the decision and carries little weight. In light of the history of the revision/amendment distinction discussed above, I very much doubt that those who framed and enacted the 1911 amendment authorizing constitutional amendment by initiative contemplated the elimination of entire constitutional provisions incorporating fundamental constitutional rights. This is particularly true because at the time of the 1911 amendment, the principle that much of the Bill of Rights is applicable to the states through the Fourteenth Amendment was still largely undeveloped. (See Tribe, American Constitutional Law (2d ed. 1988) § 11.2, p. 772, and cases cited therein.) Therefore, eliminating, for example, a prohibition of unreasonable searches and seizures in 1911 would have meant not merely shaving off extra state constitutional protections that supplemented underlying federal protections, but eliminating such protections altogether. There is no evidence, and the majority points to none, that those who enacted the 1911 amendment intended such nullification of fundamental rights to be within the reach of a simple constitutional amendment enacted by a majority of the voters.
It is true that Lance W. stands for the proposition that initiative amendments may scale back judicial remedies that implement the protection of constitutional rights, but the majority makes the far broader assertion that “the current Proposition 8 is by no means the first instance in which the California Constitution has been altered, by a constitutional amendment approved by a majority of voters, in a manner that lessens the state constitutional rights of a minority group that has been the subject of past discrimination.” (Maj. opn., ante, at p. 447.) The majority cites in support the amendment to article I, section 7, subdivision (a) of the California Constitution, which circumscribed public school busing, and Proposition 209, which curtailed affirmative action programs. (See maj. opn., ante, at pp. 447-448; Hi-Voltage Wire Works, Inc. v. City of San Jose (2000) 24 Cal.4th 537, 567-568 [101 Cal.Rptr.2d 653, 12 P.3d 1068].) Both of these measures limited remedies for discrimination, but no case has ever held that the Constitution properly may be amended to deprive a minority group of a fundamental right on the basis of a suspect classification. Unlike modifying legislative or judicially created remedies, withholding a fundamental right from a minority group on the basis of a suspect classification is inherently antithetical to the core principle of equal protection that minorities are to be *494protected against the prejudice of majorities by requiring that laws apply equally to all segments of society.6
Nor is Raven v. Deukmejian (1990) 52 Cal.3d 336, 341-343 [276 Cal.Rptr. 326, 801 P.2d 1077], the one case to invalidate a portion of an initiative on the grounds that it constituted a qualitative revision, contrary to my position. In Raven, this court invalidated the portion of Proposition 115 that amended the California Constitution “to provide that certain enumerated criminal law rights . . . shall not be construed to afford greater rights to criminal or juvenile defendants than afforded by the federal Constitution” (Raven, supra, 52 Cal.3d at pp. 342-343) because it “contemplates such a far-reaching change in our governmental framework as to amount to a qualitative constitutional revision . . .” (id. at p. 341). Relying upon the hypothetical example we posed in Amador Valley, that “an enactment which purported to vest all judicial power in the Legislature would amount to a revision without regard either to the length or complexity of the measure . . .” (Amador Valley, supra, 22 Cal.3d at p. 223), we held in Raven that “Proposition 115 contemplates a similar qualitative change. In essence and practical effect, new article I, section 24, would vest all judicial interpretive power, as to fundamental criminal defense rights, in the United States Supreme Court. From a qualitative standpoint, the effect of Proposition 115 is devastating.” (Raven, supra, 52 Cal.3d at p. 352.) The court added: “In effect, new article I, section 24, would substantially alter the substance and integrity of the state Constitution as a document of independent force and effect.” (Ibid.)
Our decision in Raven addressed whether a structural change to the Constitution was a revision, but nothing in our opinion suggests that only a structural change can constitute a revision. To the contrary, our recognition in Raven that altering fundamental rights embodied in the Constitution could “substantially alter the substance and integrity of the state Constitution as a document of independent force and effect” suggests just the opposite. *495(Raven v. Deukmejian, supra, 52 Cal.3d at p. 352.) Proposition 8 would have a similar effect by emasculating the equal protection clause of the California Constitution as a provision of independent force and effect. Any protection of a minority group recognized by this court under the equal protection clause of our state Constitution that was not recognized by the United States Supreme Court under the federal Constitution could be abrogated through the initiative process by a simple majority of the voters.
