Dexter O'SHIELDS and Hazel H. O'Shields, Plaintiffs,
v.
Robert E. McNAIR, Governor of South Carolina, O. Frank Thornton, Secretary of State of South Carolina, Edgar A. Brown, Speaker pro tem. acting President of the South Carolina Senate, and Lovick O. Thomas, Clerk of the Senate of the South Carolina General Assembly, Defendants,
James P. Mozingo, III, Intervenor. Michael J. MUNGO, Plaintiff,
v.
Robert E. McNAIR, Governor of South Carolina, O. Frank Thornton, Secretary of State of South Carolina, Solomon Blatt, Speaker of the South Carolina House of Representatives and Inez Watson, Clerk of the House of Representatives of South Carolina General Assembly, Defendants.
Civ. A. Nos. 5053, 66-17.
United States District Court D. South Carolina.
February 28, 1966.*709 Paul M. Moore, Spartanburg, S. C., Wade S. Weatherford, Jr., Gaffney, S. C., attorneys for plaintiffs Shields.
George I. Alley, Theodore W. Law, Jr., Columbia, S. C. (Law, Kirkland, Aaron & Alley, Columbia, S. C., on brief), for plaintiff Mungo.
David W. Robinson, J. Means McFadden, and Clinch Heyward Belser, Columbia, S. C., for defendants Solomon Blatt and Inez Watson.
Daniel R. McLeod, Atty. Gen., C. T. Goolsby, Jr., Asst. Atty. Gen., and James B. Ellisor, Asst. Atty. Gen., for remaining defendants.
E. N. Zeigler, Florence, S. C., amicus curiae.
Henry Hammer, Columbia, S. C., James P. Mozingo, III, D. Kenneth Baker, Darlington, S. C., for intervenor, Mozingo.
Before HAYNSWORTH, Chief Judge, Fourth Judicial Circuit, and MARTIN and HEMPHILL, District Judges.
HAYNSWORTH, Chief Judge, Fourth Judicial Circuit.
In these cases, we are asked to consider the constitutionality of a recent statutory scheme for the reapportionment of South Carolina's Senate and the provisions of the South Carolina Constitution of 1895 requiring the periodic reapportionment of the House of Representatives. We find no constitutional infirmity in the present apportionment of the House of Representatives, and we approve the statutory plan for reapportionment of the Senate as an interim measure for purposes of the next election only and for a limited term of two years.
South Carolina's Constitution of 1895, in its Article III, provided for a Senate composed of one member from each county and for a House composed of 124 members apportioned among the counties on a population basis, with a requirement of periodic reapportionment to accommodate population changes. The South Carolina Legislature is presently constituted in strict conformity with the state constitution. After Reynolds v. Sims, 377 U.S. 533, 84 S. Ct. 1362, 12 L. Ed. 2d 506, however, it was obvious that the apportionment of South Carolina's Senate did not comport with the requirement of the federal constitution that both houses of a bicameral state legislature be apportioned on a population basis.
The decision in Reynolds led to the institution of the O'Shields case bringing in issue the apportionment of the Senate, only. After a hearing on December 3, 1965, during which there was no substantial contention that the present apportionment of South Carolina's Senate could be squared with federal constitutional requirements, we declared its malapportionment, but gave the State an opportunity to devise its own reapportionment plan for its Senate. Finally, on February 3, 1966, a statute was enacted presenting two alternative plans for Senate reapportionment. Meanwhile, the Mungo case was filed. It attacks the apportionment of South Carolina's House of Representatives. After hearings in *710 O'Shields on February 14, 1966 and in Mungo on February 18, 1966, we now have for decision the appropriateness of the statutory plans for the reapportionment of the Senate and the validity, under the federal constitution, of South Carolina's constitutional plan for apportionment of its House of Representatives.
SENATE REAPPORTIONMENT
The Act of February 3, 1966 contains two sections, each of which presents a separate plan. The Section 1 plan divides the state, along county lines, into 27 election districts from which fifty Senators are to be chosen, as many as four Senators being allotted some districts. It is the plan adopted by a conference committee after the Senate had passed a Bill calling for 59 Senators and the House, one calling for 46. The House promptly accepted the conferee's compromise, 50-member plan, but a filibuster delayed action in the Senate until the Senate's original 59-member plan was tacked on as an alternative. With the acquiescence of the House, that became the Section 2 plan.
By its own terms, the Section 2 plan is not to be effective if the Section 1 plan is found to be constitutional. We accept this as a general preference of the Section 1 plan.
I
If these plans presented no substantial question under the federal constitution, we would be loath to give to either our unqualified approval, because each is facially in violation of one provision of the state constitution and arguably in violation of another.
