Concurring.—The question presented in this case is “whether the Legislature may pose to the electorate a single advisory question concerning the People’s support for a federal constitutional amendment.” (Maj. opn., ante, at p. 504.) The court answers yes to this narrow question and goes no further, “reserv[ing] for another day whether, in support of other powers not implicated here, an advisory ballot measure would be a permissible means of legislative investigation.” (Id. at p. 504, fn. 11.) With this limited holding, I concur.
*552The court’s reasoning in support of this result includes an informative survey of relevant constitutional history, and I agree that a “ ‘page of history is worth a volume of logic’ ” where, as here, the issue presented is novel and implicates basic questions about governmental structure and constitutional change. (Maj. opn., ante, at p. 505, quoting New York Trust Co. v. Eisner (1921) 256 U.S. 345, 349 [65 L.Ed. 963, 41 S.Ct. 506]; cf. National Labor Relations Bd. v. Noel Canning (2014) 573 U.S. _, _ - _ [189 L.Ed.2d 538, 134 S.Ct. 2550, 2559-2560] (Noel Canning).) But today’s opinion goes astray when it invokes or, more accurately, expands the Legislature’s investigative power in order to uphold the advisory ballot measure at issue here. (Maj. opn., ante, at pp. 498-500.) This theory is unnecessarily broad and could be read as an invitation for the Legislature to test the waters with future advisory ballot measures on a wide range of issues having nothing to do with a federal constitutional amendment. The Chief Justice would make that invitation explicit. In her view, the Legislature may use advisory ballot measures “in order to obtain the voters’ policy views with regard to any potential action that the Legislature has authority to undertake.” (Conc. opn. of Cantil-Sakauye, C. J., ante, at p. 542; accord, conc. opn. of Corrigan, J., ante, at pp. 550-551.) As I explain below, such wide-ranging use of advisory ballot measures is not authorized by our state Constitution and would potentially reshape the way electoral politics and policymaking are conducted in California. We should not take liberties with the careful structure of republican democracy that the framers of our Constitution have built and bequeathed to us.
I would not rely on the Legislature’s investigative power in reaching today’s result. Senate Bill No. 1272 (2013-2014 Reg. Sess.) is an effort by the Legislature to marshal the solemn voice of the people of California in support of a federal constitutional amendment. Article V of the United States Constitution assigns state legislatures a special role in facilitating and promoting constitutional change. I would simply hold that Senate Bill No. 1272 is a reasonable exercise of the Legislature’s implied power under article V.
I.
Senate Bill No. 1272 (2013-2014 Reg. Sess.) is titled the Overturn Citizens United Act. (Stats. 2014, ch. 175, § 1.) In enacting this statute, the Legislature declared that the United States Supreme Court’s decision in Citizens United v. Federal Election Comm’n (2010) 558 U.S. 310 [175 L.Ed.2d 753, 130 S.Ct. 876] “presents a serious threat to self-government by rolling back previous bans on corporate spending in the electoral process and allows unlimited corporate spending to influence elections, candidate selection, policy decisions, and public debate.” (Stats. 2014, ch. 175, § 2, subd. (e).) It further declared that “Article V of the United States Constitution *553empowers and obligates the people of the United States of America to use the constitutional amendment process to correct those egregiously wrong decisions of the United States Supreme Court that go to the heart of our democracy and the republican form of self-government.” (Id., subd. (l).) The statute directed the Secretary of State to submit to the voters an advisory question (Proposition 49) asking whether Congress should propose and whether the Legislature should ratify a constitutional amendment overturning Citizens United, and to report the results to Congress. (Stats. 2014, ch. 175, § 4, subds. (a), (b).) As all members of the court agree, the substantive merits of Citizens United and Proposition 49 are irrelevant to the question in this case.
Amending the federal Constitution is a difficult task, successfully accomplished only 27 times in our nation’s history. To secure approval for a constitutional amendment, a political movement must convince an extraordinary number of citizens to take the movement’s aims more seriously than they do most issues of ordinary government. This was by design. By making the process of constitutional change more ‘“unwieldy and cumbrous” than ordinary lawmaking (Barron v. The Mayor and City Council of Baltimore (1833) 32 U.S. 243, 250 [8 L.Ed. 672]), article V of the federal Constitution serves as a bulwark against the whims of bare legislative majorities and ensures that rules entrenched in the ‘“supreme law of the land” (U.S. Const., art. VI) represent the considered and collective judgments of the people of the United States. Senate Bill No. 1272 (2013-2014 Reg. Sess.) is a statute enacted in furtherance of the federal constitutional amendment process.
A.
Article V of the federal Constitution (article V) provides in relevant part: ‘“The Congress, whenever two thirds of both houses shall deem it necessary, shall propose amendments to this Constitution, or, on the application of the legislatures of two thirds of the several states, shall call a convention for proposing amendments, which, in either case, shall be valid to all intents and purposes, as part of this Constitution, when ratified by the legislatures of three fourths of the several states, or by conventions in three fourths thereof, as the one or the other mode of ratification may be proposed by the Congress . . . .” As this text makes clear, article V assigns key roles to state legislatures at the proposal and ratification stages of the federal constitutional amendment process.
Challenges to the manner in which state legislatures have exercised their article V powers have almost always been rejected. In Hawke v. Smith, No. 1 (1920) 253 U.S. 221 [64 L.Ed. 871, 40 S.Ct. 495] (Hawke), the Ohio General Assembly adopted a resolution ratifying the proposed Eighteenth Amendment *554to the federal Constitution. A voter then filed a referendum petition calling for a special election to approve or reject the amendment. A different voter brought suit to enjoin state officials from complying with the referendum petition. The high court held that the referendum could not proceed, explaining: “The only question really for determination is: What did the framers of the Constitution mean [in article V] in requiring ratification by 'Legislatures''! That was not a term of uncertain meaning when incorporated into the Constitution. What it meant when adopted it still means for the purpose of interpretation.” (Hawke, at p. 227.) Subjecting the ratification decision to a referendum would violate article V’s allocation of ratification authority to state legislatures. (Hawke, at p. 231; accord, Barlotti v. Lyons (1920) 182 Cal. 575, 577 [189 P. 282]; Prior v. Noland (1920) 68 Colo. 263 [188 P. 729, 731].)
In Leser v. Garnett (1922) 258 U.S. 130 [66 L.Ed. 505, 42 S.Ct. 217] (Leser), the plaintiff argued that the Nineteenth Amendment to the federal Constitution was invalid because the legislatures of several states had violated state constitutional provisions in ratifying the proposed amendment. The high court disagreed on the ground that state legislatures have authority to exercise their article V functions as they see fit, free from interference by the “limitations sought to be imposed by the people of a State.” (Leser, at p. 137.)
More recent decisions are in accord. Walker v. Dunn (Tenn. 1972) 498 S.W.2d 102 (Walker) involved a provision of the Tennessee Constitution prohibiting the state legislature from acting on any federal constitutional amendment unless the legislature had been elected after submission of the amendment. The Tennessee Supreme Court upheld the state legislature’s ratification of the Twenty-sixth Amendment, even though the legislature did not wait until the election cycle following submission. The court concluded that the state constitutional provision was “a limitation upon the General Assembly of Tennessee in the exercise of its federally derived power” and accordingly was invalid. (Walker, at p. 106; see Trombetta v. Florida (M.D.Fla. 1973) 353 F.Supp. 575, 577-578 (Trombetta) [invalidating similar Fla. constitutional provision].)
In Dyer v. Blair (N.D.Ill. 1975) 390 F.Supp. 1291 (Dyer), the court considered a provision of the Illinois Constitution, as well as rules adopted by Illinois’s legislature, that required a three-fifths majority vote to ratify an amendment to the federal Constitution. Each house of the legislature had approved the proposed Equal Rights Amendment by a vote of more than a majority but less than a three-fifths supermajority. The court observed that the framers of the federal Constitution had a “basic understanding that state legislatures should have the power and the discretion to determine for *555themselves how they should discharge the responsibilities committed to them by the federal government.” (Dyer, at p. 1307.) The court then explained that because the ‘“delegation [in article V] is not to the states but rather to the designated ratifying bodies,” state constitutional restrictions on the legislature’s decisionmaking were invalid. (Id. at p. 1308.) The court held, however, that the majority vote in favor of ratifying the Equal Rights Amendment was insufficient because each house had concluded for itself that only a 60 percent supermajority would suffice. (Dyer, at pp. 1308-1309.)
