Summary
concluding that the "one subject" requirements in Article IV, section 1(d), and Article IV, section 20, "should be given the same meaning"
Summary of this case from Anantha v. ClarnoOpinion
TC 86-C-10367; CA A41468; SC S33296
Argued and submitted October 7, 1986
judgment modified October 21, 1986
In Banc
On certification from the Court of Appeals.
Appeal from Judgment of Marion County Circuit Court, Val D. Sloper, Judge.
Michael D. Reynolds, Assistant Attorney General, Salem, argued the cause for Appellants/Cross-Respondents. With him on the brief were Dave Frohnmayer, Attorney General, and Virginia L. Linder, Solicitor General.
Robert D. Durham, Portland, argued the cause for Respondents/Cross-Appellants. With him on the brief was Kulongoski, Durham, Drummonds Colombo.
Richard M. Stephens, Sacramento, California, argued the cause for Respondents/Cross-Respondents. With him on the brief were Ronald A. Zumbrun, Anthony T. Caso, and Pacific Legal Foundation, Sacramento.
LENT, J.
Judgment of trial court modified as set forth herein.
Linde, J., concurred and filed an opinion in which Gillette, J., joined.
Because of the imminence of the general election to be held on November 4, 1986, all parties and the courts have taken part in the expediting of this case in an attempt to save from injury the initiative process so important to the people of this state. Many pages could be filled with discussion and decision of procedural issues presented by the record before us, but those issues must give way to the overridingly important one of whether the amendment to the state constitution proposed by the subject initiative "embrace[s] one subject only and matters properly connected therewith" as commanded by Article IV, section 1(2)(d), of the Oregon Constitution.
We have heretofore recognized "the great value ascribed to the exercise of the initiative power by the people, by the Oregon Constitution, and the courts." State ex rel Fidanque v. Paulus, 297 Or. 711, 718, 688 P.2d 1303 (1984). See also Ellis v. Roberts, 302 Or. 6, 16, 725 P.2d 886 (1986).
We do not anticipate that interested parties and the courts will address important constitutional issues concerning elections with such haste after the election at hand. Our decision in OEA v. Roberts, 301 Or. 228, 721 P.2d 833 (1986), that the Oregon Constitution requires the Secretary of State to review measures proposed by initiative for compliance with Article IV, section (1)(2)(d), of the Oregon Constitution, is now well known. We are confident that the legislature will provide a statutory time framework for orderly consideration of such decisions prior to the next election year. See State ex rel Bunn v. Roberts, 302 Or. 72, 726 P.2d 925 (1986) (decided by order on October 7, 1986, and supporting opinion on October 13, 1986) (concurring opinion by Peterson, C.J.).
I.
Defendants Phillips, Donison and Vandenberg are the chief petitioners proposing the amendment, the text of which is:
The text of the proposed amendment would repeal "Section II" of Article XI of the Oregon Constitution. There is no such section. In this respect we express no opinion on the effect of the subject measure should the voters approve it. Because the parties to the case at bar have assumed that Article XI, section 11, of the Oregon Constitution is the section to which the subject measure speaks, we shall also assume for the purpose of this opinion only that section 11 is involved.
"Be it enacted by the people of the State of Oregon:
"Section II [sic], Article XI of the Constitution of the State of Oregon is repealed, and the following section is adopted in lieu thereof.
"Section II [sic].
"(1)(a) The maximum rate of ad valorum [sic] taxes levied against any property for the fiscal year beginning July 1, 1987, shall not exceed two percent of the assessed value of such property, or the rate levied for the fiscal year beginning July 1, 1985, whichever is less.
"(b) Revenues produced by ad valorum [sic] taxes for the fiscal year beginning July 1, 1987, shall be distributed among taxing units in the same proportions as existed for the fiscal year beginning July 1, 1986.
"(c) The maximum rate of ad valorum [sic] taxes levied against any property for the fiscal year beginning July 1, 1988, and for each fiscal year thereafter, shall not exceed one and one-half percent of the assessed value of such property, or the rate levied for the fiscal year beginning July 1, 1985, whichever is less.
"(2) The limitation imposed by subsection (1) shall not apply to:
"(a) Ad Valorum [sic] taxes or special assessments levied to pay bonded indebtedness or interest thereon.
"(b) Non-operating serial levies that exist on July 1, 1986, and extend beyond July 1, 1987.
"(3) The assessed value of any property shall not increase in any one year by more than two percent above the prior year's assessed value.
"(4) All property sold, purchased, newly constructed, improved, or subject to change of ownership of [sic] eligibility for a specially assessed value subsequent to the fiscal year beginning July 1, 1987, shall be assigned the assessed value it had, or would have had in the case of newly constructed or improved property, for the fiscal year beginning July 1, 1985, adjusted for the intervening period under provisions of subsection (3).
"(5)(a) Notwithstanding subsection (1), the state, each city, county, special district, school district, or other taxing unit of or within the state may levy a new ad valorum [sic] tax rate or increase an existing ad valorum [sic] tax rate only upon approval of a majority of the legal voters of the taxing unit who vote on the question.
