Electoral and Social Order Motivations

In the most basic sense, to explain reform we need to understand under what conditions members of the legislature and executive will support it; that is, we need to understand and explain the motivations of parties and individual legislators. I contend that two main types of motivations guide reforms: (1) strategic calculations regarding potential electoral and political benefits and (2) beliefs on appropriate gender hierarchies and roles. I refer to these as electoral calculations and social order or normative motivations, respectively.

Calculations concerning electoral benefits are estimations from political actors of how reform would affect their electoral support. Incumbent parties make calculations regarding the merits of reform based on whether and how the inclusion of a new group of voters would change the existing electoral balance. They might also be concerned with how their existing constituencies stand on the issue of suffrage and respond to these interests (McConnaughy Reference McConnaughy2015). At times a government, particularly an authoritarian one, may need legitimacy rather than votes; certain reforms, such as the adoption of democratic institutions, may aid in obtaining that legitimacy from both international observers and the domestic population.

The second set of motivations is ideational. Simply put, individuals hold beliefs about what is just or unjust. These principles, emerging from deeply held normative beliefs, have what Nina Tannenwald (Reference Tannenwald1999) calls constitutive effects: they define what is acceptable. In the case of women’s suffrage, I focus on a particular set of beliefs: those related to desirable and appropriate gender roles. Some legislators were strong feminist advocates, believing that extending suffrage to women was a matter of justice. Others opposed the franchise, anticipating that gender relations would be altered as women expanded their roles in the public sphere. As the literature on male backlash has highlighted, for men who hold traditional beliefs on gender roles, changes in relative resources—such as education and income— within the household can threaten their masculine identity (see, e.g., Atkinson, Greenstein, and Lang Reference Atkinson, Greenstein and Lang2005). Extending this line of research, I argue that the franchise can also be understood as a resource that threatens men’s status both individually within the household and collectively as a social group. Overall, legislators were moved either to promote change or preserve social and not only political power relations.

Arguments about strategic electoral calculations dominate the literature on suffrage extensions and electoral reform more generally. Teele (Reference Teele2018), McConnaughy (Reference McConnaughy2015), and Przeworski (Reference Przeworski1999) emphasize the strategic considerations of incumbents and the conditions under which they will support suffrage extension; these conditions include electoral need, the strategies and alliances of women’s movements, and the ideological orientation of incumbents. Political economists studying the inclusion of lower and middle classes focus on the incumbent’s strategic assessment of whether the potential redistributive effects of extending the franchise outweigh its costs (Ansell and Samuels Reference Ansell and Samuels2014; Boix Reference Boix2003; Ziblatt Reference Ziblatt2008). The literature on gender quotas likewise stresses strategic factors such as international incentives or when parties face internal and external competition (Bush Reference Bush2011; Weeks Reference Weeks2018).

Other types of motivations play a minor role in the literature on electoral reform. Banaszak (Reference Banaszak1996) focuses on beliefs and values—but those of suffragists, not decision makers. In the literature on political regimes and democratization, ideology is usually a residual variable, and overall, there is little theorization of the role played by ideational variables and if and how they interact with strategic calculations. Important exceptions are Bateman (Reference Bateman2018) and his work on exclusion along racial lines and Amel Ahmed’s (Reference Ahmed2013) work on the choice of electoral system, claiming that right-wing parties were motivated by the existential threat that radical working-class parties posed to the social order and to the institutions of capitalism and liberal democracy.

Generally, the literature on electoral reform reproduces the problem identified by Sánchez-Cuenca (Reference Sánchez-Cuenca2008, 362), in which political scientists adopt “a narrow view of self-interest: agents are moved by selfish, material motivations.… This preference is not grounded on philosophical or empirical grounds, because both philosophers and empirical social scientists have shown the variety of human motivations.” In line with Sánchez-Cuenca, I claim that the focus on self-interest—or electoral benefits in the case under discussion—is based on methodological and not ontological grounds. Reading the historiography on women’s suffrage, the multiplicity of decision makers’ motivations is often present.

Some research programs within comparative politics have made advances to challenge the prevalence of rationalist arguments. For example, in the area of collective action and participation in protests and insurgency, the works of Wendy Pearlman (Reference Pearlman2018) and Elizabeth Wood (Reference Wood2003) highlight the importance of a moral identity and the “pleasure of agency” produced by participation itself, regardless of outcomes. In the literature on electoral reforms, in contrast, similar non-utilitarian perspectives have had a less significant impact.

I hypothesize that alignment of electoral and normative motivations—in a sufficient majority of legislators—is generally necessary for successful reform; both types of motivations are jointly necessary. Under normal circumstances, actors will not pursue a reform that they believe has negative political consequences for them. Nor, under normal circumstances, will actors go against what they believe gender relations should be. I also hypothesize that, if motivations point in opposite directions, decision makers will opt to stall or reject reform. Table 1 summarizes the motivation alignment argument. Electoral calculations can be positive or negative, whereas social order goals are progressive or conservative. A third category is labeled unclear/neutral. Here, some individuals or nonprogrammatic parties may not be committed to normative principles and ideologies on specific issues, or without being advocates of greater gender equality, they believe that suffrage represents only a limited threat to the social order. Likewise, electoral and political implications of reform may be disputed, making strategic motivations neutral or uncertain. I hypothesize that both positive/progressive motivations or one positive motivation combined with one unclear/neutral motivation will lead to reform.

Table 1 Motivation Alignment and Reform Outcome

A logical implication of this argument is that there are multiple ways in which both successful and failed reform can come about, driven by different motivations. For successful reform, political actors first pursue reform guided by a combination of both motivations. Second, reform takes place when progressive leaders can promote, usually in conjunction with women’s movements, a consensus on expanding women’s roles in a context of favorable or uncertain electoral benefits. In other cases, reform is guided by the goal of obtaining electoral benefits, provided that conservative views on women constitute a minority. Similarly, failed reform can result from conservative views on gender relations, regardless of electoral considerations, or, conversely, be the product of negative electoral considerations, which make normative beliefs irrelevant.

This argument assumes that suffrage must have sufficient political salience for most actors to develop clear motivations. This is generally the case because suffrage reform was discussed with attention to similar debates in other countries, as a response to women’s mobilization, or when committed politicians presented bills. When this assumption is not met, however, motivations may not follow theoretical predictions, as I discuss in the case of Ecuador. The next section explores the historical context in which these motivations took form in the first half of the twentieth century in Latin America, emphasizing how the axes of political competition and the forms of women’s mobilization affected decision makers’ motivations.

Historical Cleavages and Women’s Suffrage in Latin America

The motivation alignment argument is the key mechanism that explains suffrage. Here I present a historically grounded explanation for the dominant configuration of motivations among decision makers in the first half of the twentieth century. I contend that most countries in Latin America failed to enfranchise women before World War II because they shared overlapping class and religious divisions, what I call an oligarchic–anti-oligarchic cleavage.Footnote ¹ And it is this common historical pattern that put political calculations and social order beliefs at odds, making early reform rare.

Traditional landed elites who were generally close to the Catholic Church and held traditional views on gender roles comprised the oligarchic side of the cleavage. In the anti-oligarchic camp were middle and working-class parties who were usually anticlerical and favored expanded roles for women. Male elites used these cleavages to interpret the effects of the franchise both on electoral and normative terms.

Both oligarchic and anti-oligarchic actors believed that women, if enfranchised, would vote more conservatively than men. In the late nineteenth and early twentieth centuries, and before the emergence of women’s suffragist organizations, upper-class women often participated in charity work run by the Catholic Church. Women’s organizations were also leaders in efforts to defend the church from secularizing policies, including attempts by the state to undermine the church’s autonomy and to introduce divorce (Ehrick Reference Ehrick2005; Lavrin Reference Lavrin1998). Finally, women exhibited both greater rates of religious affiliation and participation. Thus, women’s initial modes of political organization often worked more to defend the church than to struggle for women’s rights. This mobilization was crucial in signaling future electoral behavior.

Feminist movements were usually weaker vis-a-vis this conservative religious mobilization, so their impact on electoral calculations was unable to counteract the potential support of a majority of women for conservative parties. Suffragist organizations were not able to offer sufficient electoral support to parties, nor did they have the capacity to enter into alliances with other organizations, as has been pointed out in other cases (e.g., McConnaughy Reference McConnaughy2015; Teele Reference Teele2018). Similarly, working-class women’s organizations that could have formed a relevant constituency for leftist parties were few in number and were linked mostly to anarchist and socialist movements that had very little presence in representative institutions in the early decades of the twentieth century (e.g., Lavrin Reference Lavrin1998).

Suffragist organizations, however, did often play a role in the second set of motivations, those related to gender roles. Strong feminist organizations that mounted regular campaigns were sometimes able to generate greater support for the expansion of women’s roles among politicians and shift public opinion (Banaszak and Ondercin Reference Banaszak and Ondercin2016). Whether this normative conviction was sufficient to lead to early reform was conditioned by the size and position of those favorable sectors. Overall, depending on the configuration of coalitions in power and some variations in historical cleavages, these different elements led to electoral or normative factors being more prominent.

Case Selection and Methodological Considerations

This article looks at successful and failed instances of reform in women’s suffrage in the early enfranchisement period in Latin America (before World War II). I focus on this early period for two main reasons. First, the central theoretical contribution of this article is that both views on social roles and electoral concerns mattered. The relative importance of electoral and normative beliefs, however, varied across cases and in time. Early debates provide the best evidence to analyze both actors committed to social change and the defense of existing gendered social hierarchies. With time, women’s suffrage became a standard, and the comparative experience, as well as the expansion of women’s roles, no longer made political participation an important threat to existing gender roles. As such, we can expect early debates to more truly reflect actual preferences. Second, the historical argument also has a temporal scope. After World War II, the configuration of cleavages and coalitions changed, affecting parties’ motivations as I discuss in the conclusion.

To assess the validity of the historical and the motivations argument, I analyze three cases: successful enfranchisement in Uruguay and Ecuador (while considering change in time by looking at the debates in two junctures) and failed reform in Peru.Footnote ² All three cases challenge the common explanation of politicians being guided by electoral calculations. The case of Uruguay exemplifies successful reform driven by progressive sectors committed to social change, which were able to push for an early consensus on women’s political inclusion. Peru, in contrast, shows a failure of reform given conservative normative opposition, despite potential electoral benefits. Finally, Ecuador was chosen for two reasons: its importance as the first enfranchiser in the region worthy of explanation on its own merit and it not clearly fitting the motivation alignment argument. A close analysis of this case brings forth the relevance of the lack of politicization of the question of suffrage, which precedes the debates in the other cases. Table 2 previews the cases by showing the configurations of incumbent parties’ dominant motivations.

Table 2 Incumbent Motivation Alignment and Reform Outcome

Using process tracing, the goal of the case studies is to provide case-grounded explanations; as such, the selection is not meant to be representative of a broader sample. Following Ryan Saylor’s (Reference Saylor2020) approach, I chose each case in relation to the theory sketched here of different theoretical ideal types of motivations guiding politicians. In this framework, every case constitutes a combination of the theorized factors—to varying degrees depending on how much they approach the ideal type—and idiosyncratic elements (1003).

The case studies, using process tracing, analyze the final reform and failed reform attempts, paying close attention to rival explanations. The most prominent rival explanation—or alternative ideal type—is a purely strategic electoral calculation. The central aim of the analysis is to look for observable implications of the argument to assess whether parties and individual legislators responded mostly to electoral or social order motivations and to determine the relative contribution of each motivation to their final decision-making process. Because motivations are not directly observable or actors may be insincere in their justifications, I triangulate multiple sources to find consistency.

The analysis is based on archival work on a multiplicity of sources, including newspapers, legislative debates, party documents, correspondence, roll-call votes, and electoral and census data. Data availability is an important concern with historical work, particularly in Latin America, and the available data vary considerably by country. First, missingness can be purposeful and can be related to the data-generation process (Gonzalez-Ocantos and LaPorte Reference Gonzalez-Ocantos and LaPorte2021). For example, there may be a lack of pronouncements in legislative debates because legislators may not want to speak publicly against women’s suffrage. Second, data are often missing due to the nature of bureaucracies and archiving procedures, as well as institutional practices such as the rare roll-call votes in legislatures. With these shortcomings in mind, I construct explanations that more plausibly account for the observed data.

Weak Cleavages and Norms Entrepreneurs in Uruguay

Uruguay was the only early case in Latin America to enfranchise women in a fully competitive context. It is also a country characterized by a weak oligarchic–anti-oligarchic cleavage in the early twentieth century. I argue that the latter factor explains why, despite having strong competition, which the literature assumes is a prerequisite for strategic decision making, electoral calculations were not particularly salient in this case. Moreover, progressive actors from a section of the Colorado Party, coupled with the suffragist movement, acted as norm entrepreneurs and turned equal suffrage into a democratic standard by the 1930s.

Women’s enfranchisement occurred in the late batllista era (1903–33), so called after President José Batlle y Ordóñez (in office between 1903–7 and 1911–15). The batllistas were the dominant faction of the Colorado Party that occupied the presidency from 1865 until 1959. The second party in this bipartisan system was the National Party. Despite the Colorado Party’s hegemony, competition increased considerably after the 1918 constitution that introduced male democracy. The Colorado and National Parties had similar liberal roots, and their differences were not based on class or ideology but emerged as disputes between independence leaders that became locked in and trickled down to society (Somma Reference Somma2015). Both were, in fact, multiclass parties, and although the National Party eventually became a relatively more conservative party, internal factions that acted as different parties complicated this characterization, which, in any case, was not based on deep sociological differences.

As Carlos Real de Azúa (Reference Real de Azúa1971, 37) puts it, the starting point of the battlista era was “the undeniable weakness that the typical constellation of powers in the continent presented in the Uruguay of the nineteenth century. The social-economic hegemony of the agrocommercial businessmen and their interlacement with the Church and the armed forces as factors of consensus and coercive backing, respectively, did not take on the same consistency that they had in the rest of the Latin American area.” The landed elite was not a consolidated political actor, and the Catholic Church, because of late colonial implantation, was materially poor and had weak links with the state. Because it was weak, the church was not able to counter the secularizing policies launched by the government, including the secularization of public health (1904–6), the divorce law (1907), the end of religious instruction in public schools (1909), the official separation of church and state (1917), and the secularization of religious holidays (1919). These policies made Uruguay the most secularized country in the region.

Although Catholics organized their own party to oppose secularization, the Catholic Union of Uruguay and later the Civic Union never were able to elect more than a couple of representatives; they acted within a very limited political space in a context of growing secularization and two strong party identities. As historians Barrán and Nahum (Reference Barrán and Nahum1986, 182) point out, the fact that Catholic sectors felt the need to organize a party meant that they felt unrepresented by the two main parties.

In this setting, a first test for women’s suffrage came in the 1916–17 constitutional convention. The resulting constitution is usually seen as the foundation of Uruguayan democracy because it consecrated male universal suffrage, the secret ballot, and proportional representation. In the 1916 elections leading to the convention, held under universal male suffrage for the first time, the government suffered its first important and unexpected electoral defeat when the conservative faction of the Colorado Party ran in a separate ticket, largely because of Batlle’s proposal to install a collegiate executive. Anti-batllistas, nationals, and Catholics formed a majority in the convention.

The two socialist representatives in the convention, Emilio Frugoni and Celestino Mibelli, proposed and strongly argued for enfranchising women.Footnote ³ Their proposal, however, did not have enough support (there was no nominal vote). My claim is that normative beliefs on women’s political participation were mostly unfavorable in this early period, particularly among nationals and anti-batllista Colorados. In the convention, two arguments against enfranchisement were prominent. First, there were claims that women’s natural place was the home and the family. Second, several representatives argued that there was no demand for suffrage, that women were not interested, and that the issue was too “advanced,”Footnote ⁴ an argument that shifted the blame to women. Both anti-batllista Colorados and Nationals made these arguments. Unlike most countries in the region, electoral arguments attributing to women a preference for a particular party were not salient.

Two years earlier, when the first bill on women’s suffrage was introduced, the official newspaper of the Catholic Church openly opposed women’s suffrage, specifically the right to be elected: “The destiny that women must fulfill, according to God’s plan and nature, is contradictory, in our view, with the exercise of the broad political function” (El Bien, July 15, 1914). A similar view was expressed in the National Party-affiliated newspaper, claiming that laws had not put men in a superior position but that such hierarchy happened naturally in all societies: “nature has imposed on her a mission too absorbing, a mission too transcendental, that fills most of her life … maternity” (La Democracia, July 11, 1914).

If views on gender roles were mostly unfavorable, strategic considerations on the effects of including women were not salient, given the small social and ideological differences between the two main parties. Additionally, given the changes to the electoral system under discussion in the constitution, the parties were probably wary of introducing any additional sources of uncertainty. And how did women’s organizing contribute to electoral calculations? The most organized group of women at the time were Catholic women, particularly the Catholic Ladies’ League of Uruguay formed in 1906 in opposition to the divorce law. This organization had links to the Civic Union Party, but because this was a small party it did not affect the calculations of the Colorado and National Parties (Barrán and Nahum Reference Barrán and Nahum1986, 183). An important feminist organization, the National Council of Women, was only founded in 1916, and although its members attended the debates in the constitutional convention, the group had yet to conduct its most important organizational and political work.

