Every one knows the trouble that follows. Mr. Sanity is "going to split the party vote." The hesitating voter is told, with considerable truth, that a vote given for Mr. Sanity is a vote given for Wurstberg. At any price the constituency does not want Wurstberg. So at the eleventh hour Mr. Sanity is induced to withdraw, and Mr. Goldbug goes into Parliament to misrepresent this constituency. And so with most constituencies, and the result is a legislative body consisting largely of men of unknown character and obscure aims, whose only credential is the wearing of a party label. They come into parliament not to forward the great interests they ostensibly support, but with an eye to the railway jobbery, corporation business, concessions and financial operations that necessarily go on in and about the national legislature. That in its simplest form is the dilemma of democracy. The problem that has confronted modern democracy since its beginning has not really been the representation of organized minorities—they are very well able to look after themselves—but the protection of the unorganized mass of busily occupied, fairly intelligent men from the tricks of the specialists who work the party machines. We know Mr. Sanity, we want Mr. Sanity, but we are too busy to watch the incessant intrigues to oust him in favour of the obscurely influential people, politically docile, who are favoured by the organization. We want an organizer-proof method of voting. It is in answer to this demand, as the outcome of a most careful examination of the ways in which voting may be protected from the exploitation of those who work elections, that the method of Proportional Representation with a single transferable vote has been evolved. It is organizer-proof. It defies the caucus. If you do not like Mr. Goldbug you can put up and vote for Mr. Sanity, giving Mr. Goldbug your second choice, in the most perfect confidence that in any case your vote cannot help to return Mr. Wurstberg.
With Proportional Representation with a single transferable vote (this specification is necessary, because there are also the inferior imitations of various election-riggers figuring as proportional representation), it is impossible to prevent the effective candidature of independent men of repute beside the official candidates.
The method of voting under the Proportional Representation system has been ignorantly represented as complex. It is really almost ideally simple. You mark the list of candidates with numbers in the order of your preference. For example, you believe A to be absolutely the best man for parliament; you mark him 1. But B you think is the next best man; you mark him 2. That means that if A gets an enormous amount of support, ever so many more votes than he requires for his return, your vote will not be wasted. Only so much of your vote as is needed will go to A; the rest will go to B. Or, on the other hand, if A has so little support that his chances are hopeless, you will not have thrown your vote away upon him; it will go to B. Similarly you may indicate a third, a fourth, and a fifth choice; if you like you may mark every name on your paper with a number to indicate the order of your preferences. And that is all the voter has to do. The reckoning and counting of the votes presents not the slightest difficulty to any one used to the business of computation. Silly and dishonest men, appealing to still sillier audiences, have got themselves and their audiences into humorous muddles over this business, but the principles are perfectly plain and simple. Let me state them here; they can be fully and exactly stated, with various ornaments, comments, arguments, sarcastic remarks, and digressions, in seventy lines of this type.
It will be evident that, in any election under this system, any one who has got a certain proportion of No. 1 votes will be elected. If, for instance, five people have to be elected and 20,000 voters vote, then any one who has got 4001 first votes or more must be elected. 4001 votes is in that case enough to elect a candidate. This sufficient number of votes is called the quota, and any one who has more than that number of votes has obviously got more votes than is needful for election. So, to begin with, the voting papers are classified according to their first votes, and any candidates who have got more than a quota of first votes are forthwith declared elected. But most of these elected men would under the old system waste votes because they would have too many; for manifestly a candidate who gets more than the quota of votes needs only a fraction of each of these votes to return him. If, for instance, he gets double the quota he needs only half each vote. He takes that fraction, therefore, under this new and better system, and the rest of each vote is entered on to No. 2 upon that voting paper. And so on. Now this is an extremely easy job for an accountant or skilled computer, and it is quite easily checked by any other accountant and skilled computer. A reader with a bad arithmetical education, ignorant of the very existence of such a thing as a slide rule, knowing nothing of account keeping, who thinks of himself working out the resultant fractions with a stumpy pencil on a bit of greasy paper in a bad light, may easily think of this transfer of fractions as a dangerous and terrifying process. It is, for a properly trained man, the easiest, exactest job conceivable. The Cash Register people will invent machines to do it for you while you wait. What happens, then, is that every candidate with more than a quota, beginning with the top candidate, sheds a traction of each vote he has received, down the list, and the next one sheds his surplus fraction in the same way, and so on until candidates lower in the list, who are at first below the quota, fill up to it. When all the surplus votes of the candidates at the head of the list have been disposed of, then the hopeless candidates at the bottom of the list are dealt with. The second votes on their voting papers are treated as whole votes and distributed up the list, and so on. It will be plain to the quick-minded that, towards the end, there will be a certain chasing about of little fractions of votes, and a slight modification of the quota due to voting papers having no second or third preferences marked upon them, a chasing about that it will be difficult for an untrained intelligence to follow. But untrained intelligences are not required to follow it. For the skilled computer these things offer no difficulty at all. And they are not difficulties of principle but of manipulation. One might as well refuse to travel in a taxicab until the driver had explained the magneto as refuse to accept the principle of Proportional Representation by the single transferable vote until one had remedied all the deficiencies of one's arithmetical education. The fundamental principle of the thing, that a candidate who gets more votes than he wants is made to hand on a fraction of each vote to the voter's second choice, and that a candidate whose chances are hopeless is made to hand on the whole vote to the voter's second choice, so that practically only a small number of votes are ineffective, is within the compass of the mind of a boy of ten.