Week 4: The secret ballot and the history of who may vote

In plain words

Week 4 of the My Vote Forward seminar covers two linked histories: why voting became secret, and how access to voting has been restricted and protected. Students read the international law standard, the Australian Electoral Commission’s account of the “Australian ballot,” a classic study of what ballot reform changed in the United States, a survey of whether people believe their ballots are secret, and the Supreme Court’s 2013 decision in Shelby County v. Holder.

Until the 1850s in Australia, people voted publicly, which left them open to intimidation and coercion. Secret ballots were the fix. But secrecy is only half the story. Rules that look neutral, such as literacy tests, were used to keep eligible people from voting. Any new way of taking part has to protect secrecy and avoid new exclusions at the same time.

Defined terms

• Secret ballot: a voting method in which no one can learn how a particular voter voted. See secrecy.
• Australian ballot: an official, government-printed ballot listing all candidates, marked in private. The Australian Electoral Commission notes it is “now known as the secret ballot.”
• Coercion: pressure that makes a person vote or express a view other than their own. See coercion.
• Preclearance: the requirement, under Section 5 of the Voting Rights Act, that covered jurisdictions obtain federal approval before changing voting procedures.
• Coverage formula: Section 4(b) of the Voting Rights Act, which defined which jurisdictions were subject to preclearance.
• See also eligibility and registration.

Learning objectives

By the end of week 4, a participant should be able to:

1. State the international legal standard for secret ballots and its source.
2. Explain what open voting allowed and why the secret ballot was introduced.
3. Explain, with Rusk’s study, why changes in voting rules must be considered before attributing changes in voting behavior to voters’ attitudes.
4. Describe what Gerber and coauthors found about beliefs in ballot secrecy and why perception matters.
5. Summarize the holding of Shelby County v. Holder and the strongest arguments for and against it.
6. Apply secrecy and access tests to a proposed remote participation channel.

Core readings

1. International Covenant on Civil and Political Rights (1966), Article 25.⁠¹
2. Australian Electoral Commission, “Australian voting history in action”.⁠²
3. Alexander Keyssar, The Right to Vote: The Contested History of Democracy in the United States, revised edition (Basic Books, 2009), selected chapters.⁠³
4. Jerrold G. Rusk, “The Effect of the Australian Ballot Reform on Split Ticket Voting: 1876 to 1908,” American Political Science Review 64(4), 1970, pp. 1220 to 1238.⁠⁴
5. Alan S. Gerber, Gregory A. Huber, David Doherty, and Conor M. Dowling, “Is There a Secret Ballot? Ballot Secrecy Perceptions and Their Implications for Voting Behaviour,” British Journal of Political Science, published online 2012.⁠⁵
6. Shelby County v. Holder, No. 12-96 (2013), syllabus and majority and dissenting opinions.⁠⁶
7. U.S. National Archives, “Voting Rights Act (1965)”.⁠⁷

Lecture notes

Why the secret ballot is a starting point

The secret ballot is the institution this course measures every other design against, and week 4 asks why. Article 25 of the International Covenant on Civil and Political Rights states that every citizen shall have the right and opportunity “to vote and to be elected at genuine periodic elections which shall be by universal and equal suffrage and shall be held by secret ballot, guaranteeing the free expression of the will of the electors”.⁠¹ The phrase that matters is the last one. Secrecy is not valued for its own sake; it is valued because it guarantees free expression. A ballot that others can see is a ballot others can pressure.

The site’s history chapter on the secret ballot traces the institution in detail. This lecture focuses on what the history teaches a designer.

What open voting allowed

The Australian Electoral Commission’s official history puts the problem plainly: “Until the 1850s, people voted publicly, which left them vulnerable to intimidation and coercion.” Its account continues that an independent electoral body was established in the 1850s “to ensure elections were run fairly, and to this end the so-called ‘Australian ballot’, otherwise now known as the secret ballot, was implemented”.⁠² The same page notes that South Australia extended the vote to male residents aged 21 and over in 1856, with Victoria following two years later.

Notice the pairing. Secrecy and an independent administrator arrived together. The lesson is that secrecy is not only a property of a piece of paper; it depends on who controls the process and whether they can be trusted not to look. Week 11 returns to this point for electronic systems, where the administrator’s control over software replaces control over the polling place.

