Opinion
Argued June 4, 1973 —
Decided July 24, 1973.
Appeal the Superior Court, Appellate Division.
Mr. David J. Goldberg and Mr. William Miller argued the cause for appellant Apportionment Commission of New Jersey.
Mr. Henry Ramer argued the cause for respondents Davenport, et al. Mr. Donald L. Berlin argued the cause for respondents Hummel, et al. ( Messrs. Lieb, Teich and Berlin, attorneys).
Mr. Alfred J. Lechner, Jr., argued the cause for respondent Dietz.
Mr. Walter F. Hoffman argued the cause for amicus curiae Mazzacca.
In Scrimminger v. Sherwin, 60 N.J. 483 (1972), we held invalid the plan for apportionment of the State Legislature adopted after the 1970 census. We remanded the subject to the Apportionment Commission for the preparation of another plan. The Commission then adopted the plan now before us. The trial court upheld the plan, with a slight modification not here involved. The Appellate Division held the primary and general elections scheduled for this year may proceed under that plan but expressed doubts as to the plan's validity. 124 N.J. Super. 30 (1973). A petition for certification was filed with us. We directed oral argument with respect to it. Certification is granted.
The Appellate Division conceived that we might not abide by Scrimminger in the light of the later case of Mahan v. Howell, 410 U.S. 315, 93 S.Ct. 979, 35 L.Ed.2d 320 (1973). Mahan does not affect the holding of Scrimminger, as we will point out in Part I below. But the Appellate Division opinion in the present case projects another question not considered in Scrimminger, and this we will develop in Part II below.
I
To judge the impact of Mahan, it is necessary to discuss Scrimminger at some length.
Our Constitution provides for a bicameral Legislature, the Senate to have 40 members and the General Assembly twice that number. A central concept of the Constitution is that the Senate districts shall consist of one or more whole counties, whereas the Assembly districts shall be portions of a Senate district except where only one Senator is to be elected in the Senate district, in which event the Senate district and the Assembly district are coterminous. We held in Scrimminger that under the demographic pattern revealed by the 1970 census, it was impossible to execute the design of our State Constitution that Senate districts shall consist of one or more whole counties, and this because the range of population deviation among the districts would exceed the deviation tolerable under the one-man one-vote concept developed by the United States Supreme Court. We concluded the Apportionment Commission had to draw 40 Senate districts, of equal population, and this in disregard of our State Constitution's mandate that Senate districts consist of one or more whole counties.
The pertinent provisions of our State Constitution are in Article IV, Section 2. Paragraph 1 reads:
"The Senate shall be composed of forty senators apportioned among Senate districts as nearly as may be according to the number of their inhabitants as reported in the last preceding decennial census of the United States and according to the method of equal proportions. Each Senate district shall be composed, wherever practicable, of one single county, and, if not so practicable, of two or more contiguous whole counties."
Paragraph 2 provides in part:
"Each senator shall be elected by the legally qualified voters of the Senate district, except that if the Senate district is composed of two or more counties and two senators are apportioned to the district, one senator shall be elected by the legally qualified voters of each Assembly district. * * *"
Paragraph 3 reads:
"The General Assembly shall be composed of eighty members. Each Senate district to which only one senator is apportioned shall constitute an Assembly district. Each of the remaining Senate districts shall be divided into Assembly districts equal in number to the number of senators apportioned to the Senate district. The Assembly districts shall be composed of contiguous territory, as nearly compact and equal in the number of their inhabitants as possible, and in no event shall each such district contain less than eighty per cent nor more than one hundred twenty per cent of one-fortieth of the total number of inhabitants of the State as reported in the last preceding decennial census of the United States. Unless necessary to meet the foregoing requirements, no county or municipality shall be divided among Assembly districts unless it shall contain more than one-fortieth of the total number of inhabitants of the state, and no county or municipality shall be divided among a number of Assembly districts larger than one plus the whole number obtained by dividing the number of inhabitants in the county or municipality by one-fortieth of the total number of inhabitants of the State."
We call attention to the last sentence in paragraph 3; it is a key provision with respect to the issue discussed in Part II of this opinion.