Article IV, § 8 of the Constitution of South Carolina provides for a classification of the Senators so that substantially half of them will be chosen every two years. This requirement of the state constitution is ignored in both plans. Instead of providing for staggered terms, each plan[1] expressly provides that all Senators shall be elected in the general election of 1966 for four-year terms and in the general election each fourth year thereafter. Each plan, in that respect, clearly conflicts with the state constitution's requirement of staggered terms of office for Senators.
This is not to suggest that the General Assembly was wholly without justification for what it did. The state's constitutional requirement that half of the Senators stand for election every two years was adopted at a time when the Senate was regarded as a continuing body and the necessity that every decade it be reapportioned on a population basis was unforeseen. When, in 1971, the General Assembly is called upon to reapportion itself in conformity to the 1970 census, the fact that the terms of office of only half of the Senators would expire in 1972 while the others would remain in office until 1974 would present a grave, if not an insurmountable, obstacle. Even a modification of the provision for staggered terms, which would terminate the terms of office of all Senators in 1972 and in each tenth year thereafter, would present difficulties complicating the reapportionment process, would dilute the virtue of the staggered term requirement, and, perhaps, would discourage candidates in the short-term election years.
The General Assembly has not addressed itself overtly to the ramifications of this problem, except that it obviously has sought to avoid it entirely by adopting a provision for the simultaneous expiration of the terms of all Senators. That course may be highly desirable or even a practical necessity, but the problem which confronts us at the moment is that it is beyond the competence of the General Assembly to effectuate it, in the face of the state constitution's requirement of staggered terms.[2]
*711 From this discussion, it seems apparent that an amendment of the state constitution is a prerequisite to the final adoption and approval of either of these plans or of any substitute thus far advanced.
II
That such an amendment is a practical prerequisite is also indicated by another problem. Article III, § 6 of the Constitution of South Carolina implicitly fixes the number of Senators at 46. In contrast to the explicit provision for 124 Representatives,[3] the numeral 46 does not appear in the constitution, but the provision for one Senator for each county required that the number of Senators be equal to the number of counties. A numerical figure was not used because the Constitution of 1895, itself, contemplated fluctuations in the number of counties. There were then only 36 counties,[4] but in approximately the next two decades ten more were created under authorizing provisions of the Constitution of 1895, to which we shall advert later. Phrased as it was, the constitutional provision fixing the size of the Senate could operate without amendment as new counties were carved out of old ones. Until the Senate's apportionment was questioned, however, there was no room for a contention that the General Assembly could vary the number of Senators. There now being 46 counties, the constitution fixed the number of Senators at 46, and the General Assembly was powerless to alter the number.
After we held on December 3, 1965 that § 6 of Article III, in its application, was invalid under federal constitutional standards because of the great discrepancies in the populations of the counties, the question arises as to what is left of the provisions of § 6 of Article III. Because the federal constitution requires a reapportionment of South Carolina's Senate, is the General Assembly now free to disregard the heretofore unquestionable, if implicit, provision of the state constitution fixing the size of the Senate? May the General Assembly now, on its own authority, enlarge the Senate from 46 to 50? To 59? To 75? If the current General Assembly may enlarge the number of Senators, may not each of its successors do so successively?
This, of course, is a state court question, but its obvious presence enhances our reluctance to give unqualified approval to either plan. Possibly, we could undertake its resolution under the pendant jurisdiction doctrine, but it is much more appropriate to await authoritative resolution of the question by the Supreme Court of South Carolina.
Still more appropriately, the entire question may be avoided by an amendment of the state constitution. As an amendment is highly desirable to alleviate the rigors of the staggered term requirement, it is even more desirable to embody in plain language in the state constitution provisions fixing the size of the Senate and its general composition. That is one of the most fundamental offices of a state constitution. If declaration of the necessity of Senate reapportionment has obscured the meaning of § 6 of Article III, it is infinitely more appropriate that the vacuum be filled by the people through the relatively flexible[5] process of constitutional amendment than that *712 narrow questions be decided by judges, subject to reconsideration later by other judges, or that the entire subject of the composition of the Senate be left to the vagaries of this and subsequent legislatures.
III
The objections to these plans are not limited to those which arise under the state's constitution. There are others arising from the federal constitution's requirement that population be the apportionment basis.
When creating voting districts for the purpose of equalizing the weight of votes, exact mathematical preciseness is not expected, for that is an ephemeral ideal. Reasonable groupings, utilizing political or geographic boundaries, cannot eliminate minimal disparities, and, if complete precision could be achieved today, it would be destroyed by normal population shifts tomorrow. The ideal may not be abandoned, however. Any redistricting must give weight to votes in reasonable conformity to available population data.