"(b) A question authorized by this subsection shall be submitted to the voters in a form specifying the reason for the new or increased tax rate, the amount of revenue it is intended to produce, and the time period during which it is to be in effect.
"(c) Elections authorized by this subsection shall be limited to the third Tuesday in May and the first Tuesday after the first Monday in November."
After the chief petitioners had filed their petition with the Secretary of State, she transmitted it to the Attorney General for preparation of a ballot title. The Attorney General certified a ballot title to the Secretary of State on January 23, 1986. On February 21, 1986, plaintiffs commenced in circuit court this case against the chief petitioners, the Secretary of State and the Attorney General, asserting that the proposed amendment offended Article IV, section 1(2)(d), of the Oregon Constitution, which provides:
"An initiative petition shall include the full text of the proposed law or amendment to the Constitution. A proposed law or amendment to the Constitution shall embrace one subject only and matters properly connected therewith."
With the acquiescence of all parties, the prosecution of the case was suspended pending final disposition of another case then pending on appeal. Final disposition of that case was achieved by decision of this court in OEA v. Roberts, 301 Or. 228, 721 P.2d 833 (1986), which became final on July 31, 1986.
After our announcement of the decision in OEA v. Roberts, supra, but before it became final, the Secretary of State certified the subject measure for the ballot, designating it Ballot Measure Number 9.
On August 15, 1986, plaintiffs wrote to the Secretary of State inquiring whether she intended to review the measure for compliance with Article IV, section 1(2)(d), of the Oregon Constitution (hereinafter subsection (2)(d)). She answered that she did not.
Plaintiffs then filed a motion for summary judgment that the measure violated subsection (2)(d) or, alternatively, that the court order the Secretary of State to review the measure for compliance with the subsection.
The Secretary of State and the Attorney General opposed plaintiffs' motion and filed their own motion for summary judgment on their affirmative defense of laches.
Defendant chief petitioners filed their written opposition to plaintiffs' motion, joining in the assertion of laches and arguing alternatively that the court should not undertake review of the measure for compliance with subsection (2)(d) without first requiring the Secretary of State to do so. Alternatively to that position, they argued that if the court were to reach the question, it should hold that the measure does not violate the constitutional command.
The trial court disposed of the matter on the cross motions for summary judgment, entering judgment as follows:
"[T]he Secretary of State is required to perform the one subject review of Ballot Measure Number 9 pursuant to Article IV § 1(2)(d) of the Oregon Constitution AND OEA v. Roberts, 301 Or. 228 (1986)."
The Secretary of State and the Attorney General promptly appealed to the Court of Appeals, and plaintiffs timely cross appealed. The Court of Appeals certified the appeal to this court pursuant to ORS 19.210 and ORAP 17.05, and this court timely accepted the certified appeal.
The Secretary of State and the Attorney General contend on appeal that the trial court erred in ordering the Secretary of State to review the subject measure and that the sole question on appeal is
"whether Ballot Measure 9 contains more than one subject, in violation of Article IV, section 1 (2)(d), and, hence, the Secretary of State's approval of the measure was invalid."
The Secretary of State and Attorney General had originally urged in the trial court that if the court concluded that their defense of laches was not well taken, the trial court should remand the matter to the Secretary of State to conduct review. On the hearing on the motions for summary judgment, however, they took the position that the court should decide whether the subject measure offends subsection (2)(d).
The Secretary of State and Attorney General had originally urged in the trial court that if the court concluded that their defense of laches was not well taken, the trial court should remand the matter to the Secretary of State to conduct review. On the hearing on the motions for summary judgment, however, they took the position that the court should decide whether the subject measure offends subsection (2)(d).
Plaintiffs answer the appeal by contending that the trial court did not err, but that the trial court
"should have addressed Secretary Roberts' second error described above [that she failed to declare the initiative invalid], and declared that Ballot Measure 9 embraces more than one subject in violation of the Oregon Constitution."
In their cross appeal, plaintiffs contend that the trial court erred in failing to instruct the Secretary of State to declare that the subject measure is invalid and in refusing to address the merits of that claim by plaintiffs. Plaintiffs have addressed the merits both by their written brief and in oral argument.
In their answering brief as respondents, the chief petitioners urge that the trial court properly disposed of the matter and that the court should not conduct the review that is the Secretary of State's initial responsibility. They argue that if this court does reach the question whether the subject measure offends subsection (2)(d), we should decide that it does not. They have fully briefed that question, as have the other parties.
Although we should much prefer that this case arose on judicial review (in some proper form) of a ruling of the Secretary of State in the first instance and that this case reached us after ordinary judicial scrutiny in the trial court and Court of Appeals, we have concluded that the circumstances call for a decision now by this court on the ultimate issue. In so proceeding, we do not intend that this approach serve as precedent for ignoring such doctrines as primary agency jurisdiction, exhaustion of administrative remedies, Administrative Procedure Act review of orders in noncontested cases, basis in the pleadings for judgment rendered, preservation of error, etc. Rather, those who would resort to the courts to contend for or against claims such as that now confronting us are warned that our allegiance to normal procedure will not ordinarily be governed by expedience.
