Opinion
Civ. A. No. 2470.
January 10, 1967.
Vincent A. Theisen and Victor F. Battaglia, Wilmington, Del., for plaintiffs.
David P. Buckson, Atty. Gen., Dover, Delaware, Frank O'Donnell, Wilmington, Del., James H. Hughes, III, Dover, Del., H. Edward Maull, Georgetown, Del., Roy S. Shiels, Dover, Del., for defendants.
Before BIGGS, Circuit Judge, and WRIGHT and LAYTON, District Judges.
I. History and Factual Background of the Present Litigation
The case at bar has a long history that requires only a résumé here for that history appears fully in the decisions cited in the footnote. It is sufficient to state here that the suit at bar was brought originally as a class action by seven citizens, taxpayers and voters of urban and suburban areas of Delaware. An adjudication was sought that Section 2, Article II of the Constitution of the State of Delaware of 1897, as amended, Del.C. Ann., and the 1963 Amendments thereto, were in violation of the Equal Protection and Due Process Clauses of the Fourteenth Amendment to the Constitution of the United States. The election officers were named as defendants and are still named as such but it should be borne in mind that certain former defendants have been relieved as parties by reason of their quitting office upon the ending of their terms and the elections of their successors who have been substituted as party-defendants. These facts have no substantial bearing on the issues presented by this litigation in its present form.
See Sincock v. Terry, D.C., 207 F. Supp. 205; D.C., 210 F. Supp. 395; D.C., 210 F. Supp. 396 (1962); Sincock v. Duffy, D.C., 215 F. Supp. 169 (1963); and Sincock v. Roman, D.C., 232 F. Supp. 844 (1964). More importantly, see the opinion of the Supreme Court in Roman v. Sincock, 377 U.S. 695, 84 S.Ct. 1449, 12 L.Ed.2d 620 (1964), affirming the judgments entered following our opinion in Sincock v. Duffy, supra, 215 F. Supp. 169. See also Sincock v. Roman, D.C., 233 F. Supp. 615 (1964).
We held that Section 2, Article II, of the Constitution of the State of Delaware and the 1963 Amendments thereto constituted a violation of the Equal Protection Clause of the Fourteenth Amendment. Sincock v. Duffy, D.C., 215 F. Supp. 169 (1963). Following the entry of our judgments an appeal was taken to the Supreme Court of the United States. Following the affirmance of our decision by the Supreme Court in Roman v. Sincock, 377 U.S. 695, 84 S.Ct. 1449, 12 L.Ed.2d 620 (1964), two new statutes, designated as (Senate Bill) S.B. 332 and S.B. 336, created a new apportionment for Delaware, an apportionment which became effective on July 13, 1964. For the sake of convenience, rather than repeating S.B. 332 and S.B. 336 verbatim here, we refer to Title 29 of the Delaware Code Annotated. S.B. 332 appears in toto in Subchapter I, Sections 601-609. S.B. 336, which sets out the districts of the General Assembly appears in toto in Subchapters II to VI, inclusive, Sections 621-622, 631-632, 641-642, 651-652 and 661. 29 Del.C. Ann.Cum.Supp. We shall quote certain portions of S.B. 332 and S.B. 336 hereinafter as need arises.
Senate Bill 332 also appears as Chapter 360 and Senate Bill 336 as Chapter 361, Volume 54, Laws of Delaware.
S.B. 332 provides for a House of Representatives to be composed of 35 members to hold office for 2 years, that the State be divided into 35 representative districts, and that each representative district choose one representative. S.B. 332 provides for a Senate to be composed of 18 members each to hold office for 4 years. The State is divided into 18 senatorial districts, each of which shall elect one senator. S.B. 332 states that the terms of office of the several senators shall be so staggered that not more than 9 senators shall be elected at each biennial general election. The Act further provides that there shall be 8 representative districts in the City of Wilmington, 16 representative districts in New Castle County outside the City of Wilmington, 5 representative districts in Kent County, and 6 representative districts in Sussex County. The Act further provides that there shall be 4 senatorial districts in the city of Wilmington, 8 senatorial districts in New Castle County outside the City of Wilmington, 3 senatorial districts in Kent County, and 3 senatorial districts in Sussex County.
Section 607 of S.B. 332 provides that "The General Assembly shall determine the boundaries of the several Representative and Senatorial Districts within the City of Wilmington and within the several counties by an act of the General Assembly in such manner as there shall be substantial equality of population among the several Representative Districts and among the several Senatorial Districts."
Section 609 of S.B. 332 provides that the newly created apportionment shall continue in effect until the official reporting by the President of the United States of the Federal Decennial Census of 1970, and that after the reporting of each following Federal Decennial Census, the General Assembly shall reapportion and redistrict the State in such a manner that the several representative and senatorial districts shall be substantially equal in population.
S.B. 332 was enacted by the General Assembly on July 6, 1964, and S.B. 336 was enacted by it on July 8, 1964. S.B. 336 designates the boundaries of representative and senatorial districts in the City of Wilmington largely street by street.
See testimony of Senator Reynolds duPont, 1394. An objection was made by the defendants to Senator duPont's testimony, 1388, insofar as it was intended to express the intent of the General Assembly. We do not regard the objection as being pertinent at this point.
(Except as hereinafter especially noted, all references to the transcript of testimony, as for example in this note, are to transcripts made at the hearings on the "Motion to Implement the Mandate of the Supreme Court". We shall refer to page numbers of those transcripts in this opinion and in its footnotes.)
S.B. 336 designates the boundaries of representative districts in New Castle County, outside the City of Wilmington, by boundaries consisting of state, county and city lines, roads, railroads, creeks (small rivers), the Chesapeake and Delaware Canal, and other boundaries which in most cases are clearly identifiable except that the 15th, 19th and 24th representative districts are defined as portions of Hundreds not included in other representative districts. The senatorial districts of New Castle County, outside of Wilmington, comprise combinations of representative districts in New Castle County, outside of Wilmington, by number, as designated in that portion of S.B. 336 relating to representative districts.
We repeat here the definition of "Hundred" contained in our opinion in Sincock v. Duffy, 215 F. Supp. at 175, as follows: "`Hundred' is used in Delaware in the same sense, we believe, as it was originally employed in English law to signify `A division of a county, which some make to have originally consisted of 100 hides of land, others of 10 tithings or 100 free families.' Bouvier's Law Dictionary, Rawle's 3rd Rev. Bouvier also states, `In Delaware the subdivisions of a county are called hundreds. They correspond to towns in New England, townships in Pennsylvania, parishes in Louisiana, and the like.' See also Webster's Collegiate Dictionary."
S.B. 336 defines the boundaries of the General Assembly representative districts in Kent County by combining the heretofore existing representative districts of Kent County. Sections 641(a) and 641(d) dealing respectively with the 25th and 28th representative districts in Kent County perhaps are typical. Section 641(a) states: "The twenty-fifth Representative District shall comprise all of the heretofore existing and constituted First, Third and Fourth Representative Districts of Kent County." Section 641(d) provides: "The twenty-eighth Representative District shall comprise all of the heretofore existing and constituted Seventh and Eighth Representative Districts of Kent County and the heretofore existing and constituted First Election District of the Sixth Representative District of Kent County."
S.B. 336 states that the senatorial districts of Kent County shall be composed of new combinations of the previously existing representative districts and of certain of the old election districts of Kent County as designated.
S.B. 336 provides that the several representative districts of Sussex County shall be composed of the previously existing representative districts and of certain old election districts of Sussex County. In this connection the provisions of Section 651(a), relating to the 30th representative district, may be deemed to be typical: "The thirtieth Representative District shall comprise all of the heretofore existing and constituted First Representative District of Sussex County and all of the heretofore existing and constituted First and Third election districts of the Second Representative District of Sussex County."
We shall discuss the phrase "heretofore existing and constituted" as referred to in Sections 641(a), 641(d) and 651(a), at a later point in this opinion.
S.B. 336 states that the senatorial districts shall be composed of combinations of the new representative districts.
Section 661 of S.B. 336, headed "Effectuation", provides: " Preparing and filing maps of districts. (a) The respective Departments of Elections shall cause to be prepared by Registered Land Surveyors maps of each county showing the boundaries of the several General Assembly districts. (b) The Registered Land Surveyors who prepare the maps shall certify on them as to their correctness and the respective Departments of Elections shall cause the appropriate maps to be recorded in the Recorders Office in the County for which the maps are prepared. Upon such recording, the maps shall be prima facie evidence in all judicial proceedings as to the correctness of the boundaries. (c) The several Departments of Elections shall cause such additional copies of the maps, as they deem necessary to properly advise the public, to be prepared and distributed." Maps referred to in Section 661 have been certified and filed and we shall deal with them at a later point in this opinion.
II. Facts Relating to the Preparation of S.B. 332 and S.B. 336
Shortly after the decision of the Supreme Court in Roman v. Sincock on June 15, 1964, a meeting or a series of meetings were held in the Capitol Building at Dover, Delaware, on or about June 17, by members of the Democratic party to discuss proposals to be put before the General Assembly to provide for reapportionment in accordance with the decision of the Supreme Court in Roman v. Sincock and the related Supreme Court decisions., On June 22, 1964, Governor Carvel held a meeting of the leaders of both parties in the Legislature. This meeting was called as a bipartisan meeting, but during the course of it the Governor stated, according to the uncontradicted testimony of Senator Reynolds duPont, "that the majority party was the majority party and it was going to do the reapportionment." In respect to S.B. 332 Senator Allen J. Cook, President pro tempore of the Senate advised the Senate in the course of the limited debate immediately preceding the passage of the bill through the Senate, that S.B. 332 was "the majority party's bill," and they were going to make the best effort "because they were the majority, to take care of the Democratic party in the redistricting." It is a permissible inference to be drawn by the triers of fact, and we draw the inference, that the system of reapportionment embodied in the new legislation which took the place of Section 2, Article II of the Constitution of Delaware and the 1963 Amendments thereto represented the partisan efforts of the Democratic members of the General Assembly as a result of the meetings hereinbefore referred to. Our finding is fortified by the fact that S.B. 332 and S.B. 336 passed the General Assembly by reason of the votes of the Democratic members of the House and Senate. The fact that the legislation was passed by Democratic majorities in both Houses does not necessarily invalidate the legislation insofar as the Equal Protection Clause of the Fourteenth Amendment is concerned, but it is a fact to be considered in respect to the allegations of gerrymandering made by the plaintiffs which will be discussed at a later point in this opinion.
See the testimony of James L. Latchum, Esquire, 376.
See the testimony of Mr. L. Winfred Hughes, 82 et seq., of Mr. Raymond V. West, 121 et seq., of Mr. F. Earl McGinnes, 189 et seq., and James L. Latchum, Esquire, 374 et seq.
See the testimony of Senator Reynolds duPont, 1403 et seq., 1471, and 1482.
There were objections made by the defendants to this testimony, in substance that the statements of individual legislators should not be employed to construe the intent of the legislature if the bill is enacted. In support of this position, Mr. (now Associate Justice) Herrmann cited Sutherland, Statutory Construction, Horick, 3rd ed., Vol. 2, sections 5014 et seq. But see Mr. Justice Frankfurter's opinion in Algoma Plywood Veneer Co. v. Wisconsin Employment Relations Board, 336 U.S. 301, 307-15, 69 S.Ct. 584, 93 L.Ed. 691 (1949). Compare also Mr. Chief Justice Vinson's opinion in the case of United States of America v. United Mine Workers of America, 330 U.S. 258, 280-283, 67 S.Ct. 677, 91 L.Ed. 884 (1947). See Hambrice v. F.W. Woolworth Co., 290 F.2d 557 (5 Cir. 1961) and Hertz v. Graham, 23 F.R.D. 17, 23-24 (S.D.N Y 1958).
As to the hearsay objection to Senator Cook's declaration as testified to by Senator duPont, infra, we hold that the objection is not a valid one. Senator duPont was testifying as to his best recollection of what was said in a public session or sessions of the Senate of the General Assembly of Delaware in which he was taking part as a participant. No stenographic record was or is kept of such debates as contrasted with the debates in our National Congress. We hold Senator duPont's statements to be admissible. If a stenographic record had been made by an official reporter, that record of course would have been admissible as the best evidence and Senator duPont's statements would not have been received in evidence. Since there was no stenographic record kept by an official reporter for the Senate, we accept Senator duPont's statements as the next best evidence. The objections will be overruled.
The vote on S.B. 332 in the Senate is illustrative of the strict party lines adhered to in the voting which passed the new legislation. Nine Democratic senators voted in favor of S.B. 332; one Democratic Senator voted against it (Senator Robbins), six Republican senators voted against the bill and one Republican senator (Senator Simpson) was absent.
The 1963 Amendments provided for 17 senators and 35 representatives.
For the details of the voting in the Senate on both bills see Court's Exs. 14 and 15. See Court's Exs. 30 and 31 as to the vote in the House of Representatives on both bills.
Prior to the enactment of the two measures, apparently between June 22 and June 29, 1964 James L. Latchum, Esquire, a member of the Bar of Delaware, was given the task of planning the reapportionment of the City of Wilmington. Mr. Latchum had been chairman of the Democratic City Committee of Wilmington, and chairman and secretary of the New Castle County Democratic Committee. He testified that he first prepared the boundaries of the 4 senatorial districts, and next those of the 8 representative districts; that in forming the senatorial districts he used census tract figures for the City of Wilmington, as set out in his affidavit. He testified that he found it necessary to split existing election districts, "some 36 or 37" of them. Mr. Latchum also testified that he prepared a map and a description of the districts and sent these to the Secretary of State of Delaware who "was heading the thing up, or at least gathering the plan in his office." He stated that in dividing the City of Wilmington, after reading "the Supreme Court opinion", he attempted to make the respective districts substantially equal in population, and that he had no other criteria than numerical equality in population in mind. He said also that he came around Wilmington "clockwise". This, of course, required him to find a western boundary for the new 2d senatorial district, which is a large one, and he testified that in order to create an area "about equal in population" he had "to take in ten census tracts." Mr. Latchum's problems were largely geographical and arithmetical, and he had available not only the census tracts for Wilmington but also the "United States Census of Housing, 1960", for Wilmington, which breaks the City down population-wise, city block by city block.
Mr. Latchum's original testimony was that he was given "figures" between July 22 and July 29, 1964, but he corrected this statement and stated that he meant the month of "June" instead of July. 378. Mr. Latchum employed the word "figures", and we presume by the use of that word he meant that the numbers of members of the Senate and of the House of Representatives to be elected from the City of Wilmington.
Latchum, 374.
Defendants' Ex. 5.
Latchum, 381-405.
Defendants' Ex. 8 and its accompanying map, Defendants' Ex. 9.
Mr. F. Earl McGinnes undertook the task of the reapportionment of Rural New Castle County, i.e., New Castle County outside of the City of Wilmington. Mr. McGinnes presently is the Budget Director of the State of Delaware, was Governor Carvel's Administrative Assistant, served a term in the House of Representatives, has been vice-chairman of the Christiana Hundred Democratic Committee, and at the time he testified was serving as chairman of the Christiana Hundred Democratic Committee. He testified that he used "population figures as prescribed by the census of the United States in 1960," with certain exceptions; viz., those situations where estimates were deemed necessary. He stated that it was his understanding that both representative and senatorial districts should be "as nearly equal in population as reasonably possible" and that "natural and historical boundaries should be followed whenever possible." He said that he sought to maintain election district boundaries where practicable and keep the new representative districts within the old Hundred boundaries where possible but he reiterated that his primary purpose was to secure equality of population in the General Assembly districts. In response to a question as to how many census enumeration districts he split in preparing the plan, Mr. McGinnes answered: "It would appear [that there were splits in] 16 different districts. Thirty-two splits show up on here and this is past one place and past the other, so it is the same district involved. I assume this means 16 districts." (Sic.) On further examination he testified that there were in fact 19 census enumeration districts split, involving 38 changes in representative districts. He testified that it was necessary to use estimates; in some instances he employed what he termed a "map estimate" or made estimates by physical inspections of split enumeration districts, and in a very few situations involving seemingly small populations he "guessed" at the numbers of people living within given areas. Mr. McGinnes also employed a ratio of registered voters to total population in making some of his estimates. He used substantially the same method in estimating ratios as did Mr. L. Winfred Hughes and Mr. Raymond V. West, who testified in these proceedings before Mr. McGinnes testified and whose evidence in this respect is set out hereinafter. The substance of Mr. McGinnes' work in respect to the reapportionment of New Castle County, outside of Wilmington, is summed up in general in the affidavit filed by him.
McGinnes, 191-92.
Id. at 195.
Id. at 196-97.
Id. at 198.
Id. at 199.
The affidavit of Mr. McGinnes contains the following statement: "It has been estimated that the population of New Castle County has increased from 211,916 [sic.] in 1960 to 250,000 at the present time. Whole housing developments and large apartment units have been built and occupied in certain areas. The simple fact is that the population changes since 1960 completely invalidate the current accuracy of the 1960 Census District data and the degree of error involved in using this data far exceeds any error that could have resulted from the above-mentioned approximations." The court takes judicial notice of the substantial accuracy of Mr. McGinnes' statement quoted above and we find that the increase in population in Rural New Castle County is at least as large as stated by him.
Defendants' Ex. 4. Mr. McGinnes in his affidavit and in his testimony inadvertently misstated the population of Rural New Castle County. The actual figure was not 211, 916, prior to the increase, but was 211, 619. This is a minor typographical error.
The considered judgment of the Bureau of Vital Statistics, Delaware State Board of Health, as of July 1, 1965, was that a reasonable approximation of the population of New Castle County (Wilmington and Rural New Castle County) was 347,084 persons, and that the population of the City of Wilmington was 94,606 persons.
Mr. L. Winfred Hughes was the president of the Kent County Department of Elections and at one time was the chairman of the Kent County Democratic Committee. He stated that he assisted in preparing the legislative redistricting plan embraced by S.B. 332 and S.B. 336 for Kent County with some aid from the chairman of the Kent County Democratic Committee. Mr. Hughes made some reapportionment estimates for Kent County prior to the passage of S.B. 332 and S.B. 336. Mr. Hughes also employed a ratio of registered voters to total population in forming some of his estimates. An example will suffice. An unknown factor, and one which of course required determination, was the number of persons in the 2d election district of the old 2d representative district. There are, of course, 1960 census figures for the entire old 2d representative district including personnel of the Dover Air Force Base living in and around the Base but there are no 1960 census figures available for the individual election districts comprising the old 2d representative district. According to the 1960 census figures there were 17,806 persons in the old 2d representative district and, according to the Department of Elections of Kent County, there were approximately 4,000 registered voters in that district. Mr. Hughes therefore employed a ratio of approximately 4 to 1 to estimate the number of persons within the 2d election district of the old 2d representative district. Knowing there were 1079 registered voters in the old 2d election district of the old 2d representative district, Mr. Hughes estimated that there were 4,000 persons within the old 2d election district of the old 2d representative district. This appears from his testimony and his affidavit. Mr. Hughes frankly conceded in subsequent testimony that the 4,000 persons as estimated by him was perhaps too high. The number of persons, United States Government personnel, living in and around the Dover Air Force Base in the old 2d representative district has been much mooted in the course of this litigation but this factor, of course, has pertinency in respect to the equality of representation in the new 26th and 27th representative districts. Mr. Hughes composed the three new senatorial districts of Kent County by using portions of the old representative districts and old election districts of Kent County. He used estimates in respect to the 14th and 15th senatorial districts as appears from his affidavit.
Hughes, 83-84.
Id. at 84-86. Court's Exs. 17, 17(a), 38, 39 and 40.
Id. at 89 et seq., and Defendants' Ex. 2.
Defendants' Ex. 2.
Mr. Raymond V. West to a large degree prepared the reapportionment of Sussex County and estimated the populations included within the General Assembly districts in relation to S.B. 332 and S.B. 336. Mr. West testified that he was employed in the personnel department of the State Highway Department and was chairman of the Sussex County Democratic Committee; that he had served in the General Assembly of Delaware, on the Sussex County Drainage Commission, and had spent seven years on a local school board. He testified that he did not draw the representative and senatorial district lines, but that he supplied the information for "someone else who did draw the lines." Mr. West also employed ratios in some instances in arriving at estimates of populations employing substantially the same method as that used by Mr. Hughes and Mr. McGinnes. He stated that the basic method of making his determination was to take the population of each representative district of Sussex County from the 1960 Federal Census as published by the Development Department of the State of Delaware and dividing this figure by the total number of registered voters in the election districts composing each of the representative districts. He received the information respecting registered voters from the Department of Elections for Sussex County. He stated and this also appears from his affidavit that with the exception of the new 32nd representative district, the remaining representative districts of Sussex County, viz., the 30th, 31st, 33rd, 34th and 35th representative districts were created by adding the population of one or more of the old election districts. This, of course, required much estimating. As will appear hereinafter, there are complicating factors in Sussex County which presented difficulties in the reapportionment of that County.
West, 121 et seq., and Defendants' Ex. 3.
In respect to the issue of gerrymandering, Mr. West was explicit. He stated: "I had gotten information relative to redistricting. I mean figures and maps and information from the Board of Election of Sussex County or the Department of Election [sic.] of Sussex County. I had a committee with legal advice. I had connections with the members of the General Assembly that told me what they wanted and what they expected. I recall that the majority leader of the General Assembly in the House passed this information along to me. He said, `I will not go along with any gerrymandering; neither in your county, in my county, or any other county; and I want a plan that can be accepted, which I can sell here, and I don't expect to be charged with gerrymandering.'" Mr. West also testified as follows: "Now, I notice * * * that I am charged with gerrymandering a certain district. I would think it would be more logical for the Democratic Party to charge me with that, because down in the southwestern area of the district, in the southwestern area of the county, is the Democratic strength. And I think the records that probably are here today will verify what I am saying, that far more Democrats are put into a Representative District there than there are Republicans in the other districts. We just did this by taking these Election Districts * * * and working them out. And it worked out very nicely, where they fitted one against the other, and they seemed to be adjacent. And when we got out we put what was left over in No. 35. Some of them are swing districts. I think there is one of these that a Democrat is fairly sure of being elected in. I think there is one that the Republicans are fairly sure of being elected in. I think the remainder of those other four will depend." (Sic.)
West, 124-25.
A "swing" district may be defined in the context in which it was used by Mr. West, as a district which might go to the Republican or to the Democratic candidate by a narrow margin depending on particular or peculiar circumstances attending that specific election.
West, 126-27.
It should be pointed out that while Mr. Latchum, Mr. McGinnes and Mr. Hughes did not make as express denials of gerrymandering as did Mr. West, nonetheless these three gentlemen asserted stoutly and positively that they were endeavoring to deal with reapportionment fairly and equitably and on the basis of equal representation.
It will be observed from the testimony of Mr. Latchum that he, in a sense at least, was the progenitor of those portions of S.B. 332 and S.B. 336 which dealt with Wilmington, and the same we believe is shown by the record insofar as Mr. McGinnes is concerned in respect to New Castle County, outside of Wilmington. It also appears from the record that Messrs. Hughes and West were to some degree at least the creators of those portions of S.B. 336 which dealt respectively with the creation of the representative and senatorial districts in Kent and Sussex Counties. Messrs. Latchum, McGinnes, Hughes and West had in mind, respectively the number of representatives and senators to be allocated to the City of Wilmington, to New Castle County outside of Wilmington, to Kent and to Sussex Counties, and the 1960 census figures as to the number of inhabitants of Delaware in each of the geographical divisions referred to. It should be noted also that once S.B. 332 was enacted a Procrustean bed was created in which it was a virtual impossibility for substantially equal representation to be given to the inhabitants of Delaware on the basis of "one person, one vote". But it should also be observed that the period of time in which the four gentlemen whom we have named had available to work out a reapportionment system for the whole of Delaware was comparatively short for the decision of the Supreme Court in Roman v. Sincock was handed down on June 15, 1964 and the primary election in Delaware that year was to be held on Saturday, August 15th, the general election being held, of course, on November 3, 1964. In the two months period first stated candidates had to announce their candidacies, slates had to be prepared by the respective political parties, and the Departments of Elections had to prepare for the primaries if a general election expressing the will of the people of Delaware was to be conducted according to law on November 3rd. Time was indeed of the essence. The tasks the four men performed were large ones and were effected swiftly.
We shall deal more particularly with issues concerning deviations in population and gerrymandering hereinafter.
III. The Work of the Committee of 39 and the Reapportionment Plans of the Committee
The Committee of 39 is a volunteer non-profit organization devoted to the purpose of effecting good government in Delaware. It began its work in respect to reapportionment in Delaware shortly after the decision of the Supreme Court in Baker v. Carr, 369 U.S. 186, 82 S.Ct. 691, 7 L.Ed.2d 663 (1962), and it cannot be doubted that the members of this volunteer organization, in particular Mr. James Weaver and Mr. Sidney W. Hess, and Bruce M. Stargatt, Esquire, its counsel, have devoted many hours in their endeavors to aid the court in understanding the problems presented by this very extensive litigation. The Committee of 39, having been allowed to intervene as amicus curiae in these proceedings, presented inter alia a plan for the reapportionment of Delaware based upon extensive data and the use of a computer. This plan is set out in an article entitled "Procedure for Non-Partisan Districting: Development of Computer Techniques." A very substantial portion of this article is devoted to what is in substance the plan of the Committee of 39 for the reapportionment of Delaware by the use of computer techniques as will appear upon examination. We entertain little doubt but that the plan advanced by the Committee of 39 would serve the State of Delaware well. However, as the Committee of 39 points out the apportionment of 35 representatives among the counties, including Wilmington, on a population basis would result in a minimum deviations between the Counties of 7% in population-per-representative. This is due to the fixed number of representatives and fixed boundaries as set out in S.B. 332 and S.B. 336 which render substantially equal representation a practical impossibility. If fixed boundaries such as County lines, Wilmington City lines, and Hundred lines were obliterated and the data appearing from certain charts set out at a later point in this opinion were made use of as will appear, a better system of reapportionment could be devised for Delaware. It is asserted that such a course would entail an almost complete recasting of county governments and county functions and those of the City of Wilmington as well. We cannot find this to be a fact for the election of representatives and senators to the General Assembly could be operated as a system apart from the election of county officers at general elections with comparatively slight difficulties.
Weaver, 548-830, 1031-37, 1412-60, and 1862-1909.
73 Yale L.J. 288-308 (1963).
Id. at 294 n. 25.
It should be noted that the plan of the Committee of 39 as submitted to the court requires disregarding County lines and City of Wilmington lines in the following respects: senatorial districts 2, 3, 10 and 12 cross Wilmington City lines; senatorial district 7 crosses the New Castle County-Kent County line; and senatorial district 15 crosses the Kent County-Sussex County line. Representative districts 2, 3, 9 and 14 cross Wilmington City lines; representative district 28 crosses the New Castle County-Kent County line; and representative district 30 crosses the Kent County-Sussex County line. The plan of the Committee of 39 provides for 17 senatorial districts instead of 18 senatorial districts, but provides the same number of representative districts for the State of Delaware as do S.B. 332 and S.B. 336.
Court's Exs. 34 and 35.
Court's Exs. 32 and 33.
Weaver, 590, 618, 657 and 666.
But the duty of this court is not to impose a system of reapportionment upon the State of Delaware, if such a task can possibly be avoided, for as we have said many times the duty of apportionment lies with the Legislature. Our duty is to adjudicate the issue presented by the instant proceedings which is whether or not the reapportionment created by S.B. 332 and S.B. 336 meets the requirements of the Fourteenth Amendment as affording equal representation to the citizens of Delaware insofar as that be practicable.
IV. Plans of the Plaintiffs for Reapportionment
The plaintiffs have also suggested plans and means for the reapportionment of the State. We dealt with the first plan submitted, albeit somewhat briefly, in the appendix to our opinion in Sincock v. Duffy, 215 F. Supp. at 192-95. The plaintiffs' plan for apportioning the State of Delaware also requires crossing the Wilmington City limits in respect to senatorial district 4 and representative district 8.,
Plaintiffs' Ex. 48.
Plaintiffs' Exs. 47 and 52.
Niederhauser, 841.
The plaintiffs have now submitted another plan providing for the reapportionment of Wilmington and we shall discuss this in some detail at a later point in this opinion.
V. Data Submitted by Witnesses of the Parties and of the Committee of 39
The record which has been accumulated in this portion of the case, based on the filing by the plaintiffs of their motion to implement the mandate of the Supreme Court following its decision in Roman v. Sincock is a large and somewhat contradictory one. It consists of evidence given by 25 witnesses and a transcript of more than 3000 pages, taken largely during the trial on the merits and on various occasions between August 3 and December 21, 1965, inclusive. In addition, there are over 125 exhibits, many of which are maps in several series designated by subnumbers or letters under the principal exhibit numbers. These maps set up the contentions of the respective parties as well as those of the Committee of 39. A very considerable portion of the evidence was received prior to our order denying the plaintiffs' motion for preliminary injunction filed on September 16, 1964. See Sincock v. Roman, 233 F. Supp. 615 (1964). At this time we reviewed our determination of the defendants' motions to dismiss. Subsequent intensive study indicated that there existed substantial deficiencies in the record. Counsel in this case were rushed for time in the preparation of the record and in filing requests for findings of fact and conclusions, and the court itself felt it necessary to reach a conclusion on the motions to dismiss and for a preliminary injunction as soon as possible to the end that the Departments of Elections of the State could proceed efficiently with the general election of 1964.
On January 18, 1965 a memorandum was submitted to counsel by the court containing a number of complicated questions to be answered by counsel or by the parties. On January 25, 1965 the opinion of the Attorney General of the State of Delaware, Document 155, respecting the status of three hold-over Senators, was filed in this proceeding, and on January 26, 1965 a conference was held by the court and counsel. At this conference the court stated to counsel its desire to have in the record maps which would show the boundaries of each representative district and each senatorial district for the City of Wilmington, for New Castle County outside of Wilmington, for Kent County and for Sussex County in accordance with the reapportionment provided for by S.B. 332 and S.B. 336, prepared by the Departments of Elections of the State of Delaware as required by Section 661 of S.B. 336. It then appeared that the Department of Elections of New Castle County had prepared such maps but had not had them certified or filed and that the two other Departments of Elections of the State of Delaware had not caused the maps to be prepared. Counsel were requested to make every effort to have the maps required by Section 661 prepared, certified and filed as soon as possible.
On May 12, 1965, the maps, required by Section 661 not having been certified to or filed, the court, in order that the case might be more promptly disposed of, ordered a further conference to be held by all counsel, and ordered the plaintiffs, the defendants, and the amicus curiae to file briefs and requests for findings of fact and conclusions of law, and set the case down for argument on July 16, 1965.
In the same order the court directed the parties "To stipulate, insofar as is possible, to the admission in evidence of maps prepared, or in the course of preparation, as required by 29 Del.C. Ann. § 661 (Supp.) * * * [viz.,] maps showing the representative districts and senatorial districts throughout the State of Delaware by counties and in Wilmington, and the election districts, and the populations of each election district as of the time that S.B. 336 became law, it being the intention of the Court, if proper standards of proof be met, to admit such maps and figures in evidence, * * *." This portion of the order was subsequently amended by striking out any reference to the election districts and their respective populations.
In accordance with the court's order, counsel by a series of conferences, by telephone calls and by other communications, with the aid of the court, endeavored to secure the filing of the maps as required by Section 661. On June 1, 1965, it appearing that the Departments of Elections of New Castle and Kent Counties had complied with the provisions of Section 661 but that the Department of Elections of Sussex County had not so complied, the court issued an order directed to the members of the Sussex County Department of Elections to show cause why they should not comply with the provisions of Section 661. This rule to show cause was made returnable on June 7, 1965 and was duly served. On June 7, 1965 it appearing that the Department of Elections of Sussex County had complied with the provisions of Section 661, the rule was discharged. On June 16, 1965 a further conference was held by the court and counsel in an endeavor to see if counsel could not stipulate as to certain disputed areas in Kent and Sussex Counties. This stipulation, Document 174, was filed on June 16, 1965, and sets out undisputed facts in respect to certain Republican Senators who, as has been indicated, assert a "hold-over" right to sit in the General Assembly, purports to correct certain errors in testimony as to population in some localities as appears on examination, and stipulates the estimated 1960 populations of representative and senatorial districts in the State of Delaware under S.B. 336 insofar as agreement was possible at that time. We have used the term "purports" advisedly in the immediately preceding sentence because the parties on March 14, 1966, filed Document 195, which is intended to correct certain population figures relating to General Assembly districts in Kent County. Document 174, therefore, is to be read in conjunction with Document 195. Document 195 for the first time supplies "Election Results under S.B. 336, Senatorial Districts", "Exhibit 6-page 2". But as pointed out hereinafter certain corrected figures as stipulated to even in Document 195 do not coincide in some respects with those prepared by this court as demonstrated in the charts set out under heading "VI.", infra. The deviations, however, are minor ones and need not greatly concern this court or the reviewing Tribunal since the decision of the issues involved herein are not substantially affected by these small deviations.
No Democratic Senator presently asserts a "hold-over" right to sit in the General Assembly.
In respect to the maps required by Section 661, certified to and filed in the respective offices of the Recorders of Deeds in the three counties of Delaware, the stipulation contained in Document 174 states the following, "9. The respective Departments of Elections have caused to be prepared by registered land surveyors' maps of each county which purport to show the boundaries of the several General Assembly Districts. The registered land surveyors who prepared the maps have certified on them as to their correctness, and the respective Departments of Elections have caused the maps to be recorded in the Recorder's Office in the County in which they were prepared in compliance with the provisions of Subchapter VI, Section 661 of Senate Bill 336." The parties agreed that these maps should be marked and put into evidence as indicated in note 38, supra.