Rusk: rules shape behavior

Jerrold Rusk’s 1970 study of the Australian ballot in the United States makes a methodological argument this seminar relies on. Rusk notes that research on voting had focused on the “sociological, psychological, and political antecedents of the vote” while neglecting “the institutional or structural properties of the electoral system itself,” such as “electoral qualification requirements, registration laws, and ballot and voting systems.” He argues that these properties “provide the framework within which the effects of other independent variables must be judged,” especially in longitudinal research. His pointed example is Walter Dean Burnham’s account of a “changing political universe” around the turn of the twentieth century, which Burnham attributed to declining party competition and voter alienation; Rusk argues it “undoubtedly could be partially, if not largely, explained by reference to the many institutional changes in voting rules which occurred during this period”.⁠⁴

The point carries into every later week. Before concluding that people have changed their minds or lost interest, ask whether the rules for expressing a view changed. A new participation channel is itself a rule change, and its results will reflect its design as much as the public’s views.

Gerber and coauthors: secrecy in the mind

A secret ballot that voters do not believe is secret may not protect them. Alan Gerber, Gregory Huber, David Doherty, and Conor Dowling added new items to a nationally representative United States survey and report that “25 per cent of respondents report not believing their ballot choices are kept secret and over 70 per cent report sharing their vote choices with others.” Their regression analysis shows that “social forces appear to have a greater effect on vote choices among people who doubt the formal secrecy of the ballot.” The broader claim is that “the intended benefits of institutional rules may not be realized if people’s perceptions of these rules differ from their formal characteristics”.⁠⁵

Two features of this finding shape how the seminar thinks about design. First, many people voluntarily share how they voted. Secrecy protects the option not to share, not a habit of silence. That is relevant to the My Vote Forward distinction between a secret vote and a voluntarily shared opinion. Second, belief matters. A system that is formally secret but hard to understand may provide less protection than one that is simple and visibly secret. That challenge is acute for any electronic or remote system, whose secrecy guarantees ordinary voters cannot inspect.

Remote voting and the limits of secrecy

The polling booth gives physical privacy. Remote voting removes it. Mail voting, which the site examines in the Oregon vote by mail case, trades the booth for convenience and relies on law and norms to protect voters at home. The site’s chapter on intimidation and the objection page can a household coerce a remote voter? take up the problem directly: no technology can fully prevent someone in the same room from watching. Week 11 reads the cryptographic literature on coercion resistance, which offers partial answers with real costs. The strongest argument for remote voting is access, especially for people with disabilities or who live far from polling places. The strongest argument against it is that it weakens the protection the secret ballot was invented to provide. Both are serious.

Exclusion through neutral rules

Secrecy addresses one way elections can fail. Exclusion addresses another. Alexander Keyssar’s The Right to Vote is described by its publisher as “a magisterial account of the evolution of suffrage from the American Revolution to the end of the twentieth century,” carried forward in the revised edition through the 2008 campaign.⁠³ The site’s chapter on access and exclusion summarizes the history. The National Archives account of the Voting Rights Act states that it was signed on August 6, 1965, and “outlawed the discriminatory voting practices adopted in many southern states after the Civil War, including literacy tests as a prerequisite to voting.” It adds that African Americans in the South had faced “poll taxes, literacy tests, and other bureaucratic restrictions,” and “risked harassment, intimidation, economic reprisals, and physical violence when they tried to register or vote”.⁠⁷

The lesson for designers is that exclusion rarely announces itself. A literacy test reads as a reasonable qualification on paper. Its effect depends on who administers it and how. Any new requirement, including identity checks for a participation channel, has to be evaluated by its effects on who can actually take part. The site’s chapters on ballot access and identity and registration apply that test.

Shelby County v. Holder

The Supreme Court’s 2013 decision in Shelby County v. Holder is the most important recent ruling on the Voting Rights Act, and students should read it with care because it is often summarized inaccurately. The Court’s syllabus explains that Section 2 of the Act, which forbids voting practices that deny or abridge the right to vote on account of race or color, “applies nationwide, is permanent, and is not at issue in this case.” Section 4 provided the coverage formula, defining covered jurisdictions as those that “maintained tests or devices as prerequisites to voting, and had low voter registration or turnout, in the 1960s and early 1970s.” Covered jurisdictions needed federal approval, called preclearance, before changing voting procedures. In 2006 the Act was reauthorized for 25 years, “but the coverage formula was not changed.” The Court held: “Section 4 of the Voting Rights Act is unconstitutional; its formula can no longer be used as a basis for subjecting jurisdictions to preclearance”.⁠⁶