II.
All parties are agreed that this court has never addressed the meaning of "one subject only and matters properly connected therewith" as used in subsection (2)(d). The Secretary of State, the Attorney General and the chief petitioners all urge that the same meaning should be ascribed to that text as this court has given to the text of Article IV, section 20, of the Oregon Constitution (hereinafter section 20), which provides in pertinent part:
"Every Act shall embrace but one subject, and matters properly connected therewith, which subject shall be expressed in the title."
Their argument is grounded in the strikingly similar text of each constitutional provision: "one subject only" as compared with "but one subject" and the absence of any "legislative history" in the adoption of subsection (2)(d) in 1968 to indicate the sponsors of the 1968 amendment meant something different.
Plaintiffs counter that the language of subsection (2)(d) ought to be construed more narrowly because of the difference in the way in which the text of an Act is fixed from the way in which the text of an initiative measure is chosen. Before directly addressing that argument of plaintiffs, we shall briefly review the cases brought to our attention construing section 20.
In Lovejoy v. Portland, 95 Or. 459, 188 P. 207 (1920), the defendant City of Portland attacked a state statute as being offensive to section 20. This court described the statute:
"This enactment * * * thoroughly covers the field of insurance regulation and supervision. It regulates the conditions under which business may be commenced and the manner in which it may be conducted. It regulates the organization of the insurance department, and it prescribes the jurisdiction and defines the powers of the state insurance commissioner. It deals with the mode of organizing local companies, the admission of foreign companies, the nature of the investments of local and foreign companies, and examinations and reports of companies. It contains various provisions to insure the solvency of insurance companies, and it affords means for excluding companies of impaired or doubtful solvency. It prescribes the qualifications of agents and provides for licensing them and for revoking the licenses of agents who violate the law or are otherwise unfit. It specifies the kinds of insurance which various companies may write, and it designates the form and manner in which policies may be written. It contains rules designed to secure equal rights and opportunities between insurance companies, to prevent discrimination in rates, and to compel insurers to treat all insured alike. It furnishes safeguards against excessive or unjust premium rates, and it contains prohibitions against combinations between companies or agents for any purpose detrimental to the public welfare. In a word, the legislature has attempted to place the insurance business completely under the supervision of the state, and to accomplish the purpose the state has adopted an elaborate and comprehensive statute covering the whole field of insurance regulation and supervision."
95 Or at 461. In holding that the statute did embrace but one subject and matters properly connected therewith, the court noted that section 20 was a provision common to most state constitutions and addressed the reason for the existence of such restraints:
"This section of the Constitution was designed to do away with the several abuses, among which was the practice of inserting in one bill two or more unrelated provisions so that those favoring one provision could be compelled, in order to secure its adoption, to combine with those favoring another provision, and by this process of log-rolling the adoption of both provisions could be accomplished, when neither, if standing alone, could succeed on its own merits. Another abuse which developed in legislative bodies was the practice of concealing from the members of the legislature the true nature of the proposed law by giving it a false and misleading title, and to prevent surreptitious legislation in this manner is one of the objects of the Constitution. These and similar abuses inspired the adoption of Article IV, Section 20: Northern Counties Trust v. Sears, 30 Or. 388, 400 ( 41 P. 931, 35 L.R.A. 188); Moore-Mansfield Construction Co. v. Indianapolis R. Co., 179 Ind. 356 ( 101 N.E. 296, Ann. Cas. 1915D, 917, 44 L.R.A. (N.S.) 816); Johnson v. Harrison, 47 Minn. 575 ( 50 N.W. 923, 28 Am. St. Rep. 382); County Commissioners v. Pocomoke Bridge Co., 109 Md. 1 ( 71 A. 462, 16 Ann. Cas. 874)."
95 Or at 465-66.
The opinion continues, relying on earlier cases, to hold that section 20 should be liberally construed to uphold legislation.
"While Article IV, Section 20, is mandatory and failure to comply with it renders a statute void, yet this section of the Constitution should be reasonably and liberally construed to sustain legislation not within the mischief aimed against: State v. Shaw, 22 Or. 287, 288 ( 29 P. 1028); 25 R.C.L. 85. Every legislative act is presumed to be constitutional, and the conflict between a statute and the Constitution should be palpable before the legislative enactment is held to be void on the ground that it embraces more than one subject or because the subject is not sufficiently expressed in the title: Pacific Elevator Co. v. Portland, 65 Or. 349, 384 ( 133 P. 72, 46 L.R.A. (N.S.) 363)."
95 Or at 466. The opinion proceeds with some further language about the meaning of the word "subject," but that language concerns the meaning of the word as used in the clause "which subject shall be expressed in the title." In this respect the court's language is not strictly pertinent to today's inquiry, and we have no occasion to endorse it in this opinion.
