*1 OF STATE OF SECRETARY KIRKPATRICK, v. PREISLER MISSOURI, et al. et al. January 13, Argued 30. No. 7, 1969* April
Decided al., *Together with Heinkel et al. Preisler et No. also appeal from the same court.
Thomas J. Downey, First Assistant Attorney General of Missouri, argued the appellants cause for in No. 30. With him on the Anderson, briefs were Norman H. se, Attorney General, pro Wood, and Louren B. Assistant Attorney General. David Collins argued cause and filed appellants a brief for in No. 31.
Irving Achtenberg argued the for appellees cause both cases. With him Preisler, on the brief was Paul W. pro se. Justice Brennan opinion delivered the
Mr. Court. Sanders, Wesberry
In v. U. S. (1964), we held that it may possible not be “[w]hile [for States] congressional draw districts with mathematical preci- sion,” id., 18, at I, 2,§ Art. the Constitution requires “as nearly practicable as is one man’s vote a con- another’s.” worth as much as is to be election
gressional cases to elucidate are these Id., required 7-8. We at practicable” standard. the “as chal- redistricting statute congressional Missouri The that State’s second resulted from in these cases lenged Wesberry redistricting since attempt congressional District Court three-judge was decided. of Missouri declared the Western District was districting Act then effect Missouri any judi- under but withheld unconstitutional Missouri State of Legislature cial “until relief prob- opportunity to deal with the has once more had Missouri, Secretary State Preisler lem . . .” . As- Supp. 187, Thereafter, 191. the General 238 F. *3 statute, a but redistricting of Missouri enacted sembly The Dis- too was declared unconstitutional. this statute any to review however, jurisdiction retained Court, trict Secretary Preisler v. might enacted. plan that further Missouri, Supp. (1966), aff’d, 953 sub 257 F. State of of Preisler, In (1967). 385 S. 450 Kirkpatrick nom. U. Assembly under enacted the statute the General 1967, Supp. 128 Stat., (Cum. 1967), here, attack Mo. Rev. c. of Missouri moved the Attorney the General the Act and sustaining for a declaration District Court order the case. dismissing data available to the Based on the best States census legislature figures, the United among Missouri’s 10 con- population equality absolute mean a gressional 431,981 districts would district. The districts created the 1967 how- by Act, each range 12,260 from this ideal within a ever, varied 13,542 below to above it. The difference between the populous 25,802. least and most was thus In districts percentage the most terms, populous district was 3.13% was populous least and the ideal, the mathematical above below.1 2.84% Assembly General found that the Court The District instead figures but on the census not fact relied had addition, less accurate data. its on plan based had Assembly General found that the District Court it which redistricting submitted to plan a rejected had population variances with smaller districts provided Court found Finally, the District among them. one counties from switching some simple device with produced plan have to another would district among districts. Based markedly variances reduced dissent- Court, judge District one findings, these meet 1967 Act not the constitu- held that did ing, representation for num- equal equal tional standard practicable,” “as and that people bers of any acceptable justification make State had failed to (1967). noted Supp. variances. 279 F. We by Act, redistricting the 1967 based on a The effected according 4,319,813 census, to the 1960 is as follows: Variation % From, Population. Ideal. District No. 439,746 One 1.80 + 436,448 1.03 Two + 436,099 Three 0.95 + - 419,721 2.84 Four - 431,178 0.19 Five - 422,238 2.26 Six *4 436,769 1.11 Seven + 445,523 Eight 3.13 + - 428,223 Nine 0.87 - 423,868 1.88 Ten 431,981 population per Ideal district. Average variation from ideal. 1.6% 1 largest 1.06 to of district. Ratio to smallest Number districts within ideal. 1.88% 25,802 largest Population between and smallest districts.. difference stayed the Court’s but District probable jurisdiction expressly authorized judgment pending appeal congressional elections under State “to conduct . . . 390 U. 1967 . . . Act S. pursuant [the] (1968). We affirm. argument is that primary Missouri’s by the 1967 Act among variances the districts created de minimis they be considered are so small that should satisfy nearly as prac- reason to the “as and for that independent justi- not require ticable” limitation and argues justifica- Missouri Alternatively, fication. tion the variances was established the evidence: Assembly provided it is contended that the General legitimate regard variances out of for such factors as the representation groups, integrity of distinct interest county compactness fines, districts, popu- lation trends within the State, high proportion military college personnel, and other students, nonvoters and the districts, political “legislative some realities of interplay.”
