IN SEVERAL ways the preceding chapter on the public and personal interests forms an essential part of the study of representation and elections. The presence of numerous overlapping and often conflicting connections among men sets the stage for the struggle for representation. These interests will cause men to seek governmental forms benefiting their interests-be the interest so large as the safety of the community itself or so small as a tariff to keep out foreign-made shoelaces. Those community influences we have discussed-such as technological interdependence and national loyalties-and those separatist traditional, local, economic, religious, and other leanings we have described combine into dynamic movements to determine the forms of government. The more enduring and basic the interest, the more likely it is that men who share it will organize and wage political battle on its behalf. It is a grave error to think of representative government - or any other form of government-as a static organization, designed to give all conceivable desires equal weight in the determination of laws. In reality, representative government originated in a conflict of interests and is perpetually maintained and changed by the conflict of interests. This chapter will reveal such facts as they are discoverable in history and today. It will show how the numerous facets of men's political characters, described in the last chapter, cause men as individuals and in groups to promote representative and elective devices that help them achieve their ends.
Preparatory to sketching the career of representative government, let us define it and its companion term "representation." Representation is a relation between an official and citizen that exists whenever an action of the official accords with the desires of the citizen. Representative government is a government that makes great use of devices to ensure representation, such as the election of officials. The representation principle is built into the state; it is regularized, provided for as a normal procedure, and given certain important powers to work with.
The origins of representative government are still somewhat unclear. Early Greek democracies were governed directly by the citizens, not by elected representatives who were authorized to pass laws. When these democracies formed leagues for religious or warlike purposes, their delegates operated under strict instructions. The Roman Republic passed into the Roman Empire without employing the principle of representation, and, so say a number of scholars, one of the reasons for the failure of the Republic lay in its inability to "invent" the principle. The Republic of Rome, built on the city-state, town-meeting pattern, had no mechanical way of representing the people of their far-flung possessions. Tenny Frank, however, in studying the semi-independent Macedonian republics set up by the Romans, has concluded with reservations that "the Macedonian state had a republican form of government that was not very different from that of more advanced republics today."
We do not know whether the Macedonian and other ancient experiments in representation were handed down into the Middle Ages. There is more evidence to show that a second Roman device persisted and had some influence on the development of representative government. This device was the corporation, a legal fiction whereby an organization could be created with an existence apart from that of its members and with powers to perform tasks in their name. Ecclesiastical bodies, such as the Dominican Order, were organized along corporate lines and in several important cases held assemblies attended by delegates of their scattered component groups.
By the beginning of the thirteenth century, this idea of the corporation was adapted to a growing secular, political pattern in western Europe. Kings were reaching out for power. They were trying to consolidate realms more extensive than the average principality. Towns were growing up and had much to gain from some form of monarchical stability, for trade prospered with wider markets and longer periods of peace. At the same time, the wealth and populations of towns could help the Kings subdue feudal competitors. Feudal ways of financing the King gradually disappeared because they were not as flexible or as generous as periodic consultations and contracts between the towns and the Kings.
In the English countryside, we find the King extending his influence through his itinerant judges, who would call up inquest juries to describe what the local law was. "Twelve men, therefore," wrote an early English historian, "were chosen to make known the provisions of their laws and customs, so far as they were able, omitting nothing and changing nothing by deception." Commenting on these events, Maude Clarke, in Medieval Representation and Consent, writes: "Thus between 1066 and 1226 the principle of representation was, by means of the inquest and jury, elaborated to such a point that the knights of the shire were brought to the frontiers of political responsibility."
Both the untitled proprietors of the land and the town leaders, therefore, came to have a certain role in relation to the King. Both were useful administratively. Both provided political support to kingly aspirations. Both were affected by the idea of consent in this new, nonfeudal situation; they felt that new practices that were not covered by the law ought to be settled by bargaining. The King was given what he wanted most of the time, but not without some bickering, protests, and refusals.
The high nobility and clergy of the late Middle Ages were already entrenched behind their privileges. In England the Magna Carta legalized their position. They were frequently consulted and already constituted an assembly. In the thirteenth century, however, the Kings called in the commons, that is, the representatives of the untitled landed interests and the commercial interests of the towns. Early in the century, parliaments of the three estates-nobles, clergy, and commons-were called in Spain, Sicily, and France. In 1265, Simon de Montfort called to the English Parliament for the first time the three estates of England. The exact method of selection in the beginning is in doubt. In 1295, the writ calling the Parliament prescribed elections. Members of these parliaments were strictly controlled by their home constituencies. The suffrage was limited, and the elections were conducted by a number of different methods in the various countries and localities. Within the parliaments, the landed interests far outnumbered and dominated the town representatives, but town membership grew gradually in numbers and power.
There is some evidence that the idea of representative government was close to realization in the fourteenth and fifteenth centuries, even though representation was based on the estates, and not on the nation as a whole. Within this estates system the assemblies had considerable power and prestige. But everywhere the estates type of representative government declined as the Kings overpowered the nobles and reduced church influence. Only in England did the representative form of government survive, though weakened and pallid under the Tudors (1485-1603). Its tenacity and durability there prepared the way for important changes in the locus of power because the government at least provided a tolerant atmosphere for the growing influence of commerce and of the towns. Elsewhere those promoting the causes of the middle classes and of the "rights of man" could find no ladder of power up which they might climb. In England they could move, slowly, it is true, and not without great effort, into positions from which they could protect their interests. Finally in 1689 a combination of landed, commercial, and religious interests in England, represented in the House of Commons, determined that the ultimate legislative authority should be the Parliament, not the Crown.
However, this form of government, although it used representation, was an oligarchy, not a representative government as we define this term-direct control of the assembly by a large electorate. The controlling forces in the Commons of eighteenth-century England were the agents of a tiny group of wealthy and often titled men.
Modern representative government arose out of bitter disputes over the relation of the representative to his constituency. One group proclaimed that the representative is a free agent, attached only by the bonds of good will and remote influence to his constituents. The opposing group demanded that the representative should be elected, restrained, and tutored by his constituency. Since the first view conformed to the existing representative situation in England, its advocates might be expected also to be defenders of the status quo.
This is what in fact occurred. Edmund Burke, the foremost spokesman of the established system, declared in 1774 that "mandates issued, which the member [of Parliament] is bound blindly and implicitly to obey, to vote and to argue for, though contrary to the clearest conviction of his judgment and conscience; these are things utterly unknown to the laws of this land, and which arise from a fundamental mistake of the whole order and tenour of our constitution."
But Burke's order had not long to live. In the last part of the eighteenth century and in the early nineteenth century, growing opposition to "irresponsible" representatives induced Parliament finally, in 1832, to reorganize the whole electoral system of England, giving many more persons the vote and evening out the number of people who stood behind each member of Parliament. Exact obedience to instructions or mandates was not demanded by law, but the representatives became much more sensitive to sentiments of their constituents than their predecessors had been.
Meanwhile the English development was surpassed in America and France. In early nineteenth-century America, the vote was given to all white males, elections of most government officials were held at brief intervals, and, in some states, the constitutions even authorized the constituents to instruct their representatives (though in practice, this was almost impossible). The Jeffersonian ideal, the belief in "the people acting directly on all things concerning them," came to be held by millions. The egalitarianism that Burke thought dangerous became a cardinal belief of the peoples of America, England, and France. Declared Jefferson, " Equal representation is so fundamental a principle in a true republic that no prejudices can justify its violation."
The twentieth century, prolonging and developing the trend of the nineteenth century, brought representative government to most parts of the world. This triumph, however, has not simplified by much our problem of analyzing governmental institutions. Not only are there still many significant differences in the structures of the governments of the world, as seen in their constitutions and laws, but there are wide differences in practices to be found among governments whose structures superficially are quite similar. In this chapter, we shall have a good deal to say about certain of these differences. We shall answer the questions: Who has the right to vote? How many people exercise their right to cast ballots? How much power does the vote give to the electorate? How are the voters grouped, that is, apportioned, for the purposes of electing officers? How is the balloting for officials conducted and how are the ballots counted?
A discussion of these questions will take us a good distance toward understanding the different forms of government and political practices. But we shall have to devote a chapter apiece to political parties and pressure groups before we can adequately understand the forces working on systems of representation and elections.
As has been shown above, much of the struggle for representation in government takes place through the fight for the vote in which different social groupings in society try to exercise disproportionate influence. Many think that every man, by virtue of being homo sapiens, ought to be allowed the vote, while others think the vote should go only to those who own land or some other form of personal wealth. Other claims have been advanced for and against women, minorities, the young, the noncitizens, the nonresident, and so on. Each new restriction and each abandonment of an old restriction shifts some amount of influence from some interests or values to others. Generally speaking, all adults of all politically organized communities who are not affected by certain limiting disabilities have the right to vote in equal measure. Some interesting exceptions to this generalization occur with respect to (1) property and taxes, (2) sex, (3) race, (4) education, (5) citizenship, (6) residence, and (7) age. Selected examples will show the things men wish to accomplish by suffrage restrictions. We will also need to discuss plural voting and compulsory voting.
In some countries the electorate is universal except for a few scattered property and tax restrictions. The poll tax in several of the Southern states of the United States is an example. Although the poll tax is usually only one or two dollars per year, many potential voters find other ways of spending their money more gratifying or forget to pay it before registration and election time. It then accumulates and twice as big a sum must be paid the next year. In many American localities, only property holders are allowed to vote on certain kinds of bond-issue proposals and other financial propositions.
All major countries now grant the vote to women. Generally, throughout the nineteenth century, liberals and radicals fought for women's suffrage on grounds of natural rights, equality, individualism, and socialist theory. But the parties of the left in the twentieth century have been cold towards women voting, believing (with some reason as we have seen) that the newly enfranchised women would vote more conservatively than the men as a whole.
Before the wars and revolutions of the twentieth century, nations composed of more than one nationality, but dominated by only one, limited the vote mainly to the dominant nationality. This was true, for example, of the Austro-Hungarian Empire. The United States and the Union of South Africa, both composed of more than one national strain, discriminate against Negroes. In the United States, the legal theory forbids such restriction, but the tradition of slavery and civil strife has made the law a dead letter in a number of Southern localities.
Repeated opinions of the Supreme Court directed against overt or semiovert discrimination have failed to give the Negro impartial treatment at the polls. Intimidation, apathy, strict enforcement of literacy requirements for the suffrage against Negroes (but not against whites), refusal of the Democratic party to tolerate Negro members, and other nonlegal but nevertheless strong social pressures have effectively blocked a large percentage of the population of certain Southern states from voting. Yet in the Northern and Western states, Negroes vote freely, and even in the South they are voting in ever-increasing numbers.
In South Africa, the whites have prevented the more numerous Negro population from voting at all. In the Soviet Union, many races vote without hindrance. In Brazil, the Negroes and Indians, who are very numerous, have the vote, and so it is with a number of other Latin American and Caribbean countries.
Most countries have abolished educational requirements and even literacy tests for voting, although a few American states, Northern as well as Southern, have literacy requirements. To require more than literacy, that is, the ability merely to read and write on a low level, would introduce new political issues, for everywhere, people of means are more educated than the poor. The former can afford an education, and education itself promotes economic and social success. To require even a high school education in most countries would disfranchise large segments of the poorer classes and weaken the political parties of the poor. Therefore, no radical or liberal party can consent to an educational test, even if it seemed abstractly desirable. Furthermore, it would be extremely difficult to demonstrate any connection between education and "political intelligence"-in political disputes one side usually regards the other as politically unintelligent.
Citizenship has almost always been a prerequisite to the right of suffrage, from the Greek city-states to modern times. Frontier or revolutionary nations provide a few exceptions. Foreigners voted and held office during the French Revolutionary Period and in the early days of the Soviet regime. Among frontier states, the United States (at one time), Argentina, Chile, and Australia have afforded examples of alien suffrage.
Many believe that, in order for a person to use the vote intelligently, he must have dwelt in the same locality for a certain length of time. Citizenship, of course, satisfies the basic need for some residence requirement. Conservatives have frequently expressed the opinion that longer residence requirements promote stability and responsibility in local and national government.
Ideological considerations aside, however, residence requirements of some minimal sort are necessary to allow time for registration lists to be prepared and checked. Otherwise "floaters" might wander from polling place to polling place on election day, voting at will. Each American state sets its own minimum residence; the period ranges from six months to two years. In addition, somewhat shorter periods of residence in the polling district itself are also required. In England only a three-month residence period is required.
A number of countries and states allow voters who are to be absent from their locality on election day to vote ahead of time or to vote by mail or proxy. When large standing armies or navies exist, how to arrange for their vote becomes a difficult problem. France before World War II denied the armed forces a vote for fear of military participation in politics. English and Soviet military forces have special arrangements to facilitate their voting. American soldiers, sailors, and airmen depend entirely on the laws of their home states.
A final and exceptional residence disqualification occurs in the District of Columbia. Residents of the American capital city are deprived of any right to vote except in so far as they retain residence in another state. The reason given is that being at the heart of the government the residents could exercise political influence beyond their numbers. The result is that, even though their number exceeds that of nine states, they cannot influence the vote at all (except those who continue to maintain residence outside the District). It is true, however, that the voting residents of Paris and London are reputed by some to rule the rest of their nations in their own interests. Still, in denying the voting right under any circumstances, the whole gamut of democratic theory must be opposed or answered. Positive evidence on both sides is available, but it is insufficient and vague.
In most countries, the age of acquiring the vote and the age of acquiring legal majority is the same-twenty-one years. A few nations require voters to be twenty-three to thirty years old. Others establish eighteen and twenty years as the voting age. The United States holds to twenty-one years with the exception of Georgia and Kentucky (and Guam) where eighteen years is the voting age, and Alaska where nineteen years brings enfranchisement. France, Britain, and Italy maintain the twenty-one-year line. The Soviet Union and several of the new governments formed since 1945 have set eighteen as the voting age, and considerable opinion in the United States and elsewhere advocates the same age. The chief argument is that military and job responsibilities begin more often at the age of eighteen than at twenty-one.
Age requirements, like most other requirements, have indirect political consequences. Younger people vote less conservatively than do older ones, according to studies of Swedish, Swiss, German, Danish, Dutch, and American elections. Opinion polls in America indicate the same trend; the youngest age groups tend more to favor the Democratic Party, whereas the oldest age groups tend to favor the Republicans.
Plural voting is the legal right to cast more than one vote in an election. A plural vote may be granted for a double residence, as formerly in England. Belgium had before World War II a law conferring an extra vote on the heads of families who possessed some wealth, education, or paid some taxes. In Prussia from 1849 to 1918, the total body of men over twentyfive were divided into three classes of voters, each class drawn to contain one third of the total taxes paid to the state. Each class could then elect one third of the members of the Land tag or parliament. As a result, about 6 per cent of the people, who paid one third of the taxes, elected one third of the representatives. In recent years there has been an attempt in France and Italy to give the head of a family more political power than single persons.
Scattered through the history of suffrage one finds other qualifications of the right to vote. Catholics were excluded in a number of Protestant countries until recent times. Jews received similar treatment in various countries: in England for a time, until 1923 in Rumania, in Germany before 1848 and during the Nazi period 1933-45, and so on. The Greek clergy were forbidden to vote before World War 11. The Mexican Constitution of 1917 prevented the clergy from voting. The clergy in the Soviet Union were likewise disfranchised. English peers, possessed of their own personal "representation" in the House of Lords, cannot vote. Persons unfavorable to the existing regime are denied the vote in the Soviet Union, the Soviet bloc of nations generally, and in other countries where elections are used to give "applause" to a powerful ruling group. For a time, men who actively supported the Confederate States of America were disfranchised by the victorious Union Congress. Persons judged morally incompetent or criminal, and insane persons, are generally prevented from voting.
In general, we may conclude that the world's representative governments today give the vote to everyone. The remaining restrictions described above are not of great practical importance to the total picture. But we must not believe that the vote in itself guarantees that total power, even of an indirect sort, rests with the public. The vote is only one of the many devices used by representative governments. Universal suffrage means only that men can participate alike in the casting of ballots.
Despite centuries of controversy over the right to vote, a large number of individuals will not vote unless driven to do so. The United States lags about a dozen points behind the countries of western Europe in the percentage of eligibles voting. In Soviet Union elections, where great official pressure is exerted to get out the vote, participation is close to 100 per cent. Something of a modern high in electoral participation in a free election occurred in republican Germany in 1933, when 88.7 per cent of the electorate voted. (This was also the German Republic's last free election before the Nazi repression.) One of the lowest records in a general election since World War I was the American presidential election of 1920, when somewhat less than half of the electorate participated. In American state and local elections, low participation is the general condition, especially in primaries.
Studies of nonvoters have given us a number of generalizations about why certain countries and certain groups within countries participate more heavily in elections. Under most circumstances, new voters vote less than those who have exercised the franchise for some time. This has been true of the working classes, the women, Negroes, and naturalized citizens of the United States. The factors at work are mainly timidity, remnants of social disapproval of the recent change, ignorance of the new procedures, and lack of interest in the new instrument of power. Larger proportions of educated individuals vote than of noneducated people, presumably because the schools of a nation stimulate interest in public affairs. More people of higher income vote regularly than do the people of lower income. Some element in this difference may arise from the feeling among higher-income earners that they have more at stake in government policy. A smaller proportion of young eligibles vote than do older eligibles except for the over-sixty age group, wherein participation declines somewhat. This may be due to a preoccupation of the young with getting a start in life and a feeling of incompetence at facing a new task.
All of these propositions must be used cautiously. For example, education, American birth, whiteness of skin, and higher income commonly are found together in the same groups in America. It is then a difficult statistical-analytic problem to assign the proper weight to each factor.
Various political and social conditions also seem to affect the amount of over-all participation in elections. Foreign or domestic crisis increases the percentage of participation. Individual excitement is expressed in the collective act of joining other citizens at the polls. The more important the elections seem to be to the voters, the greater the participation; that is, a presidential election or a British general parliamentary election will turn out more voters than local elections. Usually, elections of the top officers of the state draw out more voters than elections of lower-level officers.
The proportion of eligibles who vote in an election increases as the voters believe the race for office will be a close one. Participation is usually greater in elections the results of which definitely select the winner than in preliminary elections or primaries. Multiple-party systems draw greater participation than one- or two-party systems, perhaps because the range of choice is larger in the former systems, at least in the eyes of many voters who might not otherwise see any point in voting. Systems of proportional representation (see p. 180) increase participation.
Participation varies directly with the simplicity, intelligibility, and easy availability of the electoral machinery. "Get-outthe-vote" campaigns have a positive effect on participation if the campaign is adroit. A vigorous labor movement increases participation. The unions get out the working-class vote.
A tradition or habit of participation, formally or informally learned, tends to maintain a high level of participation. Finally, the general social evaluation of the vote as (a) precious, (b) useful in coping with the perceived political problems, and (c) authoritative affects the rate of participation. If voters are confident that an election will solve important problems, they will vote in larger numbers.
The experience of the few countries that have forced their citizens, by some mild but well-administered sanction, to vote shows that a greatly increased turnout results. A mere declaration of policy is not enough. In Australia, the adoption of a compulsory voting law in 1924 imposing a $10 fine on nonvoting brought an increase in participation from 59.4 per cent in 1922 to 91.4 per cent in 1925, 93.6 per cent in 1928, and 94.9 per cent in 1929. Belgium, Holland, Czechoslovakia, Spain, and the Argentine Republic also experimented with compulsory voting.
American opinion seems to shy away from the idea, although drives to "get out the vote" are part of every American election campaign. Professor Munro once declared the "slacker" vote was not worth getting. But that assumes a connection between intelligent voting and the act of voting that is hard to prove. In fact, in American elections, a light participation usually means a heavy proportion of machineorganized to "independent" votes. Several American colonies at one time required electoral participation. The voters of Massachusetts and the legislatures and conventions of several states have rejected compulsory voting laws in recent times.
In America, nonvoters are more common, though the difference is not striking, among the poorer, less-educated groupings. This means perhaps that compulsory voting would favor slightly the Democratic Party. In Europe, by contrast, the Conservative parties, not blessed with the natural organizations afforded left-wing parties based on trade unionism, have seen in compulsory voting a way of getting their supporters to the polls. Gosnell believes that compulsory voting would tend to lessen the power of political machines in American local elections. Primaries and many crucial general elections find only a small proportion of the potential voters coming out to choose mayors, aldermen, judges, sheriffs, and even statewide officers. Only the presidential elections in America bring out two thirds or more of the voters regularly.
How much the vote means depends in part upon how much legal authority is given the voters. Some of the devices that limit the powers of the voters are: (1) the existence of many nonelective offices, (2) constitutional limits to the events the electorate may influence, (3) bicameralism and other checks and balances on the free play of the electoral will, and (4) the principle of separation of powers to keep a popularly elected organ of government (for example, the House of Representatives) from dominating an appointed organ of government (the Supreme Court). Creating a "short ballot" for example, in the interest of "administrative efficiency," may give the electorate controls over a very few top officers but change and perhaps diminish their controls over a number of minor offices, previously elective. A constitution like that of the United States, which is hard to amend, prescribes freedom of the press, assembly, contract, religion, and other rights; the electorate is limited in the effects it may have on those rights. When one house of a bicameral assembly must pass on another's bill, the possibility of a check to the electorate is enhanced, especially when the terms of the offices of the members of one house are longer than those in the other or staggered in such a way that the appearance of a new, popularly chosen majority in one house often encounters an incumbent opposing majority in the other. The American, British, French, Italian, and other bicameral systems employ these checking devices or have used them in the past.
Despite these checks upon the electorate, the meaning of the vote over a historical period is clear. The extension of the suffrage to new groups has been associated with changes in the composition and behavior of officials of the state. We may cite Negro advances within Northern cities of the United States, the great impetus given the Labour Party in England by the suffrage reform of 1918, and the passionate opposition to the extension of the franchise to new groups by the privileged classes wherever the issue arose. However, mere multiplication of votes does not accomplish changes; new social strata must be involved. The extension of the franchise to women was followed by no recorded visible results in the United States, but it seems to have given some strength to centrist and religious parties in certain European countries. (However, even though it brought no startling changes in the legislative sphere, we should not ignore the profoundly democratic meaning of the franchise for women who were working out adjustments to their changed situation in modern life.)
Suffrage has a massive, general influence, rather than being an intelligible source of specific decision, and it deserves respect as such. The simple large fact that the vote can cause extensive movements of political personnel has been sufficient to give it a high value in the minds of that personnel and consequently in the minds of the voters. The behavior of officials is conditioned by a deep respect for it, and a large body of myth surrounds its use. An election is like the morning roll-call in the army; it brings all public officers to attention, and, though it must discipline the "good" ones, it also forces the "bad" ones into line.
The kind of ballot used in an election is also of some importance to the representative system. Although a century ago most elections were conducted viva voce, that is, by a verbal expression of preference for a candidate before the polling officials, the secret ballot is almost universal today. Even in absolute regimes that exercise complete controls over the elective process to make certain the election of the regime's candidates, some kind of "secret" ballot may be used. Then outside pressures are counted on to force the election according to the regime's wishes. The secret ballot, when printed by the government, is known as the Australian ballot from its country of origin. Its use is practically universal today, although it was not long ago that each party printed its own list of candidates for the various offices or the voter wrote up his own choices.
The state today also regulates the way in which the names appear on the ballot and the way in which a voter has to mark his ballot. The so-called Indiana ballot and the European list system ballot have a major point in common. In both it is possible for a voter to make only one mark that automatically casts his vote for every candidate of the party for which the mark is made. This encourages "party regularity," that is, adherence to the party all the way down the line. In practice, the effects of the Indiana ballot are probably quite small because a voter who is determined to be independent may easily check candidates in more than one party. The "straight-ticket" voting fostered by the Indiana ballot is not encouraged by the Massachusetts ballot, which lists the candidates in office groups in the same column, according to whether they are running for President, Vice-President, congressman, mayor, or alderman, regardless of party affiliations. Then the voter must pick and choose; if he is a party regular, he must take the trouble to search out with his pencil the candidate for each office who represents his party.
The administration of elections in most jurisdictions of the world is a technical operation, publicly operated and directed with no relationship to the parties. In the United States, however, the administration of elections is partly in public hands and partly in party hands. Three criteria are commonly used to determine the "efficiency" of systems of election administration. The "efficient" system allows the voters to participate with a minimum of personal inconvenience by providing them with convenient polling places, easy registration and identification procedures, and in some cases a legal holiday from work to cast their votes.
"Efficient" systems prevent frauds. The American system operates on the principle that by using officials of the major parties to conduct the polling, each will watch the other and fraud will be prevented. The European, including the English, systems have nonpartisan state employees doing the work.
Systems are also "efficient" if they get the job of balloting done in the shortest possible time with the least expenditure of money.' Here possible contradictions arise, however, between the different objectives. The convenience of the voter often means greater expense. Using untrained personnel as in many American jurisdictions also means greater expense. Furthermore, in order to avoid the intervention of "foreign" officers and jurisdictions, the American elective machinery is extremely decentralized.
Practically all other elective systems are extremely centralized, as in the cases of France, Italy, the U.S.S.R., and, to a lesser extent, England. In these countries the local polling places are manned and managed by persons responsible to distant authorities. Historical evidence shows that many local frauds have occurred in the American system of decentralized administration. On the other hand, in the German Republic a system that was efficient in terms of cost was converted rapidly into a perfect instrument of central control by a dominant party. However, only careful study and expert judgment can decide in any certain case whether the disadvantages of the one system are graver than those of the other.
A comparison of the American and the Soviet systems of election administration shows remarkable differences. The American is almost completely federal, while the Soviet system is directed in a stringent fashion from one Central Electoral Commission of the Soviet Union, which is appointed and instructed by the Central Executive Committee of the Soviet Union. The American states generally delegate practically the whole of the election operation to the counties, sometimes not even bothering to keep adequate records of all local elections, much to the despair of political scientists. The Central Electoral Commission of the Soviet Union, however, directs the activities of central electoral committees in each constituent republic, which in turn supervise the work of local electoral commissions in the villages and in the farm and factory soviets. The Communist Party is active at all stages of the Russian electoral process, although in most instances the elections are held without coercion and in secrecy at the polls. In America, the parties are active everywhere and have a legal responsibility in the conduct of elections that the Communist Party in Russia does not have. The Communist Party, standing allegedly outside the administration of elections, limits in advance all significant free choice of the voters at the polls.
Although the suffrage is a highly important device to produce representation for the different social groupings of a community, other devices require attention as well. Such devices are subsumed under the general subjects of electoral constituencies and balloting methods.
A wide range of possibilities for intertwining and adjusting values and interests opens up in the field of constituency construction and in the balloting process. A constituency is the group legally charged with the election of an officer. Given the qualified voters and the officers to be elected, how are the voters to be apportioned and how are the ballots to be drawn? Both processes can be carried out in many ways. We shall discuss apportionment first.
Apportionment is the division of qualified electors into constituencies. Voters may be apportioned by governmental boundaries, territorial surveys, official bodies, functional divisions of the population, or by personal or "free" population alignments. A combination of two or more is often encountered.
Governmental boundaries always form the ultimate basis of the constituency. Thus the United States as a whole is the constituency for the election of the President and Vice-President, speaking realistically. Speaking legally, the constituency of the President and Vice-President is an official body known as the Electoral College composed of men elected by the people of each state. The American governors are all legally and in fact the representatives of governmental constituencies composed of the whole state. Some of the American state senates are elected from counties, which are governmental areas of some independent powers. The American Senate, formerly elected by the state legislatures, each of which chose two Senators, is now elected by the entire electorate of the individual states, and thus is based on governmental boundaries.
Territorial surveys are the favorite method of constructing constituencies in modern times. "Artificial" areas are cut out of the map, and their populations serve as constituencies. The survey method is popular because it is the easiest way to conform to the demand that all men be treated alike, that each vote weigh as much as every other vote, and that no subgroup of the population (race or class, for example) should be differentiated from the others for the purposes of voting. This was a specific demand of the Levellers, direct democrats of 1647, who in their first "Agreement of the People" declared: "That the people of England being at this day very unequally distributed by Counties, Cities, and Burroughs, for the election of their Deputies in Parliament, ought to be more indifferently proportioned, according to the number of the inhabitants."
"Indifferently proportioned, according to the number of the inhabitants" has become the standard principle in most representative governments today. Most election districts tend to be drawn to enclose equal numbers of inhabitants (certain exceptions will be noted in this section).
Official bodies as the apportioned unit are of two general kinds, ad hoc and permanent. In a number of representative systems we find special bodies called together for the primary or sole purpose of electing another body. This is the object and function of the American Electoral College. Modeled on the American idea, which Condorcet so admired, was the French system of indirect elections of the Senate (now the Council of the Republic) by an electoral college composed of various officials and persons designated by town councils. Permanent official constituencies take many forms. The Speaker of the House of Representatives is elected by the members of the House, who thus constitute an official constituency. The French President is elected by the two chamhers of the French Parliament. The English Prime Minister is elected by the House of Commons, though nominally appointed by the King. The German city councils elect their burgermeisters, some American city councils elect the city managers, and so on.
Relatively few regions construct their constituencies by functional divisions of the population. Under the Fascist regime in Italy there existed a weak legislative body known as the Chamber of Corporations, which was based on constituencies of major occupations. The American National Industrial Recovery Administration from 1933 to 1935 gave certain powers to groups composed of the representatives of the various firms in particular industries. Russian factories until 1936 returned members to the local soviets or councils, which in turn sent members to the higher councils. English university men used to have special representation in the House of Commons. The Irish and Portuguese legislatures are based partially on functional representation.
A final way of dividing the population into constituencies is by allowing the population to create its own constituencies during the voting process; that is, the voters are told that they may vote for whomsoever they please among the candidates up for election, regardless of their place of residence, functional occupations, or other spatial or social characteristics. Of course, there is practically always some spatial limit to the electorate itself, such as national boundaries; a pure personal constituency does not exist. But within the larger boundaries, free choice is allowed. The mechanisms by which this is accomplished will be dealt with below. We need indicate here only the two basic requirements for free constituencies: more than one representative must be chosen, and a limited quota of votes must be enough to elect a candidate.
Two irregular but widespread phenomena occur and create a pattern of constituency population different from the results of the apportionments already mentioned. These are "rotten boroughs" and "gerrymandering." Rotten boroughs are historical accidents that give disproportionate weight to the votes of a thinly populated district in relation to a thickly populated one. Strictly speaking, a rotten borough is such only when the prevailing principle or belief about constituency construction is that it should be on the principle of equal populations. Thus the famous English rotten boroughs, which lasted for centuries, were originally boroughs of considerable population that declined to practically nothing while other towns of considerable size grew up as a result of the Industrial Revolution and other reasons. The declining centers kept the same representation, and the growing centers gained no new representation. To those who believed even remotely in some connection between representation and population, this situation appeared most unjust. The Reform Act of 1832 established the English representative system on the equal population principle of territorial constituencies. All subsequent legislation reinforced the idea.
Some American writers have quipped that the American Senate is the "worst rotten borough in the world." This statement is not accurate, because the Senate was never designed to be based on the principle of equal population; all states, large and small, were supposed to have equal representation in it. However, the United States has its full share of rotten boroughs in practically every state that has a considerable urban population. All states based their construction of constituencies in large part on the notion that the population of each should be the same. But the plan depends, for practical effect, on periodic reapportionments in order to adjust to population changes. As a consequence, rural regions must give up memberships in the legislatures if new memberships are to be given to the growing urban regions. This the rural representatives have been loath to do, and as a result, in states with large urban populations, such as New York, Illinois, Pennsylvania, and Minnesota, many more people are to be found in the districts of urban representatives than are to be found in those of rural representatives. In Illinois, reapportionment, supposedly mandatory each ten years according to the state constitution, was thirty years overdue in 1951. The largest district, according to a 1940 calculation, held 191.4 per cent more people than the average district, while the smallest held 54.7 per cent less than the average. In New York the situation is roughly the same; in Michigan it is a little less disparate.
Rotten boroughs are caused by a failure to reapportion according to the population principle, but gerrymandering is a positive act of malapportionment. The first violates the principle that equal populations should have equal weight, but gerrymandering violates the territorial principle that the boundaries of districts be "indifferently proportioned," that is, geometrically drawn without reference to the characteristics of the population. For gerrymandering is a process by which the dominant party in a legislature draws the boundaries of electoral districts so as to augment its strength in the assembly. By observing past electoral behavior and thus ascertaining where the consistent majorities of its opponents' voters and its own voters live, the dominant party draws the district boundaries in such a fashion as to concentrate its opponents' votes in as few districts as possible and to spread dependable majorities of its own voters over as many districts as possible. Sometimes this results in obvious distortions on the map such as the famous one committed by the party of Governor Gerry of Massachusetts when it formed a constituency that looked like a salamander. A cartoonist perceived the resemblance and called the malapportionment a "gerrymander."
The balloting process fulfills the process of the construction of constituencies that begins with apportionment. It is most convenient to group the types of balloting under systems of majority and plurality voting and systems of minority representation. Majority systems may be either expected as a matter of course or deliberately provided for. In countries having regularly only two parties running candidates for office, a majority is expected to elect its candidates and usually does. Where the presence of several parties or no parties at all means that there will be several candidates in the field contending for a single office, agitation arises for some system that will insure to the candidate elected a majority of votes. Where a third party or candidate musters a very small percentage of the total vote, say 5 per cent, and the victorious candidate wins by a plurality of 49 per cent to 46 per cent for his other competitor, the problem is not considered ordinarily to be serious. But where in similar cases the vote is regularly something like 36 per cent, 34 per cent, 30 per cent, then it is felt that steps should be taken to get the constituency's "real will," that is, to force a majority.
There are many ways to force a majority from the constituency. Three general types of forced majorities may be described. The first is the run-ofd election. In a preliminary election, several candidates contend against one another. The two with the most votes run against each other in a final election. This is found in the American South and in some localities where candidates are not allowed to list their party affiliations on the ballot. The French ballottage system allowed all candidates to run again if they pleased, but generally only the two highest pleased to do so.
A second general type of forced majority is the alternative vote. The voter is authorized to rank his choices among the candidates, giving the number 1 to his first choice on the list, the number 2 to his second choice, and so on. In the first count all the first choices of each candidate are totaled. If one candidate has a majority of all the ballots cast, he is declared elected. If he has no majority, and no one else has, the candidate with the least number of first choices is eliminated and the second choices on his ballots are distributed among the candidates who have received them. This may well result in one of the top candidates completing his majority. If so, he is declared elected. If not, the same process is repeated with the choices of the second to the bottom candidate.
Among the many systems of voting that give minorities representation, four may be presented briefly here. The simplest is the limited vote. Assuming that it is bad that a majority party should have all the voice of a district in the legislature inasmuch as it may have only won 51 per cent of the vote, a multimember district is created from which three or five representatives are to be elected instead of merely one. Then, where three representatives are to be elected, each voter is told that he may vote only for two candidates; or where five representatives are to be elected, he is told that he may only vote for three candidates. Thus a majority of the voters will under no circumstances be able to elect all three or all five members. Notice that the essential feature of minority representation in this and the other systems to follow is the "voluntary constituency." And, as we stated above (p. 177), voluntary constituencies require more than one representative to be elected from the district and allow each candidate to be elected with a quota rather than a majority of votes.
Simply providing for more than one candidate to be elected from a district is not enough. For example, the block system of election at large, under which every voter votes to elect several men from a single constituency, invariably becomes a majority or plurality election. For under such a system, the voters, instead of making up personal constituencies by scattering their votes, quickly gather or are gathered by managers behind a slate of candidates for all three, four, or however many offices are to be filled. Then the group with the largest number of voters wins all of the offices. The only regular methods of creating constituencies that are personal and free in the sense of our limited requirements here are the systems of minority representation that require only a limited quota of votes for a candidate to be declared elected.
A second type of minority representation is the cumulative vote. Three representatives are to be elected from each district. A voter is allowed to give three votes to one candidate, one and one-half votes to two, or one vote to all three, depending on whether he marks an "x" in the square of one, two, or three of the candidates in the race. Then the total points of each candidate are tallied, and the three highest are declared elected. Under this arrangement, a minority may instruct its adherents to mark only one "x" and that for their candidate. The majority party meanwhile is compelled to concentrate its strength on two candidates and, even if very strong, could hardly hope to give three times as many votes to each of three candidates as the nearest minority competitor could muster through his triple-strength supporters.
A third type of minority representation is provided by the so-called Hare or Andrae system of proportional representation (PR) by the single transferable vote. Like all other systems of minority representation here described, a multimember district is required. The desired object of this system of proportional representation is to give each group in society a number of members in the legislature proportional to its numbers in the population.
For example, let us say that five representatives are to be elected. Let us suppose for the moment that each major social group has a political party representing it. Let us also suppose that the parties rank as follows: Party A has about 20,000 voters; Party B has 10,000; Party C has 10,000; Party D has 7,000; Party E has 5,000; and Party F has 3,000. Accepting the premise that it is good to have all major groupings in the population represented in the legislature in proportion to their voting strength in the population, we would probably agree then that Party A should have two members, Party B and Party C one member each, and Party D one member, although we wouldn't feel too unhappy if Party E, instead of Party D, were represented with one member. How would we go about forcing this result?
First of all, we decide that no vote should count for more than one candidate. Otherwise we would have the block system, which is a multimember district in which the voters vote for as many candidates as have to be elected: invariably the party with the majority or plurality of the votes sweeps all the offices. We accomplish. this by having each voter designate his rank order of choices, one for his first choice, two for his second, three for his third, and so on.
We then would allow the voters to establish their personal
constituencies. We would say that any candidate receiving
over one fifth of the total vote cast ought to be elected, since
there are five men to be elected from the district. We therefore
set up a quota. The most common formula for arriving at a
quota is the Droop quota used in the Hare system of proportional
representation. The total vote cast is divided by one
more than the number of representatives to be elected, and
one is added to the quotient. The result is the number of votes
required to elect a candidate. In our example we find that the
total number of votes cast in the election is 55,000.
The number of men to be elected is 5 (plus 1 is 6). And 55,000 divided
by 6 is 9,166+-. Add 1 to this figure and the final quota is
9,167. Therefore, any candidate that gets 9,167 votes is bound
to be elected. For no matter how the vote is cast, providing
we insist that no voter's choice is to be counted for more than one candidate,
no more than five candidates can possibly get 9,167 votes.
Now we will show how the counting is done. We take a
count of first choices and find the result is as follows:
| Party A candidates | A1 | 8,000 |
| A2 | 8,000 | |
| A3 | 4,000 | |
| Party B candidates | B1 | 10,000 |
| Party C candidates | C1 | 9,167 |
| Party D candidates | D1 | 6,000 |
| D2 | 1,000 | |
| Party E candidates | E1 | 5,000 |
| Party F candidates | F1 | 3000 |
In this remarkable election, two candidates, B1 and Cl, have already made their quotas, C1 on the head, B 1 a little over. Next we take the surplus from B1, reaching into his barrel of 10,000 votes and extracting at random the difference between that number and 9,167. We examine the second choices on these 833 ballots and find they go to F1. We give them to Fi, making his total 3,833. Lacking any more surpluses and still having to elect three more representatives, we eliminate the low man. This turns out to be D2. But, as we might expect, the second choices on the D2 ballots all go to D1, giving him 7,000, but electing nobody for the moment. We go to the new low man who happens to be F1 with 3,833 votes. We take the next available choice on all of these ballots and find that 3,000 go to D1, 833 to E1. D1, with 10,000, is declared elected, while E1 has now 5,833 votes. We distribute Dl's surplus of 833 votes and find that they now go to E1, giving E1 6,666 votes. But it is too late to help him, for the next low man whose vote is to be dissolved is A3, and his 4,000 second choices go roughly 2,000 each to A1 and A2. Before they are even fully distributed, both these gentlemen are up to their quota and are declared elected. Their poll watchers have had some uncomfortable moments during the proceedings and probably have been regretting the lack of party intelligence and organization that put A3 in the field, but the results have finally come out as they should according to the party's strength. Party A elects two, B one, C one, and D one.
The fourth type of minority representation is the list system. It is a little less nerve-racking to the party managers than the single transferable vote system, for each party submits a list of its candidates and the voter, instead of wandering all over the ballot, simply designates the party of his choice. Then the quota needed to elect having been established roughly as in the Hare system, previously described, as many of the top names on the party list are declared elected as there are multiples of the quota in the total number of votes cast for the party list. In the example above, using the same quota, Party A with its 20,000 votes would again get two seats, Parties B and C would again get a seat each, but the results for Party D with 7,000 votes would be in doubt. To assign the fifth seat, following the Swiss system of list voting, we would give it to the party that would have the highest average vote per seat as a result of having it. Thus: A has 2 for 20,000; with 3 it would have an average of 6,666-I-. B has 1 for 10,000; with 2 it would have an average 5,000. C has 1 for 9,167; with 2 it would have an average of 4,583+. D with 0 for 7,000 would have 7,000 votes for its one seat, higher than any of the other averages, and would therefore receive the seat.
Generally, the majority or plurality types of balloting just discussed are found in North and South America and in England and its Dominions and are used in part in a number of European systems. The Hare system and other types of single transferable vote systems are found in scattered localities in the United States, the British Commonwealth, and the Scandinavian countries. The list system is popular in Europe but varies in details from country to country.
If we ask what kinds of values the various elective systems introduce into representative government, we can make certain partially substantiated statements. The majority election provides a working majority of members of a parliament, sharing many common views on issues. If it does not do so, the lack of a majority reflects aimlessness rather than any bitter conflict of party blocs. The systems of proportional representation or minority representation, on the other hand, when they do not produce majorities, tend to produce crystallized party blocs in the legislatures. Ordinarily a representative under a majority system has to appease a great number of his constituents, whereas a representative sent to the legislature by a certain quota of constituents can rest easy so long as he appeases that quota and may be little concerned with sparing the feelings of the rest of the community. The list system of proportional representation, on the other hand, can certainly be said to encourage strong party discipline, especially where a party determines whether a man goes on its list and assigns him his priority of election position on the list.
Before the coming of the Labour Party to power in England, it was often said that the majority election kept issues of a burning sort out of politics. That, of course, can no longer be said. But one may say that only heated issues that concern almost half or more than half the population will emerge readily in a majority system, while in a system of proportional representation any quota of the population can send its torchbearer to the legislature. A one-ring circus then becomes a several-ring circus. The exhibition that follows may be somewhat more disquieting.
It is scarcely necessary to say that no system of electing officers produces univeral satisfaction. Often a great many members of the public feel that no representative can ever be relied on to represent them adequately. At certain times in the past, this feeling has been reflected in proposals for special procedures to "return the government to the people." Among the devices favored to accomplish this task are the initiative, the referendum, and the recall. The initiative is an arrangement of government whereby a given number of people may demand that a particular proposal be acted upon by the legislature or be referred to the electorate as a whole for a referendum. The referendum is an arrangement whereby an initiated proposal or a bill of the legislature may be submitted for approval and enactment into law by the total electorate. Sometimes the legislature may itself call a referendum on a bill. At other times, any bill of a certain kind must be referred to the electorate. The recall is a procedure for allowing a prescribed number of voters to demand a special election to be held for the purpose of retaining or dismissing an elected officer whose conduct is at issue.
Hints of all three devices are to be found in American colonial history but were submerged after the American Revolution. The device of the New England town meeting, coupled with Swiss devices of the late nineteenth century, attracted many Americans who were anxious over losing control of their officers. It is difficult to generalize about the results of the three devices in a number of American governments that adopted them before 1920. However, we can say that it has not been proven that the "people" as a whole find them useful to get things done in accordance with the majority will. V. O. Key and W. W. Crouch, for example, concluded from a painstaking study of the California experience that "the initiators of propositions have usually been pressure organizations representing interests-commercial, industrial, financial, religious, political-that have been unable to persuade the legislature to follow a particular line of action." Also, Gosnell and Schmidt concluded from a general survey of "Popular Lawmaking in the United States" that participation or voting interest on issues brought up through initiative or referendum devices was generally somewhat below the interest expressed in the elections of candidates. There was also some variation according to the type of legislation proposed. More voters turned out to vote on legislation concerning morals and education, less on issues of revenue and public welfare, least on technical matters concerning governmental organization. The Swiss history of popular legislation has proved similar. A comparison of "popular" legislation with laws made by the legislature shows no striking differences in scope, novelty, moral quality, or radical majoritarianism.
Besides using many kinds of elections to gain representation, elements of the public have endeavored constantly to control the actions of political parties. If a single group of men dominated the major parties and the parties dominated the whole representative process, the universal suffrage and the many different representative devices would have small effect. Consequently, governments often force parties to admit members freely, to avoid corruption and bribery, and to allow representation to their rank and file. However, the subject of political parties is a large and important one, and we shall have to know much more about it. We ought now to turn, therefore, to the study of political parties, asking not only what they aim at and whom they represent, but also how they are formed, organized, and supported.