The majority’s reliance upon the lead opinion in People v. Frierson (1979) 25 Cal.3d 142 [158 Cal.Rptr. 281, 599 P.2d 587] (Frierson) is also misguided. That opinion stated the view of only three justices that the 1972 initiative measure that added a provision to the California Constitution stating that the death penalty did not constitute cruel or unusual punishment amended, rather than revised, the Constitution. Each of the remaining justices made it abundantly clear that they either declined to address this issue or disagreed with the lead opinion. Nevertheless, the majority treats the lead opinion as if it were a majority opinion, referring to it as “[o]ur opinion” (maj. opn., ante, at p. 430), and incorrectly referring to the lead opinion to describe what “the court concluded” (id. at p. 443). (See also id. at p. 452.)
In a footnote, the majority acknowledges that the lead opinion in Frierson “was signed by only three justices; four justices declined to join in the opinion’s discussion” upon which the majority now relies. (Maj. opn., ante, at p. 430, fn. 21.) Nevertheless, the majority attempts to justify its reliance upon this portion of the lead opinion in Frierson by noting that a majority of the court in People v. Jackson (1980) 28 Cal.3d 264, 315 [168 Cal.Rptr. 603, 618 P.2d 149], later upheld the validity of the 1977 death penalty law, saying that “ ‘[m]ost of the arguments advanced by defendant were discussed at considerable length in [Frierson] and we do not repeat them here.’ ” (Maj. opn., ante, at p. 430, fn. 21.) This cryptic reference to the lead opinion in Frierson does not establish that the court in Jackson considered whether the 1972 initiative was a constitutional amendment or a revision, and thus does not serve to transform the views of three justices in Frierson into a holding of a majority of this court.7
*496I also find unpersuasive the majority’s reliance upon the fact that “[n]o justice in Frierson, Jackson, or any other decision of this court has disagreed with the conclusion that [the 1972 initiative measure] constitutes a permissible amendment to, rather than an impermissible revision of, the California Constitution, and there can be no question that this resolution of the issue is now a firmly settled determination.” (Maj. opn., ante, at p. 430, fn. 21.) No citation to authority follows this unsupportable assertion. There is no authority that supports the view that this court’s failure to disagree with a conclusion makes it law. Rather, it is beyond cavil that “an opinion is not authority for a proposition not therein considered.” (Ginns v. Savage, supra, 61 Cal.2d at p. 524, fn. 2.)
In sum, none of our prior cases discussed above, nor any other case discussed in the majority opinion, holds that a modification of the California Constitution constitutes a revision only if it alters the structure of government. None of our prior cases considered whether an amendment to the Constitution could restrict the scope of the equal protection clause by adding language that requires discrimination based upon a suspect classification. Nor did these cases consider, as in the present situation, whether a transfer of the authority to protect the equal rights of a suspect class away from the judiciary to an electoral majority is the type of structural change that can be effected by a constitutional amendment. For the reasons discussed above, I believe this kind of change in the countermaj oritarian nature of the equal protection clause is the type of fundamental alteration that can be done only through a constitutional revision.
It is apparent, moreover, that limiting the definition of revision only to changes in the structure of government necessarily leads to the untenable conclusion that even the most drastic and far-reaching changes to basic principles of our government do not constitute revisions so long as they do not alter the governmental framework. Counsel for interveners candidly admitted at oral argument that, in his view, the equal protection clause of the California Constitution could be repealed altogether by an amendment passed by a bare majority of voters through the initiative process.
