TRADITIONAL BELIEFS AND ELECTORAL BEHAVIOR

Abstract

Spanish

Chinese

The party system literature has generally paid little attention to whether traditional beliefs have any impact on voters’ electoral behavior and the stability of party systems. The purpose of the present study shows that the stability of party systems and the pervasiveness of traditional beliefs go hand-in-hand. This article is expected to advance the scholarly understanding of the political consequences of traditional beliefs by showing that voters who hold traditional beliefs or engage in traditional practices are not simply less likely to have pro-democratic attitudes or have a greater appreciation of dictatorial rule, but also more likely to vote for ruling parties in elections. The evidence presented in the article sustains a basic claim; namely, that the pervasiveness of traditional beliefs in Togolese society should be regarded as one of the determinants of the party system's stability.

Keywords

Electoral Behavior Africa Traditional Beliefs Party System Stability Party System Change Togo Benin Tanzania Voting Behavior Witchcraft Voodoo Religion and Politics Development Democratization

The study of party systems has historically been conducted by adopting one of the following two approaches: the genetic approach (Lipset and Rokkan 1967), which claims that a party system is the expression of a cleavage structure, and the functional approach (Sartori 1976), which focuses on the functioning of the party system and its consequences. The study of party system change has, in turn, adopted one of these approaches. Some studies have attempted to explain party system change as the result of a transformation of the cleavage structure (Dalton, McAllister, and Wattenberg 2000; Bartolini and Mair 2007) on the basis of which party systems are built. And, in doing so, they have highlighted the fact that the cleavage structure changes as a result of the growing irrelevance of previously salient social divisions (cleavages), the emergence of new cleavages, or both. In this line of scholarship, change in the structural conditions (social divisions) is responsible for the transformation of the party system. Other studies have analyzed when/how a party system change occurs, how the changeability of a party system could be operationalized, and what the implications are of higher/lower levels of party system instability.

A third stream of inquiry has, more ambitiously, suggested that since many, if not all, of the party system types that had been identified by Sartori (1976) no longer exist, most of the existing party systems in advanced industrial societies, at least, could be categorized as cases of moderate pluralism; in order to preserve the ability to identify differences and similarities within the now broader family of moderate pluralism, a completely new approach for the conceptualization and operationalization of party system change had to be adopted (Mair 1997; Bértoa 2022).

Some efforts have also been made to understand whether party system change is, among other things, a function of cultural conditions. Previous studies, from Inglehart (1971) to Pelizzo and Nwokora (2022), have posited that cultural change was indeed responsible for changes in the party system. Inglehart (1971), for instance, argued that party system changes occurred because the values and attitudes of Western voters, that is their culture, had changed as a result of the socio-economic development that Western countries had been able to achieve in the wake of World War II (WWII). Development/modernization was responsible for cultural change, cultural change was responsible for the emergence of a new issue/cleavage, and the transformation of the cleavage structure led—as any scholar working in the tradition of Lipset and Rokkan (1967) would expect—to a transformation of the party system. More recently, Pelizzo and Nwokora (2022, 238) suggested that previous efforts of exploring the relationship between cultural change and party system change were vitiated by three basic problems: they reduced culture to attitudes and values, they viewed changes in attitudes and values as a response to changes in material conditions, and they explored cultural change on the basis of survey questions that may have prevented scholars from properly appreciating the preferences/positions of the respondents.¹ To avoid these possible problems, Pelizzo and Nwokora (2022) focused on cultural manifestations—both at the mass and the elite level—and suggested that the most important cultural change of the 1960s was the emergence of what they called ‘the fluid self,’ that is, “an individual who had complex, conflicting, and inconsistent preferences” (Pelizzo and Nwokora 2022). The rise of the fluid self eventually led to party system change and increasing instability of party systems because political parties, as (rigid) structures, were not able, equipped, prepared, or designed to satisfy the increasingly inconsistent/conflicting demands of their now fluid voters.

Regardless of which of these two approaches comes closer to properly appreciating the relationship between cultural change and party system change, they had something in common: both regarded party system change as a function of cultural change. The implication of both contributions was, quite simply, that in the absence of cultural change, there should not be any culturally-induced change in the party system. There is also a second implication: in more traditional countries—countries where traditional values and beliefs are more prevalent—there should be less cultural change and, ultimately, less party system change.

The purpose of the present study is precisely to explore whether and to what extent the presence or pervasiveness of traditional beliefs affects the stability of party systems. Specifically, we assume—if the claims advanced by Inglehart (1971) and Pelizzo and Nwokora (2022) are correct—that a greater pervasiveness of traditional beliefs in a society/community should be associated with more stability/less instability of the party system. We do so by analyzing the survey data that we have collected in Togo, which allows us to assess the relationship between voters’ traditional beliefs, their electoral behavior, and their desire for change. Before we proceed with this contribution to the special issue, we wish to offer a few words of explanation as to why we decided to collect survey data from Togo. The first reason is that Togo, like Chad, the Democratic Republic of Congo (DRC), Morocco, Niger, Senegal, and Tunisia, is a Francophone country—and the relationship between belief in witchcraft and fluidity of the party system was detected in Francophone countries (see Figure 1). Hence, we needed Francophone respondents, like the Togolese voters, whom we surveyed for this study, to understand whether (and possibly why) the traditional beliefs of an individual lead to stable electoral preferences and, ultimately, to the stability of the party system or whether the correlation between belief in witchcraft and the stability of the party system was an artefact of aggregation. The second reason we decided to focus on Togo is that Togo, like Chad, DRC, Morocco, Niger, Senegal, and Tunisia, has a sizeable Muslim population—and this second commonality made Togo increasingly more comparable to the countries that had been included in the survey administered by the Pew Research Center (Pew 2012; Lugo et al. 2013). The third reason is that in Togo, like in Benin, there is still a sizeable portion of the population who believes in traditional religions such as voodoo (Gilli 1980, 1986, 1987, 2004, 2006). The presence of these traditional beliefs makes Togo the ideal place to explore the traditionalism-electoral behavior nexus.² Finally, Togo is a rather interesting case for a fourth reason. Pelizzo, Turganov, and Kuzenbayev (2023) show in this special issue of World Affairs that the relationship between modernization and culture is non-linear, complex, and often ambiguous. Just as greater wealth or higher socio-economic status may have a transformative effect on the cultural dispositions of society and individuals, greater exposure to/usage of social media may be change-inducing, but its impact needs to be properly qualified. The use of social media may change the attitudes of some portions of a society, but it only does so in some ways, in some respects, and to some extent—a point that will be discussed in greater detail later on. In other words, the use of or exposure to social media, just like socio-economic development, may reshape some values and attitudes but not others—which is why voters can simultaneously be active users of social media and yet remain strongly attached to their traditional beliefs.

