The secret ballot: how private voting became a public safeguard

In plain words

For a long time, people voted in public. They called out a name, raised a hand, or handed in a paper that everyone could recognize. That made it easy for an employer, a landlord, or a party worker to check how someone voted, and to reward or punish them for it.

The secret ballot changed that. The government prints one ballot with every choice on it. You mark it alone. You put it in a locked box. Nobody can prove how you voted, so nobody can safely buy your vote or punish you for it.

Secrecy did not end every problem. Some people who used to buy votes started paying people to stay home instead. And secrecy only holds if every step, from the booth to the count, protects it. That is the lesson this chapter carries into the rest of the framework: a private vote is a whole system, not a button.

Defined terms

• Open voting (also called public voting or voice voting). Any method where others can see or hear how a particular voter chose. Examples include calling out a choice, a show of hands, or handing in a ballot that is visibly different from others.
• Party ticket. A ballot printed and handed out by a political party rather than by the government. Before reform in the United States, these were often easy to tell apart by color or design ⁵.
• Secret ballot (Australian ballot). A ballot printed by the public authority, listing all qualified choices, marked in private, and deposited so that it cannot be linked to the voter ^{2 5}.
• Ballot secrecy. The property that no one, including officials, can connect a voter to the choices on their ballot. In this framework it is a requirement of the private vote, and it is defined in the glossary.
• Coercion and vote buying. Pressure or payment intended to control a person's vote. Both depend on the ability to check what the voter did. See intimidation and coercion.

Evidence

History: open voting and its pressures

In the Australian colonies before reform, the Australian Electoral Commission notes that people voted publicly, which "left them vulnerable to intimidation and coercion" ¹. The political scientist Marian Sawer, writing for the Australian Senate's Papers on Parliament series, describes nineteenth century British elections as associated with "violence, drunkenness and bribery," and argues that the freedom to vote "without intimidation or corruption" is a less celebrated but significant part of democratic history ².

The United States followed a similar path. According to the Congressional Research Service, early American voters often announced their choice aloud or raised their hands. By the middle of the nineteenth century written ballots had largely replaced voice voting, but they did not add much privacy, because most voters used party tickets that were "readily distinguishable" and handed to officials "in public view" ⁵. That lack of privacy made it easier to confirm a purchased vote and harder to shield a choice from someone trying to coerce it ⁵.

History: the reform

Sawer dates the secret ballot to 1856, when Victoria and South Australia introduced it. She describes the system precisely: voters had their names marked off the roll at the polling place, received a printed ballot paper, marked it in a separate compartment, and placed it in a locked box watched by the presiding officer, poll clerks, and scrutineers ². She credits Henry Chapman with the Victorian ballot provisions and notes that he had seen employers intimidate workers under open voting in Canada ². She also connects the reform to Chartism, whose six points included the ballot ².

The idea moved. Sawer reports that concern for orderly elections helped pass the United Kingdom's Ballot Act in 1872 ². That statute is recorded as the Ballot Act 1872, chapter 33 ⁴. In the United States, the Congressional Research Service reports that Louisville, Kentucky, and the State of Massachusetts adopted the Australian ballot first, in 1888; thirty-eight states had it by the end of the nineteenth century, and all had it by the middle of the twentieth ⁵. Today, CRS states, secret ballot provisions "appear in all states' laws or constitutions" ⁵.

The secret ballot is also an international human rights standard. Article 25 of the International Covenant on Civil and Political Rights, adopted in 1966, says elections "shall be held by secret ballot, guaranteeing the free expression of the will of the electors" ⁸.

History: the reform was more than a ballot paper

The Congressional Research Service notes that states first tried to regulate party tickets and that parties found workarounds, for example using different shades of paper when one color was required. The more successful change was comprehensive: government funding, printing and distribution of ballots listing all nominated candidates, private booths, and rules against showing a ballot with intent to reveal one's vote ⁵. CRS calls "secret ballot" "often shorthand for a system of election procedures and technologies" ⁵.

Research results: what secrecy changed

Researchers disagree about how much secrecy itself changed behavior, because the reforms changed several things at once.

• Gary Cox and J. Morgan Kousser studied New York turnout from 1870 to 1916 and used contemporary newspaper coverage to test the claim that the secret ballot reduced reported turnout by suppressing rural corruption. Their abstract concludes that neither leading theory of the turnout decline holds up well against the evidence, and it proposes a combined explanation ⁶.
• Daniel Reed's analysis of statewide gubernatorial elections from 1870 to 1910 challenges the "vote market hypothesis." He finds that turnout declined significantly where reform ballots were organized by office ("office bloc") but not where they kept a party column layout, and he argues that ballot design explains more of the decline than secrecy does ⁷.