In recognition of those principles, it has been suggested that, in legislative apportionment, variances from the average in excess of fifteen per cent, as a rule of thumb, are impermissible or certainly suspect.[6] The suggested rule is not an absolute, however. Variations of less than fifteen per cent are not immune from scrutiny, and isolated ones of greater dimension are not mechanically beyond all possibility of justification. A rule of reasonableness, applied with faithful adherence to the ultimate objective, is not so rigid as to be universally measurable by a standard gauge which takes no account of particular circumstances.[7] When the application of the governing principles is being tested, however, rules of thumb are useful, particularly so when serious assertion of justification for departures is wanting.
Under the Section 1 plan (the 50 Senators plan) four of the twenty-seven districts vary from the population standard by more than the presumptive maximum. With an average population, according to the 1960 census, for each of fifty Senators of 47,652 people, Berkeley County is constituted as a separate election district though its population was only 38,196. It is thus over-represented by a departure from the norm of 19.84%. Lancaster County is similarly over-represented by 17.42%, Oconee by 15.63% and Calhoun and Orangeburg, which together constitute a single election district, by 15.20%.[8] On the other hand, Charleston is under-represented by 13.52%, Darlington by 11.08% and Greenville by 10.06%.
The maximum population variance ratio is 1.42 (Charleston County) to 1 (Berkeley County), and the minimum percentage of voters who may elect a majority of the Senate is 47.99%. The maximum variance ratio is not so extreme as to suggest per se invalidity,[9] while the minimum controlling percentage is probably within the range of acceptability.[10] Moreover, there may be some amelioration in the fact that the greatest variations from the norm are in the form of over-representation, not under-representation.
Nevertheless, the fact that in a number of districts the variance from the norm is substantial presents a grave question when the state has not attempted to justify those variances in the sense of proof of circumstances supporting a *713 preference of the adopted plan over other suggested plans with districts more evenly balanced. The present record does not foreclose an inference that the effort to achieve a reasonable reapportionment based on population was too much diluted by a purpose to preserve as much as possible of the old power structure.[11]
IV
The basic reapportionment question is gravely complicated by the presence in each plan of a "negative residence" clause, which, superficially at least, appears to lend credence to the suggestion that the purpose to achieve a reasonable apportionment was too greatly compromised by a purpose to preserve the offices of some of the Senators now representing some of the smallest counties.
The Section 1 plan contains a provision that no more than one Senator may be a resident of a county in a multi-county, multi-Senator district unless that county's population is 175% or more of the population base per Senator.[12] The Section 2 plan contains a similar, but even more stringent, provision, for there the minimum population requirement for a second resident Senator is 190% of the population base per Senator.[13] By virtue of the districting, this provision operates to guarantee to some small counties a resident Senator while other counties of more than twice the population have no such protected right.
One illustration suffices to illuminate the operation of the provision.
In each of the plans, Orangeburg and Calhoun Counties are combined into one election district and allotted two Senators. The population of Orangeburg County, by the 1960 census, was 68,559, while that of Calhoun was only 12,256. Despite the great disparity in the populations of the two counties the "negative residence" provision operates to require that one of the two district Senators be a resident of Calhoun County, for Orangeburg has less than 175% of the average population base for each of fifty Senators and less than 190% of the smaller average population base for each of 59 Senators. Thus in application, the statute requires a resident Senator in Calhoun, one of the smallest counties in the state.
The plaintiffs level their heaviest guns at this provision upon the assumption that it would be applied in a patently unconstitutional manner. They suppose that all aspirants from that district would run at large for the two senatorial posts with voters at liberty to vote for the first two candidates of their choice without regard to their residence. So applied, their supposition is that two or more Orangeburg men might run far ahead of one or more Calhoun men, but the leading Calhoun man would be declared elected over the second place Orangeburg candidate. Such a construction would obviously invalidate the provision, but until its application has been definitively declared, we must assume a *714 construction, in practice, most consonant with its constitutional validity.
The provision is readily susceptible of the construction that it creates two separate senatorial offices. Any number of candidates, residing in Orangeburg, may run for the resident Orangeburg senatorial office, while any number of Calhoun men may run for the senatorial office assigned to that county. Under that construction each voter in the district would vote his choice in two separate races, and no Calhoun man could be declared elected unless he received a majority of the votes cast in the district for the Senator required by the provision to be a resident of Calhoun. So construed, it is a residence requirement, which still makes election dependent upon a majority vote of the district voters participating in that electoral contest.[14]
But a construction most favorable to the validity of the provision does not obviate all of the difficulties. We may take judicial notice of the fact that the incumbent Senator from Calhoun should have little difficulty in winning one of the senatorial seats allotted the Orangeburg-Calhoun District. The effect of the negative residence clause is to exempt him from the need of campaigning, until one of his Calhoun County neighbors rises up to file against him. The provision, if effective at all, is a continuing one, so, no matter how disaffected Orangeburg voters may become with him or any of his successors in future years and no matter how many likely aspirants there may be in relatively populous Orangeburg, the Senator from Calhoun will have no opposition so long as no likely Calhoun man comes out to oppose him. In view of the great disparity in the populations of the two counties, this constitutes some restriction upon the choices available to the voters of the district.