In Eddins v. Wasco County et al, 189 Or. 184, 219 P.2d 159 (1950), a state statute authorized the defendant county to issue revenue bonds
"for the purpose of constructing, maintaining and operating a toll bridge across the Columbia River near The Dalles, Oregon, together with approaches and connecting highways, and also including the acquisition of a ferry and appurtenant properties in close proximity to the site of the bridge."
189 Or at 185. The plaintiff contended that the statute must fail for embracing more than one subject, i.e., "the construction of a toll bridge and the acquisition of a ferry." One of the plaintiff's arguments was that the county had no need to acquire the ferry when it was building a bridge to cross the river. This court held that the acquisition of the ferry was germane to and properly connected with the subject of the statute, which was transportation across the river. The opinion cited Lovejoy v. Portland, supra, as correctly stating the law on the point.
In Warren v. Marion County et al, 222 Or. 307, 353 P.2d 257 (1960), the plaintiff attacked a statute that enabled the county by ordinance to develop land use planning and provide a building code. This court upheld the statute and the ordinance, stating that the building code provisions were "related to the subject of county planning" and "properly connected therewith" within the meaning of section 20. 222 Or at 323.
Other cases have been cited for the same effect as the three just reviewed, but we find that they contain nothing that would add to our discussion.
The cases we have reviewed and the cases cited therein as authority demonstrate that section 20 has been liberally construed to uphold legislation that does not fall within the mischief at which the constitution is directed. We do not perceive that plaintiffs seriously dispute that.
Rather, as noted above, plaintiffs believe that a narrower and more strict construction should be given to subsection (2)(d) because of the differences in the nature of the legislative processes that on the one hand occur in the legislature and on the other hand occur in the initiative process. They argue that measures presented for enactment by the legislature are "fine-tune[d]" through the input of persons interested on all sides, the scrutiny of trained staff and the opportunity for amendment before an act is passed. They point to the further safeguard of possible gubernatorial veto to enforce the one-subject requirement. They point out that a measure proposed by initiative presents a "package" that must be accepted or rejected by the voters as presented. They assert:
"For this reason the `package' must present a specifically defined policy choice on one subject only in order to insure that the voters have actually expressed their will on the policy choice."
They cite Fine v. Firestone, 448 So.2d 984 (Fla 1984), which does support their argument that the initiative process does not provide that "filtering legislative process" that takes place with respect to a statute or even a proposed constitutional amendment originating in the legislature.
The chief petitioners counter that Fine v. Firestone, supra, stands alone in that position. They argue that many states have similar constitutional provisions and cite cases from Alaska, Oklahoma, North Dakota, Illinois, Arizona, Massachusetts, Missouri, Louisiana and California as standing for a liberal and nonrestrictive test for compliance with the constitution. We shall not review in these pages all of those cases. It does appear, however, that the Florida court stands alone. In Yute Air Alaska, Inc. v. McAlpine, 698 P.2d 1173 (Alaska 1985), an initiative measure was challenged, inter alia, as violating Alaska's constitutional provision requiring a proposed initiative measure to speak to but one subject. The court quoted from Gellert v. State, 522 P.2d 1120, 1123 (Alaska 1974), its standard:
They cite Gellert v. State, 522 P.2d 1120 (Alaska 1974); Boucher v. Engstrom, 528 P.2d 456 (Alaska 1974); North Slope Borough v. Sohio Petroleum Corp., 585 P.2d 534 (Alaska 1978); Yute Air Alaska v. McAlpine, 698 P.2d 1173 (Alaska 1985); In re Initiative Petition No. 319, State Question No. 563, 628 P.2d 222 (Okla 1984); SunBehm Gas, Inc. v. Conrad, 310 N.W.2d 766 (ND 1981); Great Northern Ry. Co. v. Duncan, 42 N.D. 346, 176 N.W. 992 (1919); Coalition for Political Honesty v. State Board of Elections, 83 Ill.2d 236, 415 N.E.2d 368 (1980); State ex rel Jones v. Lockhart, 76 Ariz. 390, 265 P.2d 447 (1953); Massachusetts Teachers Association v. Secretary of the Commonwealth, 384 Mass. 209, 424 N.E.2d 469 (1981); Buchanan v. Kirkpatrick, 615 S.W.2d 6 (Mo 1981); State ex rel Kemp v. City of Baton Rouge, 215 La. 315, 40 So.2d 477 (1949); Amador Valley Joint Union High School District v. State Board of Equalization, 22 Cal.3d 208, 149 Cal.Rptr. 239, 583 P.2d 1281 (1978).
"[An] act should embrace some one general subject; and by this is meant, merely, that all matters treated of should fall under some one general idea, be so connected with or related to each other, either logically or in popular understanding, as to be parts of, or germane to, one general subject."
698 P.2d at 1180-81. The court stated that this formulation was similar in breadth to those generally adopted by other state courts, citing Ruud, No Law Shall Embrace More Than One Subject, 42 Minn L Rev 389, 393-95 (1958). The Alaska court specifically rejected the Florida rule:
"The only narrower standard of which we are aware is that adopted in Florida relating to that state's one subject rule applicable to initiatives."