Figure 1.

Party System Fluidity and Belief in Witchcraft.

The organization of the rest of this article is as follows. We first show that in Francophone Africa there is a negative/inverse relationship between the percentage of respondents who believe in magic (witchcraft) and the instability of the party system or, conversely, that there is a positive relationship between the percentage of respondents who believe in magic and the stability of the party system. In addition to discussing why these variables go hand-in-hand, we note that to develop a proper understanding of why the party system is more stable in countries where a larger portion of the electorate holds traditional beliefs (such as the belief in witchcraft, the evil eye, and voodoo), we need to test whether voters with a traditional mindset are more change averse. In order to test such a claim, it is necessary to analyze the survey data and responses we collected in Togo. We then present some descriptive statistics computed by analyzing the data we collected with regard to the characteristics of the respondents (age, gender, and religion), the traditional beliefs of respondents, and their political behavior. This is followed by the results of our logistic regressions to show whether and to what extent traditional beliefs, and political interest, have an impact on the voters’ electoral choices. By doing so, the present article provides additional evidence about the political consequences of traditional beliefs. In fact, while other articles included in this special issue show that traditionally minded voters and/or superstitious voters are more inclined to have sympathetic views of Hitler and his regime or tend to prefer autocratic rule over democratic one, this study shows that traditional beliefs also affect voters’ behavior. Specifically, traditionally minded voters are less likely to vote but more likely to cast their ballots for the ruling party—which explains why in countries where a larger population holds traditional beliefs, the party system is more stable. In the final section, as is customary, we draw some tentative conclusions.

Traditional Beliefs and the Stability of Party Systems

In 2012, the Pew Research Center (Pew 2012; Lugo et al. 2013) collected survey data from Muslim communities in 27 countries across several regions—North Africa and the Middle East, Sub-Saharan Africa, Central Asia, South Asia, and Southeast Asia. Respondents were asked to report whether they believed in traditional beliefs, whether they wore amulets and/talismans, whether they used witch doctors, and so on.

Superstition has been defined as

“what subsists and survives. Popular superstitions are often relics of previous higher conceptions no longer understood and therefore degraded and subsisting, so to speak, as something mechanical and unanimated, which continues to exert a certain suggestion, to enlist irrational and instinctive forces of faith for a sort of atavism without the ability to provide an intelligible explanation anymore” (Evola 1988, 127–128).

If this understanding is correct, then the percentage of the population that holds such beliefs provides an indication of the presence and/or pervasiveness of traditional beliefs and practices in a given society. The data collected by the Pew Research Center allows one to do precisely that. More importantly, these data can also be used to assess how some aspects of culture (superstition, traditional beliefs, traditional practices), generally neglected by political scientists, may affect the political life of a country.

In the party system literature in recent years, considerable attention has been paid to the conceptualization and operationalization of party system change. Party system change has been equated to and measured by electoral change, electoral volatility (Pedersen 1979; Mozaffar and Scarritt 2005), the institutionalization of political parties (Kuenzi and Lambleft 2001; Lindberg 2007), access to government, government formulas, and changes in the pattern of competition (Mair 1997; Bértoa and Enyedi 2021). In this final respect, Nwokora and Pelizzo (2015) proposed an index of party system fluidity that captures the changeability of a party system by measuring the frequency of change, the scope of change (how different the most different party systems are in a country's political history), and the variety of change (that is, the number of different party systems/patterns of inter-party competition in a country's political history). This index has been used in a variety of studies to show how the gap between qualitative and quantitative scholars can be bridged (Pelizzo and Nwokora 2016), to map party system instability in Sub-Saharan Africa (Nwokora and Pelizzo 2015), to explore the party system instability-quality of democracy nexus (Nwokora and Pelizzo 2015; Pelizzo and Nwokora 2018), and to investigate the relationship between party system instability and good governance (Pelizzo 2020). Less attention has been paid to the factors responsible for party system change (Pelizzo 2018). With one exception (Pelizzo and Nwokora 2022) no attention has been paid to the cultural determinants of party system change—a topic that has been entirely neglected on the African continent. In other words, no study has explored whether the presence/absence of specific cultural characteristics in Sub-Saharan African countries has a stabilizing/destabilizing effect on the patterns of inter-party competition, that is, on the party systems.

By using the data collected by the Pew Research Center (see Pew 2012; Lugo et al. 2013) on the traditional beliefs of the Muslim population in a variety of countries, we can explore the relationship between cultural characteristics and party system change/stability. Specifically, by computing the index of fluidity for Francophone countries in Sub-Saharan Africa and analyzing how it relates to the pervasiveness of traditional beliefs, such as the belief in sorcery (witchcraft), we can see whether and to what extent party system stability is also affected by cultural factors such as traditional beliefs (see Figure 1).

By correlating the index of fluidity with the percentage of the Muslim population who believe in witchcraft in all the countries in Francophone Africa, we find a strong, negative, and statistically significant relationship (r = −.851, sig. = .007). In other words, in countries in which there is a higher percentage of respondents who believe in traditional beliefs such as witchcraft, there is a much higher level of party system stability (and a much lower level of party system fluidity) than in countries in which there is a lower percentage of respondents who believe in witchcraft. The high correlation between these two variables also means that 72.4 percent of the variance in the level of fluidity in Francophone Africa is explained by the percentage of the population who believe in witchcraft.