Subtler attempts at cabining the discretion of the state legislatures when performing their article V functions have also been rejected. In American Federation of Labor v. Eu (1984) 36 Cal.3d 687 [206 Cal.Rptr. 89, 686 P.2d 609], we considered various state and federal challenges to an initiative advocating a federal balanced budget amendment. Among other provisions, the initiative proposed to suspend the compensation and benefits of state legislators who would not take specific actions to support a balanced budget amendment. (Id. at pp. 692-693.) We invalidated this aspect of the initiative on the ground that it would coerce state legislators to take actions in support of a particular constitutional amendment. (Id. at p. 694.) Similarly, in Bromberg v. Jones (1999) 20 Cal.4th 1045 [86 Cal.Rptr.2d 319, 978 P.2d 1240] (Bromberg), we considered whether a voter initiative could require future ballots for the United States Congress to include the statement “ ‘Disregarded Voters’ Instruction on Term Limits’ ” next to the names of incumbents who did not support a federal term limits amendment in the previous session. (Id. at p. 1047.) We held that the initiative was “impermis-sibly coercive” and thus violated article V. (Bromberg, at p. 1060; see id. at p. 1063.)
In addition to the powers that article V expressly delegates, state legislatures have played a crucial role in achieving the popular mobilization necessary to ratify many of the amendments to the federal Constitution by issuing resolutions to Congress and to other states. Today’s opinion recounts numerous examples throughout our nation’s history, dating back to the founding era. (Maj. opn., ante, at pp. 502-504.)
State legislatures have also established state conventions when Congress has chosen that method of ratification. As today’s opinion recounts, when Congress submitted the Twenty-first Amendment to state conventions, state legislatures across the country enacted legislation establishing how delegates were to be chosen and when and where conventions would meet. (Brown, Ratification of the Twenty-first Amendment to the Constitution of the United States (1938) pp. 521-700 [collecting laws]; see Stats. 1933, ch. 149, pp. 598-602 [establishing procedures for Cal. convention to ratify the 21st Amend.].) Courts have found these actions to be within the legislatures’ *556broad discretion to act in connection with the federal constitutional amendment process. (See State ex rel. Donnelly v. Myers (1933) 127 Ohio St. 104 [186 N.E. 918] (Myers); State ex rel. Tate v. Sevier (1933) 333 Mo. 662 [62 S.W.2d 895, 898] (Sevier).)
From the discussion above, it is evident that judicial decisions in this area have given state legislatures wide latitude in carrying out their article V functions and taking actions reasonably related to the federal constitutional amendment process. This judicial posture befits the nature of constitutional change. Occasions for amending the federal Constitution are, by design, infrequent and unusual. Each is a separate and solemn moment that requires a political process calibrated to the perceived problem at hand. Courts have recognized that the actors to whom article V delegates authority have broad discretion to address each proposal for constitutional change on an individualized basis. History and precedent suggest a very narrow role for the judiciary in monitoring the federal amendment process.
B.
The texts of the federal and state Constitutions do not address whether the Legislature may place an advisory measure on the ballot asking voters whether they support a federal constitutional amendment. Although we have previously recognized that such an advisory ballot measure “does not offend article V” (American Federation of Labor v. Eu, supra, 36 Cal.3d at p. 707), no court has confronted the precise question before us. But the historical use of advisory questions in connection with the federal constitutional amendment process sheds light on the issue here. (See Dyer, supra, 390 F.Supp. at pp. 1303-1307 [consulting historical practice to illuminate the scope of state legislatures’ art. V powers].)
The practice of state legislatures consulting the voters on federal constitutional amendments must be understood in historical context. Since the founding, “the animating principle of our Constitution [is] that the people themselves are the originating source of all the powers of government.” (Arizona State Legislature v. Arizona Independent Redistricting Comm’n (2015) 576 U.S. _, _[192 L.Ed.2d 704, 135 S.Ct. 2652, 2671].) Because constitutional commitments bind future legislative majorities, cannot be easily undone, and speak on behalf of “the People of the United States” (U.S. Const, preamble), the framers repeatedly emphasized the necessity of the people’s consent in amending the federal Constitution.
At the state conventions for ratifying the federal Constitution, advocates for ratification repeatedly invoked the sovereign will of the people as the sole ground for amending its terms. In Pennsylvania, James Wilson remarked: *557“This Constitution . . . opens with a solemn and practical recognition . . . :— ‘We, the people of the United States, in order to form a more perfect union, establish justice, &c., do ordain and establish this Constitution for the United States of America.’ It is announced in their name—it receives its political existence from their authority: they ordain and establish. What is the necessary consequence? Those who ordain and establish have the power, if they think proper, to repeal and annul.” (2 Elliot, The Debates in the Several State Conventions on the Adoption of the Federal Constitution (1881) pp. 434-435 (Debates).) In North Carolina, James Iredell similarly observed: “[Ajltera-tions can without difficulty be made, agreeable to the general sense of the people. . . . Any amendments which either Congress shall propose, or which shall be proposed by such general convention, are afterwards to be submitted to the legislatures of the different states, or conventions called for that purpose, as Congress shall think proper .... By referring this business to the legislatures, expense would be saved; and in general, it may be presumed, [that] they would speak the genuine sense of the people.” (4 Elliot, Debates, at p. 177; see 3 Elliot, Debates, at p. 37 [statement of Edmund Pendleton at the Va. ratifying convention].) It is against this backdrop that state legislatures have regularly used advisory ballot measures in connection with the federal constitutional amendment process.
Today’s opinion ably recounts the history of such advisory measures, with examples from a multitude of states concerning a wide range of proposed constitutional amendments from the late nineteenth century to the present day. (Maj. opn., ante, at pp. 507-511.) Challenges to these measures have not succeeded. (Id. at pp. 510-511.) As the court recognizes (id. at p. 505), it is appropriate in this case to “put significant weight upon historical practice” (Noel Canning, supra, 573 U.S. at p. _ [134 S.Ct. at p. 2559], italics omitted)).
This robust body of shared historical practice suggests a common source of legislative authority. As noted, article V of the federal Constitution gives state legislatures a central role in the process of proposing and ratifying federal constitutional amendments. Under article V, the function of state legislatures is to convey to Congress and to other states an expression of the state’s sovereign will with respect to federal law of the most fundamental character. Sovereignty ultimately lies in “the supreme authority in each State, the authority of the people themselves” (The Federalist No. 39 (Cooke ed., 1961) p. 254 (Madison)), and the ratification function assigned to state legislatures under article V is intended to elicit “a decisive expression of the people’s will” (Dillon v. Gloss (1921) 256 U.S. 368, 374 [65 L.Ed. 994, 41 S.Ct. 510] (Dillon)). In submitting an advisory ballot measure to the electorate, a state legislature acts in furtherance of its article V function by vesting the people’s will with a degree of solemnity that cannot be achieved through an opinion poll or other means, and by ensuring and communicating to others that its *558support or opposition to a proposed amendment accurately reflects the people’s will. (Maj. opn., ante, at pp. 518-520.) It is no accident that the use of advisory ballot measures by state legislatures has historically clustered around proposals for amending the federal Constitution. “This past use of advisory questions to inform the federal constitutional process evidences a larger truth—a recognition of the particular appropriateness of consulting the polity in the course of exercising independent judgment with respect to such foundational matters.” (Id. at p. 519.) Because it is particularly appropriate for a state legislature exercising its article V function to enlist the people’s voice as an aid to expressing the state’s sovereign will, the legislature’s use of an advisory measure like Proposition 49 is properly understood as an exercise of implied power under article V.
The concept of implied powers under the federal Constitution has been well established since M’Culloch v. Maryland (1819) 17 U.S. 316 [4 L.Ed 579] (M’Culloch), which held (among other things) that Congress had the power to establish a national bank. (Id. at p. 424.) Although M’Culloch is often cited for its interpretation of the necessary and proper clause (id. at pp. 411-421), the high court’s discussion of that clause appears only after the opinion has already made an affirmative case for Congress’s implied power to create the bank (id. at pp. 401-411). M’Culloch’s primary arguments for the constitutionality of the national bank are based not on the necessary and proper clause but on the nature of the federal Constitution itself.
Chief Justice Marshall began the court’s opinion by noting that although the Constitution does not enumerate a power to establish a bank, “there is no phrase in the instrument which, like the articles of confederation, excludes incidental or implied powers; and which requires that everything granted shall be expressly and minutely described. . . . The men who drew and adopted [the 10th] amendment had experienced the embarrassments resulting from the insertion of [the word ‘necessary’] in the articles of confederation, and probably omitted it to avoid those embarrassments. A constitution . . . requires, that only its great outlines should be marked, its important objects designated, and the minor ingredients which compose those objects, be deduced from the nature of the objects themselves.” (M’Culloch, supra, 17 U.S. at pp. 406-407.) Unlike the Articles of Confederation, the second article of which had narrowly limited the powers of the Congress of the Confederation to those “expressly delegated,” the 1787 Constitution was meant to constitute a new system of government and to provide only the broad outlines of the powers of its component parts.