The majority’s argument, as stated in the syllabus, rests on the principle that the Act “imposes current burdens and must be justified by current needs,” and that a departure from the equal sovereignty of states requires a showing that the coverage formula is related to the problem it targets. Justice Ginsburg’s dissent, which students should read in full, pointed to the 2006 legislative record, which it says “filled more than 15,000 pages,” and argued that “throwing out preclearance when it has worked and is continuing to work to stop discriminatory changes is like throwing away your umbrella in a rainstorm because you are not getting wet”.⁠⁶ Both positions deserve to be stated at their strongest. One holds that a formula based on decades-old data cannot justify treating some states differently today. The other holds that preclearance was preventing discrimination, and removing it would invite the return of what it had prevented. The empirical consequences of the decision are debated, and the seminar does not resolve them.

What the history requires of new designs

Week 4 leaves the course with two tests that every later design must pass. The secrecy test asks whether a participant can express a view without anyone being able to learn it or verify it against their will, and whether participants believe that. The access test asks whether the design’s requirements, however neutral on paper, will exclude eligible people in practice. The tests pull against each other. Stronger identity checks can protect against fraud and coercion while raising barriers; easier remote access can widen participation while weakening secrecy. Weeks 11 and 12 return to that tradeoff with technical detail.

Common misreadings this week guards against

Two misreadings deserve a warning. One treats the secret ballot as a guarantee of free choice, when Gerber and coauthors show that perceptions and social pressure still shape behavior. The other treats Shelby County v. Holder as having struck down the Voting Rights Act as a whole; the syllabus is explicit that Section 2 remained in force nationwide and was not at issue, and that the holding concerned the Section 4 coverage formula.

Discussion questions

1. Article 25 ties secrecy to “free expression of the will of the electors.” Does a voluntarily shared opinion need the same protection as a vote? Why or why not?
2. Gerber and coauthors found that over 70 percent of respondents share their vote choices. If most people share, what is secrecy for?
3. Rusk warns against attributing behavior change to attitudes before checking for rule changes. Apply that warning to a recent change in turnout you know about.
4. State the strongest version of the Shelby County majority’s argument and of the dissent’s. Which empirical evidence would you need to decide between them?
5. Can a remote participation channel ever offer the protection of a polling booth? What would come closest?
6. Identify a current rule in your jurisdiction that is neutral on paper. What evidence would show whether it excludes in practice?

Assignment

Secrecy and access review (1,500 words). Choose one remote participation method actually in use: mail voting in a named jurisdiction, an online public comment portal, or an online participatory budgeting vote described in one of the site’s cases. Using official documents, describe how it handles (a) secrecy, (b) coercion at the point of participation, (c) eligibility, and (d) barriers to access. Apply Rusk’s warning and Gerber and coauthors’ finding about perception. End with one specific change that would improve secrecy without reducing access, or explain why no such change exists.

Evidence

The Article 25 text is quoted from the Covenant. The Australian history is quoted from the Australian Electoral Commission. Rusk and Gerber and coauthors are described from their published abstracts. The Voting Rights Act history is quoted from the National Archives. The Shelby County holding and background are quoted from the Court’s syllabus as published by the Legal Information Institute. All were opened on the review date.

Competing interpretations

Historians and political scientists debate how much of the late nineteenth and early twentieth century decline in American turnout reflects ballot and registration reforms and how much reflects other changes; Rusk represents the institutional side of that debate. On Shelby County, legal scholars and advocates disagree sharply about whether the coverage formula was outdated or still justified, and about the decision’s effects. On remote voting, access advocates and security and coercion researchers weigh the same tradeoff differently. The seminar asks students to state each view in terms its holders would accept.

Application to this framework

My Vote Forward separates a voluntarily shared opinion from a private vote that would have to meet the secret ballot standard. This week’s readings explain why the separation is necessary, and why it is not sufficient: a remote private vote would still face coercion at home, and any identity requirement would face the access test. The framework’s intimidation chapter states what it does not solve.

Unresolved questions

• Can a remote system produce justified belief in secrecy among ordinary participants, not only among experts?
• How should a design weigh a measurable gain in access against a harder-to-measure loss in protection from coercion?
• What evidence would show whether a new identity requirement excludes eligible people before it is adopted?