While this scatterplot and the correlation analysis show that the pervasiveness of traditional beliefs goes hand-in-hand with the stability of the party system, they do not provide much of an explanation as to why this may be the case. We believe that the correlation between these two variables could be explained in two different, but obviously complementary, ways. The first explanation is that voters who hold traditional beliefs are less likely to appreciate change. Hence, the explanatory model could be depicted as

Traditional beliefs \to aversion to change \to (stable electoral behavior) \to party system stability

The second explanation is that priests or the leaders of religious congregations in traditional settings have traditional beliefs, they may be averse to change (which could, among other things, reduce their importance and compromise their status and standing in society), and they may advocate the importance of preserving the status quo and may persuade their followers not to change their vote. Obviously, for such an explanation to hold, a condition needs to be satisfied: namely, that the voters/citizens consult their priests/religious leaders to discuss political issues—a question that we will address later on.

The data on fluidity and traditional beliefs that we have discussed so far were aggregated at the national level. In order to test these two complementary explanations, we need to switch from the macro to the micro level. This is what we plan to do by analyzing the data that we have collected in Togo.

Descriptive Statistics

From June 4 to June 11, 2022, a questionnaire of ten questions was administered to a sample of 83 Togolese respondents predominantly from Lomè. The questions were divided into three groups. The first group of questions was designed to capture some basic information about the respondents (gender, age, and place of residence). The second group of questions was designed to assess the presence and/or pervasiveness of traditional beliefs and practices in the Togolese population. In this respect, respondents were asked whether they believed in the evil eye and in witchcraft, whether they had ever witnessed exorcisms, whether they made use of traditional religious healers, and whether they knew someone who believed in voodoo.

The third group of questions was designed to assess not only whether, but also to what extent, respondents were interested in politics, whether they talked about it at home or with their religious leaders, and whether they voted in elections and actually wanted change.

From August 10 to August 15, 2022, the questionnaire was administered to a sample of 120 Togolese respondents: 30 from Lomè, 30 from Aneho, 30 from Agbodrafo, and 30 from villages in rural areas such as Adamavo, Alogavi, Anfoin, Kpessi, and Krogan. The collection of these two sets of data gave us a sample of 203 responses. Purposive sampling was employed to identify/select the 203 respondents.

The age of the respondents varies from a minimum of 18 to a maximum of 70, with an average of 28.6—a value sensibly higher than the population's average age reported by some sources such as Worldometers. The sample is almost evenly balanced in terms of gender: 53.2 percent of the respondents are male, while the remaining 46.8 percent are female.

In terms of religion or religiosity, the respondents in our sample provide a fairly accurate picture of the religious affiliations in the country. In a recent volume, Adebanwi and Orock (2021, 99) reported that 71 percent of the Togolese population was Christian, 12.4 percent were Muslim, and the remaining 16.6 percent practiced other religions. The analysis of the survey data that we collected displays fairly similar results: 70 percent of our respondents are Christian, 15 percent are Muslim, 6.25 percent are reportedly atheist, and the remaining respondents follow different/other creeds (Table 1).³

Table 1.

Religious Affiliation in Togo.

	Adebanwi and Orock (2021)	Our sample
	(%)	(%)
Christian	71	73
Muslim	12.4	15
Atheist	–	5.7
Other	16.6	6.3

The data presented in Table 2 sustain the claim that traditional beliefs and practices are fairly common in Togolese society. More than half of the respondents claim to have witnessed an exorcism and report using traditional healers, about three-quarters of the respondents believe in witchcraft and the evil eye, and more than four-fifths of the respondents know someone who believes in a traditional religion such as voodoo (Box 1).

Table 2.

Traditional Beliefs and Practices.

Question	Yes response (%)	N
Have you ever experienced or witnessed the devil or evil spirits being driven out of a person?	56.2	203
Do you or your family ever use traditional healers when someone is sick?	53.5	203
Do you believe in witchcraft or sorcery?	71.4	203
Do you believe that certain people can cast curses or spells that cause bad things to happen to someone?	79.7	203
Do you know anybody who believes in traditional religion (vodun)?	85.7	203

Box 1.

A Voodoo Tale.

“My mother told me that her grandfather died of poisoning. My grandfather was part of the Voodoo cult and according to the Voodoo rules if a person dies in an unnatural way, as in the case of my great-grandfather by poisoning, there are certain rituals to follow before burying the dead. First of all, you have to consult “the Fâ.” That means communicating with the spirits, especially that of the dead person and the ancestors, to ask the dead person how he or she would like to be buried and which items should be placed in the coffin. Depending on the wishes of the deceased that he communicated during the consultation, certain objects will be placed in his coffin, more precisely at the level of his head. For example, a knife, a machete, a wooden stick, etc. My great-grandfather had four wives. When he died, his four wives went to the Voodoo priest to perform the ritual in secret so that their husband could avenge his death. During the ritual, the Voodoo priest gives a time limit within which the dead man will take revenge on his murderers. Once the deadline arrived all the people who were involved in my great-grandfather's death started to see him. They were three men all related to him. The first one to see him was in his field cultivating when, all of a sudden, he started getting beaten with a stick. The people next to him could not see who was hitting him but he was shouting my great grandfather's name and asking for forgiveness while running away from the invisible blows of the sticks. Shortly after this event he became ill and died soon after. The second culprit also died under the same conditions. The third man was a carpenter. One day he was working on the roof of a house. He started shouting my grandfather's name and asking for forgiveness. The cries of supplication continued until he fell off the roof and died. All this happened within the time frame given by the Voodoo priest. So my great-grandfather avenged his death.” Source: An account provided by one of the respondents.”

Since the Pew Research Center conducted a similar survey among the Muslim population in 27 countries, by comparing the responses provided by the Togolese respondents with the responses collected by the Pew Research Center, it is possible to assess how common traditional beliefs in Togo are compared to how common they are in other developing countries.

With regard to the belief in the evil eye, data collected by the Pew Research Center (2012) show that such belief displays quite a lot of cross-national variation. It varies in fact from a minimum of 9.9 percent recorded in Thailand to a maximum of 89.7 percent recorded in Tunisia, with an average of 53.8 percent. The percentage of Togolese respondents who believe in the evil eye is 25.9 percent higher than the global average. Only Morocco and Tunisia record higher values, while in all the other countries for which data were collected, the percentage of respondents who believe in the evil eye is lower than in Togo.