After the constitution ratified the new electoral rules, the electorate was largely expanded and competition tightened. Figure 1 shows the percentage of votes obtained by the Colorado and National Parties in the lower chamber before women’s suffrage passed in 1932. The graph on the left considers the two main parties as a whole; on the right I disaggregate the data to include the most relevant Colorado factions. The National Party had a very stable vote share, between 44% and 49%. Within the Colorado Party, the batllistas also had relatively stable support, although with greater fluctuation as new factions emerged (mainly the classical liberal riveristas and the vieristas). Overall, there is no evidence that a change in electoral balances triggered suffrage reform. Both parties could have benefited from attracting new voters.

Figure 1 Electoral Competition in Uruguay, 1916–1931

Source: Data from Nahum (Reference Nahum2007, 68–77).

In this period, there were no clear potential winners of enfranchisement. That women as voters could favor Catholic candidates was mentioned in Uruguay as in other countries, but it was not argued as benefiting any particular party.Footnote ⁵ Moreover, a 1921 article in the batllista newspaper stated, “The female electorate is not organized, and for that to happen it will take several years. As such, it will receive more immediate advantages the party that is in better conditions to set it in motion” (El Día, July 21, 1921). According to the author, the Colorado was the party with the advantage, because of its long-term incumbency that had given it the loyalty of the bureaucracy, the army, and pensioners. Feminist Paulina Luisi, in contrast, believed that “in both traditional parties, there is an indecision regarding the orientation that the female electorate might take” (Imparcial, November 8, 1929); this had delayed their support.

As for women’s organizations having an effect on electoral calculations, the diversity of the movement meant that there were no clear winners, because multiple party factions had links with women’s organizations. The National Council of Women (NCW) and the Uruguayan Women’s National Alliance (UWNA), the main liberal feminist organization, originally comprised wealthy women associated with charitable work and the first wave of women professionals (Ehrick Reference Ehrick2005, chap. 4). In terms of partisan affiliations, although the founder of both organizations, Paulina Luisi, was close to the Socialist Party, other prominent members of the organization were batllistas (e.g., Isabel Pinto) and from the anti-batllista faction of the Colorado Party (e.g., Fanny Carrió and Sofía Álvarez Vignoli). Catholic women, in contrast, were well organized in the Catholic Ladies’ League, which had ties to both the Civic Union and later to the conservative wing of the National Party (162). This diversity meant that mobilized women were not a well-defined political constituency and that all parties could potentially attract part of the female electorate.

If electoral incentives for parties were relatively stable between 1916 and 1932, what did change was the growing acceptance of suffrage as suitable for women. This shift came in tandem with women’s greater access to secondary and higher education. After reforms in the 1910s, women went from representing 34% of secondary students in 1912 to 51% in 1937 (Nahum Reference Nahum2007, 121). There was also a growing number of professional women. In this context, I claim that two related factors were important for the increased popularity of women’s suffrage: activism by women’s organizations and the role of progressive politicians.

Liberal feminist organizations were most active during this period, particularly between 1917 and 1923 and then from 1927 to 1932. The NCW and the UWNA, created in 1916 and 1919, respectively, included among their activities the publication of the feminist magazine Acción Femenina, advocating for suffrage and other women’s rights in the mainstream press, petitioning Congress to discuss the bills presented, participating in international fora, organizing public talks, and lobbying the government for women’s access to public posts. Several leading feminists were also married to politicians, thereby exerting a direct influence (Ehrick Reference Ehrick2005, 157). Their largest campaign for suffrage came in 1929 when both organizations—after overcoming previous disputes—planned a large conference in favor of women’s political rights and later published a book with the speeches made mostly by professional women. In 1931, women once again petitioned Congress to discuss the political rights bill, backed up by more than four thousand signatures (146).

Some prominent members of the political elite also contributed to keeping suffrage on the agenda, acting as norm entrepreneurs. The first bill to extend the franchise was presented in 1914 by two Colorado congressmen, as were those presented in 1920 and 1924 (El Día, July 13, 1924). Batlle himself was a defender of women’s rights. In 1912, Batlle started publishing editorials in El Día under the pseudonym “Laura.” In one of these early pieces, he argued, “Are they not, like men, conscious beings, with rights to exercise, duties to fulfill, interests to guard? Are they not as interested as men in the good progress of the political community they belong to? And are they not capable of casting a vote as illustrated as four-fifths, at least, of the men who vote?” (El Día, March 14, 1912). President Baltasar Brum (1919–23) was also a committed feminist who in 1921 commended the study and drafting of a bill for broad civil and political rights, eliminating inequalities between men and women, and refuting arguments that women would vote for Catholic candidates (Lavrin Reference Lavrin1998, 331–32). Male politicians supporting women’s emancipation were present in other Latin American countries: the Uruguayan particularity is that these progressive voices reached the highest positions of power, having an unmatched influence.

As a result of women’s activism and progressive politicians’ support, women’s rights were regularly part of the legislative agenda and of the public debate, helping prepare the ground by refuting arguments against voting rights. The compatibility of voting with women’s role as homemakers was often addressed. “Until now, arguments presented to combat women’s votes lack solidity and one of them, which is brought up at every step, is that the female vote as incompatible with the home. This is an exaggeration” (Giménez Reference Giménez1919, 145). Women’s organizations also succeeded in achieving further inclusion of women in public administration. In 1926, a law allowed women to serve as notaries, and in 1929, after years of lobbying, Sara Rey Álvarez became the first woman named to the Council of Delinquents and Minors (Ehrick Reference Ehrick2005, 138).