In Fortson v. Dorsey, 379 U.S. 433, 85 S. Ct. 498, 13 L. Ed. 2d 401, the Supreme Court sustained a residence requirement in senatorial races in multi-district counties. Fulton County (Atlanta), Georgia was divided into a number of districts of substantially equal population. Each district was entitled to one Senator, but the statute provided that when a county was subdivided into more than one senatorial district, the voting should be on a county-wide basis. The effect was to make the County a multi-member district with a residence requirement. Obviously, each elected Senator represented the entire county for he could win election in the county-wide race though he lost to his opponent in his own district. The case seems comparable to this, except for the crucial fact that the subdivisions of the voting unit there were approximately equal in population.[15]
We are also aware of the fact that a number of municipalities in this state elect their councils under ward-residence rules. Each electoral contest is framed by opposition between residents of the same ward, but the election is determined by the results of city-wide voting. We may assume that in some of those cities, particularly those which have been operating under that system for a long time, some wards may vary greatly in population.
We may also assume the general constitutionality of such municipal electoral schemes, but that does not dispose of the present problem. Here there is no supporting history, and the record furnishes no basis for a finding of a compelling need for resident Senators in some *715 small counties but not in other counties of comparable size and in still others much larger.[16] At this stage of the case, we cannot ignore the inference that the conjunction of the particular districting and the negative residence requirement were inserted to protect the offices of particular Senators and to preserve, to the maximum possible extent, the heretofore controlling influence of the small counties. We suggest no moral criticism of such purposes, but their service cannot be permitted substantially to dilute the weight of some votes or to restrict the available choices of voters in some districts. If under other circumstances, some such provision might be acceptable, an evasive scheme substantially to avoid the consequences of reapportionment and to preserve the old order is not.
V
Finally, it is apparent that conclusive answers may not be founded on the present record.
It may be that after extended evidentiary hearings, the state might be able to justify the apportionment or the negative residence clause, or both. We only hold that justification of either is not now clearly apparent on this record.
It may be, also, that the problems arising under the state constitution might be satisfactorily answered by subsequent litigation in the state courts.
Infinitely preferable for the state would be reconsideration by the General Assembly of the entire matter, without the close time limitations of our earlier order, its adoption of a revised plan, which would avoid the substantial remaining questions of federal constitutional validity, and its submission of constitutional amendments which will validate its revised plan and formally embody in the state's constitution essential provisions relating to the size and general composition of the Senate.
With respect to the 1966 elections, any such procedure will require much more time than remains. The county conventions, the beginning of the electoral processes, are less than two weeks away. There is no time for the extended legislative consideration which appears requisite if the interest of the people of South Carolina is to be served, or, in default of that, for extended litigation in this and the state courts. An interim plan is requisite.
VI
In considering the Act of February 3, 1966 for the reapportionment of South Carolina's Senate, we have necessarily emphasized its dubious validity under the federal and state constitutions. We have also recognized the existence of the very practical problems which led the General Assembly to take the course it did, despite possible limitations of the state's constitution. Appropriately, we may, also, now record our deep understanding of the position of individual Senators, who, faced with the destruction or subordination of their sources of political power, did not rush to endorse the warrant. Our enumeration of the constitutional problems presented by the Act of February 3, 1966 is not a criticism of the General Assembly. On the contrary, with all of the difficulties which beset it, we think it produced an acceptable interim measure.
The Section 1 (50-Senator) plan[17] is an advance which approaches, if it does *716 not completely meet, the federal constitutional requirements. It is the plan of the conference committee, accepted by both Houses, which strongly commends it to our acceptance in preference to any interim plan we might devise. We are impelled to accept it in its entirety, except only the provision for a four-year term of office, as an interim plan under which the 1966 elections may be conducted.
As an interim plan, the provision for four-year terms must be suspended. If the 1966 elections are for four-year terms, the reapportionment of the Senate, as finally determined, cannot be effected until 1970. Operation under an interim plan until the general elections of 1968 is necessary, but, with all of the constitutional questions which remain presently unanswerable, a longer duration of a Senate elected under the interim plan could not be justified. By the conclusion of the 1967 session of the General Assembly, it should have had ample opportunity to determine its final plan for reapportionment of the Senate including the adoption of resolutions submitting to the people appropriate constitutional amendments.
In summary, while reserving grave doubts as to the validity under both state and federal constitutions of the Section 1 plan, we nevertheless approve its use solely as an interim measure. We reach this decision in view of the inadequacy of the present record as a basis for final adjudication of the constitutional questions, the limited time remaining before the 1966 elections, the great improvement over the existing Senate apportionment, and in the expectation that a substantially reapportioned Senate will enact an acceptable apportionment plan.