M’Culloch went on to say that although the word “bank” or “incorporation” does not appear in the federal Constitution, the enumerated powers of *559Congress include the powers “to lay and collect taxes; to borrow money; to regulate commerce; to declare and conduct a war; and to raise and support armies and navies.” (M’Culloch, supra, 17 U.S. at p. 407.) The court then explained that “it may with great reason be contended, that a government, entrusted with such ample powers, on the due execution of which the happiness and prosperity of the nation so vitally depends, must also be entrusted with ample means for their execution. The power being given, it is the interest of the nation to facilitate its execution. It can never be their interest, and cannot be presumed to have been their intention, to clog and embarrass its execution, by withholding the most appropriate means. . . . [¶] It is not denied, that the powers given to the government imply the ordinary means of execution.” (Id. at pp. 408-409.) M’Culloch’s elucidation of implied powers was based on “general reasoning” about the nature of a constitution. (Id. at p. 411; see id. at p. 407 [“we must never forget, that it is a constitution we are expounding”].) Only after upholding Congress’s implied power to create the bank did the opinion turn to refuting the objection that the necessary and proper clause is a limitation on, rather than a grant of, congressional power. (Id. at p. 412.)
M’Culloch’s reasoning is not limited to powers granted in article I of the federal Constitution. M’CuUoch rests on a general principle that each power enumerated in the federal Constitution may imply other powers necessary to the exercise of the enumerated power. (See The Federalist No. 44, supra, at pp. 304-305 (Madison) [“No axiom is more clearly established in law, or in reason, than that wherever the end is required, the means are authorized; wherever a general power to do a thing is given, every particular power necessary for doing it, is included.”].) Subsequent decisions have confirmed that this principle of constitutional construction applies to powers other than those assigned to Congress. (See, e.g., Chambers v. NASCO, Inc. (1991) 501 U.S. 32, 43-44 [115 L.Ed.2d 27, 111 S.Ct. 2123] [discussing implied powers of the federal courts]; American Ins. Assn. v. Garamendi (2003) 539 U.S. 396, 414-415 [156 L.Ed.2d 376, 123 S.Ct. 2374] [discussing implied powers of the President].)
Like articles I, II, and III of the federal Constitution, article V also conveys power as much through “what is reasonably implied” as through “what is expressed.” (Dillon, supra, 256 U.S. at p. 373; see id. at p. 376 [“As a rule the Constitution speaks in general terms, leaving Congress to deal with subsidiary matters of detail as the public interests and changing conditions may require; and Article V is no exception to the rule.” (fn. omitted)].) In the article V context, courts have recognized a variety of implied powers. The high court has held that Congress has the power to place a time limit on ratification of a proposed amendment by the states. (Dillon, at pp. 375-376.) And as noted, when Congress chose the state convention method for ratifying the Twenty-first Amendment, legislatures across the country established the *560mechanisms by which those conventions would be constituted. In challenges to such actions, state courts held that their legislatures had implied power under article V to establish conventions when Congress elects the convention method of ratification. (See Sevier, supra, 62 S.W.2d at p. 898; Myers, supra, 186 N.E. at p. 918.)
Here the Legislature’s briefing contends that ‘“[f]or the people of this state, Proposition 49 represents their only means under article V of the federal Constitution to be heard on this momentous question [i.e., whether Citizens United should be overturned by constitutional amendment]. Elections on advisory ballot questions have been held in a wide variety of other states to ascertain and to formally convey the will of the voters as to whether the U.S. Constitution should be amended in various respects, and the courts have repeatedly upheld the submission of such measures to the electorate. The California Legislature is likewise entitled to seek this input from the voters as a matter ‘incidental and ancillary’ to its constitutional power and responsibility to ratify proposed amendments to the U.S. Constitution.” I would hold that the Legislature has implied power under article V to submit Proposition 49 to the electorate because the use of an advisory ballot measure uniquely serves to aid the Legislature’s expression of California’s sovereign will.
As I explain in the rest of this opinion, the Legislature does not have general authority under the California Constitution to submit advisory questions to the electorate. But even if this means the Legislature has no authority under state law to submit an advisory measure on a federal constitutional amendment (see dis. opn., post), such a limitation must yield to the federal authority that article V confers on the Legislature. As discussed above (ante, conc. opn. of Cantil-Sakauye, C. J., at pp. 553-555), courts have consistently invalidated state law limitations, including state constitutional limitations, on a state legislature’s exercise of its federal functions under article V. (See Leser, supra, 258 U.S. at p. 137; Hawke, supra, 253 U.S. at p. 231; Dyer, supra, 390 F.Supp. at pp. 1307-1308; Trombetta, supra, 353 L.Supp. at pp. 577-578; Walker, supra, 498 S.W.2d at p. 106.)
Because article V’s ‘“delegation is not to the states but rather to the designated ratifying bodies” (Dyer, supra, 390 L.Supp. at p. 1308), state legislatures ‘“have the power and the discretion to determine for themselves how they should discharge the responsibilities committed to them by the federal government” (id. at p. 1307). It may be that certain elemental precepts of state law—for example, state constitutional provisions that create a legislature and define its membership—must remain operative when a state legislature exercises federal functions. But this case does not concern the character or legitimacy of the Legislature as the duly constituted and properly functioning legislature of California. This case concerns the source and *561validity of a substantive power that the Legislature seeks to exercise, and as this court has recognized, “a state may not, through restrictions imposed by state law, interfere with a state legislature’s ability to fulfill its function and responsibilities as contemplated by Article V of the federal Constitution.” (Bramberg, supra, 20 Cal.4th at p. 1058; see Leser, supra, 258 U.S. at p. 137 [a state legislature exercising ‘“a federal function derived from the Federal Constitution . . . transcends any limitations sought to be imposed by the people of a State”].) Accordingly, even if state law does not authorize the Legislature to submit Proposition 49 to the electorate, the Legislature may do so as an exercise of its implied power under article V.
II.
The analysis above affirms the Legislature’s broad latitude to act in the unique context of amending the federal Constitution and amply justifies today’s narrow holding. Yet the court’s opinion rests not only on the Legislature’s role under article V but also on ‘“the Legislature’s power to investigate.” (Maj. opn., ante, at p. 520.) This doctrinal move unnecessarily calls into question the narrowness of today’s holding—for if investigation by use of an advisory ballot measure ‘“is permitted as a necessary aid to the execution of other legislative powers” (id. at p. 499), and if an advisory ballot measure is permissible so long as ‘“a nexus exists between the matter investigated and some potential action the Legislature has authority to undertake” (id. at p. 500), then what is to distinguish the validity of an advisory ballot measure concerning a federal constitutional amendment from the validity of such a measure concerning any ordinary issue of public policy? Although the court expresses no view on that question (id. at p. 504, fn. 11), the Chief Justice says there is no distinction at all. She would hold that the Legislature may use advisory ballot measures ‘“with regard to any potential action that the Legislature has authority to undertake,” including matters ‘“completely unconnected with any article V issue.” (Conc. opn. of Cantil-Sakauye, C. J., ante, at pp. 542, 544.)
I find unpersuasive the court’s reliance on the Legislature’s investigative power as well as the Chief Justice’s expansive view of the Legislature’s power to use advisory ballot measures. At the outset, I acknowledge that ‘“[ujnlike the federal Constitution, which is a grant of power to Congress, the California Constitution is a limitation or restriction on the powers of the Legislature.” (Methodist Hosp. of Sacramento v. Saylor (1971) 5 Cal.3d 685, 691 [97 Cal.Rptr. 1, 488 P.2d 161].) ‘“The legislature is vested with the whole of the legislative power of the state and may deal with any subject within the scope of civil government unless it is restrained by the provisions of the constitution, and the presumption that the legislature is acting within the constitution holds good until it is made to appear in what particular it is *562violating constitutional limitations.” (Macmillan Co. v. Clarke (1920) 184 Cal. 491, 496-497 [194 P. 1030].) But limitations on the Legislature’s power need not appear in the state Constitution “expressly”; they may exist “by necessary implication.” (Methodist Hospital, at p. 691.) Below I explain why it is a necessary implication of our Constitution’s text and history that the Legislature lacks general authority to submit advisory measures to the electorate. Whether exercising its investigative power or any other power under state law, the Legislature may not statutorily alter the structure of government established by our Constitution.