With regard to the belief in witchcraft or sorcery, the data collected by the Pew Research Center show that such beliefs also display considerable cross-national variation. They vary from a minimum of 8.4 percent, recorded in both Azerbaijan and Iran, to a maximum of 88.5 percent recorded in Tunisia, with an average of 36.3 percent. The percentage of Togolese respondents who believe in witchcraft is 35.1 percent higher than the global average. Only in Algeria, Indonesia, and Tunisia is the belief in witchcraft more popular than in Togo (see Box 2).

Box 2.

Witchcraft.

“Sorcerers exist and the most common type of sorcery is black magic. This is witchcraft that aims to harm other people, make them sick, prevent their success, destroy their lives or even kill them. This type of witchcraft is fueled by envy and jealousy. The most harmful sorcerers are those who have reached a level where they destroy the lives of others for trivial reasons. For example, there was a football match between two neighboring districts and the team from the district where the match took place lost the match. The winning goal was scored by a young player with a very promising future who was soon to sign for a club. One of the old men in the neighborhood was not at all happy that their local team had lost the match and said that he was going to “show” the young man who scored the goal. So he put a spell on the young man, who a few days after the match started having pain in his feet. What started as a small mosquito bite-like pimple that he scratched then worsened to a wound that refused to heal. Eventually his leg had to be cut off and he could not play anymore.”

Source: An account provided by one of the respondents.

With regard to witnessing an exorcism, the data collected by the Pew Research Center reveal that the percentage of respondents who claim to have had this experience displays considerable cross-national variation. It varies from a minimum of 1.9 percent recorded in Azerbaijan to a maximum of 21.1 percent recorded in Afghanistan, with an average of 9.67 percent. The percentage of Togolese respondents who have had such an experience stands at a staggering 56.2 percent—which is 46.5 percent higher than the global average. But the difference between Togo and other countries in the developing world is better highlighted by the fact that while in these other countries only one respondent out of ten claims to have witnessed an exorcism, in Togo, five respondents out of ten claim to have had such an experience. In other words, experiencing an exorcism is five times more common in Togo than in the countries for which the Pew Research Center collected its data.

The use of traditional healers is more common in Togo than in most of the countries for which the Pew Research Center collected data. The use of traditional healers, according to the data collected by the Pew Research Center, varies from a minimum of 25 percent in Russia to a maximum of 65.9 percent in Tajikistan, with an average of 39.1 and a standard deviation of 12.1. The percentage of Togolese respondents using traditional healers is 14.4 percent higher than the average computed in the sample of countries included in the research of the Pew Research Center. It is important to note, as evidenced by the information provided in Box 3, that traditional healers are consulted for natural problems and for witchcraft-induced problems.

Box 3.

The Traditional Healers.

“Another story is about a woman who went to the market and wanted to eat ‘Ablo,’ an African food that is usually eaten with tomato sauce, chilli pepper, and especially fish. The Ablo seller had different kinds of side dishes and fried fish on her stand. However, the lady did not have enough money to afford the meal. She complimented the vendor and her stand so that the vendor would offer her the meal. Unfortunately, the vendor did not understand her intention and did not offer her any food. But the fact that the seller did not offer her the food made the lady angry and she bewitched the seller's child. As a result of this event, the vendor's child became ill and nothing seemed to cure the illness. The seller then went to a fetishist to find out the source of her child's illness. After questioning the spirits and the ancestors, the fetishist asked the vendor to prepare a dish of Ablo with lots of side dishes and fish and to offer it to the woman. The vendor followed the advice of the fetishist and proceeded with the offering. Sometime later the child was cured. The offering thus appeased the lady's anger.”

Source: An account provided by one of the respondents.

Finally, since in Togo, as in some other West African countries (Benin, Nigeria), a portion of the population believes in voodoo, we asked respondents whether they knew anybody who believes in this traditional religion. Interestingly, while only 3.5 percent of our respondents reported being believers in voodoo, 85.7 percent of respondents claimed to know someone who believes in voodoo. This set of data consistently shows that traditional beliefs are very common among Togolese respondents and more widespread than in the Muslim communities for which the Pew Research Center collected data.

The third set of data concerns the political orientation of the respondents, including whether they vote in elections, whether they vote for the ruling party (or not), whether they talk about political issues with their family members and priests, and whether they want change.

The data presented in Table 3 make it quite clear that the overwhelming majority of the respondents does not talk with their priests about political issues—hence it is somewhat difficult to explain the electoral success of the ruling party and the stability of the party system on the basis of the second explanation that we presented above. For clerics to be agents of stability, a simple condition has to be met, namely that clerics talk about politics with the faithful and that they provide political/electoral guidance. The data presented in Table 3 do not sustain this claim.

Table 3.

Interest in Politics—Percent of Respondents.

	Yes (%)	N
Want change	100	202
Vote in the elections	61.9	202
Talk about politics in the household	53.5	202
Talk about politics with the priest	8.5	201
Vote for the ruling party—UNIR	61.7	115

A little more than 50 percent of the respondents talk about politics at home, while nearly two-thirds of the respondents vote in the elections and about two-thirds of those who vote in the elections claim to vote for the ruling party. Interestingly, all the respondents want change, are unhappy with the (political/economic) status quo, and want Togo to make progress along the developmental path.

One of our respondents noted, in this regard, that the problem is not so much at the government level, with the president, but with lower-level officials. Lower-level officials are criticized for thinking about their own well-being. In this regard, the respondent noted that “The deputy governors should stop thinking only about their stomachs because the president himself is not aware of everything.” The same respondent, however, went on to note that opposition parties deserve a fair share of the blame because, by not doing their job properly, they fail to ensure any kind of accountability in the political system. In this respect, he observed that

“the opposing parties who do smooth talking during the electoral campaigns disappear after the elections. It's true that we want change with regard to the development of the country and the party in power, but at the same time how could we trust those who say they can change the country and respond to our tears when we no longer see them after their defeat in the elections? We keep shouting and they're not there. But they will all come back again to shout from the rooftops to vote for them and worse pay young citizens to promote themselves and vote for them.”

Other respondents, however, lamented that the country is not doing well, that the people struggle to survive and make a living, that the problem is the lack of good governance and that a change in power could go a long way in promoting good governance and creating the conditions for economic growth and development.

