Navigation – Plan du site

AccueilNuméros61-1Comptes rendus de lectureGethin AMORY, Clara MARTINEZ-TOLE...

Comptes rendus de lecture

Gethin AMORY, Clara MARTINEZ-TOLEDANO, and Thomas PIKETTY (eds), Political Cleavages and Social Inequalities: A Study of Fifty Democracies, 1948-2020

Mathieu Lizotte
p. 279-284
Référence(s) :

Gethin AMORY, Clara MARTINEZ-TOLEDANO, and Thomas PIKETTY (eds), 2021, Political Cleavages and Social Inequalities: A Study of Fifty Democracies, 1948–2020, Cambridge, Harvard University Press, 642 p.

Texte intégral

1This book will be of great interest to those who are preoccupied with the changing nature of politics and comparative analysis. Officially, the authors’ stated objective is to study how electoral choices vary according to a series of sociodemographic and socioeconomic characteristics in a systematic manner. However, the central theme essentially boils down to the issue of class versus identity factors in electoral politics. As many authors have argued, the left-right divide has become increasingly limiting to explain contemporary politics. Accordingly, we are in need of a deeper understanding of contemporary politics and in due time of a more fitting metaphor that builds upon the left-right spectrum. In a nutshell, class cleavages are not what they once were and lamenting this fact will not help us achieve a better understanding of the present. The electorate in liberal democracies appears more complex and fragmented than ever. Politicians are struggling to develop platforms that can rally these constituents into strong majority governments or coalitions, just as political scientists are trying to make sense of this new reality. This book arguably offers one of the most exhaustive demonstrations of the changing nature of electoral politics as it covers data from fifty countries from 1948 to the present. While the authors do not pretend to arrive at a definitive answer, they certainly achieved a detailed empirical investigation of electoral politics that provides much insight and sets the stage for further research and understanding.

2It must be said that this book is very descriptive, as the authors acknowledge themselves, and this might be disappointing to some readers. However, I would argue that it is descriptive in the best sense of the word. The merged dataset they have developed from a series of national post-electoral surveys is no easy feat and permits an impressive level of international and temporal comparisons. I hold the conviction that the culture of description is to be encouraged since it is the bedrock of all explanations. I suspect many researchers are reluctant to publish more descriptive work as the absence of compelling explanations or interpretations might be seen as a failure. This would be unfortunate, however, since such work can only serve to promote dialogue between researchers and ultimately lead to more compelling interpretations. Explanations are like narratives and they need to be solidly grounded to have scientific weight.

3While the length of this book is imposing, most readers may focus on the primary chapter and then read a few country-level comparisons or case studies. At first glance, it might be tempting to say that the remaining 18 chapters may be consulted almost like an encyclopedia according to one’s countries and regions of interests. This would be a mistake, however, since there are many insights scattered throughout the individual chapters. Furthermore, it is necessary to read a few of these to get a deeper understanding of the central arguments developed in the main chapter. While the structure of the chapters can be somewhat repetitive, this is necessary in order to systematically apply the analytical framework they have developed to all empirical cases (including countries from all continents).

4The theoretical framework of this book is based on the study of political cleavages, which is an established tradition in the social sciences. Notably, they cite the pioneering work of Seymour Martin Lipset and Stein Rokkan (Party Systems and Voter Alignments. Cross-National Perspectives, 1967, Free Press) in this regard. A political cleavage serves to understand how political power and influence is distributed across different groups in society. Accordingly, the authors analyse their data through the lenses of well-established cleavages, such as class (i.e. education, income), religion and religiosity, generations, rurality versus urbanity, regions and gender. They also hint at the growing importance of the ecological cleavage as can be seen indirectly by the growing influence of green parties.