Taken together, these studies support a cautious reading: the reform bundle changed elections, but the specific effect of secrecy on bribery is contested in the research literature.

Historical observation: secrecy has motives beyond fairness

The same CRS analysis notes that supporters had mixed motives. Reformers and labor organizations wanted to curb vote buying and worker intimidation. Some White Southerners saw the new ballot as a way to disenfranchise Black voters, and some party leaders wanted to shift ballot costs to government or limit local officials' power to "cut" candidates ⁵. A government-printed ballot that must be read and marked without help can exclude people who cannot read it. That connection is taken up in access and exclusion.

Competing interpretations

Voting as a public trust. John Stuart Mill, in Considerations on Representative Government (1861), argued that political voting should normally be public. He accepted that secrecy "is justifiable in many cases, imperative in some," but held that the vote is "a trust for the public," and asked: "if the public are entitled to his vote, are not they entitled to know his vote?" ³. His worry was that the ballot would teach voters that the suffrage was for "his particular use and benefit" ³. Sawer records that the novelist Anthony Trollope opposed the ballot as "unmanly," though after visiting Australia in 1872 he admitted it was popular and led to more orderly elections ².

Secrecy as protection of the weaker party. The reform tradition, from Chapman's experience in Canada to the labor organizations described by CRS, treats secrecy as a shield for people whose livelihoods depend on others ^{2 5}. On this view the accountability Mill wanted runs the wrong way: it makes the voter answerable to whoever holds power over them.

Interpretation, this site's reading. Both positions agree that a public vote can be checked. They disagree about who should be allowed to check it. Modern law, including Article 25 of the ICCPR, sided with secrecy for the ballot itself ⁸. Mill's concern does not disappear, though. It survives in a different place: people can still argue in public, explain their reasons, and be accountable for what they say, while the decisive choice stays private. That is the division this framework draws between opinions and the private vote.

Secrecy and help. Secrecy can collide with access. Federal disability guidance in the United States requires that voters with disabilities be able to vote "on the same terms and with the same level of privacy as other voters," and also that a voter "must be allowed to receive assistance from the person of their choosing" ⁹. A trusted helper is sometimes the only route to an independent vote; a helper can also be a source of pressure. Neither value simply overrides the other.

Application to this framework

These are design implications drawn from the history above, not claims that the framework has solved them.

1. Secrecy is a requirement of the whole process. CRS describes ballot secrecy as a product of "where and how ballots are cast," not only of the ballot ⁵. A future private vote would need secrecy at registration, casting, transmission, storage, counting, audit, and publication. See security requirements and audits and verification.
2. No proof of choice for anyone. The historical abuse depended on someone being able to confirm a vote. That is why My Vote Forward has no ballot share button, no public vote history, and no person to ballot research product, and why a vote's privacy cannot be switched off. Voluntary opinions are a separate, specific, consented act.
3. Remote settings reopen old problems. The polling booth separated the voter from observers. A vote cast at home or at work does not automatically have that separation. CRS notes that some security experts consider internet voting's secrecy risks hard to mitigate fully ⁵. See intimidation for household and employer pressure.
4. Published data can leak secrets. CRS reports concerns that proactively released election data could be cross-referenced with voter files to link voters to choices ⁵. This is one reason the framework uses scheduled, reviewed published releases instead of an open data feed.
5. The demo is not this system. The companion demo simulates election state. It does not cast official votes and does not claim the secrecy properties that an authorized election would need.

Hypothetical example. A company asks staff to "share your ballot" in an internal app as a show of team spirit. Even if sharing is voluntary, the request creates pressure on anyone who chooses not to share, and silence itself becomes informative. A design that offers no way to export a ballot choice removes the question entirely.

Unresolved questions

• How much of the late nineteenth century change in turnout and corruption came from secrecy, and how much from ballot layout and party behavior? The studies cited above disagree ^{6 7}.
• Can a remote voting channel ever match the separation a staffed booth provides, or should remote voting always be paired with a way to cast a later ballot that replaces an earlier one? This is an open research and legal question, and it varies by jurisdiction.
• How should assistance be designed so that a voter chooses a helper freely, and the helper cannot become an enforcer?
• Mill's question still stands for a modern framework: if the ballot is private, what forms of public reasoning keep voters and representatives accountable without exposing the individual choice?