Choice Voting America?

To home page.

American Declaration of Independence

Links to sections:

(1) Turn-out and primaries.

Low turn-out from least choice.

To elect means to choose-out. Only one candidate can be prefered with an x-vote or spot vote. All those candidates not prefered might as well be just one candidate, compared to the one candidate, who has more votes than any other single candidate. And this is what tends to happen. Hence, the two-party system, which is sustained by the spot vote's single preference, either for party candidate A or candidate B.

There are four logical possibilities of choice between two candidates: both are equally prefered; both are equally disliked; A is prefered to B; B is prefered to A.

An election can only take place when someone is actually elected or chosen-out, and that only covers the two latter possibilities, that is half the logical possibilities. Logically, one would only expect half the electorate to vote, because the other half have no preference between only two candidates.
Indeed, in 1980 US presidential elections, 50% of the electors didnt vote. If this is apathy, it is a logical apathy. US non-voters are doing what should be logically expected of them, given their restriction to a minimum choice. But in so many other American elections, turn-out drops alarmingly.

Prof. Douglas Amy, in Real Choices, New Voices cites studies that some 62% 'no show' voters have least education and income, and, therefore, least cause to be content.

American reformers paraphrase Bill Clinton's election-winning advice, 'It's the economy, stupid,' to: It's the electoral system, stupid.
His appointee, Prof. Lani Guinier had started a debate on the subject, when that appointment was withdrawn. She remarked:

our level of participation is an embarrassment. Some may say that reflects contentment with the status quo. I think it represents...rational behavior by voters who realize their votes don't count.

Some encouragement for this view may be gleaned from the French 1981 presidential elections. Several party candidates had quite respectable levels of support but no real hope of winning. Basically, there were four serious contenders from four reasonably well-matched main parties, in the first round of France's Second Ballot system. The two more centre candidates, from the Right and the Left, went thru to the second round.

In the Second Ballot, the turn-out increased, from 80%, in the first round, to 86%. This latter figure is almost exactly the turn-out to be expected from adding up the logical possibilities of choice for four main candidates in France, instead of two main candidates, in the USA.
For two candidates, there are four logical possibilities of choice. For four candidates, there are sixteen possibilities.

( The number of logical possibilities of choice for candidates is obtained by the binomial theorem. It is two to the power of the number of candidates. In the above examples, two to the power of two equals four possibilities of choice between two candidates; and two to the power of four equals sixteen possibilities of choice between four candidates. )

Of the sixteen possibilities, two of them will be non-elective: all four candidates being equally prefered is one possibility, and all four candidates being equally disliked is another possibility. So, one would expect two-sixteenths or one-eighth of the electorate, who have no preference between the four candidates, not to vote. One eighth is twelve and a half per cent, leaving eighty-seven and a half per cent, as the number of voters. This is a good approximation to the 86% French turn-out quoted above.

The French Second Ballot seems to work, even tho it only effects a ranking of two out of four choices of candidates. In the first round, the single preference usually goes between one of two rightist or one of two leftist candidates. In the second ballot of one's vote, a single preference is enough to decide between the one remaining left or right candidate.

One could decide the election, in one round, with a so-called Supplementary Vote, giving one's second ballot choice, in a combined ballot. This is the same as having a first choice and a second choice. For example, on the French Left, one's first choice might be a communist candidate and one's second choice the socialist. If one's communist choice lost against the socialist, in the first round of counting, then one's second choice for the socialist would help the socialist against the winning right wing candidate, in the second round.

Still, a mere double preference vote of first and second choice allows no information about one's preference between the other main two candidates, to say nothing of minor party candidates. This could encourage a rigid Left-Right divide.
It used to be joked that the French voted either for Marx or Jesus.

North American electoral reformers, including Canadians, dont seem to favor the Second Ballot, as such, for single seat elections such as of mayor or president. Rather, they speak of a more thoro version, offering more than two orders of preference. Voters are given a preference vote, to rank their order of choice for candidates.

The candidates with least first preferences are gradually eliminated till some candidate wins by achieving an over-all majority, or over half the votes. There is no need for the voters to go to the polls twice. Hence, the Americans call this 'instant run-off voting' ( or simply, run-off voting ). In other countries, it has been known as 'the alternative vote'.
( I discussed the short-comings of AV in my review of the Kerley report. )


Short history of solving basic problems in preference voting.

To top.

The modern history of voting method perhaps begins with the French philosophs of the Enlightenment. Condorcet pointed out an apparent paradox in preference voting. Normally the winner of a preference vote of ranked choices, first, second, third etc, is decided by eliminating the candidate with the least first preferences. His vote is then re-distributed according to his voters' second preferences. These re-distributed votes may take some other candidate over the winning line of an over-all majority.

But the Condorcet paradox shows that even the candidate with the least first preferences may be elected if the second preference of most of the voters. This is if no candidate has an over-all majority of first preferences and if all the candidates in turn are paired against each other, with the help of re-distributed votes from the other candidates.

The Chevalier de Borda answered the Condorcet paradox by saying it did not take into account the relative importance of order of choice. So, he proposed that 1st, 2nd, 3rd etc choices should be weighted in the count. If there were four candidates, first preferences should count four times as much as fourth preferences, second preferences would count three times as much, third preferences twice as much.

This and other systems of weighting the preference vote, Ive discussed ( in detail in part one of my Scientific method of elections and on the page of simple examples of STV. )

But it is of interest that Laplace endorsed it with one of his typically involved proofs. Laplace often didnt bother to give proofs, saying a thing was too obvious. His American translator said, whenever he saw this remark, he knew he had a hard night's work ahead of him, demonstrating 'the obvious'. Laplace is rated among history's half-dozen greatest mathematicians.
( J F S Ross discussed Borda's method in Elections and Electors. ) We now know, or should do, that weighting the preference vote is, indeed, the essential means to resolving the Condorcet paradox.

Nevertheless, Borda's method left two unresolved problems, whose solution would be provided by the definitive means of weighting the preference vote in the count. Whatever mathematical series one uses to weight preferences in order of importance, it is an essentially arbitrary business. The example given above is Borda's original suggestion of the arithmetic series. Ross said this gave too much weight to lesser preferences, especially if there were many candidates. He suggested the geometric series to offset this. Others have suggested the harmonic series as a compromise. It is still guesswork

Whichever weighting series one uses, later weightings count against earlier ones, so that the more preferences one expresses the more one is voting against ones first preference.

It wasnt till the middle of the nineteenth century that the way was made open to remedy these faults, tho it was not realised for another quarter of a century. Carl Andrae and Thomas Hare independently proposed that there should be a proportional count of preference voting. Representatives were equitably elected on each winning a required proportion of the votes in a multi-member constituency.

Some candidates would get more first preferences than they needed to achieve an elective quota of votes. The surplus votes would not be wasted but transfered to those voters' second preferences. The question was: whose second preferences, amongst all the voters for the already elected candidate, would make up the transferable surplus vote?

The simplest answer was to make the surplus vote a random sample. Give the ballot box a good shake, so its contents are properly mixed or at random, like a prize draw. Then draw out a representative sample of votes, to the number of surplus votes to be transfered.

There is a more exact way, discovered by an Australian called J B Gregory. Gregory's method came to be called 'the Senatorial rules', after its use in Commonwealth senates. And it gives definitive answer to the out-standing problems of Borda's method. ( See above references. )

The point is, the weighting of second or lesser preferences is no longer arbitrary but in proportion to the size of the transferable surplus. Also, later preferences do not count against earlier preferences. The first preference has already been elected before the second preference comes into play to help elect another representative to a multi-member constituency.

This 'single transferable vote' ( STV ) of surpluses usually left some seats needing to be filled by redistributing the votes of candidates with least first preferences. And this was a residual anomaly in the traditional hand-counted version of the system.
But computers make possible a systemmatic flow chart of counting that further reduces the minimal possibilities for well informed voters to work the system with a contrived ordering of their preferences.

Cambridge, Massachusetts, local government recently automated their count after some six decades use of STV or 'choice voting', as it's called in America. There, the term 'preference voting' is another synonym for STV. But this system also uses a proportional count of preference voting. Bearing that in mind, I shall keep to the terms, choice voting or STV.


Presidential and congressional primaries by choice voting.

To top.

For the reasons outlined ( or refered to ) above, the Second Ballot or successive ballots, which can be rendered compactly by preference voting, gives way, by itself, to paradox and anomaly. But the proportional counting of preference votes, with the transferable voting system, resolved the main logical problems.

This system, however, involves multi-member constituencies. So, it has been claimed that STV or choice voting doesnt apply to single vacancies, like the presidency. But this involves some misleading assumptions.
Of course, if there are only two candidates chasing one seat, then there is no question of votes being transfered from a third candidate. The only possible preference is a first preference ( whether marked on the ballot paper by a cross, a punch hole or the number one ). Likewise, the only possible majority is one majority over the remaining minority. ( e.g. 51 voters out of 100.

It has not been appreciated that two, three, or more, majorities are possible over a remaining minority ( e.g. two majorities of 51 votes each, out of 150 voters; three majorities each out of 200 voters ). This involves the so-called Droop quota, used to count STV.

As soon as there is more than a single majority to count, the voter needs more than a single preference. Second and third preferences help decide which candidates attract enough votes to make up second and third majorities in two or three member constituencies.
One, two, three,.. preferences elect one, two, three,..majorities. This is the logical correspondence of the vote to the count, which must consistently sum up degrees of individual choice to degrees of community choice.
The transferable vote translates the preference votes' measure of greatness of individual choice into a measure of greatness of community choice among candidates.

Carl Andrae and Thomas Hare understood, that preference voting and proportional counting imply each other, when they independently invented this generalised electoral system. But most 'modern democracies' illogically impose electoral laws that allow preference voting only in single member constituencies ( the Second Ballot or Alternative Vote ) or only proportional counting ( party lists or additional member systems ).

A spot vote in a single member constituency is only the most limited choice that the transferable vote boils down to, for the final round of a contest for a single seat. STV is not a different system, it is a generalised system of 'choice voting'.
When someone says STV or choice voting doesnt apply to single-seat contests, all they are really saying is that only a final round of choice is allowed the public.

The American presidential election also confuses the issue. If the presidential primaries were truly comprehensive elections, then the final vote would truly be a show-down election between the two final presidential candidates, and no other candidates would be permitted to run. Instead of that, the third candidate is decried as a 'spoiler' of the popular choice of the finalists. And he is punished financially with huge campaign expenses and a lost deposit.
This is unjust, as the real blame rests with the democratic inadequacies of the primaries system.

G K Chesterton said that if a thing is worth doing at all, it is worth doing badly. Tho, he later was disparaging of his own 'mal mot'. Things usually have to be done badly before they can be done at all -- like the one megawatt of fusion power produced for the first time. But science, as such, tries to improve. Whereas, politics tends to be fundamentalist.

The problem with American primaries is that the voting system is too primitive for them to work properly. In the democratic spirit, some elections between candidates seeking to be their party's official nomination, were thrown open to non-party members. But rival party members would vote for their opponents' worst candidates. This was a version of Gresham's law, that bad money drives out good: bad candidates drive out good.

You could say that sums up the whole party system of career obedience to out-moded dogmas dictated from the top. Certainly this would be true of more rigidly partisan orthodoxies than in the United States. It would apply to British politics, for instance, even before the introduction of party list systems, whereby candidates are appointed to parliaments by the party bosses who draw up the lists -- under the false pretence this proportional partisanship is 'proportional representation'.

PR was intended to promote democracy, which has been surreptitiously undermined by this debased form of it, using party lists. ( That is what Prof. Hermens meant by PR as 'the Trojan horse of democracy'. But Hermens threw out the baby with the bath water by culpably not distinguishing the original PR of choice voting, which greatly improved democratic method. )

Likewise, the irony of American primaries is that they threatened to undermine democracy, in the attempt to enhance it. However, the solution to both problems of democratic degradation is the same. Choice voting gives more equitable representation, than first past the post, without denying individual representation. Indeed, it gives much greater freedom of choice and, in so doing, effectively solves the primaries problem.

The primaries problem is whether primaries should be open or closed. Open primaries allow everyone to choose who will be a party's official candidate. But this choice is typically abused by partisan opponents. So, parties have wished to conduct closed primaries to all but their members. This, in turn, may be challenged as less democratic.

The dilemma has caused endless wrangling over the constitutional rights and wrongs of the matter. But it is actually a logical problem in electoral method, whose solution ( the short history explains above ) led to the transferable voting method. For instance, the Oxford professor of Constitutional Law, Vernon Bogdanor has explained why STV could produce a more democratic and efficient system of American primaries.

With STV or choice voting, everyone can vote, and the candidates most prefered are elected in a congressional multi-member constituency. Say, there are five seats, then the two most prefered out of five Republicans and the two most prefered out of five Democrats might be elected.
Another one-sixth of the voters might opt for an Independent ( as in Vermont ) or a Green or a Reform party candidate. ( This would leave less than a sixth of the voters unrepresented here, for a proportional representation of five-sixth the voters. )

With choice voting, primaries are built into the general election itself, because the most prefered of the party candidates are elected in a multi-member constituency. This is done by all the voters. And even open primaries cannot expect the turn-out of a general election.

Prof. Hugh Bone's study, of New York's use of choice voting, observed:

Some of the most able councilmen were non-organization Democrats and Republicans...who undoubtedly would never have won the primary in a single member district because of opposition from the district machine.

Moreover, the parties can still hold primaries, private to their own members, that is closed primaries, that no longer need be contested, in the constitutional courts, as extremist or factional 'conspiracies' against the public will. This is especially the case because all parties need to field a good spread of candidates, rather than a narrow segment of opinion, to pick up as many seats as possible in a multi-member constituency. The quality of the candidates is essential, also, as personal preference is decisive, with choice voting.

The same sort of reasoning applies to presidential primaries. Each party could put up to, say, five candidates, in a given state. Independents could stand, too, without being discriminated against by a party list system.
( List counting allows votes for one party candidate to go to another on the same list, in a proportional count, without the voters' consent, which gives an unfair advantage against Independent candidates not belonging a list. Advocates of party lists, as 'fair', only mean between parties and uncritical partisans, not the rest of us. )

At any rate, the last but one stage of a presidential election would probably be fought as a two-member election ( like the Senate's two member constituencies ). All but two of the presidential hopefuls, getting over one third the votes each, would be left in the running. The final vote would be a straight fight for over half the votes. This would avoid the undemocratic effect of a candidate winning due to a spoiler splitting the vote of a candidate, who is more prefered to the ( official ) winner.
( Also, this would avoid run-off voting, with its irrational non-proportional re-distribution of preferences from arbitrarily eliminated candidates. )

In a comic thriller, two ex-presidents, played by Jack Lemmon, as a Republican, and James Garner, as a Democrat, end-up by standing for re-election on a 'united' Republican-Democrat ticket. The film ends with them privately squabbling over precedence, as they start their joint campaign.

Choice voting could allow the public settle the squabble. Imagine, again, the choice voting presidential primaries have reached the last stage but one. There may be the two leading Democrats and the two leading Republicans still in the running, and, say, an Independent, like John Anderson, or a Green, like Ralph Nader.

We supposed these five contending for two remaining places. The two most prefered candidates are likely to be a Republican and a Democrat. But assume now that America is in a period of national emergency, in which there is a strong feeling for political unity. Then, the Democratic and Republican parties could agree that whichever of their candidates wins the final round, the straight fight, becomes president, and the other party's candidate becomes vice-president.



(2) Against corporatism.

Proportional representation and partisanship.

Choice voting would much more efficiently effect primaries, and thereby has a potential for hugely increasing voter turn-out. These considerations are of especial prominence in the American system. But choice voting has other democratic functions of comparable importance.

The achievement of proportional representation between the parties has already been touched on. There is no doubt that simple majority elections constrain the voters into a two-party system. For instance, Irish and Prothro admit this in their textbook on American democracy. To the other parties struggling to get a foot-hold in the system, this is all-important. The small party mentality is responsible for much of the exaggerated claims for electoral systems with a proportional count for political parties.

'Proportional representation' means the election of representatives on a proportion of the votes. The point is that each representative, in a multi-member constituency, is elected on the same number of votes, which is an elective proportion or quota of the total constituency votes. Thus, equal representation is ensured by the proportional count.

This system requires a preference vote, so the voters can elect the candidates, to the winning proportions of votes, in order of choice: 1, 2, 3,...etc on the ballot paper. In this way, if your first preference has more votes than needed to achieve her quota, the extra or surplus votes are transferable to second or next choices of candidate. That way, a range of the most popular candidates are elected.

This method is called the single transferable vote; 'choice voting' in the USA and the Hare-Clark system in Australia. STV was originally called Hare's system after Thomas Hare, who thought of it in the mid 19th century -- independently of Carl Andrae, a few months before. This is the original and literal form of PR.

However, when most people speak of proportional representation or PR, they are really talking about the parties getting their 'fair' share of parliamentary seats for votes. Systems, that work on this basis, give an X-vote for a 'party list'. Then the parties get seats in proportion to their respective totals of party votes.
The snag is -- as Enid Lakeman ( How Democracies Vote ) pointed out -- that your X or spot vote, for an individual candidate on a party list, may count towards the election of some candidate on that list, that you didnt vote for.

Altho such systems are still called proportional representation, they are really based on a principle of proportional partisanship. It is a popular fallacy to confuse the two. The former implies the latter but the latter does not imply the former.

These partisan systems have been around in Europe, more or less since the turn of the 1900s. This doesnt stop some electoral reformers describing them as 'modern' or allow them to worry that their contribution to European history may leave something to be desired.

All is forgotten in the desire to remove 'first past the post', except when used in combination with list systems, when simple majorities suddenly and mysteriously become alright again: as long as the smaller parties are served, nothing else seems to matter to such reformers.
In recent years, party lists have spread further, notably when used in combination with the traditional 'first past the post', as 'additional ( list ) member systems' ( AMS ). Usually, this involves two X-votes, one for a single member and one for a party list: it is sometimes called 'the Double Vote.'

First past the post, as of an election 'race', means that the candidate, with more votes than any other, is elected. This usually takes place in single member constituencies, when it is also called a 'winner takes all' system, because the candidate with this 'simple majority' may have a quite small fraction of the constituency vote; much less than an over-all majority of votes, that would show a clear winner. Yet the rep. of the largest faction monopolises the constituency, only having a single seat.

'In July 1997 ( the Center for Voting and Democracy ) predicted the winners in 83% of the November 1998 U.S. House races.' One party held the balance of power in most districts. Its candidate can be bought or 'sponsored' by special interests.
Michael Moore asked why we bother with the party representatives, instead of dealing direct with their big business backers.
The Center said America's 'no-choice' elections must have a demoralising effect on turn-out:

Legislators are already sharpening their knives to carve up the electorate into a new round of safe districts in the redistricting of 2001.

The Center's 1997 prediction was less than twenty years after the National Campaign for Fair Votes made a similar close guess of the results of a British general election. That, by the way, appears to be the inspiration of the 'Fair Votes' campaigns in both Canada and the USA.
North Americans, please note that it once seemed inconceivable that 'the British system' would ever change. But 2000 saw the British mainland with half a dozen different systems, proportional, as well as majoritarian, and all of them bad.

I believe that list counting proportional systems are not democratically acceptable. The fact that they are widespread does not make them right. Many abuses are widespread and party lists are an abuse of individual liberty. List systems are where I disagree with all those 'Fair Votes' campaigners, who believe implicitly that fairness need mean only fairness between parties -- fairness only within the self-appointed political class.

Party list systems, 'fairly' speaking, also, should be called 'no-choice' elections, in that their vote, for a party, over-rides individual representation. To partisans, especially small partisans, that doesnt matter, so long as parties get their 'fair' share of seats for votes to their candidates.

Take the contradiction inherent in the much-advocated form of AMS, the Double Vote. Proportional representation, properly speaking, is about power-sharing but the single member system is about monopolising power. This contradiction hints at further inconsistencies in the system ( explained on my web page, HOW NOT TO DO IT -- a phrase, shouted in capitals, in the reforming novel Little Dorrit by Charles Dickens. )

The 'doublespeak' of the Double Vote is not logical, or even just, but power has its own logic and rough justice, and this cynical business is what you may see much of, in the struggle for and against certain types of electoral reform.


Party lists are an irresponsible corporatism.

To top.

New parties, such as the Greens and the Reform party, have arisen in America, because they feel the two main parties are not properly dealing with vital issues. I happen to believe they are right. In their frustration with the system, they are liable to believe that the ends justify the means. Party list 'PR' would certainly get small parties a place in government. But it would do so at too heavy a price.

For one thing, the lists would be filled by the activists you might not want to represent you. You might be Green but prefer some other candidates -- or some other order -- than the party slate offers.
Choice voting means your vote goes to candidates in the order of your choice.

Party lists are ordered by the organisation man or boss, the sole holder of any preference voting, for which there is no 'universal suffrage.' In Europe, and no doubt North America and elsewhere, there are already small party activists who have sold their souls to oligarchy by lists, so they may come to power by proportional partisanship.

Nor need we be surprised at this. Nineteenth century socialism was supposed to herald a New Jerusalem, by replacing the capitalist with the bureaucrat. The world has still not recovered from the grotesque tragedy and criminal folly of 'the dictatorship of the proletariat' and state socialism.
The socialists have left a power vacuum to be filled by greedy exploitation. Revisionists recognised political democracy, but even they opted for economic bureaucracy. Consequently, economic democracy has struggled to be an issue at all.
( Ive discussed 'Constitutional Economics' and an economic parliament -- or occupational congress -- for 'equality of lobbying', on other web pages. )

Now the Greens are stepping in, not only to safeguard society but nature. And like the socialists, before them, they are choosing bad means to good ends. The support of undemocratic reforms, like party list systems, or their hybrids, can be guaranteed to have unfortunate consequences for the common weal.

Much of the blame for degradation of the eco-system goes to the inequitable nature of corporate law, that treats a company as an 'individual', so that no individual directors are held responsible for profitable vandalism. This may be Capitalism's worst corruption. br /> Yet, party lists are a form of political corporatism, in which the people are expected to vote for a 'party', as if it were an individual. The same irresponsibility is being fostered in politics as in economics.
In logic, this is the fallacy of confusing a member with its class.

Moreover, party lists make political ghettos of an issue, such as the environment. The false implication of the list system is that only the Green party is green, only the Reform party reformist, and so on. This may encourage ruling politicians to regard green politics or reform as a marginal issue.

It is not hard to guess that Green partisans may be seduced by the dream that they shall sweep all before them, electorally, and rapidly become the governing party. But this is a gamble and the Greens are supposed to be against gambling with the future of life on this planet.
By 2000, the world's governments were unwilling to take even minor preventive action against global warming.

And there is a much better way, that could be so easily implemented.
Choice voting would allow voters to prefer the most environmentally-friendly candidates from all parties, if they so wished. The issue neednt be marginalised, by just having an X-vote to vote for a Green party list. That might not suit some green activists, but certainly, it would be a way for public interest issues in general to permeate representative bodies.


PR and holding to democratic account.

To top.

The separation of powers in the US constitution shifts the emphasis on which features of choice voting are most important for its democracy, but does not lessen that importance. In countries ruled by parliamentary cabinet governments, the little understood value of STV is that it can pick out a particular coalition, prefered by a majority of voters, if one party falls short thereof.

Choice voting allows one to extend one's preferences from candidates of one's most prefered party to those of one's next prefered party. The combination prefered by a majority of voters will have the majority of seats in parliament to form a government.

In America, the president forms the government, so STV is less needed in its role of forming the democratic choice of coalition. On the other hand, we have seen that the American system needs choice voting for more effective primaries.
STV works, for unity in diversity, both with parliamentary coalition-forming and presidential primaries or primacy-forming.

State or federal congresses need choice voting to ensure that the group composition of their majorities are made up the way the majority of Americans prefer. They may be legislatures, rather than executives. But the president or state governor still depends on a friendly majority to see thru his legislation. It may not matter too much, for the American system, if he does not have that friendly majority. What matters is that the majority, friendly or unfriendly, should be representative of the voters.

The unrepresentative nature of first past the post is as evident in the USA as other countries.
For example, it turned Ulster into a one-party state governed by the Unionists for over fifty years. Eventually, the Northern Irish Catholics emulated the peaceful civil rights movement, in the USA, and went on the march. This was not tolerated and a sort of drawn-out civil war followed, in which crimes against humanity were committed, by various factions, including against their own communities.

Likewise, the USA found itself with 'the one-party south' after the civil war. The continued oppression of the Afro-American might have been mitigated by a proportionate voice in government, there.

First past the post fails democracy when it fails to adequately represent minorities, because effective democracy depends on a strong opposition to keep the ruling group in line. First past the post gives the biggest party a monopoly of representation, so they can do what they like, without parliamentary criticism, to give authoritative challenge.

The single member system is also prone to gerrymandering, the drawing of constituency boundaries to include a safe majority of a given candidate's supporters, so that the result of the election is effectively decided before it's held.
Whether the boundaries are drawn with fraudulent intent or not, single member constituencies are monopolistic. And the calculating side of human nature is bound to challenge the process on the grounds that an 'unfair' fraction of some candidate's natural constituency may be left out of bounds to his candidacy.

At any rate, single-member boundary-drawing is especially contentious. But the two interested parties have the motive to carve up the country between them, so they each have their 'fair share' of safe seats, whose bounds are a sort of lassoo round the representatives' majorities and unwilling minorities, in every constituency.

A study by the League of Women Voters ( Seattle ) says that gerrymandering is widespread, mentioning two kinds: fragmenting splits up residential blocs of support for the rival party into different districts; the sweet-heart gerrymander is an agreement between the two parties to have the boundaries drawn so they each get safe seats.

'Affirmative gerrymandering' was an attempt to make the system work for minorities, such as the blacks. Enid Lakeman called it an exercise in futility. The idea was to make this inequitable or unfair system work unfairly to the advantage of the unfairly unrepresented minorities!
The blacks rarely have a majority in any constituency, but salamander-like boundaries could be drawn to artificially include a black majority.

Later, the Supreme Court decided gerrymandering, affirmative or otherwise, was unconstitutional, with regard to a black Congress-woman's constituency. To which, coming from the citrus state of Georgia, Cynthia McKinney made the memorable reply: Today, they have given us lemons, but we shall turn them into lemonade.

The realisation is spreading in America that first past the post cannot deliver equality, and counter-acting inequality is no remedy.
Proportional representation seems all set to be a plank in the platform to make racial equality a reality, as well as a campaign objective for organisations like The Alliance for Democracy, co-operating with The Center for Voting and Democracy.

Perhaps the most notorious example in local government was in New York, controlled by Tammany Hall, which became a by-word for corruption. Less remarked are its antics, afterwards, to suppress proportional representation.
Tammany Hall held three referendums, in a row, before it could rid New York local elections of PR, even with the big money and publicity on its side.
Needless to say, there was no attempt here to inform public opinion or respect the public will, only to impose on the public, to get the electoral law changed in Tammany's own office-holding interest.

Bringing back first past the post effectively abolished the opposition and all criticism. The Tammany Democrats had a free hand to do what they did: mismanage, misappropriate and bankrupt the city, which they then handed over to big business.


Of course, Cambridge is the great success story of choice voting PR in local government. This is the home of the famous Massachusetts Institute of Technology.
( MIT announced, after the 2000 Presidential election controversy, it was taking part in the preparation of a new standard of balloting technology. )
It stretches credibility that if choice voting was not an especially good system, that those critical minds would not have rejected it.

The League of Women Voters have been studying alternative voting methods. The Pasadena chapter refered to a guest columnist. David Sullivan had much experience of PR in Cambridge:

Cambridge is a diverse city of about 100,000 residents, including many ordinary working families as well as the university population for which it is better known. It is a melting pot of many races and cultures (literally dozens of languages are spoken at our high school), including a large African-American population.

The genius of PR is that it successfully represents all these varying interests on our local governing bodies in proportion to their strength in the electorate. For example, our six-member school board (the Mayor casts the seventh vote) now consists of four women, including two African-Americans and a Latina who is a union organizer, and two white men from the more traditional Italian- and Irish-American neighborhoods.

Without PR, I am sure that much of this diversity would be lost from our elected bodies. Yet, these different individuals work remarkably well together, and in my opinion have served as a force for civic unity among the interests they represent. For example, last year our School Board hired a new superintendent with very little rancor or divisiveness. In general, that has been our experience with PR over the years, so that no one now seriously proposes to change or get rid of it.

Hopefully, Cambridge choice voting will be the model for electoral change in America. Choice voting or the single transferable vote is a general theory of choice, which uniquely applies to elections in general, political or non-political, representative or referential.

First past the post initiatives and referendums in the USA are liable to impose the will of some larger faction. This has been the case with successful initiatives to ban the teaching of evolution or 'Darwinism' in schools, naturally to the annoyance of a good many American citizens!
( Ive discussed this question on my web page on referendums -- and to a lesser extent on my first web page on the Kerley report. )


Conclusion: back to basics.

The democratic world, such as it is, is in a hopeless mess of conflicting and varying voting systems, because the first principles of electoral logic are scarcely understood. As J F S Ross said, in Elections and Electors, an election consists of a vote and a count. The vote is for the individual, the count is its aggregation to a community decision. For this to be achieved, the magnitude of choice must carry over from vote to count.

With a spot vote, the voter can only express a greater choice for one candidate over another. And this can only consistently carry over, in the count, as a greater choice, or majority, of the community for one candidate over another.

Choice voting or STV is the system that consistently generalises greatness of choice, from this special case, in both the vote and the count. That means that a spot vote, for one candidate over another, is generalised to a preference vote, that gives ranked choice of candidates, for a generalised count, by way of relative majorities, in a multi-member constituency. ( The Droop quota count, used with STV, is simply a rationalisation of an over-all majority count. Again, I have discussed these things more fully on other web pages. )

Choice voting is the consistent logic of electoral choice.
Failure, to realise this, is why the world is awash with obsolete or half-witted voting systems. Some rely on first past the post, no longer remotely adequate for a choice of more than two candidates.
Some rely on exhaustive ballots or preference votes without a proportional count, and vice versa. Tho, relative greatness of choice in a preference vote is meant to translate into relatively great majorities, in a proportional count.
An electoral system that uses a preference vote, not proportionally counted, is like a question without a rational answer. And a system that uses a proportional count without establishing voters' order of choice, in a preference vote, is like an answer that begs the question.

In the latter case, the parties have already dogmatised that the voters are all obedient partisans voting for a party. And the question who shall be the voters' individual representatives is decided by the preference vote ballot papers, exclusive to a few party bosses and called party lists.

Everyone may have a vote, but universal suffrage is far from being achieved, if preference voting is exclusive to party bosses and proportional counting exclusively for party lists: one social group's oligarchic control of the count, for their leaders' choice of individual political officials.


Richard Lung



To top.

To home page.