Direct democracy is not possible in a nation as populous as the United States is now, or even as it was in 1787 when the Constitution of the United States was drafted. Accordingly, the objective was then, and is now, to devise and implement as fair and effective a plan of democratic representation as possible.
The idea of fair and effective representation at each level of government was not new in 1787. The search for such a formula lies at the center of Anglo-American political thought. In 1690 john locke sought to abolish England's rotten boroughs by urging that, "it being the interest as well as the intention of the people, to have a fair and equal representation, whoever brings it nearest to that is an undoubted friend to and establisher of the government, and cannot miss the consent and approbation of the community."
Although Britain did not put an end to its rotten boroughs until near the middle of the twentieth century, the issue of how best to structure a truly representative government was very much alive at the time the various proposals for the American Constitution were being debated. At last a compromise was struck in the constitutional convention of 1787, giving equal representation to each state in the Senate and representation based on population in the House of Representatives. Article I, section 2, of the Constitution provides that "Representatives … shall be apportioned among the several states … according to their respective numbers …," with recomputation of the apportionment every ten years, and each state to have at least one representative regardless of population. But the task of fixing the formula for the apportioning process was left to Congress, and no directions at all were established to guide the states in the parallel function of allocating seats in the state legislature or in local governmental bodies. We are not, however, left entirely in doubt about what Congress thought appropriate for apportionment in the states. The northwest ordinance of 1787 provided that representation in the territorial legislatures to be created in that area should be based on population. In general, the states accepted the principle of reasonably equal population among legislative districts, but the principle was often modified by assurances of at least one representative from each county or township or municipality. Departures from population equality may not have been egregious in this time of mostly rural dispersal; but by the late nineteenth and early twentieth centuries what had once been minor deviations became major divergences.
john quincy adams observed in 1839 that the division of sovereign powers between the states and the nation, as set out in the Constitution, gave us "the most complicated government on the face of the globe." The twentieth century has proved how right he was. The interaction between increasingly potent national and state governments, frequently aggravated by friction arising out of competition for power, has produced a delicately balanced division of power and a complexity of relationships probably unsurpassed in the history of governmental institutions.
Yet it is the proud boast of federalism in the United States that the governments of the fifty states and that of the nation can work together in common purpose rather than in a relationship of competition and mistrust. Moreover, it is a basic premise of representative democracy in the United States that the people are entitled to representation somewhat in proportion to their numbers, at every level of government. The tradition of majority rule cannot otherwise be attained. Neither the division of sovereign powers prescribed in the federal system nor the fairness of legislative representation formulas can long be left unattended. Vigilant superintendence by an informed electorate is essential.
Even the wisest political scientists have difficulty in defining the precise meaning of representative democracy. There is, however, general agreement that representative democracy in the United States includes something of liberty, equality, and majority rule. Even though these qualities are scarcely less abstract, it can surely be said that representative democracy relates to the processes by which citizens exert control over their leaders. From the time of the Constitutional Convention debate has centered on the extent to which, and the ways in which, majority control over leaders should be exercised. Congress has wrestled with the issue, with inconclusive results. In 1842, for example, Congress required each state to establish compact, contiguous, single-member congressional districts as nearly equal in population as possible. These criteria, however, lapsed in 1911. In any event, no enforcement method had been established, and the courts considered the issue none of their business.
Not until more than a hundred years after the ratification of the constitution in 1789 did such states as California, Illinois, Michigan, New York, Ohio, and Pennsylvania, responding to new pressures, abandon the equal-population principle in one or both houses. So widespread had been the original acceptance of the equality concept that no fewer than thirty-six of the original state constitutions provided that representation in both houses of the state legislature would be based completely, or predominantly, on population. Between 1790 and 1889 no state was admitted to the Union in which its original constitution did not provide for representation principally based on population in both houses of the state legislature.
To speak of the equal-population principle as the basis for apportionment of those nineteenth-century legislatures is not to say that there was mathematically precise equality among the districts at that time. The western states, for example, commonly relied on county lines in drawing their apportionment formulas. The distortions that resulted from assuring each county at least one representative, for example, or from grouping whole counties to form election districts, were much less pronounced in agricultural and rural America than in present-day industrial and urban America. The population of the United States, outside the few great commercial centers in the East, was spread thinly across the face of the country.
The drift from relative equality to substantial inequality would have moved at about the same pace as the shift in population from rural to urban America; and that would have been bad enough. But some states accelerated the trend away from the equality principle by other devices as well. As state legislatures were enlarged, additional seats were granted to the areas of new growth without diminishing representation of the declining population areas. As the population of rural areas declined, state legislatures abandoned even the formal acceptance of equal population as a controlling principle, typically guaranteeing each county (or township) one representative. Some states, unable or unwilling to change the constitutional requirement for equality among districts, simply ignored the mandate for decennial change. (Tennessee is a good example; the state constitutional requirement of reapportionment every ten years was ignored between 1901 and 1961, giving rise in 1962 to baker v. carr.)
The consequence of these factors, singly or in combination, was by the middle of the twentieth century a remarkable skewing of voter impact, ordinarily giving the less populated areas of a state a disproportionate influence in legislative representation. The impact was most marked at the state and local legislative levels, but not without considerable influence on congressional districting as well.
By the middle of the twentieth century the disparities in legislative representation were marked. Thus, in the then ninety-nine state legislative chambers (forty-nine bicameral legislatures plus the Nebraska unicameral legislature), thirty-two relied in large part on population; eight used population, but with weighted ratios; forty-five combined population and area considerations; eight granted representation to each unit; five had fixed constitutional apportionments; and one (the New Hampshire Senate) was based on state tax payments. These conclusions somewhat understate the actual disregard of population as the basis of representation because this summary is drawn exclusively from the state constitutional requirements, without adjustment for violation of those provisions.
The time has come to ask: what is (re)apportionment and what is (re)districting? The question is well put, for the terms are sometimes (confusingly) used interchangeably. But there is a difference. Apportionment is ordinarily described as the allocation of legislative seats by a legislative body to a subordinate unit of government, and districting as the process of drawing the final lines by which each legislative district is bounded. Thus, Congress apportions the number of congressional districts to which each state is entitled, based on population figures disclosed at each decennial census. Each state legislature then draws lines that divide the state into as many congressional districts as have been allocated to it by Congress.
State legislatures, on the other hand, both apportion the distribution of state legislative seats and draw the district lines that determine which voters will make each selection. Therein lies the problem, clearly rooted in the political ambition of each political group to overcome its opposition, before the voting begins, on the basis of the dispersion of voters eligible to vote for one candidate rather than another.
By the early 1960s the act and the impact of malapportionment were everywhere apparent, typically to the apparent disadvantage of individual voters in heavily populated districts and to the apparent advantage of voters in sparsely populated districts. Despite the fact that many state constitutions required reapportionment every ten years and included formulas requiring approximate population equality, no legislative chamber came closer to that goal than two to one, and the disparity between most populous to least populous district was in many states more than ten to one and in several more than one hundred to one. To put the matter another way, in twelve states fewer than twenty percent of the voters lived in districts that elected a majority of the state senators, and in seven states fewer than thirty percent of the voters lived in districts that elected a majority of the members of the lower house.
State courts occasionally acted to deal with the most egregious abuses, but the federal courts, until 1962, adamantly refused to intervene. Although the Supreme Court had long recognized the right of citizens to vote free of arbitrary impairment by state action when such impairment resulted from dilution by false tally or by stuffing of the ballot box, the Court had declined to deal with apportionment and districting abuses on the grounds that the issue was not justiciable, that is, not appropriate for federal judicial intervention. As Justice felix frankfurter said in colegrove v. green (1946), "Courts ought not to enter this political thicket."
Finally, the case of interference with the exercise of the franchise was made so clearly that a majority of the Court was persuaded that only federal judicial intervention could put an end to this denial of equality. The case that triggered this change in attitude provided a dramatic illustration of flagrant abuse of voter rights by a state legislature that had openly flouted its own state constitution for more than half a century.
The Tennessee Constitution had required, since 1870, that the number of senators and representatives in the general assembly "be apportioned among the several counties or districts, according to the number of qualified electors in each.…" Moreover, the state constitution required reapportionment in accordance with the equal-population standard every ten years. Between 1901 and 1961, however, the legislature had not acted on the matter. As a result, thirty-seven percent of the Tennessee voters lived in districts that elected twenty of the thirty-three senators, and forty percent of the voters lived in districts that elected sixty-three of the ninety-nine members of the lower house. The federal court challenge was brought by voters in urban areas of the state, who invoked the Constitution of the United States and claimed that they had been denied the equal protection of the laws, "by virtue of the debasement of their votes."
The resulting Supreme Court decision, Baker v. Carr, did not rule on the substance of the equality claim, but did hold that the issue was properly within the jurisdiction of the federal courts and was justiciable. Only Justices Frankfurter and john marshall harlan dissented.
Within two years the Supreme Court signaled how it would decide the equality issue. gray v. sanders (1963), while not strictly an apportionment case, involved the closely related issue of voter discrimination. The election practice there challenged was the Georgia "county-unit" system, as it applied to statewide primaries: a candidate for nomination who received the highest number of popular votes in a county was considered to have carried the county and to be entitled to two votes for each representative to which the county was entitled in the lower house of the general assembly. The majority of the county unit vote was required to nominate a candidate for United States senator or state governor, while a plurality was sufficient for nomination of candidates for other offices. Because the most populous county (Fulton, with a 1960 population of 556,326) had only six unit votes, while the least populous county (Echols, with a 1960 population of 1,876) had two unit votes, "one resident in Echols County had an influence in the nomination of candidates equivalent to 99 residents of Fulton County."
Georgia argued that, because the electoral collegepermitted substantial inequalities in voter representation in a "winner-take-all" system, parallel systems should be permitted to the states. Moreover, the state argued that because United States senators represent widely divergent numbers of voters, the same should be permitted in one house of a state legislature. But the Supreme Court rejected all such analogies as inapposite: "The inclusion of the electoral college in the Constitution, as the result of specific historical concerns, validated the collegiate principle despite its inherent numerical inequality, but implied nothing about the use of an analogous system by a State in a statewide election. No such specific accommodation of the latter was ever undertaken, and therefore no validation of its numerical equality ensued."
While conceding that states "can within limits specify the qualifications of voters both in state and federal elections," the Court denied that a state is entitled to weight the votes "once the geographical unit for which a representative is to be chosen is designated.…" Accordingly, the Court concluded: "The conception of political equality from the declaration of independence, to Lincoln's Gettysburg Address to the fifteenth, seventeenth, and nineteenth amendments can mean only one thing—one person, one vote." The fatal defect in the Georgia plan was that the votes were weighted on the basis of geography as an expression of legislative preference for rural over urban voters.
The next franchise case decided by the Supreme Court with full opinion, wesberry v. sanders (1964), was also not a state legislative apportionment case; it was a congressional districting case not very dissimilar from Colegrove v. Green—except in result. Plaintiffs were qualified voters of Fulton County, Georgia, entitled to vote in the state's fifth congressional district, which had a 1960 population of 823,680, as compared with the 272,154 residents of the ninth district.
After the decision in Baker v. Carr, the Court had little difficulty deciding that such issues were justiciable in federal courts. The substantive ruling, however, came as something of a surprise. Plaintiffs had argued principally that the gross population disparities violated the equal protection clause of the fourteenth amendment. The Supreme Court, however, adopted what had been a subordinate contention, that the Georgia arrangement violated Article I, section 2, of the Constitution, which prescribed that representatives be chosen "by the People of the several States." Justice hugo l. black, writing for the majority of six, stated that this provision, when construed in its historical context, means "that as nearly as practicable one man's vote in a congressional election is to be worth as much as another's.… To say that a vote is worth more in one district than in another would not only run counter to our fundamental ideas of democratic government, it would cast aside the principle of a House of Representatives elected "by the People,' a principle tenaciously fought for and established at the Constitutional Convention." That result, at first surprising in view of the nonspecific constitutional text, was well supported in the Court's review of the relevant history. For example, at the Constitutional Convention james wilson of Pennsylvania had said that "equal numbers of people ought to have an equal number of representatives," and representatives "of different districts ought clearly to hold the same proportion to each other, as their respective constituents hold to each other."
Reliance on section 2 of Article I rather than on the equal protection clause has had significant consequences. From that date forward the Court has been less tolerant of population variations among congressional districts than of those in state legislative districts, as to which the population-equality principle has, since reynolds v. sims (1964), been based on the equal protection clause of the Fourteenth Amendment.
Reynolds v. Sims and its five companion cases completed the original round of apportionment and districting cases and constituted the foundation on which all subsequent litigation has built. On June 15, 1964, the Court invalidated the state legislative apportionment and districting structure in Alabama (the Reynolds case), Colorado, Delaware, Maryland, New York, and Virginia. One week later the Court struck down the formulas in nine additional states, foretelling a complete reapportionment revolution.
Reynolds v. Sims was illustrative. The complaint in that case alleged that the last legislative reapportionment in the state had been based on the 1900 federal census despite a state constitutional requirement for decennial re-apportionment. Accordingly, because the population growth had been uneven, urban counties were severely disadvantaged by the state legislature's failure to reapportion every ten years and by the state constitution's provision requiring each of the sixty-seven counties to have at least one representative in the lower house with a membership of 106. The Supreme Court of the United States ruled unequivocally in favor of the equal-population principle: "We hold that, as a basic constitutional standard, the Equal Protection Clause requires that the seats in both houses of a bicameral state legislature must be apportioned on a population basis. Simply stated, an individual's right to vote for state legislators is unconstitutionally impaired when its weight is in a substantial fashion diluted when compared with votes of citizens living in other parts of the state."
The decisions in Wesberry and Reynolds required adjustment of congressional districting practices in all states (except the few states with only one representative each) and of all state legislative districting practices. Despite considerable adverse reaction in the beginning and substantial litigation to determine the full significance of the decisions, by and large compliance was secured; and further adjustments were made after the results of the 1970 and 1980 censuses were determined.
Two principal types of questions remained to be worked out after the first decisions were announced: how equal is "equal" in congressional districting and in state legislative apportionment and districting? and to what extent does the equal-population principle apply to the thousands of local governmental units and the even larger number of special districts that serve multitudinous quasi-governmental purposes?
Despite criticism of the Reynolds decision based on an assumption that the Court had demanded mathematical exactness among election districts, the Court explicitly acknowledged the permissibility of some variation: "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 important obligation is for each state to "make an honest and good faith effort to construct districts, in both houses of its legislature, as nearly of equal population as is practicable."
From the beginning the Reynolds Court acknowledged that states could continue to place some reliance on political subdivision lines, at least in drawing the lines for state legislative bodies. "Since almost invariably there is a significantly larger number of seats in state legislative bodies to be distributed within a state than congressional seats, it may be feasible to use political subdivision lines to a greater extent in establishing state legislative districts than in congressional districting while still affording adequate representation to all parts of the State." A further reason for at least limited adherence to local political subdivision lines is the highly pragmatic proposition that, to do otherwise, "[i]ndiscriminate districting, without any regard for political subdivisions, may be little more than an invitation to partisan gerrymandering."
Acknowledging the principle that population deviations are permissible in state districting implementation of rational state policy, the Supreme Court has recognized that de minimis numerical deviations are unavoidable. Maximum deviations in Connecticut of 7.83 percent among house districts and 1.81 percent among senate districts were upheld in Gaffney v. Cummings (1973). Texas deviations of 9.9 and 1.82 percent respectively among house and senate districts were similarly approved in White v. Regester (1973). In mahan v. howell (1973) the Court upheld a Virginia plan despite a maximum deviation of 16.4 percent, on the grounds that the plan could "reasonably be said to advance the rational state policy of respecting the boundaries of political subdivisions," but cautioned that "this percentage may well approach tolerable limits."
The requirement of population equality is far more exacting in the drawing of congressional district lines. In Kirkpatrick v. Preisler (1969) the Court struck down Missouri's 1967 Redistricting Act despite the fact that the most populous district was 3.13 percent larger and the least populous 2.84 percent smaller than the average district. In explanation the Court stated, "Since 'equal representation for equal numbers of people [is] the fundamental goal for the House of Representatives,' the 'as nearly as practicable' standard requires that the State make a good faith effort to achieve precise mathematical equality. Unless population variances among congressional districts are shown to have resulted despite such effort, the State must justify each variance, no matter how small." In Karcher v. Daggett (1983) the Supreme Court invalidated a deviation of less than one percent among New Jersey congressional districts because the state had failed to make "a good-faith effort to achieve precise mathematical equality" in population among its congressional districts. In sum, because local units of government are less important as factors in the representation of relatively large numbers of persons in the Congress than for smaller numbers of persons in the state legislature, population deviations among congressional districts are strictly scrutinized, while a more tolerant review is accorded state districting. But even in state districting the excesses of the past are no longer tolerable; above the de minimis level deviations must be held within narrow limits and must be justified in terms of preservation of political subdivisions, compactness and contiguity of districts, and respect for natural or historical boundaries.
No matter how close the judicial superintendence of population equality, one problem remains. In congressional and state legislative districting alike, even the most exact adherence to the equal-population principle does not assure protection against legislative line-drawers who seek partisan advantage out of the process. " gerrymander " is the term used to describe such efforts to preserve partisan power or to extend such power through manipulative use of the process. The term originated in 1812 in Massachusetts, where political maneuvering had produced a salamander-shaped district which was named after elbridge gerry, then governor. From that time forward the gerrymander has been an altogether too-common fact of American political life. Nevertheless, despite repeated attempts to persuade the Supreme Court to enter this new "political thicket," the Court has denied that there is any constitutional ground for superintending the apportionment and districting process other than the equal-population principle. Accordingly, the states remain free, so far as the United States Constitution is concerned, to construct congressional and state legislative districts that resemble salamanders or other equally peculiar creatures. And many state legislatures have done just that, particularly where one party is in secure control of the state legislative process. Where party control of the two houses of a bicameral legislature is divided, or where the governor is of a different party, the drawing of congressional and state legislative district lines is likely to be worked out by political compromise or, that failing, by the courts.
More seemly alternatives are possible, but they are not often adopted in the absence of judicial review over the process except as to population equality among districts. Congress has the authority to set standards of compactness and contiguity that would avoid the worst abuses and could be enforced in the courts. State legislatures could adopt similar standards to control the process within their own states, but few political leaders are willing to relinquish the prospect of present or future partisan advantage to be secured out of the districting process.
Like state legislative districting, the districting of counties, municipalities, or other local governmental units is constitutionally permitted to deviate to some extent from full equality if it can be demonstrated that the governmental unit has made "an honest and good faith effort" to construct districts "as nearly of equal population as practicable." Local governments may use multimember districts if there is a history of such representation and if such plans are not part of a deliberate attempt to dilute or cancel the voting strength of racial or political elements in the governmental unit. Despite that limitation, local governments, like states, may use ethnic or minority population data in constructing districts designed to elect representatives of that minority or ethnic group. (See united jewish organization v. carey.)
Supreme Court intervention in the apportionment and districting process has unquestionably restructured congressional and state legislative representation. Gross population disparities among election districts have been evened out so that the democratic promise of fair representation has been made possible of realization. But no court, even so powerful a body as the Supreme Court of the United States, can assure democratic representation. The ultimate test of the democratic process will depend upon the level of concern of the voters and their willingness to insist that their legislative representatives take whatever action is necessary to prevent excesses.
There are two principal types of gerrymandering, both of which should be controlled. The bipartisan or "incumbent survival" plan is designed to assure as far as possible the reelection of incumbents, sometimes regardless of party affiliation; the technique is to distribute party registration or proven party supporters to the legislators who will benefit most. The partisan plan is designed to maintain or increase the number of seats held by the majority party; the technique is to "waste" votes of the opposition party either by concentrating the voters loyal to that party in as few districts as possible, or by dispersing the opposition voting strength among a number of districts in which it cannot command majorities. Control of these abuses is not likely to come from party leadership. Voters concerned with the integrity of the process must demand an end to such practices, calling for state constitutional amendments or statutes requiring that districts be compact and contiguous.
Redistricting should be a matter of special concern for ethnic and racial minorities, many of whom are concentrated in urban centers. Typically, minority spokesmen claim that fair representation requires districts that will elect members of their own groups. When legislatures act to meet such demands, other groups are likely to feel disadvantaged. That issue was litigated to the Supreme Court in United Jewish Organization v. Carey. In that case a New York redistricting plan had been modified to bring it into compliance with the voting rights act of 1965. In the process the act divided a community of Hasidic Jews in order to establish several substantially nonwhite districts. The Court upheld the plan, ruling that such a use of racial criteria was justified in fulfillment of congressional legislative policy in the Voting Rights Act.
Somewhat related to the issue just discussed is the question whether municipalities and other local legislative bodies should be permitted to require at-large elections for all the seats in the legislative unit. Such a practice may make it impossible for a minority group in the community to secure representation, even though one or more members of that minority might be elected if single-member districts were used. The Supreme Court held, in mobile v. bolden (1980), that multimember district elections would be tolerated, even where the impact on minority groups was demonstrated, unless it could be shown that the plan was adopted with racially discriminatory intent. However, the Voting Rights Act of 1982 overturned that ruling; under the act, invidious intent need not be shown if impact disadvantageous to identifiable minorities can be established.
In the era before the application of computer technology to politics, it was common for politicians and their staffs to spread out maps on office floors, using adding machines for their arithmetic, slowly building new districts from census tracts and precinct figures. Because most redistricting decisions must be made sequentially—one boundary change requires another, which requires yet another—the computer is perfectly designed to speed the process and allow for more sophisticated analysis. The computer not only makes available numerical population counts, voter history, and party registration, but also permits a graphic display of the areas represented.
These technical advances have resulted in what may be styled the second reapportionment revolution. They place in the hands of those responsible for redistricting a vast array of information for use in drawing district lines. It follows, for better or for worse, that the computer's twin features of speed and accuracy can advance the goal of "fair and effective representation" as well as engineer the nearly perfect gerrymander.
At the time of the reapportionment decisions of the early 1960s, commentators speculated about the decisions' likely impact on the representational process. The most common prediction was that the urban areas would dominate state legislatures, with a general tendency toward liberal legislative policies. It is by no means clear that this prediction has come true. Enlarged influence of the suburbs, often with a conservative representation and not infrequently allied with rural representatives, has been the more typical reality. The one thing that can be said with confidence is that adoption of the equal-population principle has ended the worst abuses and assured basic fairness in the most important features of the democratic process.
Robert B. Mc k ay
Adams, Bruce 1977 A Model State Reapportionment Process: The Continuing Quest for "Fair and Effective Representation." Harvard Journal on Legislation 14:825–904.
Common Cause 1977 Reapportionment: A Better Way. Washington, D.C.: Common Cause.
Dixon, Robert G. 1968 Democratic Representation: Reapportionment in Law and Politics. New York: Oxford University Press.
Grofman, Bernard et al., eds. 1981 Representation and Redistricting Issues. Lexington, Mass.: D. C. Heath.
Mc Kay, Robert B. 1965 Reapportionment: The Law and Politics of Equal Representation. New York: Twentieth Century Fund.
Polsby, Nelson, ed. 1971 Reapportionment in the 1970s. Berkeley: University of California Press.