Within a reasonable time after the adjournment of the General Assembly in 1967, the Attorney General will file with the court a report disclosing what has been done. The court will then consider its acceptability under established constitutional principles. Meanwhile, any party may have leave to apply to the court for any further orders which may seem necessary or appropriate.
THE HOUSE OF REPRESENTATIVES
The question of the apportionment of South Carolina's House of Representatives is uncluttered by collateral problems. The question is simply whether her constitutional provision for apportionment by population, utilizing the counties as voting entities, meets the requirements of the Federal Constitution. There is a secondary question of the validity of the provision allotting at least one Representative to each county, but, as we shall see, the effect of that provision is so minimal that it does not hinder our conclusion that the House of Representatives is apportioned in accordance with federal constitutional requirements.
South Carolina's Constitution of 1895, by its Article 3, § 3, provides that the House of Representatives shall consist of 124 members apportioned among the counties in accordance with their populations. Each county is a separate election district, and the section requires the General Assembly to reapportion the House in 1901, and in each tenth year thereafter. In that connection, it is authorized to use the immediately preceding United States census as an accurate enumeration of the inhabitants of the state and of each county. By Article 3 § 4, the General Assembly is required to allot one Representative to each county for each 124th part of the state's total population, to allot one Representative to each county which contains less than one 1/124th part of the state's total population, and to allocate the remaining number of the total of 124 Representatives to those counties having the largest remaining fractional part of 1/124th of the state's total population.
There were thirty-six counties in the state in 1895, but Article 3 § 3 of the Constitution contemplates that new counties may be established, and, upon such events, the General Assembly is required to reapportion itself.
The procedure for the formation of new counties is prescribed in Article 7 §§ 1 and 2. Article 7 § 3 requires that any new county must have a minimum *717 area of 400 square miles, a minimum population of 1/124th of the state's total population, and assessed taxable property of at least a specified amount. By § 4, no new county can be carved out of an old one, unless the remaining part of the old county is at least 500 square miles in area, contains at least 15,000 people (that being more than 1/124th of the state's total population at that time) and have a minimum assessed valuation of taxable property.
By such provisions, then, the Constitution of 1895 sought to protect the integrity of its apportionment requirement by forbidding the formation of new counties which would not be entitled to a Representative on a population basis, or which would leave the remaining part of the old county unentitled to representation on a population basis.
Subsequent shifts in population have, to some extent, distorted the apportionment plan of 1895 and rendered its safeguards, to some extent, unavailing. McCormick County today has only 8,629 people, but in 1920, the year of the first census after its formation, it had a population of 16,444, more than 1/124th of the total population of the state at that time. Since then, approximately forty per cent of that county has been acquired by the United States, which, with the general shift in population from rural to urban areas, has occasioned a substantial loss in population.
In every decade since the adoption of the Constitution of 1895, the South Carolina General Assembly has conscientiously reapportioned the House of Representatives in accordance with the Constitution's requirements. This was last done in 1961 on the basis of the 1960 census figures, and that reapportionment controlled the elections of 1962 and 1964. The apportionment was mathematically accurate and in strict conformity to the Constitution's requirements. In respect to the reapportionment of the House of Representatives, South Carolina's General Assembly is one of those which has faithfully complied with its constitutional obligations.[18]
The requirement of Article 3 § 4 that every county be allotted at least one Representative, though the population be less than 1/124th of the total population of the state, affects only one Representative. If all of the 124 Representatives were allotted among the counties on a population basis, without regard for the requirement that each county have at least one, the apportionment would be exactly as it is, except that McCormick County, the smallest, would have no Representative and Clarendon County would have two instead of one. A number of the counties which would thus be allotted one Representative contain less than 1/124th of the state's total population, but they, nevertheless, would be allotted one Representative under the major surplus fraction rule. Allocating one Representative for each whole 1/124th of the state's total population cares for the allocation of 101 Representatives. The remaining 23, under the requirement of Article 3 § 4, without regard to the one Representative for each county rule, are then allocated among those counties having major unrepresented fractions of a full 1/124th part of the state's total population. In 1960, 1/124th of the state's total population was 19,214. Barnwell County, for instance, then had 17,659 people, less than a full 1/124th of the state's total. It could not participate in the state's allocation of the 101 Representatives, one for each whole fractional part, but it would be allotted a Representative because its 17,659 people are one of the larger surplus fractions, entitling it to recognition at that stage of the allocation process.[19]
*718 Since the requirement of one Representative for each county regardless of population demonstrably affected the present apportionment of only one of the 124 Representatives, we think its present effect is minimal. If there are strong reasons for treating counties as election districts, and the population apportionment is, otherwise, in accordance with the federal constitutional population requirement, the allocation of this one Representative to McCormick would not, in our opinion, be objectionable on constitutional grounds. McCormick's 8,629 people are not shockingly fewer in number than the surplus fraction of 10,276 which would occasion the allocation of that Representative to Clarendon County if the one Representative county requirement were disregarded. Furthermore, since the allocation of 123 of the 124 Representatives is strictly upon a population basis, a variant of 1/124th would not deprive the apportionment of its essential population basis.