“For me, Today when it comes to change in this country, my point of view is more on the governance side. For more than 50 years, a single family has governed us, from father to son. All we had as a change was the manipulation of the letters for the creation of an acronym for the party in power. RPT (by the father) which becomes UNIR (for the son). Abuse of power, political manipulation is all there is in this country. If we can have another party, if only for a mandate, it would be a new experience of under governance for us Togolese.”

Similar concerns were echoed by another respondent who lamented that the corruption of the ruling party had corrupted not only the integrity of the electoral process but also the integrity of opposition parties:

“has been in power for decades. And even the opposing parties are also corrupted by UNIR. The stories of elections here in Togo are just stagings (sic) that unfortunately put the lives of young people in danger who think they are starting to fight on the left side. Elections have always been rigged in this country and when an opponent really imposes himself and demands real results, he is called as not respecting the government.”

Beyond Description

The descriptive statistics discussed so far have shown that the Togolese population is young, traditional beliefs are pervasive, there is a general consensus on the fact that Togo needs change (though there is some disagreement as to what this change may entail), and there are fairly low levels of interest in politics and political participation. The question that the descriptive statistics were not able to answer is whether traditional values are in some ways responsible for respondents’ political apathy and/or demand for change.

The first hypothesis we had formulated to explain why party systems are more stable or display lower levels of fluidity in countries in which traditional beliefs and superstitions are more common stated that voters with traditional beliefs are change averse. Change, in turn, can be understood as our respondents indicated as either political change or a change in the economic conditions of the country.

As we noted earlier, while a sizeable portion of the population has traditional beliefs or engages in traditional practices, all of the respondents reportedly want change. This finding may be of some interest to modernization theorists. While modernization theory posits that modernization leads to cultural change—from traditional to modern, rational, secular culture—our data reveal that citizens in the developing world may desire, strive for, and demand development and modernization in spite of the fact that they hold traditional beliefs. The quest for change and modernization is not hampered by the presence of traditional values and beliefs. Despite the fact that an overwhelming majority of Togolese respondents believe in the evil eye and witchcraft, it does not prevent them from desiring change, development, or modernization. All our respondents want change in the political and economic realms, but at the same time, they are deeply embedded in their traditional beliefs as our data have revealed.

Second, our data reveal that Togolese voters are not very active politically—they do not talk much about politics, and more than one-third of our respondents reported not participating in the elections. Only 61.9 percent of our respondents voted in the elections—this value is in line with the values reported by Tobolka (2014) and Kohnert (2015). To make matters worse, as Kohnert (2015, 14) observed, the turnout rate varies considerably within the country: the turnout rate is much higher in the UNIR-dominated North and much lower in the South, where the opposition is (expected to be) stronger. Obviously, if the supporters of the ruling party go to the polls in larger numbers and the supporters of the opposition go to the polls in smaller numbers, political change in the form of government alternation will not occur.

Third, even though all our respondents are from the South, where UNIR is less popular, more than 60 percent of those who were willing to provide some information as to what party they vote for said they vote for UNIR. Again, voting for the ruling party is obviously a legitimate choice, and it may be somewhat consistent with the hope that the ruling party may eventually modernize the country, but it is entirely inconsistent with the desire to engender and promote political change and alternation in government.

Do voters with a traditional mindset prefer the political status quo? In order to answer this question, we decided to run a logistic regression to assess the probability that a respondent may vote in the elections and, second, that she may vote for the ruling party. By performing such analyses, we found that the probability of voting in the elections is lower for voters who believe in the evil eye, while the probability of voting for the ruling party, UNIR, is higher for voters who believe in the evil eye (Table 4).

Table 4.

The Cultural Determinants of Togolese Voting Behavior.

Togo	(1)	(2)
	Vote	Vote for ruling party
Age	0.0372	0.0234
	(0.0306)	(0.0300)
Muslim	−0.131	0.545
	(0.421)	(0.631)
Evil eye	−0.673*	1.194**
	(0.405)	(0.487)
_cons	0.0257	−1.044
	(0.922)	(0.948)
N	190	108

Notes: Standard errors in parentheses.

*p < .10.

**p < .05.

***p < .01.

The implication of these findings is that traditionally minded voters are more likely, by voting for the ruling party, to support the status quo and to contribute to the stability of the party system. Conversely, more modern-minded voters are more likely to vote for the opposition parties, to support political change and to create the conditions for a party system change.

To properly appreciate the meaning or the significance of these findings, it is best to see them from a comparative perspective. When we compare, Sartori (1970) noted, we need to be aware of the fact that there may be a trade-off between the precision (in intension) with which we apply a specific concept and its extension. Something similar can be detected also with regard to the generalizability of our findings. If the key conclusion of our research is that traditional beliefs matter, then we can assert that the evidence generated by the analysis of data collected in other countries largely sustains this claim. Pelizzo and Kuzenbayev (2023) show in this special issue that Muslim voters with a traditional mindset are less supportive of democratic rule than those Muslim voters who have a less traditional mindset. If the key conclusion of our article is that traditional beliefs affect voters’ electoral behavior, then we can say that the evidence generated by Harakan, Pelizzo, and Kuzenbayev (2023), also in this special issue, largely sustains this claim. Yet, once we try see the ways in which traditional beliefs shape electoral behavior, we find that there are some cross-national differences. In Tanzania, for instance, respondents who report to have witnessed an exorcism are less likely to vote in the elections but their willingness to vote for the ruling party is not affected by whether they have witnessed an exorcism or not (see Table 5).⁴

Table 5.

Traditional Beliefs and Electoral Behavior in Tanzania.

Tanzania	(1)	(2)	(3)
	Vote	Disclose	Vote for ruling party
Age	0.0576***	0.00831	−0.00111
	(0.0117)	(0.0107)	(0.0133)
Muslim	−0.322	0.854**	0.391
	(0.335)	(0.334)	(0.423)
Exorcism	−0.575*	0.860***	0.178
	(0.322)	(0.323)	(0.433)
_cons	−1.765***	−0.293	0.789
	(0.459)	(0.463)	(0.676)
N	200	200	137

Notes: Standard errors in parentheses.

*p < .10.

