One of the most critical priority for any democracy is to improve its electoral system and start using a much better voting method.
Single Choice Voting is the most prevalent and most underestimated threat to democracy. It is present in virtually every electoral system on earth — in different forms, under different names, dressed in the language of tradition and simplicity — and it is almost never identified as a problem. That invisibility is part of what makes it dangerous.
The mechanism is structural: by restricting voters to a single mark on a ballot, the system mechanically distorts the political landscape, suppresses genuine diversity of opinion, and produces a predictable family of pathologies known as the Duverger Syndrome. These pathologies in turn create the conditions that corrupt actors exploit — the polarisation, the captured institutions, the manufactured loyalty — giving rise to the Tweed Syndrome.
The mechanism was identified as far back as the 1950s, when Maurice Duverger documented the gravitational pull of single-winner voting toward two-party systems. The knowledge has existed for decades. The reform has not followed. This is a stubborn instance of the Knowledge-Application Gap — humanity knowing the answer and failing to apply it.
Understanding Single Choice Voting — what it is, what it does, and why the conventional distinction between plurality and proportional systems misses the point — is one of the central objectives of the Pildem Framework. The ballot constraint is the root cause. Everything downstream follows.
Single Choice Voting is the umbrella term for any electoral method in which the voter is restricted to selecting one and only one option — whether that option is a candidate, a party, or a list. The seat-allocation formulas differ. The ballot instruction does not.
Four main variants share this constraint:
Many candidates may appear on the ballot — sometimes dozens — representing the full range of political opinion in a constituency. The candidate with the most votes wins, whether that is 51% or 31%. No majority required. Used in the United States, the United Kingdom, Canada, India, and most former British colonies.
The voter marks one name. Every other candidate — however acceptable, however preferred, however closely aligned with the voter's actual views — receives nothing from this voter's expression. A single mark must carry the full weight of political judgment.
Party lists may be numerous, each representing a distinct set of priorities and values. Seats are distributed to parties in proportion to their vote share — a fairer allocation than plurality, but built on the same constrained foundation.
The voter may declare support for one party. A voter who agrees with Party A on the economy, Party B on the environment, and Party C on criminal justice has no way to say so. The full texture of political opinion — the agreements, the partial sympathies, the considered trade-offs — is compressed into a single party name and discarded.
The first round may present a genuine field — five, eight, a dozen candidates. The second round always reduces that field to two. Used in France and in many presidential elections worldwide.
In neither round may the voter do more than name one. Not this one and that one; not broadly acceptable to three — only: this one. The decisive round is always a forced binary, regardless of how many candidates the voter actually wanted to weigh in on. Two chances to express a preference; both of them single.
Multi-seat districts, one ballot mark per voter. The top N candidates win the N available seats. Used historically in Japan and in Taiwan's legislative elections before the 2005 constitutional reform.
The single-choice constraint does not soften in this multi-seat setting — it compounds. Because multiple candidates from the same party compete for shares of a vote pool that cannot be split, the party's rational response is to override voter preference entirely: routing sub-groups of supporters to specific candidates to optimise seat yield. The voter's individual judgment becomes an obstacle to be managed rather than a preference to be honoured. One vote, many seats available — and less genuine voter choice than any other variant.
Political science draws a sharp distinction between plurality systems and proportional systems, treating them as fundamentally different answers to the question of how to count votes. And in terms of seat allocation, they are different — in how governments form, in how large or small parties can survive, in whether a 10% vote share earns 10% of seats or none.
But from the voter's perspective — from the perspective of democratic expression — they share the same critical flaw: the voter cannot express an opinion on more than one option. All of the diversity of a voter's actual views, all of the agreement they might have across different candidates and parties, collapses to a single mark on a single line. The system then treats that mark as a complete statement of political preference.
It is not. It is the best approximation a constrained instrument allows. And the constraint, applied across millions of voters, produces systematic distortions — the same family of distortions, in every country that uses any variant of single-choice voting.
