their influence has come about as a result of the social arrangements which the past has evolved. Another class, equally large, are dissatisfied with their lot. They see inconsistency or injustice in social arrangements. They desire change, reform. Society makes progress through the ebb and flow of these two fundamental groups.

But this natural grouping of free individuals is far different from the iron-bound classification imposed by the modern highly-developed party machine. Individuals, if left to themselves, will be continually forming new groupings as new political questions arise. But the machine having control of nominations, as has already been shown, maintains itself in power against the natural regroupings of the voters. Freedom for the machine, then, means, first, power on the part of the voters to control the nominations of their party; and, second, power to defeat obnoxious candidates of their own party without endangering the success of the party. Both of these advantages are provided for in the proposed bill-first, by the provision for cumulation of votes, and, second, by the provisions for independent tickets.

In the first place, let us suppose the nominations have all been made and voters have come to the polls ready to cast their ballots. Suppose there are fifteen candidates to be elected, and that tickets have been nominated as in the sample ballot given on page 161. It is known beforehand very nearly how many candidates each party will elect. The Democrats, for example, are certain to elect not less than six, and perhaps not more than eight. They have accordingly nominated ten candidates. The provision for cumulation now enables any voter to cast his entire fifteen votes for the party of his choice, but at the same time to select among the ten candidates that one who best represents him and to "plump" his fifteen votes for him. The party convention is powerless in the matter. It is compelled to put up nine or ten candidates instead of one. The voter may be so disgusted with some of the nominations that he would willingly stay at home or vote the opposing ticket if there were no further choice for him. This is what he is forced to do in the single-membered district. But with ten candidates there will likely be one or two at least with whom he is satisfied. For them he can give his entire strength. And if there are a sufficient number of voters like-minded with himself, they can elect those candidates who are least subjected to the machine rule.

I believe that herein will be found a very decided advantage over the Swiss method of proportional election. The Swiss ticket is prepared in exactly the same way as the one here proposed, and voting is conducted in exactly the same way, with the exception of this one feature of cumulation. The Swiss voter is not allowed to cumulate. If he votes a party ticket, it must be a straight ticket. He can give but one vote to each candidate. He is therefore constrained by the party machine, not as much as under the district system, but more than seems necessary.

According to the true theory of proportional representation one representative should be elected by every group of voters whose number is equal to the quotient obtained by dividing the total number of votes by the number of representatives to be elected. If 200 voters elect ten representatives, any group of twenty voters should be able to elect one, i.., each elector should have zoo part of the total influence. But the Swiss method by giving electors as many votes as there are representatives, reduce the influence of the elector in his vote for any individual candidate in exactly the inverse ratio. That is to say if 200 electors have ten votes each the total votes will be 2,000 and it will require 200 instead of 20 to elect. If the elector can give only one of his ten votes to any one candidate he has 2000 part of the total ifluence, i.e., only the influence he rightly should have.

The simplest plan of cumulative voting is now employed in England in the election of members of the school boards. It is also in operation in Illinois in the election of members of the Lower House of the State Legislature. There have been more or less valid objections raised to those schemes for cumulative voting which have been heretofore proposed, namely (1) that they lead to a loss of voting power, (2) that they increase the power of political machines, and (3) that they give too much power to the minority. The present plan differs from all others in such a way that I believe these objections will not hold.

Sir John Lubbock presents conclusively the first objection to this system in the example he gives of the Marylebone election for members of the school board. (November, 1870.)*

There were seven members to elect and the votes were as follows:—

"It will be seen that Miss Garrett received no less than 47,858 votes, while under the circumstances 8,000 would have elected her. Nearly 40,000 votes out of the 48,000 were therefore wasted, and it is obvious that if Miss Garrett's supporters hal known their strength, they would have desired to vote so as to secure the return of other candidates sharing their opinions. The Marylebone election was certainly an extreme case, but there have been many others in which the same phenomenon has been repeated."

Out of this fact grows the second objection. If there is danger of wasting votes the only way in which votes can be properly economised is for the voters to know their strength before election. This they can do only through the party organisation. The machine therefore will designate the candidates to be voted for and those of the party voters who do not follow the instructions of the machine will throw their votes away.