A.
We previously addressed the issue of advisory ballot measures in American Federation of Labor v. Eu, supra, 36 Cal.3d 687, where concerned citizens sought to place on the ballot a nonbinding resolution urging Congress to propose a federal balanced budget amendment and directing the Secretary of State to apply for a constitutional convention. We invalidated this advisory ballot measure on the ground that it was not authorized by the initiative power. (Id. at pp. 714-715.) Observing that the initiative “is the power of the electors to propose statutes and amendments to the Constitution and to adopt or reject them” (Cal. Const., art. II, § 8, subd. (a)), we explained that a measure that merely adopts a resolution or declaration of policy is one that “fails to adopt a statute” (American Federation of Labor v. Eu, at p. 715).
This precedent makes clear that a citizen-initiated advisory measure asking the electorate whether the Legislature should increase the gasoline tax, for example, would be unconstitutional. The court in American Federation of Labor v. Eu reasoned that the initiative power does not authorize the people to place an advisory measure on the ballot. But upon a moment’s reflection, it is evident that our decision suggests a deeper structural principle.
Suppose citizens who support an increase in the gasoline tax qualify the following initiative for the general election ballot: “Proposition X. The People of the State of California hereby enact the following statute: In the next general election, the Secretary of State shall place on the ballot an advisory question asking the people of California whether the Legislature should increase the tax on retail sales of gasoline by 5 cents per gallon.” If Proposition X were to pass, would the statute it enacted be constitutional? Of course not; otherwise, our decision in American Federation of Labor v. Eu would be easily evaded and reduced to a virtual nullity. But the invalidity of Proposition X does not arise from the fact that it is an advisory ballot measure like the advisory initiative at issue in American Federation of Labor v. Eu. Proposition X is not an advisory measure; it enacts a statute that directs the Secretary of State to place an advisory measure on the ballot at the next election.
*563The invalidity of Proposition X must be explained by reference to its substance, not its form. Although Proposition X enacts a statute, the statute it enacts asks voters to do something they lack authority to do: to adopt a resolution with no binding legal consequence in their official capacity as the people of California. We recognized this principle in American Federation of Labor v. Eu when we said the provisions of the balanced budget initiative “adopt, and mandate the Legislature to adopt, a resolution which does not change California law and constitutes only one step in a process which might eventually amend the federal Constitution. Such a resolution is not an exercise of legislative power reserved to the people under the California Constitution.” (American Federation of Labor v. Eu, supra, 36 Cal.3d at p. 694.) In other words, the California Constitution reserves to the people the power to enact laws by initiative or referendum, but it does not reserve to the people the power to adopt nonbinding resolutions.
If Proposition X is invalid, then an identical statute enacted by the Legislature would be invalid as well because “[t]he electorate’s legislative power is ‘generally coextensive with the power of the Legislature to enact statutes.’ ” (Professional Engineers in California Government v. Kempton (2007) 40 Cal.4th 1016, 1042 [56 Cal.Rptr.3d 814, 155 P.3d 226]; see Santa Clara County Local Transportation Auth. v. Guardino (1995) 11 Cal.4th 220, 253 [45 Cal.Rptr.2d 207, 902 P.2d 225]; Legislature v. Deukmejian (1983) 34 Cal.3d 658, 675 [194 Cal.Rptr. 781, 669 P.2d 17].) In resisting this conclusion, the court cites our recognition that “ ‘the reserved powers of initiative and referendum do not encompass all possible actions of a legislative body.’ ” (Maj. opn., ante, at p. 516, quoting American Federation of Labor v. Eu, supra, 36 Cal.3d at p. 708.) But we made that statement in the context of explaining that the Legislature, but not the citizenry, has the authority to adopt or reject “a resolution which merely expresses the wishes of the enacting body.” (American Federation of Labor v. Eu, at p. 708.) Nothing in American Federation of Labor v. Eu suggests that the Legislature has authority that the citizenry lacks to place an advisory measure on the ballot.
Distinguishing Proposition X from its legislatively enacted twin would require us to hold that our state Constitution does not authorize the citizenry, on its own initiative, to vote on an advisory measure, but does authorize the citizenry to do so when directed by the Legislature. Yet it seems dubious to say that our Constitution privileges the ordinary lawmaking process over the initiative qualification process as a gatekeeper for the people’s exercise of their putative advisory voice. After all, the Legislature is the agent of the people, not the other way around.
As I explain more fully below, the infirmity of Proposition X is not that the electorate, as opposed to the Legislature, has authorized the electorate to vote *564on a resolution. It is that the people, having reserved to themselves only the powers of initiative and referendum while vesting in the Legislature all other “legislative power of this State” (Cal. Const., art. IV, § 1), have no constitutional authority to adopt a resolution. Such authority cannot be conferred by statute, whether enacted by the people or by their representatives.
B.
When the California Constitution of 1849 was ratified, article IV, section 1 provided: “The legislative power of this State shall be vested in a Senate and Assembly, which shall be designated the Legislature of the State of California .. . .” This language affirmed what the people had accomplished by forming the Legislature: The people gave up the “legislative power of this State” and vested it in their duly elected representatives. (Ibid.) The only exceptions were provided expressly: The people retained the power to approve or reject proposed constitutional amendments and proposed assumptions of debt. (Cal. Const, of 1849, arts. VIII, X.) By ratifying the 1849 Constitution and creating the Legislature, the people of California established a republic. They divested themselves of authority to exercise certain forms of power, they authorized the Legislature to exercise those powers on their behalf, and they restrained their ability to take those powers back without amending the state Constitution. Most prominently, the people gave up the authority to enact laws and vested that authority in the Legislature. That is why the initiative and referendum had to be established by constitutional amendment in 1911 and why the Legislature does not have authority to submit ordinary statutes for the citizenry to enact by ballot. (See post, at pp. 565-566.)
Apart from the power to enact laws (what I will call “lawmaking power,” as distinguished from the more encompassing term “legislative power”), what powers were included in the people’s grant of “legislative power” to their representatives under article IV, section 1 of the 1849 Constitution? The basic contours of the legislative power are established by its historical roots in the powers of parliament. In Ex parte D. O. McCarthy (1866) 29 Cal. 395 (McCarthy), we explained: “A legislative assembly, when established, becomes vested with all the powers and privileges which are necessary and incidental to a free and unobstructed exercise of its appropriate functions. These powers and privileges are derived not from the Constitution; on the contrary, they arise from the very creation of a legislative body . . . . [¶] What powers and privileges ... a legislative assembly takes by force and effect of its creation, are to be ascertained by a reference to the common parliamentary law.” (Id. at p. 403.) We then cited Luther Stearns Cushing’s 1856 treatise, Elements of the Law and Practice of Legislative Assemblies in the United States of America, for examples of “necessary and incidental” powers. (McCarthy, at pp. 403-404.)
*565As relevant here, one of the traditional powers characteristic of a parliamentary body was the power to adopt nonbinding resolutions. A resolution “expresses the wishes of the enacting body.” (American Federation of Labor v. Eu, supra, 36 Cal.3d at p. 708.) As Cushing explained, “When the house expresses any opinion, with reference to any subject before it, either public or private; or its will to do something at a given time, (not incidental to the ordinary course of business); or declares its adoption of general orders relative to its proceedings; in all these cases, it expresses itself in the form of resolutions.” (Cushing, Elements of the Law and Practice of Legislative Assemblies in the United States, supra, at p. 314, par. 799.) Historically, resolutions served as a means for one chamber of a legislative body to express an opinion directed at another chamber, for a committee to express an opinion directed at the chamber as a whole (id. at p. 778), for the legislature to express an opinion directed at the executive or the “crown” (id. at p. 359, par. 905), or for a state legislature to express an opinion directed at the federal government (see, e.g., 4 Elliot, Debates, supra, at pp. 540-545 [Ky. resolutions of 1798 & 1799]; id. at pp. 528-529 [Va. resolutions of 1798]).
In vesting “[t]he legislative power of this State” in the Legislature (Cal. Const, of 1849, art. IV, § 1), the people gave up their authority to exercise the power to adopt resolutions—that is, to express an opinion in their official capacity as the people of California—and they restrained their authority to take that power back. That power could not be returned to or shared with the people without a state constitutional amendment or some other source of authority such as article V of the federal Constitution. Under the California Constitution of 1849, the people were not authorized to adopt a nonbinding resolution, just as they were not authorized to enact ordinary law.