The whole county concept led to diverse treatment of voters, depending upon whether the voter's county itself constituted a Senate district to which only one Senator was apportioned, was grouped with one or more whole counties to create a district, or was itself a multi-member district. To repeat from Scrimminger, 60 N.J. at 486-487:
"We emphasize some aspects of the foregoing constitutional plan. A central concept is that the Senate districts shall consist of whole counties, and of only one whole county if practicable. The Senators are to be elected by the whole Senate district with this exception, that if two Senators are to be elected from a multi-county Senate district, each Senator shall be elected from a constituent Assembly district. As we have construed the Constitution, the theme of that exception also applies if more than two Senators are to be elected from a multi-county Senate district, so that the candidates must run in Assembly districts, rather than at large in the Senate district. Jackman [ Bodine], 49 N.J. [406] at 416.
"Hence there are diversities with respect to the representation of whole counties in the Senate. If a whole county constitutes a Senate district entitled to one Senator, the voters of that county of course will elect him. If a whole county constitutes a Senate district entitled to more than one Senator, all the voters of the county elect all of the Senators. But if any county, whether it would be entitled by population to one or to more than one Senator if it were a separate district, is joined with one or more counties to constitute a district, then the Senators will be severally elected within Assembly districts, which may consist of either a part of one county or parts of more than one county."
With this design in view, we dealt in Scrimminger with the question whether our State constitutional objective that Senate districts consist of whole counties could be realized under the federal one-man one-vote doctrine. We concluded it could not. The reason, simply put, is that our State, with a population of 7,170,634 has but 21 counties, ranging in population from 59,554 to 929,986. With that spread among so few counties, the range of deviation depended upon the number of Senate districts drawn and a tolerable range could be had only by reducing the number of Senate districts to a point where the objective of a whole voice for a whole county was simply lost. Scrimminger set forth in the dilemma and the Apportionment Commission's attempt to deal with it, 60 N.J. at 488-489:
"The Apportionment Commission noted that the Constitution's mandate for apportionment of 40 Senators among Senate districts `as nearly as may be according to the number of their inhabitants' conflicts with the further mandate that `Each Senate district shall be composed, wherever practicable, of one single county, and, if not so practicable, of two or more whole counties.' This is so because greater population equality can be achieved as more counties are joined into fewer Senate districts but that process necessarily reduces the number of counties standing alone as Senate districts. The Commission tried to strike a balance between those mandates. It created 15 Senate districts, of which 10 counties constituted separate districts, with the following consequences. Of those 10 counties, two elect one Senator each; one elects two Senators; five elect three; and two elect five. If any of those 10 counties had been made a part of a larger district, that county would not have been able to elect a Senator at large, for the candidates would have to run within subdistricts, i.e., the Assembly districts. As to the five districts the Commission created by joining two or more of the remaining 11 counties, four districts consist of two counties while the remaining district consists of three counties. As to the two-county districts, two districts are allotted one Senator, one is allotted two Senators, and the other is allotted three Senators. The three-county district is allotted four Senators.
"In terms of deviations from ideal representation, the results are these:
Senate Districts Assembly Districts
Maximum % above Ideal +13.29% Maximum % below Ideal -15.54% Range of Deviations 28.83% Population Ratio 1:1.34 Maximum % above Ideal +10.66% Maximum % below Ideal -15.54% Range of Deviations 26.20% Population Ratio 1:1.31
It will be noted that the population ratios of 1:1.34 and 1:1.31 are within the limit of 1:1.5 set forth in the State Constitution's provision that no Assembly district shall contain less than 80% nor more than 120% of one-fortieth of the total number of inhabitants ( Art. 4, § 2, ¶ 3), but for the reasons which follow it is now plain that this limit in our State Constitution exceeds what the Federal Constitution permits."
As we pointed out, 60 N.J. at 495 n. 2, the Apportionment Commission considered only plans with 10 or more Senate districts. The record in Scrimminger showed the progressive increase in the range of deviation between the most populous and the least populous districts as the number of the districts increased. With 10 districts, the minimum range of deviation was 7.90%; but there were only 3 single-county districts and 2 of the 3 were multi-member districts. With 11 districts, the minimum range of deviation was 14.33%; but there were only 4 single-county districts and 2 of them were multi-member. With 13 districts, the range of deviation advanced to 27.41%; but there were only 7 single-county districts, of which 5 were multi-member. The best 14-district plan had the slightly higher range of deviation of 27.61%; but there were only 9 single-county districts of which 7 were multi-member. The best 16-district plan had a range of deviation of 31.83%; but it contained only 12 single-county districts of which 9 were multi-member. The best 15-district plan is the one involved in Scrimminger, and it is described in the excerpt above from the opinion in that case and involved a range of deviation of 28.83%.