These objections seem valid against the simple plan of cumulation. But it will be readily seen that the bill proposed in this essay does away with these objections by reason of its double provision for ticket votes and candidate votes. The elector cannot possibly throw his votes away, because if too many votes have been cumulated on one candidate, as in the case of Miss Garrett in the Marylebone election, the surplus votes as well as the necessary votes for the candidate go to swell the aggregate of votes for the ticket, and thus help to elect other candidates on the same ticket. In this way, too, as is already shown, the power of the machine is greatly minimised instead of increased.

But suppose the elector finds that his party ticket has been put forward wholly in the interests of the machine, that there is no candidate offered to him whom he deems worthy of election. Under the district system, as has been shown, he must vote for the party nominee or else stay at home or cast his vote in such a way as to benefit the opposite party. Proportional representation enables him to "bolt" the party ticket and not bolt the party.

Of course, in order to do this, he must join with other protestants against the machine and nominate a new candidate, and a new ticket. The bill provides readily for a movement of this kind by a petition signed by fifty names for every candidate to be elected from the given State. This would be about fifty signatures for every 30,000 voters in the State. In the case of the simple ballot on page 161 the petition would require 750 signatures.

The bill now provides that all tickets shall be printed on a single ballot. This is the well recognised principle of the official ballot, introduced into this country from Australia, and now adopted by States representing of the population of the country. The bill provides that this ballot shall be extended into all States in so far as congressional elections are concerned. Indeed, the highest success of proportional representation could not be obtained were it not for the admirable features of the Australian ballot. It gives the voter complete independence of choice, enables him to scatter his votes among different

tickets, and to easily pick out from a large number of candidates an individual for whom he may wish to cumulate. The system could readily be applied to the voting machine which has just been introduced into elections in New York State and which possesses many features superior even to those of the Australian ballot.

The independent voter has therefore before him the ticket nominated by his party organization and another ticket of the same party nominated by petitioners outside the machine. Suppose he votes for the independent ticket. He does not endanger the success of his party ticket as a whole, but only of one candidate out of the entire ticket. And he is not running the risk of throwing his vote away, providing only a single quota of the voters of the State cast their ballot for the independent. Under the district system his vote would be wasted unless the voters for the independent were a mojority or a plurality of all the voters in the district

Returning to the sample ballot on page 161, the Republicans have nominated nine candidates, expecting to elect six or seven. The Independent Republicans have nominated two, expecting to elect one. From the final result it will be seen that the Republicans could under no circumstances have elected more than six candidates, because the Democrats have elected seven and the People's two, out of the fifteen. But the straight Republicans and the Independent Republicans together have elected the six can didates to which the Republicans were entitled. The Independents have in no way endangered the success of their party, but they have succeeded in putting in a Republican who, standing for the party, nevertheless opposel the machine. The party as a whole gains its share of representatives, but the independent wing of the party has also secured its just share. Had the election been conducted under the district system, the Independent Republicans in order to be elected would have required a majority or plurality of the votes in his district. And in the effort to reach this result, the Democrats would probably have elected their candidate. But in proportional representation the independent is elected if he polls an average of one-fifteenth of the votes in all the districts, and the Democrats do not gain an iota through the disaffection of the Republicans. The present system, in other words, pens up a minority of independent voters in narrow districts under the whip of their party machine, but proportional representation tears down the fences and enables them to combine throughout the State, without risking the success of the opposing party's machine.

That this feature of proportional representation strikes at the ralical evil of present politics, is evidenced by the current literature on all hands The evil of absenteeism on the part of the intelligent and business classes is everywhere alarming. It can not be too loudly deplored. In an article favoring compulsory voting Mr. F. W. Holls says: * "Toe extent to which this duty is shirked is easily ascertained, at least approximately. The interest which centers in a presidential contest is generally sufficient to bring out the fullest vote obtainable without compulsion, and a comparison of the total vote in a presidential year with that in an "off" year, shows almost the entire number of shirks. In the state of New York 300,375 persons who voted in 1880 remained away from the polls in 1889, and 286,278 did so in 1890 In the last mayoralty election in New York city over 35,000 men who had even registered abstained from voting, with the result that the city was once more turned over to an organised gang of plunderers. A more deliberate and extensive betrayal of trust would be difficult to find. In Massachusetts the total vote of 328,588 in 1888 fell to 260,798 in 18.00, a difference of 67,790. In Chicago the figures are even more startling. In the spring election of 1887 less than 72,000 votes out of a possible 138,000 were cast-66,000 citizens failing in their duty-while in June o f the same year, at the judiciary election for the choice of judges for a city of almost a million of souls, the total vote was 44,074, less than one-third of the number of qualifie voters."