698 P.2d at 1181. The opinion goes on to state that the court does not believe the Florida rule to be "an apt one."
With respect to proposed amendments to the state constitution, Article XII, section 2 (b), of the Missouri Constitution provides:
"All amendments proposed by the general assembly or by the initiative shall be submitted to the electors for their approval or rejection by official ballot title as may be provided by law. * * * No such proposed amendment shall contain more than one amended and revised article of this constitution, or one new article which shall not contain more than one subject and matters properly connected therewith." (Emphasis added.)
Article III, section 50, of the Missouri Constitution provides:
"Petitions for constitutional amendments shall not contain more than one amended and revised article of this constitution, or one new article which shall not contain more than one subject and matters properly connected therewith * * *." (Emphasis added.)
Article III, section 50, of the Missouri Constitution provides:
"Petitions for constitutional amendments shall not contain more than one amended and revised article of this constitution, or one new article which shall not contain more than one subject and matters properly connected therewith * * *." (Emphasis added.)
In Buchanan v. Kirkpatrick, 615 S.W.2d 6 (Mo banc 1981), the measure challenged for violation of the one-subject rule was described by the court as follows:
"Generally stated, the central purpose of Amendment No. 5 is to limit taxes by establishing tax and revenue limits and expenditure limits for the state and other political subdivisions which may not be exceeded without voter approval. Amendment No. 5 is popularly described as `the tax and spending lid' amendment, words which also reflect its central purpose. To accomplish the central purpose, the amendment authorizes certain formulas for establishing the limits and provides a method for the repayment of taxes collected in excess of the limit. The amendment further seeks to prohibit the state from avoiding the defined limit or limits by the shifting of governmental responsibilities or the shifting of responsibility for payment for either existing or newly created governmental responsibilities. Provision is made for emergencies. Provision is made in section 23 of the amendment to give taxpayers and political subdivisions standing to enforce the amendment in the courts."
615 S.W.2d at 13. The court rejected the challenge, holding that the measure did not deal with separate subjects:
"They are subjects not only properly, but vitally connected with the central or primary purpose of the amendment to control taxes and expenditures. The same is true of provisions relating to the repayment of excess revenues collected and provisions relating to standing in court to enforce the act. All of these items are properly connected to the single controlling purpose of the amendment: to limit taxes and governmental expenditures within the state of Missouri."
As will be seen in part III of this opinion, we do not believe that our constitutional text allows analysis in terms of proper connection of "subjects"; we quote from the Missouri Supreme Court's decision because the topic of the measure there examined is so similar to that in the case at bar.
We find the above decisions from Alaska and Missouri to be representative of the other decisions cited by the chief petitioners, and we agree that the Florida court stands alone in its reasoning that there should be a stricter construction of the constitutional provision where an initiative measure is involved. Indeed, the Alaska and Missouri courts take exactly the opposite tack, reasoning that where the people, whose power in the final analysis is plenary, propose a constitutional amendment, or any other measure, more leeway is to be given in this kind of challenge than when the legislature and its trained staff make such a proposal.
Prior to the adoption of subsection (2)(d) in 1968, this court had ruled that section 20 was just as applicable to initiative measures as it was to those adopted by the legislature. Anthony et al v. Veatch et al, 189 Or. 462, 501, 220 P.2d 493, 221 P.2d 575 (1950).
We find nothing in the extremely minor difference in wording of subsection (2)(d) and section 20 to indicate that the two texts should be construed or interpreted to have different meanings. There is no legislative history to indicate that the proponents of the 1968 amendment meant something different. The reasoning of the Florida court that the rule should be different does not persuade us. We conclude, accordingly, that the two provisions should be given the same meaning.
III.
What is that meaning? The constitutional text limits a proposal for amendment to "one subject only and matters properly connected therewith." A measure must first be scrutinized to determine whether it embraces more than one subject. If it does, it offends the constitutional limitation even if the subjects are "properly connected," and that is the end of the inquiry. If it does not, the single subject must be identified. When that is done, and if the proposal embraces no other matters, there is no need to inquire into proper connection. A ready example would be a proposal to amend the state constitution by repealing Article I, section 30. Another might be a proposal to amend Article V, section 7, by changing the term of office for the Governor from four years to some other period. In each of these examples there is obviously only one subject, and there are no other matters embraced by the proposal.
"No law shall be passed prohibiting emigration from the state." Or Const, Art I, 30.
If the proposal embraces one subject only and also other "matters," then, and only then, it must be determined whether those other matters are properly connected with the subject.
The chief petitioners urge that this is a property tax relief proposal. This it may do, and this may be their purpose, but purpose is not to be confused with subject. We gather that the subject of this proposal is ad valorem tax limitation. Indeed, the subject of present Article XI, section 11, of the Oregon Constitution has been given the lead line "Tax limitation" in volume 5 of the Oregon Revised Statutes ever since that section was adopted by the people as a constitutional amendment in 1962.
IV.
We come now to consideration of plaintiffs' contention that the subject measure offends the one-subject rule. They argue that the subject measure both abandons the present "dollar-based" limitation on property tax collections and substitutes a "rate-based" limitation. They argue that as a result a municipality might be able actually to increase the dollar amount it could collect without an authorizing vote of the people that would be required for such an increase under present law. That might be so; we express no opinion as to the accuracy of the assertion, but we do not conclude as do plaintiffs that the change from a dollar limit to a percentage of value limit embraces matters not connected with ad valorem tax limitation.
Plaintiffs argue that the proposed change in the method of limitation is a case of the kind of log-rolling that the constitution seeks to outlaw. They argue:
"Those who favor the present dollar-based limitation system may wish to oppose a new rate-based tax limitation system, but may also desire to approve a limitation on the annual increases in the assessed valuation of property. Because Ballot Measure 9 joins a repeal of the present dollar-based property based tax limitation system together with a new rate-based limitation system and a limitation on assessed valuation increases, the voter will be forced to approve the new rate-based system in order to gain a limitation on annual assessed value increases on his property. The compulsion on such a voter is precisely the evil which the one subject limitation in the Oregon Constitution was designed to prevent."
We do not agree that this constitutes log-rolling. Both of these "matters" are not only "properly connected" with ad valorem tax limitation, but are complementary methods to attempt putting beyond statutory reach circumvention of the limitation.
Plaintiffs argue that the provision for distribution of taxes for fiscal year 1987-88 is foreign to the rest of the subject measure. It strikes us that assuring the voters how revenues are to be distributed for the first fiscal year following the changes proposed by the subject measure is a matter properly connected with the major portions of the measure concerning limitation on revenue to be achieved through the property tax. To argue, as do plaintiffs, that log-rolling will occur because voters will be forced to approve the rate limitation in order to obtain approval of the revenue distribution is to put the cart long down the road from the following horse.
Plaintiffs likewise attack section 5 of the subject measure, contending that the "override" provision could force a voter who favored the existing dollar-based limitation system to give up that position in order to gain the override feature described in section 5. We see the override provision as a matter properly connected with the chief petitioners' attempt to place more control in the hands of the electorate over the amount of property taxes to be collected. The feature is germane to ad valorem tax limitation.
Plaintiffs argue that the limitation of elections to the dates fixed under present law for the primary and general elections violate the constitution because a voter who might favor most of the measure will have to take with the good parts an undesired modification of the authority of local governments to conduct elections regarding property tax issues. That this may well be true may present that voter with a difficult choice, but if the limitation on elections were not found in the measure, another voter might well face the difficult choice of desiring to vote for most of the measure but not wanting to do so unless it also contained such a limitation. Most measures dealing with a complex and emotional subject will present those kinds of choices, but that does not mean that the measure embraces matters not connected with its subject.
We agree with the Secretary of State and the Attorney General that such competing choices as may inhere in the subject measure are not of such a nature as to force voters strongly opposed to some provision of the measure to vote for this particular measure just to attain ad valorem tax limitation.
V.
The technical disposition of the case arrives. The trial court ordered the Secretary of State to conduct one-subject review. In light of our decision, there is no necessity for that course of conduct. On the other hand, we cannot say that the Secretary of State had no duty to conduct one-subject review under our recent decisions discussed above and, therefore, are hard put to say that the circuit court erred and should be reversed.
The claims advanced in plaintiffs' amended complaint below were for a declaratory judgment and injunctive relief, judicial review under the Administrative Procedures Act, review under the election laws and a petition for a writ of mandamus. All claims, in the final analysis, depended on whether the measure violated the one-subject rule. We have here declared that it did not. We believe the proper disposition of the case now is to describe our decision as a modification of the judgment of the trial court.
The judgment of the trial court is modified as set forth herein.
A proposed initiative must "embrace one subject only and matters properly connected therewith." Or Const, Art IV, § 1 (2)(d). I agree that the proposal before us passes that test. I write separately to draw attention to some questions about the usefulness of the test.
The quoted clause employs two distinct terms: "subject" and connected "matters." The text implies that connected "matters" are something less than a "subject." It does not say "one subject only and other subjects properly connected therewith." The connected "matters" cannot be separate "subjects," because then the proposal would have two subjects. From this one might infer that, whenever "matters" included in a proposal could intelligibly be enacted on their own, without another "subject" to which they are connected, these matters themselves are a separate subject and must be initiated as such. "Matters" means provisions ancillary to accomplishing the one substantive policy that is the "subject" of the measure.
Under this reading of the constitution, it is doubtful that the proposal before us qualifies to be put on the ballot, because the proposed limitation on increased assessments, the ceiling on tax rates, and the shift from a dollar-based limitation to a rate-based limitation of property taxes could each be a subject of an independent proposal rather than only a "matter properly connected" with a subject. It would make as much sense to describe the subject of this measure as a shift from an expenditure-based to an assessment-based limitation on property taxes, and the specified percentages as a "matter properly connected" with such a shift, as the reverse.
It therefore would focus on the wrong question, as the court's opinion notes, to examine only whether the percentage rate limitations and the limitation on assessed values are "properly connected." To repeat, the constitution does not allow combining two or more subjects in one measure if the subjects are "properly connected"; the constitution allows only one subject and ancillary "matters." But characterizing as a separate "subject" any provision that could be enacted by itself rather than only as an ancillary "matter" would make the "one subject" rule too restrictive, far more restrictive than the experienced political leaders who put the rule into state constitutions throughout the country could possibly have intended by the phrase, because it would make impractical demands on most ordinary legislation. Although ability to stand alone is not the proper test, sponsors of initiative measures, the Secretary of State, and this court cannot escape the problem of identifying the "subject" of a measure by focusing on how well its separate parts are "connected." But I question whether characterizing the "one subject" of a measure is a usable legal test for its constitutionality, or whether it simply compels endless conceptual manipulation, controversy, and litigation unless it is refined.
In the spiritual drawn from Ezekiel 37:1-15, whether the thigh bone is connected to the backbone and it to the neck bone or vice versa, I suppose the narrowest definition of the song's "subject" is the skeleton. Its intended subject is rather loftier.
The Linnean system of classifying plants and animals offers a familiar illustration. A measure to control the anopheles mosquito deals with a particular genus. Try to control gypsy moths as well as mosquitos, and your "subject" skips past families and orders to the class of insects. Add rattlesnakes and ragwort, and the measure simply vaults past the separate plant and animal kingdoms to the "subject" of controlling "dangerous organisms," or perhaps "pests." Measures to preserve "endangered species," "the environment," or "public health," or to prevent "pollution" are similar examples. "Regulation of insurance," as in Lovejoy v. Portland, 95 Or. 459, 188 P. 207 (1920), is another.
The indeterminacy of the term "subject," one of the most abstract words in the language, cannot be overcome by synonyms, paraphrases, and tautological formulas. It can either be replaced or supplemented by a formula that sets out some concrete goals or operational directives, or measures will be left to continual case-by-case decisions under standards so meaningless that it is difficult to avoid ad hoc administrative or judicial reactions to the merits of individual measures.
In fact, it is difficult to find proposals that could not be verbally brought under one "subject," though common sense rebels. The majority opinion says, quite plausibly, that the subject of this measure is "ad valorem tax limitation." Counsel at the oral argument, reaching for an outrageous example, suggested that the initiative to decriminalize marijuana offenses, also on the November ballot, could not have been combined with the tax limitation measure in one "subject." But what if a proposed tax limitation measure provided that the lost revenue would be made up by another tax, say a tax on an activity now illegal, such as gambling or the cultivation and sale of marijuana, and in a "connected" step the measure repealed the existing prohibition of the activity? Would that really be so different from the combination sales tax, property tax and school finance measure referred to the voters in 1985?
The constitutional words being too abstract, courts rightly have sought their meaning in what the limitation to one subject was intended to accomplish or to prevent. See 1A Sutherland Statutory Construction 7-20, §§ 17.02-.04 (4th ed 1985). The majority quotes from Lovejoy v. Portland. There, the court examined the requirement of Article IV, section 20, that acts of the legislature have a single subject "which subject shall be expressed in the title." The court identified two distinct purposes: one to curb what it called "log-rolling," the other to avoid misunderstanding and concealment of the contents of an act if it dealt with two or more subjects. But the judicial task under a "one subject" formula is hardly eased by a test that calls on judges to discern whether a measure confused legislators who voted for it or whether a measure gained passage only by combining supporters of several provisions that could not muster majorities on their own. When long after a measure is enacted such an attack is leveled by a party more concerned with escaping the effect of the law than with the political process as such, it is not surprising that this court in 127 years has never invalidated a law under the "one subject" rule of Article IV, section 20.
Rules governing the germaneness of provisions combined in a single bill are appropriate constraints within the legislative process. They are likely to be routinely respected in daily practice, and objections can be and often are raised by points of order in the legislative body. When a majority overrides the objection, at least it acts with notice of the contents of the bill and upon majority support for combining its several parts. It is doubtful whether a court later should invalidate such an institutional decision of the legislature, and Oregon courts in fact have not done so.
The Commission on Constitutional Revision in 1962 proposed substituting "bill" for "act" in the "one subject" requirement of Article IV, section 20. See State ex rel Fidanque v. Paulus, 297 Or. 711, 720, 688 P.2d 1303 (1984) (Linde, J., dissenting). As I there wrote:
"The rule limiting proposed laws to one subject is not concerned with constitutional limitations on the substance of public policies, such as tax limitations or the guarantees of individual rights. It is concerned with the lawmaking process itself. It aims to enhance the likelihood that distinct policies will be judged rationally on their individual merits rather than being packaged to attract support from legislators or constituencies with special interest in one provision and no worse than indifference toward other unrelated ones.
"Once the legislative process is completed, however, it is at least debatable whether an otherwise valid provision of law should be vulnerable to challenge years after enactment, by someone whose personal interest is not in the political process but in escaping the effect of the law. * * *"State ex rel Fidanque v. Paulus, supra, 297 Or at 721.
The initiative process, however, is quite different from that in the Legislative Assembly, and it poses greater institutional risks both for the internal rationality of measures and for the comprehension of the citizens who must vote on them. First, once the sponsors have drafted and submitted the text of a measure, there is no further opportunity to correct, refine, or clarify that text, even to fit it into existing statutes, no matter what obscurity, errors, or unintended implications the sponsors themselves or anyone else may discover in it. Second, there is nothing comparable to the hearings of legislative committees to raise questions and allow the presentation of different viewpoints in order to eliminate unacceptable aspects of a proposal and to produce an improved or at least defensible compromise with the help of staff members and other knowledgeable persons. Third, efforts to secure or to defeat passage of an initiative measure by the voters are bound to rely on slogans and oversimplified appeals that can be broadcast to the public at large by advertising techniques rather than on the give and take of debate among legislators chosen to represent, and sensitive to, constituents with divergent interests.
See, e.g., Rogers v. Roberts, 300 Or. 687, 717 P.2d 620 (1986)(Lent, J., concurring)(sponsors' stated purpose cannot override what their chosen language would accomplish); State v. Quinn, 290 Or. 383, 623 P.2d 630 (1981) (declaring Oregon's 1978 death penalty statute unconstitutional due, in part, to its failure to harmonize with the then existing criminal statutory scheme).
Several states provide that before an initiative proposal is put on the ballot, the legislature has an opportunity to consider and enact a legislative version of the proposed measure, or to put its version on the ballot along with the version proposed by the initiative. See, e.g., Me Const, Art IV, Pt 3, § 18; Mass. Const, Art XLVIII, Pts II, III, IV.
These differences are so substantial that they raised doubt whether under some circumstances replacement of the legislative process by the initiative would be inconsistent with the "Republican Form of Government" required for every state by Article IV, section 4, of the United States Constitution. See Kiernan v. Portland, 57 Or. 454, 111 P. 379, 112 P. 402 (1910), Kadderly v. Portland, 44 Or. 118, 74 P. 710, 75 P. 222 (1903). It is these differences, also, that recently led the Florida Supreme Court to hold an initiative proposal to a tighter "one subject" standard than a proposal that has gone through a normal deliberative process. Fine v. Firestone, 448 So.2d 984 (1984). Although Oregon takes pains to provide carefully written explanatory ballot titles and to include both explanatory and argumentative statements about each measure in the official Voters Pamphlet sent to registered voters, the risks of mistake, confusion, and emotional responses to a single, yes-or-no choice divorced from the context of other laws or alternative measures cannot wholly be avoided.
Judicial examination of this issue has been rare since the United States Supreme Court held that federal enforcement of the requirement was lodged in Congress rather than in the federal courts. Pacific States Teleph Teleg Co. v. Oregon, 223 U.S. 118, 32 S Ct 224, 56 L Ed 377 (1912). But see VanSickle v. Shanahan, 212 Kan. 426, 511 P.2d 223 (1973) (reviewing Kansas constitutional amendment under US Const, Art IV, § 4.)
"It is apparent that the authors of article XI realized that the initiative method did not provide a filtering legislative process for the drafting of any specific proposed constitutional amendment or revision. The legislative, revision commission, and constitutional convention processes of sections 1, 2 and 4 all afford an opportunity for public hearing and debate not only on the proposal itself but also in the drafting of any constitutional proposal. That opportunity for input in the drafting of a proposal is not present under the initiative process and this is one of the reasons the initiative process is restricted to single-subject changes in the state constitution. The single-subject requirement in article XI, section 3, mandates that the electorate's attention be directed to a change regarding one specific subject of government to protect against multiple precipitous changes in our state constitution. This requirement avoids voters having to accept part of an initiative proposal which they oppose in order to obtain a change in the constitution which they support. An initiative proposal with multiple subjects, in which the public has had no representative interest in drafting, places voters with different views on the subjects contained in the proposal in the position of having to choose which subject they feel most strongly about."
Fine v. Firestone, 448 So.2d 984 (1984).
I doubt that, if these risks are to be avoided, the "one subject" formula of Article IV, section 1(2)(d), standing alone, can accomplish it. Nevertheless, we cannot relieve the Secretary of State, or whoever may be given this responsibility by the election laws, from deciding whether a proposed measure contains provisions aiming at such disparate substantive objectives that to bring them within one abstract "subject" subordinates political reality to a verbal fiction. I have no serious disagreement with the majority's conclusion that the formulas stated in Article IV, section 1(2)(d) and section 20, are so similar that the difference is not worth pursuing in this case, though perhaps their different functions could be fleshed out by legislation. The question for others than this court to consider is whether the objectives sought by the present "one subject" formulas are sufficiently clear to be more precisely stated in terms that sponsors of initiatives, the Secretary of State and, if this is deemed necessary, courts can apply.
Gillette, J., joins in this concurring opinion.