I. reject argument We Missouri’s there is a fixed percentage population numerical or variance small enough to be considered de minimis satisfy and to with- out question nearly the “as practicable” as standard. The whole thrust of nearly the “as as practicable” ap- proach is inconsistent with adoption of fixed numerical standards which excuse variances without regard to the circumstances of each particular case. The extent to which equality may practicably be achieved may differ from State to State and from district to dis- trict. “equal representation Since for equal numbers of people the fundamental goal [is] for the Rep- House of resentatives,” Sanders, supra, at 18, the “as practicable” standard requires that the State good-faith make a effort to precise achieve mathematical *5 Sims, Reynolds 377 U. S. equality. See among congres- variances (1964). population Unless despite such sional districts are shown to have resulted variance, each no matter effort, justify the State must how small. de minimis rejecting other reasons for
There are nonarbitrary way pick a approach. We can see no to suddenly population at which variances point cutoff become de minimis. to consider a certain Moreover, de minimis would range encourage legislators variances range equality to for rather than for as strive practicable. found, example, The District Court for legislator it leading that at least one Missouri deemed proper a attempt to to achieve level of variance 2% rather than seek equality. to Equal representation equal people numbers of is a principle designed to prevent voting debasement of power to representa- and diminution of access elected tives. Toleration of even small deviations detracts from these purposes. Therefore, the command of I, § 2, Art. that States create districts which provide equal representation equal numbers of people permits only the limited variances which are un- despite good-faith avoidable effort achieve absolute equality, justification or for which is shown.
Clearly, among variances the Missouri congressional districts were not Indeed, unavoidable. seriously is not contended that the Missouri Legislature equality might came as close to as it have come. The that, Court found contrary, District the two efforts of the reapportionment Legislature Missouri since leadership “the of both political parties in the given the House were nothing Senate and better to work produced with than a makeshift bill by what has been recognized to no candidly be more than ... an expedient
political compromise.” F. Supp., 966. Legisla- *6 at the frankly conceded Act 1967 of the proponents tive simple device resort to the hearing that District Court popu- known subdivisions political transferring entire produced have would contiguous districts lation between The District equality. numerical to much closer districts Legislature the Missouri moreover, found, Court constructing districts, the data inaccurate relied on would plan which consideration without rejected the among population variances markedly reduced have popu- inconceivable that simply it is Finally, districts. in the Missouri magnitude found of the disparities lation apportionment York The New were unavoidable.2 plan absolute districts of almost divided into regions plan Rockefeller, in Wells v. described equality a state striking evidence that 545-546, provides post, complete almost which tries can achieve legislature In the districts. equality among all State’s numerical have the State could quite obvious that sum, “it seems providing equal popu- to districts of much closer come Adams, Swann v. 385 U. S. than it did.” lation (1967). 445 record question turn to whether the therefore the
We any legally acceptable justification the establishes variances. It was the burden the State acceptable . reasons the variations present “to . . various . . . districts . among populations of the . . .” Adams, supra, at 443-444. Swann v.
2 Contrary
appellants’ assertion,
we have not sustained
any congressional districting
constitutionality
plan
popula
with
magnitude
plan.
of the
found in the Missouri
tion variances
only
presented
Johnson,
(1967), the
issue
Connor v.
II. We agree with the District Court that Missouri has not satisfactorily justified variances among the districts.
Missouri contends that variances were necessary to avoid fragmenting areas with distinct economic social interests and thereby representa- diluting effective tion of those interests in But to Congress. accept popu- lation variances, large small, or order to create districts specific with interest orientations is antithetical to the premise basic provide constitutional command to *7 representation equal for equal people. numbers of either history alone, nor economic or other sorts “[N] of group interests, permissible are factors in attempting disparities to justify population-based from representa- Citizens, tion. not history or economic interests, cast Sims, Reynolds supra, at 579-580. votes.” v. also See Mann, Davis (1964). 377 U. S. reject
We also argument Missouri’s that reason- “[t]he population ableness in differences the congressional districts under review must in ... viewed the con- of legislative interplay. text The legislative leaders all testified that in question act was in opinion their legislative compromise. reasonable ... It must be practical political remembered . . . problems that are in inherent enactment congressional reapportion- legislation.”3 agree ment We with the District Court that “the rule is one of ‘practicability’ rather than polit- ” ical ‘practicality.’ F. Supp., 989. Problems by partisan politics created cannot justify apportion- pass ment which does not otherwise constitutional muster. we do not find Similarly, legally acceptable the argu- justified ment that variances are if they necessarily result attempt from a to avoid State’s fragmenting political Appellants Brief for 37-38. along district lines by drawing congressional subdivisions county, municipal, political or other subdivision existing con- constructing boundaries. The State’s interest manner, districts in this it is is to gressional suggested, opportunities partisan minimize the gerrymandering. justi- But an that deviations from are argument equality from legislators engaging parti- fied order to inhibit gerrymandering4 san is no more than a variant rejected, prac- argument, already that considerations politics population disparities. justify tical can population Missouri further contends certain legislature’s taking variances resulted from the account eligible among of the fact that voters percentage the total differed significantly from district disproportionately to district —some districts contained personnel numbers of large military stationed at bases by maintained the Armed Forces and students attend- may ance at universities or There colleges. question be a except whether distribution of seats accord- ing permissible to total can ever be under assuming deciding Art. But without I, appor- § 2. tionment be based on may eligible voter rather than population, plan total the Missouri is still unac- ceptable. attempt Missouri made no to ascertain the *8 any temptation gerrymander It is dubious in event that the to inhibited, legislature would be much since the would still be free subdivisions, to choose political which of several all with their own complexion, particular congressional in Besides, to include a district. opportunities greatest gerrymandering are when there is freedom unequally populated to artistry construct districts. “[T]he political cartographer put highest is to its test when he must work equal with population. times, constituencies of At such his skills compared can surgeon, of a to those for both work under fixed and However, mapmaker arduous rules. if the is free to allocate varying populations districts, to different then the butcher's cleaver replaces scalpel; sharply the and the results reflect the difference operation.” Hacker, Congressional Districting in the method of A. (1964 ed.). rev. number of in eligible apportion voters each district and to accordingly. haphazard adjustments At best it made scheme population: overpopulation a based total in the Eighth explained by pres- District was the away in ence that district of a base and military university; a no attempt presence was made to account for in universities other districts or the disproportionate of newly numbers arrived and short-term residents City Even as to Eighth District, St. Louis. there is no indication that population the excess allocated to that district corresponds to the alleged extraordinary additional numbers noneligible voters there.
Missouri also argues that population disparities be- tween some of its districts result from the legislature’s attempt projected popu- take into account lation shifts. recognize We that a congressional district- ing plan will usually be in effect for at least 10 years five congressional elections. may Situations arise where substantial period shifts over such a can be anticipated. Where predicted these shifts can be with high a degree of States accuracy, redistricting that are may properly consider them. By open this we mean to no avenue for subterfuge. as to Findings trends must be thoroughly applied documented and throughout hoc, State a systematic, not an ad manner. attempted Missouri’s justification of the sub- underpopulation stantial in the Fourth and Sixth Dis- tricts falls short of far this standard. The District Court found “no evidence . . . that . . . General Assembly adopted any policy projection in devising any Districts 4 or6, other district, enacting the 1967 Act.” 279 Supp., F. at 983.
Finally, Missouri that some of claims the deviations equality consequence from were legislature’s at- tempt to ensure that each congressional district would be geographically compact. Sims, Reynolds However, *9 im- developments said, “Modern
supra, we make and communications transportation provements that mid-1960’s, in the most claims hollow, rather representation from can population-based deviations solely geographical considerations. validly based allowing for such in order insure Arguments deviations areas and to representation sparsely settled effective becoming large so that legislative districts from prevent availability representa- of access citizens to their part, is are for the most uncon- impaired today, tives compactness claim vincing.” any event, In Missouri’s solely upon appearance is of the based unaesthetic congressional that from map of boundaries would result attempt changes to effect some of the district lines according to the lower would which, court, achieve greater pleasingly A equality. preference State’s hardly justify population districts can shaped variances.
Affirmed. dissenting Harlan, Mr. Justice opinion of see [For post, p. 549.] dissenting opinion Mr. Justice see
[For White, post, p. 553.] Fortas, concurring.
Mr. Justice I judgment cases, concur these Court but I cannot near-perfection subscribe the standard of obligatory which the upon Court announces state legislatures facing problem reapportion- the difficult ment for elections. Sanders, S. this (1964), U. Court may
recognized “it possible that not be to draw con- gressional precision,” districts with mathematical requires they held Constitution be drawn that, practicable,” representative so “as as is each *10 of a vote on behalf of the same number should cast people. only interprets prac-
The Court now not “as ticable” to mean that the State is to “make a required good-faith precise equal- effort to achieve mathematical ity,” population but it also that requires any remaining disparities “no matter small,” justified. how be It then seriatim, proceeds to reject, every type justification of that has one possibly, every that could be—advanced. been —
I agree legislatures that the state should be required good-faith to make “a effort to achieve” a result that allocates the or the residents1 State roughly in equal numbers to each district, upon based orderly some objective my method.2 In view, the State could properly figures popu- arrive at for current by taking lation making latest census returns and modifications to allow for movements since (which the last census the Court accept- seems to find able). my It in also, opinion, could discount the census figures to take presence significant account or transient particular nonresident in areas (an adjustment as to which the Court indicates doubt). If proceed the State should appropriate some popu- constitutionally entirely
1 I would find acceptable for a State apportionment to base its residents, on numbers of rather than total population, in each district at the time the districts are established. permit adjustments This account, would to example, take resulting large distortions from numbers of military nonresidents at colleges installations or in an area. 2 Avery County, Midland (1968), U. S. I argued dissenting opinion in a disparate consideration local might appropriate respect defining interests types with certain government exercising of local governmental units powers. limited there, however, I justify noted the same factors could not departing man, from theory the one one legislatures— vote state or, might add, I now general districts —because of the performed. and basic nature of the function producing approx- lation such as I have suggested, basis imately my trial equal districts, courts, judgment, justified declining disapprove would be the result merely disparities, because of small the absence of of gerrymandering- evidence deliberate and arbi- —the trary populations distortion boundaries and district partisan personal political or purposes.
In considering “approximated” whether the State has division and allocation of I equal the population, *11 sympathize majority’s with the view that a de minimis disparities rule of allowable to theory tends demean in in practice the constitutional objective because it suggests necessary that it is not even to aim equality. On the other to de hand, reject minimis as a state- of ment the limits on the equality rule of should not us to toss recognition lead aside the wise of the inscrut- ability of imperfection fact and the of man which is implicit the standard: “as nearly prac- as phrase ticable.” This does merely not refer to arith- metical possibilities. Arithmetically, is possible to achieve division aof State into precisely districts of equal size, as by measured the decennial any census or other population carry base. To out this theoretical pos- sibility, legislature a however, might have ignore to the boundaries of common running sense, the district fine down the middle of the corridor of an apart- ment house or even dividing the of single- residents family house between two districts. The majority opin- ion does not suggest so practical extreme a application of its teaching, and I only mention it because the ex- ample may dramatize the fallacy of inflexible insistence upon mathematical exactness, with no tolerance for reality.
Whatever might be the merits of insistence on absolute if it equality could be attained, the majority’s pursuit of precision is a search for a will-o’-the-wisp. The fact is any solution apportionment the and districting problem is at because it based approximation best is always degree upon figures which are some obsolete. on which purpose by precision No is served an insistence imprecisions in is unattainable because of the inherent must be based. districting data which legislature The base to which Missouri’s should have is precisely, according adhered to the majority, legislature’s 1960 decennial census. The here under plan perfect preci- review was in 1967. Assuming enacted for the when taken,3 by 1967, sion because census of population movement within the State The basic enumeration error the census—that is the variation which would be observed between successive enumerations very surveys areas, same area —is low. Second selected conducted by specially enumerators, produced by varying only trained counts regular about for the whole from the counts of the 1% particular groups population, enumerators. For in the the variance significantly larger. was Census, U. S. See Bureau Evaluation Program Population and Research U. S. Censuses of “Accuracy Housing, 1960, Population of Data on Characteristics as by Re-interviews,” ER-60, Measured Ser. (1964), No. 4 Table p. 22. *12 significant
Far more than variations between successive enumera- virtually undercountings tions are all produced are errors — —which by system. enumerating the inherent limitations of the A Census only Bureau estimate indicates that the 1960 census counted 96.9% population, of people being the whole of the not found and 3.1% by counted the Undercounting evenly enumerators. was not dis- population. tributed over Instead, the whole members of certain groups, notably young Negroes, likely adult were far more to be by missed the For age enumerators. nonwhites in groups all the population by census was estimated to understate the actual 9.5%. young Negro males, For undercounting adult reached 20% five-year age groups. generally, some See Siegel, Completeness Coverage Population of of the Nonwhite in the 1960 Census and Estimates, Implications, Current and Some Report, Conference on City the (Washington, Social Statistics and C., 22-23, 1967) D. June (Heer ed., 1968). heavily the groups Because undercounted are evenly not country, distributed over the the differential rates of undercounting produce divergences pop- between the actual relative particular ulations of by areas and those indicated the census. had disparities substantial migration, as in-and-out well of real distribution the arisen between here 1960 census base reflected in the and that State by the Court.4 zealously protected so indicating be taken as I have said should Nothing that requires that the standard I do not believe the demon- correspondence of between degree a high the district figures or residence and strated relatively this fix, would I least at divisions. Nor effort, percentage a reapportionment of the early stage variation.5 permissible figure agree judgment I that the cases, however, present history The be affirmed. of the District Court should of the failure reapportionment legislature’s and of this of the patient directions comply plain with the and of the legisla- District and the failure three-judge Court census or other figures ture to use either accurate 1960 for all the districts— systematically figures obtained District Court’s refusal strongly support factors these counties, Missouri’s 114 Bureau has estimated that of The Census gained. 1966, while 64 The 50 lost between 1960 7.7%; County city 57,900, Louis independent Louis lost or St. St. City County, Louis gained 146,000 or Outside St. 20.8%. County change ranged 22,100 from increase in St. Charles a absolute County. percentage change Dunklin The 7,100 to a decrease in County to ranged from increase in St. Charles a a 21.4% 41.7% Population County. in Holt Estimates Counties: decrease Population Reports, Popula July (Report 3), 1, 1966 No. Current (Bureau Estimates, P-25, Census, No. 407 October tion Ser. 1968) 11-13. Reynolds (1964): Sims, 533, 578 Cf. 377 U. S. expedient attempt spell present, “For the we deem it not to out any marginally permissible in precise constitutional is tests. What may unsatisfactory another, depending par- one State body Developing ticular circumstances the ease. doctrine *13 case-by-case appears provide satisfactory on a basis to us to the most requirements arriving means of detailed constitutional area legislative apportionment.” of state that on true It is plan. accept the Missouri from what only variation was there average, 1.6% (meaning the “ideal” calls majority quaintly was 10 districts of the only three figures) census true it is also more, or there a variation 2% regardless But finding gerrymandering. no there is might range this within that variances possibility of the Wes- tolerable within be considered in some situations sustain I that we should standard, agree berry’s the his- light plan of the rejection Court’s District preparation. plan’s and the record the cases tory of