Opinion

We are called on in this case to construe the provisions of article V, section 5, subdivision (b), of the California Constitution (hereafter section 5(b)), which sets forth the requirements for confirmation by the Legislature of the Governor’s nominee to fill a vacancy in the office of Treasurer and other constitutional offices. Specifically, we shall decide whether under that provision a nominee may be viewed as having been confirmed by the Legislature even though he has been confirmed by only one house and his nomination has been rejected by the other house.

Jesse Unruh, who had been elected Treasurer, died on August 4, 1987. Elizabeth Whitney, his chief deputy, became Acting State Treasurer, as provided by section 1775 of the Government Code. Thereafter, on December 1, 1987, Governor Deukmejian appointed Congressman Daniel Lungren to fill the office of Treasurer and forwarded his name to the Senate and Assembly for confirmation, as required by section 5(b). There ensued several days of hearings on the nomination by committees of both houses, as well as debate on the floor of each chamber. On February 25, 1988, the Assembly voted to confirm Lungren’s nomination, but the Senate voted to deny confirmation.

The Governor declared that confirmation by one house was sufficient under section 5(b) and that he therefore viewed Lungren as having been confirmed. However, the Governor, in a letter to Lungren, stated that he would not issue a commission to enable him to assume the office because to do so under the circumstances would have “serious negative consequences” for marketability of state bonds, and Lungren’s ability to carry out the duties of Treasurer would be compromised by a legal challenge to the appointment.

Lungren then filed an original petition for a writ of mandate in this court to enforce his right to assume office. He named as respondents the Governor, the Secretary of State, the Controller, and Acting Treasurer Whitney. The President Pro Tem. of the Senate was named as real party in interest. The petition alleges that Lungren has satisfied all requirements necessary to assume the office of Treasurer under section 5(b), but that the Governor has refused to issue him a commission, and even if he were to do so, the remaining respondents would refuse to carry out the duties imposed on them by law to allow Lungren to assume office. Lungren seeks a writ of mandate commanding respondents to comply with these duties and to allow him to assume office, and asks that Whitney be directed to proceed immediately with all pending bond sales until Lungren assumes office.

Because of the importance of the issues and the need for their speedy resolution, we retained the petition and granted an alternative writ. (Joli coeur v. Mihaly (1971) 5 Cal.3d 565, 570, fn. 1 [ 96 Cal.Rptr. 697 , 488 P.2d 1 ].) We also granted the motions of the Senate to intervene as a real party in interest, and of the Attorney General to file a complaint in intervention. For ease of reference, the Attorney General, the Senate, and the respondents named in the petition, with the exception of the Governor, will be referred to collectively as “respondents,” even though each of them does not make every argument attributed to “respondents.” The Governor has filed a brief in support of Lungren, making substantially the same arguments as those advanced by him.

Before we reach the merits of the issues raised by the parties, we must discuss respondents’ claim that mandate is not available to Lungren because he does not have a present interest in the office of Treasurer and none of the state officials he has joined in the action has a present duty to him. Mandate will not lie unless the applicant for the writ has a present interest in the remedy he seeks and the respondent has a present duty to perform the acts the applicant seeks to compel. (Treber v. Superior Court (1968) 68 Cal.2d 128, 134 [ 65 Cal.Rptr. 330 , 436 P.2d 330 ]; Communist Party v. Peek (1942) 20 Cal.2d 536, 540 [ 127 P.2d 889 ]; Northridge etc. Water Dist. v. McDonell (1958) 158 Cal.App.2d 123, 127-128 [ 322 P.2d 25 ].) The petitioner’s right and the respondent’s duty are measured as of the time the proceeding is filed. (Christ v. Superior Court (1931) 211 Cal. 593, 600 [ 296 P. 612 ]; McGinnis v. Mayor and Common Council (1908) 153 Cal. 711, 715 [ 96 P. 367 ].)

In support of their claim that Lungren does not have a present interest in the office of Treasurer, respondents cité article VII, section 7, of the Constitution. It provides, “A person holding a lucrative office under the United States or other power may not hold a civil office of profit." The purpose of the provision is to prevent “dual office-holding by one person under two separate and distinct governments, and the separation of the allegiance justly due one by its officers from that due to ánother power.” (McCoy v. Board of Supervisors (1941) 18 Cal.2d 193, 196 [ 114 P.2d 569 ].) Lungren is a United States Congressman, and is thus prohibited by this provision from holding that position and the office of Treasurer simultaneously. He acknowledges this, but explains that he will resign his congressional seat if we decide in his favor, and therefore he will not “hold” the two offices at the same time.