As the Legislature’s briefing explains, our state Constitution is now more democratic and less republican than it was in 1849. In 1911, amid widespread perception that the Legislature had become corrupt and self-serving, Governor Hiram Johnson and a wave of representatives were elected on the promise of amending the state Constitution to give some power back to the people. (Cal. Com. on Campaign Linancing, Democracy by Initiative: Shaping California’s Lourth Branch of Government (1992) pp. 36-42.) Those representatives proposed, and the people ratified, amendments to the California Constitution that created the broad outlines of our current structure of government. Article IV, section 1 of the California Constitution now provides: “The legislative power of this State is vested in the California Legislature which consists of the Senate and Assembly, but the people reserve to themselves the powers of initiative and referendum.”
Importantly, the departures from republican government introduced in 1911 were specific and circumscribed. “When the people established the Legislature, they conveyed to it the full breadth of their sovereign legislative powers. *566[Citation.] When they adopted the initiative power in 1911, they restored to themselves only a shared piece of that power.” (Maj. opn., ante, at p. 516.) The powers ‘“reserved” to the people did not ‘“encompass all possible actions of a legislative body. Those powers are limited, under article II, to the adoption or rejection of ‘statutes.’. . . [I]t does not include a resolution which merely expresses the wishes of the enacting body . . . .” (American Federation of Labor v. Eu, supra, 36 Cal.3d at p. 708.) After 1911, as before, the people retain no authority to adopt a resolution. (Id. at p. 694 [‘“Such a resolution is not an exercise of legislative power reserved to the people under the California Constitution.”].) That power continues to reside exclusively in the Legislature, and it cannot be restored to the people without a constitutional amendment. The Legislature may not ask the people to vote on an advisory measure because the Legislature may not statutorily authorize the people to exercise a power that the people have constitutionally vested in the Legislature and divested from themselves.
Of course, this does not mean the Legislature may not investigate the views of the citizenry through committee hearings, town halls, opinion polls, meetings with constituents, social media, and myriad other information-gathering mechanisms aided by modern technology. It just means the Legislature may not use the ballot for this purpose. The framers of the 1849 Constitution knew well the ballot’s singular political significance and the importance of circumscribing the purposes for which it could be used. The words ‘“election,” ‘“electors,” and ‘“ballot” appeared in the original charter 69 times. Collectively, these usages contemplated only three ways in which the ballot would be employed: to approve the assumption of state debt (Cal. Const, of 1849, art. VIII), to ratify state constitutional amendments (id., art. X), and to elect constitutional officers. The terms of our state Constitution today continue to authorize use of the ballot for the solemn and legally binding tasks of electing public officials and approving limited and specific types of laws. Nothing in our Constitution contemplates use of the statewide ballot—the traditional means for solemn expression of the people’s will—to gauge public opinion on a ‘“purely precatory” nonbinding resolution. (American Federation of Labor v. Eu, supra, 36 Cal.3d at p. 708.)
In sum, the legislative power—i.e., the full range of parliamentary power, including the power to make laws and the power to adopt resolutions—was vested in the Legislature and divested from the people in the 1849 Constitution. That is why the 1911 amendment to article IV, section 1 of the California Constitution was needed to authorize the people, not just the Legislature, to exercise general lawmaking power. And that is why the people do not have general authority to vote on advisory ballot measures. No constitutional amendment has returned that power to the people or authorized the Legislature to do so.
*567C.
This understanding of how legislative power has been vested, divested, and shared throughout our constitutional history draws further support from the indirect initiative process that existed between 1911 and 1966. In addition to reserving to the people the powers of initiative and referendum, the 1911 amendment to article IV, section 1 of the California Constitution provided for “direct” and “indirect” initiatives, with the “indirect” initiative process authorizing the Legislature to place on the ballot, alongside a voter initiative, an alternative measure for the electorate to approve or disapprove. (Cal. Const., art. IV, former § 1, as amended Oct. 10, 1911 [“The legislature may reject any measure so proposed by initiative petition and propose a different one on the same subject . . . .”].) The specific authorizing language of this provision—“The legislature may . . . propose a different [measure] on the same subject”—would have been unnecessary if the Legislature already had general power to submit ordinary statutes for the people to enact by ballot. The indirect initiative confirms that the people originally vested in the Legislature and divested from themselves the power to enact ordinary laws, and then later authorized the Legislature to share a limited portion of that vested power with the electorate. No similar measure has ever authorized the Legislature to share the power to adopt nonbinding resolutions with the electorate. It remains a power that the people have vested in the Legislature and divested from themselves.
The Center for State and Local Government Law at the University of California Hastings College of the Law, in an amicus curiae brief, contends that the Legislature has always had general power to submit statutes for the voters to enact and, because “[t]he greater power includes the lesser,” also has power to submit advisory measures. According to amicus curiae, the 1911 indirect initiative provision “acknowledged” rather than authorized the Legislature’s power to submit to the voters an alternative measure on the same subject as a citizens’ initiative. For this proposition, amicus curiae relies on a report by the California Constitution Revision Commission (Commission) discussing 1966 and 1968 revisions that included elimination of the indirect initiative. The “Introduction” to the report said: “The Legislature can exercise all of the State’s legislative power and can act upon any subject unless the power has been delegated to the federal government or exercise of the power is forbidden by the State or Federal Constitution. It therefore is unnecessary to grant power in the Constitution which the Legislature inherently possesses. It is appropriate, however, to prohibit or compel by constitutional provision certain exercises of the Legislature’s inherent power. [Citation.] [¶] Our Constitution is encumbered by this type of unnecessary grant of power to the Legislature and by provisions which the Legislature is competent to enact by statute. Relying upon the rule of inherent legislative power, the Commission often has formulated recommended revisions which free our Constitution *568from these encumbrances.” (Com., Proposed Revision (1968) p. 7; see cone, opn. of Cantil-Sakauye, C. J., ante, at pp. 543-544 & fn. 35.)
But this quotation, which states the Commission’s general objective of eliminating unnecessary grants of power to the Legislature, does not illuminate whether the specific revision eliminating the indirect initiative was an instance of eliminating an unnecessary grant of power. As the Legislature’s briefing notes, the Commission’s specific rationale for eliminating the indirect initiative was that a separate proposed revision had ‘“reduced the percentage of signatures required for an initiative statute from 8 to 5 percent, the same as required for the indirect initiative. The indirect initiative merely adds an additional step to accomplish the same result that can be accomplished under the initiative generally. Further, the indirect initiative has been used only four times, and only once successfully. Accordingly, it was determined that the indirect initiative could be deleted without impairing the right of the people to propose laws through the initiative procedure.” (Com., Proposed Revision (1966) p. 52; see People v. Kelly (2010) 47 Cal.4th 1008, 1040 & fn. 55 [103 Cal.Rptr.3d 733, 222 P.3d 186] [discussing history of elimination of indirect initiative].) Nothing in this explanation suggests that the indirect initiative provision merely acknowledged the Legislature’s extant power to submit statutes to the electorate. The provision is most naturally read to grant the Legislature a power it did not previously have.
Amicus curiae also contends that article II, section 12 of the California Constitution—“No amendment to the Constitution, and no statute proposed to the electors by the Legislature or by initiative, that names any individual to hold any office, or names or identifies any private corporation to perform any function or to have any power or duty, may be submitted to the electors or have any effect” (italics added)—confirms that the Legislature has authority to place ordinary statutes on the ballot. But this provision became part of our Constitution in 1950 (Cal. Const., art. IV, former § 26, as adopted Nov. 7, 1950, renumbered June 8, 1976), when the Legislature possessed the power of indirect initiative. Moreover, well before 1950 and ever since, our state Constitution has required the Legislature to submit to the people any statute that issues state bonds above a certain amount (Cal. Const., art. XVI, § 1) or that amends or repeals an initiative statute (id., art. II, § 10, subd. (c)). Thus, even after the indirect initiative was eliminated in 1966, article II, section 12’s reference to a “statute proposed to the electors by the Legislature” has served to ensure that statutes authorizing state debt and statutes amending or repealing initiative statutes do not name any individual to hold office or any private corporation to perform any function or duty. It does not necessarily imply that the Legislature has general authority to propose statutes to the voters. (Cf. Czesla, Review of Article IV of the Cal. Constitution *569(1964) p. 3 [art. II, § 12’s prohibition “apparently resulted from the enactment in 1948 of an initiative constitutional amendment which made the Director of Social Welfare an elective office and named the first director”].)