The majority wisely does not embrace this extreme view, but it does not explain how it avoids it, simply stating that “there is no need for us to consider whether a measure that actually deprives a minority group of the entire protection of a fundamental constitutional right or, even more sweepingly, leaves such a group vulnerable to public or private discrimination in all areas without legal recourse [citation], would constitute a constitutional revision . . . .” (Maj. opn., ante, at p. 446.) But the possible basis for limiting the broad rule adopted by the majority is not apparent. If a change in the Constitution that leaves a minority group vulnerable to discrimination in all *497areas might be a revision, why not a change that leaves that group subject to discrimination in most areas, or a change like Proposition 8 that requires discrimination based upon a suspect classification in one very important area?8
Thus, under the majority’s view, it is not clear what sorts of state constitutional constraints limit the power of a majority of the electorate to discriminate against minorities. As petitioners point out, “imagine if Perez v. Sharp, 32 Cal.2d 711 [198 P.2d 17] (1948), striking down California’s ban on interracial marriages, had been decided on state constitutional grounds rather than federal constitutional grounds. And imagine if a bare majority had attempted to overturn that landmark ruling by enshrining the ban into the Constitution.” Other equally unattractive hypotheticals suggest themselves. Under the majority’s reasoning, California’s voters could permissibly amend the state Constitution to limit Catholics’ right to freely exercise their religious beliefs (Cal. Const., art. I, § 4), condition African-Americans’ right to vote on their ownership of real property (id., § 22), or strip women of the right to enter into or pursue a business or profession (id., § 8). While the federal Constitution would likely bar these initiatives, the California Constitution is intended to operate independently of (art. I, § 24), and in some cases more broadly than (see, e.g., Fashion Valley Mall, LLC v. National Labor Relations Bd. (2007) 42 Cal.4th 850, 857-858 [69 Cal.Rptr.3d 288, 172 P.3d 742]), its federal counterpart.9 The majority’s holding essentially strips the state Constitution of its independent vitality in protecting the fundamental rights of *498suspect classes. And if the majority does not avow that such broad constitutional changes could be made by amendment, but only more “limited” ones, then I disagree with such an implicit distinction. As discussed, denying gays and lesbians the right to marry, by wrenching minority rights away from judicial protection and subjecting them instead to a majority vote, attacks the very core of the equal protection principle.
The majority criticizes petitioners’ position because “under petitioners’ approach, the people would have the ability—through the initiative process—to extend a constitutional right to a disfavored group that had not previously enjoyed that right, but the people would lack the power to undo or repeal that very same extension of rights through their exercise of the identical initiative process.” (Maj. opn., ante, at p. 451.) Whether or not the above accurately characterizes petitioners’ position, it does not accurately describe mine. The scenario of a majority of the electorate giving and then taking away rights does not implicate my objections in the present case: that Proposition 8 entirely undermines the countermaj oritarian nature of the equal protection clause and usurps the judiciary’s special constitutional role as protector of minority rights. Therefore, without deciding cases not before us, my reasons for concluding that Proposition 8 attempts a constitutional change that can only be accomplished through revision do not apply to a situation in which an electoral majority grants and then repeals rights.
I realize, of course, that the right of gays and lesbians to marry in this state has only lately been recognized. But that belated recognition does not make the protection of those rights less important. Rather, that the right has only recently been acknowledged reflects an age-old prejudice (Marriage Cases, supra, 43 Cal.4th at pp. 821-822, 846, 853) that makes the safeguarding of that right by the judiciary all the more critical. As the Supreme Court of Iowa recently observed: “[G]ay and lesbian people as a group have long been the victim of purposeful and invidious discrimination because of their sexual orientation. The long and painful history of discrimination against gay and lesbian persons is epitomized by the criminalization of homosexual conduct in many parts of this country until very recently. [Citation.] Additionally, only a few years ago persons identified as homosexual were dismissed from military service regardless of past dedication and demonstrated valor. Public employees identified as gay or lesbian have been thought to pose security risks due to a perceived risk of extortion resulting from a threat of public exposure. School-yard bullies have psychologically ground children with apparently gay or lesbian sexual orientation in the cruel mortar and pestle of *499school-yard prejudice. At the same time, lesbian and gay people continue to be frequent victims of hate crimes. [Citation.]” (Varnum v. Brien, supra, 763 N.W.2d 862, 889.)10
Proposition 8 represents an unprecedented instance of a majority of voters altering the meaning of the equal protection clause by modifying the California Constitution to require deprivation of a fundamental right on the basis of a suspect classification. The majority’s holding is not just a defeat for same-sex couples, but for any minority group that seeks the protection of the equal protection clause of the California Constitution.