When voters can only pick one candidate, every additional candidate risks splitting the vote among people who share broadly similar preferences. The rational response — taught to voters by experience — is to abandon the preferred candidate and vote for whoever can defeat the candidate one most wants to stop. This is tactical voting: not expressing what you want, but calculating what you can survive losing.
Under sustained tactical pressure, smaller parties are gradually squeezed out. The wasted-vote psychology tells their supporters to defect before the vote is wasted. Over election cycles, the field narrows. Two dominant parties emerge and entrench. This is the mechanism behind Duverger's Law — the empirical observation that plurality voting generates two-party systems — and it is one component of a much larger family of pathologies.
The binary structure then produces secondary distortions: safe seats that insulate incumbents from general-election accountability; party primaries that hand nomination power to the most committed partisans; spoiler effects where a third candidate's presence determines the winner without winning themselves.
Proportional seat allocation softens the wasted-vote problem — a party with 10% of the vote wins roughly 10% of seats, rather than near-zero seats as under plurality. More parties survive. But the single-choice constraint remains, and with it a distinct set of distortions.
A voter who agrees with several parties across different policy areas must still reduce that agreement to a single mark. In closed-list systems — the most common form — party leadership decides the rank order of candidates on the list, giving voters no direct say over which individuals their vote elevates. The constraint falls on the voter twice: once in the choice of party, and once in the delegation of candidate selection to the party apparatus.
Coalition negotiations happen after the vote, in rooms the voter never enters. Voters cannot express preferences about which coalition they want to see formed — the combinations emerge from post-election bargaining between party leaderships. When coalitions collapse — as they do with regularity in Italy, Israel, Belgium, and the Netherlands — the government falls not because voters changed their minds but because the post-election bargain broke down.
At the macro level, proportional systems still tend toward two dominant blocs: a centre-left alliance and a centre-right alliance. The adversarial dynamics that define plurality politics reappear at the bloc level. The binary survives; it is merely elevated one tier.
Two rounds appear to dissolve the binary problem: more candidates compete in the first round, and voters get a second opportunity to choose. In practice, the second round is always binary — two survivors compete one-on-one — and strategic calculations migrate to round one.
Voters in two-round systems face a version of the same wasted-vote dilemma: should they vote sincerely for their preferred candidate in round one, risking the elimination of a candidate they might have helped advance? Or should they vote strategically to shape which two candidates reach the decisive round? The first round becomes a negotiation with hypothetical outcomes, not a sincere expression of preference.
Two rounds also do not prevent extreme candidates from reaching the decisive choice. In France's two-round presidential system, the far-right candidate reached the final round in both 2017 and 2022 — becoming, each time, one of the only two options the entire electorate had left. The first round delayed the reckoning; the second round made it inescapable.
The constraint is structural, not natural. It was designed into the ballot, and it can be redesigned out of it.
Approval Voting replaces "pick one" with "pick all you approve of." The ballot looks identical. The instruction changes. Voters can support every candidate they find acceptable, without splitting the vote or sacrificing honest preference to tactical calculation. Counting is equally simple — add up the approvals, highest total wins.
Score Voting goes further: voters assign a numerical rating to each candidate. The full range of preference becomes expressible — not just approval or rejection, but degree. The candidate with the highest average score wins.
Informed Score Voting introduces two departures from Score Voting. First, the rating scale becomes symmetric around zero — negative and positive ratings mirror each other (for example, −5 to +5) — so that voters can express opposition, not merely varying degrees of support. Second, a dedicated "I don't know" option is added, distinct from any numerical rating. Voters should not be forced to score candidates they have no basis to evaluate; uninformed ratings distort outcomes, and the "I don't know" option removes the pressure to manufacture a judgment one does not hold.
All three methods share the property that dissolves the core pathology: voters can express their opinion on every candidate simultaneously. The spoiler effect vanishes. Strategic voting is dramatically reduced. The structural pressure toward two-party dominance dissolves. The Duverger Syndrome — the full family of pathologies that single-choice voting mechanically produces — depends on the ballot constraint. Remove the constraint, and the syndrome loses its mechanism.