Most telling, the Legislature in its briefing does not claim it has the power to submit ordinary statutes to the electorate. When pressed on this point at oral argument, counsel for the Legislature declined to assert the Legislature has such power. This hesitation undermines the broad contention in the Legislature’s briefing that no action of the Legislature may be found “inconsistent with the ‘structure’ of the Constitution, without it violating some explicit prohibition contained therein.” Even though our Constitution does not explicitly prohibit the Legislature from submitting ordinary statutes to the electorate, the Legislature refuses to claim it has such authority.
And for good reason: Apart from the powers of initiative and referendum, the people have vested lawmaking power in the Legislature, thereby divesting it from themselves, and the Legislature may not delegate that power to the people without specific constitutional authorization. Such authorization has occurred at various times through various provisions of our Constitution, demonstrating that when our Constitution contemplates shared lawmaking authority between the people and their representatives, it has said so explicitly. Article I’s declaration that “[t]he people have the right to instruct their representatives” is another provision that expressly authorizes shared accountability in lawmaking. (Cal. Const., art. I, § 3, subd. (a); but cf. dis. opn., post, at pp. 584-585 [the people’s right to instruct cannot empower the Legislature to submit advisory questions that the people themselves cannot place on the ballot].) Absent such authorization, the Legislature may not statutorily alter the constitutional allocation of lawmaking power between the people and their representatives. And by the same logic, the Legislature may not statutorily authorize the people to adopt a nonbinding resolution because that power is constitutionally vested in the Legislature and not retained by the people. This arrangement, with its clear lines of accountability, is not the only way to structure a well-functioning lawmaking process. But it is the way the people of California have chosen.
D.
The Chief Justice contends that advisory ballot measures have long been used by states and localities, and that this historical practice suggests that such measures are legally valid. (Conc. opn. of Cantil-Sakauye, C. J., ante, at pp. 524-535.) But it is unclear how probative the examples from other jurisdictions are. For one thing, the “scores” (and how many “scores” are there really?) “of legislatively initiated advisory ballot measures that have been submitted to statewide voters, both in California and throughout the *570country from this nation’s inception” (id. at pp. 549-550, fn. 39), do not seem all that impressive when considered in the context of state constitutional histories that, in the aggregate, span several thousands of years. In addition, we cannot assume that the constitutions and laws of the several states—with their unique texts, structures, and histories—are uniform with respect to the issue before us. For example, one important aspect of the California Constitution is that the people, by ratifying the initiative and referendum provisions in 1911, returned part of the legislative power to themselves, but not the power to adopt resolutions. (American Federation of Labor v. Eu, supra, 36 Cal.3d at pp. 694, 708.) By contrast, at least three states from which the Chief Justice draws examples authorize voters to place advisory measures on the ballot. (Yute Air Alaska, Inc. v. McAlpine (Alaska 1985) 698 P.2d 1173, 1175-1177; Simpson v. Cenarrusa (1997) 130 Idaho 609 [944 P.2d 1372, 1376-1377]; 10 Ill.Comp.Stat. 5/28-9.) Moreover, the Chief Justice identifies only one state high court that has ever addressed an issue like the one before us. (Cranmer v. Thorson (1896) 9 S.D. 149 [68 N.W. 202].) But that precedent predated South Dakota’s constitutional amendment providing for the power of initiative and referendum, and since then, the South Dakota Supreme Court has been narrowly divided on whether the legislature has authority to submit questions to the voters. (Wyatt v. Kundert (S.D. 1985) 375 N.W.2d 186, 191; id. at p. 198 (dis. opn. of Wuest, J., joined by Fosheim, C. J.).)
What is clear is that historical practice in California provides scant support for the broad thesis that the Legislature has general authority to place advisory measures on the statewide ballot. Although the Chief Justice cites a wide range of local advisory measures, such measures are specifically authorized by statute (Elec. Code, § 9603) and do not illuminate the constitutional question presented here. As to the seven examples of statewide advisory measures identified by the Chief Justice, three implicated the Legislature’s role under article I, section 3, or article V of the federal Constitution of expressing the state’s sovereign will concerning the election of United States Senators. (See maj. opn., ante, at pp. 507-508.) That leaves four advisory measures that did not concern any function assigned to the Legislature by the federal Constitution.
Among those four measures, the Chief Justice places greatest emphasis on an 1879 advisory measure that asked the electorate whether it was “ ‘[f]or’ ” or “ ‘[a]gainsf ” Chinese immigration. (Stats. 1877, ch. 5, § 1, p. 3.) According to the Chief Justice, this measure, which resulted in a 96 percent majority voting against Chinese immigration (Certificate of Vote on ‘“An Act to Ascertain and Express the Will of the People of the State of California on the Subject of Chinese Immigration” (1879)), shows that the delegates to the 1878-1879 Constitutional Convention ‘“obviously assumed that the Legislature possessed and retained authority to submit such an advisory measure to *571the statewide voters” (conc. opn. of Cantil-Sakauye, C. J., ante, at p. 536; see id. at p. 543 [reiterating this point]; id. at p. 544 [again]; id. at p. 546 [and again]).
But this example should not guide our constitutional inquiry. As the Chief Justice acknowledges, the 1879 advisory measure was motivated by ‘“virulent and racist views” (conc. opn. of Cantil-Sakauye, C. J., ante, at p. 536), and this court recently observed that ‘“[a]nti-Chinese sentiment was a major impetus for the California Constitutional Convention of 1879” (In re Chang (2015) 60 Cal.4th 1169, 1172 [185 Cal.Rptr.3d 1, 344 P.3d 288]). Against this historical backdrop, it is fanciful to posit that the constitutional delegates—a large portion of whom were members of the Workingmen’s Party, whose slogan was ‘“The Chinese Must Go!” (ibid.)—had a well-considered view of the constitutionality of the anti-Chinese advisory measure.
Scarcely a decade after ratification of the Fourteenth Amendment to the federal Constitution—section 1 of which says no state shall ‘“deprive any person of life, liberty, or property, without due process of law; nor deny to any person within its jurisdiction equal protection of the law”—those delegates wrote the infamous former article XIX (titled Chinese) into the 1879 Constitution, expressly prohibiting employment of the Chinese, authorizing the Legislature and localities to remove the Chinese from their jurisdictions, and directing the Legislature to ‘“discourage their immigration by all the means within its power.” (Cal. Const., art. XIX, as ratified 1879; see In re Chang, at pp. 1172-1173.) The delegates also constitutionalized the denial of the right to vote to any “native of China.” (Cal. Const., art. II, § 1, as ratified 1879.) Given the fact that the anti-Chinese advisory measure received “the blessing of the constitutional delegates” (conc. opn. of Cantil-Sakauye, C. J., ante, at p. 536) alongside these other blatantly unconstitutional provisions, I do not think the 1879 example sheds much light on the constitutionality of statutes placing advisory measures on the ballot. For the same reason, I would not assign any weight to the Legislature’s submission of another anti-Chinese advisory measure in 1891 asking whether English literacy should be a requirement for voting. (Stats. 1891, ch. 113, pp. 704-705.) The fact that an advisory ballot measure, as a procedural mechanism, “is not by its nature invariably discriminatory or offensive” (conc. opn. of Cantil-Sakauye, C. J., ante, at p. 549, fn. 39) does not explain why we should think that legislators who did submit advisory measures that were patently discriminatory during an era of “xenophobia” (In re Chang, at p. 1172) thought or cared much about the constitutionality of such measures.
The two remaining examples are a pair of 1933 measures asking voters whether the Legislature should divert gasoline tax funds to pay down debt on outstanding highway bonds. (Stats. 1933, ch. 435, § 3, p. 1126.) The voter *572guide explained that the measures proposed an alternative means of paying down bond debt that had been previously approved by the voters, as was (and still is) required by article XVI of our state Constitution. (Ballot Pamp., Special Elec. (June 27, 1933) argument in favor of Props. 9 & 10, p. 11.) The Legislature may well have believed that the alternative payment plans, like the original debts, required voter approval. (Cal. Const., art. XVI, § 1 [Legislature may not create debt of over a certain amount unless a law enacted by a two-thirds vote in both houses and approved by a majority of voters “provide[s] ways and means” for payment of interest and principal.].) In any event, the legality of these measures was never tested; they were placed on the ballot just one month before the election, and both were soundly defeated at the polls. (Stats. 1933, p. xc, Propositions Submitted to Vote of Electors, Special Election (June 27, 1933) Measures Defeated, Nos. 9, 10.)