We speak, of course, in terms of the application of the present apportionment. If McCormick County's population continues to dwindle while the state average increases, a different problem may be presented. This is particularly true if, in future years, other counties should find themselves in McCormick's position, dependent upon the one Representative for each county requirement for the allocation for any Representative to it.
The apportionment of Representatives among the counties in accordance with their population produces some substantial variances in the lower register. This is inevitable when the counties are made voting districts and they have disparate populations. The greatest variances, of course, occur at the boundary line between those counties allotted one representative and those having two. It is also substantial within the list of the counties having one Representative. Thus a county, the population of which is just under enough to have two, may seem quite unequally represented when compared to another county which has just enough people to qualify for two.[20] So, also, there is a substantial disparity between the smallest county entitled to one Representative and the largest county entitled to only one but which comes close to the line which divides the one Representative counties from the two Representative counties.
The result of this is that 19 of the 46 counties vary from the population norm by more than 15%. In some instances, the variance is only slightly in excess of 15%. In other instances, it is substantially greater. As would be expected, each of these 19 counties has only one or two Representatives.[21] There are no *719 such substantial variances among the larger or intermediate counties.
While these disparities cannot be ignored and would not be permissible if the use of counties as election districts did not have strong justification, the disparities, themselves, are obliterated when the counties are reasonably grouped. Thus, the 17 single Representative counties contain 13.80% of the state's population, and they elect 13.70% of the House of Representatives. Viewed collectively, the single Representative counties are very slightly below the norm, but very close to it. Certainly, they are not grossly overly represented.
There are 16 counties having two Representatives each. They have 25.25% of the population and they elect 25.80% of the House of Representatives.
Two counties are assigned three Representatives each. They have 4.70% of the population and elect 4.83% of the House of Representatives.
Six counties are assigned four Representatives each. They have 19.13% of the population and they elect 19.35% of the House of Representatives.
The five largest counties, Charleston and Greenville, having eleven Representatives each, Richland, ten, Spartanburg, eight, and Anderson, five, contain 37% of the population and elect 36.28% of the House of Representatives.
From these figures, it is readily apparent that the apportionment of the House of Representatives among the small, large and intermediate counties is an almost precise reflection of population data. This further appears in the minimum controlling percentage. The 37 smallest counties have 49.6% of the state's total population, and they can elect 63 of the 124 members of the House of Representatives. A closer correspondence to the theoretical ideal could hardly be expected.
In light of the close correlation between population and representation among all the counties when considered in their natural groupings, we now turn to the question of whether the disparities among individual smaller counties are justified by the state's interest in utilizing the counties as voting districts.
In South Carolina, the counties are major political subdivisions of the state. They levy and assess taxes, build and maintain those roads which are not incorporated into the national or state highway systems, and supply a multitude of other services to their residents, and to other subdivisions such as school districts. Each county has a clerk of court, an office for the registration of mesne conveyances, a probate court, a treasurer, an auditor, a sheriff, a supervisor, and other administrative officials. They are selected by popular elections in the counties. Within each county, there is a strong and clearly recognized community of interest, which does not cross county lines.
There has been no general delegation of legislative authority to the counties. With the exception of a few counties, the legislative authority in county affairs is still vested in the General Assembly. In practice, it is exercised by the county delegation in the General Assembly. When a particular county delegation reaches a conclusion on a county legislative matter, "local" bills are introduced in the General Assembly which are routinely passed in both houses without scrutiny by other members of the General Assembly. Practically, the county's delegation in the General Assembly is its legislative body.
The plaintiff in this case attacks as archaic the system for exercising legislative authority. He thinks county councils should be established with delegated legislative authority. What he says may have merit when viewed as a question of political science or of politics, but until some such system is established for the exercise of legislative authority respecting the affairs of the individual counties, it is highly important that each county have some representation in the General Assembly. By virtue of the reapportionment of the Senate now being *720 effected, many counties will be left without representation in that body. If such a county were also now left without representation in the House of Representatives, the existing system would break down. From the plaintiff's point of view, if this resulted in an earlier creation of county councils, it is a consequence to be desired, but we think the General Assembly is entitled to determine what, in its opinion, is the most desirable method of handling the legislative affairs of the counties. Until some other method is adopted, consideration of the need of individual counties for representation in the House of Representatives is not impermissible.