This could not have been the intent of those who devised and enacted the initiative process. In my view, the aim of Proposition 8 and all similar initiative measures that seek to alter the California Constitution to deny a fundamental right to a group that has historically been subject to discrimination on the basis of a suspect classification, violates the essence of the equal protection clause of the California Constitution and fundamentally alters its scope and meaning. Such a change cannot be accomplished through the initiative process by a simple amendment to our Constitution enacted by a bare majority of the voters; it must be accomplished, if at all, by a *500constitutional revision to modify the equal protection clause to protect some, rather than all, similarly situated persons. I would therefore hold that Proposition 8 is not a lawful amendment of the California Constitution.
The petition of petitioners Karen L. Strauss et al., and City and County of San Francisco for a rehearing was denied June 17, 2009, and the opinion was modified to read as printed above.
I agree with part VI of the majority opinion that Proposition 8 does not invalidate same-sex marriages entered into before its passage. (See maj. opn., ante, at p. 392.) I also agree with the majority opinion that Proposition 8 does not entirely repeal or abrogate a same-sex couple’s substantive state constitutional right to marry as set forth in the Marriage Cases, but rather carves out an exception by “reserving the official designation of the term ‘marriage’ for the union of opposite-sex couples.” (Maj. opn., ante, at p. 388.)
The equal protection clause was added to the California Constitution, article I, section 7, upon the recommendation of the California Constitution Revision Commission, as part of 1974’s Proposition 7, a ballot measure proposed by two-thirds of both the Senate and the Assembly, which, according to the Legislative Counsel’s Digest, “[rjevises, renumbers and specifically provides for various constitutional rights of persons.” (Legis. Counsel’s Dig., Assem. Const. Amend. No. 60 (1973-1974 Reg. Sess.) 2 Stats. 1974, Summary Dig., p. 275.)
The majority contends that “when the entire pertinent passage of the Livermore decision is considered, it appears reasonable to conclude that the court in Livermore itself would have recognized that a measure such as Proposition 8 constitutes a constitutional amendment, because in describing the type of measures that would constitute an amendment, the court in that case noted that ‘some popular wave of sociological reform, like the abolition of the death penalty for crime, or a prohibition against the manufacture or sale of intoxicating liquors, may induce a legislature to submit for enactment, in the permanent form of a constitutional prohibition, a rule which it has the power itself to enact as a law, but which [as such] might be of only temporary effect.’ [Citation.] In adding to the California Constitution a provision declaring that marriage shall refer only to a union between a man and a woman, Proposition 8 would appear to constitute just the type of discrete ‘popular’ and ‘sociological’ amendment that the Livermore decision had in mind.” (Maj. opn., ante, at p. 454, fn. omitted.) Yet it is clear from reading the “entire” passage, that the majority’s interpretation is dubious, because Livermore speaks in terms of enacting in “permanent form” “a rule which [the Legislature] has the power itself to enact as a law, but which [as such] might be of only temporary effect.” (Livermore v. Waite, supra, 102 Cal. at p. 119.) What is at issue in this case is an alteration in the Constitution that the Legislature would have no power to enact, and is therefore fundamentally distinguishable from the type of amendment contemplated by Livermore in the above passage.