5The analytical framework that is developed and applied throughout this book is a typology of party systems. The purpose of this typology is to provide a global view of the main configurations of party systems observed among the countries. This typology is composed of four ideal-types: 1) class-based party systems; 2) identity-based party systems; 3) multidimensional party systems; and 4) the depolarized or unstable party system. The class-based party system is characterized by a strong and persistent socioeconomic divide whose influence on electoral behavior is particularly important. This was the case for the majority of post-war Western democracies, where the working class strongly supported left-wing parties, but fewer countries correspond to this type today (e.g. Brazil and Argentina today). At the other end of the spectrum, the identity-based party system corresponds to the situation where socioeconomic cleavages are virtually eclipsed by other types of political conflicts which are fundamentally more defining for electoral behaviour (e.g. ethnicity, religion, etc.). It is not surprising that countries such as Iraq, South Korea and Taiwan correspond to this type given their political and geopolitical situation. Although the authors observe a deep tendency for the growing importance of identity-based conflicts that inhibit class divides in many countries, these do not necessarily correspond the most to identity-based party systems.

6Most democracies reviewed in this book correspond to the multidimensional party system to a varying degree, which the authors argue should be seen as a middle way between the class-based and identity-based party systems. This type corresponds to societies where socioeconomic divides are still important, but they are not the only political conflicts that structure electoral outcomes. The clearest example of this is what the authors call multi-elite party systems where two dimensions of social class (i.e. education and income) no longer align politically (i.e. top incomes continue to support right-wing parties, while the highly educated have progressively gone through a reversal and now overwhelmingly support left-wing parties). The last type is the depolarized or unstable party system which corresponds to the situation where there are no clear and stable cleavages or stable partisan affiliations. This type of party system corresponds to countries like the Philippines, Indonesia, Peru and Japan where electoral politics are often dominated by a charismatic leader.

7This typology serves as the main analytical framework throughout the comparative and case studies following the main chapter. Obviously, a country’s party system evolves and changes not only between types but also by the degree to which they conform to the ideal-type. It appears that the transitions are mostly gradual, however, abrupt changes can also be observed (e.g. when a country’s party system temporarily shifts to an unstable party system). There is certainly room for the authors to expand on this framework in future work in at least two ways. Firstly, since the bulk of countries have multidimensional party systems, further analysis will most likely reveal some pertinent subtypes linked to the nature of their political cleavages and conflicts. Secondly, there are also more connections to be made how political conflicts and inequality affect party systems.

8The dataset should be considered a contribution itself. The collaborative efforts of nearly two dozen researchers have managed to assemble and merge several hundred post-electoral surveys conducted in fifty democracies throughout the world for the 1948 to 2020 period. Post-electoral surveys, as the name implies, are conducted after an election to gather information about the attitudes and behaviors of voters, such as reasons for voting and demographic and socioeconomic characteristics. While I understand that methodological chapters are not necessarily the most exciting, probably for readers and editors alike, a more in-depth discussion regarding the pros and cons of post-electoral surveys would have been welcome. There are limits and disadvantages to using post-electoral surveys (e.g. limited scope, sample size, selection bias, etc.) and while these are acknowledged to some extent, they are certainly not discussed to any great length.

9The absence of a detailed methodological discussion does cast some shadow of doubt on the validity of some findings. For instance, voter turnout and electoral systems are potentially big issues. Most countries do not have mandatory voting, which means that a sizable and variable percentage of the population do not vote. How are these non-voters treated in the analysis? Are they simply excluded or are they included in the sense of which political party they would have voted for? In Europe alone, the national election turnout ranges approximately from 30% to 90%. If non-voters are excluded, this implies that the analysis, at least for some countries and for some given years, only reflects a minority of legal voting-age citizens (i.e. and not the whole party system). Furthermore, voter turnout also varies in countries with compulsory voting, typically ranging from 60% to 90%. These types of methodological issues are further complicated in the context of international datasets where survey methodologies and practices must not only vary between countries but also by year. The main concern here is that the analyses performed may not quite represent the whole party systems.

10Whether this is possible with currently available data, the issue deserves to be explored further in some way. The self-disenfranchised, for lack of a better term, offer a promising key to better understand party systems and their underling relationship with inequality. Who are these individuals and from what walks of life are they from? Beyond the apathetic and the disillusioned, they could be the constituents of ideologies and political projects that are poorly represented by the political supply. Also, what conclusions could be drawn between the self-disenfranchised and electoral rules? Are there more self-disenfranchised citizens in countries with first-past-the-post (FPTP) electoral systems than those with proportional representation? How are non-voters related to political conflicts, the level of inequality and political platforms?