We accept Lungren’s representation in this regard. His future resignation would remove the bar of article VII, section 7, of the Constitution, but it does not follow that he has a present right to assume the office of Treasurer. Clearly he does not: the fact that he is a Congressman precludes him from assuming the office of Treasurer. Because he cannot now assume that office, it follows that respondents owe him no present duty to facilitate his assumption of it.

Thus, it is clear to us that Lungren is not presently entitled to the remedy he seeks. Nevertheless, the same considerations that led us to exercise our original jurisdiction convince us that we should decide the merits of his claim that he has been confirmed as Treasurer. According to the allegations in the petition, a prolonged dispute over whether Lungren is entitled to assume the office could imperil the state’s ability to market bonds and adversely affect the state’s bond rating. Respondents deny that the state’s bond rating is at risk pending adjudication of his rights, but they agree that expeditious resolution of the dispute on its merits is in the public interest. No one can know with assurance whether the state’s fiscal position will be adversely affected by the failure to speedily resolve the dispute. Nevertheless, we are of the view that there is sufficient uncertainty regarding the matter to justify deciding the entitlement of Lungren to the office on its merits in this proceeding rather than postponing a decision until he resigns his congressional seat and thereafter demands that respondents perform the acts purportedly required of them.

We proceed, then, to a discussion of the merits of his claim. Section 5(b), adopted by the people as an assembly constitutional amendment in 1976, provides in part, “Whenever there is a vacancy in the office of the. . . Treasurer . . ., the Governor shall nominate a person to fill the vacancy who shall take office upon confirmation by a majority of the membership of the Senate and a majority of the membership of the Assembly and who shall hold office for the balance of the unexpired term. In the event the nominee is neither confirmed nor refused confirmation by both the Senate and the Assembly within 90 days of the submission of the nomination, the nominee shall take office as if he or she had been confirmed by a majority of the Senate and Assembly . . . .”

The first sentence of the section is clear: the nominee takes office if he is confirmed by a majority of each house of the Legislature. This is the normal means by which a bicameral legislative body takes action. Unless otherwise expressly provided by law, both houses must concur in enacting legislation or taking any other action. (Cal. Const., art. IV, § 1.) A measure fails of passage if either house does not approve it, whether such failure arises from an express negative vote or from inaction. The second sentence of section 5(b) was evidently intended to state an exception to this normal mode of operation. Our task is to determine the scope of the exception.

We shall conclude that the language of the second sentence of section 5(b), standing alone, is susceptible to the construction offered by both respondents and Lungren, but that, when the section is read as a whole, it supports respondents’ view that a negative vote on the confirmation by either house of the Legislature results in disapproval of the nomination. Further, that the legislative history of section 5(b), as derived from the ballot pamphlet submitted to the voters at the election in which the provision was adopted, confirms the latter view as to its meaning. Finally, we shall reject an alternative legislative history argument advanced by Lungren in support of his position.

I. The Language of Section 5(b)

According to respondents, the only effect of the second sentence is to prohibit the Legislature from rejecting a nominee by failing to act on a nomination. Since the result of such inaction under this view is that the nominee is deemed to be confirmed, the Legislature has an incentive to take a vote on the nomination within 90 days.

Lungren, while agreeing that the failure by the Legislature to act would result in confirmation, asserts that the second sentence has an additional effect: i.e., unless both houses actually vote to reject the nominee, he is deemed confirmed after 90 days from the time the nomination is submitted to them. Under this view, confirmation occurs if the two houses vote on the nomination but one approves and the other disapproves, as happened here, or if one votes (whether it approves or rejects) and the other fails to vote. In all these situations, according to Lungren, the nominee has been “neither confirmed nor refused confirmation by both the Senate and the Assembly” within 90 days, and he is therefore entitled to take office as though he had been confirmed by both.

Lungren claims that the construction he advances is the only reading that gives effect to the word “both” in the second sentence, and that if we do not adopt his interpretation, that word would be read out of the sentence. This is plainly incorrect. The word “both” may be viewed as referring only to a situation in which both houses fail to vote on the nomination: in that event, the nominee would be “neither confirmed nor refused confirmation by both the Senate and the Assembly,” in the words of the second sentence. This is essentially the “pocket veto” theory of respondents. Indeed, this meaning appears to flow more naturally from the language of the second sentence than the interpretation offered by Lungren, since the failure by the Legislature to confirm or refuse confirmation suggests inaction by the two houses rather than inconsistent action. The issue, then, is whether the provision of the second sentence that the nominee will be deemed to have been confirmed if “both” houses do not refuse or approve the nomination was meant to refer only to inaction by both houses, or whether, as Lungren claims, it applies also to inconsistent action by the two houses or inaction by one house alone.

It is arguable that if read literally, the second sentence will bear the more expansive interpretation offered by Lungren. In support of his position, he cites the so-called “plain meaning” rule. Words used in a statute or constitutional provision should be given the meaning they bear in ordinary use. (In re Rojas (1979) 23 Cal.3d 152, 155 [ 151 Cal.Rptr. 649 , 588 P.2d 789 ]; Great Lakes Properties, Inc. v. City of El Segundo (1977) 19 Cal.3d 152, 155 [ 137 Cal.Rptr. 154 , 561 P.2d 244 ].) If the language is clear and unambiguous there is no need for construction, nor is it necessary to resort to indicia of the intent of the Legislature (in the case of a statute) or of the voters (in the case of a provision adopted by the voters). (In re Lance W. (1985) 37 Cal.3d 873, 886 [ 210 Cal.Rptr. 631 , 694 P.2d 744 ]; State Board of Education v. Levit (1959) 52 Cal.2d 441, 462 [ 343 P.2d 8 ].)

But the “plain meaning” rule does not prohibit a court from determining whether the literal meaning of a statute comports with its purpose or whether such a construction of one provision is consistent with other provisions of the statute. The meaning of a statute may not be determined from a single word or sentence; the words must be construed in context, and provisions relating to the same subject matter must be harmonized to the extent possible. (Dyna-Med, Inc. v. Fair Employment & Housing Com. (1987) 43 Cal.3d 1379, 1386-1387 [ 241 Cal.Rptr. 67 , 743 P.2d 1323 ].) Literal construction should not prevail if it is contrary to the legislative intent apparent in the statute. The intent prevails over the letter, and the letter will, if possible, be so read as to conform to the spirit of the act. (People v. Belton (1979) 23 Cal.3d 516, 526 [ 153 Cal.Rptr. 195 , 591 P.2d 485 ]; Amador Valley Joint Union High Sch. Dist. v. State Bd. of Equalization (1978) 22 Cal.3d 208, 245 [ 149 Cal.Rptr. 239 , 583 P.2d 1281 ].) An interpretation that renders related provisions nugatory must be avoided (People v. Craft (1986) 41 Cal.3d 554, 561 [ 224 Cal.Rptr. 626 , 715 P.2d 585 ]); each sentence must be read not in isolation but in the light of the statutory scheme (In re Catalano (1981) 29 Cal.3d 1, 10-11 [ 171 Cal.Rptr. 667 , 623 P.2d 228 ]); and if a statute is amenable to two alternative interpretations, the one that leads to the more reasonable result will be followed (Metropolitan Water Dist. v. Adams (1948) 32 Cal.2d 620, 630-631 [ 197 P.2d 543 ]). These rules apply as well to the interpretation of constitutional provisions. (Stanton v. Panish (1980) 28 Cal.3d 107, 115 [ 167 Cal.Rptr. 584 , 615 P.2d 1372 ].)

It is immediately apparent from an examination of section 5(b) that the first sentence represents the main premise of the section, i.e., that both houses must confirm a nominee. This is the normal manner in which the Legislature acts. This conclusion is evident not only from the language used, but also, as we shall see, from its origin and legislative history. The second sentence is a proviso, i.e., an exception to or limita tion on the operation of the first. As such, it is to be strictly construed. (People ex rel. S.F. Bay etc. Com. v. Town of Emeryville (1968) 69 Cal.2d 533, 543 [ 72 Cal.Rptr. 790 , 446 P.2d 790 ]; Kirkwood v. Bank of America (1954) 43 Cal.2d 333, 341 [ 273 P.2d 532 ]; Johnson v. Board of Supervisors (1929) 208 Cal. 282, 285 [ 281 P. 57 ].) In the words of one court, provisos are “qualifiers, not nullifiers.” (Commonwealth of Pennsylvania v. Brown (E.D.Pa. 1966) 260 F.Supp. 323, 351 .) One reason to reject Lungren’s position is that it ascribes an unreasonably expansive meaning to the second sentence—the proviso—and concomitantly restricts the operation of the first sentence—the premise—of the section.