Support slowly increased for suffrage. After a lengthy discussion of a new party platform, the 1925 party manifesto of the Colorados included women’s civil and political rights as key issues, moving from the support by a few individuals to an institutional position. In 1930, the National Party followed suit. The support of both parties was necessary because there was a two-thirds majority requirement.Footnote ⁶ They also both presented bills in the new decade, the batllistas in 1930 and 1932, and the Nationals in 1931. Catholic women also openly supported suffrage, after some reluctance during the 1917 debate. The League had started recruiting university students, defending the pursuit of higher education if it had solid “moral grounds” (Ehrick Reference Ehrick2005, 167); this indicated that women’s new roles as professionals and potential voters were gaining acceptance across social and political sectors. In the press, during the final suffragist campaign in 1929, little resistance was expressed. The batllista-affiliated newspaper El Ideal (November 8, 1929) editorialized about women’s suffrage as a “moral imperative,” even if women’s incorporation translated into a “political upheaval.”

When a suffrage bill was debated in 1932, it received almost unanimous support. The few interventions centered on claiming long-time support for women’s suffrage or blaming the opponent for the delay in incorporating women into the electorate. But even if women’s enfranchisement became a normative standard for a majority, its potential to upset gender roles remained a concern for some. It was often brought up in the press and even in the final legislative debate. Although only 2 of 86 legislators rejected the proposal in the lower chamber, a few others said they only supported the bill following party discipline, voting for it while citing concerns that women might abandon their domestic duties.Footnote ⁷ Women’s political rights had gained enough support to expand the contours of acceptable gender roles at the same time that social hierarchies remained largely in place—for example, through the low representation of women—reflecting how, through adaptation, gender hierarchies remain resilient while some new rights are incorporated.

Contradictory Motivations in Peru’s Failed Reform

Women’s enfranchisement in Peru was debated in the context of a constitutional congress elected in 1931. Following a suffrage campaign by a small but influential group of women led by writer Zoila Cáceres—the daughter of a former president—and amidst a reformist environment, the expansion of the franchise was included in the constitution draft presented to the congress. Full and restricted rights were analyzed as alternatives before settling on the adoption of the municipal vote.Footnote ⁸ I argue that the Unión Revolucionaria (Revolutionary Union; UR) government party, which had a majority, was divided in its support for reform because of contradictory electoral and social order motivations among its members. Although led by a non-elite populist figure, this party was associated with the oligarchy and had reason to believe that women would support them to a greater degree than the opposition. However, enough of its members voted against the expansion of suffrage for the provision to fail.

After the 11-year rule of Augusto Leguía (1919–30), a military coup and a transitional government translated into a brief political opening that led to presidential and constitutional convention elections in October 1931. José Miguel Sánchez Cerro, who conducted the military coup that ousted Leguía, would go on to win the presidential election with 50.7% of the vote. In the constitutional congress, three main forces were represented: Sánchez Cerro’s UR with 43% of the seats, the Alianza Popular Revolucionaria Americana (APRA) with 21%, and the Partido Descentralista (Decentralist Party) also with 21%.

Table 3 shows how parties voted in the constitutional convention. APRA was the main actor in the anti-oligarchic side during the debate to enfranchise women. It emerged as a noncommunist revolutionary party that succeeded in capturing labor votes (Drinot Reference Drinot2012), particularly from the north of the country. APRA exhibited moderate anticlericalism during its early years; its program only proposed the separation of church and state. Still, the party made efforts to refute its depiction as an iconoclast party (APRA, September 1, 1931, 10).

Table 3 Party Votes for Broad Women’s Suffrage in Peru, Constitutional Congress 1931

Sources: República del Perú (1931); Jurado Nacional de Elecciones (Reference de Elecciones2016).

Theoretical predictions from my argument state that, as a progressive actor, APRA should have been favorable to transforming gender relations to some degree while at the same time opposing women’s suffrage on electoral grounds, which I argue is the case. Universal suffrage was part of the party’s program, a position that was frequently restated by the party leader and other members, both in party publications and in private correspondence—which hints that it was not only a show for the broader public (Partido Aprista Peruano Reference Peruano1931).Footnote ⁹ The inclusion of equal political rights was advocated by the women’s section of the party, led by Magda Portal, who was a founding member and part of the national executive committee. Moreover, APRA was the first Peruvian party—and among the pioneers in Latin America—to have women in important positions within the party structure and to have a women’s section.

Despite this support for women’s political participation, APRA’s defense of the franchise became qualified as the debate grew concrete. The central argument was that, although attaining equal rights for men and women was the goal, the conditions were not yet adequate because women were unduly influenced by religious beliefs: “the secret vote in the current moment, exercised by the woman not yet off the homely prejudices and the tutorship of the priest, would go to increase, we repeat, the banks of the reactionary conservatism” (APRA, April 2, 1931, 10). In consequence, APRA representatives voted squarely against broad women’s suffrage. Instead, they proposed a restricted franchise for economically independent women, which did not receive sufficient support. After the franchise expansion was rejected, the official party magazine claimed, “The civilismo [the old elite backing Sánchez Cerro] would have never dared to grant the unrestricted vote to women had it not been convinced of the conservatism it entails” (APRA, January 7, 1932, 14; emphasis in original). Thus, APRA’s electoral calculations led them to reject women’s suffrage, despite holding favorable normative views.

The second important actors in the assembly were the decentralists, based in the southern region of Arequipa. The party was organized by a provincial middle class in opposition to centralized political elites (Klarén Reference Klarén2004, 330; Vergara Reference Vergara2015, 126–27). This movement associated religion with tradition against the modernizing forces in Lima (Klaiber Reference Klaiber1992, 167). Based on this religious cleavage, the argument predicts such a party would hold conservative views on gender roles and as such, oppose women’s suffrage. Decentralist Manuel Bustamante de la Fuente reflected this view, claiming, “The field of action of women is in the home. The function that nature has commended her is a strictly conservative function: conservation of the species, of the home, of the family and its traditions” (República del Perú 1931, 568). He went on to argue that if women voted differently from their husbands, there would be important consequences: “fights, tears, abandonment of small children, separation, divorce, etc. And that is why I am for the unscathed preservation of the rights of the home and the family” (569). Overall, only four decentralist representatives voted in favor of the unrestricted vote, and a strong majority voted against it.

Decentralists’ electoral calculations are less straightforward. Given their strongly Catholic character at a time when religious lay organizations were starting to gain salience with women as protagonists (Ara Goñi Reference Goñi and Angel2019), they could have expected to benefit from women’s votes. However, in their speeches and their votes, it seems the decentralists had no such expectations. One possible explanation is that they did not think women as voters would have a strong enough impact to outweigh their concerns about transforming gender roles. First, there was no significant women’s mobilization in the provinces. Second, the decentralists had their best electoral performance in the departments of Apurimac, Áncash, Puno, and Cusco, which had very low literacy rates (a requirement to vote) in general and among women in particular. In these departments women represented only 22%, 35%, 20%, and 31% percent of potential voters, respectively, indicating the low potential electoral impact of their incorporation (República del Perú 1944, 264–65). In contrast, in Arequipa and Lima, which were UR strongholds, women represented 43% and 45%, respectively, of potential voters. Third, except for Apurimac where they were the most popular party, in the other departments where they had a good electoral performance, the decentralists came in second to UR. They might have estimated that women’s votes would favor UR, be similar to men’s, or again, even if benefiting them, would not affect the electoral results because of their small number. Thus, if electoral calculations were mostly neutral (estimating no impact), other considerations must have dominated their decision-making process, considerations that I argue had to do with maintaining gender hierarchies.

UR, the government party, was created just months before the elections as a personalist vehicle for Sánchez Cerro (Molinari Morales Reference Tirso2006). It encompassed a diverse coalition that included different factions of the traditional elites and a popular base of support due to Sánchez Cerro’s humble social origin and mestizo character. UR represented the different opponents of the leguismo, and during the campaign, it also became an anti-aprista coalition (Klarén Reference Klarén2004, 334–35). Because what brought them together was an external adversary, there was no ideological consistency within representatives of the party; this heterogeneity was reflected in their vote. Members of UR split their votes for the majority proposal of unrestricted women’s suffrage, as indicated in table 3. Thirty-one members supported the proposal, 13 voted against it, and 12 did not vote or were absent. The proposal needed 65 favorable votes, so had a larger majority of UR members voted in favor, the bill would have passed.

Sanchecerrismo did not show interest in women’s suffrage before the debate in the assembly. When asked directly about the issue, some figures such as Luis Eguiguren, the former mayor of Lima who joined Sánchez Cerro’s coalition, expressed only lukewarm support.Footnote ¹⁰ However, women actively participated in Sánchez Cerro’s campaign, speaking at rallies and joining support clubs, as well as forming women-only committees. Yet these women’s organizations’ agendas did not include suffrage or other women’s rights (El Comercio, December 4, 1931).

During the debate, only six UR representatives presented their views, all in support of women’s rights. Most, however, held traditional views on gender relations, showing that for many, accepting women’s suffrage required only a mild expansion of acceptable gender roles: “I agree that, in fact, the intellect of the woman is not developed in the same way as that of men”; “It may be that technical or markedly scientific considerations of the problem make it inconvenient for women to exercise government.”Footnote ¹¹ For this group of legislators, enfranchising women seems to have been based on an electoral calculation, which became particularly relevant as APRA emerged as a threatening force. In line with this electoralist decision making, several representatives mentioned the fact that women were already participating in politics, in reference to the ladies’ committees mentioned earlier. Although suffragist mobilization was weak, women had mobilized most notably in support of Sánchez Cerro, particularly in Lima, so it seemed a logical conclusion that enfranchising them would benefit UR. Sánchez Cerro himself acknowledged the importance of women mobilizing during the presidential campaign: “I am a convinced and enthusiastic supporter of female suffrage.… During the electoral campaign, she has been one of the more efficient factors of the triumph of Revolutionary Union” (El Comercio, December 30, 1931). José de la Riva-Agüero y Osma, an important intellectual and conservative politician, argued that women’s suffrage should be achieved by conservative sectors and that women’s votes would translate into the triumph of ideas of order and “traditional sentiments,” likewise hinting at electoral considerations.Footnote ¹²

The three parties were running for the first time. Because this was a new scenario with emerging political actors and new electoral rules, we cannot evaluate how competition and electoral volatility affected politicians’ calculations beyond this one election. However, even though Sánchez Cerro received an absolute majority in a four-way race, it is fair to say that electoral uncertainty was high. APRA’s leader Víctor Raúl Haya de la Torre competed for the presidency and obtained an important 35% of the votes, signaling the emergence of a strong anti-oligarchic actor. For many sectors of Peruvian society, APRA’s level of organization and popular support represented a threat. Sánchez Cerro eventually opted for a repressive strategy in early 1932; by mid-year, sectors of APRA attempted a revolution in the northern city of Trujillo, and as a consequence, the party was outlawed for more than a decade. Before the electoral competition route was discarded, there were strong incentives for the oligarchic coalition to increase their electoral support as a tool in the anti-aprismo fight when women’s suffrage was debated in late 1931.

There is no record of the arguments of those UR representatives who voted against the unrestricted franchise or did not vote, so it is worth looking in some detail at the territorial distribution of UR’s opposition to the franchise. The party dominated in the center and the south of the country. Whereas in the former region, 65% of its legislators voted in favor of women’s enfranchisement, only 54% did so in the south of the country, pointing to the relevance of the regional cleavage. Lima long represented modernizing and liberal tendencies, with the small anticlerical currents within universities and intellectual circles concentrated there. And it was in Lima where UR was founded. Of the 13 UR representatives who were founding members and close to Sánchez Cerro, none voted against suffrage (10 were positive votes, 2 did not vote, and one legislator was sick; Molinari Morales Reference Tirso2006, 31). Similarly, of eight representatives identified by contemporary historian Jorge Basadre (Reference Basadre2014, 164) as defending the government position, seven voted in favor of women’s suffrage. We can infer, then, that those representatives who were part of UR’s core were more invested in defending the party’s electoral success by incorporating new voters.

In the rest of the country and in the south in particular, many legislators who ran on the UR list had a looser relationship with the party. And in these regions—the ones where the Decentralist Party was most successful—Catholicism remained a highly conservative and traditional force (Klaiber Reference Klaiber1992, 209). Of the 13 nay votes, 8 were from representatives of the departments of Cusco and Junín, two regions where traditional values dominated, especially in comparison to Lima. Moreover, some figures were strongly linked to religious organizations. For example, Felix Cosío, one of the UR representatives who voted against suffrage, was initially nominated as a candidate by a committee organized by Cusco’s Catholic Action (217). Another representative from Cusco who rejected the proposal, Luis Velasco Aragón, is identified by Basadre (Reference Basadre2014, 164) as an independent, indicating that he might have left the party after getting elected on its list or that he had a more distant relationship with the party initially. A final piece of information is that all the negative votes came from departments where UR legislators split their vote; thus, representatives who were facing very similar scenarios in terms of competition voted differently. These clues point to the importance of individual-level factors: representatives who had a loose relationship with the party were less invested in its mid-term electoral success, prioritizing their own views on women’s suffrage.

The reservations on women’s participation only receded in time, as more women entered universities and expanded their public roles while women’s suffrage became the norm in the region. In 1955, Peru became the second-to-last Latin American country to enfranchise women, under the initiative of authoritarian leader Manuel Odría.

Low Politicization and Early Suffrage in Ecuador

In 1924, Matilde Hidalgo—the first woman to obtain a doctorate in medicine in Ecuador—registered to vote in the city of Machala. She argued that the constitution’s generic ciudadanos included both men and women. The president of the municipal council requested the opinion of the State Council, and the latter gave a favorable pronouncement.Footnote ¹³ Hidalgo was able to vote in the legislative elections that year. Reports indicate that other women also voted in this election (Prieto and Goetschel Reference Prieto, Goetschel and Prieto2008, 319). A few years later, the 1929 constitution made explicit mention of women’s voting rights.

The agency of this one woman is key to understanding why Ecuador was the first country in Latin America to enfranchise women. The fact that she encountered little resistance, I argue, is related to the lack of politicization of the issue. According to the cleavage argument, the incumbent Liberal Party should have raised concerns about including women based on strategic calculations and how women’s suffrage might benefit the opposing Conservative Party. But because of the low politicization of suffrage, those calculations were not salient until a decade after women’s right to vote had been ratified in the constitution.

Ecuador had a liberal–conservative political division rooted in a religious and (to a lesser extent) class cleavage. Conservatives had their economic base of power in the large landholdings of the highlands. They were also strong defenders of the Catholic Church. During the government of García Moreno (1860–75), which strengthened the role of the church in public affairs, a liberal anticlerical sector took form, propelled by the development of a commercial and financial sector as a result of the cocoa export boom (Ayala Mora Reference Ayala Mora1996). Liberals reached power in 1895 in a revolution led by Eloy Alfaro, initiating a period of liberal hegemony (1895–1930) during which, through fraud and repression, the Conservative Party had almost no representation in the executive and only limited participation in the legislature. True party organizations only emerged in 1925; before then the Liberal Party was more a loose collection of caudillos and followers that, according to the leader, swung the party between more radical and moderate positions. Likewise, only then did the Conservative Party mount a territorial organization and develop clearer ideological principles (Hurtado Reference Hurtado2010, 165–72).

What differentiated liberals from their conservative opponents was a religious division. With the liberal revolution’s ascent to power, multiple secularizing measures were adopted related to civil marriage, cemeteries, divorce, and lay education. Additionally, the church was stripped of much of its land, and in 1906 the separation of church and state was formalized. Traditional landowners of the highlands closed ranks behind the church, and the relationship between both sectors was strengthened and remained in place for decades to come (Ayala Mora Reference Ayala Mora1996).

Based on this configuration of social and political forces, the fact that the liberal-dominated State Council and the 1928–29 constitutional assembly ratified women’s enfranchisement is somewhat surprising. I contend that this result was due to the low politicization of the issue. Hidalgo’s actions were not preceded by any relevant discussion on women’s suffrage. In the literature that has analyzed these debates and in my review of the press, there is no mention of organizations partly or completely devoted to promoting women’s political participation (Estrada Reference Estrada2015; Prieto and Goetschel Reference Prieto, Goetschel and Prieto2008; Rodas Morales Reference Rodas Morales and Morales2009). Debates on suffrage among political parties were also practically nonexistent, and the Catholic Church likewise failed to mention any discussion in its regular ecclesiastical bulletins. Because the prevailing social order was very traditional, this is not a story about early progressive actors. During the first third of the century, women started occupying public posts, but it was mostly as educators, a profession that was seen as an extension of maternal roles (Goetschel Reference Goetschel2007). Women’s presence in liberal professions was very limited; Hidalgo became the first female doctor in 1921, whereas Paulina Luisi achieved the same title in 1908 in Uruguay, and Eloisa Díaz in 1887 in Chile.

The lack of politicization of the issue was due in part to the early “solution” of introducing a generic term (ciudadanos) regarding qualifications for suffrage. Both the 1897 and 1906 constitutions included this term, and even though there was no important debate at the time, the implications were clear. For example, in June 1897, liberal deputy Gonzalo Córdova argued they had gone “to the extreme of having granted her citizenship rights and therefore leaving her in aptitude to occupy any public office.”Footnote ¹⁴ And even if there were diverging legal interpretations about the inclusiveness of the term (Prieto and Goetschel Reference Prieto, Goetschel and Prieto2008, 305), the possibility of voting was not pursued until Matilde Hidalgo raised the issue almost three decades later.