In concluding that the constitutional plan for Senate districts could not be achieved under the 1970 census, we stated the considerations which led to that view, 60 N.J. at 495-498:
"The first is that the supposed justification for adhering to county lines cannot be realized. The principal justification for the deviations is the public advantage gained by assuring each county a separate voice in its relations with the State. That objective is not achieved in the present plan since only 10 of the 21 counties are represented by Senators elected by the voters of the county. The result is that the counties are dealt with unequally in what, by hypothesis, is an important aspect of the legislative scene. And failure to achieve that objective will become more pronounced and the treatment of the counties more disparate with every effort to reduce the range of deviation, for as we noted earlier, greater population equality can be achieved only if the number of Senate districts is reduced, and as more counties are thus combined to achieve smaller deviations, the fewer will be the counties which will have their own representatives in the Senate. Thus the smaller the deviation, the smaller will be the justification for any deviation at all.
"The second consideration militating against the value of adhering to county lines is the phenomenon of the multi-member district. Of the 10 counties which severally elect their Senators, eight elect between two and five Senators at large rather than within sub-senate districts. Thus 8 of the 21 counties have multiple voices in the Senate. This raises the troublesome problem of multi-member districts, discussed at length in Whitcomb v. Chavis, 403 U.S. [124] at 142-149, 91 S.Ct. [1858] at 1868-1872, 29 L.Ed.2d [363] at 375-379. Opponents of the multi-member district contend that each voter in such a district has the edge over a voter in a single-member district in that he has the chance of casting the decisive vote with respect to the election to each office to be filled whereas the voter in a single-member district can hope to do so only as to one office. That chance is too remote mathematically to be exciting, but there seems to be substance to the claim that the voters of a multi-member county have a larger influence in the Senate by virtue of multi-representation. If this is correct, then our Constitution's plan induces this further element of inequality in its insistence upon adherence to the county as the election district.
"Furthermore, a multi-member election means that the winner may take all. That result does not clash with the one-man one-vote ideal since, as we have said, at-large elections advance voter equality, but there is a clash with the overriding concept that representative government calls for elections by clusters of population to the end that sundry interests may thereby be heard. Members of a minority group may well feel they will fare better when the population clusters are smaller, notwithstanding the political reality that a substantial minority group is not ignored when a multi-member slate is assembled. Although Whitcomb v. Chavis refused to hold that the multi-member district is necessarily invalid, nonetheless the Supreme Court said four days earlier that `We agree that when district courts are forced to fashion apportionment plans, single-member districts are preferable to large multi-member districts as a general matter.' Connor v. Johnson, 402 U.S. 690, 692, 91 S.Ct. 1760, 1762, 29 L.Ed.2d 268, 271 (1971). This was repeated in Connor v. Williams, supra, 404 U.S. [549] at 551, 92 S.Ct. [656] at 658, 30 L.Ed.2d [704] at 707. We of course are not discussing the constitutionality of the multi-member district. We are speaking only of the element of inequality among voters which may ensue when only some of the Senators are elected on a multi-member basis, and we speak of this subject only to the extent that it relates to the larger question whether population deviations among districts is made worthwhile because of advantages thought to inhere in the preservation of the county in drawing district lines.
"For these reasons, the use of county lines under the present distribution of population in our State yields little to compensate for the deviations it produces among Senate districts. And we repeat, the supposed advantage will shrink even more if more counties are combined as they must be to reach a tolerable range of deviation. Moreover if the county is ignored in drawing district lines, its interests will not go unrepresented. A Senator must be mindful of the interests of the county or counties in which his constituents live. True, there may be exceptional situations in which there are conflicting interests within a county, but when that is so, something may well be gained by letting those diverse interests have independent voices through Senators elected in smaller one-man districts. And finally the use of one-man Senate districts would obviate the present problem of the odd number of Assemblymen, for the deviations among one-man Senate districts should be so small as to permit realization of the intent of the Constitution that two Assemblymen shall be allotted per Senator.