In sum, the Legislature’s past uses of advisory ballot measures hardly demonstrate a ‘“well accepted” practice (conc. opn. of Cantil-Sakauye, C. J., ante, at p. 535), let alone one whose constitutionality has been implicitly affirmed. As Justice Chin notes, ‘“the most recent of these advisory measures occurred over eight decades ago, long before the development of modern polling techniques, the Internet, and other methods of ascertaining the people’s wishes in political matters.” (Dis. opn., post, at p. 587.) If anything, the paucity of historical examples, with none in the last 80 years, tends to cast doubt on the existence and validity of the power claimed here.
E.
No member of the court suggests that American Federation of Labor v. Eu was wrongly decided or that our holding in that case may be evaded by an initiative statute that directs the Secretary of State to place an advisory measure on the ballot. But today’s opinion rejects the argument that ‘“if, under American Federation of Labor v. Eu, supra, 36 Cal.3d 697, the people are limited to placing on the ballot only proposed laws, then the Legislature must be too.” (Maj. opn., ante, at p. 516.) Noting that the Legislature’s powers are broader than the people’s initiative power (American Federation of Labor v. Eu, at p. 708), the court contends that the Legislature’s broader powers authorize it to pass a statute placing an advisory measure on the ballot even though the people may not do so by initiative. I agree with this thesis where, as here, the Legislature is exercising its implied power under article V of the federal Constitution. But I do not agree when it comes to advisory ballot measures on ordinary matters of public policy.
The court, like the Legislature, does not rest its reasoning “on the syllogism that the legislative power includes the power to enact statutes, *573Senate Bill No. 1272 takes the form of an enacted statute, and thus for that reason alone the bill and Proposition 49 are within a constitutional source of power.” (Maj. opn., ante, at p. 498.) Instead, the court accepts the Legislature’s claim that a nonlawmaking power—the power to investigate— encompasses the authority to ‘“enact a statute placing an advisory question before the voters.” (Ibid.) This invocation of the investigative power is unpersuasive for several reasons.
As an initial matter, it is odd to characterize Senate Bill No. 1272 (2013-2014 Reg. Sess.), which the Legislature labeled the Overturn Citizens United Act (Stats. 2014, ch. 175, § 1), as an effort to investigate the electorate’s views. The Legislature’s findings include declarations that Citizens United ‘“presents a serious threat to self-government” (Stats. 2014, ch. 175, § 2, subd. (e)); that ‘“[a] February 2010 Washington Post-ABC News poll found that 80 percent of Americans oppose the ruling in Citizens United” (id., subd. (k)); and that ‘“Article V of the United States Constitution empowers and obligates the people of the United States of America to use the constitutional amendment process to correct those egregiously wrong decisions of the United States Supreme Court that go to the heart of our democracy and the republican form of self-government” (id., subd. (l)). This is the language of public mobilization, not investigation, and it fits comfortably within the Legislature’s role under article V in marshaling vigorous and solemn expressions of California’s sovereign will on whether to amend our nation’s basic charter. It blinks reality to suggest that the Legislature—plainly aware of opinion polls showing that broad majorities of Americans are opposed to Citizens United—enacted Senate Bill No. 1272 in order to investigate the citizenry’s views.
More fundamentally, the court’s invocation of the investigative power in this context is a novel expansion of how we have historically construed this power. Our case law on the investigative power uniformly consists of disputes concerning the Legislature’s prerogative to establish committees, boards, commissions, and agencies, and to vest such entities with authority to compel evidence and testimony and to hold noncompliant witnesses in contempt. Those issues are far afield from the question before us. As a review of the cases shows, nothing in this court’s treatment of the Legislature’s power to investigate suggests it encompasses the authority to submit an advisory measure to the electorate.
In McCarthy, supra, 29 Cal. 395, a newspaper editor was held in contempt by the state Senate and jailed after he refused to answer questions posed by the Senate in the course of investigating charges of bribery against its members. (Id. at pp. 397-399.) The petitioner challenged his detention, claiming the Senate lacked power to investigate the bribery charges. We *574rejected the claim and held that the Senate’s power to investigate included the power to compel testimony from witnesses and to hold persons in contempt for refusing to testify. (Id. at pp. 403-407.)
In In re Battelle (1929) 207 Cal. 227 [277 P. 725] (Battelle), the state Senate adopted a resolution authorizing a committee of senators to investigate a possible price-fixing conspiracy among cement manufacturers. The committee subpoenaed witnesses and required them to bring certain records and documents within their control, but various witnesses refused. The Senate adopted a resolution holding the witnesses in contempt. The petitioner was arrested and sought habeas corpus relief in this court. In addressing the petitioner’s claims, we said that “in many instances, in order to the preparation of wise and timely laws the necessity of investigation of some sort must exist as an indispensable incident and auxiliary to the proper exercise of legislative power.” (Id. at p. 241.) We further explained that “the inherent and auxiliary power reposed in legislative bodies to conduct investigations in aid of prospective legislation has already been held to carry with it the power in proper cases to require and compel the attendance of witnesses and the production of books and papers by means of legal process and to institute and carry to the extent of punishment contempt proceedings in order to compel the attendance of such witnesses and the production of such documentary evidence as may be legally called for in the course of such proceedings, whether conducted by the legislative body or a branch of it, directly or through properly constituted committees thereof.” (Ibid) We rejected the claim that the use of such compulsory process meant the Legislature was exercising a judicial function in violation of the separation of powers. (Id. at pp. 241-244.) But we ultimately granted the petitioner relief on the ground that the contempt order did not adequately explain how the questions the petitioner refused to answer were pertinent to the issue the Legislature was investigating. (Id. at pp. 246-247.) Like McCarthy, Battelle had nothing to do with advisory ballot measures.
The other cases cited in today’s opinion are similarly far afield. Both Swing v. Riley (1939) 13 Cal.2d 513 [90 P.2d 313] and Special Assembly Int. Committee v. Southard (1939) 13 Cal.2d 497 [90 P.2d 304] (Southard) addressed issues concerning the proper appointment of legislative committees. Parker v. Riley (1941) 18 Cal.2d 83 [113 P.2d 873] held that a statute creating the California Commission on Interstate Cooperation did not violate separation of powers. The key passage in that case discussing the power to investigate gives context and substance to this power: “ ‘The ascertainment of facts in its essence is not a legislative function. It is simply ancillary to legislation. It may be accomplished in divers ways. While it may be done by the Legislature itself, it is a responsibility not infrequently placed upon committees and individuals. . . . Frequent illustrations of this practice also are found respecting permanent boards or commissions. . . . The ascertainment of *575pertinent facts for legislation is within the power of the lawmaking department of government. When a legislative body has a right to do an act it must be allowed to select the means within reasonable bounds. It is not precluded from delegating incidental powers which it may exercise itself in aid of its primary functions. . . . Familiar methods are by appropriating the results of studies already made by itself or by others, by conducting an inquiry through a committee of its members, or by utilizing an existing commission or board to make and report the results of a research.’ ” (Id. at p. 91.) Amid these mentions of committees, boards, commissions, studies, and research, there is not the slightest hint that the Legislature may submit an advisory measure to the electorate pursuant to its investigative power.
The novelty of the court’s reasoning has an even deeper dimension. As noted, neither the Legislature nor any member of this court has invoked the Legislature’s ordinary lawmaking power as its source of authority for enacting Senate Bill No. 1272 (2013-2014 Reg. Sess.), even though a duly enacted statute (as opposed to, say, a mere committee resolution or executive order) is required to place Proposition 49 on the ballot. The court instead resorts to the Legislature’s incidental power to investigate. Yet in every case cited by the court regarding the scope of the Legislature’s incidental powers, the Legislature acted without passing a statute. (See Battelle, supra, 207 Cal. at pp. 230-240; Southard, supra, 13 Cal.2d at pp. 498-502; Swing v. Riley, supra, 13 Cal.2d at pp. 514-517; McCarthy, supra, 29 Cal. at pp. 397-399; Parker v. Riley, supra, 18 Cal.2d at pp. 89-91.) Today’s opinion holds for the first time that the Legislature’s investigative power is itself a font of authority, separate and apart from its ordinary lawmaking power, for the Legislature to enact statutes. This holding implies that the Legislature may enact certain statutes pursuant to its investigative power that it could not enact pursuant to its ordinary lawmaking power. After all, if the Legislature’s ordinary lawmaking power were sufficient in this case, presumably the court’s analysis would have stopped there.