In Californians for an Open Primary v. McPherson (2006) 38 Cal.4th 735 [43 Cal.Rptr.3d 315, 134 P.3d 299] (Californians for an Open Primary), I attempted in my concurring opinion to explain why the Legislature was subject to the requirement of article XVIII, section 1 of the California Constitution that when the Legislature proposes an amendment of the state Constitution, “[e]ach amendment shall be so prepared and submitted that it can be voted on separately,” while the Legislature is not subject to the separate-vote requirement when it submits a constitutional revision to the electorate as per Proposition 7. In accounting for this seeming incongruity, I reasoned that one of the primary purposes of the separate-vote requirement was to prevent “logrolling." (Californians for an Open Primary, supra, 38 Cal.4th at p. 789 (conc. opn. of Moreno, J.).) I further reasoned that the danger of logrolling was significantly diminished in the case of an authentic constitutional revision because “[a] constitutional revision, by its very nature and purpose—systematic, comprehensive constitutional renovation and reform—appears to be inherently contrary to the practice of logrolling motivated by political expediency.” (Id. at p. 790.) The majority cites part of the above statement to suggest that I endorsed a view that a constitutional revision consists only of “ ‘systematic, comprehensive constitutional renovation and reform.’ ” (Maj. opn., ante, at p. 426, italics omitted.) But when taken in context, it is clear that all that was intended was that one aspect of a legitimate constitutional revision is that it not be used to circumvent the separate-vote rule and engage in logrolling, and that historically the Legislature has not used the revision process in that manner. (38 Cal.4th at pp. 790-791 (conc. opn. of Moreno, J.).) Nothing in my concurring opinion in Californians for an Open Primary considers whether depriving a suspect class of a fundamental right may be accomplished through a constitutional amendment.
Article I, section 13 of the California Constitution follows closely the text of the Fourth Amendment to the United States Constitution, stating: “The right of the people to be secure in their persons, houses, papers, and effects against unreasonable seizures and searches may not be violated; and a warrant may not issue except on probable cause, supported by oath or affirmation, particularly describing the place to be searched and the persons and things to be seized.”
The majority also cites in support Proposition 14, a state constitutional amendment adopted in 1964 that repealed a statutory provision barring racial discrimination in the sale or rental of housing. As the majority states: “Although Proposition 14 subsequently was held invalid under the federal Constitution (Mulkey v. Reitman (1966) 64 Cal.2d 529 [50 Cal.Rptr. 881, 413 P.2d 825], affd. sub nom. Reitman v. Mulkey (1967) 387 U.S. 369 [18 L.Ed.2d 830, 87 S.Ct. 1627]), [it] was [not] found to constitute an impermissible constitutional revision under the state Constitution.” (Maj. opn., ante, at pp. 447-448, italics omitted.) But in Mulkey v. Reitman, Proposition 14 was not even challenged on the ground that it constituted an improper revision of the California Constitution, and its patent violation of the United States Constitution made such a challenge unnecessary. If “an opinion is not authority for a proposition not therein considered” (Ginns v. Savage (1964) 61 Cal.2d 520, 524, fn. 2 [39 Cal.Rptr. 377, 393 P.2d 689]), the fact that certain arguments were not raised at all carries even less weight. Moreover, as the majority acknowledges, the issue was raised in the related case of Hill v. Miller (1966) 64 Cal.2d 757 [51 Cal.Rptr. 689, 415 P.2d 33]. (Maj. opn., ante, at p. 448, fn. 26.)
In emphasizing the limits of Frierson, I do not in any sense call into question the constitutionality of California’s death penalty law. Rather, I share Justice Mosk’s view that People v. Anderson (1972) 6 Cal.3d 628 [100 Cal.Rptr. 152, 493 P.2d 880], which held that the death penalty violated the state’s constitutional prohibition against cruel or unusual punishment, was erroneously decided. (Frierson, supra, 25 Cal.3d at p. 189 (conc. opn. of Mosk, J.).) I therefore find it unnecessary to address the argument of some of the petitioners that the state’s cruel or unusual punishment clause is distinguishable from the equal protection clause because the former is not as inherently countermajoritarian as the latter and, therefore, may be amended by initiative.