Stipulation Ex. 11, map of Wilmington, showing representative districts, Wilmington.
Stipulation Ex. 12, map of Wilmington, showing senatorial districts, Wilmington.
Stipulation Ex. 10A through 10H, inclusive (8 maps), New Castle County, showing representative and senatorial districts, as follows:
10A shows the 9th and 10th representative districts and the 5th senatorial district.
10B shows the 11th and 12th representative districts and the 6th senatorial district.
10C shows the 13th and 14th representative districts and the 7th senatorial district.
10D shows the 15th and 16th representative districts and the 8th senatorial district.
10E shows the 17th and 18th representative districts and the 9th senatorial district.
10F shows the 19th and 20th representative districts and the 10th senatorial district.
10G shows the 21st and 22nd representative districts and the 11th senatorial district.
10H shows the 23rd and 24th representative districts and the 12th senatorial district.
Stipulation Ex. 14, map of Kent County, showing senatorial districts.
Stipulation Ex. 13, map of Kent County, showing representative districts.
Stipulation Ex. 9, map of Sussex County, showing senatorial and representative districts (single map).
The maps of Wilmington and of New Castle County show the populations per representative district and senatorial district, as demonstrated by the records of the Department of Elections of New Castle County. The maps of Kent and Sussex Counties do not show population figures. This is because the parties were unable to stipulate as to the population figures in some districts.
The stipulation further provides that "the parties also agreed that the population figures referred to in Paragraph 1(b) of this Stipulation, and appearing on Exhibit S(1) attached to this Stipulation, represent the agreed 1960 population of the Representative Districts in New Castle County, including Wilmington. They also agree that the Plaintiffs' contention with regard to the population of Representative Districts 25, 28, and 29 are correct in Kent County, and that there remains in dispute only the populations of Representative Districts 26 and 27. The Exhibit sets forth the respective contentions of the Plaintiffs and Defendants with regard to those Districts.
"The Defendants' figures are derived from population estimates in five out of the six Sussex Representative districts, viz., 30, 31, 33, 34 and 35. As to these districts, Plaintiffs wish to make it plain that they do not agree that Defendants have presented satisfactory proof of the precise populations within said districts. However, for the purpose of this hearing, Plaintiffs will now indicate that they do not propose to present evidence in opposition to the populations.
"As to Representative District 32, the population of this District was derived from reliable census data, and Plaintiffs concede that the Defendants' contention as to the population of this District is factually correct.
"The parties all agree that the population figures referred to in Paragraph 1(b) of this Stipulation and appearing on Exhibit S(2) attached to this Stipulation represent the 1960 population of the Senatorial Districts of New Castle County, including Wilmington. With regard to Kent County, the parties agree with respect to the 13th Senatorial District, but disagree with respect to the 14th and 15th Senatorial Districts and have set forth in Exhibit S(2) their respective contentions with regard to the population of those Districts in 1960.
"With regard to the Senatorial Districts of Sussex County, the Plaintiffs incorporate herein the comments made with respect to the correlative Representative Districts.
"The parties stipulate that the maps prepared on behalf of the New Castle County Department of Elections, including Wilmington, correctly and accurately reflect the boundaries of the Representative Districts and the Senatorial Districts for that area. They further stipulate that the maps prepared for the Departments of Elections of Kent and Sussex Counties represent conscientious efforts on the part of the Departments of Elections of those Counties to present as accurately as possible on a single county map the District boundaries in accordance with the Representative and Senatorial lines employed by the Departments in the last five General Elections. Several of these Election Districts and Representative Districts are defined by map or written description only in terms of the boundaries of the traditional `hundreds' in both of these counties. In many instances, unambiguous descriptions of the hundreds do not exist; and it is impossible to determine in Sussex County the exact location of the boundary markers and lines which comprise the Representative and Senatorial Districts as they `heretofore' existed.
"With regard to the Sussex County map, it was determined that the Department of Elections had theretofore had a map which contained a significant deviation in boundaries between the present 34th and 35th Representative Districts. This deviation was continued in the map dated May 1, 1965, prepared by Messrs. Simpson and Burton, had no significance with regard to population and has been corrected in the map which they prepared and which is dated June 29, 1965."
The parties further stipulated that "after these corrections, the Representative District boundaries on these maps correspond to the boundaries used in conducting the last five General Elections for the corresponding Election Districts."
Following the admission of the maps into evidence and the filing of the stipulation, Document 174, the brief of the defendants and their proposed findings of fact and conclusions of law were filed on August 3, 1965. One day later the post-trial brief of the plaintiffs was filed. The plaintiffs stood upon their prior requests for findings of fact and conclusions of law as set out in their brief in support of their motion for a preliminary injunction and did not file new requests. But the "Conclusion" of the plaintiffs' "Post-Trial Brief" contains a series of prayers for relief which we will treat as an amendment to the pleading and as a part of the plaintiffs' motion to implement the mandate of the Supreme Court. These are discussed briefly later under heading "XVI. The Remedy."
See Plaintiffs' Post-Trial Brief, pp. 84-88.
On August 9th the maps hereinbefore designated in note 38, supra, were introduced in evidence and argument was had by the parties in respect to all issues deemed pertinent by them in the instant proceedings. On August 16th the counsel for the defendants filed a letter brief on the issue of gerrymandering, a brief which was replied to by the letter brief of the plaintiffs filed on August 30th. Despite the actions taken by the court and counsel as set out above, the court still found the record to contain certain deficiencies in respect to the maps required by Section 661 of S.B. 336 and on December 1, 1965 held a further conference with counsel and informed them as to what the court deemed to be some of the deficiencies. Subsequently, on December 15, 1965, counsel for the defendants filed a motion to reopen the record as will appear from their motion. The motion to reopen was not opposed by the plaintiffs. On December 20-21 the court held a further hearing to receive evidence on issues presented by the maps. The maps required by Section 661 were admitted into evidence as appears in note 38, supra, subject to the minor deficiencies referred to under heading "XIII.", infra.
One other exhibit requires mention and that is plaintiffs' Ex. 60. This exhibit purported to show "1956-1962 Voting Records for State Representatives and State Senators Arranged in Total by Legislative Districts". The original figures for this exhibit were prepared for the Committee of 39, by Mr. Donald Niederhauser, from returns published by the "News-Journal newspapers", the Wilmington "Morning News" and the Wilmington "Evening Journal", and not from the returns made to the respective Prothonotaries of the three Counties in accordance with Section 4978, Title 15, Del.C. Ann. The figures contained in the original exhibit, therefore, were not official figures and the court deemed it necessary that the exhibit be corrected to the end that the official figures might be before the court. The present exhibit is entitled "Plaintiff's [sic.] Exhibit 60 Substitute" and was filed on March 14, 1966.
With the filing of Document 195 correcting Document 174 and supplying the court with the election results under S.B. 336 as to senatorial districts, "Exhibit 6 — page 2", the evidence in this protracted case at long last seems to be complete.
The court endeavored to digest and understand this extensive and difficult record. It contains many, many population figures which the parties and the Committee of 39 and its counsel have attempted, in many instances successfully, to tie to specific geographic areas such as representative districts and senatorial districts, both old and new, to election districts, and in respect to the City of Wilmington even to block-by-block locations. But many of these figures are conflicting and to detail the testimony of the various witnesses herein would create an opinion of greater than permissible length. To the end that all the issues may be brought into focus, insofar as we are able to do so, a series of charts are set out in the next heading of this opinion which we think will serve to clarify the record and, we hope, will give perspective to the issues which we must adjudicate. To the end that the complete picture may be presented we have included in certain of the charts referred to facts brought upon the record during the first trial of this case; viz., Sincock v. Duffy, supra, 215 F. Supp. 169.
VI. Tables
Table I below shows the House of Representatives by political subdivisions. Explanations of its abbreviations and symbols are set out in the footnote. TABLE I. House of Representatives by Political Subdivisions 1960 Pop. ÷ by No. of Rep. Act. Pers. Deviations Pct. Dev. If Wilm. had 7 R.N.C. 17 Rep. Subdivision Census 12,751 Allotted Per Rep. from Mean from Mean Act. Pers. Dev. fr Pct. 12,751 Per Rep. Mean Dev.
Column 1 of Table I indicates the geographic subdivisions of Delaware; viz., "Wilmington", "R. New Castle" for Rural New Castle County, "New Castle (Combined)", indicating the City of Wilmington and New Castle County outside of Wilmington combined, and "Kent" and "Sussex" indicating, of course, Kent and Sussex Counties, and the word "Total" signifies the total population of the State.
Column 2 entitled "1960 Census" gives to the political areas as designated the respective population figures per area as shown by the 1960 Census with a total of 446,292 persons.
At the bottom of "Table I" is shown the total population of Delaware as indicated by the 1960 Census, viz., 446,292, divided by 35, the number of representatives provided by S.B. 332, and the figure of 12,751 as the "Mean or Perfect Representative District", if perfect apportionment were possible. In respect to Table I, and indeed in respect to all the tables in this opinion where "Mean" or "Perfect" is used in connection with the word "district", under the circumstances at bar, the words "Norm" or "Average" could be employed with equal accuracy.
Column 3 entitled "Pop. ÷ by 12,751" shows the population for each geographic subdivision by dividing the total population of each such subdivision by the mean or perfect representative district population of 12,751.
Column 4, entitled "No. of Rep. Allotted", shows the number of representatives allotted per political area as indicated in Column 1 pursuant to S.B. 332 and S.B. 336, and Column 5 entitled "Act. Pers. Per Rep." means the actual number of persons which would be in each representative district if the number of representatives allocated to the particular area was divided into the total census population found by the 1960 Federal Census to exist in that particular designated area.
Column 6, "Deviations from Mean 12,751", needs no explanation.
Column 7, entitled "% Deviation from Mean", shows plus or minus wise the deviation from the mean, viz., the percentage deviation from the mean or perfect representation based on the census figures.
Columns 8, 9, and 10 show the actual percentage per representative, the deviation from the mean and the percentage of deviation assuming that Wilmington had seven and Rural New Castle County had seventeen representatives. The reason for the figures contained in the last three columns is because of the allegation to be dealt with hereinafter that S.B. 332 and S.B. 336 allocate to Wilmington one more representative than it should and allocate to New Castle County one less representative than it should, treating these two respective geographic areas as wholes.
Table II below deals with the Senate by political subdivisions and performs substantially the same functions with regard to population figures for senatorial districts as does Table I for representative districts. TABLE II. Senate — By Political Subdivision 1960 Pop. ÷ by Act. No. Persons Pct. Subdivision Census 24,794 Allotted Per Senator Deviation Dev.
At the bottom of Table II is shown the total population of Delaware as indicated by the 1960 Census, viz., 446,292, divided by 18, the number of senators provided by S.B. 332, and the figure of 24,794 as the "Mean or Perfect Senatorial District", if perfect apportionment were possible. As to the meaning to be given to the words "Mean" or "Perfect", see the third paragraph of note 40, supra.
Table III below demonstrates the allocation of voting power per political subdivision in relation to populations. TABLE III. Allocation of Voting Power Per Political Subdivisions in Relation to Population Political 1960 % of State Pop., No. of Rep. % of Power No. of Sen. % of Power No. in % of Power Deviation of % of Subdivision Census 1960 Census S.B. 332 in House S.B. 332 in Senate Gen. Assem. (Total) in Gen. Assembly Power Minus % of Pop.
Column 1 states the political subdivisions of Delaware substantially as does Column 1 in "Table I" and "Table II." Column 2 sets out the figures supplied by the 1960 Federal Census. Column 3 states the percentage of state population according to the 1960 Census by geographic locality as indicated. Column 4, as appears, shows the number of representatives per geographic area in accordance with S.B. 332. Column 5 shows the percentage of voting power in the House of Representatives according to S.B. 332 in the respective geographic areas. Column 6 shows the number of senators allocated to each geographic area by S.B. 332. Column 7 shows the percentage of voting power in the Senate in the respective geographic areas. Column 8 shows the total number of members, both representatives and senators, in the General Assembly for each geographic area. Column 9 shows the percentage of voting power in the General Assembly for both representatives and senators in the General Assembly. Column 10 is headed "Deviation of % of Power Minus % of Pop." This legend means that Column 10 shows the deviation of voting power percentage-wise for each geographic area. These figures are arrived at by subtracting in each instance the percentage of population as shown in Column 3 from the percentage of voting power in the General Assembly as shown in Column 9.
It should be observed that if the figures in Column 3 are larger than the figures in Column 9, a negative figure is arrived at, as in the case in respect to Rural New Castle County.
Table IV shows the allocation of voting power per political subdivision in relation to population if Wilmington had 7 representatives and Rural New Castle had 17. TABLE IV. Allocation of Voting Power Per Political Subdivisions in Relation to Populations if Wilmington had 7 Representatives and Rural New Castle County had 17 Representatives Political 1960 % of State No. of Rep. % of Power No. of Sen. % of Power No. in Gen. % of Power in Deviation of % of Subdivision Census Pop., — 1960 (Hypo.) in House S.B. 332 in Senate Assem. Gen. Assem. Power Minus % of Pop.
Only one of the columnar headings of Table IV perhaps requires explanation. Column 4 is designated "No. of Rep. (Hypo.)". The word "Hypo." indicates that the number of representatives set out in Column 4 is hypothetical, based on the assumption of 7 representatives for the City of Wilmington and 17 representatives for New Castle County, outside of Wilmington.
Table V shows the population of Delaware by political subdivisions for the years 1930, 1940, 1950, 1960, and in the single instance of Rural New Castle County, i.e. New Castle County outside of Wilmington, the population for 1964. TABLE V. Populations by Political Subdivisions 1930-1960 Political 1930 1940 Inc. Pct. Inc. 1950 Inc. Pct. Inc. 1960 Inc. Pct. Inc. 1964 Inc. Pct. Subdivision Pop. Pop. (Dec.) (Dec.) Pop. (Dec.) (Dec.) Pop. (Dec.) (Dec.) Pop. (Dec.) Inc.
In the single instance of New Castle County, as appears from the chart itself, the population figure for 1964 is arrived at by adding the figure of 38,000 persons as shown by Mr. McGinnes' affidavit to the number of persons shown by the 1960 census figure. The population figures with the exception of the 1964 population figure in Column 12, relating to New Castle County, appear as set out in our opinion in Sincock v. Duffy, 215 F. Supp. at 174-75.
The legend at the head of Column 4 shown as "Inc. (Dec.)" means a possible increase or decrease though it should be noted that the figures in Column 4 all demonstrate increases. Column 5 shows the percentage of increases or decreases in population over the 1930 federal census. Column 7 shows that Wilmington's population had decreased by 2,148 persons since the 1940 federal census but all the other figures in the column show increases. Column 8 shows the percentage of increases or decreases in the populations as of the 1950 federal census and demonstrates that the population of Wilmington has decreased 1.9% while there had been substantial increases as indicated in the other geographical areas with which we are concerned. The other column heads speak for themselves but it should be noted that the decrease of population in Wilmington is shown in Column 11 as 13.2% in 1960 and that the last three columns show that the population of Rural New Castle in 1964 was 250,000 persons indicating an increase of 38,381 persons over the population shown by the 1960 census, an increase of 18.1%. The figures mentioned in respect to the last three columns, 12, 13 and 14, of Table V are supported by the affidavit of Mr. McGinnes, Defendants' Ex. 4.
Source — Affidavit of F. Earl McGinnes, Defendants' Ex. 4.
Table VI below shows estimated populations in relation to the House of Representatives and the population of Delaware set up pursuant to S.B. 336, using the norm or mean figure of 12,751 persons. TABLE VI. House of Representatives, Districts (S.B. 336) and Populations (Norm or Mean 12,751) District Est. Dev. from % Comm. of 39 Corr. Dev. from % Pop. Norm or Mean Dev. Indet. Min. Indet. Max. Pop. Norm or Mean Dev.
The representative districts are set out in Column 1; the estimated populations in Column 2. The figures from Column 2 are taken from the affidavits, respectively, of Mr. Latchum, Defendants' Ex. 5; of Mr. McGinnes, Defendants' Ex. 4; of Mr. West, Defendants' Ex. 3; and of Mr. Hughes, Defendants' Ex. 2. Column 3 shows arithmetically the deviation of the population figures from the norm or mean. Column 4 shows the percentages of deviations. Column 5 shows the figures of the Committee of 39 which used indeterminate minimums and indeterminate maximums in a manner and for the reasons which will be discussed later in the opinion. Wherever a blank appears in that portion of Column 5 dealing with indeterminate minimums and indeterminate maximums, it means the Committee of 39 followed the actual data of the 1960 Federal Census and that there were no split enumeration districts.
For the sake of clarity we have cited the affidavits in the order in which the figures are employed in Table VI.
Table VII shows the estimated populations in relation to senatorial districts and the population of Delaware pursuant to S.B. 336, using the norm or mean of 24,794 persons. TABLE VII. Senatorial Districts (S.B. 336) and Populations (Norm or Mean = 24,794) District Est. Dev. from % Comm. of 39 Corr. Dev. from % Pop. Norm Dev. Indet. Min. Indet. Max. Pop. Norm or Mean Dev.
The format of Table VII is substantially the same as that of Table VI except for differences required by the fact that we are dealing with senatorial districts instead of representative districts. Again we point out that where blanks appear in that portion of Column 5 dealing with the figures of the Committee of 39 which uses indeterminate minimums and indeterminate maximums, it means that the Committee of 39 followed the actual data of the 1960 Federal Census and there were no split census enumeration districts.
Table VI and Table VII are to be considered in conjunction with Court's Ex. 41, entitled "New District Populations for Delaware's General Assembly Under Senate Bill 336." This exhibit presents a valuable analysis of S.B. 336 and has enabled the court to test the accuracy of the population estimates as asserted by the defendants.
Putting the matter briefly the apportionment plan created by the Legislature by S.B. 332 and S.B. 336 resulted in some splitting of census enumeration districts. In the case of Delaware, a census enumeration district is the smallest unit in which census figures are collected or maintained, outside of Wilmington. In Wilmington, census data was available for the most part by city blocks. In the instant case a split census enumeration district occurred when a portion of the geographical area of a census enumeration district was placed in one representative district or senatorial district while the remaining portion of it was placed in another representative or senatorial district. In one or two instances census enumeration districts were split and were added to one or more representative or senatorial districts.
When a census enumeration district is split to form representative or senatorial districts it is impossible, unless block census figures are available as for the City of Wilmington, to tell from census figures how much of the population of each portion of the split census enumeration district is included in each representative or senatorial district. For example, assume that census enumeration district A-1 contains a total population of 900 persons and assume further that census enumeration district A-1 is split between representative district No. 1 and representative district No. 2, it cannot be told from the census figures how many of the 900 persons specified are included in district No. 2. Therefore, an estimate is necessary. The problem increases in difficulty when a representative district contains portions of several split enumeration districts. When this occurs, the possibility of error necessarily is enlarged because the number of estimates is increased.
The accuracy of such estimates can, however, be checked within a limited range. For example, assume that hypothetical Representative District No. 1 is composed of the following hypothetical census enumeration districts with the following hypothetical populations:
A-2 5,000
A-3 3,000
A-4 2,000 ------ Total 10,000
Assume further that Representative District No. 1 also contains portions of the following census enumeration districts with the following total populations:
A-1 2,000
A-5 3,000 ----- Total 5,000
Now, as indicated, we cannot tell from the available census material how many of the 5,000 persons contained in the two split census enumeration districts reside in Representative District No. 1. But, we do know that it can contain no fewer than 0 and no more than 5,000 persons from these two split census enumeration districts. Therefore, we can calculate a range for Representative District No. 1 of a minimum of 10,000, viz., 10,000 known population plus zero from split census enumeration districts, to a maximum of 15,000, viz., 10,000 known plus 5,000 from the two split census enumeration districts.
Assuming the accuracy of these figures, it is a mathematical impossibility for Representative District No. 1 to contain fewer than 10,000 or more than 15,000 persons. Therefore, if the estimated population for this representative district was 9,000 persons, or 16,000 persons, we would know at once that the estimate was inaccurate and must be in error by at least 1,000 persons.
Absolute minimums and maximums by representative and senatorial districts were computed by the Committee of 39 and are shown in Court's Ex. 41. These figures, "Indeterminate Minimum" and "Indeterminate Maximum" have been transposed respectively into Column 5 of Tables VI and VII.
Table VIII, set out below, shows the population deviations within the City of Wilmington created by S.B. 332 and S.B. 336 providing for 8 representatives and 4 senators.
When we use the term "deviations" we mean deviations from the norm or mean within the City of Wilmington. In so stating, we point out that the state-wide norm or mean per representative is 12,751 persons and the state-wide norm or mean per senator is 24,794 persons.
In connection with the figures of this table and other data we will discuss the plaintiffs' plan for the apportionment of the City of Wilmington as submitted by their Counsels' letter of August 30, 1965 at a later point in this opinion.
TABLE VIII. City of Wilmington Population Deviations (House of Representatives Norm or Mean equals 95,827 divided by 8 equals 11,978) (Senate Norm or Mean equals 95,827 divided by 4 equals 23,957)=============================================================== District Population Deviation Pct. from Norm or Deviation Mean --------------------------------------------------------------- Rep. Districts #1 11,710 -268 -2.24 #2 11,853 -125 -1.04 #3 11,890 — 88 — .73 #4 11,834 -144 -1.20 #5 12,917 939 7.84 #6 12,592 614 5.13 #7 11,917 — 61 — .51 #8 11,114 -864 -7.21 --------------------------------------------------------------- Total 95,827 =============================================================== Senate Districts #1 23,563 -394 -1.64 #2 23,724 -233 — .97 #3 24,630 673 2.81 #4 23,910 — 47 — .20 --------------------------------------------------------------- Total 95,827 ===============================================================
Ratios: Largest Representative District (5th) to Smallest (8th) ............................... 1.16
Largest Senatorial District (3rd) to Smallest (1st) ............................... 1.05
Table IX below shows the population deviations within Rural New Castle County created by S.B. 332 and S.B. 336 providing for 16 representatives and 8 senators.
We reiterate that the deviations from the norm or mean shown are deviations within Rural New Castle County itself, and we point out again as we did in connection with Table VIII that the state-wide norm or mean per representative is 12,751 persons, and the state-wide norm or mean per senator is 24,794 persons.
============================================================== Senate Districts #5 26,860 408 1.54 #6 27,715 1263 4.77 #7 26,618 166 .63 #8 25,686 -766 -2.90 #9 26,470 18 .07 #10 25,414 -1038 -3.92 #11 26,960 508 1.92 #12 25,896 -556 -2.10 -------------------------------------------------------------- Total 211,619 ==============================================================
Corrected population estimate, see Table VII, Column 6.
Corrected population estimate, see Table VII, Column 6.
Includes 1,610 auditor votes for 20th Rep. Dist. where no Republican ran.
Ratios: Largest Representative District (12th) to Smallest (20th) = 1.15
Largest Senatorial District (6th) to Smallest (10th) = 1.09
Table X shows the population deviations within Kent County created by S.B. 332 and S.B. 336 providing for 5 representatives and 3 senators.
We reiterate that the deviations from the norm or mean shown in this Table are deviations within the County itself, and we point out again as we did in connection with Tables VIII and IX the state-wide norm or mean per representative is 12,751 persons and that the state-wide norm or mean per senator is 24,794 persons.
Ratios: Largest Representative District (28th) to Smallest (29th) — 1.25
Largest Senatorial District (15th) to Smallest (13th) — 1.15
Table XI, set out below, shows the population deviations within Sussex County created by S.B. 332 and S.B. 336 providing for 6 representatives and 3 senators.
We reiterate that the deviations from the norm or mean shown are deviations within the county itself. We point out, as we have previously, that the state-wide norm or mean per representative is 12,751 persons and that the state-wide norm or mean per senator is 24,794 persons.
Ratios: Largest Representative District (32nd) to Smallest (34th) = 1.22
Largest Senatorial District (16th) to Smallest (18th) = 1.12
Table XII, Stipulation " Exhibit 3 (Exhibit S-1", and Stipulation " Exhibit 4 (Exhibit S-2)": We have referred several times to Document 174, the stipulation filed July 16, 1965. It is now necessary to refer to it again. The parties with the aid of the Committee of 39, and in particular with that of Mr. Weaver, acting for the Committee of 39, filed with Document 174 an exhibit designated as " Exhibit 3 (Exhibit S-1)", the latter designation, that enclosed in parentheses, meaning Stipulation Exhibit 1. Though this is designated as a "Proposed Stipulation" and as such was not approved by the Court when filed, nonetheless we find it was the intention of the parties that it be treated as binding upon them. The extent to which it will be approved by the court is stated later in this opinion. Stipulation Exhibit 1 is based on checking census totals and compromises reached by the parties following a meeting on February 3, 1965 and following other meetings, and includes allegedly "slight changes" from certain figures set up in the defendants' motion for summary judgment, Document 128, and from the figures set up in the plaintiffs' proposed findings of fact, Document 149, p. 11, as appears.
Cf. Orange Theatre Corp. v. Rayherstz Amusement Corp., 130 F.2d 185, 186 (3 Cir. 1942). See, United States v. Reading Co., 289 F.2d 7, 9 (3 Cir. 1961). Compare Derewecki v. Pennsylvania R. Co., 353 F.2d 436 (3 Cir. 1965).
The representative districts are set out for New Castle County, including Wilmington, in Column 1, as are those for Kent and Sussex Counties. The population figures asserted on behalf of the defendants are set out in the second column and the populations asserted by the plaintiffs are set out in the third column.
Document 174 also contains a "Proposed Stipulation" designated by the parties as " Exhibit 4 (Exhibit S-2)", the latter designation, that enclosed in parentheses, meaning Stipulation Exhibit 2 which we will treat in pari materia with Exhibit 3 (Exhibit S-1) insofar as admissibility is concerned and hence deem it to be part of the record in this case and as expressing the intent of the parties. Exhibit S-2 displays a setup as stipulated to by the parties for senatorial districts alike to that stipulated to by the parties in respect to representative districts as shown by Stipulation Exhibit 1.
It will be noted that in respect to Document 174 a dispute existed between the parties as to the populations of the 26th and 27th representative districts in Kent County. A portion of paragraph 9 of the stipulation, pp. 5-6, states the parties "agree that the Plaintiffs' contention with regard to the population of Representative Districts 25, 28 and 29 are correct in Kent County and that there remains in dispute only the population of Representative Districts 26 and 27. The exhibit sets forth the respective contentions of the Plaintiffs and Defendants with regard to these districts."
The stipulation, paragraph 9, pp. 6-7, states in respect to the disputed senatorial districts in Kent County the following: "The parties all agree that the population figures referred to in Paragraph 1(b) of this Stipulation and appearing on Exhibit S(2) attached to this Stipulation represent the 1960 population of the Senatorial Districts of New Castle County, including Wilmington. With regard to Kent County, the parties agree with respect to the 13th Senatorial District, but disagree with respect to the 14th and 15th Senatorial Districts and have set forth in Exhibit S(2) their respective contentions with regard to the population of those Districts in 1960.
"With regard to the Senatorial Districts of Sussex County, the Plaintiffs incorporate herein the comments made with respect to the correlative Representative Districts."
It must be noted also that Document 174 has been corrected, pursuant to the agreement of the parties as set out in Document 195 supplementing Document 174.
The last page of Table XII, marked stipulation 1 continued, is an elaboration of the respective contentions of the parties in regard to the populations of representative and senatorial districts of Kent County. We have included it because it reduces the controversies in respect to populations and geographic areas as asserted by the parties to definite cognizable form.
The last page of Table XII, marked "(S-1 cont.)" is included in the exhibit as a second page to Stipulation Exhibit 1, but since this document deals with the respective contentions of the parties in regard to the populations of both representative and senatorial districts in Kent County, we deemed it more appropriate to put the so-called continuation of Stipulation 1 at the end of Table XII and have done so.
TABLE XII. " Exhibit 3 (Exhibit S-1)"
(Estimated 1960 populations of Representative Districts in accordance with S.B. 336, using figures from the "Proposed Stipulation" based on checking census totals and compromises reached at meeting of 2/3/65 as set out in the text following the heading "Table XII". The heading of the exhibit has been rephrased for the sake of clarity.) Representative District Defendants Plaintiffs New Castle 1 11,710 11,710 2 11,853 11,853 3 11,890 11,890 4 11,834 11,834 5 12,917 12,917 6 12,592 12,592 7 11,917 11,917 8 11,114 11,114 9 12,932 12,932 10 13,928 13,928 11 13,791 13,791 12 13,924 13,924 13 13,842 13,842 14 12,776 12,776 15 12,826 12,826 16 12,860 12,860 17 13,609 13,609 18 12,861 12,861 19 13,223 13,223 20 12,741 12,741 21 13,213 13,213 22 13,197 13,197 23 12,896 12,896 24 13,000 13,000 ------- ------- Subtotal 307,446 307,446Kent
25 13,161 13,161 26 13,806 15,486 27 13,125 11,445fn_ 28 13,748 14,209 29 11,811 11,350 ------ ------ Subtotal 65,651 65,651
State auditor vote used in districts where no Republican candidate ran.
Sussex
30 12,798 12,798 31 12,719 12,719 32 13,458 13,458 33 11,534 11,534 34 11,017 11,017 35 11,669 11,669 ------ ------ Subtotal 73,195 73,195
TOTAL 446,292 446,292
" Exhibit 4 (Exhibit S-2)"
(Estimated 1960 populations of Senatorial Districts in accordance with S.B. 336, using figures from the "Proposed Stipulation" based on checking census totals and compromises reached at meeting of 2/3/65 as set out in the text following the heading "Table XII". The heading of the exhibit has been rephrased for the sake of clarity.) Senatorial District Defendants Plaintiffs New Castle1 23,563 23,563 2 23,724 23,724 3 24,630 24,630 4 23,910 23,910 5 26,860 26,860 6 27,715 27,715 7 26,618 26,618 8 25,686 25,686 9 26,470 26,470 10 25,964 25,964 11 26,410 26,410 12 25,896 25,896 ------- ------- Subtotal 307,446 307,446Kent 13 20,743 20,743 14 21,005 19,325fn_ 15 23,903 25,583fn_ ------ ------ Subtotal 65,651 65,651
Sussex 16 25,517 25,517 17 24,992 24,992 18 22,686 22,686 ------ ------ Subtotal 73,195 73,195
TOTAL 446,292 446,292
Kent County STIPULATION EXHIBIT NO. 1 S-1 (cont.)Representative Districts 25th 13,161 (census figure)
26th 17,806 (census) less 4,000(a) = 13,806 (defendants' estimate) 17,806 less 2300(b) = 15,506 (plaintiffs' estimate)
27th 9,125 (census) plus 4,000(a) = 13,125 (defendants' estimate) 9,125 plus 2,300(b) = 11,425 (plaintiffs' estimate)
28th 14,209 (census figure)
29th 11,360 (census figure)
Senatorial Districts 13th 20,753 (census figure)
14th 17,005 (census) plus 4,000(a) = 21,005 (defendants' estimate) 17,005 plus 2,300(b) = 19,305 (plaintiffs' estimate)
15th 10,097 (census) plus 13,806(a) = 23,903 (defendants' estimate) 10,097 plus 15,506(b) = 25,603 (plaintiffs' estimate)
(a) For defendants' estimate see testimony of L. Winfred Hughes, Tr. pp. 109-13, 1008-09.
(b) For plaintiffs' estimate see testimony of Donald Niederhauser, Tr. p. 1702.
Table XIII shows the Democratic and Republican votes cast for representatives and senators in the 1962 election by seats in accordance with the geographic areas as specified: i.e., the "Representativeness" of the election. TABLE XIII. Representativeness of 1964 Election (Republican Seats vs. Republican Votes) Republican Votes Seats % Dem. Rep. Total Dem. Rep. Total Votes Seats REPRESENTATIVES fn_ SENATE
"Outer New Castle Cy." means Rural New Castle County outside of Wilmington. "New Castle County" includes the vote of Rural New Castle County and Wilmington. The second sheet of Table XIII states the same information in respect to the 1964 general election. The third and fourth sheets of Table XIII show the election results under S.B. 336, the votes for representatives, Democratic and Republican, by representative districts for the general elections in 1960, 1962 and 1964.
Includes 246 Independent votes.
1960 and 1962 results combined because only half of election districts voted for state senator in each election.
Includes 4,676 auditor votes for 2nd and 4th Sen. Dist. where no Republican ran.
Includes 612 Independent votes.
Includes 7,485 auditor votes for 10th and 11th Sen. Dist. where no Republican ran.
Corrected December 21, 1965 to agree with official Tally Sheets, Boards of Canvass, Prothonotaries' Offices of Kent, Sussex and New Castle Counties.
ELECTION RESULTS UNDER S.B. 336 Vote for Representative fn__
------------------------------------------------------------------------ Representative 1964 1962 1960 --------------------------------------------------------- District R D R D R D ------------------------------------------------------------------------ # 1 (Wilm.) 2469 3887 2 " 1921 3142 3 " 539 3163 4 " 921fn_ 3333 5 " 910fn_ 3443 6 " 2459 3157 7 " 1276 3940 8 " 2392 3270 ------- -------- Wilm. Total 12,887 27,335
9 (Rural N.C.) 3732 2853 10 " 3009 3154 11 " 4970 3014 12 " 5202 2612 13 " 6980 2850 14 " 5119 3044 15 " 3051 3339 16 " 2701 3010 17 " 2113 3729 18 " 2286 3578 19 " 2458 4056 20 " 1610fn_ 2955 21 " 1260 4080 22 " 2466 3779 23 " 2376 3290 24 " 2598 2953 -------- -------- -------- -------- R.N.C. Total 51,931 52,296 41,457 33,009
25 (Kent) 1925 3220 1976 2655 2564 2667 26 " 1106 1431 776 1111 1196 1243 27 " 2032 2418 1887 1551 2227 1740 28 " 2205 2643 1771 2230 2269 2466 29 " 1819 2754 1763 2347 2064 2868 ------- -------- ------- ------- -------- -------- Kent Total 9,087 12,466 8,173 9,894 10,320 10,984
30 (Sussex) 2279 2831 2149 2220 2569 2502 31 " 2178 3204 1866 2292 2636 2776 32 " 2506 3259 1722 2716 2242 3292 33 " 2333 2797 1769 2638 2295 2576 34 " 2352 2799 2680 1861 2915 2189 35 " 2500 3239 2534 2509 3123 2623 -------- -------- -------- -------- -------- -------- Sussex Total 14,148 18,129 12,720 14,236 15,780 15,958 ======== ========= ======== ======== ======== ======== State Total 88,053 110,226 67,040 71,182
ELECTION RESULTS UNDER S.B. 336 SENATORIAL DISTRICTS
Vote for State Senator --------------------------------------------------------- Senatorial 1964 1960-1962 fn__ -------------------------------------- District R D R D --------------------------------------------------------- 1 (Wilm) 4492 6586 2 " 1352fn_ 6617 Not Available Split Elect. Dists. 3 " 3207 6286 4 " 3324fn_ 7384 -------- -------- Wilm. Total 12,375 26,8735 (Rural N.C.) 6589 6087 6 " 9817 6019 7 " 12670 5424 Not Available 8 " 6525 5607 Split Election Dists. 9 " 4430 7256 10 " 3997fn_ 7161 11 " 3488fn_ 8171 12 " 4929 6285 -------- -------- Rural N.C. Total 52,445 52,010
13 (Kent) 3088 4939 3480 4619 14 " 3308 3871 2893 2582 15 " 2721 3708 2346 2984 ------- -------- ------- -------- Kent Total 9,117 12,518 8,719 10,185
16 (Sussex) 4522 6002 4262 5211 17 " 4854 6019 4090 5549 18 " 5924 5009 5766 4233 -------- -------- -------- -------- Sussex Total 15,300 17,030 14,118 14,993 ======== ========= State Total 89,237 108,431
Source: Official Tally Sheets, Boards of Canvass, Prothonotaries' Offices of Kent, Sussex and New Castle Counties.
VII. The Plaintiffs' Plan for Apportioning Representative Districts in the City of Wilmington
Document 181 is a letter dated August 30, 1965, delivered to the court by counsel for the plaintiffs, Vincent A. Theisen and Victor Battaglia, Esquires. The letter addresses itself to a plan for the City of Wilmington which would avoid large deviations in the representative districts in the City. It is apparent that Mr. Latchum in preparing the representative districts for the City of Wilmington largely ignored the historic boundaries which ran easterly and westerly in the City of Wilmington, and it is alleged that he reversed the course of the districts so that they now run northerly and southerly, thereby splitting and eliminating pockets of Republican strength contained in the old 7th and 9th Wards of the City of Wilmington and compressed 41% of the total Negro population into one district consisting of 84% Negro population. The defendants assert that this is simply the result of Mr. Latchum having followed census tract boundaries. One of the issues presented by Document 181 is gerrymandering. This will be dealt with at a later point in this opinion. We are presently concerned with aspects which relate primarily to differences in population of representative districts within the City of Wilmington. The plaintiffs assert that the tools were and are at hand with which proper allocations of populations to representative districts easily could have been made and that this is a potent factor to be considered in respect to representation as equal as practicable in order that invidious discrimination be avoided as required by the Fourteenth Amendment.We set out immediately below as Table XIV the chart contained in Document 181 in its precise words and figures.
TABLE XIV. City of Wilmington — Population of Representative Districts
Adjusted District Lines to S.B. 336 Reduce Population Differences Pop. Dev. from Ideal Pop. Dev. from Ideal
#1 11,710 -268 11,953 -25 #2 11,853 -125 12,021 +43 #3 11,890 -88 11,972 -6 #4 11,834 -144 11,917 -61 #5 12,917 +939 11,995 +17 #6 12,592 +614 11,961 -17 #7 11,917 -61 12,031 +53 #8 11,114 -864 11,977 -1
Ratio of Largest to Smallest 1.16:1 1.01:1 Ideal Population: 11,978
The first column shows the representative districts in the City of Wilmington by number. The second column shows the population by representative districts under S.B. 336. The third column, entitled "Dev. from Ideal" represents the number of persons by which each representative district deviates from the ideal population which we have described earlier as the "norm" or "mean". The fourth column shows Mr. Theisen's plan's adjustments of representative district lines to reduce the population differences, and the fifth column entitled "Dev. from Ideal" shows the number of persons by which each district deviates from the ideal population which we have described as the "norm" or "mean".
To the end that it may be demonstrated how Mr. Theisen's plan works, we have set out in Table XV in note 56, infra, representative district by representative district in the City of Wilmington, the number of the block census and the total population by block census, and Mr. Theisen's final population figure as contrasted with Mr. Latchum's population figure, as well as a statement as to the number of persons under or over the "ideal", or mean, population figure for each representative district accompanied by brief descriptive formulae. We also have appended to Table XV, representative district by representative district, boundaries of added or subtracted areas.
VIII. Kent County in respect to the Dover Air Force Base and Its Environs and Delaware State College
A substantial portion of the evidence in this case and several maps have been devoted to the issues presented by the Dover Air Force Base and Delaware State College and the voting areas surrounding them. As we have indicated at an earlier point in this opinion, Mr. L. Winfred Hughes in preparing the re-districting of Kent County combined old representative districts to comprise new representative and new senatorial districts, but we lay emphasis on the fact that old election districts were not split. It was stated that the reason that old election districts were not split was that the boards of elections in the various election districts lacked the addresses of the registered voters in many of the more rural portions of Kent County. It followed that if new election districts were created it would have been impossible to determine which registered voters lived in the new election districts in time to hold the primary election in August 1964, or even in time for the 1964 general election in November. If new election districts were created it would have been necessary to have had a general re-registration and it was concluded that the limited time available would not permit this to be accomplished. It is clear, however, that the 1960 Federal census enumeration districts followed the old representative district lines and that therefore no census enumeration district boundary crossed the boundary of any old representative district, i.e., as such existed in 1960. But the census enumeration districts did not respect old election district boundaries and the census enumeration district boundaries did cross old election district lines so that the result of following old election district boundaries in Kent County was to split many census enumeration districts, thereby rendering it impossible to follow the census data in the newly-created representative and senatorial districts. This, of course, made population estimates necessary. We have already set out the method employed by Mr. Hughes in making his estimates.Dr. C. Harold Brown, called by the plaintiffs as an expert, testified that generally voter registration ratios are inaccurate because the level of voter registration will vary from area to area depending on the social, economic and ethnic complexions of the populations. Pinpointing the precise problem in Kent County, he testified that the county contained the Dover Air Force Base and Delaware State College. It cannot be doubted that persons living at the Dover Air Force Base and at the College have residencies of a more transient character than those of persons regularly living in those areas and that the former do not tend to register or vote from what may be called their "temporary" homes as frequently as members of the latter group. Dr. Brown suggested means of making estimates, other than physically counting population, which could yield more accurate results than Mr. Hughes' estimates, such as analyses of utility company data, of postal information, and actual house counts in sparsely settled areas. These methods, of course, would have been time consuming.
Dr. Brown is a sociologist of the Division of Urban Affairs of the University of Delaware. He is an associate professor of sociology at the University. He testified that he was a research analyst for the "Farm Population Division" of the Farm Bureau of the United States Department of Agriculture and that he had done population analyses while at the Agricultural and Mechanical College of Texas, College Station, Brazos County, Texas for two years, and that while at the University of Delaware he has done research for the Greater Wilmington Development Council and that his work involved population analyses of Wilmington and New Castle County.
Mr. Hughes stated that he was aware of the fact that voter registration is likely to be lower in an area in which military personnel lives; nonetheless, he made only one adjustment on account of this factor and he accomplished this by a guess. Mr. Hughes split the old 2d representative district between the new 26th and the new 27th representative districts. In estimating the number of persons in the old 2d election district of the old 2d representative district now contained in the new 27th representative district, the voter registration ratio employed by Mr. Hughes yielded a result of 4701 persons. Mr. Hughes, however, in the exercise of his judgment, struck off 701 persons.
The new 26th representative district contains the Dover Air Force Base and the new 26th and new 27th representative districts contain the areas in and around the base. It is in these areas that the plaintiffs challenge most seriously the population estimates made by Mr. Hughes.
It will be noted that Column 5 of Table VI shows a minimum population for the new 26th representative district of 9,779 persons, while Court's Ex. 41 shows a minimum figure of 15,986 persons for the same representative district. The difference is asserted to be attributable to the testimony of Mr. Weaver of the Committee of 39 that upon rechecking S.B. 336 and the apportionment thereunder, he found that more census enumeration districts were split in this area than he had calculated at first. He testified that the range (maximum — minimum) for the 27th representative district remained the same as indicated on Court's Ex. 41. But insofar as we can ascertain from the record this statement does not clarify the situation for the population of the 27th representative district remained at 10,006 minimum, and if Mr. Weaver did take splits from enumeration districts out of the 26th representative district, they must of necessity have been put by him into the 27th representative district. Therefore, the indeterminate maximum would have been increased while the indeterminate minimum would remain the same. If, however, we turn to the stipulation, "S-1 (con't.)" Table XII, supra, we see that the defendants' estimate for the 26th representative district is 17,806 persons, the census figure, less 4,000 persons, Mr. Hughes' estimate, or a net of 13,806 persons, while the plaintiffs' estimate is arrived at by taking the census figure of 17,806 persons and subtracting therefrom 2,300 persons according to the testimony of Mr. Niederhauser, one of the plaintiffs, which leaves a net figure of 15,506 persons.
Weaver, 766-67.
Hughes, 109-13.
Niederhauser, 1702.
In respect to the 27th representative district, the stipulation "S-1 (con't.)" the defendants' estimate starts with the census figure of 9,125 persons plus 4,000 persons, deducted from the 26th representative district as appears above, or a net of 13,125 persons. The plaintiffs' estimate begins with the census figure of 9,125 persons plus 2,300 persons, taken from the 26th representative district as set out above, leaving a net figure of 11,425 persons.
The plaintiffs' figure is again based on the testimony of Niederhauser, see n. 60, supra.
In respect to the 14th senatorial district of Kent County, the defendants' estimate starts with 17,005 persons, the census figure, and adds thereto 4,000 persons for a total of 21,005 persons. The plaintiffs' estimate starts with 17,005 persons, the census figure, adds 2,300 persons for a total of 19,305 persons.
Hughes, 1021-22.
Again relying on Mr. Niederhauser's testimony, see n. 60, supra.
In respect to the 15th senatorial district, the defendants' estimate starts with the census figure of 10,097 persons and adds 13,806 persons for a total of 23,903 persons. The plaintiffs' estimate starts with the census figure of 10,097 persons and adds 15,506 persons for a total of 25,503 persons.
Again relying on Mr. Hughes' testimony, see n. 62, supra.
Again relying on Mr. Niederhauser's testimony, see n. 60, supra.
We have no substantial evidence before us upon which we as the triers of fact can rely to resolve these figures. There was testimony by Mr. Niederhauser that Dr. C. Harold Brown had made a house count in the old 2d election district of the old 2d representative district but this "house count", if it was made, was not offered in evidence and, of course, what Mr. Niederhauser might have been able to say about its contents would have been inadmissible as hearsay evidence. No direct testimony was offered as to the number of persons, civilian and military, employed at the Dover Air Base or where they lived. Nor do we have any direct evidence as to the number of persons living at or near Delaware State College either as students or as teachers. Though the shortness of time before the ensuing primary election and the general election required haste in the preparation of the plan of apportionment embodied in S.B. 332 and S.B. 336, we cannot help but conclude that a comparatively small effort might have brought more realistic figures upon the record in regard to the actual number of persons employed in or about the Dover Air Base and the number of persons living at or near Delaware State College as students and teachers.
Niederhauser, 1702.
In addition to the foregoing, Mr. Hughes testified that since 1960 the largest growth area in Kent County is that comprised of the old 5th representative district. We take judicial notice of the fact that this statement is correct, and this growth tends to counteract the miscalculation in the old 2d election district of the 2d representative district, tending to bring the population of the new 27th representative district into closer parity with the new 26th representative district.
Hughes, 1059.
The new 28th representative district population figure contains an error of approximately 461 persons. This was caused by the fact that Mr. Hughes thought he was splitting a census enumeration district when in fact he was not. Therefore, the stipulation, Document 174, aside, the corrected population for the 28th representative district would be 14,209. The 29th representative district should be decreased by 461 persons to take into account the correction just stated. In addition, there appears to be a minor error of 10 persons in the 29th representative district and the 13th senatorial district.
See column 5 of Tables VI and VII.
IX. Sussex County Population Figures and Estimates
The plaintiffs charge that many of the estimates made by Mr. West in the apportionment of Sussex County are incorrect for the same reasons which caused inaccuracies in the apportionment figures for Kent County. We point out that in Sussex County as in Kent election district lines were respected but that this required the splitting of many census enumeration districts. It follows that numerous population estimates were necessary and these were made by Mr. West by using voter registration ratios in the manner which we have explained at an earlier point in this opinion. Nothing in Sussex County presented such problems as those caused by the Dover Air Force Base and the Delaware State College in Kent County. It is also a fact that the voter registration ratio figures as presented by Mr. Weaver of the Committee of 39 do not indicate as wide a disparity in the defendants' population figures as is presented by the Kent County population figures. While the plaintiffs' evidence does cast some doubt on the reliability of the estimated population figures as computed by the defendants, we find these doubts to be of less serious nature than those we entertain in respect to Kent County, but we point out that the population could have been computed more accurately by the means suggested by the plaintiffs' witness, Dr. C. Harold Brown, though this would have taken more time than the method employed by Mr. West.X. Analysis of Reapportionment
A. Comparison of Population Figures by Representative and Senatorial Districts with the Mean or Norm
Focusing on deviations from the mean or norm in general Table VI, supra, shows that there are five representative districts, the 8th, 12th, 28th, 29th and 34th, that deviate from the norm by more than 10%. Table VII, supra, shows there are three senatorial districts, the 6th, the 13th and the 14th, that deviate from the norm or mean by more than 10%. The figures shown by Table VII insofar as they relate to the 6th and 13th senatorial districts are in accordance with the stipulation, " Exhibit 4 (Exhibit S-2)". In respect to the 14th senatorial district in Kent County, the stipulation gives as the defendants' population figure 21,005 persons and the plaintiffs' figure is shown as 19,325 persons, but whether the defendants' figure or the plaintiffs' figure be taken, the percentage deviation is in excess of 10%. If the plaintiffs' figure be taken the deviation is 22.4% and if the defendants' figure be taken the deviation is 15.28%.
Aside from the figures of the exhibits and Tables VI and VII showing the larger deviations, it will be noted that in respect to both representative and senatorial districts, whether the plaintiffs' figures or the defendants' figures be employed, the mean or norm is rarely satisfactorily met. In the case of representative districts, there are only 10 districts out of the 35 where the populations created by S.B. 332 and S.B. 336 substantially meet the mean or norm of 12,751 persons, and the deviations, particularly in view of the comparatively small populations involved in the respective representative districts, a fact which renders the creation of geographic divisions easy, cannot be deemed to be entirely insignificant. In respect to the senatorial districts, leaving out the disputed 14th and 15th districts, it would appear that there are only 2 out of the 16 districts in which the mean or norm of 24,794 persons is substantially met.
B. Deviations in Representative and Senatorial Districts within the City of Wilmington
Again referring to the stipulations designated under subheading "A.", supra, and Tables VI and VII, we find deviations within political subdivisions. We shall deal first with Wilmington.
We find from Table VI that the smallest representative district, the 8th, has an estimated population of 11,114, compared to the largest representative district, the 5th, which has an estimated population of 12,917. The 8th representative district has a deviation from the mean or norm of -12.84%, and the 5th representative district has a deviation of +1.30%. While the two districts designated are not contiguous, they are separated only by the 6th representative district. The difference in populations between the 5th and 8th representative districts is 1803 persons presenting a ratio of 1.16-1. It will also be noted that the 1st, 2d 3rd, 4th, and 7th representative districts of Wilmington have substantial representational advantages over the larger districts, viz., the 5th and 6th.
Coming now to the deviations in senatorial districts within Wilmington we find that the deviation between the 3rd senatorial district, 24,630 persons, the largest, and the 1st senatorial district, 23,563, the smallest, is 1067 persons. The difference in populations between the 3rd senatorial district and the 1st senatorial district has a ratio of 1.05-1. Table VIII, supra, gives the deviations from the mean or norm and also gives the population figures for the senatorial districts for the City of Wilmington. The differences in population between the 3rd senatorial district and the 2d and 4th senatorial districts is about the same as to the number of persons involved.
Table VIII gives the population figures within the City of Wilmington and the deviations in number and percentage from the norm or mean within the City of Wilmington.
Mr. Latchum chose the course of first creating the senatorial districts for the City of Wilmington, and then proceeded to create the representative districts of Wilmington. There are four of the former and eight of the latter. Mr. Latchum did not compose the senatorial districts by combining representative districts as did Mr. McGinnes and Mr. West respectively for New Castle County outside of Wilmington and for Sussex County. No logical reason appears as to why Mr. Latchum did not first compose the representative districts for Wilmington and then follow what would appear to be the natural method of creating Wilmington senatorial districts by combinations of representative districts. We shall deal with this subject again on the issue of gerrymandering.
Mr. Hughes could not compose senatorial districts in Kent County from combinations of representative districts in Kent County because Kent County has five representative districts and only three senatorial districts.
See stipulation exhibit 11, map of Wilmington, showing the representative districts of Wilmington pursuant to S.B. 336.
We also feel constrained to point out that there are block figures available under the 1960 census, with very few exceptions, for the whole of Wilmington. We can perceive no valid reason why Mr. Latchum could not have made use of these block figures to create representative and senatorial districts as nearly equal in population as those set out in the plaintiffs' plan for the City of Wilmington, Table XIV, and its accompanying supporting data as contained in Table XV. It should be borne in mind that Mr. Latchum was in substance the progenitor of the new apportionment plan for the City of Wilmington.
C. Deviations within the Representative and Senatorial Districts of Rural New Castle County
Making use of Tables VI and IX and " Exhibit 3 (Exhibit S-1)" of Document 174, we find that the stipulation shows the smallest representative district in New Castle County, outside of Wilmington, is the 20th, with a population of 12,741. Table VI, Column 2, shows the same figure, while Column 6 of Table VI and Table IX show a population figure of 12,191. The stipulation shows that the largest representative district in New Castle County, outside of Wilmington, is the 10th, with a population of 13,928. Table VI, Column 2, shows the same figure as the stipulation but the corrected figure in Column 6 of Table VI and Table IX show a population figure of 14,046 for the 12th representative district. If the figure last referred to of Tables VI and IX be correct the 12th representative district is of course the largest population-wise in Rural New Castle County.
The figure last quoted is arrived at by correcting an error made by Mr. McGinnes concerning the 167th census enumeration district. Due to an incorrect map he inadvertently split the 167th census enumeration district between the 20th and 21st representative districts. See McGinnes, 247-49.
The figure last quoted is arrived at by correcting an error made by Mr. McGinnes concerning the 31st census enumeration district. Mr. McGinnes' map showed that the 31st census enumeration district was split between the 11th and 12th representative districts. Actually there was no split. See 234-35.
Taking the figure from " Exhibit 3 (Exhibit S-1)" and treating the 10th representative district as having the largest population in Rural New Castle County, we find the difference between the largest and the smallest, the 20th representative district, to be 1187 persons, presenting a ratio of 1.09-1. If we take the mean or norm for the representative districts of Rural New Castle County as a whole, aside from the state-wide norm or mean of 12,751, which is 13,226, we find that of the sixteen representative districts there are only six which approximate the mean or norm, of Rural New Castle County. The deviations are approximately the same as the deviation between the populations of the 10th and the 20th representative districts.
Table IX.
Turning now to the senatorial districts in Rural New Castle County and comparing them with one another, employing the Stipulation " Exhibit 4 (Exhibit S-2)" and Tables VII and IX, we find that the largest is the 6th senatorial district with a population of 27,715. In respect to the 6th senatorial district the tables and the stipulation display no difference. According to the stipulation the smallest senatorial district in Rural New Castle County is the 8th, with a population of 25,686. According to Column 6 of Table VII, the smallest senatorial district is the 10th, with a population of 25,414. The same figure is shown in Table IX. We will employ the stipulated figures. According to these figures, the difference in population between the 6th senatorial district with a population of 27,715 and the 8th with a population of 25,686, is 2,029 persons. The ratio between the the two districts is 1.079-1. We point out that the norm or mean population within Rural New Castle County, according to Table IX, is 26,452, compared with an average or norm mean for senatorial districts throughout the State of 24,794 persons. Only 4 of the 8 senatorial districts in Rural New Castle County approximate the norm or mean for Rural New Castle County.
D. Comparison of Deviations of Populations of General Assembly Districts of Wilmington with Those of Rural New Castle County
Table VI shows the state-wide norm or mean per representative district to be 12,751 persons. The stipulation, " Exhibit 3 (Exhibit S-1)", shows that the smallest representative district in the City of Wilmington is the 8th with a population of 11,114 persons, and that the largest representative district in New Castle County outside of Wilmington is the 10th with a population of 13,928 persons. The 8th representative district, therefore, contains 1637 persons less than the norm or mean and the 10th representative district contains 1177 persons more than the state-wide norm or mean. The percentage of deviation of the 8th representative district from the norm or mean is -12. 84%, and the percentage of deviation of the 10th representative district from the norm or mean is +9.23%. See Table VI, Column 4. If we compare the 8th representative district with the 10th representative district, we find the 10th representative district contains 2,814 more persons than the 8th representative district. The ratio between the 8th representative district and the 10th is 1.25-1.
If we compare the populations of the senatorial districts in Rural New Castle County with the populations of the senatorial districts in Wilmington, we find that according to Tables VII and IX and the stipulation referred to, the smallest senatorial district is the 1st with a population of 23,563 persons, and the largest senatorial district is the 6th with a population of 27,715 persons. This is a difference of 4152 persons. The ratio between the 1st senatorial district, the smallest, and the 6th senatorial district, the largest, is 1.17-1.
E. Deviations in Representative and Senatorial Districts within Kent County
Turning now to Table X and again to Document 174, Stipulation, " Exhibit 3 (Exhibit S-1)", we find that the smallest representative district in Kent County is the 29th, containing 11,350 persons, and that the largest is the 28th, which contains 14,209 persons. These two representative districts are contiguous. The difference between these two figures is 2859 persons. The ratio between the two representative districts is 1.25-1.
It will be observed that the Stipulation, " Exhibit 3 (Exhibit S-1)" gives different totals for the 28th and 29th representative districts, the "Defendants" column stating that the populations, respectively, are 13,748 and 11,811, and the "Plaintiffs" column stating, that the populations, respectively, are 14,209 and 11,350. It is stipulated, Document 174, that the correct figures are 14,209 and 11,350.
It should be noted that the norm or mean figure for the populations of the representative districts in Kent County is 13,130 persons as distinguished from the norm or mean for the State of 12,751 persons. As to the Kent County mean, it will be observed that only two representative districts in Kent County, the 25th and the 27th representative districts, approximate the mean or norm for Kent County, and that none of the Kent County representative districts approximates the mean or norm of the State of 12,751 persons.
As we have stated in the text of this opinion, two representative districts approximate the mean or norm for Kent County. The 25th representative district contains 13,161 persons pursuant to the census figures and the 27th representative district, assuming the defendants estimate to be correct, contains 13,125 persons. See Document 174, Exhibit 3 (Exhibit S-1).
As to the senatorial districts in Kent County it is stipulated that the smallest senatorial district is the 13th containing 20,743 persons. This figure corresponds with that set out on Table X and is but 10 more than the figure appearing in Column 2 of Table VII. The largest senatorial district in Kent County, whether we employ the defendants' figures or the plaintiffs' figures is senatorial district 15. See the stipulation " Exhibit 3 (Exhibit S-1)". The stipulation shows as the defendants' population figure 23,903 and the plaintiffs' population figure as 25,583. Table X shows a population figure of 23,903 and Column 2 of Table VII shows the identical figure. We have made use of the defendants' population figures in the past. We shall do so in this instance. The difference between the smallest senatorial district in Kent County, the 13th, and the largest, the 15th, using the defendants' figures of 23,903 and 20,743, therefore is 3,160 persons. The ratio between the 13th senatorial district and the 15th senatorial district is 1.15-1.
Column 2 of Table VII shows 10 more persons than does Table IX and the stipulation referred to, but this is a simple mathematical error. We believe the smaller figure is correct.
The norm or mean for the population for senatorial districts within Kent County is 21,884 persons. None of the senatorial districts approximates this norm. The norm or mean for state-wide senatorial districts is 24,794 persons, and none of the Kent County senatorial districts approximates this norm.
F. Deviations in Representative and Senatorial Districts in Sussex County
Again making use of Document 174, " Exhibit 3 (Exhibit S-1)", and Table XI, we find that the smallest representative district in Sussex County is the 34th containing 11,017 persons and that the figures in Table XI coincide with the stipulation. The largest is shown to be the 32d containing 13,458 persons and again the figures of the defendants and the plaintiffs in the stipulation coincide and no differences are shown by Table XI. The difference between the largest and the smallest representative district in Sussex County, therefore, is 2441 persons. The ratio between the two representative districts, the smallest, the 34th containing 11,017 and the largest, the 32d containing 13,458 is 1.22-1. The norm or mean for Sussex County per representative district is 12,199 and the state-wide norm is 12,751.
As to the senatorial districts in Sussex County, it is stipulated, Document 174, " Exhibit 4 (Exhibit S-2)", that the smallest senatorial district is the 18th containing 22,686 persons. There is no disparity in the stipulation between the defendants' figures and the plaintiffs' figures, nor does any discrepancy in respect thereto appear in Table XI. The largest senatorial district as appears from the stipulation and from the table is the 16th, containing 25,517 persons. The difference in population between the smallest and the largest districts is 2831 persons. The ratio, therefore, is 1.12-1.
The norm or mean for the senatorial districts within Sussex County is 24,398 persons, and only the 17th senatorial district approximates this norm. The norm or mean for senatorial districts throughout the state is 24,794 persons, and again only the 17th senatorial district approaches this norm. It should be noted that the three senatorial districts of Sussex County are contiguous.
G. A View of Representative and Senatorial Districts of the State of Delaware as a Whole
The largest representative district in the State is the 28th representative district in Kent County embracing 14,209 persons. This population figure is stipulated to. " Exhibit 3 (Exhibit S-1)." See also Table VI. The smallest is the 34th representative district in Sussex County which contains 11,017 persons. This also is stipulated to and also appears on Table VI. The difference in population is 3,192 persons. The ratio between the largest and the smallest representative districts is 1.289-1. But an examination of the stipulation and the map exhibits and the Tables will demonstrate that many other representative districts throughout the State are almost equally out of line.
The largest senatorial district in the State is the 6th in Rural New Castle County which has a population of 27,715 persons. This figure is stipulated to, " Exhibit 4 (Exhibit S-2)". See also Table VII. The smallest senatorial district is the 13th in Kent County, having a population of 20,743 persons. This also is covered by the stipulation and by Table VII. The ratio between the largest and the smallest senatorial districts, therefore is 1.33-1. The difference in populations is 6,972 persons. While the difference of 6,972 persons is numerically the largest difference created in the State whether between representative districts or senatorial districts, and the ratio of 1.33-1 is the highest ratio, nonetheless an examination of the stipulation and the exhibits will show a substantial number of senatorial districts almost equally lacking in parity and giving ratios close to that last stated.
There are other factors to be considered. One of these is brought into focus by Table I which shows, as we have said, geographic voting areas and their respective populations. If we accept the deviations from the norm or mean and the percentages of such deviations as shown by Columns 6 and 7 of Table I, and view the four geographic subdivisions of the State as shown in Column 1 against the background of the State as a whole, we find that even if Wilmington boundaries and the dividing lines between the Counties be respected, the deviations in actual numbers of persons and percentages are still comparatively large. On the other hand, if we examine "New Castle (Combined)", i.e., New Castle County including Wilmington, we find that the deviation from the mean by way of number of persons to be only 59 more than the norm or mean of 12,751 persons and percentage-wise this amounts to only +.5%. If an election at large were ordered for New Castle County including Wilmington, it would result in almost perfect representation of these political units in respect to each other.
Looking now to the senatorial districts shown by Table II, we find more substantial deviations. Kent County shows a deviation of -2910 persons and a percentage deviation of -11.74%. Sussex County shows a deviation of but -396 persons and a percentage deviation of only -1.6%. On the other hand Wilmington shows a deviation of -837 persons and a percentage deviation of -3.38%, whereas Rural New Castle County shows a deviation of +1658 persons and a percentage deviation of +6.69%.
Column 8 gives the total number of representatives and senators in the General Assembly by political subdivision, combining the seats of representatives and senators. In Table III, we find the smallest deviation is in Kent County with 15.1% of seat-power in the General Assembly and a percentage deviation of only +.4%. The largest deviation is in Rural New Castle County which is shown as having 45.3% of seat-power in the General Assembly with a deviation of only -2.1%.
The defendants have laid emphasis on the testimony of Dr. Malcolm E. Jewell. Dr. Jewell testified that under S.B. 332 and S.B. 336, and employing 1960 census figures, 46.9% of the population of Delaware could elect 9 senators or one-half the Senate, and that 52.4% of the population could elect a majority of the Senate or 10 senators. Dr. Jewell also testified that 48.9% of the population could elect 18 representatives or a majority of the House of Representatives. He testified he arrived at these conclusions in respect to the House of Representatives, conclusions which we find to be correct, by adding together the 18 representative districts "smallest" in population, and it would appear that he arrived at his conclusions respecting the Senate by adding together the 9 senatorial districts smallest in population. Dr. Jewell also testified that no other State, in his judgment, was as mathematically correctly apportioned as is Delaware.
Dr. Jewell is an associate professor of Political Science at the University of Kentucky. He has a Bachelor of Arts degree from Harvard University, a Master of Arts degree from Columbia University, and a degree of Doctor of Philosophy from Penn State University. He majored in political science. He testified he has taught political science since 1958 at the University of Kentucky and served one year as a visiting professor at Duke University. He is a member of a number of learned societies engaged with political science and has edited a book titled "The Politics of Reapportionment," Atherton Press, 1962, published shortly after the decision in Baker v. Carr, 369 U.S. 186 (1962). He has published a number of pamphlets and articles respecting reapportionment in the National Civic Review.
Representative District Population County
34 11,017 Sussex 8 11,114 Wilmington 33 11,534 Sussex 35 11,669 Sussex 1 11,710 Wilmington 29 11,811 Kent 4 11,834 Wilmington 2 11,853 Wilmington 3 11,890 Wilmington 7 11,917 Wilmington 6 12,592 Wilmington 31 12,719 Sussex 20 12,741 Rural N.C. 14 12,776 Rural N.C. 30 12,798 Sussex 15 12,826 Rural N.C. 16 12,860 Rural N.C. 18 12,861 Rural N.C. ------- 218,522
Senatorial District Population County
13 20,743 Kent 14 21,005 Kent 18 22,686 Sussex 15 23,903 Kent 1 23,563 Wilmington 2 23,724 Wilmington 4 23,910 Wilmington 3 24,630 Wilmington 17 24,992 Sussex -------- 209,156
It should be noted, however, as appears from the representative and senatorial districts designated in notes 80 and 81 supra as the smallest in population, that these are widely distributed throughout the State and that the chance of these districts voting Republican or Democratic in unison, so to say, as is required if Dr. Jewell's assumption as to percentages is to be effective is an unlikely one unless as was the case in 1964 there is a general election involving a presidential candidate of great popularity.
See Table XIII.
We come now to a vexing issue. As shown by Table I, referring to the City of Wilmington vis-a-vis Rural New Castle County, i.e., New Castle County outside of Wilmington, and the number of representatives allocated to each geographical unit by S.B. 332 and S.B. 336, we find that Wilmington is given 8 representatives and Rural New Castle County is allotted 16 representatives. The same table shows that Wilmington has a population of 95,827 persons according to the 1960 census and that Rural New Castle has a population according to the 1960 census of 211,619 persons. As we have pointed out earlier, Column 5 of Table I shows that the actual number of persons per representative allotted to Wilmington was one representative per 11,978 inhabitants, and the same column shows that in Rural New Castle one representative is allotted per 13,226 inhabitants. As shown by Column 7 of Table I, under S.B. 332 and S.B. 336, Wilmington is the most over-represented political subdivision in the State and Rural New Castle County is the most under-represented political subdivision in the State. If Wilmington were given 7 representatives and Rural New Castle County 17 representatives, the deviations from the mean or norm in populations and the percentage of deviations would be not very much larger, as shown by Columns 8, 9, and 10 of Table I, than if Wilmington were to retain its 8 representatives and Rural New Castle County its 16. But it will be noted that if Wilmington be given 7 representatives and Rural New Castle County 17 representatives, the deviation from the state-wide mean or norm of 12,751 rises from -773 persons to +939 persons and the percentage of deviation from the mean or norm rises from -6.06% to +7.36% for Wilmington. For Rural New Castle County the deviation from the state-wide mean or norm population-wise falls from +475 persons to -303 persons and the percentage of deviation from the mean falls from +3.73% to -2.38%. We reiterate that Wilmington is the most overrepresented political unit in the State and this statement is made by us without regard to the 38,000 persons referred to in the McGinnes affidavit as augmenting the population of Rural New Castle County, which is dealt with immediately hereinafter.
This is another important factor which must be considered and carefully weighed. Mr. McGinnes stated in his affidavit, hereinbefore quoted, that the population in Rural New Castle County had increased by approximately 38,000 persons between 1960 and the date of the execution of his affidavit, July 13, 1964. We find Mr. McGinnes' statement to be correct and therefore we find that the increase in population between 1960 and 1964 in Rural New Castle County was approximately 38,000 persons.
Defendants' Ex. 4.
Using the census figures alone, as employed by the defendants' witness, Dr. Jewell, Wilmington with 8 representatives should have but 7.4 representatives and Rural New Castle which has but 16 representatives should have 16.7 representatives., We do not suggest that representatives can or should be physically divided, and the differences may seem to be small indeed but this is not so in reality even if no effect be given to the increase in population in New Castle County as set out in the McGinnes affidavit for no proper or logical reason appears as to why Wilmington should be favored over Rural New Castle County and Rural New Castle County denigrated for the excess of population was and is substantially in favor of Rural New Castle County. As shown by Table I, census-wise there were 13,326 persons per representative in Rural New Castle County and but 11,978 persons per representative in Wilmington. Furthermore, employing the figures of Column 3 of Table II, we find that Wilmington was entitled to 3.86 senators and Rural New Castle to only 8.5 senators.
See 304 et seq., and note 80, supra.
The figures from Table I, Column 3, indicate that Wilmington should have 7.51 representatives and Rural New Castle County should have 16.59 representatives.
If we go further and take into consideration the increase of approximately 38,000 persons over the previous total of 211,619 persons in Rural New Castle County, as is demonstrated by the McGinnes affidavit, we find it to be even clearer that the benefit of any differential should have been given to Rural New Castle County and not to Wilmington for 38,000 persons is about 18% of 211,619 persons and about 40% of 95,827 persons, the population of Wilmington as shown by the 1960 census. If these differentials be employed, assuming no loss of population in Wilmington, Rural New Castle County would be entitled to 17.3 representatives and Wilmington to 6.6 representatives. It should be borne in mind, moreover, that Wilmington has lost population in every census from 1930 to 1960, inclusive, as demonstrated by Table V.
Though we are unable to state how many persons, if any, have quit Wilmington since the 1960 census, since 24,794 persons is the mean or norm or perfect population for a senatorial district, we can and do say that Wilmington is entitled to less than 4 senators, viz., 3.3 senators and Rural New Castle County to 8.6 senators. We believe that our figures are accurate, but even if they are not entirely so, the larger disparity in representation between Wilmington and Rural New Castle County in respect to senatorial representation is obvious. We will make further findings of fact, in respect to the increase of population of 38,000 persons in Rural New Castle County as seems appropriate at later points in this opinion, but there seems to be no logical explanation insofar as arithmetical considerations or other permissible factors are concerned which justifies the disparities dealt with in the immediately preceding paragraphs. This is peculiarly the case in view of the admonition of Section 607 of S.B. 332 which expressly provides that the boundaries of representative and senatorial districts shall be determined by the General Assembly in such a manner that there shall be substantial equality of population among the several representative and senatorial districts. The plaintiffs assert that the disparities are due to conscious gerrymandering.
XI. Analysis of Gerrymandering
A. As to the City of WilmingtonIn preparing the reapportionment of Wilmington Mr. Latchum first created four senatorial districts and next created eight representative districts. He did not compose the senatorial districts of Wilmington by combinations of representative districts, new or old. Mr. McGinnes, it will be recalled, first composed sixteen representative districts for Rural New Castle County and then combined them to form eight senatorial districts. Mr. West followed the same plan for Sussex County. Mr. Hughes could not follow this plan since he had five representative districts and three senatorial districts to compose for Kent County. Mr. Hughes, however, did create the representative districts of Kent County first. The course pursued by Mr. Latchum in first creating senatorial districts for the City of Wilmington rather than first composing representative districts, in our opinion was somewhat unusual, and we can perceive no sound logical basis for the course which he pursued. We find that he could have elected the course of first creating eight representative districts for Wilmington and then have composed senatorial districts from them. If he had done so, certain, at least, of the odd shapes and sizes of some representative and senatorial districts in Wilmington could have been avoided and some of the charges of gerrymandering leveled by the plaintiffs, at least insofar as General Assembly Districts in Wilmington are concerned, could have been obviated.
See Stipulation Ex. 11, map of Wilmington showing the representative districts pursuant to S.B. 336.
For example, the 8th representative district as created by Mr. Latchum has a population of 11,114 persons according to the 1960 census and lies on the western side of Wilmington. Its boundaries begin at the northwest corner of the Wilmington City line and proceed eastward along the Brandywine River to Riddle Avenue, thence approximately southeast to Kentmere Parkway, thence west to Union Street, thence west by southwest to Sixth Street, thence east to DuPont Street, thence south along DuPont Street to Lancaster Avenue, thence west on Lancaster Avenue to Union Street, thence south on Union Street curving slightly to the southwest to the Wilmington City line on the south, thence along the Wilmington City line various courses north, west and east to the place of beginning.
The 8th representative district as composed by Mr. Latchum, as will be observed from the map, Stipulation Ex. 11, runs longitudinally down the western portion of Wilmington from north to south and in effect splits two pockets of Republican strength in the 7th Ward of Wilmington, two pockets which had for some time elected Republican representatives to the General Assembly. Under the Latchum apportionment the pocket of Republican strength to the west was retained in the new 8th representative district, and the pocket of Republican strength to the east was put into the new 6th representative district.
Niederhauser, 1600 et seq. and 1649 et seq.; Harrison, 2032. See Stipulation Ex. 11 and Plaintiffs' Ex. 46, and its overlay, Plaintiffs' Ex. 50. See also Court's Ex. 17 and its overlay, Court's Ex. 17a, and Court's Ex. 11(c) admitted in evidence 5/16/63, in the proceeding sub nom. Sincock v. Duffy, 215 F. Supp. 169. See also "1956-1962 Voting, Records Arranged by Legislative Districts (State Representative and State Senator)", Plaintiffs' Ex. 60.
The same result was achieved in respect to the 4th senatorial district by S.B. 336, a district which, it will be noted, had a population of 23,910 persons as shown by the 1960 census. The eastern boundary of the 4th senatorial district runs north and south along Broom Street which lies six blocks west of Union Street, the western boundary of the 8th representative district. The 4th senatorial district is the second largest in Wilmington.
This senatorial district as composed by Mr. Latchum consists of a large part of the 7th Ward, a slice of the 5th Ward, a substantial part of the 12th Ward and a portion of the 11th Ward. Two pockets of Republican strength which, for a considerable period of time, had elected a Republican senator to the General Assembly were split and combined with areas of greater Democratic strength, one pocket, that to the west, being included in the 4th senatorial district, and the other, that to the east, being put in large part in the 3rd senatorial district.
See Niederhauser, 1600 et seq., and 1649 et seq. See Stipulation Ex. 12 and Plaintiffs' Ex. 46, and its overlay, Plaintiffs' Ex. 51. See also Court's Ex. 18 and its overlay, Court's Ex. 17a. See also Court's Ex. 11(c), admitted in evidence 5/16/63 in the proceedings reported sub nom. Sincock v. Duffy, 215 F. Supp. 169. See also "1956-1962 Voting, Records Arranged by Legislative Districts (State Representative and State Senator)", Plaintiffs' Ex. 60.
The 1st senatorial district, as composed by Mr. Latchum, starts from a point at Wilmington's northwest corner and proceeds east along the City line, at Rockwood Road, thence south, west, east along the City line to the Governor Printz Boulevard, thence southwest along the Governor Printz Boulevard to the Brandywine River, thence along the middle line of the Brandywine River to the point where the latter intersects the Wilmington City line, and thence northwest along the Wilmington City line to the place of beginning. This senatorial district is composed of the 9th Ward with the exception of the 1st, 2d 3rd and 10th election districts.
See the exhibits referred to in note 88, supra.
As has been stated, the 1st senatorial district was the first composed by Mr. Latchum. It lies entirely in the 9th Ward but it is not coextensive therewith, as indicated in the last sentence of the previous paragraph. We perceive no sound basis for any charge of gerrymandering insofar as the composition of the 1st senatorial district is concerned, though our previous statement as to the rather unusual course chosen by Mr. Latchum in creating senatorial districts before representative districts should be borne in mind.
Latchum, 374 et seq.
The 1st and 2d representative districts, however, were formed by Mr. Latchum by dividing the 1st senatorial district by a line running roughly northwest to southeast along Concord Avenue to Baynard Boulevard, thence north on Baynard Boulevard to 26th Street, thence east by south along 26th Street to Northeast Boulevard, Governor Printz Boulevard. In respect to both the new 1st and the new 2d representative districts, containing respectively populations of 11,710 and 11,853 persons as shown by the 1960 census, it is apparent that the traditional Republican strength in the eastern part of the 9th Ward was divided between these two new districts. Each portion of the Republican strength was combined with areas of Democratic strength in the western parts of the two representative districts. Thus, a substantial single pocket of Republican strength was divided into two smaller pockets.
Niederhauser, 1654. See all of note 87, supra, save the first sentence.
Mr. Latchum composed the 2d senatorial district in an odd manner. Beginning at the northeast corner of the new 1st senatorial district he drew a line southwest to the Brandywine River, thence west along the middle line of the Brandywine River to Adams Street, thence southwest along the west line of the Wilmington and Brandywine Cemetery, thence southeast along Delaware Avenue, cutting off a portion of the cemetery, to Jefferson Street, thence southwest along Jefferson Street to 9th Street, thence east along 9th Street to West Street, thence southwest along West Street to 7th Street approximately two blocks, thence east along 7th Street for one block to Tatnall Street, thence southwest along Tatnall Street to 2d Street, thence two blocks west along 2d Street to Washington Street, thence two blocks southwest to Lancaster Avenue, thence approximately two blocks west to Maryland Avenue, thence southwest along Maryland Avenue to the southwest boundary of Wilmington, thence along the Wilmington boundary by various courses to the east, north and west, to the place of beginning, as appears from Stipulation Ex. 12. The result is a peculiar many-sided figure embracing an area which in size is somewhat larger than the rest of Wilmington. After stating this, however, we point out that at least two-thirds of this large area includes much marshland, the Wilmington Marine Terminal, the Christiana or Christina River, the Delaware shore of the Delaware River and various small streams debouching therein, and part of the Pennsylvania Railroad freight classification yard. The total population of the 2d senatorial district, despite its large area, is 23,724 persons, according to the 1960 census. It should be pointed out also in respect to the 2d senatorial district that Mr. Latchum followed census enumeration tracts precisely. Insofar as the creation of this senatorial district qua senatorial district is concerned, we can perceive no sound basis for a charge of gerrymandering but there can be no doubt that the district is lacking in compactness.
Stipulation Ex. 12.
A different story appears, however, on examination of the 3rd and 4th representative districts which are included in the 2d senatorial district. The 4th representative district is a 28-sided figure which the plaintiffs have called "grotesque" and designated as the "galloping horse". This appears graphically on examination of Stipulation Ex. 11 and the Appendix to this opinion. The district can be described by courses as follows: beginning at the extreme southeast corner at a point where the Wilmington City line impinges on the easterly side of the Delaware River, thence along the City line west by north, following various courses to the west and to the south to Maryland Avenue, then northeast along Maryland Avenue to Front Street, Lancaster Avenue, thence east along Front Street to Washington Street, thence along Washington Street one block north to 2d Street, thence along 2d Street two blocks east to Tatnall Street, thence along Tatnall Street north to 7th Street, thence one block west along 7th Street to West Street, thence along West Street north to 9th Street, thence along 9th Street two blocks west to Jefferson Street, thence along Jefferson Street north to Delaware Avenue, thence along Delaware Avenue northwest to Adams Street, thence northeast along Adams Street to the middle line of the Brandywine River, thence east along the middle line of the Brandywine River to Market Street, thence along Market Street south by southwest to 15th Street, thence along 15th Street east to King Street, thence along King Street south two blocks to 13th Street, thence along 13th Street east two blocks to Walnut Street, thence south along Walnut Street to 4th Street, thence east along 4th Street to Christina Avenue, thence along Christina Avenue southeast to the Christiana River, thence along the middle line of the Christina River northeast by southeast to the easterly side of the Delaware River, thence south along the easterly bank of the Delaware River to the point of beginning. See Appendix.
Stipulation Ex. 11 is indefinite in one immaterial particular. Actually the easterly border of Wilmington extends to the low-water mark of the Delaware River on the New Jersey side. See note 103, infra, which refers to the twelve-mile circle and to the decision in State of New Jersey v. State of Delaware, 291 U.S. 361, 366, 54 S.Ct. 407, 78 L.Ed. 847 (1934).
But in creating the 4th representative district Mr. Latchum used parts, but only parts, of the 1st, 2d 3rd, 4th, 5th, 6th, 7th and 11th Wards and the new representative district contains parts, but only parts, of all of the five previously existing representative districts of Wilmington.
Niederhauser 849 et seq. See also Court's Ex. 11(c), referred to in notes 87 and 88, supra, and Plaintiffs' Ex. 29, admitted in evidence in the proceedings sub nom. Sincock v. Duffy, 215 F. Supp. 169.
The 3rd representative district, of course, lies in large part immediately north of the larger part of the 4th representative district and appears graphically on Stipulation Ex. 11. It can be described by courses as follows: beginning at a point on the easterly bank of the Delaware River at a point where the Christina River debouches into the Delaware River, thence along the middle line of the Christina River to a point where its middle line intersects Christina Avenue, thence cutting across Christina Park to Church Street, thence west by north along 4th Street to Walnut Street, thence north on Walnut Street to 13th Street, thence west by north along 13th Street to King Street, thence north by east along King Street to 15th Street, thence along 15th Street approximately one-half block west to Market Street, which at this point runs approximately due east and west, thence along Market Street to the middle line of the Brandywine River to a point just past Kirkwood Park, thence approximately due east and west along Northeast Boulevard to the Wilmington City boundary at the north, at the point where the legend "Governor Printz Boulevard" appears on Stipulation Ex. 11, and thence south by east along the boundary line of the City of Wilmington to the easterly bank of the Delaware River, thence along the easterly bank of the Delaware River to the place of beginning. The district contains 11,890 persons according to the 1960 census.
The boundary lines of the 3rd representative district, as drawn by Mr. Latchum and as shown on Stipulation Ex. 11, created a district in which 84% of the population are Negroes and the Negro population of this representative district comprises 41% of the total Negro population of Wilmington. The plaintiffs contend that as the lines were drawn one pre-dominantly Negro district was created and the representation of this racial minority has been limited to but one representative out of the eight allotted by S.B. 332 for Wilmington. The Negro population, as the plaintiffs point out, constitutes more than 25% of the total population of Wilmington.
See the testimony of Mr. Clayton S. Harrison, Jr., 2056 et seq. Mr. Harrison is the chairman of the Republican State Committee and has served as chairman of the Republican Wilmington City Committee and as secretary of the Republican New Castle County Committee.
Mr. Harrison testified that Negroes constituted 7% of the population of the 1st representative district; 15% of the 2d representative district; 84% of the 3rd representative district; 46% of the 4th representative district; 36% of the 5th representative district; 6% of the 6th representative district; 5% of the 7th representative district, and 7% of the 8th representative district. 2056 et seq.
But certain additional facts must be considered. Forty-six per cent of the population of the 4th representative district is comprised of Negroes and 36% of the population of the 5th representative district is comprised of Negroes. The representative districts with small proportions of Negro populations are the 1st, 6th, 7th, and 8th. The population of the 2d representative district is 15% Negro. See note 96, supra. We have difficulty in perceiving how the large Negro population percentages in the 3rd and 4th representative districts could be dispersed effectively among those representative districts having smaller percentages of Negro populations unless representative districts were created in narrow slices extending longitudinally from north to south through the City and even if this were done, tremendous problems of equality of representation still would be presented, problems which in our view would be almost insoluble. Moreover, we take judicial notice that in the general election of 1964 the voters of the City of Wilmington elected one Negro senator, a perfect result percentage-wise, and two representatives, an almost perfect result percentage-wise. We are mindful, of course, of the very substantial Democratic majorities of the 1964 election as demonstrated in Wilmington, in Delaware, and in the Nation, that those election results cannot, perhaps, be deemed to be typical but nonetheless the results in Wilmington as shown by the election of a Negro senator and two Negro representatives compel a finding of fact that the plaintiffs have not established a factual basis for racial gerrymander in the apportionment of the 3rd representative district or elsewhere in the City of Wilmington or in the State of Delaware. We will discuss the pertinent authorities on this point briefly at a later point in this opinion.
Turning now to the 3rd senatorial district which, as hereinbefore stated, Mr. Latchum composed before he drafted the boundaries of the representative districts, we find that he ignored the pattern he had pursued when he divided the 1st senatorial district, as a practical matter, into half by a dividing boundary line following 26th Street to form the 1st and 2d representative districts, and also when he divided the 2d senatorial district substantially in half by a dividing boundary line following the Christiana River to form the 3rd and 4th representative districts. Instead Mr. Latchum composed the 5th, 6th, 7th, and 8th representative districts by a complete new set of boundaries running in various courses through the western part of Wilmington. The plaintiffs assert that if Mr. Latchum had pursued the design followed by him in the creation of the 1st, 2d 3rd, and 4th representative districts, the result would have been a Republican representative district in the northern half of the 3rd senatorial district.
The plaintiffs argue that substantial equality in populations would have been achieved had the 3rd senatorial district been cut in half by a boundary line running northeast by southwest along 6th Street to create two representative districts which we will designate for the sake of clarity as "(a)" and "(b)". If Mr. Latchum had followed his original pattern the pocket of Republican strength in the northern half of the 3rd senatorial district would have been preserved in representative district "(a)" for such a representative district would have contained strong Republican election districts had the voting followed the Democratic-Republican party lines of the 1960 and 1962 general elections. We find that those assertions of the plaintiffs relating to Mr. Latchum's change of pattern and the pocket of Republican strength to be based on adequate proof and to be correct. See the chart that is set out in note 99, infra, which is based upon the evidence set out in note 98, supra. But this does not prove gerrymandering for if we take the new 6th representative district and add up the Democratic and Republican votes as cast in the 1960 and 1962 elections shown on the chart set out in note 100, infra, we see that the Democratic majority in 1960 was 390 and in 1962 was 106.
3rd Senatorial District
North of Fifth Street
Census Tracts Blocks Population 11 3028 15 3355 16 4318 21 ----
1 122 13 73 14 208 28 178 29 109 30 97 37 18 22 ____
1 195 15 143 16 130 29 75 30 48 31 35
South of Fifth Street
21 (minus blocks 1, 3441 13, 14 28, 29, 30, 37) (4246-805)
22 (minus blocks 1, 3727 15, 16, 29, 30, 31) (4353-626) ______
Niederhauser, 1656 et seq. Stipulation Exs. 11 and 12, Plaintiffs' Exs. 43, 46, 50 and 51.
The following Chart shows the results of the general elections of 1960 and 1962 in respect to the election districts indicated as shown by Plaintiffs' Exs. 43 and, in particular 46 as applied to hypothetical representative district "a". The red color on Ex. 46 indicates that the election district went Republican in both elections. The blue color indicates that the election district went Democrat in both elections. The districts shown as striped red or striped blue went Republican in one general election and Democratic in the other election. The word "split" on the left hand side of the Chart indicates that the election district was split by the boundary line of a representative district. Republican Republican Democratic Republican Majority Democratic Republican Majority Democratic Democratic Democratic Republican Majority Democratic Republican Majority Republican Democratic Red Majority Majority 1960 1962 Red (13) 14th of 7 152 360 208 132 308 176 (14) 13th of 7 99 312 213 88 251 163 (15) 9th of 7 78 104 26 42 67 25 (16) 11th of 7 111 302 191 88 288 200 Split (17) 12th of 7 130 207 77 92 164 72 (18) 8th of 7 121 156 35 27 57 30 (19) 12th of 5 135 135 ____ 70 86 16 Total _____ _____ 750 682 1960 1962 Blue Split (1) 17th of 7 133 106 27 101 75 26 (2) 10th of 7 113 86 27 60 52 8 (3) 7th of 7 153 133 20 122 77 45 (4) 11th of 5 168 80 88 139 57 82 (5) 10th of 5 189 132 57 135 79 56 Split (6) 13th of 5 91 81 10 65 57 8 Split (7) 6th of 5 228 152 76 170 124 46 (8) 5th of 5 172 98 74 82 54 28 (9) 4th of 5 129 75 54 74 30 44 (10) 3rd of 5 270 103 167 161 62 99 (11) 2nd of 5 183 63 120 105 37 68 Split (12) 1st of 5 197 95 102 141 65 76 ____ _____ Total 822 586 Striped Split 6th of 7 132 165 33 87 85 2 1960 1962 822 682 -783 -588 ___ ___ 39 Democratic majority 94 Republican majority
This Chart demonstrates how the voting would have eventuated, Democratic and Republican, in the 1960 and 1962 elections had the new sixth representative district been then constituted as it is under S.B. 332 and S.B. 336.
These figures, however, are based upon the assumption that all of the voters in the split election districts voted in the new 6th representative districts which, of course, was not the case. If the split election districts are excluded, the Democratic majority in 1960 would have been but 219, and in 1962 would have been exactly zero. And the plaintiffs have not seen fit to furnish us with figures which demonstrate that the 13th, 14th, 15th, 16th, 17th, 18th and 19th election districts of the 7th Ward, shown in red on Plaintiffs' Ex. 46, were more potently arrayed in the apportionment of Wilmington prior to the reapportionment created by S.B. 332 and S.B. 336. Treating the election districts just described, which lie in the eastern part of the old 3rd representative district, apart from that portion of that representative district now constituting the new 8th representative district lying in the western portion of Wilmington, we cannot say that the plaintiffs have proved political gerrymander here as that is defined and described hereinafter.
But in respect to the creation of the new 6th and 8th representative districts from what was substantially the old 3rd representative district, it is clear that Mr. Latchum divided two pockets of Republican strength as appears from Plaintiffs' Ex. 46.
The old 3rd representative district had returned Republican majorities for a representative in both the 1960 and 1962 elections. In the 1960 election there were 4524 Democratic votes and 5001 Republican votes, giving a Republican majority of 477, and in the 1962 election there were 3241 Democratic votes and 3822 Republican votes, giving a Republican majority of 581. See Plaintiffs' Ex. 60. By including these pockets of Republican strength in the new 6th and 8th representative districts respectively, the Republican strength in the old 3rd representative district was divided. Does this constitute a basis from which this court should find political gerrymander as that term is defined hereinafter? What other factors are to be considered? Mr. Latchum formed the 1st, 2d 3rd and 4th representative districts in large part by boundary lines running approximately northwest to southeast, but in forming the four remaining representative districts this pattern was not used and though there are boundary lines running approximately northwest to southeast, boundary lines running approximately northeast to southwest largely created the remaining four representative districts. Mr. Latchum also disregarded the historical boundaries of the Wards, and, as we have said, he adopted the rather unusual course of first creating senatorial districts and then creating representative districts. But there was an advantage gained by the Democratic Party by Mr. Latchum's reapportionment if we test it by the 1960 and 1962 election returns. In 1960 and 1962 the Republican Party elected one representative out of five available seats and in 1962 elected one senator, there being only one seat available in the latter election year. The Republican percentage in Wilmington in the 1962 election was 39.6% and in 1964 was 38.9%. See Document 195, "S 4 A" and "S 4 B". But in 1964 the Republican Party failed to elect even one representative when eight seats were available and failed to elect even one senator when four seats were available. In short the Republican Party in the 1964 election, with approximately the same percentage of votes as in the 1962 election, failed to elect either a representative or a senator although the number of representative and senatorial seats for the City of Wilmington had been increased by three and two, respectively. Should the court draw the inference from these facts that the Latchum reapportionment constituted a political gerrymander of the City of Wilmington?
It is also true, as the plaintiffs assert, that Mr. Latchum, in creating both the senatorial and representative districts of the City of Wilmington, largely disregarded the traditional boundary lines of the Wards and old representative and senatorial districts.
In respect to compactness and contiguity of the representative districts of Wilmington, upon viewing all of the evidence relating to this aspect of the case, and in particular the location and shape of the head of the "galloping horse", can it be said that the tests of compactness and contiguity have been met, or did the grotesque figure result from political gerrymander as defined hereinafter? Upon consideration of the representative districts of Wilmington and their relation to each other as prepared by Mr. Latchum and as set out in S.B. 336, we are forced to the conclusion that the 4th representative district, the "galloping horse", does not meet the test of compactness and is a result of political gerrymandering within the City of Wilmington.
B. As to Rural New Castle County
The plaintiffs assert that Mr. McGinnes' apportionment of Rural New Castle County, New Castle County outside of Wilmington, resulted in representative and senatorial districts grossly lacking in compactness and contiguity causing gerrymandering. Mr. McGinnes stated that he used the criteria of equality of population and the preservation of historic boundaries but that he put greater emphasis upon the factor first stated. As we have said, Mr. McGinnes cut the boundaries of some 1960 census enumeration districts and made large use of estimating procedures as hereinbefore stated. He also cut through some 20 election districts and in several areas disregarded old Hundred lines.
The plaintiffs point out that in creating the new 24th representative district, Mr. McGinnes departed from the historic boundary between White Clay Creek Hundred and Pencader Hundred, cutting some census enumeration district lines and some election district lines. In drawing the boundaries of the 24th representative district he proceeded along the boundary of Pencader Hundred to Marrows Road, thence a short distance north to Newark Road, then along Newark Road to the southwest to Chestnut Hill Road, then northwest to Christina Road, thence southwest along Newark Road to Ogletown Road, thence to the town of Red Lion, thence southwest to the village of Kirkwood by various courses to a point along the Wildlife Area of the Chesapeake and Delaware Canal, thence approximately west along the line of the Canal to the Delaware-Maryland boundary, and thence north to the place of beginning. These boundary lines caused to be included in the 24th representative district comparatively small parts of White Clay Creek Hundred and of New Castle Hundred.
As to the foregoing see Stipulation Ex. 10 H and Court's Ex. 19. See also Niederhauser, 1643 et seq.
The new 13th representative district, as drawn by Mr. McGinnes, includes the larger part of Christina Hundred but the boundaries of this representative district follow none of the Christina Hundred boundaries except where that Hundred borders on Wilmington.
See Stipulation Ex. 10 C and Court's Ex. 19.
The 7th senatorial district, as prepared by Mr. McGinnes, extends in an arc for approximately 21 miles along the Delaware-Pennsylvania border to the north. This is a curving line, part of a circle drawn from a point in the Town of New Castle in New Castle Hundred and based on a radius of 12 miles. The 7th senatorial district may be described as follows: beginning at a point on the Delaware-Pennsylvania border at the point where the Foulk Road crosses the border, thence proceeding westerly to that point where the 12 Mile Circle intersects Hopkins Road, thence due west to the juncture of the Delaware-Pennsylvania-Maryland boundaries, thence due south along the line of the Maryland-Delaware border to the point where the Pencader Hundred line cuts the Delaware-Maryland border, thence approximately due east to Elkton Road, an extension of Capitol Trail, thence cutting through Newark and turning due east on Main Street for a short distance, thence north along North College Avenue in Newark following the northern boundary of Newark, thence along Creek Road by various courses to the east, thence along the line of White Clay Creek to Harmony Road, thence north on Harmony Road to Kirkwood Highway, thence along Kirkwood Highway to Limestone Road, thence northwest along Limestone Road, thence northeast along Milltown Road to its intersection with Duncan Road, thence northwest along Duncan Road to Faulkland Road and thence following Faulkland Road east to the boundary of Wilmington, thence various courses along the boundary of Wilmington to Concord Avenue, thence along Concord Avenue, thence north to Talleyville, thence east by north and south by east along Silverside Road to Foulk Road, and thence north along Foulk Road to the place of beginning.
For a description of the scope of the Twelve-Mile Circle see State of New Jersey v. State of Delaware, 291 U.S. 361, 366, 54 S.Ct. 407, 78 L.Ed. 847 (1934), and note 1 cited to the text at the stated page.
See also "The Bounds of Delaware", The Star Publishing Company, Wilmington, Delaware, 64 et seq., Dudley Lunt, Esquire, of the Bars of Delaware and of Maine.
Stipulation Ex. 10C and Court's Ex. 20.
The plaintiffs contend that in his plan for reapportionment of Rural New Castle County Mr. McGinnes took care to arrange Republican strength in such a way as to concentrate it in the 7th senatorial district, thereby eliminating the possibility of electing more than one Republican senator in Mill Creek and Christina Hundreds. There is no doubt that Christina and Mill Creek Hundreds are strongly Republican and this has been the case for a substantial period of time. We take judicial notice of these facts. The new 13th and 14th representative districts were Republican even in the 1964 election which, as we all know, resulted in a Democratic victory of near landslide proportions. In creating the new 7th senatorial district Mr. McGinnes combined in one senatorial district not only the two representative districts largest in geographical area but also put into a single unit for the election of one senator those two representative districts in the State having the largest Republican majorities.
See Court's Exs. 19 and 20 and Document 174, Stipulation "Election Results under S.B. 336", "Exhibit 6".
The accuracy of our statement is demonstrable if we make use of the figures shown by "Exhibit 6", note 105, supra. The 13th representative district in the 1964 election delivered 6,980 Republican votes against 2,850 Democratic votes for a Republican majority of 4,130 votes, and the 14th representative district delivered 5,119 Republican votes against 3,044 Democratic votes for a Republican majority of 2,075 votes. Combining the Republican and Democratic votes of the 13th and 14th districts we find that there were 12,099 Republican votes against 5,894 Democratic votes for a Republican majority of 6,205 votes. The 7th senatorial district comprises the 13th and 14th representative districts. At the 1964 election according to official count the votes cast for senator in the 7th senatorial district were as follows: 12,670 votes for the Republican candidate and 5,424 for the Democratic candidate for a Republican majority of 7,246 votes.
Document 195, Exhibit 6-page 2.
The plaintiffs contend that Mr. McGinnes in effect gerrymandered the Republican strength in Mill Creek and Christina Hundreds so that one Republican senator and only one could be elected and that this resulted in the creation of a very large senatorial district with long borders which cannot be described as compact although it is internally contiguous. The plaintiffs point out that various other geographical combinations of representative districts were not only possible but were indeed more suitable and would have effected a more rational result and one which would have caused a more equal division of voting strength between the two major political parties. They state, for example, that if the 13th representative district had been combined with the 17th representative district a Republican senator would have been elected in 1964. Again taking the figures from "Exhibit 6", note 105, supra, we find that the 13th representative district returned 6,980 Republican votes against 2,850 Democratic votes for a Republican majority of 4,130 votes and that the 17th representative district returned 2,113 Republican votes against 3,729 Democratic votes for a Democratic majority of 1,616 votes but that of the respective voting returns, Republican and Democratic of the 13th and 17th representative districts be added there would be 9,093 Republican votes against 6,579 Democratic votes and a Republican majority of 2,514 votes would result. The 14th representative district returned 5,119 Republican votes against 3,044 Democratic votes for a Republican majority of 2,075 votes, and the 15th representative district returned 3,051 Republican votes against 3,339 Democratic votes for a Democratic majority of 288 votes. But if the voting of the two districts be combined there would be a total of 8,170 Republican votes against a total of 6,383 Democratic votes for a Republican majority of 1,787 votes. The voting for senatorial candidates followed with reasonable exactitude the voting for representatives. It would follow that if the combinations suggested in this paragraph had been employed by Mr. McGinnes, one additional Republican senator would have been elected.
See Niederhauser, 1664 et seq., and Plaintiffs' Ex. 53.
The foregoing can be demonstrated as follows: Using the combinations as suggested by Mr. Niederhauser, we find that the 5th senatorial district would be composed of the 9th and 10th representative districts and would have elected a Republican senator in the 1962 election. The 6th senatorial district would be composed of the 11th and 12th representative districts and also would have elected a Republican senator. The 7th senatorial district would be composed of the 13th and 17th representative districts and would have elected a Republican senator. The 8th senatorial district would be composed of the 14th and 15th representative districts and would have elected a Republican senator. The 9th senatorial district would be composed of the 18th and 19th representative districts and would have elected a Democratic senator. The 10th senatorial district would be composed of the 20th and 21st representative districts and would have elected a Democratic senator. The 11th senatorial district would be composed of the 16th and 24th representative districts and would have elected a Republican senator. The 12th senatorial district would be composed of the 22nd and 23rd representative districts and would have elected a Democratic senator. Thus, if the 1962 election patterns had been followed, 5 Republican senators rather than 4 would have been elected in 1964. We find it to be a fact, therefore, that if Mr. McGinnes had employed the combinations suggested by Mr. Niederhauser in his testimony and as demonstrated by the exhibits in notes 107 and 108, one more Republican senator would have been elected in 1964, a very strong Democratic year. Does this constitute proof of gerrymandering?
Plaintiffs' Exs. 53 and 60.
The 7th senatorial district is the largest senatorial district in Rural New Castle County. We have described it at an earlier point in this opinion and we need not repeat here what we have said previously. It is sprawling and oddly shaped. Mr. McGinnes, when he was on the witness stand, demonstrated a thorough knowledge of Rural New Castle County and the political significance of the election districts, representative districts and senatorial districts therein. We can entertain no doubt that he was aware of the results party-wise of the combination of representative districts employed by him to compose the 7th senatorial district, and when we weigh the additional fact that if the 1962 senatorial district patterns had been followed, even in the heavily Democratic 1964 election, five Republican senators rather than four would have been elected, we find that Mr. McGinnes' composing the 7th senatorial district as he has constituted a political gerrymander as described hereinafter. The 7th senatorial district does not meet the test of compactness and its elongated shape is the result of the political gerrymander. In connection with the 7th senatorial district, we call attention to the very odd shapes of the 5th and 6th senatorial districts, geographic oddities compelled by the form of the 7th senatorial district.
The plaintiffs assert that there was also gerrymandering by Mr. McGinnes in the creation of the 10th representative district which embraces the easterly part of Brandywine Hundred. They assert that the only four election districts in Brandywine Hundred that voted Democratic in all four general elections, 1956 to 1962 inclusive, were grouped together to create a "swing" representative district, i.e., the new 10th representative district, stretching the new district far enough to the south to include the old 7th election district, which was strongly Democratic. By this means, say the plaintiffs, a Democratic representative was elected in the 1964 election in the new 10th representative district. In support of this contention the plaintiffs asserted that the new 10th representative district is the most populous in the State, containing 13,928 persons according to the 1960 census. This assertion subsequently was withdrawn and the plaintiffs stipulated that the 28th representative district in Kent County, with a population of 14,209 persons as shown by the 1960 census, was the largest in the State. The 10th representative district is, however, the second most populous in the State. It commences at the low-water mark on the New Jersey side of the Delaware River along the arc of Delaware's northern boundary hereinbefore referred to, thence to the point where the right-of-way of the Baltimore Ohio Railroad cuts the arc on the north, thence southwest along the line of the right-of-way of the Railroad to Holly Oak Creek, thence southeast along Holly Oak Creek to a point where Holly Oak would cross Harrison Avenue if Harrison Avenue were projected to the northeast, thence southwest along Harrison Avenue to Holly Oak Road, thence along Holly Oak Road to the Delaware River and across the Delaware River to the low-water mark along the New Jersey shore, and thence to the low-water mark at the easterly bank of the Delaware River in New Jersey, and thence northeast following the low-water mark of the Delaware River to the place of beginning.,
See Document 174, Stipulation " Exhibit 3 (Exhibit S-1)".
See note 103, supra, State of New Jersey v. State of Delaware, note 1, cited to the text at 366, 54 S.Ct. 407, 78 L.Ed. 847.
See Stipulation Ex. 10A, Plaintiffs' Ex. 60, and Appendix.
The plaintiffs assert that if Mr. McGinnes had placed the lower boundary of the 10th representative district at Holly Oak Creek, the population of the 10th representative district would have been 12,905 persons, a figure close to the ideal population of 12,751 persons state-wide and also close to the mean or norm of 13,226 persons for Rural New Castle County. We find these assertions of the plaintiffs to be factually correct. Moreover, if Mr. McGinnes had followed the course just stated, the 10th representative district would not have cut through any election district nor through any census enumeration district.
See Table IX.
Mr. Niederhauser asserted that no census enumeration district would be cut. 1636 et seq. We conclude that his statement was correct. Compare Court's Ex. 21 which shows no cutting with Court's Ex. 19 and its accompanynig overlays, Court's Exs. 19A and 19B. We find that Court's Ex. 21 is correct and that the other three maps are in error in this particular, which of course is a comparatively small one.
Table XVI, set out immediately hereinafter, shows the voting by parties in the old election districts now comprising the new 10th representative district. We think the chart speaks for itself and that no further explanation of its words and figures is necessary.
By including the 7th election district in the 10th representative district, Mr. TABLE XVI. Voting in Old Election Districts now Comprising the New 10th Representative District Old Old 1956 1958 1960 1962 1964 Representative Election District District D. R. D. R. D. R. D. R. D. R.
6 1 321 304 232 184 378 239 227 118 249 152 6 5 352 336 354 284 483 432 349 274 507 402 6 8 176 362 199 265 315 337 202 244 393 336 6 18 128 196 134 163 221 196 105 97 249 182 6 23 303 283 258 203 315 273 256 196 302 225 6 26 115 332 134 260 158 294 128 255 225 345 6 34 339 572 160 148 255 254 167 147 261 225 6 37 136 227 162 209 278 322 213 294 272 366 6 47 ___ ___ 205 332 392 520 306 432 421 552 Sub Total 1870 2612 1838 2048 2795 2867 1953 2057 2879 2783 6 7 273 259 240 170 292 254 190 164 277 219 Total 2143 2871 2078 2218 3087 3121 2143 2221 3156 3002 Number of extra Democratic votes from 7th Election Dist. 14 70 38 26 58 McGinnes did cut two census enumeration districts, the 12th and the 37th, thus requiring him to estimate populations by using voting ratios, as we have described. The plaintiffs' contention is that the addition of the 7th election district to the 10th representative district made the 10th representative district a "swing" district, but as can be perceived from the bottom legend of Table XVI, "Number of Extra Democratic Votes from the 7th Election District", the added Democratic margins are indeed narrow ones. The contention of the plaintiffs that the inclusion of the 7th election district in the 10th representative district constitutes gerrymandering finds insufficient support in the evidence. The election districts comprising the 10th representative district returned Republican majorities in the elections of 1956, 1958, 1960 and 1962. It is true that in the 1964 election the 10th representative district returned a Democratic majority of 154 votes, but as we have stated, the election in 1964 was of landslide proportions in Delaware as in many other states. We point out that in the year 1964 the number of Democratic votes gained by adding the 7th election district to the 10th representative district numbered but 58.We find, viewing Rural New Castle County as a whole, that Mr. McGinnes, by creating representative districts as we have indicated above, concentrated the Republican vote in Rural New Castle County and created five safe Republican districts, viz., the 9th, 11th, 12th, 13th and 14th; at least four "safe" Democratic districts, and probably six when party voting runs normally, viz., the 17th, 19th, 20th, 21st, 22nd and 23rd; and five "swing" districts, viz., the 10th, 15th, 16th, 18th and 24th. This last creation was in an area that from 1956 to 1962 ordinarily voted Republican. In the 1964 election in which the vote was heavily Democratic, Democratic candidates in the fifteen districts of Rural New Castle County which were contested, captured ten seats or substantially two-thirds of the representatives in Rural New Castle County with slightly less than 50% of the vote. The plaintiffs tender the foregoing as proof of gerrymandering in Rural New Castle County. Viewing that area as a whole, we think the proof is insufficient to sustain this charge.
There was no Republican candidate for representative in the 20th representative district.
The representative and senatorial districts in Rural New Castle County are contiguous and, with the exceptions hereinbefore noted, they can be said to be reasonably compact. We state, however, that the new representative and senatorial districts do not follow historical lines or boundaries to any great degree and we find that the 7th senatorial does not meeting the test of compactness.
Delaware is a small state, narrow in compass, as any atlas will demonstrate.
C. As to the City of Wilmington and Rural New Castle County
We dealt at some length at a previous point in this opinion, under heading "X.", "Analysis of the Reapportionment", subheading "G.", with the vexing issue of the arithmetical disparity that was created by giving Wilmington eight representatives and Rural New Castle County sixteen representatives, and by giving Wilmington four senators and Rural New Castle County but eight senators, particularly in view of the statements in Mr. McGinnes' affidavit, which we have found to be correct, that there has been an increase in the population of Rural New Castle County of approximately 38,000 persons since the 1960 census. The disparity in populations is, of course, asserted by the plaintiffs as a proof of gerrymandering in favor of the Democratic party. There is no doubt that the City of Wilmington, as it voted in the 1962 election, was heavily Democratic, for out of a total of 32,743 votes cast for representatives there were 19,775 Democratic votes as against 12,968 Republican votes, the latter figure including 3,282 auditor votes in the 4th and 5th representative districts where no Republican candidate ran for office. The Democratic majority in Wilmington in that election therefore was 6,807 votes.
In the 1964 election Wilmington again went Democratic, electing eight Democratic representatives to the eight seats available. Democratic votes numbered 27,335 against 12,887 Republican votes. The Democratic party elected four senators out of the four senatorial seats available. The Democratic party secured 26,873 votes and the Republican party secured 12,375 votes. In Rural New Castle County the Democratic party elected eleven representatives and the Republican party five representatives out of the total sixteen seats available. There were 52,296 Democratic votes and 51,931 Republican votes.
We conclude that there was a substantial favoring of Wilmington over Rural New Castle County. We take judicial notice that Democratic leaders of the 122nd General Assembly were men of large political experience, possessing adequate knowledge of salient population figures at least of the larger geographical areas of the State such as Wilmington and Rural New Castle County, and were aware of the result of the apportionments created by S.B. 332 and S.B. 336, i.e., a favoring of Wilmington, believed to be Democratic, over Rural New Castle County, believed to be Republican. Should we find that the unjustified apportionments of Wilmington and Rural New Castle County in relation to each other were the result of political gerrymandering? Our answer, in the light of all the circumstances, is in the affirmative. Moreover, since the population of Rural New Castle County has increased substantially since the 1960 census — a fact generally known throughout the State — the gerrymander is all the more apparent. Mr. Latchum and Mr. McGinnes also are men of large political experience and knowledge of the populations of the City of Wilmington and of Rural New Castle County. They must be deemed to have been aware of the results of the apportionments of Wilmington and Rural New Castle County in relation to each other, created by S.B. 332 and S.B. 336, and implemented them as hereinbefore described.
Document 195.
D. As to Kent County
In Kent County, as we have said, Mr. Hughes split no election district, and the new representative and senatorial districts were formed from combinations of existing election districts. We find that Mr. Hughes adhered to historical boundaries insofar as he could do so. He created the new 27th representative district from the old 5th representative district. In the election of 1962, the old 5th representative district was composed of four election districts, viz., the 1st, 2d 3rd and 4th. Of these four election districts, only two had Republican majorities. They were the 3rd election district, in which the voting for representative was 438 Republican votes to 312 Democratic votes giving a Republican majority of 126 and the 4th election district, in which the voting for representative was 305 Republican votes to 269 Democratic votes giving a Republican majority of 36. In the 1st election district, the voting for representative was 283 Republican votes to 297 Democratic votes giving a Democratic majority of 14 votes. In the 2d election district, the voting for representative was 365 Republican votes to 414 Democratic votes giving a Democratic majority of 49 votes. Taking the votes in all the districts together, the Republican majority was 99 votes.
In the 1960 election there were three election districts in the old 5th representative district, viz., the 1st, 2d and 3rd. The 4th election district was added because of the increase in population in West Dover. All three election districts returned Republican majorities. In the 1st election district there were 445 Republican votes to 356 Democratic votes, giving a Republican majority of 89. In the 2d election district there were 779 Republican votes to 772 Democratic votes, giving a Republican majority of 7 votes. In the 3rd election district there were 364 Republican votes to 313 Democratic votes, giving a Republican majority of 51 votes. Taking the votes in all the districts together, the Republican majority was 147 votes.
Niederhauser, 1829; Hughes, 1059.
Mr. Hughes added the old 2d election district of the old 2d representative district to the old 5th representative district to complete the new 27th representative district. The point asserted by the plaintiffs is that the old 2d election district of the old 2d representative district returned the largest Republican majorities of any election districts of Kent County in both the 1960 and 1962 elections and that by the addition of the old 2d election district as stated, S.B. 336 limited the possibility of the election of Republican representatives from Kent County to one representative.
The following facts appear in support of the plaintiffs' contention. In the old 2nd election district in 1960 there were 639 Republican votes cast against 299 Democratic votes, giving a Republican majority of 340. In 1962 there were 496 Republican votes cast to 259 Democratic votes, giving a Republican majority of 237 votes. Had the election districts now constituting the 27th representative district been aligned in 1960 and 1962 as they are presently aligned by S.B. 336, the representative district would have given a return of 1740 Democratic votes against 2227 Republican votes, or a Republican majority of 487 votes in the 1960 election, and would have given a return of 1551 Democratic votes against 1887 Republican votes, or a Republican majority of 336 votes in the 1962 election. The plaintiffs assert that what S.B. 336 did was to concentrate the Republican strength in one single representative district. This is the converse of the argument which they have made in respect to Wilmington.
Plaintiffs' Ex. 60 Substitute.
There were, of course, no Republican representatives elected in Kent County in the 1964 election, which, as we have stated repeatedly, was of landslide proportions in Delaware. But such unusual election results do not invalidate the plaintiffs' contentions. Two Republican representatives were elected in Kent County in 1962. We take judicial notice of the fact that the election of Republican representatives in Kent County has ebbed and flowed in accordance with national and local political tides. There is no doubt that S.B. 336 does concentrate the Republican strength in Kent County in one district. We take judicial notice of the fact that in the 1960 election the Republican party elected 4 representatives to the ten seats available in Kent County in the 1958 election, 2 representatives; and in the 1956 election, 3 representatives.
It must be borne in mind that the 27th representative district is the "lop-eared rabbit" district, the shape of which is most unusual. We conclude that the 27th district does not meet the test of compactness as that term is used. See Appendix. There are no population figures in the record for the election districts adjacent to the 27th representative district and the census tracts of Kent County do not follow the election districts, but there appears to be no compelling reason for the precise combination of election districts effected by Mr. Hughes in creating the 27th representative district.
Court's Exs. 38, 39, 40.
Mr. Hughes, however, did observe the historic boundaries of Kent County to a substantial degree. It is a correct assertion of the plaintiffs that he concentrated Republican strength in the 27th representative district where there was in fact a Democratic majority of 386 votes in the 1964 election. Nonetheless in the adjacent 28th and 26th representative districts, where it is not alleged by the plaintiffs that there was any concentration of Republican strength, the Democratic majorities were, respectively, only 438 votes and 325 votes. We perceive no evidence sufficient to sustain a charge of gerrymandering here.
Document 174, "Exhibit 6".
The plaintiffs also assert that in the 1964 election because of gerrymandering the Democratic party made "a clean sweep of the three senatorial seats" in Kent County. The plaintiffs' argument here is largely statistical. They point out that the Republican votes in the election of 1964 amounted to 42.1% of the total and that in the election of 1962 the Republicans elected one senator out of the three then running for office. These are facts. We state, however, that in the 1962 election the Republican party received approximately 48.1% of the total vote cast in Kent County for state senators. The difference between 42.1% and 48.1% is not unsubstantial but we think the fact that the Republican party carried one senatorial seat in 1962 in Kent County is due to the fact that the party was far stronger in Kent County in 1962 than it was in 1964 as the percentages stated show. The evidence does not support the plaintiffs' contention.
Plaintiffs' Ex. 60 Substitute.
E. As to Sussex County
In Sussex County 6 representatives were elected in the 1964 election. The plaintiffs contend that the Republican strength was deliberately concentrated by Mr. West in the 34th and 35th representative districts and the 18th senatorial district formed by combining these two representative districts and that this constituted gerrymandering. They assert that as a consequence it is unlikely that the Republican party could elect more than 2 out of the 6 representatives in Sussex County, or more than 1 out of the 3 senators, even with 50% of the vote. We point out that the Republican party elected a senator from the 18th senatorial district in the 1964 election. The substance of the plaintiffs' contention here is substantially the same as it was in regard to the asserted gerrymandering in Kent County, viz., that S.B. 336 concentrated the Republican strength in a single area obviating the possibility of any large Republican representation from Sussex County in the House or Senate.
Mr. Niederhauser testified that the two strongest Republican districts were the old 1st election district of the old 7th representative district and the old 3rd election district of the old 10th representative district. He stated that the old 3rd election district of the old 10th representative district, which includes Rehoboth, "is by far the larger", meaning, we believe, to state in substance that ordinarily the old 10th representative district returned the largest Republican majorities in Sussex County. In the 1960 and 1962 elections, the old 3rd election district of the old 10th representative district had the largest Republican majorities in Sussex County. Mr. Niederhauser asserted that the Republican strength was concentrated in the new 34th representative district which, in his opinion, reduced the chances of the Republican party "carrying additional [representative] districts, or many additional [representative] districts."
Niederhauser, 1710 et seq.
The old 1st election district of the old 7th representative district, in the 1960 election, returned 549 Democratic votes against 780 Republican votes for a Republican majority of 231 votes. In the 1962 election, it returned 427 Democratic votes against 688 Republican votes for a Republican majority of 261 votes. The old 3rd election district of the old 10th representative district in the 1960 election returned 985 Democratic votes against 1400 Republican votes for a Republican majority of 415 votes, and in the 1962 election it returned 796 Democratic votes against 1357 Republican votes, for a Republican majority of 561 votes. To these strong Republican election districts Mr. West added the old 8th representative district which in 1960 returned 655 Democratic votes against 735 Republican votes for a Republican majority of 80 votes, and in 1962 returned 638 Democratic votes against 635 Republican votes for a Democratic majority of 3 votes. Totalling the three election districts which now comprise the new 34th representative district, we find that in the 1960 election there would have been returned 2189 Democratic votes against 2915 Republican votes for a Republican majority of 726 votes, and that in the 1962 election there would have been returned 1861 Democratic votes against 2680 Republican votes for a Republican majority of 819 votes.
The 35th representative district as composed by S.B. 336 is made up of the old 9th representative district and the old 1st and 2d election districts of the old 10th representative district. Totalling the three election districts which now comprise the new 35th representative district for the 1960 election we find that 2623 Democratic votes would have been returned as against 3123 Republican votes for a Republican majority of 500, and in the 1962 election 2509 Democratic votes would have been returned as against 2534 Republican votes for a Republican majority of 25. It would appear that should the new 35th representative district follow the pattern established by the 1960 and 1962 elections, Mr. West created a "swing" district. But when the new 34th and new 35th representative districts were combined to form the new 18th senatorial district, there would have been Republican majorities of 1226 votes in the 1960 election and 844 votes in the 1962 election.
The plaintiffs assert, as we have said, that Republican strength was concentrated by Mr. West in the 34th representative district. We point out, however, that in the 1964 election the 34th representative district returned a Democratic majority of 447 votes. But the 33rd representative district, adjoining it, where no Republican concentration is alleged by the plaintiffs, returned a Democratic majority of only 464 votes. In the 35th representative district in the 1964 election a Democratic majority of 739 votes was returned. If we add the Democratic majorities in the 34th and 35th representative districts in the 1964 election we find that they total 1186 votes. But the plaintiffs, in support of their contention of the concentration of Republican strength, point to the fact that the 18th senatorial district returned a Republican majority of 915 votes despite the landslide proportions of the Democratic victory in Delaware in 1964. We think this fact was due to the unusual strength of the Republican candidate for senator in the 18th senatorial district, then the chairman of the Republican County Committee for Sussex County, a man repeatedly elected to public office, and not because of a concentration of Republicans in the 18th senatorial district composed as it is by the 34th and 35th representative districts. Indeed the facts look the other way. We find that the plaintiffs' contention of gerrymandering on this aspect of the case cannot be sustained.
Part of the plaintiffs' argument respecting gerrymandering in Sussex County is statistical. They point out that the Republican party polled 43.8% of the total vote for representatives in Sussex County in the 1964 election but elected 0 out of 6, and that in the 1962 election the Republican party elected 3 representatives out of the 10, securing 47.2% of the votes cast for representatives. The plaintiffs also state correctly that in 1962 3 senators were elected in Sussex County, that the Republican party elected 1 senator and that the total Republican vote cast for senators amounted to 47.3%.
We have difficulty in grasping the force of the plaintiffs' statistical contentions. It is true that in the landslide year of 1964 the Republican party elected 1 senator. Statistically speaking, to elect 1 senator out of a total of 3 by 47.3% of the vote would not seem inequitable for the majority party by the plaintiffs' own reasoning should have elected and did elect 2 senators.
The plaintiffs assert that because the Republican party secured 43.8% of the votes cast in Sussex County the Republican party should have done better in the election and that this constitutes proof of gerrymandering. The truth of the matter is that the Republican party did not fare well in the 1964 election for members of the General Assembly in Sussex County. The answer to the plaintiffs' contention here is that our form of government is representational and not statistical.
No charge of gerrymandering in respect to Sussex County can be sustained on the proof made by the plaintiffs in support of their statistical argument.
F. Assertions of the Plaintiffs Based on State-wide Statistical Results
We refer again to Exhibits "S4A" and "S4B" of Document 174 relating, respectively, to the "Representativeness" of the 1962 and 1964 elections for members of the General Assembly. The plaintiffs point to the fact that the Republican party elected only 5 of the 35 members of the House or one-seventh of the total number of representatives in 1964 despite the fact that the Republican candidates received 44.4% of the vote, and that in the Senate the Republican party elected only 5 of the 18 candidates despite the fact that those candidates received 45% of the votes cast. They contrast these showings with the 1962 election where the Republican candidates won 5 out of the 10 senatorial seats which were contested but the plaintiffs fail to refer to the fact that in the 1962 election the Republican candidates received 54.2% of the vote. The plaintiffs contend that this demonstrates that the apportionment plan of S.B. 332 and S.B. 336 "has operated efficiently to minimize and cancel out the voting strength of the members of the minority party", and that "Under the development of the law as it now stands, the [present apportionment] scheme cannot pass constitutional muster and should be held invalid in its entirety."
Plaintiffs' post-trial brief, 73-74.
The plaintiffs have placed in their brief a table setting out the 1964 election returns by representative districts, arranged by voting totals in order of size of vote. This table is set out below.
Plaintiffs' post-trial brief, 74-75.
1964 Election Results Representative Districts Arranged by Voting Totals
Vote for Representative
Size of District No. of Vote for Representative (Total Vote) Districts % of State Total %R
2,537-5,000 8 16.7 33.2 5,000-5,500 8 21.0 37.4 5,500-6,000 9 26.0 42.7 6,000-7,000 6 19.3 44.9 7,000-9,830 4 17.1 65.9
The "1964 Election Results", note 125, supra, does show that the 7 largest representative districts in the State, according to vote totals, are in Rural New Castle County, and that the 3 of the 4 smallest representative districts are in Wilmington. The 1964 Election Results also show, as the plaintiffs point out, that the 4 largest representative districts in the State are in Rural New Castle County, and that in the 1964 election these representative districts cast from 7,000 votes to 9,830 votes averaging 65.9% Republican. We agree with the contention of the plaintiffs that these facts possess significance in respect to the division of representatives between New Castle County outside of Wilmington, i.e., Rural New Castle County, and the City of Wilmington as we have indicated previously. We shall deal with this matter again at a later point in this opinion.
The plaintiffs also point out that the 26th representative district of Kent County cast the smallest vote of any representative district in the 1964 election, viz., 2537 votes, and that this figure casts doubt upon the accuracy of the estimate arrived at by Mr. Hughes in calculating the population of the old 2d election district of the old 2d representative district, comprising part of the new 27th representative district, by the use of the ratio method heretofore described.
As was stated at an earlier point in this opinion, Mr. Hughes knew there were 1079 registered voters in the old 2d election district of the old 2d representative district. Using, roughly, the ratio of 4-1, he estimated there were 4,000 people in the area designated, though he conceded in subsequent testimony that the 4,000 persons estimated by him was perhaps too high.
The plaintiffs asserted that a substantial number of the persons inhabiting the Dover Air Force Base, which lies in the old 2d representative district referred to and is now in the 26th representative district, do not vote. The small number of votes cast in the 26th representative district for representative in the 1964 election bears out the plaintiffs' contention. On the present record the correct population of the old 2d election district of the old 2d representative district cannot be ascertained, but it is clear that Mr. Hughes' estimate was too high.
We find that the proof submitted by the plaintiffs in support of their contentions under this subheading, "F.", is too speculative to be sustained. We point out, however, that the malapportionments described in foregoing headings do of course affect the totality of the reapportionment of Delaware under S.B. 332 and S.B. 336. We have also found under previous subheadings instances of gerrymandering but we conclude that the proof of gerrymandering brought forward by the plaintiffs under this subheading, "F.", is insufficient.
XII. The Law as to Deviations in Population and as to Gerrymandering
A. The Law as to Deviations in Population
We will discuss under this heading the law dealing with deviations in populations as they have been set out hereinbefore in this opinion, and when this has been completed, under the next subheading "B.", we will discuss the law insofar as it applies to the issue of gerrymandering.
In our opinion in Sincock v. Duffy, 215 F. Supp. 169, 190, we stated the view that a ratio not exceeding 1.5-1 between geographic voting units of the State of Delaware would not constitute too great a disparity. In so stating we relied on a number of state court decisions and suggested examination of The Yale Law Journal, Vol. 72, No. 1, p. 93 and Appendix B, pp. 101-04. On appeal to the Supreme Court, however, Mr. Chief Justice Warren, delivering the majority opinion affirming our judgment in Sincock v. Duffy, sub. nom., Roman v. Sincock, 377 U.S. 695, 710, 84 S.Ct. 1449, 1458, 12 L.Ed.2d 620 (1964), stated: "Our affirmance of the decision below is not meant to indicate approval of the District Court's attempt to state in mathematical language the constitutionally permissible bounds of discretion in deviating from apportionment according to population. In our view the problem does not lend itself to any such uniform formula, and it is neither practicable nor desirable to establish rigid mathematical standards for evaluating the constitutional validity of a state legislative apportionment scheme under the Equal Protection Clause. Rather, the proper judicial approach is to ascertain whether, under the particular circumstances existing in the individual State whose legislative apportionment is at issue, there has been a faithful adherence to a plan of population-based representation, with such minor deviations only as may occur in recognizing certain factors that are free from any taint of arbitrariness or discrimination."
In note 21 cited to the text of Mr. Chief Justice Warren's opinion it is stated: "The court below suggested that population-variance ratios smaller than 1 1/2-to-1 would presumably comport with minimal constitutional requisites, while ratios in excess thereof would necessarily involve deviations from population-based apportionment too extreme to be constitutionally sustainable. See 215 F. Supp., at 190."
Though the language employed by Mr. Chief Justice Warren quoted above does not explicitly disapprove the suggested ratio of 1.5-1 as the maximum allowable deviation, we conclude that our suggested ratio did not meet with favor in the eyes of the Supreme Court and had our statement been relevant to our decision and not merely dictum, the disapproval might well have been express instead of implied. Following the decision of the Supreme Court in Sincock v. Roman, reliance nonetheless has been placed by the defendants, both in oral argument and upon their brief, upon the ratio of 1.5-1 referred to. We deem this reliance to be unwarranted.
Many factors, of course, must be considered in determining whether the plan of apportionment created for Delaware by S.B. 332 and S.B. 336 meets the standard of "one person, one vote". That population deviations require careful consideration is demonstrated by the statement from Reynolds v. Sims, 377 U.S. 533, 581, 84 S.Ct. 1362, 1392, 12 L.Ed.2d 506 (1964) as follows: "[C]areful judicial scrutiny must of course be given, in evaluating state apportionment schemes, to the character as well as the degree of deviations from a strict population basis." See Davis v. Mann, 377 U.S. 678, 84 S.Ct. 1441, 12 L.Ed.2d 609 (1964).
The Chief Justice also stated in Reynolds v. Sims, supra, at 577, 84 S.Ct. at 1390: "[W]e mean that the Equal Protection Clause requires that a State make an honest and good faith effort to construct districts, in both houses of its legislature, as nearly of equal population as is practicable. We realize that it is a practical impossibility to arrange legislative districts so that each one has an identical number of residents, or citizens, or voters. Mathematical exactness or precision is hardly a workable constitutional requirement." The Chief Justice pointed out, id. at 580-81, 84 S.Ct. at 1391, that there could be justification for some deviations from population-based representation in state legislatures in order to insure some voice "to political subdivisions, as political subdivisions", and further "[A] State may legitimately desire to construct districts along political subdivision lines to deter the possibilities of gerrymandering."
The Supreme Court therefore has laid heavy emphasis upon the factor of rationality of apportionment. Our first inquiry, therefore, must be as to the rationality of the Delaware apportionment plan now sub judice.
It is, of course, the fact, as already pointed out, that Mr. Latchum, Mr. McGinnes, Mr. Hughes and Mr. West had to put together an apportionment plan very quickly. The time elapsing between the decision of the Supreme Court in Roman v. Sincock on June 15, 1964 and the primary and general elections of 1964 was short. We have this in mind, but in our opinion these gentlemen and the General Assembly, facing a hard task, found a hard way to solve it. We agree, however, with the statement of Dr. Malcolm E. Jewell who testified that the Delaware plan in his judgment was the best plan of apportionment that any State had evolved as of the time of his testimony, viz., August 5, 1964. The members of the 122nd General Assembly of Delaware from the rural areas deserve credit and honor for voluntarily depriving themselves of a large part of the very great power they held in controlling legislation in Delaware and in giving away a very large part of that power to Wilmington and upstate areas. We do not intend to denigrate their efforts. Nevertheless we must view the plan in the light of the decisions of the Supreme Court and of our own consciences.
Jewell, 305-06.
What the progenitors of the plan under scrutiny here did was to cut the suit of clothes and sew it together without having any very sound measurements of the stature of the man who had to wear the garment. The four tailors also cut and sewed their portions of the garment with but little collaboration. Once S.B. 332 was passed, there were inescapable built-in mathematical disparities population-wise which could not be avoided in S.B. 336. Deviations were inescapable so long as the Wilmington City lines were adhered to and the number of senators and representatives were divided among the respective political areas as allocated by S.B. 332.
We cannot perceive why the General Assembly seized upon 18 as the number of senators. If S.B. 332 had provided for 19 senators; 4 for Wilmington, 9 for Rural New Castle County, 3 for Kent County, and 3 for Sussex County, the percentages of political power in the General Assembly would have been as follows: Wilmington 21.05, Rural New Castle County 47.36, Kent County 15.78, and Sussex County 15.78. Compare the percentages cited with columns 3 and 7 of Table III. It will be observed that the inequalities of political power which are not unsubstantial as demonstrated by Table III would be practically obliterated. Nor can we perceive any sound reason for adhering so strictly to Wilmington City boundaries. In the plaintiffs' plan, the plaintiffs' 4th senatorial district crosses the Wilmington City line to the northeast and includes a small area of Brandywine Hundred but with a very equitable result.
See Plaintiffs' Exs. 48 and 49.
Referring now to that portion of our opinion designated as "X. Analysis of Reapportionment, * * * B. Deviations in Representative and Senatorial Districts within the City of Wilmington", we reiterate that we found the difference in populations between the 5th representative district, the largest, containing 12,917 persons, and the 8th representative district, the smallest, containing 11,114 persons, to be 1803 persons, and the ratio between them to be 1.16-1.
In respect to senatorial districts, we have found that the difference in population between the 3rd senatorial district, the largest, containing 24,630 persons, and the 1st senatorial district, the smallest, containing 23,563 persons, to be 1067 persons, and the ratio between the largest and the smallest to be 1.05-1. Ordinarily these would be considered to be comparatively minor disparities and we would not deem them to be of significant consequence were it not for the fact that Wilmington is a comparatively small city, easily divided, geographically, and the additional fact that the Wilmington block census figures were in Mr. Latchum's hands. In addition to these very pertinent circumstances the ratios could have been reduced without difficulty to the very low ratio of 1.01-1 as is demonstrated by the plaintiffs' plan for Wilmington, Table XIV.
Latchum, 378.
It must be borne in mind that the political units within Wilmington are very small. Moreover it is clear that the representative and senatorial districts in Wilmington, as created by S.B. 332 and S.B. 336, were laid down without regard to historical boundaries, save only the bounds of Wilmington itself, which have no relevance here other than in delineating the political unit of Wilmington. The ease with which a better plan for Wilmington than that presented by S.B. 332 and S.B. 336 could have been evolved assuredly is a factor to be considered in the light of the principles quoted in this portion of this opinion from Reynolds v. Sims going to the issue of what is practicable in the making of political apportionments. The deviations in population and the disparities of the ratios are much smaller, however, than those thus far condemned by any court insofar as we have been able to ascertain. Cf. Baker v. Carr, 247 F. Supp. 629, and Baker v. Clement, 147 F. Supp. 886 (M.D.Tenn. 1965). It will be observed that the district figures with which the Tennessee Three-Judge Courts were dealing are large indeed compared to those which we have before us for the Delaware districts. While it is not our duty to compel the General Assembly of Delaware to create a better plan of apportionment, it is our task to determine whether the plan before us transgresses the Equal Protection Clause of the Fourteenth Amendment. In the light of the foregoing, we conclude that the Wilmington apportionment plan fails to meet the test of Reynolds v. Sims, quoted above, in that it does not construct districts "as nearly of equal population as is practicable."
In respect to Rural New Castle County and its representative and senatorial districts, we have set out the relevant statistics under heading "X.", "C.", supra. The 10th representative district, the largest, is shown as containing 13,928 persons, and the smallest, the 20th representative district, as containing 12,741 persons, creating a difference of 1187 persons, according to " Exhibit 3 (Exhibit S-1)", and the difference of populations was shown to be at a ratio of 1.09-1. While these ratios are comparable in size to those ratios described in the preceding paragraph in respect to the Wilmington representative and senatorial districts, should we find these deviations to be of such magnitudes in and of themselves as to constitute invidious discrimination? We bear in mind the fact that Mr. McGinnes did not have available block census figures and in view of the apportionment provided by S.B. 332 had to proceed by splitting numerous census enumeration districts, employing ratios and other means of estimating as described in this opinion. Assuming that Mr. McGinnes was entitled to make use of the methods necessary to effect the apportionment that he made, i.e., splitting census enumeration districts and making estimates, can we conclude that the results that he achieved, regarding Rural New Castle County as a unit, gave a result of "one person, one vote" insofar as was practicable, thereby meeting the standards of Reynolds v. Sims? We conclude that Mr. McGinnes did not construct representative districts as nearly of equal population as was practicable under the circumstances, and, therefore, this apportionment does not meet the test of Reynolds v. Sims.
Referring again to that portion of our opinion last designated but dealing now with the senatorial districts in Rural New Castle County, as we stated the largest senatorial district is the 6th with an estimated population of 27,715 persons and the smallest is the 8th senatorial district with an estimated population of 25,686 persons. The difference was found to be 2029 persons and the ratio between the two districts to be 1.079-1. Assuming again that Mr. McGinnes was entitled to make use of the method of apportionment employed by him, we find that the result that he achieved, treating Rural New Castle County as a unit, did not effect the result of "one person, one vote" insofar as was practicable.
In respect to the deviations between representative and senatorial districts in Wilmington respectively vis-á-vis representative and senatorial districts in Rural New Castle County, we refer again to the statistics in that portion of our opinion previously designated. We found the smallest representative district to be the 8th in Wilmington, with an estimated population of 11,114 persons, and the largest to be the 10th representative district in Rural New Castle County with an estimated population of 13,928 persons. The difference is 2814 persons. The ratio between the two districts we found to be 1.25-1. The state-wide norm for the representative districts is 12,751 persons. The difference between the state-wide norm and the 8th representative district is -1637 persons, and the difference between the statewide norm and the 10th representative district is +1177 persons. The percentage of deviation in the case of the 8th representative district is -12.84% from the norm, and in the case of the 10th representative district is +9.23% from the norm. These deviations are not insubstantial under the circumstances of this case. The question which confronts us is, are they sufficiently large to break the standard of "one person, one vote"? We are of the opinion that they are too large and do not conform to the standard of "one person, one vote" insofar as is practicable.
As to senatorial districts, we found that the largest was the 6th senatorial district in Rural New Castle County, with an estimated population of 27,715 persons, and the smallest was the 1st in Wilmington with a population of 23,563 persons, and that the difference was 4152 persons. The ratio here was found to be 1.17-1. The state-wide norm for senatorial districts is 24,794 persons. The difference between the state-wide norm and the 6th senatorial district in Rural New Castle County is +2921 persons, and the difference between the state-wide norm and the 1st senatorial district is -1231 persons. The 1st senatorial district in Wilmington is contiguous to and runs with a portion of the southern boundary of the 6th senatorial district in Rural New Castle County. This is too high a ratio and, even imposing the stricture of observing the boundaries of Wilmington, we cannot consider that it provides an equality of population as nearly as is practicable.
The background statistical material of the problem of the over-representation of the City of Wilmington in relation to Rural New Castle County has been set out under headings "X.", "G." and "XI.", "C". We also refer again to column 3 of Tables I and II. Under the 1960 census Rural New Castle County is entitled to 16.59 representatives and to 8.5 senators while Wilmington is entitled only to 7.51 representatives and 3.86 senators on the basis of the 1960 census, without regard to the 38,000 persons who have moved into Rural New Castle County since that census according to Mr. McGinnes' affidavit. It appears clearly from the foregoing that Rural New Castle County has been the victim of invidious discrimination. If we add to the population of Rural New Castle County, as shown by the 1960 census, the 38,000 people who have moved into Rural New Castle County since that census it is clearly demonstrated that the inhabitants of Wilmington, as a unit, have a 30+% advantage over the inhabitants of Rural New Castle County, as a unit. Conversely the apportionments have decreased the representation of the inhabitants of Rural New Castle County, treating them as a unit, by about 18%.
See again the statements from the McGinnes' affidavit referred to on page 5 of this opinion and our finding.
The advantage given to the inhabitants of Wilmington can be stated in another manner. We have found that the 8 representatives of Wilmington each represent a mean or norm of 11,978 persons, according to Table VIII. If we add 38,000 persons to the population of Rural New Castle County as shown by the 1960 census, the total population is approximately 250,000 persons. Each of the 16 representatives elected in Rural New Castle County represents therefore about 15,625 persons, whereas each representative elected in Wilmington represents about 11,978 persons. The difference demonstrates an increase in representation per each representative elected in Rural New Castle County of 3,647 persons. The advantage of an inhabitant of Wilmington over an inhabitant of Rural New Castle County is thus demonstrated to be approximately 30+%, as we stated. A fortiori, it becomes all the more manifest that Rural New Castle County has been the subject of invidious discrimination.
The situation presented in respect to Rural New Castle County vis-a-vis Wilmington is directly opposite to that which was before the Three-Judge Court in Baker v. Carr, 247 F. Supp. 629 (M.D.Tenn. 1965). One of the stated purposes of the Tennessee Reapportionment Act, was "to equalize in one house any inequalities in the representation of certain areas found to be necessary in the other house." Id. at 633. No equalizing purpose appears on an examination of S.B. 332 and S.B. 336.
1965 Public Chapter No. 3, Extraordinary Session (84th General Assembly), Senate Bill No. 10, signed by the Governor on May 28, 1965.
Respecting senatorial representation, we find that if the 8 senators be divided into a Rural New Castle County population of 250,000 persons, the representation is 31,250 persons per senator, whereas in Wilmington each senator continues to represent 23,957 persons. The difference is 7,293 persons, which again results in a 30+% advantage for the inhabitants of Wilmington over the inhabitants of Rural New Castle County.
The percentages, of course, remain the same.
The disparities demonstrated are too large to be acceptable. We conclude that they are constitutionally impermissible.
We dealt with the statistical information respecting deviations in senatorial and representative districts in Kent County under the heading "X.", "E." of this opinion. We found that the largest representative district in Kent County, the 28th, was estimated to contain 14,209 persons, and that the smallest, the 29th, was deemed to contain 11,350 persons. The difference in population between the 28th and 29th representative districts was found to be 2,859 persons and the ratio was determined to be 1.25-1. An issue of law is, of course, presented. Are these deviations such as to be constitutionally impermissible within the rule of Reynolds v. Sims? They are, of course, substantially smaller than any deviation which has been held unconstitutional in any decided case that we have been able to discover.
We have pointed out that Mr. Hughes was faced, as were Mr. Latchum, Mr. McGinnes and Mr. West, with the necessity of very quickly preparing an apportionment plan in the short period available prior to the primary and general elections of 1964. Mr. Hughes testified that if the old election districts were not to be employed and adhered to in creating the reapportionment, a new registration would have been necessary and there was insufficient time to effect a new registration. We refused to enjoin the holding of the 1964 elections. Sincock v. Roman, 233 F. Supp. 615 (1964). We stated in the cited opinion that it was necessary to examine the entire apportionment of Delaware under S.B. 332 and S.B. 336 before any definite conclusions could be reached as to the constitutionality of the apportionment. The disparity in populations in representative districts within Kent County is the highest intra-county in the State. Upon consideration of all pertinent factors, including the circumstances that the pressure of time is certainly not as great now as when S.B. 332 and S.B. 336 were enacted and that the units involved in Kent County are small indeed, we are forced to the conclusion that the representational disparity is too large to be permissible. The Kent County reapportionment in the respects indicated does not meet the test of practicability.
In respect to senatorial districts in Kent County, as we stated previously, where the populations of senatorial districts are in dispute we employ the defendants' figures. According to the defendants' figures, the largest senatorial district in Kent County is the 15th which is asserted to have a population of 23,903 persons, and the smallest is the 13th, which, according to the figures of the stipulation, has a population of 20,743 persons. The difference was found to be 3,160 persons. The ratio population-wise between the two senatorial districts was determined to be 1.15-1. Assuming that the method of apportionment employed by Mr. Hughes to be a permissible one and bearing in mind that we are dealing here with the larger units of senatorial districts and not with representative districts, we find this deviation to be consequential and significant and conclude that it is of a sufficient degree or size to fall within the constitutional prohibition.
As to deviations in the representativeness of the General Assembly Districts in Sussex County, the statistical data is set out under the heading "X.", "F." of this opinion. The largest representative district was shown to be the 32nd, containing 13,458 persons, and the smallest was the 34th, containing 11,017 persons. The stipulated figures coincide with those set out in our Table XI. The difference was 2441 persons. The ratio between the two representative districts was determined to be 1.22-1. We find that a ratio of 1.25-1 in Kent County was impermissible, and the ratio here is but .03 less. Our legal conclusion is the same as it was in respect to the disparate representative districts in Kent County.
As to the senatorial districts, the largest senatorial district in Sussex County, as appears from the stipulation and Table IX, was found to be the 16th with a population of 25,517 persons, and the smallest was found to be the 18th, estimated to contain 22,686 persons, according to the stipulation and Table IX. The difference in population between the largest district and the smallest district was found to be 2,831 persons. The ratio was found to be 1.12-1. Assuming that the apportionment method employed by Mr. West was permissible, we conclude that this deviation is of sufficient significance and size to raise a constitutional issue.
As to state-wide disparities in representative and senatorial districts, the statistical background will be found under the heading "X.", "G." of this opinion. We found the largest representative district in the State to be the 28th of Kent County, estimated as having a population of 14,209 persons. The smallest representative district was found to be the 34th in Sussex County, which was found to contain 11,017 persons. The figures cited are in accordance with Documents 174 and 195 and with the area figures set out in Table VI. The difference in populations was found to be 3,192 persons. The 8th representative district in Wilmington, found to contain 11,114 persons, Stipulation Ex. 11, is almost as small as the 34th representative district. The ratio between the largest and the smallest representative districts in the State, the 34th and the 28th, was determined to be 1.289-1. As stated previously under the heading designated at the beginning of this paragraph, an examination of Documents 174 and 195, the map exhibits and the tables will demonstrate that the populations of several other representative districts throughout the State are almost equally out of balance.
In respect to senatorial districts statewide, we found the largest senatorial district in the State to be the 6th in Rural New Castle County which has a population according to the 1960 census of 27,715 persons, and the smallest senatorial district to be the 13th in Kent County, which has a population of 20,743 persons. These figures were stipulated to and also are in accord with Table VII. The difference in populations is 6,972 persons. The ratio between the largest and the smallest senatorial districts was found to be 1.33-1. As we said under the heading designated as "X." "G." while a difference of 6.972 persons is numerically the largest created in any General Assembly district in the State and the ratio of 1.33-1 is the highest ratio, an examination of the stipulation and the exhibits will show several senatorial districts almost equally lacking in parity and these will supply ratios fairly close to that last stated.
Taking into consideration the state-wide showing in respect to disparate populations in both representative and senatorial districts and considering all of the circumstances, including the speed with which the reapportionment plan for Delaware had to be put together, nonetheless we are compelled to the conclusion that the state-wide disparities pointed out are constitutionally impermissible. It is true that if we contrast the deviations in the present Delaware apportionment with those of the States set out in Appendix "B" to the opinion of the District Court in Baker v. Carr, supra, 247 F. Supp. at 642, the Delaware disparities seem small, but it must be borne in mind that the entire population of Delaware hardly equals those of many federal representative districts and that the Delaware land mass is the second smallest of the States being but 1978 square miles. Evaluating the Delaware apportionment in its totality we cannot avoid the conclusion that there has been an insufficient compliance with the Equal Protection Clause. Lucas v. General Forty-Fourth Assembly of State of Colorado, 377 U.S. 713, 735, n. 27, 84 S.Ct. 1459, 12 L.Ed.2d 632 (1964). Delaware reapportionment is not a fit subject for glib answers.
In conclusion upon the subject of General Assembly districts disparate in populations, considering them not only upon a state-wide basis but also as they exist within the individual political subdivisions of the State, we reiterate that there was but little coordination of effort in creating the reapportionment of Delaware by the four gentlemen who prepared the representative and senatorial districts in the respective political divisions. Moreover, the record fails to demonstrate any studied coordination of districts or the populations therein by the General Assembly. While we are aware of the presumption of constitutional validity that attends every legislative enactment, we are compelled to adhere to our conclusion that the apportionment of Delaware, as created by S.B. 332 and S.B. 336, fails to meet the constitutional test in that these statutes do not adhere to the principle of "one person, one vote" insofar as practicable as required by Reynolds v. Sims.
The whole subject of disparities in the present Delaware apportionment can be summed up by quoting the language of Mr. Chief Justice Warren in Roman v. Sincock, supra, 377 U.S. at 710, 84 S.Ct. at 1458: "[T]he proper judicial approach is to ascertain whether, under the particular circumstances existing in the individual State whose legislative apportionment is at issue, there has been a faithful adherence to a plan of population-based representation, with such minor deviations only as may occur in recognizing certain factors that are free from any taint of arbitrariness or discrimination." The reapportionment provided by S.B. 332 and S.B. 336 does not the test quoted above.
B. The Law as to Gerrymandering
We must now consider the law in respect to gerrymandering, including gerrymandering based on race, i.e., gerrymandering such as was before the Supreme Court in Gomillion v. Lightfoot, 364 U.S. 339, 81 S.Ct. 125, 5 L.Ed.2d 110 (1960). The plaintiffs assert that racial gerrymandering is clearly before this court on the instant record. The pertinent facts are set out above under the heading, "XI. Facts Relating to the Issues of Gerrymandering. A. As to the City of Wilmington". It is the contention of the plaintiffs, and we have stated that contention more particularly under the heading last referred to, that Mr. Latchum endeavored to create and did create a predominantly Negro district in Wilmington, the 3rd representative district, and that the representation of the Negro racial minority thereby has been limited to but one representative out of the eight allocated by S.B. 332 to Wilmington, though Negroes constitute more than 25% of the total population. But as we pointed out, according to the 1964 election, the voters of Wilmington elected one Negro senator, a perfect result percentage-wise, and two representatives, an almost perfect result percentage-wise. It is abundantly clear to us that there was no racial gerrymandering. We conclude that the decision of the Supreme Court in Wright v. Rockefeller, 376 U.S. 52, 84 S.Ct. 603, 11 L.Ed.2d 512 (1964), governs here.
We now consider the law in respect to gerrymandering other than gerrymandering based on race. For want of a more apt phrase we will refer to the type of gerrymandering which we will now proceed to consider as "political gerrymandering". It is the contention of the defendants that the established principle of "one person, one vote" does not encompass political gerrymandering as distinguished from racial gerrymandering. The plaintiffs assert that political gerrymandering was exercised by apportioning Delaware's General Assembly districts in such a way as to enable the Democratic party to obtain control of, or a more dominant position in, certain General Assembly districts by concentrating Republican strength in certain localities or by dividing it in other areas in order to limit the election of Republican senators and representatives.
As we understand political gerrymandering it effects a diminution of the qualitative as distinguished from the quantitative basis of an individual's vote. We refer to "Apportionment and Representative Government", by Alfred de Grazia. Mr. de Grazia described gerrymandering as follows: "Gerrymandering * * * is a positive act of malapportionment. Its special character emanates not from being a biased apportionment (for all apportionments are "biased"), but from its disharmony with the legal order. Any system of apportionment may be gerrymandered, although some systems, such as the free apportionment of proportional representation, are more difficult to gerrymander than others.
"The gerrymander most commonly observed in American politics understandably affects the prevailing system of apportionment by territorial survey. In North Carolina, for example, one finds certain long bacon-like strips that slice up the Republican vote into pieces manageable by the dominant Democrats. A territorial gerrymander may violate the stipulated condition that district populations be equal or that district boundaries be drawn solely with reference to enclosing a contiguous and compact area and population or both. The usual objective of a gerrymander is to maximize the number of districts returning safe majorities for the apportioning group and to minimize the number of districts returning safe majorities for the opposition party or faction. The gerrymandering group seeks to draw district boundaries in a way that will concentrate its opponents' votes in as few districts as possible and will spread its own dependable majorities over as many districts as possible; or it seeks to disperse the majorities of its opponents and gather together its own partisans to create new majorities. Hence the success of a gerrymander depends upon an accurate knowledge of past voting behavior in the various constituencies and a reliable prediction about voting behavior in the future in both old and new constituencies.
"The temptation to gerrymander or encourage rotten boroughs is always present in a legislature. Extreme gerrymanders, maximizing the predictive ability of the dominant faction, are usually prevented not only by various legal provisions but also by the erring legislators' consciences that provide generally some modicum of shame and by their fears of exciting too great a popular resentment against transgressions of traditional and legal rules. The gerrymander and rotten boroughs are perversions of the accepted traditional and legal order, it is well to emphasize here, and can be distinguished readily from the more general struggle to achieve social values, of which the process of apportionment forms part."
"Apportionment and Representative Government", 66-67, Alfred de Grazia, Frederick A. Praeger, New York, 1962. See also Niederhauser, 1545-46.
We point out that there are at least two conceivable answers to the legal problem here posed. One answer might lie in holding that where malapportionment results from gerrymandering the principle of "one person, one vote" would be applicable and would prohibit the malapportionment resulting from the gerrymandering. This answer is in reality based on malapportionment. Another answer lies in holding that purely political gerrymandering, as distinguished from gerrymandering based on race, and without regard to malapportionment, falls within the interdiction of the Fourteenth Amendment. An example will clarify what we have just written: a situation is conceivable where every general assembly district in a state could be balanced perfectly population-wise by an equal number of inhabitants in each general assembly district yet be so set up or arranged geographically that the majority party could greatly reduce or even obliterate minority party representation.
Though the plaintiffs have referred to decisions or opinions of lower courts as supporting the theory that political gerrymandering is cognizable, the fact is that no lower court opinion or decision supports their contentions with the exception of the dicta in our opinions in Sincock v. Duffy, supra, 215 F. Supp. at 188-89, and Sincock v. Roman, supra, 233 F. Supp. at 620. Nonetheless we will review the lower courts' decisions.
In our opinion in Sincock v. Duffy, 215 F. Supp. 169, 188 (1963), we stated: "As to the 2 additional senators allotted to Kent and Sussex Counties superimposed on existing senatorial districts, viz., Districts 6 and 7 in Kent and Sussex Counties, so that voters in Kent and Sussex Counties will elect two senators while the voters of New Castle County will continue to vote for but one senator, this seems to involve a hidden gerrymander that cannot be countenanced and if the reapportionment created by the 1963 Amendment met the requirements of the Equal Protection Clause in all other respects, though it does not, this feature alone, in the absence of any substantial possibility of severability, would require us to adjudge the whole Amendment unconstitutional."
In our opinion in Sincock v. Roman, 233 F. Supp. 615, 620 (1964), we stated: "As to the issue of gerrymandering, the plaintiffs have offered evidence in support of their charge while the defendants vehemently deny any intent to gerrymander on the part of the General Assembly and that the new apportionment will result in any unfairness, partisan or otherwise, to any political party in the election of members of the General Assembly. We shall not attempt now to analyze the evidence presented on this issue in this opinion."
The plaintiffs can gain no support from the decision of the Supreme Court of Pennsylvania in Butcher v. Bloom, 415 Pa. 438, 467-68, 203 A.2d 556, 573 (1964). In a more recent decision in the case last cited, the Supreme Court of Pennsylvania itself reapportioned Pennsylvania in a per curiam opinion, 420 Pa. 305, 216 A.2d 457 (1966). Mr. Justice Musmanno dissented, stating as his view that both as to senatorial and representative districts the Supreme Court of Pennsylvania had itself gerrymandered the Pennsylvania General Assembly districts and reached the conclusion that gerrymandering was constitutionally prohibited. Id. at 465-69. Mr. Justice Eagen also dissented upon the ground that the Supreme Court of Pennsylvania had gerrymandered the senatorial districts of the State and he concluded also that this course was constitutionally prohibited.
The plaintiffs quote from the opinion in Butcher v. Bloom, 415 Pa. 438, 467-68, 203 A.2d 556, 573 (1964) in which the Supreme Court of Pennsylvania stated: "We would agree with the district court, however, that in the absence of any reasonable justification (historical or otherwise), such districting might be the result of gerrymandering for partisan advantage and, in that event, would be arbitrary and capricious."
The reference is to the decision of the Three-Judge District Court in Drew v. Scranton, 229 F. Supp. 310, 326 (M.D.Pa. 1964), wherein Circuit Judge Maris, referring to the disparate treatment accorded to three quite homogeneous adjoining counties in Pennsylvania, Berks, Lancaster and York, stated: "In the absence of any legislative history or other explanation justifying it, and we have found none, we can only conclude that this districting is either the result of gerrymandering for partisan advantage, as was suggested at the hearing, or that it is wholly arbitrary and capricious." But Judge Maris' statement does not sanction the doctrine that political gerrymandering offends a right protected by the Constitution of the United States.
We reiterate that we can find no decision of the lower courts which clearly supports the plaintiffs' position in respect to purely political gerrymandering, as we have described it, and none has been cited to us.
We turn now to consideration of the opinions of the Supreme Court of the United States. Representation strictly according to population is a comparatively new concept. Before Baker v. Carr, 369 U.S. 186, 82 S.Ct. 691, 7 L.Ed.2d 663 (1962), the political effectiveness of an individual's vote was considered to be beyond the scope of constitutional protection. See, e.g., Mr. Justice Frankfurter's dissent in Baker v. Carr, supra, at 301, 82 S.Ct. at 755. Following Baker v. Carr the Supreme Court enunciated the rule of "one person, one vote" in Wesberry v. Sanders, 376 U.S. 1, 84 S.Ct. 526, 11 L.Ed.2d 481 (1964). In Reynolds v. Sims, supra, 377 U.S. at 565, 84 S.Ct. at 1383, 12 L.Ed.2d 506, the Supreme Court stated: "Full and effective participation by all citizens in state government requires, * * * that each citizen have an equally effective voice in the election of members of his state legislature. Modern and viable state government needs, and the Constitution demands, no less." But the Supreme Court in Reynolds v. Sims was not faced with an issue involving political gerrymandering.
In Fortson v. Dorsey, 379 U.S. 433, 439, 85 S.Ct. 498, 501, 13 L.Ed.2d 401, 402 (1965), Mr. Justice Brennan stated: "It might well be that, designedly or otherwise, a multi-member constituency apportionment scheme, under the circumstances of a particular case, would operate to minimize or cancel out the voting strength of racial or political elements of the voting population. When this is demonstrated it will be time enough to consider whether the system still passes constitutional muster. This question, however, is not presented by the record before us." The Supreme Court in the cited case was dealing, as stated by Mr. Justice Brennan, with "a multi-member constituency."
But in WMCA, Inc. v. Lomenzo, 238 F. Supp. 916, 925 (S.D.N Y 1965), the Three-Judge Court, ruling upon New York apportionment statutes, stated: "Plaintiffs, supported by Nassau County and New York City, claim with great vehemence that Plan A violates the XIV Amendment of the United States Constitution, because it allegedly gerrymanders the state for partisan political advantage.
"At the start of the hearing on January 21, 1965, we ruled that these attacks `do not raise questions under the Federal Constitution, and consequently we request of you that you devote none of your time in addressing us with reference thereto.'"
The decision of the Three-Judge Court in the cited case was affirmed per curiam by the Supreme Court, 382 U.S. 4, 86 S.Ct. 24, 15 L.Ed.2d 2 (1965), with a concurring opinion by Mr. Justice Harlan, id. at 5-6, 86 S.Ct. at 25-27. Mr. Justice Harlan stated: "In WMCA, Inc. v. Lomenzo, D.C., 238 F. Supp. 916, the three-judge court found that Plan A satisfied this order; in so doing it rejected contentions that apportioning on a basis of citizen population violates the federal Constitution, and that partisan `gerrymandering' may be subject to federal constitutional attack under the Fourteenth Amendment. In affirming this decision, this Court necessarily affirms these two eminently correct principles."
In Burns v. Richardson, 384 U.S. 73, 86 S.Ct. 1286, 16 L.Ed.2d 376 (1966), however, the Supreme Court passed on the Hawaii Legislature's interim reapportionment plan based on the number of registered voters rather than on total populations. It was alleged that the plan was designed to cancel out or minimize the voting strength of both racial and political elements, and the Supreme Court stated that no sufficient demonstration was made from the record that a multi-member reapportionment scheme effected invidious discrimination. Mr. Justice Brennan stated: "But the Equal Protection Clause does not require that at least one house of a bicameral state legislature consist of single-member legislative districts. See Fortson v. Dorsey, 379 U.S. 433 [ 85 S.Ct. 498, 13 L.Ed.2d 401]. Where the requirements of Reynolds v. Sims are met, apportionment schemes including multi-member districts will constitute an invidious discrimination only if it can be shown that `designedly or otherwise, a multi-member constituency apportionment scheme, under the circumstances of a particular case, would operate to minimize or cancel out the voting strength of racial or political elements of the voting population.' * * * Once the District Court had decided, properly, not to impose its own senate apportionment but to allow the legislature to frame one, such judgments were exclusively for the legislature to make. They were subject to constitutional challenge only upon a demonstration that the interim apportionment, although made on a proper population basis, was designed to or would operate to minimize or cancel out the voting strength of racial or political elements of the voting population.", and, finally, "Use of a registered voter or actual voter basis presents an additional problem. Such a basis depends not only upon criteria such as govern state citizenship, but also upon the extent of political activity of those eligible to register and vote. Each is thus susceptible to improper influences by which those in political power might be able to perpetuate underrepresentation of groups constitutionally entitled to participate in the electoral process, or perpetuate `a ghost of prior malapportionment.'" Id. at 88, 89 and 92-93, 86 S.Ct. at 1294, 1297. In the cited case the Supreme Court was concerned, as in Fortson v. Dorsey, with a multi-member constituency apportionment scheme. It is argued that the language of Fortson v. Dorsey and Burns v. Richardson, in contrast to that of WMCA, Inc. v. Lomenzo, by broad, or perhaps very broad interpretation is susceptible of being construed as looking to political gerrymandering as if it were a cognizable issue under the Fourteenth Amendment.
We have omitted the footnote references of Mr. Justice Brennan's opinion.
On November 21, 1966, the Supreme Court handed down a Per curiam opinion in "In the Matter of the Apportionment of the Michigan Legislature, Badgley, et al., appellants, No. 505, October Term 1966", an appeal from the Supreme Court of Michigan, 377 Mich. 396, 140 N.W.2d 436. Badgley is a Michigan elector. The appeal was taken pursuant to Section 1257(2), 28 U.S.C. Motions to dismiss or affirm were filed by the appellee, Hare, Secretary of State of Michigan, and others. The motions seem to have raised the issue as to whether political gerrymandering was cognizable under the Fourteenth Amendment. The Supreme Court dismissed the appeal, 385 U.S. 114, 87 S.Ct. 338, 17 L.Ed.2d 207 (1966), stating: "The motions to dismiss are granted and the appeal is dismissed for want of a substantial federal question."
A brief statement of the facts relating to the cited case is necessary here. After a prolonged political battle a new Constitution was adopted for Michigan, known as the Constitution of 1963. Article IV of this Constitution provided that a bipartisan commission of eight members should be set up to reapportion the State. The Commission could not agree and divided four-four on purely partisan lines. The 1963 Constitution, Article IV, Section 6, subparagraph (7), provided that if the Commission could not agree, various projected plans could be submitted by the Commission to the Supreme Court of Michigan for its approval and that any elector could object thereto. See subparagraph (8) of Section 6. One of the plans submitted was the Austin-Kleiner plan. This plan was ultimately approved by the Supreme Court of Michigan over an elector's objections on March 8, 1966. The appeal to the Supreme Court of the United States followed.
An examination of the jurisdictional statement, the motions and Badgley's brief submitted to the Supreme Court of the United States seems to demonstrate that the judgment of the Supreme Court of Michigan appealed from was a final judgment and that the issue of gerrymandering was raised albeit the Supreme Court of Michigan would not allow discovery as to gerrymandering, treating the allegations of gerrymandering as "conclusory" and indeed as irrelevant to any issue of constitutionality. The history of Michigan reapportionment is indeed complex. See the earlier cases relating to this subject: 376 Mich. 410, 138 N.W.2d 16; 376 Mich. 410, 137 N.W.2d 495; 373 Mich. 250, 128 N.W.2d 721 (2 cases), and 372 Mich. 418, 126 N.W.2d 731. Nonetheless it seems that the decision of the Supreme Court of the United States of November 21, 1966, in the Michigan reapportionment case, when considered in conjunction with its decision in WMCA, Inc. v. Lomenzo, requires the conclusion that political gerrymandering as defined in this opinion is not cognizable under the Fourteenth Amendment. We so hold.
It is necessary to add, however, that in view of the absence of an express declaration by the Supreme Court that political gerrymandering is not cognizable as an issue under the Fourteenth Amendment, we are not certain of the correctness of our position and accordingly we have made findings of fact as to gerrymandering as set out in this opinion. If we are in error in our ruling, the necessary findings of fact will be available to the Supreme Court so that our mistake, if it be such, can be corrected on appeal.
XIII. The Maps as Required by Section 661 of S.B. 336
We have previously quoted Section 661 in respect to its requirements, subparagraph (2), that maps be prepared and filed by the Departments of Elections showing the boundaries of "the several General Assembly districts," and, subparagraph (b), that the "Registered Land Surveyors who prepare the maps" should "certify them in the Recorder of Deeds offices in the respective counties for which they were prepared." Subparagraph (b) also contains the declaration: "Upon such recording the maps shall be prima facie evidence in all judicial proceedings as to the correctness of the boundaries." It is stipulated by the parties, Document 174, that "[T]he respective Departments of Elections have caused to be prepared by registered land surveyors maps of each county which purport to show the boundaries of the several General Assembly districts," and that "The registered land surveyors who prepared the maps have certified them as to their correctness, and the respective Departments of Elections have caused the maps to be recorded in the Recorder's Office in the County in which they were prepared in compliance with the provisions of Subchapter VI, Section 661 of Senate Bill 336. The parties agree that these maps should be marked into evidence in this case."
The parties have stipulated that the maps prepared "on behalf of the New Castle County Department of Elections including Wilmington, correctly and accurately reflect the boundaries of the representative districts and the senatorial districts for that area." This portion of the stipulation will be approved to the extent indicated at a later point in this opinion and an order to such effect will be entered. We find the stipulation to be in accordance with the facts insofar as New Castle County as a whole is concerned. The political areas specified on the maps set out in note 138 are clearly recognizable and present no problem of identification, but as soon as we leave the City of Wilmington and Rural New Castle County and enter Kent and Sussex we encounter difficulties as will appear hereinafter.
Document 174, at 7.
See "Map Stipulation", Ex. 11, map of Wilmington, showing representative districts, Wilmington.
"Map Stipulation", Ex. 12, map of Wilmington, showing senatorial districts, Wilmington.
"Map Stipulation", Ex. 10a through 10h, inclusive (8 maps), New Castle County, showing representative and senatorial districts, as follows:
10a shows the 9th and 10th representative districts and the 5th senatorial district.
10b shows the 11th and 12th representative districts and the 6th senatorial district.
10c shows the 13th and 14th representative districts and the 7th senatorial district.
10d shows the 15th and 16th representative districts and the 8th senatorial district.
10e shows the 17th and 18th representative districts and the 9th senatorial district.
10f shows the 19th and 20th representative districts and the 10th senatorial district.
10g shows the 21st and 22nd representative districts and the 11th senatorial district.
10h shows the 23rd and 24th representative districts and the 12th senatorial district.
Compare note 38, supra.
Document 174 also provides that "They [the parties] further stipulate that the maps prepared for the Departments of Elections of Kent and Sussex Counties represent conscientious efforts on the part of the Departments of Elections of those Counties to present as accurately as possible on a single county map the District boundaries in accordance with the Representative and Senatorial lines employed by the Departments in the last five General Elections. Several of these Election Districts and Representative Districts are defined by map or written description only in terms of the boundaries of the traditional `Hundreds' in both of these counties. In many instances, unambiguous descriptions of the hundreds do not exist; and it is impossible to determine in Sussex County the exact location of the boundary markers and lines which comprise the Representative and Senatorial Districts as they `heretofore' existed.
"With regard to the Sussex County map, it was determined that the Department of Elections had heretofore had a map which contained a significant deviation in boundaries between the present 34th and 35th Representative Districts. This deviation was continued in the map dated May 1, 1965, prepared by Messrs. Simpson and Burton, had no significance with regard to population and has been corrected in the map which they prepared and which is dated June 29, 1965.
"It is further stipulated that after these corrections, the Representative District boundaries on these maps correspond to the boundaries used in conducting the last five General Elections for the corresponding Election Districts.",
"Map Stipulation", Ex. 14, map of Kent County, showing senatorial districts.
"Map Stipulation", Ex. 13, map of Kent County, showing representative districts.
"Map Stipulation", Ex. 9, map of Sussex County, showing senatorial and representative districts (single map).
The maps of Wilmington and of New Castle County show the populations per representative district and senatorial district, as demonstrated by the records of the Department of Elections of New Castle County and as shown in the office of the Prothonotary. The maps of Kent and Sussex Counties do not show population figures. This is because the parties were unable to stipulate as to the population figures in some districts.
See note 38, supra.
Document 174, at 7-8.
In respect to the maps of Kent County and Sussex County purporting to show the representative and senatorial districts we think it will aid the reader if we repeat in part what we stated under heading "I." Turning to the descriptions of the districts set out in Sections 641, 642, 651 and 652 of S.B. 336, we find the frequent use of the specific language that a representative district or a senatorial district "shall comprise" certain of the "heretofore existing and constituted" representative districts of Kent County as designated. In respect to Kent County, Section 641(a) is typical and provides, " Twenty-fifth Representative District: The twenty-fifth Representative District shall comprise all of the heretofore existing and constituted First, Third and Fourth Representative Districts of Kent County." This formula is repeated again and again throughout Section 641 in respect to representative districts in Kent County, and in Section 651 in respect to the representative districts in Sussex County. The senatorial districts in Kent County and Sussex County, as created respectively by Sections 642 and 652, are composed of "the heretofore existing and constituted" representative districts of Kent and Sussex County respectively, except where smaller units are used. Section 642(b), creating the 14th senatorial district in Kent County is typical of Kent County creation and is as follows, " Fourteenth Senatorial District: The Fourteenth Senatorial District shall comprise all of the heretofore existing and constituted Fifth and Seventh Representative Districts of Kent County and all of the heretofore existing and constituted Second Election District of the Second Representative District of Kent County." The formula in Sussex County for the creation of representative and senatorial districts is the same.
Court's Ex. 16 is the brochure entitled "Delaware Election District Boundaries", issued by Mr. Ernest E. Killen, who was the Commissioner of Elections in 1958. It is designated as an "Appendix", at 233 entitled: "Election District Boundaries" and it states that "Under authority of § 4103, Title 15, Delaware Code, each Department of Elections is granted authority to divide election districts which contain a greater number of voters than can conveniently vote in such district." The brochure goes on to say that the "[F]ollowing list gives the boundaries of the various election districts as of February 1, 1958." First comes "New Castle County — Wilmington". Next comes Kent County commencing at 311 of the brochure followed by Sussex County commencing at 322. The references are, of course, to old election districts and to old representative and senatorial districts.
The present Delaware Code Annotated was enacted by the General Assembly and was approved by the Governor on February 12, 1953. Page V, Publisher's Preface.
The Departments of Elections of Kent and Sussex Counties, pursuant to the authority of Section 4103 of Title 15 of the present Delaware Code Annotated, have proceeded to recast some of the election districts in Kent and Sussex Counties. The plaintiffs, on their post-trial brief, state: "With respect to the maps required to be prepared by the respective departments of elections, maps have been prepared for the City of Wilmington, New Castle County, Kent County and Sussex County. With respect to Sussex County, several of the election districts and representative districts are defined only in terms of the boundaries of the traditional `hundreds' and it has been impossible to determine in Sussex County the exact location of the boundaries, markers and lines which comprise the representative and senatorial districts as they `heretofore existed'."
Plaintiffs' post-trial brief at 7-8.
The parties stipulated in respect to Sussex County that while the Department of Elections of Sussex County heretofore had a map prepared by Messrs. Simpson and Burton, registered surveyors, on May 1, 1961, which contains a significant deviation in boundaries between the 34th and 35th representative districts, this deviation had no significance with regard to population and was corrected in the map which Messrs. Simpson and Burton prepared, dated June 29, 1965. But in respect to the old election districts and old representative districts of Kent and Sussex Counties, the stipulation candidly states that several of these election districts and representative districts are defined "by map or written description" only in terms of the boundaries of "hundreds", seemingly as they existed in 1953, in these counties, and that unambiguous descriptions of some of the hundreds of Kent County and Sussex County do not exist, and that, therefore, in some instances it was impossible to determine the exact location of boundary markers and lines of some of the representative districts as they "heretofore existed". Despite the stipulation that the difference in population between the 34th and 35th representative districts in Sussex County is minimal, which we find to be correct, our finding cannot render vague district lines less vague and indefinite.
Document 189.
This map is that referred to in note 38, supra, as "`Map Stipulation', Ex. 9, Map of Sussex County."
It must be borne in mind that in respect to Kent and Sussex Counties no use was made of census enumeration tracts. The reason given for this, and in support of the apportionment created by S.B. 332 and S.B. 336, was that there was insufficient time to rearrange the election districts according to census enumeration districts in view of the short period before the oncoming primary election and the general election of 1964. But the stipulation in respect to the election districts and representative districts in Kent and Sussex Counties, and, therefore, necessarily, the senatorial districts of these Counties was vague and ambiguous. Following a study of the record in November 1965 a question presented itself to the court as to whether the contents of the stipulation were such as to overcome the statutory prima facie presumption that the maps of Kent and Sussex Counties filed pursuant to Section 661 were valid. The court was of the opinion that since the geographic boundaries of certain General Assembly districts in Kent and Sussex Counties were described as "heretofore existing" by Sections 641, 642, 651 and 652 of S.B. 336, and the boundaries of several hundreds were described by ancient landmarks, Court's Ex. 16 describing election districts, now non-existent or so vague as to be presently unascertainable, that it followed perhaps that many of the General Assembly districts in Kent and Sussex Counties were described so indefinitely as to render their apportionment constitutionally invalid. See Jordan v. De George, 341 U.S. 223, 71 S.Ct. 703, 95 L.Ed. 886 (1951), Adler v. Board of Education of the City of New York, 342 U.S. 485, 72 S.Ct. 380, 96 L.Ed. 517 (1952), and American Communications Association, CIO v. Douds, 339 U.S. 382, 70 S.Ct. 674, 94 L.Ed. 925 (1950).
Having these possible imperfections of the record in mind, on December 20 and 21, 1965, as stated at an earlier point in this opinion, the court had a further hearing respecting the maps required by Section 661 of S.B. 336, and numerous witnesses who had previously testified were recalled. Additional witnesses were heard and additional exhibits were put in the record, as appears from the transcript.
See transcript of hearing of December 20-21, 1965, Document 192, 2224-2551, and index to witnesses and exhibits.
Dr. C. Harold Brown was recalled to the stand and testified that he had made available to Mr. McGinnes copies of maps and actual counts of populations as shown by census enumeration districts of the State of Delaware. He testified in particular that the old representative districts were treated by the United States Bureau of the Census as minor civil divisions and that no census enumeration district boundary crossed the boundaries of old representative districts. He made it clear that he himself had prepared a set of maps showing the census enumeration district boundaries in each of the three counties and had given these to Mr. McGinnes. Mr. McGinnes, when recalled as a witness, testified that he had turned the maps over to Mr. Raymond West and to Mr. Maurice Hartnett, a member of the Bar of Delaware, who has not been identified adequately on this record. Further identification is unnecessary, however, as will appear hereinafter. A map showing the census enumeration districts in Sussex County was admitted into evidence. Maps showing the Kent County census enumeration districts and the Sussex County census enumeration districts were admitted at the trial in the original case. There was also introduced in evidence photographs of pertinent portions of a Beer's Atlas of the year 1868 and a small photographic reproduction of a large map which was in the Kent County surveyor's office.
Brown, 2236.
Defendants' Ex. 14 admitted at 2248. See also Defendants' Ex. 18, "Geographic Identification Code Scheme, Delaware, District of Columbia". In the record upon which our decision in Sincock v. Duffy, 215 F. Supp. 169 (1963), was based, a map showing the Kent County enumeration districts was introduced as "CX 5", and a map showing the Sussex County enumeration districts was introduced as "CX 6" and, of course, are presently available here.
Brown 2272, et seq. The testimony of C.C. Brown, the registered land surveyor, should not be confused with that of Dr. C. Harold Brown who took the stand just prior to C.C. Brown at the hearing of December 20, 1965. See Court's Exs. 45 and 46.
Mr. C.C. Brown, a registered land surveyor who prepared the Stipulation Exs. 13 and 14, Maps for Kent County, testified that he was able generally to discern the General Assembly districts though many difficulties confronted him when he attempted to ascertain old representative district boundaries. It will be observed that the Beer's Atlas maps are studded with family names indicating houses and farms, which Mr. Brown candidly admitted made the running out of old representative district lines very difficult. He testified nonetheless in substance that while the task was not an impossible one there were some areas which he was unable to ascertain with complete definiteness. We point, for example, to the map description "K-18" designated as being in South Murderkill Hundred, Court's Ex. 16, which was anciently "Murderkill Hundred" and at some date which does not appear from the record or from any other source which we have been able to ascertain, was divided into North Murderkill Hundred and South Murderkill Hundred. It should be borne in mind as will be seen from the Constitution of the State of Delaware of 1897, Article II, Section 1, Const.Del.C. Ann. Vol. 1 that the Hundreds played a large part in the apportionment. The same is true of prior Constitutions of Delaware the provisions of which for the sake of brevity we will not cite here. The difficulty in locating census enumeration tracts, old representative districts and hence old senatorial districts, and Hundreds, was well exemplified by Mr. Brown when he stated: "I think I do know that the former Representative and Senatorial Districts, Hundreds of Kent County, have been set up as of at least 100 years ago from common knowledge, from my knowledge of research of deeds in the Kent County Court House I know that Kent County has been laid out in these Hundreds since before the Revolution. Some of the Hundreds have been subdivided in later years." The court takes judicial notice of these facts.
See note 38, supra.
Brown, 2275.
Brown, 2281-82.
Mr. L. Winfred Hughes, recalled as a witness, testified that he was with the Department of Elections in Kent County for eight years and that within that time became familiar with the representative and the election districts of that county; that the Department of Elections for many years had had a map giving the boundary lines of all of the election districts and which had been used by the Department of Elections of Kent County and, to the witness' knowledge, had never been in dispute. He testified: "In setting up your Election Districts, if there is anyone who wants to know where they live and in what Election District they determine it by that map, and it has been in use for many years." He stated that when he created the Kent County apportionment he used the map referred to in combining election districts to form the new representative districts and that he received his information as to population from the published pamphlet of the United States Bureau of Census. Mr. Hughes was then shown the maps of Kent County, Stipulation Exs. 13 and 14, and was asked to state whether he could tell from an inspection of these exhibits whether the new General Assembly districts embodied in S.B. 336, as shown by the maps, were formed by combining old election districts. The question was objected to and the objection was sustained. Mr. Hughes was then asked if he could inform the court as to what were the populations of the General Assembly districts as shown on the Stipulation Exs. 13 and 14. He said that he could do so and that his figures were derived from the 1960 census. Using the new 25th representative district as an example, he testified that this district consisted of three complete old representative districts in which the total population is given by the 1960 census, though necessarily he had made estimates as to the populations of other districts as previously indicated in this opinion. He said in substance that he could ascertain all necessary geographical boundaries of the new General Assembly districts by examining the map in the office of the Department of Elections of Kent County. A map of the old Kent County election districts was produced in court, Defendants' Ex. 6, and Mr. Hughes testified that this exhibit was identical with the map in the possession of and "hanging on the wall" of the Department of Elections of Kent County in Dover. Mr. Hughes testified that he himself drew the original maps for the new General Assembly districts in Kent County and, after examining Stipulation Exs. 13 and 14 again, he stated that the maps agreed exactly in respect to "lines" with the map used by the Kent County Department of Elections for many years. Mr. Hughes' testimony is well tied in to Stipulation Exs. 13 and 14, and we find his statements to be correct.
Hughes, 2285, et seq.
For this pamphlet, see Plaintiffs' Ex. 2 to the Deposition of Dr. Paul Dolan. Record in Sincock v. Duffy, 215 F. Supp. 169 (1963). See note 148, supra. See also Defendants' Ex. 18.
Defendants' Ex. 6 also is part of the record upon which our decision in Sincock v. Duffy, 215 F. Supp. 169 (1963), was based. Defendants' Ex. 6 is the same as Joint Ex. 1, p. 2296.
Hughes, 2334-35.
Registered Surveyor Brown was subjected to a searching cross-examination as to the Kent County districts in respect to the election districts in Kent County which were not susceptible of precise definition. He indicated that the description of "K-3A" on Court's Ex. 16, the brochure entitled "Delaware Election District Boundaries," issued by Commissioner of Elections Killen, was not clear and that there was "a variance of a few hundred feet". As to the description of "K-10" Mr. Brown indicated that where Court's Ex. 16 referred to "State House" there was a typographical error and it should be "Station House". As to "K-16" he was in doubt as to the exact locality of "Beaver Dam Branch", a very small part of the boundary; and that the descriptions of "K-18" and "K-19" of Court's Ex. 16 were clear except as to a lane between the lands of Amos Cole and Benjamin Harrington, again a very small portion of a boundary.
Brown, 2308.
Brown, 2310.
On redirect examination Mr. Hughes was examined as to certain of the areas which could not be exactly ascertained. He reiterated his testimony that Stipulation Exs. 13 and 14 possessed, to the best of his knowledge, the precise geographical lines as those on the map used in the Kent County Department of Elections for many years. He was examined as to individual areas on the maps and, as an example, in regard to the boundary of a disputed area that involved the point at which the Delaware Division of the Pennsylvania Railroad crossed Murderkill Creek. The dispute concerned which branch of the creek the boundary line of the old 1st election district of the old 8th representative district followed. Mr. Hughes testified that perhaps four to six families were involved in the disputed area. All disputed areas seem small and apparently involve minimum populations and it appears affirmatively from Mr. Hughes' testimony that the employees of the Bureau of the Census working in Kent County made a copy of the map in the possession of the Department of Elections of Kent County. Mr. Hughes stated: "I assume [that they, the Census Bureau employees] used [that map] in gathering the census for each Representative District."
Hughes, 2235-37.
Hughes, 2344.
Mr. C.C. Brown, the surveyor, supplemented Mr. Hughes' testimony, after an examination by him of Plaintiffs' Ex. 39, the Bureau of the Census enumeration maps for Kent County, and Joint Ex. 1, the Maintenance Map of Kent County showing the old election districts as they existed under the 1897 Delaware Constitution. He testified that with three exceptions, Joint Ex. 1 coincided with the census enumeration districts as shown on Plaintiffs' Ex. 39; i.e., he testified in substance that the census enumeration district populations were allocated with geographical correctness to the Kent County maps as required by Section 661 of S.B. 336, with but three exceptions. Mr. Brown testified in some detail respecting these exceptions, stating that he found them to be "[A]n enumeration area indicated as 1-1, another area 1-2, and an area here, 1-S5 but [said] that the boundaries of these areas or their total boundaries coincide with the boundaries of the old first [representative] district bounded by New Castle County on the north, taking in the Town of Smyrna, going a bit to the west of Smyrna and southerly to Duck Creek. Therefore it is apparent that these three enumeration districts I have just mentioned plus the Town of Smyrna do comprise the old Representative District No. 1". He went on to say in respect to the three differences between the maps: "The one major difference is that on the Census Enumeration map the Town of Smyrna was indicated on the map as a whole, as a unit, apparently as an enumeration district. The City of Dover was similarly portrayed. Whereas * * * on * * * Joint Exhibit No. 1, in some cases, such as Dover, the old 5th [representative] district and the old 2d [representative] district, the dividing line is shown." He pointed out, however, that no similar dividing line was shown on Plaintiffs' Ex. 39. Making use, however, of Plaintiffs' Ex. 2 to Dr. Dolan's deposition, Document 29, upon examination and comparison it is apparent that the enumeration districts on Plaintiffs' Ex. 39 do coincide as Mr. Brown put it, "with certain or all or each of the old Representative Districts". Mr. Brown testified in substance, as we understand the tenor of his evidence, that the small divergences were of no substantial significance. As triers of the fact we reach the conclusion that substantially everyone included in the census enumeration districts was properly allocated to General Assembly districts of Kent County and that but few, if any, persons entitled to inclusion therein were excluded therefrom.
Article II, Section 2.
Brown, 2479.
We state again that this document was before us in the record on which we based our decision in 215 F. Supp. 169 (1963) and, is, of course, part of the record in this case.
Brown, 2481-84.
Surveyor Brown testified that he found three differences between Plaintiffs' Ex. 39, the map showing census enumeration districts, and Joint Ex. 1, the map showing the old representative districts in Kent County. The first of the differences concerns the town of Smyrna which was indicated on the census enumeration map, Plaintiffs' Ex. 39, as a unit, "as an Enumeration District". This difference is of no consequence, however, since the town of Smyrna lies entirely within the new 25th representative and the new 13th senatorial district.
The second difference concerns a small triangle of land near McNett's Corner which was part of the boundary line between the old 2d election district of the old 6th representative district and the old 1st election district of the old 9th representative district. This triangle appears on Plaintiffs' Ex. 39 to be in a different census enumeration district than as is allocated on Joint Ex. 1. But this divergence also is of no consequence because the land in question is included entirely within the new 29th representative district and the new 13th senatorial district.
The third difference concerns the City of Dover which, similar to the Town of Smyrna, is treated "as an Enumeration District" on Plaintiffs' Ex. 39, whereas on Joint Ex. 1 the City of Dover is divided between the old 5th representative district and the old 2d representative district. However, Surveyor Brown assumed this did not destroy the correctness of the population estimates. He stated: "We get down to the division between the old Fifth Representative and the old Second Representative and the tabulation gives a population figure for the old Fifth apparently exclusive of that part of it within the City of Dover, but below it adds a part of the City of Dover, certain population. [Sic.] Therefore I am assuming that we can correctly say that a combination of certain Enumeration Districts on Exhibit 39 does coincide with certain or all or each of the old Representative Districts."
Brown, 2483.
This assumption would seem to be justified by the record. Population figures for the City of Dover are broken into two parts by Plaintiffs' Ex. 2 to Dr. Dolan's deposition, Document 29. According to the 1960 census the part of Dover which constituted a part of the old 2d representative district contained a population of 1317 persons, and the remainder of Dover which constituted a part of the old 5th representative district contained 5933 persons. Joint Ex. 1, setting out the old representative districts, shows the City of Dover as split, as does Stipulation Ex. 13. There is also a map, page 9-3 of Plaintiffs' Ex. 2 to Dr. Dolan's deposition entitled, "Delaware-Minor Civil Divisions Representative Districts" which shows the division of the City of Dover between the old 2d representative district and the old 5th representative district. The division line between those two districts as drawn upon the map last referred to seems to correspond with that shown on Joint Ex. 1. There are, therefore, population figures and maps which seem to support Surveyor Brown's assumption. It would appear, therefore, that the census enumeration districts shown on Plaintiffs' Ex. 39 correspond with the old representative districts on Joint Ex. 1, and we so find.
We observe that on Plaintiffs' Ex. 39 the population of Dover is stated as being 7250 persons which is in accord with the 1960 census, but that on Stipulation Ex. 13 the estimated population of Dover is given as 12000 inhabitants. The stipulation maps are dated 1963.
Mr. Raymond West, recalled to the witness stand, stated that he had prepared a map, Joint Ex. 3, showing the old representative districts of Sussex County and that this corresponded in all respects with the official map of the Department of Elections of Sussex County made by a surveyor of the State Highway Department. Mr. West said that "That [Joint Ex. 3] * * * [was] an exact copy. I traced it line by line for all Election Districts. It shows all the 26 Election Districts of Sussex County." He was then asked if he knew of his own knowledge whether or not the lines shown on the map had been used in the past to determine where a person was eligible to vote. He testified that he knew of but one case in which the map was so used and that case involved "the only dispute" that he knew of. He also stated that the last change in the official map, i.e., Defendants' Ex. 17, the map used by the Department of Elections of Sussex County, had been made eight or ten years before. He also testified that he had received his working data and a map from Mr. McGinnes and had gotten the census enumeration district figures from the Department of Urban Affairs of the University of Delaware. He also stated that the census enumeration district figures, or as he put it, the "tracts" "dovetailed" with the old representative districts.
West, 2408-09.
West, 2409. The change was a very minor one.
West, 2413.
Mrs. Frances Hitchens testified that she was the chief clerk of the Department of Elections of Sussex County and had been employed as such for about ten years. She produced a map which was the map used by the Department of Elections of Sussex County which had been enclosed in glass in a frame and hung on the wall of the office of the Department of Elections and she stated that this map showed the old election districts. This map was introduced as Defendants' Ex. 17 and Mrs. Hitchens testified that the map was identical with Joint Ex. 3 and was accurate with but one small exception, i.e., that a line was shown between the 4th election district of the old 3rd representative district and the 3rd election district of the old 3rd representative district which should not have been there, and that she had had put on the glass covering the map a correcting line with crayon. Mrs. Hitchens testified that the line should have followed the line of the Delaware Division of the Pennsylvania Railroad thus dividing the Town of Seaford in such a way as to indicate that a part of the 4th election district of the old 3rd representative district was within the town. This correction has no bearing on the new representative and senatorial districts since the two lines are wholly within the new 31st representative district and the new 16th senatorial district. Mrs. Hitchens' evidence makes it clear that in her view there were no substantial deviations in General Assembly districts in Sussex County as demonstrated by maps in evidence.
Hitchens, 2438-41.
Hitchens, 2441-42.
See Stipulation Ex. 9.
Mr. Perry T. Burton, a registered surveyor and an engineer employed by the Delaware State Highway Department testified that he with Mr. Edward L. Simpson, another registered surveyor, had made up the map, Stipulation Ex. 9. He testified that he used a map of the old election districts, the Beer's Atlas of 1868 and Court's Ex. 16, the brochure describing the Delaware election district boundaries as issued by Mr. Ernest E. Killen. He testified that he and Mr. Simpson employed S.B. 336 and other maps of Sussex County, in particular Joint Ex. 3. Mr. Burton testified that difficulties were encountered in finding representative district boundaries. Mr. Burton stated: "There were, I think, three locations in which we had some difficulty in establishing the boundary lines. We found no place where these Hundred lines were defined." These three discrepancies are discussed immediately hereinafter.
Actually Perry, stated we believe erroneously in the record as Harry. See Burton, 2348.
Burton, 2350.
The first vague area is just south of Cool Spring in the new 35th representative district, approximately eight miles east of Georgetown, the County Seat of Sussex County. Here exists confusion as to which side of Route 291 the disputed area lies. The surveyors again made use of Beer's Atlas but it threw small light upon the subject. The area was estimated to consist of approximately six acres and to contain some "30 people, seven or eight families".
A second area of vagueness involves the head of Sowbridge Branch, involving two small streams running into Sowbridge Branch. The surveyors testified that there was difficulty in ascertaining the dividing line between the new 30th and the new 35th representative districts, but the area involved is a portion of what is known as Ellendale Forest, sometimes called "Great Ellendale Swamp". Mr. Burton stated that there were no persons living in the disputed area, that it was marshland and most of it was State Forest.
A third area of vagueness allegedly arises in respect to a small area between the new 33rd and the new 35th representative districts. There was a suit in the Superior Court of Sussex County respecting the election of a member of the Levy Court of Sussex County, as to whether or not the elector was entitled to run from the district from which he was apparently elected. Only two persons were involved: the plaintiff in the suit and his wife. The question was never settled and the suit was dropped. The record does not demonstrate the area, or the number of people involved, and Mr. Burton gave as his testimony that there was no vague area here. We accept his conclusion.
The Levy Courts of the respective counties of Delaware are administrative bodies with limited power. See 9 Del.C. Ann. §§ 301-79.
Burton, 2350-61.
The fourth difference appears upon comparing the map designated as Stipulation Ex. 9 with Joint Ex. 3. It is not clear from the maps as to where the boundary, approximately one mile long, runs between the old 9th and old 8th representative districts along Route 30. We take judicial notice of the fact that the boundary in dispute is approximately four miles east of Georgetown, in deep farming country, and the number of persons involved is miniscule.
See also the testimony of Surveyor Simpson, 2396-2400.
The complete picture of the vague areas of General Assembly districts of Sussex County was summed up by Mr. Burton in response to a question asked by the court. He stated that the number of inhabitants involved did not exceed 32 persons and we find that his estimate is correct.
As we have stated, the stipulation maps filed pursuant to Section 661 of S.B. 336, pursuant to the terms of subsection "(b)" when recorded "shall be prima facie evidence in all judicial proceedings as to the correctness of the boundaries" of the General Assembly districts. We have also stated that S.B. 336 provides that certain new representative districts shall comprise former representative districts as "heretofore existing and constituted". See, for example, the 27th representative district in Kent County as defined by Section 641(c). As we have said, the Constitution of 1897, Article II, Section 2, Del.C. Ann. describes the original General Assembly districts and some of these have been changed as set out in Court's Ex. 16, "Delaware Election District Boundaries" issued by Commissioner of Elections Ernest E. Killen. We reiterate that the appendix to the brochure states the authority of each of the Departments of Elections of Delaware "to divide election districts which contain a greater number of voters than can conveniently vote in such districts". The authority lies in Section 4103, Title 15, Del.C. Ann. The brochure lists and describes the General Assembly districts in the three Counties of Delaware. We need not go back of the authorities cited. We conclude that, with the comparatively minor exceptions hereinbefore discussed under this heading, the requirements of law have been met, and the few vague areas are not sufficiently important to invalidate the map provisions of S.B. 336. See Jordan v. De George, supra, and the authorities cited with it. Moreover, every state statute is clothed with the presumption of constitutional validity. Davis v. Department of Labor, 317 U.S. 249, 63 S.Ct. 225, 87 L.Ed. 246 (1942); Salsburg v. State of Maryland, 346 U.S. 545, 74 S.Ct. 280, 98 L.Ed. 281 (1954). The plaintiffs have failed to overcome that presumption.
We cannot, however, view the situation presented by the maps with equanimity. It is clear from the evidence that the starting point of the registered surveyors in the preparation of the maps is the Beer's Atlas of 1868, and that they had substantial difficulties in many instances in establishing General Assembly district lines because many of the landmarks from which the lines ran no longer existed in fact and had to be developed from local recollection. The situation created is not unlike that which existed in ancient England where some boundaries were established by asking in respect to them: "What do old men say when seated around the fires at night?" Our point is amply demonstrated by the cross-examination of the registered surveyors by plaintiffs' counsel and the substantial difficulty which this court has had in ascertaining the existence of election districts in respect to the General Assembly districts as set up by S.B. 332 and S.B. 336.
We think the citizens of Kent County and Sussex County deserve better of the State of Delaware than this. Our State is a forward-looking and prosperous one. It is our view that a survey should be made of Kent and Sussex Counties and that clearly defined boundaries of General Assembly districts should be established so that our individual citizens' right of franchise may be clearly established. Moreover, a census of the two lower Counties, allocated to the survey, would be of great assistance in making sure that the fundamental principle established by the Supreme Court of "one person, one vote" is not impaired or abrogated.
XIV. Issue as to the Constitutionality of the Failure of S.B. 332 and S.B. 336 to Provide for the Continuation of the Terms of Certain Members of the Senate of the General Assembly of DelawareThree Senators are involved in this issue. All three were elected to the 122d General Assembly of the State of Delaware at the general election of 1962 for a four-year term pursuant to Section 2, Article II of the Constitution of Delaware of 1897, as amended, Del.C. Ann. Const. The three senators involved are: Senator Evelyn M. Lord, elected from the old 1st senatorial district of New Castle County, Senator J. Donald Isaacs, elected from the old 8th senatorial district of New Castle County, and Senator Walter H. Simpson, elected from the old 3rd senatorial district of Kent County. The Senate of the 123rd General Assembly refused to seat them.
It is possible that the issue is moot as to Senator Lord for we take judicial notice of the fact that she has moved out of the State of Delaware, but since the facts relating to her status are the same as those relating to those of Senator Isaacs and of Senator Simpson, our ruling in respect to the two last named senators will be equally apposite to Senator Lord's status, if her case is not moot.
The parties have stipulated that each of the three Senators has consistently claimed his right to office and has done nothing to waive that right. Paragraph one of Section 2, Article II provides that the terms of members of the Senate shall be for four years. We find that each of the three Senators was elected, was certified as elected, and upon taking the oath of office was seated as a member of the Senate of the General Assembly of Delaware pursuant to Section 8, Article II of the Constitution of Delaware of 1897. All three Senators voted against the enactment of S.B. 332 and S.B. 336.
There is no specific provision in the Constitution of Delaware or the laws of Delaware relating to the qualifications of members of the Senate but Section 3 of Article II, which provides that "No person shall be a Senator who shall not have attained the age of twenty-seven years and have been a citizen and inhabitant of the State three years next preceding the day of his election and the last year of that term an inhabitant of the Senatorial District in which he shall be chosen, unless he shall have been absent on the public business of the United States or of this State."
See Court's Exs. 14 and 15.
The plaintiffs point to Section 9, Article II of the Constitution of Delaware and state that that section provides the only procedure pursuant to which either house of the General Assembly may "expel" duly qualified members. Section 9 also provides that each House of the General Assembly may not only expel a member but "shall have all other powers necessary for a branch of the Legislature of a free and independent State." The plaintiffs also call our attention to Section 4 of Article XV of the Delaware Constitution which provides that "No law shall extend the term of any public officer nor diminish his salary or emoluments after his election or appointment." This provision has been interpreted in Delaware and in other states as intended to safeguard the terms of public officers whether they hold office under Constitutional provisions or not, and to deny to a legislature the power to deprive an incumbent of an office to which he has been duly elected. Cf. du Pont v. Green, 8 W.W.Harr. 566, 571, 195 A. 273, 114 A.L.R. 1184 (1937), and in particular the authorities cited at 276. Compare State ex rel. Biggs v. Corley, Lieutenant Governor, 6 W.W.Harr. 135, 146, 172 A. 415, 420 (Court in Banc of Delaware, 1934). We point out that the three Senators with whom we are concerned were not expelled from the General Assembly, and we think it is obvious that the expulsion provisions of Section 9 relate only to expulsion for misconduct.
We agree with the plaintiffs that the Constitution of Delaware, under the authorities cited, secured the tenure of those three Senators insofar as their tenure could be made secure by provisions of the Delaware Constitution, but we have here an adjudication by the Supreme Court of the United States holding that the provisions of the Delaware Constitution pursuant to which the three Senators were elected were unconstitutional because they were in violation of the Fourteenth Amendment. In Roman v. Sincock, supra, 377 U.S. at 708, 84 S.Ct. at 1457, 12 L.Ed.2d at 629, Mr. Chief Justice Warren stated: "Neither of the houses of the Delaware General Assembly, either before or after the 1963 constitutional amendment, was so [i.e., constitutionally] apportioned." We hold therefore, as we did in Sincock v. Duffy supra, 215 F. Supp. at 191, that senators elected pursuant to the provisions of Section 2, Article II of the Delaware Constitution as it existed prior to the 1963 amendment were de facto members of the General Assembly and it follows necessarily that the three Senators with whom we are concerned were de facto officers. If the provisions of the Constitution under which the three Senators were elected had not been held invalid as unconstitutional and S.B. 332 and S.B. 336 had not been enacted, their terms would have expired on the first Tuesday of January, 1967, i.e. on January 3, 1967.
Section 4, Article II, Constitution of Delaware of 1897, as amended, Del.C. Ann. Const.App.
The second paragraph of Section 3 of the "Schedule" appended to the 1897 Constitution of Delaware provides staggered terms for Senators. Del.C. Ann. Const. at 345.
We note that the 1963 Amendments to the Constitution of Delaware provided the first Senators elected from the Sixth Senatorial Districts of Kent and Sussex Counties should serve for a two-year term only, thereafter their successors should serve for a full four-year term.
Section 602 of S.B. 336 provides staggered terms for senators.
The 122d General Assembly, however, following the adjudication by the Supreme Court that Section 2, Article II of the Constitution of Delaware was unconstitutional as it existed both before and after amendment, passed S.B. 332 and S.B. 336 and created the apportionment under which members of the General Assembly were elected at the 1964 General Election. The new statutes, however, made no provision for the continuation in office of the three Senators, and the 123rd General Assembly refused to seat them. The General Assembly under the Constitution of Delaware is the judge of the qualifications of its members. In State ex rel. Biggs v. Corley, supra, 6 W.W.Harr. at 148-53, 172 A. at 422-23, the Court in Banc of Delaware so held. Section 3, Article II of the Delaware Constitution was not held to be invalid under the Fourteenth Amendment for it presents no instance of invidious discrimination. Its provisions are separable from those of Section 2 of Article II. See 2 Sutherland, Statutory Construction, Sections 2403-04. Nor could the fact that the 122d General Assembly enacted S.B. 332 and S.B. 336 affect the rights of the three Senators to hold their offices until the end of their terms on January 3, 1967, were it not for the overriding considerations effected by the application of the Fourteenth Amendment, both by the judgment of the Supreme Court in Roman v. Sincock, supra, and by the judgment of this court to be entered following the handing down of this opinion holding S.B. 332 and S.B. 336 to be unconstitutional. Vide the Supremacy Clause, Clause 2 of Article VI of the Constitution of the United States. We do not pass upon the rights of the three Senators under the law of the State of Delaware insofar as the emoluments of their offices may be concerned. It is unnecessary that we do so for this is a matter for determination by the State Courts.
See Document 155, Opinion of Attorney General of Delaware, dated December 17, 1964.
It is clear that the members of the 122d General Assembly had the power and exercised it to enact S.B. 332 and S.B. 336 as de facto members of that General Assembly. Mann v. Davis, 238 F. Supp. 458 (E.D.Va. 1964). The fact that one group of de facto members of the General Assembly supplanted another, i.e., the fact that de facto members of the 122d General Assembly were supplanted by de facto members of the 123rd General Assembly and the fact that de facto members of the 123rd General Assembly were in turn supplanted by members of the 124th General Assembly, the present General Assembly, is of no importance in the light of the overriding interdictions of the Constitution of the United States. The members of the 124th General Assembly, the present General Assembly, have the power as de facto members to enact an adequate constitutional apportionment statute for Delaware. Cf. Fruit v. Metropolitan School District of Winchester, Etc., 241 Ind. 621, 172 N.E.2d 864 (1961). Substantially similar issues presented by situations analogous to those at bar, though less complex, have been held not to affect the validity of remedial legislation enacted to reapportion General Assembly districts when prior existing apportionment statutes have been invalidated. See Hughes v. WMCA, Inc., 379 U.S. 694, 85 S.Ct. 713, 13 L.Ed.2d 698 (1965), and Davis v. Mann, 379 U.S. 694, 85 S.Ct. 713, 13 L.Ed.2d 698 (1965).
In view of the foregoing, we hold that there is no sound basis for a constitutional attack upon S.B. 332 and S.B. 336 upon the ground asserted by the plaintiffs that the two Acts failed to provide for the continuation of the terms of the three Senators.
XV. Matters of Procedure, Objections to Evidence, Motions to Strike and Stipulations
A. (1) As to Objections to Evidence and Motions to Strike Evidence
During the course of these protracted proceedings, based on the plaintiffs' motion "To Implement the Mandate of the Supreme Court", many objections to questions propounded by their adversaries were made by the parties, and many of these objections were coupled with motions to strike testimony. The court ruled upon some of these motions from the bench but took many others under advisement subject to later determination when the complex issues presented could be decided upon a full record. On December 20 and 21, 1965, as has been stated, the last hearing was had and testimony of witnesses and exhibits were received. On December 29, 1965, the court ordered each party to prepare a list of all objections to testimony and motions to strike testimony which each party desired to press to the court in this proceeding, requiring these to be filed no later than February 7, 1966. The court further ordered that any objection to testimony or motion to strike testimony not specifically listed by the parties would be deemed to have been waived. On February 7, 1966 the plaintiffs filed a memorandum in support of their motion to strike certain portions of the testimony and the defendants filed a "Motion to Renew Motions to Strike Certain Testimony" and an accompanying memorandum of law.
We will deal first with the objections and motions of the plaintiffs. The plaintiffs object to a question asked Dr. Malcolm E. Jewell, as follows: "Doctor, based on your education, your experience, and your study of Senate Bills 332 and 336 of the Delaware General Assembly, and as a political scientist, do you have an opinion as to how the * * * Bills meet the one man-one vote test as laid down by the United States Supreme Court and the decisions in Baker versus Carr, Reynolds versus Sims, Sincock v. Roman, etcetera?" (Sic.)
Jewell, 299-300.
Dr. Jewell was permitted to answer the question over objection. In his answer he testified at some length respecting the Delaware plan of apportionment, in substance stating his opinion as a political scientist that the percentages of the apportionment and ratios provided by S.B. 332 and S.B. 336 were adequate, proper, and rational, pointing out that "Mathematical exactitude in apportionment cannot be expected." As has been stated, Dr. Jewell's evidence was received subject to plaintiffs' motion to strike.
Jewell, 300-15.
The plaintiffs also objected to a question asked Dr. Jewell as follows: "Doctor, based on your education, your experience and your study of Senate Bills 332 and 336, and again as a political scientist, do these Bills 332 and 336 result in apportionment by substantial equality of population as that term has been used by the United States Supreme Court in Baker v. Carr and the cases which have been decided since that case on the same subject matter?" Dr. Jewell's answer, or perhaps more properly speaking, his answers, were in large part based upon facts of record in the proceeding, and he concluded that the apportionment was a rational one. It is true, however, that he stated his conclusion in the light of "his understanding" that the Supreme Court intended the Decennial Census to be the basis for apportionment.
Jewell, 318-19.
Jewell, 319-26.
The plaintiffs moved to strike the answers to the two questions quoted above. As to both questions, the plaintiffs asserted in substance that the answers were conclusions of law by the witness. Richard T. Green Co. v. City of Chelsea, 149 F.2d 927 (1 Cir. 1945), note 5 cited to the text at 930, cert. den., 326 U.S. 741, 66 S.Ct. 54, 90 L.Ed. 443 (1945); 20 Am.Jur. (Evidence) Section 782, at 653 et seq. The answer to the second question was attacked on the further ground that the question was improper because it was designed to elicit a conclusion of law. 7 Wigmore on Evidence, Sections 1918 and 1952 (3rd ed. 1940). We take the substance of the motions to strike the answers to be that the witness gave his conclusions as to whether or not the Delaware reapportionment constituted invidious discrimination and hence falls within the interdiction of the Fourteenth Amendment.
The issue of whether or not the Delaware apportionment is unconstitutional must be decided by this court. In giving some of his answers Dr. Jewell did state his ultimate conclusions on this issue. Nonetheless by far the greater portion of those answers dealt directly with facts which are before us for determination. These portions of his testimony are clearly admissible. The dissection of the inadmissible portions of Dr. Jewell's answers from the admissible parts is our function. The court sits without a jury and the exercise of that function under the circumstances presents no substantial problem. In making the findings of fact set out in this opinion we have disregarded all conclusions of law expressed not only by Dr. Jewell but also by any other witness whether testifying for or against the validity of the apportionment set up by S.B. 332 and S.B. 336. In view of what we have just written we deem it unnecessary to enter a detailed order specifically dissecting out of Dr. Jewell's whole testimony his conclusions of law. An order will be entered striking the latter from the record but the balance of Dr. Jewell's testimony will be retained and will be given due weight.
We will treat all conclusions of law stated by the witnesses and the motions to strike these conclusions from the record on the same basis as we have treated Dr. Jewell's testimony. What we have said applies, of course, to the motion to strike made by the plaintiffs to the evidence given by Dr. C. Harold Brown in answer to the question asked by the defendants on cross-examination set out in the note below.
"Are you prepared to state as a professional opinion Dr. Brown, that Senate Bills 332 and 336 do not create an apportionment of the State Legislature based as nearly as practicable on equal population?" 519.
(2) As to Objections to Evidence and Motions to Strike Evidence as Made by the Defendants
The motion to renew motions to strike testimony made by the defendants presents a more complicated situation. They have moved (1) to strike all testimony "with regard to alleged racial discrimination" on the ground that the plaintiffs did not have the necessary status to raise the question because they are not Negroes, and (2) to strike all testimony with respect to alleged racial discrimination on the ground that such discrimination has not been found to be unconstitutional by the United States Supreme Court. We have found under heading "XII., B" above, that there is not sufficient evidence to sustain a charge of racial gerrymandering. We deem it unnecessary, therefore, to embark in this already too lengthy opinion upon a discussion to decide whether or not the plaintiffs, who are Caucasians, have locus standi to raise the issue of racial gerrymandering. But the reviewing Court may take a different view of the facts presented than do we. There should be no question of the availability to the higher Tribunal of the evidence which the defendants seek to strike. For this reason, we will decline to grant the first motion referred to in this paragraph.
As to the second motion referred to in the preceding paragraph, as we have just stated, we cannot agree with the assertion that racial discrimination has not been found to be unconstitutional by the Supreme Court. The ruling in Go-million v. Lightfoot, supra, in our view is clearly to the contrary albeit based upon the Fifteenth Amendment and not on the Fourteenth. Again the reviewing Tribunal may disagree with us and the evidence to which the second motion is directed should be available to that Court. It, therefore, must remain in the record. We will deny the motion to strike the testimony relating to alleged racial discrimination.
The defendants also request that the court should take judicial notice that "President Lyndon B. Johnson swept * * * [the State of Delaware] in an unprecedented landslide carrying Democratic candidates into General Assembly seats notwithstanding the prognostication of experienced politicians * * * [and hence] the present make-up of the General Assembly is in an unusual condition * * * not predicted by anyone in this litigation and was not the fault of Senate Bill 336 [sic]." We have taken judicial notice in this opinion that the Democratic victory in Delaware in the 1964 election was of unusual proportions. This is relevant only to the issue of political gerrymandering as we have found it to exist in the case at bar.
Objections were made or were presumed by the court to have been made to all testimony given by any witness in respect to what we have termed political gerrymandering. The defendants have moved to strike the evidence given by Mr. Clayton S. Harrison, Jr., by Senator Eugene D. Bookhammer, by Mr. Niederhauser, by Mr. Christopher L. Perry, a correspondent for the News-Journal papers, by Senator duPont, and by plaintiff Richard Sincock, all germane and relevant to the issue of political gerrymandering. If we are correct in our conclusion that the issue of political gerrymandering is not cognizable, all of the foregoing evidence is irrelevant and inadmissible. Mr. Perry's evidence also is inadmissible as hearsay. In respect to the objections first made to Senator duPont's testimony our ruling that it is admissible is set out in note 7, supra. This testimony seems not to be referred to by the defendants in their present listing of objections to evidence and motions to strike, albeit it was of a character similar to that ruled on in note 7. To clear the record, however, and to make our position plain to the reviewing Court in respect to the objection, Senator duPont's testimony on the issue of gerrymandering will not be stricken from the record since we deem it to be apposite if political gerrymandering as we have described it is cognizable.
Harrison, 2029, et seq.; Bookhammer, 1994, et seq.; Niederhauser, 1542, 1593 and 1921, et seq.; Perry, 1516, et seq.; duPont, 1385, et seq.; and Sincock, 1535, et seq.
If in the event of an appeal the Supreme Court should rule that the issue of political gerrymandering is cognizable in the present proceedings, the evidence which the defendants have moved to strike would be admissible. If on appeal, the issue of political gerrymandering is held to be relevant, there must be no question of the availability to the Supreme Court of the evidence under discussion and our findings in respect thereto. Consequently we will refuse to strike all of the testimony referred to with the exception of that given by Mr. Perry, as indicated. This ruling covers the issue raised by paragraph "7" of the defendants' "Motion to Renew Motions to Strike Certain Testimony."
B. As to the Stipulation, Documents Nos. 174 and 195
The contents of Document 174 have been referred to frequently in this opinion, for example, in "VI.", supra. Document 174 has been supplemented and amended by Document 195, filed on March 14, 1966. We have dealt with the contents of these two documents at an earlier point in this opinion and need not repeat what we have said. It is necessary to reiterate, however, that Document 174 referred in paragraph 6 to the methods employed by Messrs. McGinnes, Hughes and West in estimating the populations of reapportioned General Assembly districts, and in paragraph 9 referred to the maps, set out in note 38, supra, prepared by the registered surveyors showing the boundaries of the General Assembly districts. Paragraph 9 of Document 174, however, states in substance that there were significant deviations in certain boundaries of some of the General Assembly districts in Kent and Sussex Counties and though "conscientious efforts * * * [had been made by] the Departments of Elections of those Counties" to bring the General Assembly district boundaries "as accurately as possible" upon the maps nonetheless "unambiguous descriptions of the `hundreds' do not exist and it is impossible to determine", at least in Sussex County, "the exact location of the boundary markers and lines which comprise[d] the Representative and Senatorial districts as they `heretofore' existed." The last reference is, of course, to the language of Section 661 of S.B. 336, repeatedly referred to and quoted in "XIII.", supra.
It will be remembered that under "XIII." the court checked out the boundaries of the General Assembly districts and found with certain very minor and insignificant exceptions that they were as "heretofore existing and constituted". In short, by the efforts of the parties the language of paragraph 9 of Document 174 quoted above has been rendered incorrect and inapplicable. It is, therefore, impossible to approve the portion of the stipulation referred to.
It will also be observed that Document 195 amends certain incorrect figures in Document 174 insofar as they relate to population figures of Kent County. No issue, therefore, is presently presented as to these figures insofar as Document 174, as amended by Document 195, is concerned in respect to such figures. But another question remains. It will be observed on comparing the population figures of Document 174 even as amended by Document 195 with the figures set out in our opinion, both in the text and in the charts, that the stipulation figures differ in some respects from those found by the court. We have carefully examined all of the figures presented by the evidence and compared them with those contained in Document 174 as amended by Document 195, as well as those to be found in Document 195 "Exhibit 6 — page 2", "Election Results under SB 336, Senatorial Districts." This sheet brings upon the record in columnar form the election returns for the senatorial districts in 1964 and insofar as they are available the election returns for the general elections of 1960 and 1962. The differences between figures as stipulated by the parties and those found by the court are of very small significance and the adjudication of the issues presented by the proceedings at bar in no wise can turn upon the differences indicated. Nonetheless, the court finds its figures to be correct and therefore insofar as the stipulated figures differ therefrom, the stipulation cannot be approved.
In all other respects, however, the stipulation will be approved. Rule 16, Fed.R.Civ.Proc., 28 U.S.C.
XVI. The Remedy
The plaintiffs have prayed, since we treat the "Conclusion" of their "Post-Trial Brief" as containing prayers for relief under their "Motion to Implement", that this court declare S.B. 332 and S.B. 336 to be unconstitutional and invalid under the Fourteenth Amendment; that we enjoin the defendants from conducting, canvassing, or proclaiming the results of any general election for members of the General Assembly of Delaware held pursuant to the provisions of S.B. 332 and S.B. 336; that this court reapportion the General Assembly of Delaware in accordance with the plaintiffs' plan or, as an alternative, in accordance with the plan presented by the Committee of 39, or as another alternative, that this court appoint a special master, under Rule 53, Fed.R.Civ.Proc. 28 U.S.C. to develop a plan to apportion Delaware with 21 senatorial districts and 45 representative districts, each district to be compact and composed of contiguous territory only and to be of substantially equal populations with but minimum deviations from population means or norms and to be set up without regard to the boundary lines of the Counties or of the City of Wilmington; the special master in connection with the accomplishment of the foregoing to give consideration to the feasibility of using computer techniques; the special master's fees and the cost of the supporting personnel to assist him to be taxed as costs against the defendants or their successors in office in their official capacities; that this court declare that the terms of the three Senators referred to under "XIV", supra, are unimpaired and that they are entitled to all the rights, powers and emoluments of their office until the expiration of their terms of office; and finally, that this court declare the members of the present General Assembly to have de facto status only.
We reiterate, as we have in the past, our firm conclusion that it is the duty of the General Assembly to apportion Delaware in such a way as to conform with the Fourteenth Amendment and not transgress its prohibitions, and that only as a last resort should this court embark upon the task of apportioning this State. We reiterate, as we have in the past, our confidence that the General Assembly of Delaware, which has taken a long and unselfish step forward in enacting S.B. 332 and S.B. 336, will again undertake promptly such revisions of the present apportionment as may be necessary, or that the General Assembly may see fit to bring forth an entirely new plan which will meet the requirements of the Constitution of the United States.
As we have previously stated, one of the difficulties which has been presented is that the four gentlemen dealing most immediately with the apportionment, Messrs. Latchum, McGinnes, Hughes and West, by virtue of the circumstances, were compelled to work separately and without substantial guidelines laid down by the General Assembly. The General Assembly itself adopted a two-bill technique, enacting first S.B. 332 and then S.B. 336, thereby putting, as we have said, the General Assembly districts of the State into a Procrustean bed, a bed in which in our opinion the citizens of the State cannot be permitted to rest in view of constitutional interdictions. Perhaps a more equitable result and one which would meet Constitutional requirements could be achieved most expeditiously by the appointment of a small non-partisan or bipartisan commission which could, if funds were made available by the General Assembly, employ disinterested experts to aid it in its difficult task. Delaware is a forward-looking State and the right of unimpaired suffrage is dear to all of us. The citizens of the First State deserve no less. The right of equal suffrage is not a matter of belonging to a particular political party.
We point out, with all respect for the General Assembly, that a survey of the two lower Counties, aided by a census which would involve perhaps not more than 150,000 persons, could be achieved at a comparatively small cost, and that such a census and survey of the General Assembly districts of Kent and Sussex Counties would solve many of the problems with which this court necessarily has struggled.
With the view that the General Assembly will promptly take such steps as may be required to effect a reapportionment in accordance with the Fourteenth Amendment, we will at this time deny, without prejudice, the plaintiffs' prayers that the court reapportion the General Assembly in accordance with the plaintiffs' plan, or as an alternative, in accordance with the plan of the Committee of 39, or as another alternative, appoint a special master to reapportion the State as prayed for by the plaintiffs. We will retain jurisdiction of these proceedings and if no adequate action be forthcoming, we will grant such relief as we deem to be required such as, for example, sua sponte, requiring the next general election or subsequent general elections to be held at large in the State or in certain geographic or political areas thereof.
As we have stated, the record in this case was not completed until March 14, 1966. In so stating we do not intend to imply any criticism of the diligence of counsel though the "Motion to Implement the Mandate of the Supreme Court" was filed on July 14, 1964, and approximately two and one half years have transpired since the date last mentioned. The circumstances of this litigation, particularly the difficulties faced and overcome in the preparation of the maps required by Section 661 of S.B. 336, have rendered the expiration of the period of time referred to inevitable.
We will enter a decree adjudging S.B. 332 and S.B. 336 to be unconstitutional and invalid as transgressing the Equal Protection Clause of the Fourteenth Amendment of the Constitution of the United States and enjoining the defendants and their successors in office from conducting, canvassing or proclaiming the results of elections for members of the General Assembly held pursuant to the provisions of S.B. 332 and S.B. 336 and will declare also that the members of the General Assembly as presently constituted following the 1966 elections have de facto status only. We will set January 10, 1968 as the date by which a statute providing a new apportionment of the State shall be enacted, an apportionment in accordance with the rulings of this opinion.
Findings of fact and conclusions of law are set out in this opinion in accordance with Rule 52(a), Fed.R.Civ.Proc., 28 U.S.C.
A decree may be submitted in accordance with this opinion.
APPENDIX
TABLE XV. 1st Representative District
Census Tract 7 Mr. Theisen's FigureBlocks Total Population
1 — 2 — 3 3 4 — 5 — 14 182 15 226 16 — 17 — 18 — ---- ----- 411 411
Mr. Latchum's population figure 11,710 11,710-168 (transferred to 2nd District) = 11,542 + 411 (subtracted from 3rd District) = 11,953. 11,978-11,953 = 25
Under ideal population 25
2nd Representative District
Census Tract 3 Mr. Theisen's FigureBlocks Total Population
2 20 3 30 4 48 5 57 6 — 13 13 ---- ----- 168 168
Mr. Latchum's population figure 11,853 11,853 + 168 (added to 2nd District, subtracted from 1st District) = 12,021.
Exceeds ideal population 43
3rd Representative District
Census Tract 1 Mr. Theisen's FigureBlocks Total Population
74 58 75 — 76 40 77 62 78 70 79 126 80 — 81 82 82 55 ---- ----- 493 493
Mr. Latchum's population figure 11,890 11,890-411 (transferred to 1st District) = 11,479 11,479 + 493 (subtracted from 4th District) = 11,972 11,978-11,972 = 6
Under ideal population 6
4th Representative District
Census Tract 16 Mr. Theisen's FigureBlocks Total Population
17 161 18 82 33 134 34 118 Census Tract 21 29 109 30 97 31 — 32 122 33 199 34 142 35 70 36 152 37 18 ----- ------ 1404 1404
Mr. Latchum's population figure 11,834 11,834-493 (transferred to 3rd District) = 11,341 11,341 + 1,404 (subtracted from 5th District) = 12,745
Census Tract 1 Mr. Theisen's Figure
Blocks Total Population1 29 2 97 3 103 4 153 Census Tract 10 1 — 2 137 3 100 4 71 5 138 ---- ----- 828 828
12,745-828 (transferred to 5th 6th Districts) = 11,917 11,978-11,917 = 61
Under ideal population 61
5th Representative District
Census Tract 1 Mr. Theisen's FigureBlocks Total Population
1 29 2 97 3 103 4 153 Census Tract 10 3 100 ---- ----- 482 482
Mr. Latchum's population figure 12,917 12,917-1,404 (transferred to 4th District) = 11,513 11,513 + 482 (subtracted from 4th District) = 11,995 11,995-11,978 = 17
Exceeds ideal population 17
6th Representative District
Census Tract 10 Mr. Theisen's FigureBlocks Total Population
2 137 4 71 5 138 ---- ----- 346 346
Mr. Latchum's population figure 12,592 12,592 + 346 (subtracted from 4th District) = 12,938
Census Tract 12
7 155 8 82 9 126 10 18 11 — 12 5 Census Tract 14 5 19 6 54 17 60 18 82 19 120 34 121 35 135 ---- ----- 977 977
12,938-977 (transferred to 8th District) = 11,961 11,978-11,961= 17
Under ideal population 17
7th Representative District
Census Tract 23 Mr. Theisen's FigureBlock Total Population
13 114 114
Mr. Latchum's population figure 11,917 11,917 + 114 (subtracted from 8th District) =12,031 12,031-11,978 = 53
Exceeds ideal population 53
8th Representative District
Mr. Latchum's population figure 11,114 11,114 + 977 (subtracted from 6th District) = 12,091 12,091-114 (transferred to 7th District) = 11,977 11,978-11,977 = 1Under ideal population 1
1st Representative District
Boundaries of added tract (from 3rd Representative District)
N. — Northern city limits E. — Bowers Ave. S. — 28th Street W. — North East Blvd.
Boundaries of substracted tract (added to 2nd Representative District)
N.W. — Northern city limits E. — Baynard Boulevard S.W. — Concord Ave.
2nd Representative District
Boundaries of added tract (from 1st Representative District)
N. — Northern city limits E. — Baynard Boulevard S.W. — Concord Ave.
3rd Representative District
Boundaries of added tract (from 4th Representative District)
N. — 13th Street E. — Walnut Street S. — 9th Street W. — King Street
Boundaries of subtracted tract (added to 1st Representative District)
N. — Northern city limits E. — Bowers Ave. S. — 28th Street W. — North East Blvd.
4th Representative District
Boundaries of added tract (from 5th Representative District)
N. — 9th Street E. — West Street N. — 7th Street E. — Tatnall Street S. — 2nd Street W. — Washington Street
Boundaries of subtracted tracts (added to 3rd Representative District)
N. — 13th Street E. — Walnut Street S. — 9th Street W. — King Street
(added to 5th and 6th Representative Districts)
N. — Brandywine Creek E. — Washington Street S. — 9th Street W. — Jefferson Street S.W. — Delaware Avenue W. — Adams Street
5th Representative District
Boundaries of added tract (from 4th Representative District)
N. — 13th Street E. — Washington Street S. — 9th Street W. — Jefferson Street S.W. — Delaware Avenue W. — Madison Street N. — 12th Street W. — Jefferson Street
Boundaries of subtracted tract (added to 4th Representative District)
N. — 9th Street E. — West Street N. — 7th Street E. — Tatnall Street S. — 2nd Street W. — Washington Street
6th Representative District
Boundaries of added tract (from 4th Representative District)
N. — Brandywine Creek E. — Washington Street S. — 13th Street E. — Jefferson Street S. — 12th Street E. — Madison Street S.W. — Delaware Ave. W. — Adams Street
Boundaries of subtracted tract (added to 8th Representative District)
N. — Delaware Ave. E. — Lincoln Street S. — 6th Street W. — Union Street S.W. — Pennsylvania Ave. W. — Union Street
7th Representative District
Boundaries of added tract (from 8th Representative District)
N. — 6th Street E. — DuPont Street S. — 5th Street W. — Scott Street
8th Representative District
Boundaries of added tract (from 6th Representative District)
N. — Delaware Ave. E. — Lincoln Street S. — 6th Street W. — Union Street S.W. — Pennsylvania Ave. W. — Union Street
Boundaries of subtracted tract (added to 7th Representative District)
N. — 6th Street E. — DuPont Street S. — 5th Street W. — Scott StreetRepublican Republican Democratic Republican Majority Democratic Republican Majority Democratic Republican Democratic Democratic Republican Democratic Majority Majority 1960 1962 Red 14th of 7 152 360 208 132 308 176 13th of 7 99 312 213 88 251 163 9th of 7 78 104 26 42 67 25 11th of 7 111 302 191 88 288 200 12th of 7 130 207 77 92 164 72 8th of 7 121 156 35 27 57 30 12th of 5 135 135 ___ 70 86 16 ___ ___ Total 750 682 ___ Split 19th of 7 166 188 22 132 137 5 772 687 1960 1962 Blue 17th of 7 133 106 27 101 75 26 18th of 7 228 90 138 197 53 144 16th of 7 210 151 69 145 92 53 21st of 7 278 110 168 209 81 128 Split 20th of 7 202 142 60 160 115 45 15th of 7 262 147 115 195 108 87 Split 14th of 5 290 157 133 199 133 66 7th of 5 308 97 211 204 94 110 6th of 5 228 152 76 170 124 46 13th of 5 91 81 10 65 57 8 5th of 5 172 98 74 82 54 28 4th of 5 129 75 54 74 30 44 10th of 7 113 86 27 60 52 8 ---- ---- Total 1162 793 1960 1962 1162 793 -772 -687 ----- ---- 390 Democratic Majority 106 Democratic Majority
I agree with the majority that Senate Bills 332 and 336 are unconstitutional. In so doing, however, I base my conclusion upon the 29% variation between the populations of the 28th Representative District in Kent County as compared with the 34th Representative District in Sussex County, and the 33% variation between the populations of the 6th Senatorial District in Rural New Castle County and the 13th Senatorial District in Kent County. Such wide deviations in population, in the absence of a satisfactory explanation, are not permissible under the Equal Protection Clause of the federal Constitution. Reynolds v. Sims, 377 U.S. 533, 568, 84 S.Ct. 1362, 12 L.Ed.2d 506. Nor should the conclusion here reached be taken as indicating that, aside from the districts specifically mentioned, deviations between the populations of a number of other districts are to be regarded as having my approval. I merely prefer to rest my decision on the narrower grounds above stated.
Turning now to the question of gerrymandering, I cannot say that it presents a constitutional issue. In the first place, I think it unnecessary even to consider the question at this point for several reasons, not the least of which is the fact that we have already declared the plan unconstitutional on other grounds. Alma Motor Co. v. Timken-Detroit Axle Co., 329 U.S. 129, 67 S.Ct. 231, 91 L.Ed. 128. Moreover, while the Supreme Court has declared reapportionment to be a constitutional issue, it has not spoken on the question of gerrymandering. And until it does, it is my view that this important question should continue to be treated by the lower courts as a political, not a constitutional, issue. See the dissenting opinion of Mr. Justice Harlan in Reynolds v. Sims, 377 U.S. at 590, 84 S.Ct. at 1396.
I find it necessary to write a separate opinion, concurring in part and dissenting in part. First, however, I wish to make it clear that I accept without reservation my brother Biggs' findings of fact as set forth in his opinion. I also accept my brother's conclusions of law, except those relating to partisan gerrymandering.
My brothers agree that Senate Bills 332 and 336 violate the Equal Protection Clause of the Fourteenth Amendment because of malapportionment — constitutionally impermissible deviations in population between various Representative and Senatorial Districts. With this holding I heartily concur.
However, as I understand my brothers they agree that partisan gerrymandering is a "political question", and hence does not present an issue cognizable by this Court. But my brothers are not in complete agreement as to the content they ascribe to the phrase "political question." My brother Biggs believes that the Supreme Court has held that partisan gerrymandering presents a nonjusticiable controversy as that term was defined in Baker v. Carr, 369 U.S. 186, 82 S.Ct. 691, 7 L.Ed.2d 663 (1962). My brother Layton, on the other hand, believes that the Supreme Court has not yet spoken on the question, and, that absent a specific mandate from the High Court, this Court should eschew this sensitive issue. With both of these views I must respectfully disagree.
My brother Biggs' position is that partisan gerrymandering has been held, by the Supreme Court, to present no constitutional problem under the Fourteenth Amendment or otherwise. Badgley v. Hare, 87 S.Ct. 338, 17 L.Ed.2d 207 (1966); WMCA, Inc. v. Lomenzo, 382 U.S. 4, 86 S.Ct. 24, 15 L.Ed.2d 2 (1965). Both these citations contain but a summary treatment of a significant problem. In Badgley the appellants raised the question of whether partisan gerrymandering constituted a constitutional question. The Supreme Court dismissed the appeal for want of a substantial federal question. In WMCA the three-judge lower court refused to accept argument on the question of partisan gerrymandering, and found the apportionment plan constitutional. The Supreme Court, per curiam, with Mr. Justice Harlan concurring, affirmed the lower court. Mr. Justice Harlan's opinion contains language to the effect that in affirming the Supreme Court is holding that partisan gerrymandering does not violate the Fourteenth Amendment. However, Mr. Justice Harlan's opinion is but a concurrence; it was not accepted as the majority opinion. Under these circumstances, I find it difficult to conclude that the Supreme Court has spoken on the issue. I would place greater reliance than does my brother Biggs upon the Supreme Court's dicta in Fortson v. Dorsey, 379 U.S. 433, 85 S.Ct. 498, 13 L.Ed.2d 401 (1965) and in Burns v. Richardson, 384 U.S. 73, 86 S.Ct. 1286, 16 L.Ed.2d 376 (1966).
Nor can I agree with the position taken on partisan gerrymandering by my brother Layton in his concurring opinion. As I understand my brother's position it lauds the merits of judicial abstention from touchy questions. No doubt such a course is productive of serenity, but I cannot adhere to a jurisprudence which would require the Supreme Court to speak on these important questions before the lower courts are to confront them.
The Supreme Court in Baker v. Carr, supra, dealt extensively with the doctrine of the "political question." The Court's remarks indicate that the political question doctrine, as set forth in cases such as Colegrove v. Green, 328 U.S. 549, 66 S.Ct. 1198, 90 L.Ed. 1432 (1946), is reserved for cases involving a conflict between the federal judiciary and coordinate branches of the federal government, or cases where judicially manageable standards are lacking. The doctrine is not meant to be a catch-all to avoid the necessity of grappling with sensitive issues. The political question doctrine does not immunize state legislative action from scrutiny by the federal judiciary to determine whether federal constitutional standards have been complied with. As Mr. Justice Brennan said for the majority:
"The question here is the consistency of state action with the Federal Constitution. We have no question decided, or to be decided, by a political branch of government coequal with this Court. Nor do we risk embarrassment of our government abroad, or grave disturbance at home if we take issue with Tennessee as to the constitutionality of her action here challenged. Nor need the appellants, in order to succeed in this action, ask the Court to enter upon policy determinations for which judicially manageable standards are lacking. Judicial standards under the Equal Protection Clause are well-developed and familiar, and it has been open to courts since the enactment of the Fourteenth Amendment to determine, if on the particular facts they must, that a discrimination reflects no policy but simply arbitrary and capricious action." Baker v. Carr, 369 U.S. at 226, 82 S.Ct. at 715 (emphasis added).
The same Mr. Justice Brennan, who wrote for the majority in Baker v. Carr, went on to strongly intimate that, given a sufficient record, partisan gerrymandering would violate the Equal Protection Clause. Fortson v. Dorsey, 379 U.S. 433, 85 S.Ct. 498, 13 L.Ed.2d 401 (1965); Burns v. Richardson, 384 U.S. 73, 86 S.Ct. 1286, 16 L.Ed.2d 376 (1966).
"It might well be that, designedly or otherwise, a multi-member constituency apportionment scheme, under the circumstances of a particular case, would operate to minimize or cancel out the voting strength of racial or political elements of the voting population. When this is demonstrated it will be time enough to consider whether the system still passes constitutional muster." 379 U.S. at 439, 85 S.Ct. at 501.
"Where the requirements of Reynolds v. Sims [ 377 U.S. 533, 84 S.Ct. 1362, 12 L.Ed.2d 506] are met, apportionment schemes including multi-member districts will constitute an invidious discrimination only if it can be shown that `designedly or otherwise, a multi-member constituency apportionment scheme, under the circumstances of a particular case, would operate to minimize or cancel out the voting strength of racial or political elements of the voting population.'" 384 U.S. at 88, 86 S.Ct. at 1294.Fortson and Burns both involved multi-member constituency apportionments, whereas the Delaware apportionment calls for single member districts, but if partisan gerrymandering is proscribed in one instance then logic requires that it be proscribed in the other. Perhaps partisan gerrymandering may be more discernable in one instance than the other, but difficulties of proof should not contract the scope of constitutional protection.
Concededly, gerrymandering is fairly deep in the "political thicket." Nevertheless, to allow a legislature to deprive any group of fair representation in any manner would be to condone invidious discrimination of the sort condemned in Gomillion v. Lightfoot, 364 U.S. 339, 81 S.Ct. 125, 5 L.Ed.2d 110 (1960). Further the underlying rationale of Baker v. Carr, supra, is that voters aggrieved by apportionment discriminations may appear in Federal Court to vindicate their rights. Since the discrimination worked by partisan gerrymandering is as sinister as that worked by malapportionment — both operate to nullify the voting power of certain elements of the citizenry — it would seem that the rationale of Baker v. Carr, supra, requires that those whose votes are debased by partisan gerrymandering be afforded the protection of the Fourteenth Amendment.
For the aforementioned reasons, regardless of the delicate nature of the issues presented, I am unable to conclude that partisan gerrymandering, sufficiently demonstrated, is permissible under the Fourteenth Amendment.
A brief examination of the record in this case as it relates to gerrymandering will demonstrate the flagrant abuses of partisanship. On June 22, 1964 Governor Carvel called a meeting of the leaders of both parties. At this meeting the Governor stated, according to Senator Reynolds duPont, that "the majority party was the majority party and it was going to do the reapportionment." Record at 1407. Accordingly, Mr. Latchum, a former chairman of the Democratic City Committee was asked to reapportion the city of Wilmington. Mr. McGinnis, Governor Carvel's Administrative Assistant, was asked to reapportion rural New Castle County. Mr. Hughes, President of the Kent County Department of Elections and former chairman of the Kent County Democratic Committee, was requested to reapportion Kent County. And, Mr. West, Chairman of the Sussex County Democratic Committee, was asked to reapportion Sussex County.
Subsequently, at the commencement of the debate on Senate Bills 332 and 336, Senator duPont introduced Senate Resolution 131 which provided for recording the proceedings of the Senate during the debate on the reapportionment bills. The resolution was defeated on a straight party vote, and so the verbatim remarks of the participants to the reapportionment debates are not available. However, according to Senator duPont, Senator Allen Cook, president pro tempore stated, "in effect, that this was the majority party's Bill, and that they were going to make their best efforts, because they were the majority, to take care of the Democratic party in the redistricting." Record at page 1471.
I believe that these isolated and fragmentary remarks bespeak an intent to maximize the participation of the majority party in the legislature. My brother Biggs' findings of fact demonstrate that this intent was carried out in Wilmington and in rural New Castle County. The conclusion is inescapable that the "policy" behind Senate Bills 332 and 336 was to maximize the number of districts returning members of the majority party. To this end certain district boundaries were drawn with regard to the political affiliations of the inhabitants. Such a "policy" is what the Supreme Court had in mind in Baker v. Carr when it spoke of a discrimination reflecting "no policy, but simply arbitrary and capricious action." To concentrate on the political affiliations of the citizenry in drafting a scheme of reapportionment is to employ an impermissible standard.
Accordingly, I would agree with my brothers that the reapportionment plan before the Court violates the one man-one vote formula of Reynolds v. Sims, 377 U.S. 533, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964). However, I would also hold the Bills unconstitutional because of partisan gerrymandering in Wilmington, in rural New Castle County, and with respect to the number of Representatives and Senators assigned to Wilmington vis-à-vis rural New Castle County.