But to say that the investigative power itself encompasses a power to enact a statute, whether or not the statute could be enacted pursuant to the Legislature’s ordinary lawmaking power, appears to contravene the settled principle that the Legislature’s ordinary lawmaking power under our state Constitution is “plenary.” (California Redevelopment Assn. v. Matosantos (2011) 53 Cal.4th 231, 254 [135 Cal.Rptr.3d 683, 267 P.3d 580].) In other words, the Legislature’s lawmaking power is “[f]ull,” “complete,” and “entire.” (Black’s Law Diet. (9th ed. 2009) p. 1273, col. 1.) This court has never held that the Legislature’s authority under the California Constitution to make laws may be grounded in anything other than its lawmaking power, and we have never suggested that the investigative power or any incidental power subsumes a portion of the Legislature’s power to make laws. Of course, the Legislature may exercise its lawmaking power to pass laws that accomplish *576investigative aims. (See, e.g., Ed. Code, §§ 10600-10601.5; id., § 10600, subd. (a) [requiring state Dept, of Education to maintain an information system that “make[s] complete, current, and reliable information relating to education available to the Legislature”].) But that is quite different from saying that the investigative power itself authorizes the Legislature to pass such laws.
Today’s opinion cautions that ‘“[t]he investigative power is not unlimited” (maj. opn., ante, at p. 499) and holds only that this power authorizes an advisory ballot measure as ‘“a reasonable and lawful means of assisting the Legislature in the discharge of its article V-related functions” (id. at p. 523). Whether this holding can be limited to the article V context remains to be seen. If all that is required is ‘“any reasonable connection” between the advisory measure and the Legislature’s potential exercise of any power it lawfully possesses (id. at p. 521), then I find it difficult to discern any meaningful justiciable limit on the Legislature’s use of advisory ballot measures.
In sum, today’s application of the Legislature’s investigative power finds no support in our precedent and disturbs the settled understanding that the Legislature’s lawmaking power is plenary. It is hard to believe that the people of California—having vested ‘“[t]he legislative power of this State” in the Legislature (Cal. Const., art. IV, § 1), including the power to adopt a resolution (American Federation of Labor v. Eu, supra, 36 Cal.3d at p. 708)—at the same time contemplated that the Legislature, by exercising an incidental power, could restore to the people a portion of that constitutionally vested legislative power.
III.
The court’s reliance on the investigative power is unnecessary to reach today’s limited holding. As explained above, Senate Bill No. 1272 (2013-2014 Reg. Sess.) is a valid exercise of the Legislature’s implied power under article V of the federal Constitution. Although the Chief Justice would uphold advisory ballot measures ‘“with regard to any potential action that the Legislature has authority to undertake” (conc. opn. of Cantil-Sakauye, C. J., ante, at p. 542), the court wisely declines to go so far. A decision of this court opening the door to advisory ballot measures on virtually any subject would potentially transform the way electoral politics and policymaking are conducted in California.
I do not doubt that advisory ballot measures can provide the Legislature with valuable information about the electorate’s views on public policy. But such prosaic uses of advisory measures do not exhaust the possibilities. Lor *577example, advisory measures can be used to undercut the people’s initiative power, a concern expressed by Citizens in Charge, an “advocacy group dedicated to protecting and expanding citizens’ initiative, referendum, and recall rights throughout America,” in an amicus curiae letter supporting the original emergency petition for writ of mandate in this matter. It is a common tactic for opponents of an initiative to qualify a competing measure on the ballot, thereby creating voter confusion. (Garrett, Direct Democracy in Research Handbook on Public Choice and Public Law (Farber & O’Connell edits., 2010) p. 155.) A legislative majority opposed to an initiative could swell the ballot with advisory measures on the same subject. Such an approach would be much easier than qualifying a competing initiative but no less effective in confusing and exhausting the electorate. (See dis. opn., post, at p. 585 [“the potential, and the temptation, for legislative interference with the people’s power of initiative is real”].) As many California voters would agree, it is already hard enough to cast an informed vote on the statewide ballot given the multiple propositions and competing propositions each year. The prospect of also having to vote on advisory measures can only worsen voter confusion and fatigue.
In addition, advisory measures could be used to influence voter sentiment about a qualified initiative. Suppose, not unrealistically, that one political party controls majorities in the Legislature and the governorship, and that the majority party opposes tax cuts. Suppose further that concerned citizens qualify an initiative proposing broad-based tax cuts. If the Legislature had general authority to put advisory measures on the ballot, then presumably the Legislature could pose the following questions on the same ballot as the tax-cutting initiative:
Proposition A. To pay for any future tax cuts, should the Legislature consider cutting funding for police departments?
Proposition B. To pay for any future tax cuts, should the Legislature consider cutting funding for state parks?
Proposition C. To pay for any future tax cuts, should the Legislature consider cutting funding for higher education?
Moreover, advisory ballot measures could be used to entrench a political majority. Consider, for example, the 2010 gubernatorial race between Jerry Brown and Meg Whitman. Whitman, a wealthy business executive, faced controversies over her alleged inside dealings with an investment bank and the retention of a housekeeper whose undocumented status she allegedly knew. Suppose Democrats had controlled majorities in the Legislature and the governorship during the 2010 campaign season. If the Legislature had general *578authority to put advisory measures on the ballot, then presumably the Legislature could have posed the following questions on the 2010 ballot:
Proposition D. To promote fairness, should the Legislature consider a law that would prohibit business executives from earning incomes more than 100 times the incomes of their average employees?
Proposition E. To discourage corruption, should the Legislature consider a law that would more closely monitor business executives involved in inside deals with investment banks?
Proposition F. To restore jobs to hard-working Americans, should the Legislature consider a law that would heavily penalize those who knowingly employ undocumented immigrants as housekeepers?
In these examples above, the advisory measures would be designed to cue voters to defeat the tax-cutting initiative (Propositions A, B, and C) or to defeat Meg Whitman (Propositions D, E, and F). I am not sure whether these measures would bear a “reasonable connection” to the Legislature’s lawful power to act under today’s opinion. (Maj. opn., ante, at p. 521.) But if we are not to “inquire into underlying motives” (ibid.), and if the Legislature may “obtain the views of the voters concerning all manner of subjects reasonably within a legislative body’s authority to act” (conc. opn. of Cantil-Sakauye, C. J., ante, at p. 524), then on what basis could this court foreclose the Legislature from placing such measures on the ballot?
It is easy to dismiss the possibility of such canny tactics as speculative or remote. But even if advisory measures in local elections have not been used in such shrewd ways, there is little assurance that the same would be true in statewide elections, where matters of broader scale and impact are typically at stake, and where much greater sums of money and higher levels of political sophistication are brought to bear on electioneering tactics. Moreover, because advisory measures provide an avenue for influencing voters right at the moment of voting, they are likely to be far more effective—and certainly far less expensive—than running ads on television or the Internet, or using mail, e-mail, text messages, or social media to reach voters. In a hard-fought political battle, the ability to prime voters with messages on the ballot itself presents an enormously attractive option. And there would be no point in calling such tactics abusive, for they would be perfectly legal and fair game if this court were to hold that the Legislature has general authority to place advisory measures on the ballot. As the proponents of Proposition 49 well know, the judiciary plays a critical role in setting the rules of politics, and it is no surprise when political actors seek every possible advantage within those rules.
*579It is true that the Legislature has not used such tactics in modern times. But in my view, that is evidence that the claimed power does not exist and that our political system has not needed it. Of course, I cannot say for sure whether a judicial decision recognizing such a power would actually change the landscape of California politics in the way suggested above. But the risk is not one that our Constitution requires us to tolerate.
The California Constitution, like the United States Constitution and other state constitutions, establishes the fundamental structure of government. It is within this structure that the Legislature exercises its power. No issue is more basic to the structure of government than what types of matters may be placed on the ballot for a vote of the citizenry. Our state Constitution establishes the rules by which voters may propose initiatives and referenda, and it requires the Legislature to submit constitutional amendments and certain statutes to the voters for approval. This constitutional structure constrains the power of legislative majorities. Just as the Legislature does not have general authority to submit statutes for voters to approve, it does not have general authority to use advisory ballot measures to poll the electorate. As a matter of state law, the Legislature may not alter its constitutional role as the people’s accountable representative body by directing the people, at the Legislature’s convenience, to exercise a merely hortatory, nonbinding voice in ordinary public affairs. If this court were to revise the balance our Constitution has struck between direct and representative democracy, it would have potentially serious consequences for our political system.
The case before us is different. Here the Legislature is playing a role that originates not in our state Constitution but in article V of the federal Constitution. Under article V, the Legislature has wide latitude to marshal the electorate’s views in order to express California’s sovereign will to Congress and to other states on whether to amend the nation’s fundamental charter. The court properly confines today’s holding to advisory measures concerning federal constitutional amendments. Such a measure is all that is at issue here, and I concur in the judgment upholding its placement on a statewide ballot.