11The dataset is openly available online (<>) and the website interface allows some data analysis by party, election, and political cleavages across countries and over time, etc. We can expect interesting scholarship to come from this dataset. While the set of variables in the dataset is rather limited (a common characteristic for post-electoral surveys), this disadvantage is compensated by the sheer number of countries and the large period available. This commitment to transparent research is a step in the right direction, especially in the context of the replication crisis which is ongoing in both science (medicine) and the social sciences (psychology, economics). As it is well known, the difficulty encountered in replicating many seminal studies is casting doubt in the scientific endeavour, with issues such as the publication bias towards positive findings, p-hacking, etc. Mistakes are inevitable, but data transparency allows us to uncover and address them. Science is supposed to be cumulative, but it is perilous to pursue avenues based on assumptions that are falsely or very provisionally validated. This move towards data transparency can only be good in the long-run, although it may be painful in the short-run.

12The authors conclude as rightly as humbly that much work remains to be done, despite the considerable work they did achieve. In this regard, they acknowledge the need to mobilize other sources and methods to build upon their results. What is firmly established, however, is that the new political landscape is no longer one strictly of redistribution (i.e. to the extent that it ever was). Accordingly, the authors emphasize that nostalgia for class-based politics is not a good guide to understand the nature of contemporary politics. Nonetheless, the authors affirm their preferences for class-based conflicts since they are arguably more soluble. This is a common sentiment among many scholars (myself included), which is often framed as a sort of paradise lost, as if the rise of identity-based politics suddenly opened a Pandora’s box of irresolvable contention. Upon reflection, however, it is more likely that many of these conflicts were already present in some form, but they did not get the same attention. Many issues were surely minimized, neglected, and ignored (i.e. there is a parallel here to be drawn between demand and effective demand). Several arguments have been made to explain the rise of identity-based demands over class-based ones (i.e. the rise of post-industrial values, pent-up grievances, etc.), but whatever the reason, as this book clearly documents, identity-based cleavages are on the rise and cannot be ignored. Since most countries presently correspond to multi-elite party systems, class cleavages are not dead, but they co-exist with identity-based conflicts. The problem, however, is that intersectionality does not necessarily imply solidarity between fragmented groups.

13Not only are identity-based conflicts on the rise, but it does not seem realistic that citizens will accept to put aside these concerns in order to deal solely with economic inequality. To do so, if I may speculate for a moment, would require some notable events, such as massive economic insecurity, grassroot mobilization, great political leadership, a shift in political culture, a large-scale coalition, etc. Furthermore, whether they can put these concerns aside does not necessarily mean they should. This, of course, echoes the ongoing debate within liberal thought between the difference-blind and the difference-sensitive approach. Theoretically, it does appear like a convincing argument that more progress is to be made by addressing inequality through redistribution policies within the classic difference-blind liberal framework. However, the problem is that the difference-blind policies, if applied too blindly, can lead to neglecting the experiences of marginalized groups and thus fail to address structural inequality. In this sense, the rise of identity-based political demands may not be false consciousness per se (or whichever synonym is preferred) as much as a sign heralding the essoufflement of the difference-blind approach.

Haut de page

Pour citer cet article

Référence papier

Mathieu Lizotte, « Gethin AMORY, Clara MARTINEZ-TOLEDANO, and Thomas PIKETTY (eds), Political Cleavages and Social Inequalities: A Study of Fifty Democracies, 1948-2020 »Revue européenne des sciences sociales, 61-1 | 2023, 279-284.

Référence électronique

Mathieu Lizotte, « Gethin AMORY, Clara MARTINEZ-TOLEDANO, and Thomas PIKETTY (eds), Political Cleavages and Social Inequalities: A Study of Fifty Democracies, 1948-2020 »Revue européenne des sciences sociales [En ligne], 61-1 | 2023, mis en ligne le 01 mars 2023, consulté le 01 juin 2023. URL : ; DOI :

Haut de page


Mathieu Lizotte

University of Ottawa, School of Sociological and Anthropological Studies

Articles du même auteur

Haut de page

Droits d’auteur


Creative Commons - Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International - CC BY-NC-ND 4.0

Haut de page
  • Logo Cahiers Vilfredo Pareto
  • OpenEdition Journals
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search