**p < .05.

***p < .01.

In Benin, the belief in the evil eye increases the probability that a voter will cast her ballot for the ruling party but has no detectable impact on whether they participate in the elections or not, while witnessing an exorcism does not affect the voters’ propensity to vote in the elections or not but decreases their propensity to vote for the ruling party (see Table 6).⁵

Table 6.

Traditional Beliefs and Electoral Behavior in Benin.

Benin	(1)	(2)	(3)	(4)
	Vote	Vote for ruling party	Vote	Vote for ruling party
Age	0.110***	−0.0333	0.109***	−0.00982
	(0.0377)	(0.0341)	(0.0374)	(0.0364)
Muslim	−0.178	0.383	−0.154	0.173
	(0.594)	(1.000)	(0.595)	(1.105)
Evil eye	−0.149	2.152*
	(0.424)	(1.173)
Exorcism			0.196	−1.413**
			(0.341)	(0.594)
_cons	−2.970***	−0.986	−3.191***	1.049
	(0.987)	(1.489)	(0.865)	(1.084)
N	172	55	172	55

Notes: Standard errors in parentheses.

*p < .10.

**p < .05.

***p < .01.

So, while the evidence presented so far sustains the claim that traditional beliefs matter, in the sense that they affect electoral behavior, the way in which traditional beliefs do so varies from country-to-country depending on country-specific contextual conditions. The way in which traditional beliefs influence electoral behavior in Togo (and/or Benin) is rather different from the way which they do so in Tanzania or Indonesia.

A possible objection to what we have said so far is that the Togolese party system, just like the Tanzanian party system, is uncompetitive and electoral reforms and a more competitive party system could weaken the linkage between traditional beliefs and the vote for the ruling party. Such an objection is obviously plausible. If voters were given a wider set of viable options, they might vote differently. Yet, the evidence that Harakan, Pelizzo, and Kuzenbayev (2023) have presented in this special issue with regard to the Indonesian case and the evidence pertaining to Benin raise some doubts as to how compelling this objection may actually be. The party system in Benin and Indonesia is considerably more competitive than it is in Togo or, for that matter, in Tanzania—and yet the evidence we presented above and the evidence presented by Harakan, Pelizzo, and Kuzenbayev (2023) also in this special issue shows consistently that traditionally minded voters are more inclined to vote for government parties even in countries where the inter-party competition is more competitive.

Togo is not an exceptional case. The presence and pervasiveness of traditional beliefs in the country contribute to the stability of the party system. The question, however, is whether the Togolese cultural landscape can be reshaped by the advent/usage of new communication technologies and social media. Specifically, one may wonder whether the use of social media may contribute to the cultural modernization of the Togolese electorate and eventually pave the way for a transformation of the party and political system.

Our data do not allow us to answer this question in a conclusive fashion, but it is nonetheless possible to formulate an educated guess: namely, that at the time being the use of social media is unlikely to have a transformative impact on the political culture of Togolese voters, to make traditional beliefs obsolete and to promote a change in the voters’ electoral behavior and the party system for at least three reasons.

The first reason is that internet usage and the usage of social media, albeit growing, are still very limited. In 2022 the internet penetration rate stood at 25.9 percent, which represented a 2.4 percent increase over the previous year. This internet penetration, in spite of its upward trend, is lower than in neighboring countries—in Ghana it stands at about 58 percent.

The use of social media is also rather limited. Digital 2022 reports that, in 2022, there were 956,700 social media users in the country or about 11 percent of the country's population (Digital 2022: Togo—DataReportal—Global Digital Insights), while Digital 2023 reported that the number of social media users had dropped to 775,600 in 2023 or 8.7 percent of the population (Digital 2023: Togo—DataReportal—Global Digital Insights). Regardless of whether the change indicates a drop in the social media usage or simply reflects the adoption of a new methodology to estimate the popularity/usage of social media, the evidence presented here sustains the claim that the use of social media is fairly limited, that only a small portion of the population is using such media, and that it is not entirely clear whether social media are becoming increasingly more popular in the population. Hence, it is somewhat unlikely for social media to engender a value change and have a transformative impact on Togolese society and culture.

Second, these social media are used for the same reasons as they are used elsewhere: to connect, to be in touch with friends, and to let friends and acquaintances know what is happening in one's life, but they are also used for what could be loosely regarded as cultural purposes. In recent years, a group of young writers—Francis Kev and Tony Hemrix, among others—have started using a wide range of social media and portals to showcase and promote their (erotic) writings.⁶ Instagram, WhatsApp, Facebook, and Wattpad are used by these young writers to connect with their readers. And the bond between these young writers and their readers is particularly strong, not simply because these young writers have thousands of readers and followers, but because readers are keen to provide authors with some feedback which feeds into—as one of these authors mentioned to us—the writing of new installments of their novels. Tony Hemrix's Facebook page, where his novels or ‘chroniques’ are serialized, has more than 17,000 followers (228 Hemrix ChroniQ's | Facebook), while the page of Francis Kev has more than 9000 followers (Les Chroniques de Francis Kev | Facebook). On Instagram, Francis Kev has more than 2,000 followers (Francis-Kev (@francis_kev) • Instagram photos and videos), while Tony Hemrix has about 1,000 followers (Tony Hemrix (@tonyhemrix) • Instagram photos and videos). This means not only that the content of Togolese literature is changing and that young readers are using new (social) media—channels in the semiotic sense (Jakobson, Sebeok 1960)—to connect with their readers, but also that there is a sizeable portion of the Togolese readership that is keen to read a literature that one would not immediately associate with the tastes of readers imbued with/by traditional values. But even assuming that these writers or cultural entrepreneurs may be able, eventually, to promote cultural and political change in the country, this kind of change may occur over a relatively long period of time—hence, it is not realistic to expect social media to reshape Togolese society and culture any time soon.

The third reason is that traditional beliefs remain quite common and deeply ingrained not only in the most conservative or traditionally minded segments of society but also in the mindset of forward-looking, progressive, modern young Togolese—from young computer scientists to erotic writers, to students who have joined the Togolese diaspora and have eventually moved to industrially advanced, Western societies. The content of the three boxes presented above that testifies to the persistence and pervasiveness of traditional beliefs was not provided by old, illiterate, (wo)men living in rural settings, but was provided instead by young, highly educated, urban residents. It is not uncommon in the country for young and highly educated individuals to be simultaneously tech savvy, active on social media, and traditionally minded.

Conclusions

The purpose of the present article was to deepen our understanding of countries in which traditional beliefs (such as belief in witchcraft) are pervasive, and the party system is more stable than in countries in which traditional beliefs are less widespread. We hypothesized that the (possibly puzzling) relationship we had detected between the stability of the party system and the pervasiveness of traditional beliefs could be explained in different, though, possibly, complementary ways. First, we hypothesized that the link between party system stability and traditional beliefs could be explained by the fact that individuals who hold onto traditional beliefs (in spirits, magic, the evil eye, etc.) may be more averse to change than individuals who instead have abandoned such beliefs. Second, we hypothesized that in countries with widespread traditional values and beliefs, priests (who may benefit from the status quo) may be consulted by their followers and may advise them not to vote against the status quo and to seek change.

The evidence presented in the present article shows that Togolese voters are not terribly active, that they generally do not talk about politics at home, and that they talk about politics with their priests or religious leaders. Only 8.5 percent of the respondents talks about political issues with their priests—a percentage sufficiently low to suggest that, even if priests and religious figures wanted to be as agents of stability, they may not be able to do so because their opinions and advice are not sought after in political matters. Hence, the evidence we have collected does not allow us to accept the second hypothesis.

With regard to the relationship between the traditional mindset and the aversion to change, the evidence that we have gathered in the course of the present study shows unequivocally that the desire for change is shared by each of the respondents who participated in our survey regardless of whether they held traditional beliefs or engaged in traditional practices. Hence, the evidence at our disposal does not allow us to accept the first hypothesis.

Our data, however, reveal that while respondents are not change-averse, they are more inclined to vote for the ruling party than for the opposition forces and that certain cultural traits, such as belief in the evil eye, affects voters’ electoral behavior. Voters who believe in the evil eye are less likely to participate in elections but more likely to vote for the ruling party and, by doing so, they contribute—among others—to the stability of the party system. When we regress the vote for the ruling party against the belief in the evil eye, we find that such a belief has a fairly strong, positive, and statistically significant impact on the probability to vote for the ruling party—which is what the analysis of data on fluidity and belief in witchcraft, aggregated at the country level, had also suggested. The fact that electoral behavior in Togo is influenced by the belief in the evil eye corroborates a claim that Pasquino and Pelizzo (2022, 151) recently advanced: that in explaining political phenomena “culture matters and, second that some aspects of culture are more important than others.” Therefore, if we want to develop a more nuanced understanding of the impact of ‘culture’ on the functioning of political systems, greater attention should be paid to the fact that to properly capture ‘culture’ in its manifold manifestations, it is necessary to go beyond a conceptualization that reduces culture to religion or religiosity and appreciate what other facets of ‘culture’ make political systems work the way they do.

Footnotes

About the Authors

Riccardo Pelizzo is a Professor and the Vice Dean for Academic Affairs in the Graduate School of Public Policy at Nazarbayev University. A comparative politics specialist, he has extensively explored party system change in Sub-Saharan Africa. With Gianfranco Pasquino he recently published The Culture of Accountability (Routledge, 2023). He can be contacted at riccardo.pelizzo@nu.edu.kz

Yao Moise Koepko is an independent researcher from Lomè, Togo. His writings regularly appear in several online and traditional media.

Nygmetzhan Kuzenbayev is a PhD candidate at Nazarbayev University, Astana, Kazakhstan. His research interests lie at the intersection of economics, psychology, and political science. He can be contacted at nygmetzhan.kuzenbayev@nu.edu.kz

Abel Kinyondo holds a PhD from Monash University, Australia. He is a senior lecturer at the University of Dar es Salaam (DUCE) and is currently a visiting scholar at York University in Canada. Previously, he was the head of Economics and Geography Department at DUCE. He has also worked for the United Nations Development Program. A seasoned Development Economist, he has over 15 years of experience having widely published over 70 articles, books, and book chapters in various reputable outlets. He can be contacted at kinyondoabel@yahoo.com

ORCID iDs

Riccardo Pelizzo

Nygmetzhan Kuzenbayev

Abel Kinyondo

Notes

References

Adebanwi

Wale

Orock

Rogers

, eds. 2021. Elites and the Politics of Accountability in Africa. Ann Arbor: University of Michigan Press.

Bartolini

Stefano

Mair

Peter

. 2007. Identity, Competition and Electoral Availability: The Stabilisation of European Electorates 1885–1985. Colchester, UK: ECPR Press.

Bértoa

Fernando Casal

. 2022. “The Problem of Party System Change Revisited: The 2022 Peter Mair Lecture.” Irish Political Studies, 1–29. https://doi.org/10.1080/07907184.2022.2161034.

Bértoa

Fernando Casal

Enyedi

Zsolt

. 2021. Party System Closure: Party Alliances, Government Alternatives, and Democracy in Europe. Oxford: Oxford University Press.

Dalton

Russell J.

McAllister

Ian

Wattenberg

Martin P.

. 2000. “The Consequences of Partisan Dealignment.” In Parties Without Partisans: Political Change in Advanced Industrial Democracies, edited by Dalton

Russell

, 37–63. Oxford: Oxford University Press. https://doi.org/10.1093/0199253099.003.0003.

Ekwem

Nalugala

Okiya

D. O.

. 2020. “The Effects of Witchcraft Practices on the Electoral Process in Nigeria: A Case of Anambra State.” The Strategic Journal of Business & Change Management 7 (3): 1471–1483. https://www.academia.edu/download/77071392/1681.pdf (accessed April 22, 2023).

Evola

Julius.

1988. Simboli della Tradizione Occidentale. Carmagnola, Italy: Edizioni Arktos.

Gilli

Bruno.

1980. “Approcci al Vodù: Tra l’uomo e Mawu.” Nigrizia 98 (1): 49–52.

Gilli

Bruno.

1986. “Vodù: Scuola di Vita.” Nigrizia 104 (11): 29–36.

10.

Gilli

Bruno.

1987. Heviosso et le Bon Ordre du Monde: Approches d’une Religion Africaine. Lome: Editions HAHO.

11.

Gilli

Bruno.

2004. Bambini Vodu: la Nascita nella Religione dei Wacì del Togo. Bologna: EMI.

12.

Gilli

Bruno.

2006. “Togo: Vodù e Inculturazione nel Regno di Hebiesso.” Nigrizia 124 (5): 54–56.

13.

Harakan

Ahmad

Pelizzo

Riccardo

Kuzenbayev

Nygmetzhan

. 2023. “Traditional Beliefs and Electoral Behavior in Indonesia.” World Affairs 186 (3). https://doi.org/10.1177/00438200231203005.

14.

Hemrix

Tony

. 2020. Chaos Sensuel . Editions La Calabasse. https://www.amazon.com.au/Chaos-Sensuel-French-Tony-Hemrix-ebook/dp/B0897RJQ7L (accessed April 22, 2023).

15.

Inglehart

Ronald

. 1971. “The Silent Revolution in Europe: Intergenerational Change in Post-Industrial Societies.” American Political Science Review 65 (4): 991–1017. https://doi.org/10.2307/1953494 (accessed April 22, 2023).

16.

Jakobson

Roman

Sebeok

Thomas A.

. 1960. “Closing Statement: Linguistics and Poetics.” In Semiotics: An Introductory Anthology, edited by Innis

Robert E.

, 147–175. Bloomington: Indiana University Press.

17.

Kohnert

Dirk

. 2015. “Togo: Recent Political and Economic Development.” 1–35. https://mpra.ub.uni-muenchen.de/63411/8/MPRA_paper_63411.pdf (accessed April 22, 2023).

18.

Kuenzi

Michelle

Lambright

Gina

. 2001. “Party System Institutionalization in 30 African Countries.” Party Politics 7 (4): 437–468. https://doi.org/10.1177/1354068801007004003 (accessed April 22, 2023).

19.

Kuzenbayev

Nygmetzhan

Pelizzo

Riccardo

. 2023. “Political Stability, Confidence in the Future, and Values.” Social Sciences 12 (2): 1–16. https://doi.org/10.3390/socsci12020082.

20.

Laguerre

Michel S.

1989. Voodoo and Politics in Haiti. New York: Palgrave Macmillan.

21.

Lindberg

Staffan I.

2007. “Institutionalization of Party Systems? Stability and Fluidity Among Legislative Parties in Africa's Democracies.” Government and Opposition 42 (2): 215–241. https://doi.org/10.1111/j.1477-7053.2007.00219.x (accessed April 22, 2023).

22.

Lipset

Seymour Martin

Rokkan

Stein

, eds. 1967. Party Systems and Voter Alignments: Cross-National Perspectives. New York: Free Press.

23.

Lugo

Luis

Cooperman

Alan

Bell

James

O'Connell

Erin

Stencel

Sandra

. 2013. The World's Muslims. Washington DC: Pew Research Center.

24.

Mair

Peter.

1997. Party System Change: Approaches and Interpretations. Oxford: Oxford University Press.

25.

Mozaffar

Shaheen

Scarritt

James R.

. 2005. “The Puzzle of African Party Systems.” Party Politics 11 (4): 399–421. https://doi.org/10.1177/1354068805053210 (accessed April 22, 2023).

26.

Nwokora

Zim

Pelizzo

Riccardo

. 2015. “The Political Consequences of Party System Change.” Politics & Policy 43 (4): 453–473. https://doi.org/10.1111/polp.12124.

27.

Pasquino

Gianfranco

Pelizzo

Riccardo

. 2022. The Culture of Accountability. London: Routledge.

28.

Pedersen

Mogens N.

1979. “The Dynamics of European Party Systems: Changing Patterns of Electoral Volatility.” European Journal of Political Research 7 (1): 1–26. https://doi.org/10.1111/j.1475-6765.1979.tb01267.x.

29.

Pelizzo

Riccardo

. 2018. “The Determinants of Party System Change in Sub-Saharan Africa.” Otoritas: Jurnal Ilmu Pemerintahan 8 (2): 124–129. https://journal.unismuh.ac.id/index.php/Otoritas/article/viewFile/1161/1287 (accessed April 22, 2023).

30.

Pelizzo, Riccardo. 2020. “Good Governance and Party System Change.” Politics & Policy 48 (2): 265–287.

31.

Pelizzo

Riccardo

Kuzenbayev

Nygmetzhan

. 2023. “Traditional Practices and Support for the Strongman.” World Affairs 186 (4): 951–977. https://doi.org/10.1177/00438200231203094.

32.

Pelizzo

Riccardo

Nwokora

Zim

. 2016. “Bridging the Divide: Measuring Party System Change and Classifying Party Systems.” Politics & Policy 44 (6): 1017–1052. https://doi.org/10.1111/polp.12188.

33.

Pelizzo

Riccardo

Nwokora

Zim

. 2018. “Party System Change and the Quality of Democracy in East Africa.” Politics & Policy 46 (3): 505–528. https://doi.org/10.1111/polp.12255.

34.

Pelizzo

Riccardo

Nwokora

Zim

. 2022. “Commentary: The Cultural Determinants of Party System Change.” Policy Studies Yearbook Annual Review 2021 12: 235. https://www.psoyearbook.org/wp-content/uploads/2022/04/PSY-12.1-full-issue.pdf#page=247 (accessed April 22, 2023).

35.

Pelizzo

Riccardo

Turganov

Daulet

Kuzenbayev

Nygmetzhan

. 2023. “Modernization, Superstition and Cultural Change.” World Affairs 186 (4): 869–895. https://doi.org/10.1177/00438200231203012.

36.

Pew Research Center. 2012. World’s Muslims Data Set [Dataset]. https://www.pewresearch.org/religion/dataset/the-worlds-muslims/ (accessed April 22, 2023).

37.

Sartori

Giovanni.

1970. “Concept Misformation in Comparative Politics.” American Political Science Review 64 (4): 1033–1053. https://doi.org/10.2307/1958356.

38.

Sartori

Giovanni

. 1976. Parties and Party Systems: A Framework for Analysis. New York: Cambridge University Press.

39.

Tobolka

Radim

. 2014. “Togo: Legislative Elections of July 2013.” Electoral Studies 35: 389–394. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2489315 (accessed August 2, 2023).