The above illustrations would, in themselves, be sufficient to kill any new system of representation. If the most ingenious critic could bring against proportional representation one-hundredth part of the case I have here brought against the existing plan of locality representation, Sir John Lubbock's society would not have a single adherent. As it is, gentlemen of every shade of opinion have not hesitated to join his society, because, as Mr. Mill has declared, the principle of proportional representation is a principle of fair play to all parties and opinions without distinction; it helps no one party or section to bear down others, but is for the benefit of whoever is in danger of being borne down. It is therefore a principle in which all parties may concur, if they prefer permanent justice to a temporary victory.'

But something more is wanting than the alleged perfection of a rival scheme if we are to attack with effect a vicious principle of Parliamentary election which is defended with vehemence by statesmen of the type of Mr. Bright, on the sole ground that it has been established for upwards of six hundred years. Proof must be forthcoming as to the actual existence of great and recognised defects in the working of the principle, and not only as to the possibility of their occurrence. There is no difficulty in obtaining the required proof, the only difficulty is in the making of selections from the evidence at our command. The statement of the United States senator," wherever you go in the United States you find gross misrepresentation of the people of the United States in that House which was peculiarly intended to represent them,' is true all the world over wherever the system of locality representation prevails. We have our home examples, the most striking of which are well known, i.e. the general election of 1868, when in Lancashire a minority of the electors secured two representatives for every one obtained by the majority,

and, more important still, the general election of 1874, which gave us a Parliament with a Conservative majority of 50, although the Liberals polled 1,400,000 against the Conservative total of 1,200,000. More striking examples, however, can be quoted from abroad.

The defects inherent in the principle of community as opposed to individual representation have been shown most conspicuously by recent events in Belgium. In that country the representatives are elected by a general ticket,' or scrutin de liste'-i.e. each voter votes for as many candidates as there are representatives to be elected, giving to each one vote.

It does not appear likely that there will be any serious attempt to

apply the principles of the general list system to the coming Redistribution Bill. There are, however, not a few politicians who agree with Mr. Leake's recent declaration when he said, 'Rather than break the political unity of Manchester into individual communities, I would accept in one block the whole allotment of members to Manchester, whatever the number which was allotted might be;' and who further believe that the difficulties of representative government will be most easily surmounted by the adoption of this system.

It will therefore not be out of place if I refer to the working of the system as illustrated by Belgium, in order to show how important it is that we, profiting by their experience, should avoid their mistakes.

The result of the Parliamentary elections which were held in June 1882 was to give

An indignant protest against a system capable of producing such a result was at once issued by 'L'Association Réformiste Belge'— a young society which had been set on foot the year before, with the object of substituting the principle of proportional for majority representation. The Liberals, who had profited by this unfair representation, met this protest issued by the advocates of true representation with the taunt that they were infected with the taint of Clericalism, and unworthy of the attention of all true Liberals.

The recent elections, however, which were held last June have caused the Liberal party to alter their attitude towards proportional representation, for if the principle of majority representation helped them in June 1882, it helped their opponents still more in June 1884. Their eyes, which were shut to the unfairness of the system so long as they reaped the advantage, have been opened now that they are in their turn the sufferers; for the result of the late Parliamentary election held last June has been to give

i.e. that party which polled within 3,000 of half the votes received only one-twenty-fifth part of the representation obtained by that party which polled less than 3,000 more than half the votes; or, in other words, every Conservative vote counted for as much as nineteen Liberal votes !

The following table shows the results of the elections of June 1884 :

Thus at Brussels the Conservatives won sixteen seats with 9,300 votes: the Liberals with 7,900 got none; at Antwerp the Conservatives with 6,800 votes won eight seats, and the Liberals with 5,400 got none; at Nivelles a majority of eighty-seven in a poll of 3,223 was sufficient to turn out four Liberals and put in four Conservatives; and in addition to Brussels, Antwerp, and Nivelles, where the Liberals were unable to obtain a single seat, at Louvain, Bruges, Ypres, Namur, Dinant, and Philippeville the Liberals were utterly demolished and absolutely blotted out from the representation.

A month after the election for the Chamber followed the election for the Senate, with an exactly opposite result, owing to the change of a few votes from one side to the other.

The results of the July election for the Senate are stated in the following table :

Now what do these two elections show? It must be remembered that they followed each other within the space of one month.

In June Brussels elected sixteen Conservatives and not a single Liberal for the Chamber, and in July Brussels elected eight Liberals and not a single Conservative for the Senate. Brussels has now sixteen Conservatives in the Chamber and eight Liberals in the Senate. The same constituency has thus instructed the deputies which it sends to the Senate to undo the work of its representatives which it sent only a month before to the Chamber. On the 10th of June the Conservatives have a majority of 1,387 and sweep the board. On the 15th of July the Liberals have a majority of 542 and in their turn sweep the board.

It will also be observed by a reference to the above table that the question which party should obtain the majority of the representation in the Senate rested with 275 Brussels voters. If at the July election 275 Brussels voters had voted for the Conservative list instead of the Liberal list, Brussels would have sent to the Senate eight Conservatives in place of eight Liberals, and the balance of parties in the Senate would have been completely reversed, 27,432 Conservatives carrying twenty seats, while the majority of 28,195 Liberals would have won only eleven seats: i.e. a minority of electors would have secured two-thirds of the representation.

67

It would be difficult to imagine any plan of election which could put greater power into the hands of a minority than the system which enables 275 Brussels electors--i.e. th of the Brussels voters, orth of the whole number of voters to decide whether the majority or minority shall rule. A still more flagrant and scandalous example of the utter absurdity of community representation as a means of providing a security for true representation and against the undue power of minorities, was afforded by the elections for the Chamber three years ago. Parties were then so evenly balanced in the Chamber that the question which party should be in office depended on the complexion of the contingent sent by the city of Ghent to the popular assembly. If Ghent sent eight Liberals, the country would enjoy the benefits of a Liberal Administration; if, on the other hand, it sent eight Conservatives, a Conservative Government would be installed in power. The Liberals of Ghent carried their list by a majority of forty, and an era of Liberal Administration was secured. But if twenty-one electors had changed sides, the Conservatives would have won the Ghent election, and the majority in the Chamber would have been Conservative instead of Liberal! Surely a system which is capable of producing such results as these cannot be defended by those who affect to wish for the true representation of opinion. The fact that the change of a few votes from one side to the other may entirely alter the whole result of the election, necessarily imports into election contests every artifice which

may win over from one side to the other a solitary vote. An election contest thus becomes a battle for life and death. Corruption, intimidation, unscrupulous flattery, reckless promises, are the necessary features of a contest in which the victors gain everything and the losers are frappés de mort politique.'

Let those who seem to think that because we are obliged to legislate by a majority it is also necessary (for some strange unexplained reason) that the majority should alone be represented, and who seem to see in local majority representation a security against the undue power of minorities -let them recall the Ghent example and remember how, under a system of majority representation, twenty electors in Ghent, and these probably the most corrupt of the whole population, had the power of deciding which party should be in office.

I have referred at length to these examples because the system of election in Belgium is exactly similar to that which some of our politicians wish to establish here. Reference to the table of elections for June will show that the constituencies resemble in their variety the constituencies that will be established in England, if those politicians who are in favour of community representation and against the subdivision of large towns into wards succeed in forcing the adoption of their views upon Parliament: e.g. Ostende, Arlon, Marche, Neufchâteau, and Virton return one member, Philippeville and Dinant return two members, Bruges and Ypres three, Nivelles and Namur four, Louvain five, Antwerp eight, and Brussels sixteen, the very same sort of arrangement as would obtain in England if redistribution is to be on the basis of population, and the towns are not to be broken up into wards.

The obvious abuses of the system existing in Belgium led three years ago to the formation of L'Association Réformiste Belge, a society which aims at substituting the principle of proportional representation in the place of the existing principle of majority representation. M. Albert Nyssens, the secretary of the association, at their annual meeting last year said:

If two years ago the question had been asked, What is your opinion as to proportional representation? ninety out of a hundred would have answered that they had no opinion at all. It had never occurred to the generality of electors to doubt that if at an election 4,000 votes got eight representatives and 3,999 got none, that everything was not for the best in the best of all possible electoral systems.

In January 1882 they started a fortnightly paper called La Représentation Proportionnelle. They now issue a most admirable monthly pamphlet under the same title, which is well worth the study of all those who are interested in the question, How can we obtain true representation? A large number of the most eminent men in Belgium, representing the most different opinions, and coming from all parts of the country, have joined the association. A great part of the press supports it, and there is every reason to believe that before long the