Prof. A. B. Hart + asserts in a careful review of election statistics that the voting population is one-fourth of the total population, and that in presidential elections five sixths of this voting strength is cast. In New York, in 1880, the vote was 1,104,605, ̄ being 23 per cent. of the total population and 95 per cent. of the voters, but in 1891 it was only 259,425. In New York city, in 1888, the vote was eighteen per cent. of the population; in 1890 it was 11.6 per cent. and in 1891 it was 13.2 per cent.

* Annals of Amer. Acad. Pol. and Soc. Sci., April, 1891.

+ Political Science Quarterly, vol. vii., page 307.

The New York Nation of April 13, 1893, says, pertinently

"The Government of Russia has been described as 'despotism tempered by assassination.' In like manner the government of large cities in America may be termed 'bummer government tempered by uprisings.' Nevertheless, we believe that both in Chicago and New York the better element is really' in a majority, and could, if it chose, retain the government of the municipality permanently in its hands. This is certainly true of New York, for the bummer element here has never yet polled a majority of the registered vote. Take, by way of illustration, the important election of 1888, at which Tammany got possession of the city. Tammany polled in that year 114,000 out of a total registered vote of 286,000. In 1890 it polled 116,000, or only 2,000 more, which may be called the natural increase of the Bummer Element. We think it is quite fair to set down as Better Element all voters of every description who do not vote the Tammany ticket. This Better Element, then, in 1888, registered 172,000 votes; in 1890, after two years' experience of Bummer rule, 129,000 votes. If the full registered vote in either of these two years had been cast against Tammany, Tammany would have been defeated and the Better Element would now be in possession of the city. But, in 1890, after a full trial of the kind of government the new Bummer régime was prepared to furnish, 30,000 of the Better Element stayed away from the polls and allowed Tammany to retain the city. Why did they stay away? Any one who could answer this question would explain the failure of popular government in American cities.

Now, if it were possible to go around among these 30,000 and ask them severally why, having registered, they failed to vote, we should in all probability get a perfectly intelligent answer from nine out of every ten of them. Not one would say that he did not vote because he was not "organized;" that if anybody wanted him to vote, he must "organize" him. Nor probably would any of them say that they preferred Bummer Government to Citizens' Government. Some would have said, doubtless, that they did not think the election of Scott would be enough of an improvement on Grant to make it worth their while to go to the polls to bring it about; others, that Scott was "Grace's man," and they hated Grace; others, that they would never, under any circumstances, vote for a Democrat; others, that they hated Mugwumps, and that Scott was a Mugwump invention; others, that Scott's nomination was a contrivance for breaking up the Republican party in this city; others, that they wanted to keep Tammany in power as an example of Democratic rule."

But appeals to the voters to go to the polls are of little consequence. What can they accomplish by going? Says a recent writer:* "It is only the fear of wasting their votes on good men who have no chance of winning, which deters the people from voting against bad candidates who are forced upon them by the regular machine. Neither would compulsory voting meet the real difficulty. To quote the words of Prof. Hart, in the article already referred to, "To compel men to vote against their will is to tighten the control of party managers. Honest voters are indifferent or refuse to vote because they feel their impotence to affect their own party management. Yet they support their party management because experience shows that the men who fight it must make great exertions and sacrifices or be set out of politics; and further because permanent political results can be brought about only through strong and persistiug parties. Compulsory voting supplies no new motives, and would not alter those political habits of the American people which are the real evil. Compulsory voting cannot create interest in local affairs, or break up the practice of adhesion to unfit leaders."