A second reason is that the construction he offers would virtually read the first sentence out of the section. Perhaps in recognition of the vulnerability of his position to such a charge, Lungren attempts to harmonize the first and second sentences in the following manner: he suggests that the two sentences offer alternative and equally valid means to confirm a nominee. Under the first sentence, both houses of the Legislature may vote to confirm within 90 days of the time the nomination is submitted to them; in that case, the nominee takes office immediately on confirmation by both houses. Alternatively, if such action does not occur within 90 days and both houses have not rejected the nominee, then under the second sentence he is entitled to assume office in any event on the 91st day. In that case, the nominee is deemed confirmed (because both houses neither approved nor rejected the nomination) no matter whether, as here, one house approved the nomination and the other rejected it, or whether one house voted (either to approve or reject) and the other failed to vote.

As we conclude above, however, the first and second sentences do not represent alternative and equal means of confirmation. Rather, the first provides the primary thrust of the section, while the second is a proviso limiting the operation of the first, and hence must be read narrowly.

More important, under Lungren’s interpretation there is a fundamental inconsistency between the first and second sentences. The first sentence would mean not that both houses must confirm, but that confirmation by either would suffice—or even that confirmation would result if one house voted to reject and the other failed to act —provided only that the nominee waited 90 days. The sole function of the two-house confirmation requirement of section 5(b) would thus be reduced to possibly allowing a nominee to take office a few days or weeks earlier than he would if only one house had acted on the nomination. Even this negligible effect would disappear if, as happened in this case, both houses acted at the end of the 90-day period. After 90 days, the nominee would be deemed confirmed if only one house had voted on the confirmation within that period, whether the vote was to approve or to reject, or if the two houses had voted inconsistently with each other. It is difficult to conceive of a more distorted interpretation of the straightforward command of the first sentence that the Senate and the Assembly must confirm a nominee.

Moreover, the 90-day waiting period required by the second sentence is incomprehensible if, as Lungren contends, confirmation by one house is sufficient. There would be no rational reason for a nominee to wait 90 days before assuming office following an affirmative vote on his nomination by one house if he was entitled to the office by virtue of that vote alone. Any action by the second house would ipso facto be rendered meaningless.

Finally, to accept Lungren’s position would seriously degrade the power and dignity of one house of the Legislature in the confirmation process: an express rejection of the nominee by that house would be rendered a nullity. So far as we are aware, there is no precedent for a parliamentary procedure that calls for a legislative body to vote on a measure and then ignores its negative vote. In a bicameral Legislature such as ours, the legislative power rests in both houses. (Cal. Const., art. IV, § 1.) We are not free to depart from this basic tenet of our constitutional system unless compelled to do so by the clear command of some other provision of law. In our view, the language of section 5(b) does not authorize such a drastic departure.

II. The Legislative History of Section 5(b)

A. The Ballot Pamphlet

The foregoing conclusion is bolstered by the legislative history of the section, which supports respondents’ assertion that a nomination is rejected if either house of the Legislature votes to reject the nominee within 90 days of the time the nomination is submitted by the Governor, and that the purpose of the second sentence is to prevent rejection by mere inaction.

Section 2 of the 25th Amendment was added to the United States Constitution in 1967 to provide for filling a vacancy in the office of Vice President. It is strikingly similar to the first sentence of section 5(b). In 1970, the Council of State Governments recommended that states adopt a similar provision to fill vacancies in the office of lieutenant governor, and such measures were adopted by a number of states in the years following. (Council of State Governments, Suggested State Legislation (1970) Model Exec. Art. 14-10-00, § 4, p. 5); Ind. Const., art. V, § 10, subd. (b), adopted 1978; La. Const., art. IV, § 15, adopted 1974; Md. Const., art. II, § 6, subd. (d), adopted 1970; S.D. Const., art. IV, § 6, adopted 1972; Wis. Const., art. XIII, § 10, subd. (2), adopted 1979.)