In Korematsu v. United States (1944) 323 U.S. 214 [89 L.Ed. 194, 65 S.Ct. 193], Justice Jackson in dissent decried how the court’s carefully limited opinion in Hirabayashi v. United States (1943) 320 U.S. 81 [87 L.Ed. 1774, 63 S.Ct. 1375] sustaining an order imposing a curfew on Japanese-Americans had led the court to uphold the internment of Japanese-Americans, stating: “[I]n spite of our limiting words we did validate a discrimination on the basis of ancestry for mild and temporary deprivation of liberty. Now the principle of racial discrimination is pushed from support of mild measures to very harsh ones, and from temporary deprivations to indeterminate ones.” (Korematsu v. United States, supra, at p. 247 (dis. opn. of Jackson, J.).) Justice Jackson observed that once a judicial opinion establishes a principle, “[t]he principle then lies about like a loaded weapon .... All who observe the work of courts are familiar with what Judge Cardozo described as ‘the tendency of a principle to expand itself to the limit of its logic.’ ” (Id. at p. 246, fn. omitted.)
In Romer v. Evans (1996) 517 U.S. 620 [134 L.Ed.2d 855, 116 S.Ct. 1620], the high court invalidated on equal protection grounds an amendment to the Colorado Constitution that would have prohibited the enactment of any law designed to protect homosexuals, repeating Justice Harlan’s admonition in his dissent in Plessy v. Ferguson (1896) 163 U.S. 537, 559 [41 L.Ed. 256, 16 S.Ct. 1138], that the Constitution “neither knows nor tolerates classes among citizens” and adding: “It is not within our constitutional tradition to enact laws of this sort. . . . ‘ “Equal protection of the laws is not achieved through indiscriminate imposition of inequalities." ’ [Citation.] Respect for this principle explains why laws singling out a certain class of citizens for disfavored legal status or general hardships are rare.” (Romer v. Evans, supra, 517 U.S. at p. 633.)
The majority quotes dicta in the decision in Varnum v. Brien that recognizes that “the power of the constitution flows from the people, and the people of Iowa retain the ultimate power to shape it over time.” (Varnum v. Brien, supra, 763 N.W.2d 862, 876.) The majority gleans from the Iowa court’s citation of a provision authorizing amendments to the Iowa Constitution that “even as the Iowa high court emphatically declared in Varnum v. Brien that a statute limiting marriage to opposite-sex couples violated a fundamental principle embodied in the constitution of that state, the court at the same time acknowledged the ultimate power of the people to alter the content of the state constitution through a constitutional amendment.” (Maj. opn., ante, at p. 462, fn. omitted.)
It is not remarkable that the Iowa Supreme Court recognized that the people retain the ultimate power to shape the constitution. As I stated above, “[t]here is no doubt that the ultimate authority over the content of the California Constitution lies with the people.” (Ante, at p. 487.) And even if we assume that the Iowa court’s citation of a provision authorizing amendments to the Iowa Constitution was intended to express the view that its own decision regarding marriage equality could be overturned by constitutional amendment, that dicta has no bearing on whether Proposition 8 was a proper amendment to the California Constitution, because the process for amending the Iowa Constitution differs substantially from the process for amending the California Constitution. In Iowa, the people cannot directly initiate a constitutional amendment, but can only vote on an amendment after it has been approved by the legislature, then reapproved by a new legislature after the next general election. (See Iowa Const., art. X, § 1.) The Iowa Constitution can only be revised through a constitutional convention. (Id., § 3.) The procedure for amending the Iowa Constitution, therefore, resembles one of the procedures for revising the California Constitution, requiring approval both by more than a simple majority of the legislature (in California by a two-thirds majority, in Iowa by a majority of two successive legislatures) and by a majority of the people. Accordingly, the above quoted passage from Varnum, even when read expansively, does not support the majority’s position that a simple majority of the electorate can amend the California Constitution to deprive a suspect class of a fundamental right.