Our decision, however, does not turn upon the legislative need of county representation in the House of Representatives, for we think there are other factors which give each county such a strong community of interest as to justify the treatment of each as a separate election district, so long as the disparities between individual counties are not greater or more numerous than they are and so long as a comparison of population and representation for every reasonable grouping of counties shows the close correlation which is now present.
In Reynolds v. Sims, 377 U.S. 533, 580-581, 84 S. Ct. 1362, there is recognition that use of political subdivisions, such as South Carolina's counties, as election districts may be appropriate in at least one house of a bicameral legislature. The Supreme Court there indicated its awareness of the importance of the responsibilities discharged by such political subdivisions and of their legislative needs. It also recognized that their use as election districts may deter gerrymandering. Its dictum was that such considerations may justify use of political subdivisions as election districts, even though deviations from the population base result which exceed in number or degree deviations which, otherwise, would be impermissible. Since South Carolina's reapportionment of its House of Representatives has been accomplished promptly after each decennial census, since it has been done accurately and on a strictly population basis, save only the allocation of one representative to McCormick County, and since a legitimate state interest is served by treating the counties as election districts, we think South Carolina's apportionment of her House of Representatives among the counties is not inconsistent with the federal requirement of apportionment on a population basis.
Population, as the basis of apportionment of South Carolina's House of Representatives, has not been submerged. It is the reigning principle. The ratio between the number of members to be apportioned (124) and the number of counties (46) is quite adequate to permit an effective play of the population principle. Scrupulously applied, as it has been, the principle shines brightly in application. The inevitable distortion, vis-a-vis each other only, among the smaller counties, as in the smaller states having representation in the Congress, is not a fatal objection when considered in the light of the ratio of the allotted places to the counties and the congruity of representation and population data when the House is viewed as a whole.
We conclude that the apportionment of South Carolina's House of Representatives meets the requirements of the Constitution of the United States of apportionment on the basis of population.
CONCLUSION
In each of these cases, orders will be entered in conformity with this opinion.
ADDENDUM
Since this opinion was prepared, the Supreme Court has reversed, per curiam without a hearing, acceptance of Florida's legislative reapportionment as an interim plan. Swann v. Adams, 382 U.S. 210, 86 S. Ct. 767, decided February 25, 1966. That proceeding was commenced in 1962. Under judicial prodding, Florida's legislature enacted a legislative reapportionment *721 plan on June 29, 1965. Though the District Court found the plan unconstitutional, and, apparently, in the Supreme Court, there was no contention that it was not, the District Court nevertheless authorized its use on an interim basis, promising to review the situation again after the legislative session of 1967. In summarily reversing, the Supreme Court expressed the opinion that, since the action was commenced in 1962, a further delay in effecting a valid reapportionment until 1969 (the commencement of the terms of office of the successful candidates in the elections of 1968) could not be justified.
Here, the circumstances are quite dissimilar. This action was not commenced until September 23, 1965. The court moved as expeditiously as orderly procedure would permit, holding on December 3, 1965 that the existing Senate apportionment was invalid. The General Assembly moved with reasonable promptness, enacting a reapportionment plan on February 3, 1966. Final enactment of the reapportionment statute was delayed for some weeks by a filibuster, but that fact does not infect the good faith of the legislative majority in adopting the plan we find acceptable on an interim basis and which was the object of the attack of the filibuster. Finally, the less than six-month interim between the commencement of this action and March 7, 1966 is clearly not long enough to permit a final solution of, at least, some of the constitutional questions that arise. The county conventions will be held on March 7, 1966, and at those conventions, one of the major parties will nominate candidates for legislative offices. If it is not then known what offices are to be filled, there would be a grave disruption of the 1966 nominative-elective processes.
Under the circumstances, we think Swann v. Adams does not require a change in our conclusion in the O'Shields case here. See Drum v. Seawell, M.D. N.C., 250 F. Supp. 922, decided February 18, 1966.
NOTES
[1] Subsection (c) of Section 1; subsection (b) of Section 2.
[2] It is readily conceivable, of course, that South Carolina's Supreme Court might hold that the declaration of invalidity of the apportionment of the Senate on the basis of one Senator for each county also struck down such closely related provisions as that for staggered terms of office. In the absence of such a decision and in the present posture of the case, we think such a question is not now before us, if, under any circumstances, it might be for initial determination by us. Under the circumstances, it is appropriate that we treat the staggered term requirement as one of present validity, without, however, undertaking an adjudication of the matter.
[3] Article III, § 3.
[4] See Article III, § 3 of the Constitution of 1895.