Section 5(b) of our Constitution, adopted by the voters as Proposition 9 at the General Election in 1976, was clearly a part of this movement, sparked by the 25th Amendment, to provide for the filling of vacancies in state offices. The ballot argument in favor of the measure informed the voters that it was modeled after the 25th Amendment, “which provides that a nominee for the Office of Vice President must be approved by the Senate and House.” (Ballot Pamp., Gen. Elec. (Nov. 2, 1976), argument in favor of Prop. 9, p. 38 [hereafter ballot argument].) There can be no doubt that under the 25th Amendment and the constitutional provisions of the other jurisdictions cited, the requirement that both houses must confirm a nominee, embodied in the first sentence of section 5(b), means that an express refusal by one house to approve the nomination or the failure of either house to vote thereon results in rejection. Lungren’s claim that the first sentence means only that the nominee might possibly take office sooner than if he had been rejected by one house is particularly unpersuasive in light of the close similarity between the first sentence of the section and the 25th Amendment, and the fact that the section was expressly modeled after the amendment which, the voters were told, requires confirmation by both houses of Congress.

Indeed, the ballot pamphlet repeated at least half a dozen times that the measure requires confirmation by both houses. The point was made in the ballot title and the analysis by the Legislative Analyst, and in the arguments to the voters. These statements are incompatible with the view that confirmation would result if one house rejected the nominee.

But neither the 25th Amendment nor the constitutional provisions in the other jurisdictions referred to above include a provision similar to the second sentence of section 5(b). As just observed, in these jurisdictions the failure of either or both houses to vote on the nomination would result in rejection by inaction. The ballot title as well as the arguments to the voters by both the proponents and the opponents of Proposition 9 make it clear that the second sentence was intended to cover inaction by the Legislature and to prevent it from rejecting a nominee merely by failing to act on the nomination.

The ballot title explicitly stated that if the Legislature “does not act within 90 days of Governor’s nomination . . . appointees may take office as if confirmed.” (P. 739, ante, fn. 13, italics added.)

The argument against the measure, written by Assemblyman Antonovich, described the situation that would exist if the two houses could not agree on whether to confirm the nominee: “Proposition 9 would also result in a political football game between the Legislature and the Governor. As the Governor would be subject to the whims of either the Senate or the Assembly, he could be rendered virtually powerless. The simple act of filling a vacancy could assume monstrous proportions if the Senate or Assembly could not reach agreement regarding a candidate that would be acceptable to both. Hence, the appointment could bounce back and forth between the Governor and Legislature, with each rejection involving more time wasted. In turn, the vacancy would remain unfilled and unproductive, while the Legislature becomes further embroiled in political maneuvering.” (Ballot argument, p. 39, italics added.)

The proponents, in response, impliedly acknowledged that the nomination would fail if the two houses did not agree, since they rebutted the foregoing argument with the statement that if there was a delay in filling the office because of a “genuine disagreement on the qualification of a candidate” the “tasks of the vacant office [would be] performed by the appropriate deputy.” (Ibid.)

We do not see how it can be denied that the statements by both proponents and opponents advised the voters that disapproval by either house would result in rejection of the nomination, requiring the Governor to submit another nominee to fill the vacancy.

Lungren argues that the ballot pamphlet is inconclusive as to the intention of the voters because some language in it supports his view as to the meaning of the section, and because it is impossible to know whether the voters accepted the arguments made by one side or the other in approving the measure. As to the first of these assertions, the portions of the pamphlet relied on by Lungren do not bear him out. For example, he cites a statement by the Legislative Analyst as follows: “If the Senate and Assembly neither accept nor reject the person designated to the vacancy by the Governor within 90 days, the person automatically assumes office.” (Ballot Pamp., Gen. Elec. (Nov. 3, 1976), analysis of Prop. 9 by Legislative Analyst, p. 36.) But as we make clear above, this language is subject to the construction advocated by respondents, i.e., that the reference to the Legislature’s acceptance or rejection of the nominee relates only to its failure to act on the nomination. There was no such failure here.

Lungren’s second contention, that it cannot be determined whether the voters agreed with the argument of one side or the other in approving the measure, misses the point; whether one side or the other was more persuasive, the fact is that the proponents and the opponents both premised their arguments on an interpretation of the proposal that is inconsistent with Lungren’s position, and that premise was communicated to the voters. Furthermore, Lungren’s claim proves too much; the assertion that the voters’ motivation cannot be determined from the ballot argument could be made in every case involving a measure adopted by vote of the people.

B. Lungren’s Legislative History Argument

Lungren advances a complicated and unpersuasive legislative history argument on his own behalf. Proposition 9, which embodied section 5(b), originated as Assembly Constitutional Amendment No. 94 (hereafter ACA 94), introduced by Assemblyman (now Senator) Lockyer, who was also one of the persons who later argued in favor of the proposition in the election ballot pamphlet. ACA 94 in turn had its genesis in a bill introduced by the assemblyman in 1974 (Assem. Bill No. 253, hereafter AB 253) which, after undergoing various amendments, passed both houses, but was vetoed by the Governor in 1975. AB 253, in its final version, and ACA 94 were substantially similar to Proposition 9.

Lungren relies on the legislative history of AB 253, arguing that it demonstrates the Legislature intended that the bill would require that both houses must reject a nominee in order to defeat the nomination. He places his reliance largely on a statement made in a staff memorandum drafted while the measure was in the Assembly. In January 1975 AB 253 provided, like the first sentence of section 5(b), that the Governor’s nominee for Treasurer must be confirmed by both houses, but it did not include what is now the second sentence of the section. The bill was approved in this form by the Assembly Committee on Governmental Organization and reported out to the Assembly. However, the Assembly referred the bill back to the committee. A staff report prepared after the referral stated as follows: “It is understood that the author will submit amendments to permit the nominee to take office on the 91st day after nomination unless both Houses vote to reject the nominee.” Thereafter, AB 253 was amended by Assemblyman Lockyer to add the second sentence as it now appears in section 5(b). Lungren contends that the statement in the staff report strongly suggests that in passing AB 253, the Legislature intended to require that both houses must disapprove a nominee in order to reject the nomination.

We disagree. The statement in the staff report represents neither the intent of the Legislature that drafted the section (California Teachers Assn. v. San Diego Community College Dist. (1981) 28 Cal.3d 692, 700-701 [ 170 Cal.Rptr. 817 , 621 P.2d 856 ]), nor that of the electorate in adopting it (People v. Castro (1985) 38 Cal.3d 301, 311-312 [ 211 Cal.Rptr. 719 , 696 P.2d 111 ]).

The first of the cited cases stands for the proposition that in construing a statute we will not consider as evidence in favor of a particular meaning the opinions of individual legislators who voted for the bill nor those of the author, on the ground there is no assurance that other legislators shared such opinions. A fortiori, the “understanding” of an unnamed staff member of a legislative committee, derived from an unnamed source, as to the anticipated contents of a forthcoming amendment to a bill, is not admissible as an indication of the Legislature’s intent in ultimately enacting the measure. This is obviously the correct rule where, as here, it appears that the report was distributed only to the members of the committee that was considering the bill and its author, and was not before the general membership of either house of the Legislature.

Moreover, in Castro, supra, 38 Cal.3d 301 , we rejected as evidence of the voters’ intent in adopting an initiative measure the majority and minority reports of a legislative committee on the ground that these reports did not represent either the intent of the drafters or of the electorate in approving the measure. We stated that because the reports were not included in the voters’ pamphlet they were not helpful in interpreting the intent of the voters, and “we can only speculate on the extent to which the voters were cognizant of them.” ( 38 Cal.3d 301, 312 .) We can state with confidence in this case that the voters were not aware of the obscure staff memorandum now relied on by Lungren.

We conclude that section 5(b) requires confirmation by both houses of the Legislature unless it fails to act on a nomination within 90 days, in which case the nominee is deemed to be confirmed under the second sentence of the provision. When, as here, one house votes to disapprove a nominee, the nomination is rejected, just as in other matters requiring legislative action. This construction fulfills both the major and minor premises of the section: i.e., that both houses must confirm a nominee, but that he may not be rejected merely by the Legislature’s failure to vote on the nomination. Under this interpretation, the nomination process moves expeditiously by virtue of a 90-day time limit on the Legislature’s opportunity to vote and by allowing the Governor to appoint another person upon rejection of his nominee by one house. Only this construction is consistent with the voters’ intention in adopting the constitutional provision.

The alternative writ is discharged and a peremptory writ is denied.