"We therefore must conclude that the mandate in the State Constitution for recognition of county lines will not justify any deviation under the demographic pattern of the 1970 census. Since the county cannot now serve as the basis of districting and since the multi-member district is contemplated in the Constitution only as an incident of county representation, it follows that the Senate districts to be created without reference to counties must be single-member districts. Municipal lines should be observed, if possible, for if they are followed, dividends may be expected in terms of furthering the relationship of these political subdivisions and the State and also in terms of restraining to some extent the opportunities for drawing lines for partisan advantage. Municipalities are thus appropriate building blocks for the creation of districts. The boundaries of the larger municipalities will of course have to be reached, and in this regard, the Commission may have to depart from the direction in Art. 4, § 2, ¶ 3, concerning the division of a municipality. The requirement for contiguity will obtain. So also will the requirement for compactness, which may serve to justify a deviation or to curb the quest for partisan gain, although, as we have noted before, compactness may be of limited utility in the light of the odd configurations of our State and its municipalities. Jackman, 49 N.J. at 419. * * *"
With this background we turn to the question whether Mahan v. Howell affects the judgment in Scrimminger. It does not for two reasons. The first is that Mahan does not suggest that a range of deviation of 28.83% in the Senate or of 26.20% in the General Assembly would be constitutionally tolerable. The second reason is that even with those deviations, the objective of our State Constitution would not be achieved. Unlike Mahan where presumably the statutory plan was achieved within a range of 16.4%, our constitutional plan cannot be realized even within the larger ranges of deviations we recounted above. On the contrary, the plans fall so short of that goal as to generate the discriminations among whole counties and disquieting unevenness as among voters detailed in the excerpts above from Scrimminger. Mahan did repeat the proposition in Reynolds v. Sims, 377 U.S. 533, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964), that there is more leeway in districting for elections to the State Legislature than for elections to the Congress, and thus Mahan erased doubts which intervening opinions had raised. But in Scrimminger we deemed to be viable the proposition that "adherence to the lines of political subdivisions will justify a greater deviation from mathematical equality in the case of elections for the State legislature than in the case of an election for the Congress." 60 N.J. at 492. Thus we acted upon a thesis agreeable with Mahan. But we concluded the deviations which are unavoidable under our Constitution's plan were intolerable because of size, and Mahan does not question that conclusion.
And we should add that opinions handed down since the present matter was argued before us do not undercut the holding in Scrimminger although they do reveal a greater legislative latitude in other respects. We refer to White v. Regester, 412 U.S. 755, 93 S.Ct. 2332, 37 L.Ed.2d 314 (1973); Gaffney v. Cummings, 412 U.S. 735, 93 S.Ct. 2321, 37 L.Ed.2d 298 (1973); White v. Weiser, 412 U.S. 783, 93 S.Ct. 2348, 37 L.Ed.2d 335 (1973).
In White v. Regester, involving the Texas House of Representatives, the Court held that variation of 9.9% between the largest and the smallest district did not make out a prima facie case of a denial of equal protection calling upon the State to show a rational State policy. It reached the same conclusion in Gaffney, involving the Connecticut Legislature, where the deviation range in the Senate was 1.81% and in the House 7.83%. But it is significant that Gaffney again described as "unacceptable," 412 U.S. at 744, 93 S.Ct. at 2327, 37 L.Ed.2d at 307, the range of deviations found excessive in Swann v. Adams, 385 U.S. 440, 87 S.Ct. 569, 17 L.Ed.2d 501 (1967) (25.65%), Kilgarlin v. Hill, 386 U.S. 120, 87 S.Ct. 820, 17 L.Ed.2d 771 (1967) (26.48%), and Whitcomb v. Chavis, 403 U.S. 124, 91 S.Ct. 1858, 29 L.Ed. 2d 363 (1971) (28.20% and 24.78%). Thus the decisions we relied upon in Scrimminger remain unaffected.
Although the case at hand does not turn upon some further views expressed in Gaffney, it is pertinent to note the restrictions placed upon judicial intervention under the equal protection clause. The Court deplored the idea that a judge may strike down a plan merely because someone comes up with a plan somewhat better, saying, 412 U.S. 750, 93 S.Ct. at 2330:
"* * * And what is to happen to the master's plan if a resourceful mind hits upon a plan better than the master's by a fraction of a percentage point? Involvements like this must end at some point, but that point constantly recedes if those who litigate need only produce a plan that is marginally `better' when measured against a rigid and unyielding population equality standard."