[5] If the course of constitutional amendment is followed, the size of the Senate becomes purely a political question. If some expansion in the number of Senators will contribute materially to a resolution of the practical problems of reapportionment, the people may freely act in recognition of that fact; neither judges nor legislators have such unrestricted powers as the people acting through the amendatory process.
[6] See Toombs v. Fortson, N.D.Ga., 241 F. Supp. 65, 70; H.R. 5505, 89th Cong. 1st Sess. But see Roman v. Sincock, 377 U.S. 695, 710, 84 S. Ct. 1449, 12 L. Ed. 2d 620.
[7] Roman v. Sincock, supra note 6.
[8] A number of other election districts are over-represented by close to fifteen per cent: Aiken by 14.97%, District 17, consisting of Dorchester, Jasper, Hampton and Colleton Counties by 14.11%, and Williamsburg by 14.10%.
[9] See, Lucas v. Forty-Fourth Colorado General Assembly, 377 U.S. 713, 727, 730, 84 S. Ct. 1459, 12 L. Ed. 2d 632; Sims v. Baggett, D.Ala., 247 F. Supp. 96; Schaefer v. Thomson, D.Wyo., 240 F. Supp. 247.
[10] Ibid.
[11] That purpose may have greatly influenced the adoption of the alternative Section 2 (59-Senator) plan. Its variances from the norm are less than those of the Section 1 plan, but the difference is what one reasonably would expect from an increase in the number of Senators to 59. Its maximum population variance is 1.37 to 1, and its minimum controlling percentage is a good 48.54%. Still, it leaves two districts over-represented by more than fifteen per cent (Kershaw, 16.83% and Chesterfield, 16.51%) and it enlarges the number of counties so under-represented as to create borderline problems (Pickens, under-represented by 13.98%, District 21 (Darlington, Florence), 13.41%, District 6 (Allendale, Barnwell, Bamberg), 12.17%, Beaufort, 9.42% and District 25 (Richland, Fairfield), 9.36%.) Counsel for some of the prospective beneficiaries of the 59-Senator plan has urged that it is so far superior to the 50-member plan as to require a declaration that the latter is constitutionally invalid while the former is acceptable. We find no such marked differences between the two plans, however, particularly in light of the alternate plan's much greater departure from the norm of 46 Senators and its apparent purpose.
[12] Section 1(e).
[13] Section 2(d).
[14] This construction is supported by the statute's requirement that in single county, multi-member districts, the offices be numbered and that candidates run, not generally, but for a specifically numbered senatorial office. Clearly, that provision contemplates an individualization of senatorial contests in multi-member districts.
The plaintiffs also question the numbering provision. We perceive there no question of constitutional dimension. The argument is principally addressed to the wisdom of the requirement, but that is a matter for legislative determination.
[15] See Swann v. Adams, S.D.Fla., No. 186-62, December 23, 1965, petition for cert. filed 383 U.S. 210, 86 S. Ct. 767, 15 L. Ed. 2d 707.
[16] In dealing with the apportionment of the House, infra, we recognize the importance of the counties in South Carolina and the desirability, until some other means for their governance is provided, of their each having some representation in the General Assembly. Such considerations, however, do not necessarily establish justification of a provision which assures resident Senators for some small counties, but not for others.
[17] We do not further consider the Section 2 (59-Senator) plan. We do not hold the Section 1 plan "constitutional," but we do not hold it "unconstitutional" either. Since the General Assembly has directed that the Section 2 plan is inoperative if the Section 1 plan is held "constitutional" and each is no less vulnerable to constitutional attack than the other, we accept the legislative preference. If the choice were ours, it would be the same. See footnote 11.
[18] Cf. Reynolds v. Sims, 377 U.S. 533, 84 S. Ct. 1362, 12 L. Ed. 2d 506; Maryland Committee for Fair Representation v. Tawes, 377 U.S. 656, 84 S. Ct. 1429, 12 L. Ed. 2d 595; Baker v. Carr, 369 U.S. 186, 82 S. Ct. 691, 7 L. Ed. 2d 663.
[19] The surplus fractions resulting in actual allocations of Representatives in the 1961 reapportionment ranged from 18,982 people in Berkeley to 10,605 people in Horry.
[20] Decennial reapportionment may tend to equalize the situation as between counties close to the borderline between one and two Representatives.
[21] The same phenomenon is apparent in the apportionment of the national House of Representatives among the states. As between the smaller states, substantial disparities inevitably exist, as shown by the following table:
No. of Population per Percentage
State Representatives Representation Variance*
Alaska 1 226,167 -45%
Nevada 1 285,278 -31%
Wyoming 1 330,066 -20%
Vermont 1 389,881 - 5%
Delaware 1 446,292 + 8%
New Hampshire 2 303,460 -26%
Montana 2 337,383 -18%
* * *
Maine 2 484,632 +18%
* The national average population per representative is 412,237.