Hidalgo’s actions generated a reaction in the press that denoted bewilderment. There was a consensus in terms of the legality of women’s registration and exercise of the franchise, but voices were raised as to the “convenience” of its implementation. For example, an editorial stated, “In this point, as in many others, our legislators have sinned by excess; they have given us laws that assume … a social state more cultured and advanced” (El Comercio, June 11, 1924). Another article made a similar argument that, despite not being organized, laws are very “generous” with women. Some contrary arguments stated that women’s participation would mean abandoning the home (El Comercio, June 12, 1924).

In the 1928 constitutional assembly, again there was no important debate on women’s suffrage: it was assumed it was already an existing right and that making it explicit would not signify a fundamental change. This time around there was some debate in the press. One journalist argued,

In another piece, the same journalist mentioned an anecdote in which a leftist liberal politician had claimed that women’s suffrage was dangerous in Ecuador, given women’s submissiveness to the church, and that their participation should not be encouraged (El Día, December 7, 1928). The fact that liberals quickly dismissed contrary electoral arguments—in addition to the added difficulty of rolling back an existing right already exercised by some women—indicates the low politicization of the issue.

The low politicization might have led liberals to dismiss the impact of women’s suffrage for two reasons. First, women were already citizens according to many interpretations; making an explicit mention was not a fundamental change, and it would not lead women to register in mass. These estimations were not wrong: in the year after explicit enfranchisement (1930), 9.5% of those registered were women, which increased only to 12% in 1933 (Quintero Reference Quintero1980, 245). Second, the assessment relied on the lack of a suffragist movement, which limited the threat to existing gender roles felt by both liberals and conservatives. This is in line with Teele’s (Reference Teele2018, 38–40) argument that women’s movements provide information to parties as to their potential behavior as voters and that without movements parties have no incentives for enfranchisement.

The Ecuadorian case indicates that the lack of a movement might actually promote women’s suffrage by decreasing opposition. In this sense, Ecuador resembles other cases of early democratization and incorporation where time plays an important role. For example, Ruth Collier (Reference Collier1999, 58–59) argues that democratization in Switzerland was largely preemptive because the working class was barely organized in 1848. In these cases, reform follows somewhat different processes than some later cases.

Without strong electoral calculations, reform was guided by equality arguments. Unlike in Uruguay, there were no important proponents of progressive feminist views and discussions of suffrage as relevant for upending gender roles, but a more general belief that women should have equal rights was often presented as a principle of the liberal revolution. Alfaro granted women access to public employment, divorce by mutual consent was established in 1910, and the economic emancipation of married women was introduced in 1911 (Rodas Morales Reference Rodas Morales and Morales2009). Without producing a deep questioning of the sexual division of labor, women came to be seen as citizens by the liberal intelligentsia and the state (Goetschel Reference Goetschel2007). The 1923 Liberal Party Program and Statutes specified, “The Liberal Party will impulse the cultural development of women to elevate her to equal conditions as men, in her legal status and the unfolding of her political, economic, and social activities” (Partido Liberal Ecuatoriano Reference Ecuatoriano1923, 5). During the constitutional debate, writer César Emilio Arroyo defended the liberal principle of equality as a guide for the new constitution, stating that women should have equal rights (El Día, October 9, 1928).

The fact that after enfranchisement the issue did become politicized along the liberal–conservative division when a reversal of reform was debated lends additional support to the politicization argument. Women’s suffrage reemerged with greater strength when the dictatorship of Federico Páez (1935–37) decreed that the 1929 constitution, which explicitly acknowledged women’s voting rights, was no longer in effect and that a new electoral law was required, prompting the debate as to whether women would be included in it. In this context, an editorial in the liberal newspaper El Día stated, “We already know that women, in their immense majority, followed the routes imposed by their sentimental and mystical determinations: as such, they constituted an electorally decisive force in favor of traditionalist politics.” The author goes on to specify “the shackles that today and forever have restricted women’s freedom: that of eternal religious submission” (El Día, March 12, 1937).

These claims stemmed from the end of the liberal hegemony and the first electoral triumph of conservatives (with Neptalí Bonifaz and later José María Velasco Ibarra in 1931 and 1933, respectively), which coincided with women’s exercise of the franchise. Provincial registration data for 1931 can be used to evaluate the claim that women supported conservatives. The Guayas province, where the city of Guayaquil is located, has traditionally been home to the liberal financial and export elites. Only 3.5% of voters in this province were women. In contrast, in the northern highland provinces of Chimborazo, Imbabura, and Pichincha, strongholds of conservatives and Catholicism, women represented 13%, 17%, and 16% of those registered, well beyond the national average of 9.5% (El Día, June 7, 1937). This could indicate that conservatives were more successful at mobilizing women. A 1929 publication of the Conservative Electoral Committee of Riobamba (in the province of Chimborazo) called on women to register to give the homeland “her Christian patriotism,” seeking to restore the Catholic Ecuador of García Moreno.Footnote ¹⁵ It also celebrated both the number of women who were registering to vote and the formation of multiple women’s electoral committees. Registration numbers, however, do not mean that, foreseeing this support, conservatives had promoted suffrage a decade earlier, as Rafael Quintero argues (Reference Quintero1980, 243). Conservatives neither initiated the debate nor had the votes to enfranchise women.

Women’s suffrage ended up not being revoked. The debate in the 1930s shows that it was not the absence of the cleavage that led to early enfranchisement but the low politicization of the issue in the 1920s, which caused parties not to worry about strategic considerations and to favor a normative belief in equal rights. Once women’s suffrage became a reality, those electoral considerations were clearly present, with the claim around women’s religiosity a central element of the debate. Had Matilde Hidalgo not registered to vote in 1924, women’s suffrage would likely not have materialized for years.