Of equal interest is Gaffney's acceptance of the proposition that the apportionment plan may be drawn with an awareness of the respective political strength of the major parties and the political consequences of the lines that are drawn, provided the deviations are acceptable and that racial and political groups are not "fenced out of the political process and their voting strength invidiously minimized." At 754, 93 S.Ct. at 2332. Upon this same subject, the Court said in White v. Weiser, 412 U.S. at 791, 93 S.Ct. at 2352, 37 L.Ed.2d at 343, which involved congressional districting, that it will not disparage "a policy frankly aimed at maintaining existing relationships between incumbent congressmen and their constituents and preserving the seniority the members of the State's delegation have achieved in the United States House of Representatives," adding that "We have, in the context of state reapportionment, said that the fact that `district boundaries may have been drawn in a way that minimizes the number of contests between present incumbents does not in and of itself establish invidiousness,'" and citing Burns v. Richardson, 384 U.S. 73, 89 n. 16, 86 S.Ct. 1286, 16 L.Ed.2d 376, 389-390 (1966).
But although these cases suggest greater leeway in the legislative decision, they do not affect our holding in Scrimminger that, under the demographic pattern of the 1970 census, there must be 40 individual Senate districts, drawn necessarily without regard to the "whole county" theme of the State Constitution.
II
When in Scrimminger we spoke of "adherence to county lines" we used that expression as the equivalent of adhering to the whole county in setting up the Senate districts. We did not have in mind adherence to less than all of the lines of a whole county. In other words, there was not before us the question whether the 40 new districts should be so drawn as to adhere to as many county lines, as such, as possible even though the whole county concept plan was not to be followed. The Appellate Division opinion projects the question whether an apportionment plan must place as many of the 40 districts as possible somewhere within the borders of a county.
The issue seemingly was suggested by a plan proposed by one of the plaintiffs, which involved a range of deviation of 9.76%. This exceeds the deviations of the Commission's plan, which are +2.854% and -1.389% for a range of 4.24%. Thus we have the unusual proposition that the Commission should be ordered to produce a plan with increased deviations. It would seem clear that a court cannot so order unless it finds a positive violation of some legal mandate. We gather that the Appellate Division itself advanced the question. In any event the briefs filed below do not adequately explore the issue.
Without intending to limit the parties, we note two theses which could arguably be advanced for the proposition that the Constitution commands that there shall be placed within whole counties as many of the 40 Senate districts as can be. The first is that this mandate is a lesser-included mandate within the Constitution's direction that the Senate districts consist of whole counties, i.e., that this mandate would tend to achieve the same constitutional policy which cannot be achieved in full under the 1970 demographic pattern. The other possible thesis would be that the provision in Article 4, § 2, ¶ 3, which deals with Assembly districts rather than Senate districts, compels the same or a similar result even though that provision was included on a hypothesis that failed, i.e., that it would be applied in creating sub-districts within Senate districts consisting of one or more whole counties. We refer to the last sentence of Art. 4, § 2, ¶ 3, which we flagged earlier and which reads:
"* * * Unless necessary to meet the foregoing requirements, no county or municipality shall be divided among Assembly districts unless it shall contain more than one-fortieth of the total number of inhabitants of the state, and no county or municipality shall be divided among a number of Assembly districts larger than one plus the whole number obtained by dividing the number of inhabitants in the county or municipality by one-fortieth of the total number of inhabitants of the State."
In deciding whether the Constitution demands that result, it may be helpful to know how many districts can be placed within whole counties, at what deviations, and with what pluses and minuses as between the voters of Senate districts so placed and the voters of Senate districts not so placed.
The record is inadequate for a decision upon this issue. We assume that a series of plans could be prepared to indicate what results could be achieved and the deviations involved. The parties may file pertinent material within 30 days from the date this opinion is filed and briefs 30 days thereafter. Any party may move for additional directions. A date for further argument will be fixed. We of course do not disturb the Appellate Division's determination that the general election scheduled for this year may proceed under the plan prepared by the Commission.
For affirmance as modified — Chief Justice WEINTRAUB, and Justices JACOBS, PROCTOR, HALL, MOUNTAIN and SULLIVAN — 6.
For reversal — None.