The Treasurer receives, keeps, and approves state bonds (Gov. Code, §§ 12320, 12332), decides the conditions under which bonds will be sold (id., § 16754.3), awards bids for their sale (id., §§ 16752-16754), and pays, cancels and redeems bonds (id., §§ 16770, 16771, 16774). Bonnie J. Graham, a citizen, taxpayer and registered voter joined in the action as a petitioner. It is alleged that the Secretary of State would refuse to attest to the commission or accept Lungren’s bond or file his oath, that the Controller would refuse to draw a warrant to pay his salary, and that Whitney would refuse to acknowledge Lungren’s authority to act as Treasurer or to proceed with pending bond sales. Respondents do not claim that both houses must fail to act before the automatic confirmation provision of the second sentence operates. They posit that failure of one house to act would result in confirmation if the other house has voted to confirm. Their view is that in this event (as well as when neither house votes on the nomination) the nominee has been “neither confirmed nor refused confirmation.” But the use of the word “both” in the second sentence more readily comports with a construction requiring inaction by both houses. Whether respondents’ position is supported by other considerations we need not now decide, because we are concerned here with inconsistent action of the two houses rather than the failure of either or both to vote on the nomination. Respondents assert that the word “both" was added to the second sentence to require that each house of the Legislature must act separately in the confirmation process rather than in joint session. They rely, by way of analogy, on the debates of the United States Senate on the 25th Amendment to the federal Constitution, and on a provision of the California Constitution of 1849 that the Treasurer was to be elected by joint vote of the two houses. (Cal. Const, of 1849, art. V, § 20.) We have serious doubts as to the merits of this argument. Respondents cite only these rather unconvincing analogies to support their claim; neither the language of the second sentence nor its history suggests that it was designed to prevent confirmation by a joint session of the Legislature. The words, “in the event,” which begin the second sentence, are the equivalent of a proviso. (Gunn v. United Air Lines. Inc. (1982) 138 Cal.App.3d 765, 766 [ 188 Cal.Rptr. 302 ]; Brown Newspaper Publishing Co. v. Superior Court (1981) 114 Cal.App.3d 462, 467 [ 170 Cal.Rptr. 611 ]; Young v. Conejo Valley Sanitary Co. (1969) 1 Cal.App.3d 976, 979-980 [ 82 Cal.Rptr. 52 ].) Lungren attempts to distinguish Gunn and Young on the ground that they involved construction of contracts rather than a statute or constitutional provision. However, he does not explain why this distinction would justify attribution of a different meaning to the words “in the event.” He also challenges some of the authorities relied on by respondents for the rule that provisos are strictly construed, but advances no support for his assertion that these cases are distinguishable. Lungren argues that we should narrowly construe section 5(b) because the Governor has the general power to fill vacancies in state office under subdivision (a), section 5, article V of the Constitution, and that power should not be transferred to the Legislature absent express language to that effect. Subdivision (a) provides that the Governor may fill a vacancy by appointment “[u]nless the law otherwise provides.” Clearly, the Governor’s power to fill vacancies is subject to legislative action. A provision similar to subdivision (a) has existed in our Constitution since 1849. (Cal. Const. of 1849, art. V, § 8.) Beginning in 1863, and until the amendment of section 5(b) in 1976, statutes have given the Governor the power to fill vacancies in the office of Treasurer for the balance of the unexpired term without assent of the Legislature. (Stats. 1943, ch. 134, p. 971; Code Amends. 1875-1876, ch. 474, § 1, p. 23.) Section 1775 of the Government Code, which embodied that principle, was amended in 1977 to conform to section 5(b). (Stats. 1977, ch. 96, § 1, p. 517.) Since the Governor enjoys no general power of appointment under our Constitution, the rule cited by Lungren, that restrictions on general powers granted by the Constitution should be narrowly construed, is not applicable here. To bolster his view as to the meaning of the first sentence, Lungren points to the single instance in which section 5(b) has been applied since its adoption in 1976. The section applies to the filling of vacancies in several state-wide offices in addition to Treasurer, including vacancies on the Board of Equalization. In 1982, the Governor nominated a member to fill a vacancy in that body. The nominee was confirmed by both houses of the Legislature within one week of the time his nomination was submitted to them, and was sworn in immediately. We do not view this as an affirmation of Lungren’s interpretation of section 5(b). None of the parties disputes that a nominee confirmed by both houses within 90 days is entitled to enter into office immediately upon his confirmation. The disagreement relates to the consequence of a failure to obtain a favorable vote by both houses within that time. Lungren readily concedes that the logic of his argument leads inevitably to the conclusion that the nominee would be confirmed even if one house rejected the nomination and the other failed to act, because in that event he would have been neither confirmed nor rejected by both houses. Lungren denies this inference and argues that the reason for waiting 90 days, in spite of the fact that approval by a single house is sufficient for confirmation, is to allow the house that acted to change its vote within the 90-day period. The action of the house that had not voted on the nomination would still have some significance within the 90-day period, he argues, because hearings before the second house could reveal problems with the nominee which might cause the first house to change its vote. Lungren would thus reduce the role of the second house to that of possibly revealing information in its hearings on the nomination which might induce the first house to change its vote. The actual vote of the second house, or its failure to vote, would be irrelevant. Again, this is a highly unlikely construction of section 5(b). Furthermore, if the measure was drafted with a view to allowing one house to change its vote within 90 days, it is difficult to see why the same option was not afforded if both houses voted to confirm within that time. Unfavorable information regarding a nominee could surface within the 90-day period whether one or both houses voted on the nomination. Lungren concedes that in this situation the nominee would take office immediately without waiting 90 days. Indeed, as we make clear above, under his interpretation of the section, this ability to assume office immediately without waiting 90 days is the primary function of the first sentence. Section 2 provides: “Whenever there is a vacancy in the office of the Vice President, the President shall nominate a Vice President who shall take office upon confirmation by a majority vote of both Houses of Congress.” The Constitutions of Indiana, Louisiana and South Dakota provide that a majority of each house must confirm. Maryland’s Constitution states that confirmation shall be “by the affirmative vote of a majority of all members of the General Assembly in joint session,” whereas the Wisconsin Constitution provides for confirmation “by the senate and by the assembly.” The ballot title stated: “Requires confirmation by Legislature before Governor’s appointees to fill vacancies in office . . . of. . . Treasurer . . . may take office. If Legislature does not act within 90 days of Governor’s nomination . . . appointees may take office as if confirmed.” The analysis of the Legislative Analyst was as follows: “The State Constitution currently authorizes the Governor to fill vacancies in the offices of. . . Treasurer . . . without approval of the Legislature. “This proposal would require the Governor’s appointee to a vacancy . . . to be approved by a majority of the Senate and Assembly. If the Senate and Assembly neither accept nor re ject the person designated to the vacancy by the Governor within 90 days, the person automatically assumes office.” (Ballot Pamp., Gen. Elec. (Nov. 2, 1976) p. 36.) The rule that the ballot pamphlet is an important aid in determining the intent of the voters in adopting a constitutional amendment or statute is too well settled to require extensive citation of authority. (See, e.g., Long Beach City Employees Assn. v. City of Long Beach (1986) 41 Cal.3d 937, 943, fn. 5 [ 227 Cal.Rptr. 90 , 719 P.2d 660 ]; Brosnahan v. Brown (1982) 32 Cal.3d 236, 248 [ 186 Cal.Rptr. 30 , 651 P.2d 274 ]; White v. Davis (1975) 13 Cal.3d 757, 775 [ 120 Cal.Rptr. 94 , 533 P.2d 222 ].) Prior to the time it was reported out to the Assembly, the bill contained a second sentence providing that the nomination was defeated if the Senate and Assembly did not confirm the nominee within 90 days. This sentence was deleted and the second sentence of the present section 5(b) was inserted in its place. Ultimately, the Assembly passed the bill without the second sentence, but the sentence was restored by the Senate. The Assembly thereafter approved the bill with the second sentence, and it was sent to the Governor. Respondents have submitted a declaration by Parke D. Terry, the staff member of the Assembly Committee on Governmental Organization who wrote the statement on which Lungren relies. He declares that his analysis containing the statement in question was distributed only to the members of that committee and Assemblyman Lockyer, and was not considered on the floor of either house, that it is not uncommon for drafts of proposed amendments to be changed prior to a committee hearing, and that the fact a committee staff member anticipates a certain amendment will be forthcoming does not necessarily mean that the amendment will in fact be submitted to the committee. Further, he states that when Assemblyman Lockyer introduced the amendment to AB 253, it was substantially different from the wording that Terry had anticipated.