Suffrage and voting secrecy in general elections

Adam Przeworski

    Research output: Chapter in Book/Report/Conference proceedingChapter


    INTRODUCTION Voting was almost everywhere public when first national elections took place, early in the nineteenth century. Yet the adoption of secret ballots was steady and inexorable. To my best knowledge, only Bhutan and Iran utilized public voting as of 2000. Why, then, did voting tend to be public early on and became almost universally secret over the past two hundred years? One argument was that secret voting is a second-best option. Independent citizens, endowed with reason and virtue, should bear the responsibility of making their political choices known to fellow citizens. The impediment is independence. If the electorate contains citizens who are not only unequal but embedded in relations of social, economic, or political dependence, publicity of the electoral choice would make those who are dependent vulnerable to sanctions by their masters and thus exposed to intimidation (See Elster, 2013 for arguments in England, France, and the United States in the 1780s, Tau Anzoategui and Martire: 587 on Argentina, Baland and Robinson 2006 on Chile.) Hence, they must be protected by the secrecy of the vote. If this is the reason voting became secret, it should be true that voting would be public when the electorate is homogeneous and it would be secret when it includes workers as well as their employers, peasants as well as landlords, or perhaps women as well as their husbands or fathers. The purpose of this chapter is to examine whether this is a plausible explanation of the historical evolution of voting secrecy rules. HISTORICAL BACKGROUND The raw information concerning numbers of elections that took place under different suffrage criteria and different secrecy rules is provided in Table 4.1 If we group all the cases in which suffrage was restricted by criteria other than gender separately from manhood or universal rules, the relation between the extent of suffrage and the voting rule shows not to be accidental (ρ = 0.52). This grouping, however, is questionable. A closer inspection of Table 4.1 shows that voting tended to be public when suffrage was restricted by property or income criteria but less frequently when restrictions entailed literacy.

    Original languageEnglish (US)
    Title of host publicationSecrecy and Publicity in Votes and Debates
    PublisherCambridge University Press
    Number of pages11
    ISBN (Electronic)9781316015360
    ISBN (Print)9781107083363
    StatePublished - Jan 1 2015

    ASJC Scopus subject areas

    • General Social Sciences


    Dive into the research topics of 'Suffrage and voting secrecy in general elections'. Together they form a unique fingerprint.

    Cite this