Introduction

Democratic innovations have expanded globally over the last thirty years, becoming a recognised practice of public administrations at different levels of the polity. The multiple crises facing representative democracy have triggered growing interest in these procedures, not least because of the ways in which they can expand the number and diversity of voices heard in political decision-making, open new channels of citizen influence and foster democratic skills of citizens (Geissel & Newton 2012; Smith 2009). These procedures, with their many democratic potentials, promise to increase the responsiveness of public administrations, cushion resistance to controversial policy decisions, improve the provision of services and, consequently, increase the legitimacy of public policies (Warren 2009).

The worldwide diffusion of democratic innovations has occurred in overlapping waves, which can be related to different models (and theories) of democracy. Around the turn of the century, for example, we saw the global expansion of participatory budgeting (PB) (Sintomer et al. 2016), which was strongly associated with the ideas of participatory democracy. In the last decade, we have witnessed the diffusion of deliberative mini-publics (DMPs), which has been closely connected to principles of deliberative democracy (Smith & Setälä 2018). The rapid expansion of DMPs, mostly across highly industrialized nations, has led the OECD (2020) to identify an ongoing ‘deliberative wave’. While a significant amount of research has been undertaken on the potential and limits of DMPs (e.g., Grönlund et al. 2014; Reuchamps et al. 2023), we have scant knowledge of the actors who promote them and, most importantly, on the influence different political parties may have on the commissioning of deliberative procedures (Gherghina & Jacquet 2022).

The role political parties play in the expansion of DMPs is crucial to understand whether the ‘deliberative wave’ is the product of a particular ideological orientation. Political parties with distinct ideological affiliations – and consequently different values and interpretations of democracy (Heywood 2012) – represent different understandings of what citizen participation is, its democratic purposes and its centrality to the functioning of representative democracy (e.g., Kittilson & Scarrow 2003; Verge 2007). Left-wing parties have traditionally been associated with advocating a greater connection of political decision-making with the citizenry (Fung & Wright 2001), an idea that is reinforced by evidence that leftist politicians have a more favourable attitude towards deliberative procedures than right-wing ones (e.g., Jacquet et al. 2022; Junius et al. 2020; Rangoni et al. 2021). Yet, with the exception of a few studies (Kübler et al. 2020; Saradín et al. 2022), the influence of party ideology on the actual commissioning of DMPs has been largely overlooked.

The importance of this variable in the diffusion of democratic innovations, however, has been confirmed in the expansion of PB. Research on PB shows that the leftist ideology of the governing party was crucial in its emergence and initial expansion, but became less relevant as the procedure diffused across the globe and became standardised (Pogrebinschi 2023; Ramírez & Welp 2011). Although the importance of party ideology seems to fade over time, ideology has continued to influence the characteristics of PBs in terms of design (Ganuza & Francés 2012; Sintomer et al. 2016). Where commissioners are left-wing parties, the procedures tend to have an institutional design that fosters greater deliberative quality and a greater capacity to empower participants. Identifying the effects of party ideology on DMPs thus appears crucial for enhancing our understanding of the development of the ‘deliberative wave’. It also provides valuable insights into how political actors interact with deliberative procedures and, more broadly, democratic innovations.

This article investigates the influence of governing party’s ideology in the commissioning of deliberative mini-publics. Are political parties with particular ideological affiliations more involved in the commissioning of DMPs? Does the institutional design of DMPs vary according to the ideology of governing parties? To begin to answer these questions, we briefly review the existing literature on party ideology and democratic innovations, before moving to an analysis of the OECD dataset of all known DMPs commissioned by public authorities in Europe over the last four decades. Our results, which provide the first descriptive insights of a large-N dataset of DMPs, show that the ideological pattern of adoption is less clear cut compared to other democratic innovations such as participatory budgeting. But stronger ideological differentiation emerges when we pay close attention to the design features of DMPs implemented.

Does Ideology Matter?

The political will of public officials often comes to the fore when the causal factors explaining the promotion of democratic innovations are discussed (e.g., Kübler et al. 2020; Ryan 2021). Although it is often suggested that politicians have strong reasons not to adopt these procedures or, at the very least, to distrust them (Thompson 2019), the literature on political representatives’ attitudes towards DMPs has shown that they are not all equally reluctant: left-wing politicians are more willing to launch these procedures, while those on the right are more reluctant to do so (e.g., Jacquet et al. 2022; Junius et al. 2020; Rangoni et al. 2021). This seems not only related to ideological self-placement. Research analysing party discourses has found equivalent results, with left-wing parties being the most vocal about increasing the means for citizen participation in policy-making (e.g., Kittilson & Scarrow 2003; Verge 2007). Leftist parties tend to emphasise the potential beneficial effects that these procedures could have on democratic goods such as inclusiveness, popular control and considered judgement – in contrast to the efficiency or legitimacy, which are more emphasised by right-wing parties. Despite this evidence, very few studies have investigated the extent to which the ideology of political parties influences actual commissioning of deliberative mini-publics. Rare examples include Kübler and colleagues (2020) who found no relationship between Left/Green vote share in Switzerland and DMP commissioning. They show that the prevalence of DMPs is higher in contexts with greater party fragmentation. In Prague (Czech Republic), Saradín et al. (2022) find that the commissioning of democratic innovations (including among them DMPs) is more related to the differences between traditional and new parties than to their ideology.

The results of these studies seem to suggest the lack of influence of party ideology on the commissioning of DMPs. However, these are very particular contexts. What would the results of a more comprehensive analysis look like? Furthermore, the main vein of scholarship around the diffusion of democratic innovations – focussed on the study of PBs – has shown the influence of the ideology of governing parties rather than electoral vote share. Drawing on the PB literature allows us to build some hypotheses about what we may find for the case of the commissioning of DMPs by ideologically different governing parties, while being aware of the potential differences that exist between different ‘families’ of democratic innovations (Smith 2009) and the potential complexities of the relationship between party ideology and the commissioning of these procedures that PB scholarship exposes.

The most well-studied set of PB procedures is those in Brazil. Several scholars demonstrate that the leftist Partido dos Trabalhadores’ (Workers’ Party, PT) played a crucial role in increasing the adoption of the PB at the local level (e.g., Avritzer & Wampler 2008; Spada 2014). The relationship between leftist parties and the commissioning of PB has also proved relevant in other Latin-American countries (Dias 2014). However, over a number of years, administrations affiliated to other ideologies increasingly responded to the diffusion of PB. According to both Ramirez and Welp (2011) and Pogrebinschi (2023), the left has lost the ‘monopoly’ of participatory democracy in Latin America. They show how, since the 2000s, an increasing number of political parties from the centre to the right (and without major distinctions between Neoconservatives and Liberals) have adopted various democratic innovations in the course of their administrations. While ideological commitment to participation is certainly not meaningless, time plays a moderating role on this relationship. International agencies and experts gradually joined the ‘PB wave’, standardising criteria for the implementation of these procedures and promoting them across governments of all ideological persuasions. Toolkits and professionals who design and implement these procedures enabled widespread adoption. In such a context where participation is widely facilitated by laws, international agencies and participatory consultants, the dissemination of procedures may surpass ideological commitments (Baiocchi & Ganuza 2017). In Latin America, the relevance of left-wing parties is therefore more evident in the initial stages of PB diffusion.

The adoption of participatory budgeting in Europe shows a similar pattern. In most European countries, the first steps in the diffusion of PB tended to be led by left-leaning parties. This is the case for Spain, Italy, and France, for example, where mostly leftist local authorities introduced PB (Sintomer et al. 2016). Once established, ideology gradually ceases to be important. In Italy, left-wing parties conducted the introduction of PB in the early 2000s. As they lost the subsequent municipal elections, PB almost disappeared, but by 2009 the ideology of municipal governments was no longer a decisive factor (Allegretti & Stortone 2014). A similar trend can be found in Spain, where leftist parties introduced PB in the early 2000s, and slowly conservative parties followed in their wake (Ganuza & Francés 2012). Exceptions to the general trend exist. Germany is an outlier: both Conservative and Liberal local governments were early adopters (Röcke 2014). Translating these findings to the case of deliberative mini-publics, and considering the existing research that suggests more positive attitudes of left-wing politicians to these procedures (e.g., Jacquet et al. 2022; Junius et al. 2020; Rangoni et al. 2021), we expect to find a comparable situation: that left-wing governments will have a greater presence in the early adoption of these procedures.

Literature on PB also shows us that it is not enough to simply look at the ideological position of the parties driving diffusion of a named democratic innovation; it is also necessary to look at the specific institutional design characteristics of commissioned procedures. Even with increasing standardisation of implementation (Baiocchi & Ganuza 2017), Sintomer and colleagues (2016) highlight differences in the PB models implemented in Europe and the differential role that political families have had in their commissioning. They identify six ‘ideal-types’, each of which differ in terms of origin, the organisation of meetings, the nature of deliberation and the weight of civil society in the procedure. In their analysis, they find that the ‘participatory democracy’ model is clearly linked to left-wing governments, both Social-Democratic and New Left. This model is characterised by openness to all citizens, with a strong role for them in decision-making and an emphasis on good-quality deliberation. The model aims to build links between conventional and non-conventional politics to enhance political impact and to help build a strong and autonomous civil society. The remaining models have been implemented almost equally among the different political families. These tend to have a more ‘administrative’ vocation – citizens are conceived of as consumers and participation as a way of scoping their preferences, with little opportunity to mobilise civil society. Hence, beyond its influence on the greater or lesser adoption of democratic innovations, the ideology of governing party can also influence the design characteristics of the procedures implemented. This is crucial in terms of the democratic goods of these procedures and for analysing DMPs’ potential impact on the functioning of our democracies. Unlike in the case of PB, no studies exist that have addressed how ideology influences the democratic characteristics of implemented DMPs.

Overall, we can thus say that the left seems more inclined to experiment with new procedures such as PB, while parties on the right of the political spectrum tend to replicate those (successful) democratic innovations. Once a given democratic innovation is established and its relative success makes it an attractive governance option, its adoption tends to be less influenced by ideological positions. As for the features of democratic innovations, following the evidence from PB research, the left appears to be more focussed on those procedures that have a high deliberative quality and a clearer impact on public policies; on those procedures that realise the democratic goods of considered judgement and popular control (Smith 2009).

Given the paucity of existing knowledge, this paper takes the significant step of offering the first large-N study on the influence of governing party’s ideology in the commissioning and institutional design of DMPs. Departing from the insights provided by the scholarship on PB, and aware of differences in context and design of the two democratic innovations, we seek to explore whether left-leaning parties across Europe influence DMPs’ commissioning, at least in the first stage of its diffusion, and whether party ideology affects the features of European deliberative procedures.

Results

Party ideology and time

How has the commissioning of deliberative mini-publics been influenced by party ideology over time? Figure 2 provides a general descriptive picture. Not only have most of these procedures been commissioned – in aggregate terms – under Social-Democratic governments (as also seen in Table 1), but early commissioning also appears to be influenced by governing party. Until the 2000s, most DMPs were clearly developed under Social-Democratic governments (26 against the 16 developed by Christian-Democrats), even more than the sum of all other political families. While a large number of DMPs have also been commissioned under Christian-Democratic governments, in most years their number is lower compared to that of Social-Democrats. The exception lies in the period 2001–2010 (Figure 3). This relative increase of Christian-Democrats cases appears to be more related to a lack of commissioning on the part of the Social-Democrats than to an increase by Christian-Democratic governments.

Figure 2

Commissioning of DMPs in Europe by party family and year.

Source: Own elaboration with data from OECD Database of Representative Deliberative Processes and Institutions (2021) (N = 226).

Figure 3

Evolution of DMPs commissioning (party family relative weight by 5-year periods).

Source: Own elaboration with data from OECD Database of Representative Deliberative Processes and Institutions (2021) (N = 226).

Figure 3 suggests three main periods of commissioning of DMPs in relation to party ideology.⁶ First, between 1979 and 2000, Social-Democratic governments in particular and Christian-Democratic ones to a lesser extent dominate the field of activity, with almost all DMPs commissioned under their rule. Second, between 2001 and 2015, Social-Democrats lose weight in relative terms, with the appearance of the Liberals (with most cases concentrated between 2001 and 2005). Finally, from 2016 onwards, Social-Democrats gain weight again, but with Liberals surpassing Christian-Democrats as the second highest party family and cases under New-Left/Green and Conservative rule growing from a low base.

Party ideology and time: Country specificities

But a closer look at the country level reveals the particularities of the diffusion process in different contexts. We show this by focusing on four countries: Germany, Denmark, Austria and the United Kingdom. Both Germany and Denmark took the lead in the first European ‘wave’ of commissioning (1979–2005, Figure 1), with 37 and 14 (out of 68) cases, respectively. Austria and the UK both developed DMPs at a later stage. Deliberative procedures have not been developed at the same pace or at the same time everywhere. To interrogate these national patterns and how they are influenced by party ideology, we look in more detail at these four countries.⁷

In Germany (Figure 4), we find a mixed picture, with almost the same number of DMPs developed under both Social-Democratic and Christian-Democratic governments (28 vs. 27 respectively). Before 2005, commissioning was higher among Christian-Democratic governments (21 cases) compared to Social-Democrats (15). All these cases are Planning Cells, designed by sociologist Peter Dienel at the University of Wuppertal (Hendriks 2005). After 2005, the numbers are inverted, as most DMPs were commissioned by Social-Democrats, with New-Left/Green governments commissioning equal numbers to Christian-Democrats (6). During the entire period analysed, all DMPs commissioned under Social-Democratic rule were Planning Cells, while Christian-Democrats experimented with other models after 2005 (e.g., Citizens’ Dialogues and Citizens’ Assemblies). New-Left/Greens have also experimented with diverse designs (e.g., Citizens’ Juries and Planning Cells). Nonetheless, most DMPs developed in Germany were Planning Cells (53 out of 61), despite some diversification in the second period.

Figure 4

DMPs commissioning per year and party family: Germany (N = 61).

The other early adopter, Denmark, shows a clear influence of Social-Democrat commissioning. This is true when considering early-commissioning and the whole period studied (19 cases, 63%) (Figure 5). Liberals are present throughout but with less weight. The Danish case is similar to the German in that almost all procedures commissioned before 2005 correspond to a specific DMP model developed in that country: in this case, the Consensus Conference invented by the Danish Board of Technology (Hendriks 2005). Most Consensus Conferences have been commissioned under Social-Democratic rule (10 out of 14). After that period, Citizens’ Dialogues gained more presence, with Liberal governments behind early cases of this sub-type of DMP. Conservative ruling parties are related to only one Consensus Conference and one Citizens’ Dialogue, but in both cases, they represent the first adoption of each model in Denmark.

Figure 5

DMPs commissioning per year and party family: Denmark (N = 30).

In Austria (Figure 6), a different pattern of adoption emerges, with Christian-Democratic governments commissioning the majority of DMPs during the entire period. The most frequent model implemented in the country is Citizens’ Councils (28 out of 30) – a model also known as Wisdom Councils, which were developed by Jim Rough in the US and then exported to Austria (Asenbaum 2016).

Figure 6

DMPs commissioning per year and party family: Austria (N = 34).

The UK is a late adopter (Figure 7). Although three DMPs were commissioned under different governing parties between 1999 and 2006, almost all procedures have been developed from 2017 onwards. Social-Democratic governments are major adopters, with almost two-thirds of total cases commissioned under their rule (19 cases, 61%). Four different models make up those cases, with Citizens’ Juries (9) and Citizens’ Assemblies (7) being the most common. Conservative and Liberal governments make up the remaining cases.

Figure 7

DMPs commissioning per year and party family: United Kingdom (N = 31).

In addition to these four countries, and among the countries that developed at least 10 DMPs, only France presents a clear party influencing early adoption and commissioning of such deliberative procedures: Social-Democratic governments, which have commissioned 63% of all DMPs.

Discussion

This paper offers the first large-N insights into how the ideological affiliation of governing parties influences the commissioning of DMPs, contributing to the emerging body of research on the influence of political parties on the implementation of deliberative procedures (Gherghina & Jacquet 2022). In particular, we have explored two types of influences: first, whether left-leaning parties across Europe are more involved in the commissioning of DMPs than other parties, at least in the initial stage of their diffusion; second, whether party ideology affects the design characteristics of European deliberative procedures.

In terms of diffusion of DMPs in general, we find that governments ruled by parties along all points of the ideological spectrum have, to a greater or lesser extent, commissioned DMPs. In aggregate terms, most procedures in Europe have been developed under Social-Democratic governments and they were the early adopters across the continent. At this level of generality, the influence of party ideology on the ‘deliberative wave’ is similar to that found for other democratic innovations – particularly for PB, for which we have most evidence. But as with other studies of the characteristics of the commissioning of democratic innovations (Baiocchi & Ganuza 2017; Dryzek & Tucker 2008), national context plays a mediating role.

Our results show that the diffusion of DMPs in Europe takes very different national forms when it comes to the influence of ideology. Denmark, the UK, and France reinforce the Social-Democratic prevalence – although not in all cases as early adopter. But Germany and Austria display a different pattern. In Germany, Christian-Democrats were a significant player alongside Social-Democrats, particularly in the early diffusion of the procedure. In Austria, Christian-Democratic rule has dominated the field. The status of Germany as an outlier has some resonance with PB diffusion, where Christian-Democrats were a leading adopter, running against the general trends across Europe (Sintomer et al. 2016). Compared to the diffusion of PB, we find less comprehensive ideological trends for early commissioning of DMPs in each country. The ‘deliberative wave’ has not been instigated everywhere by left-wing parties, which helps to understand why previous studies on DMPs diffusion have shown that the influence of ideology is limited (Kübler et al. 2020; Saradín et al. 2022). Our country-level analysis also helps illustrate the notable contribution that particular policy entrepreneurs have made to the dissemination of these procedures. For instance, the practice of DMPs in Germany was driven by the creation and diffusion of Planning Cells by Peter Dienel; in Denmark by Consensus Conferences developed and popularised by the Danish Board of Technology (Hendriks 2005). Future studies will need to address the role of these entrepreneurs in the country-specific patterns of commissioning of deliberative procedures, as well as that of other elements such as civil society mobilisation (e.g., Felicetti 2023) or previous participatory traditions (e.g., Röcke 2014). This would help to contextualise the patterns found here and to better understand when and why the ruling party’s ideology is influential.

Unlike in the expansion of PB, where New-Left/Green parties played a prominent role both as governing parties and as members of coalitions alongside Social-Democrats (Sintomer et al. 2016), we do not find that they influenced the ‘deliberative wave’ in the same way. This finding may be determined by three limitations of the data we use. First, we code party ideology based on the main political party in each jurisdiction. