Cyclopædia of Political Science, Political Economy, and the Political History of the United States

Edited by: Lalor, John J.
Display paragraphs in this book containing:
First Pub. Date
New York: Maynard, Merrill, and Co.
Pub. Date
Includes articles by Frédéric Bastiat, Gustave de Molinari, Henry George, J. B. Say, Francis A. Walker, and more.
876 of 1105



PROPORTIONAL REPRESENTATION. Democracy has as its basis the right of the individual to be represented in the government of his country. This right has been distorted into the alleged right of a majority of men to be so represented, and to deny a like power to all others. Three causes have led to this: first, the fact, that all government rests at last upon superior physical force, and that in every civilized country a majority of its men is a stronger force than all the rest of the community; second, the belief that legislative bodies would find it too difficult to do even the little done now if the minority of the voters was fully represented (the common phrase of a "working majority" condenses this idea); and, third, the practical impossibility of representing all minorities, with the illogical deduction of the uselessness of representing any.


—The first cause would justify the seizure of a country and the overthrow of its institutions by any one who could persuade one more than half of its adult males to join him. The second is due to the existence in legislatures of a foolish partisan spirit, based on a desire to use public positions as party plunder, and to the non-existence of reasonable rules of procedure. Civil service reform will destroy absurd partisanship. Experience will create a proper procedure. It would be difficult, if not impossible, to represent all minorities. If 1,000,000 voters elect 100 representatives, one voter whose views differ wholly from those of all the rest, can not well be represented. But nearly all minorities can be represented, and should be, unless government of the people by the people is wrong.


—A crude form of minority representation often prevails, for a time, when a legislature is composed of two chambers. Whenever there is a liberal ministry in England and the house of lords dares to use its constitutional rights, this is true. The senate of the United States has not infrequently occupied the same position, notably in 1876, and again in 1883. Such a condition of affairs is often a good one in checking hasty legislation and insuring that public opinion, rather than public passion, shall be reflected in the statutes. But it is politically wrong. For it gives the minority more than proportional representation. It gives it a veto on the measures of the majority. Such a power, persistently used, would lead to revolution in any free country.


—A few figures may serve to show the injustice of the existing methods or election in the United States and in Great Britain.


—In the latter, two men have sat in the same house of commons, one of whom received 18,292 votes, and the other 69; ten successful candidates have polled 159,650 votes, while ten other successful ones polled 1,873, and ten defeated ones, 83,117; ten millions of the English people have elected 302 members, when twelve millions returned 187; of the lucky ten millions, 1,850,000 sent 81 members, and 3,008,000 sent 22; and 952,000 persons have returned 120 members, while 7,500,000 returned 96.


—In the United States, about 8,000,000 men voted, in 1882, for candidates for the forty-eighth congress. More than 3,500,000 of them, nearly 44 per cent, of the whole, voted for unsuccessful candidates, and therefore have no representation in the present congress. That is, the American system of majority representation practically disfranchised forty-four out of every hundred men to whom American laws gave the franchise. If we take the votes cast for Grant and Greeley in November, 1872, and divide each by the number of congressmen elected by the party in question, we find that a successful republican candidate required, on an average, 18,076 votes, while a liberal, to insure success, had to get 30,474. That is, majority representation made one republican vote worth one and three-fourths liberal votes. In 1866, when the fortieth congress was elected, one republican vote equaled two and one-fourth democratic votes. In 1872 the administration party received 55.93 per cent of the popular vote; the opposition, 44.07 per cent. But the respective strengths of the two parties in congress were 68.15 and 31.85 per cent. That is, majority representation added nearly forty-forty per cent, to the just congressional power of the majority. In 1880 it took 29,500 votes to elect a republican congressman, and 33,500 to give him a democratic colleague. In 1882 the successful democrat, got, on an average, less than 21,000 votes, while the successful republican had to poll more than 28,000. Of the 8,000,000 persons who voted in 1882, 1,792,000 elected 163 members of Congress, a majority of the whole body. If these 183 vote together on any question (which is merely improbable, not impossible), they can carry it , though they represent less than one-fourth of the voters who took part in the election, and less than one-fifth of the voters in the country. A minority of less than one-fourth would then rule the nation, and perhaps dictate its policy for a term of years. To prevent the possibility of minority rule, minority representation must be granted.


—The ideal sought by all systems of proportional representation is this: Every vote cast at the polls for a candidate for membership in a law-making assembly should count in every vote taken in that assembly, whether or not the particular person voted for is elected. Seven methods of reform in representation have been suggested. These are, the proxy, the limited, the cumulative and the double vote, the free list or registered ballot, the Andræ (or Hare) system, and totality representation. The limited, the cumulative and the Andræ plans have been tried on a large scale.


—The proxy vote regards every vote for a legislative candidate as an informal power of attorney, and authorizes him, if elected, to cast as many votes as were cast for him. The fatal objection lies in the "if." If a candidate is not elected, his supporters have no representation whatever. This plan merely makes the power of local majorities greater than it is now, and so offers a standing reward for the fraudulent increase of such majorities.


—The limited vote applies only to elections in which three or more places are to be filled. Some English boroughs choose members of parliament in this fashion. Every elector can cast as many votes as there are vacancies, less one. If three men are to be chosen, he can vote for two; if four, three. But he can not give more than one vote to one man. This plan fails to give representation to any but a very large minority. Suppose 100,000 electors, and three places to be filled. A minority of 39,997 can elect no body, for the majority of 60,003 can cast 120,006 votes, which, divided among three candidates, will give each 40,002. In case of an accidental vacancy, under this system, a direct majority vote must decide the succession.


—The cumulative vote gives every elector as many votes as there are places to be filled and allows him to concentrate or scatter them as he will. This plan also recognizes the rights only of large minorities. It is not proportional. Let the number of vacancies equal x. A minority of less than (1/x+1) + 1 loses all representation. If there are 100,000 electors and three vacancies, 25,001 electors can secure one member (which is more than their share), but any less number must go unrepresented. Again, a very large minority does not get enough representation. If the 100,000 electors stand 50,001 to 49,999, the minority can get but one member, provided their opponents concentrate on two. The cumulative vote usually involves a great waste. In the first election of the London school board, conducted on this plan, the leading candidate received nearly 50,000 votes, while her colleagues were elected by from 8,000 to 13,000, and 50,000 were wholly lost. This system also makes no provision for the filling of accidental vacancies. Under the last constitution of the state of Illinois, adopted in 1870, the members of the lower house of the state legislature are chosen in this way, three from each district. This has been moderately successful. Parties have been better balanced in the legislature, and some better men have been sent there from the country districts. In the cities and towns, however, the most marked result has been to make king caucus more of a monarch than ever. A premium is put on the bargaining of party managers at the cost of party voters. By limiting the number of candidates on each side, the managers practically run a joint ticket, a proceeding which is rarely conducive to the public welfare. A similar provision in the same constitution in regard to the election of the officers of private corporations seems to have had no results.


—The double vote requires two elections. The first tests the relative strength of the several parties, and so determines how many representatives each shall have. If, with a legislature of 100 members, 600,000 persons vote the liberal ticket and 400,000 the conservative, the liberals become entitled to sixty members, and their opponents to forty. At a supplemental election each party selects its representatives. The double vote involves a waste of time, trouble and money, but it solves to a certain extent the problem of filling an accidental vacancy, and is worth more attention than other systems far better known.


—The free list or registered ballot scheme, provides that a certain number of citizens can make nominations by registering a list of names, the number of which shall not exceed the number of places to be filled. At the ensuing election only the lists thus registered can be voted for. The number of votes needed to elect is found by dividing the whole number of votes by the number of vacancies. If there are 100,000 voters, and ten places to be filled, the quota is 10,000. Suppose four tickets to be nominated, which receive respectively 35,000, 30,000, 27,400, and 7,600 votes. As the first has three times the quota, the first three men on it are elected. The second ticket also has three men elected, and the third two. The largest two remainders are those for the fourth and the third ticket, 7,600 and 7,400. The first nominee on the fourth ticket and the third on the third are therefore chosen. There is a waste of votes here. The men elected on the four tickets represent, respectively, 11,666, 10,000, 9,133 and 7,600 voters. Since 7,600 ballots suffice to elect a candidate, 24,000 ballots, nearly one-fourth of the whole, have been wasted. Again, in the event of the death, resignation or expulsion of a legislator, and a consequent special election, the free list can not be used.


—The Andræ system of proportional representation is commonly known in Great Britain and America as the Hare system, but Mr. Andræ introduced the practice of it in Denmark two years before Mr. Hare called attention to the theory of it in England. Under this system the quota of votes needed to elect a candidate is found as it is under the free list. Every elector puts as many names as he pleases on his ballot, numbered one, two, etc. As the ballots are taken from the box, each is credited to the name which is first upon it. If the electoral quota has already been cast for this first name, the ballot is credited to the second name upon it, and so on till all the full quotas have been ascertained. The largest fractions of quotas then elect, as under the free list system. This plan is somewhat complex, but not unduly so. It reduces the waste of votes almost to a minimum, except in the case of a special election or of an unusual number of candidates. The gravest objection to it is, that in transferring votes the real wishes of very many electors may be wholly ignored; chance may conquer choice. Suppose 100,00 electors, and two men to be elected; A is everybody's first choice; B and C each stand second on 50,000 papers. It makes a great difference which 50,000 ballots are counted for A. Chance or cunning, not choice, will elect B or C, as the case may be. Again, suppose B to stand second on 74,500 papers, and C on 25,500. If all the ballots counted for A have B as second choice, B's remaining 24,500 votes are eclipsed by C's 25,500, and C is elected, although B's real majority over him is 49,000.


—The formula of totality representation is this: after every general election of a law-making assembly, let the aggregate number of votes cast by each party be ascertained; divide this by the number of representatives elected by the party in question; the quotient will be the number of votes which each of those representatives is entitled to cast.


—Suppose that of 8,000,000 voters, who choose a congress of 300 members, 4,500,000 belong to one party and 3,500,000 to the other. It is quite possible that the congress thus chosen would stand 200 to 100. This estimate gives the majority less proportional weight than it has had in several congressional elections. While the parties in the nation were as nine to seven, they would be in the house as two to one. The legal majority in the latter would be 100; the equitable, 38. But apply the plan here proposed. Each of the majority has (4,500,000 200—) 22,500 votes; each of minority has (3,500,000 100—) 35,000 votes. The end sought is attained. The strength of each party in the house is a precise index to its strength in the nation. There is not an unrepresented man in the country.


—Under totality representation, an independent legislator would cast the number of votes he received. The ballots thrown for the man he defeated would be credited to that man's party. If an independent candidate were defeated his supporters' votes could be credited to other independents or go to swell the sum total of one of the two great parties. His constituents could express their wishes in this respect on their ballots.


—Fraud would be diminished by diminishing its usefulness. If we take our hypothetical figures of 8,000,000 voters, divided into two parties of 4,500,000 and 3,500,000, represented by 200 and 100 members of congress respectively, 50,000 fraudulent votes in favor of the majority would doubtless ensure the return of ten more members. The party in power would then have 210 to the opposition's 90. But under this system the administration would have a voting strength of 4,550,000 to its opponents' 3,500,000. In the first case, the fraud would increase the party majority by 20 per cent.; in the second case, by 5 per cent.


—At first the process of recording the votes of the legislature might be a trifle slow, but after two or three days' experience under the appointment which would follow each general election, a clerk could reckon the result of a doubtful vote about as quickly as if it were taken by yeas and nays.


—The totality representation system would make every vote cast at the polls at an election of a law-making assembly count in every vote taken in that assembly, whether or not the particular person voted for was elected. It would bring many habitual absentees to the polls, by giving every vote its proper weight, and would thus maintain a healthy public interest in politics. Its introduction would involve no sweeping changes, either in electoral districts or in modes of election. The ignorant citizen could vote as before, without being perplexed by new methods. All the necessary calculations would be made for him after the election. The system would stop "gerrymandering" by making it useless. A vote, whereever cast, would count. Finally, totality representation would allow an accidental vacancy to be filled at once, without depriving the minority in the particular district of its due representation. The new figures from this district would be substituted for the old ones in the aggregate vote of each party; each aggregate would be divided as before; and the quotient would be the number of votes which each representative of the party in question would be entitled to cast.


—See Memorandum on the History, working and Results of Cumulative voting, by Thomas Hare, 1857;The Election of Representatives, parliamentary and Municipalby Thomas Hare;Considerations on Representative Governement, by John Stuart Mill, 1862;On Representative Governement, by Simon Sterne, 1871;Essays and Lectures, Political and Social, by Henry Fawcett and Millicent G. Fawcett, 1872;Mac Millan's Magazine, November, 1872;Minority or Proportional Representation, by Salem Dutcher, 1872; Proportional Representation, by Charles R. Buckalew, 1872;The New Englande, July, 1874;The Science of Politicsby Sheldon Amos, 1883.


876 of 1105

Return to top