Compulsory voting would not stimulate independence. It is but natural that a well known machine Governor of New York should have recommended it to the legislature in one of his annual messages. Professor Giddings has shown that the fringe of mobile voters who change from one party to another, is seldom more than 5% of the maximum total vote in a presidential year, and "the number of voters liable to be decisively influenced by mere opinion, apart from personal, class or sectional interests, is not more than 2 or 3% of the whole." Compulsory voting might possibly change this proportion slightly, but there is little reason to hope for the good results which its advocates claim

for it.

* Charles Richardson in American Academy of Political and Social Science, March, 1892, page 86 + Political Science Quarterly, Vol. VII., page 116, 124.

That proportional representation would bring out a full vote at every election is not to be asserted. There is no experience from which to draw conclusions. The system has been in practice in Switzerland for only one year. But it certainly removes one of the most potent reasons for staying at home. In this respect the election of representatives offers advantages over the election of executives, judges, and other single officers, where, of course, there is no question of proportion. These advantages consist, as already shown, in the fact that there are several candidates to be elected on one ballot, and a small number of voters does not hold the balance of power. If the party vote is reduced even so much as a full quota,―i. e. 15 to 20%-the party loses only one candidate out of a possible five, ten, or fifteen, and this one candidate is not gained by the opposing party but by the independents of their own party. A similar defection in the district system would usually throw the majority to the other party, since, as is well known, the two great parties are closely divided in the majority of districts. A change of 14% in twenty-one districts is estimated by Mr. Barry as sufficient to give the control of the Massachusetts senate to the opposite party.

As a result of the freedom which would be given by proportional representation to the rank and file of the party voters, the machine would usually be compelled to consult their wishes in nominating its candidates. At present it knows that, whatever their threats, they will not bolt and thus elect the common enemy. But if they can bolt without committing such direful treason, they will be inclined to do so, and the machine must listen. Candidates will be of larger caliber and better reputation and more in harmony with the opinions of the voters.

Numerous writers have called attention to the importance of the primaries, and have emphasised the duty of citizens to attend their party primaries and conventions. In the view here presented of the probable workings of proportional representation, the primary loses much of its significance. A party nominates candidates to win. Though voters do not attend the caucus, yet the machine has the probable action of these voters in mind when it nominates its candidates. The significance of the primary to-day is due to the impossibility of acting independently of machine dictation.

Besides, politics is a business. It requires time and strength. The politician does the very least part of his work in the primary. The real work is done beforeband. America has no leisure class who can afford to give themselves to this work. They must leave it to the professionals. The latter are tacit attorneys. They sound their clients, learn their wishes, and act accordingly. If proportional representation should bring forward an abler and purer class of politicians, more in harmony with the best wishes of the people, the latter could leave the primaries to them.

At the same time, primaries and conventions are the sources of power. They must be recognized as such. They must be brought under legal regulation. Perhaps the most serious evil of primaries and conventions and the one which gives the machine entire control is the practice of exclusive majority rule. It is the almost universal rule to elect committees and delegates by a majority vote on the principle of the general ticket, or else to authorize the chairman to appoint them. The true purpose of a primary, as representing all sections of a party is thereby defeated. It is proper that the majority should elect the chairman or nominate single candidates for offices, but why should the majority be alone represented on committees and delegations? Plainly here is the need for a further application of the proportional rule. A plan is required which will be simple and quickly wor ed. Such a plan has been proposed by Dr. L. B Tuckerman, and is described in the Review of Reviews for November, 1891, as follows:

The Tuckerman plan provides for weighing the choices of each elector. If there are

five offices to be filled the elector writes on his ballot the names of five candidates in the order of his preference. Then the tellers, in counting the ballots, allot to each name on the ballot a weight of choice corresponding to the position held by that name on the ballot. Thus if the candidates A, B, C, D, E, are written on a single ballot in the order given, candidate A will have five units credited to him, candidate B will have four units, C three units, D two units, and E one unit. After all the ballots are counted the units opposite the names of the candidates are added up, and the five having the highest number of units are declared elected. Thus only one ballot is required to elect the five officers. Continuing the example given, suppose the candidates A, B, C, D, E, are voted for in the der named by each of the fifty-five delegates. The weight of choice would be as follows: