The letters to editor that NY Times published in response to H.Dean pro-IRV op-ed

[NY Times readers discuss the risks and benefits of instant runoff ranked voting, as proposed by Howard Dean in October 2016. See also our own response containing point by point corrections of errors by Dean.]

To the Editor: Re "How to Get Beyond Two Parties" (Op-Ed, Oct. 8): I fully support Howard Dean's proposal for ranked voting (a k a instant runoff, a k a the Australian voting method). This would be especially useful in primary elections, where we often have more than two candidates.

Often, one is forced to vote for the "second best" for fear of tilting the election to the worse candidate. In this method, voters can vote for their first choice without risking "giving the vote away" and unwittingly supporting the candidate they really do not want. No voting method is perfect, but this proposal is definitely a more democratic alternative than the current winner-take-all approach.

DANIEL SZYLD Philadelphia [The writer is a professor of mathematics at Temple University.]

Comment: Instant runoff is used to elect members of the Australian Federal House, but not its Senate.

To the Editor: Howard Dean's objective is laudable, but his choice of the ranked-choice voting system is not only unlikely to achieve his objective, but also has many serious faults. He notes that the system has been in use in national elections in Australia and Ireland – but it hasn't got them beyond two parties.

Voting according to my preferences, or even choosing to vote at all, should not decrease the likelihood of a desirable election outcome. Yet each of these undemocratic faults can occur in ranked-choice voting, and not uncommonly in close elections.

Mr. Dean's objectives would be better served with approval voting, a simpler system that voids such faults, in which voters can vote for all the candidates they want to support, the winner being the candidate with the most support.

ROBERT Z. NORMAN Hanover, N.H. [The writer is emeritus professor of mathematics at Dartmouth.]

To the Editor: Kudos to Howard Dean for supporting ranked voting. There is little question that if ranked voting were in place today, nationwide, and if Bernie Sanders were on the ballot, he would be elected president rather than an unpopular Hillary Clinton or Donald Trump.

But here's the conundrum, which is not dissimilar to the Security Council veto that can be changed only by vote of the five countries that wield it. How likely is it that ranked voting will be supported by the two major parties, since it will weaken the very political stranglehold they perpetuate, whose victim in this election, like Mr. Dean in an earlier one, was the nontraditional and independent-minded candidate, Bernie Sanders?


Comment: I have poll data and it does not support Saltarelli's evidence-free contention that Instant Runoff would have elected Sanders "with little question." Indeed it is probable it still would have come down to Clinton vs. Trump in the IRV final round, with the victor among those two being unclear. However, both Sanders and Kasich were far more approved than either Clinton or (especially) Trump, i.e. clearly would have defeated them using approval voting (and Sanders was the most-approved candidate in the USA). There also is poll-based evidence they would have defeated Clinton and Trump using score voting.

To the Editor: Voters already have "more than two options" when campaigns begin. You can vote for your favored option in the primaries. If by Election Day most voters have spurned your favored option, that does not mean you're being deprived of a chance to vote your conscience; you already did.

To accept the constraints of the majority's decision is not to endorse it but merely to uphold majority rule. Upholding the basic tenet of democracy should not strain your conscience.

ILYA SHLYAKHTER Cambridge, Mass.

Comment: Actually, the reason Trump won the Republican primary was flaws in the present plurality voting sytem used in those primaries, e.g. vote-splitting and "fear of wasting your (honest) vote." Trump would not have been the Republican nominee with approval or score voting. Polls show these facts. Therefore, Shlyakhter is incorrect to say "you can vote for your favored option in the primaries" for the same reasons he already agrees that is untrue in the second (general election) round; and he also is incorrect to imply (if he does) that the voting system itself is not a problem.

To the Editor: Howard Dean extols ranked-choice voting without mentioning its antidemocratic features. The most serious is that it is "non-monotonic," which means a voter can raise the ranking of a candidate, even to first place, and by doing so cause him to lose because of anomalies in the transfer of votes as candidates are sequentially eliminated. This is not a rare event but can happen surprisingly often in close elections. If there is anything antithetical to a democracy, it's that giving more support to a candidate should hurt rather than help.

Mr. Dean mistakenly thinks that ranked-choice voting supports the principle of majority rule. But this is not true, even when the sequential elimination of weaker candidates whittles the number down to two. The reason is that voters who supported weaker candidates can have all their preferred candidates eliminated, so in the end these voters are not counted in the contest between the final two.

These and related problems cropped up in the 2009 mayoral race in Burlington, Vt., which had adopted ranked-choice voting. Burlington voters voted to repeal it, as have voters in other cities that had similar unfortunate experiences. Moreover, countries like Australia that have long used ranked-choice voting have remained essentially two-party systems.

It makes no sense to adopt a fundamentally flawed voting system when there are simpler alternatives, like approval voting, that do not suffer from such failures and would facilitate the entry of new candidates and parties.

STEVEN J. BRAMS New York [The writer is a professor of politics at New York University.]

To the Editor: Howard Dean's suggestion that the United States should adopt ranked-choice voting is long overdue. Australia, whose electoral system I have studied extensively, has used ranked choice for almost a hundred years.

Beyond giving voters the ability to take a chance on a third party, ranked choice would immediately defuse the toxic polarization of our national political scene by psychologically reframing how we see the choices we face. Replacing our red-versus-blue Armageddon with something more nuanced would make us a less angry country almost overnight.

Ranked choice would also make voting more complicated; this could overwhelm some voters and depress turnout. Australia deals with this problem effectively by combining mandatory voting with some ingenious procedures that simplify the decision process while preserving an enviable level of freedom of choice. We would do well to learn from Australia's example.

MATTHEW G. NAGLER Livingston, N.J. [The writer is a professor of economics at CUNY.]

Comment: "Defuse" being a subjective word, we cannot definitely refute Nadler's claim IRV would "defuse" the USA's "toxic [2-party] polarization." However, there is reason to doubt this: Australia despite nearly 100 years of IRV evolved to a state of massive 2-party domination in its IRV-elected house, and over 85% of Australian IRV voters appear to vote in a strategic and dishonest "naive exaggerated" manner re the top two parties, which in turn assures continuation of that 2-party domination, by theorem. In contrast, with score voting, naive exaggeration voter behavior does not force 2-party domination, and indeed in the only governments to have used approval and score voting, it appears 2-party domination was wholy or partly avoided, usually wholy.

To the Editor: Why does Howard Dean promote a new system of voting – ranked choice – when a system has long existed that gives third parties a real voice in government?

A proportional representation system, used throughout much of the world, allows third parties representation equal to the percentage of the popular vote they obtain.

If, say, Socialists get 10 percent, they get 10 percent of the legislative seats. Then the minority party can, as needed, forge coalitions with dominant parties, force change and be heard.

What Mr. Dean proposes is the illusion of choice, a Burger King menu of candidates that, by a process of elimination, ends up with the winner-take-all results we have now. Such a system guarantees the marginalization of third parties regardless of their merit.

NEIL MULLIN Montclair, N.J.

Comment: Proportional representation is inherently not capable of electing single-winner seats such as US President, US State Governors, and (each year's) US senators. If however the USA completely rewrote the constitution (for example, getting rid of the President) and changed its system of government entirely, then it could adopt PR.

Return to main page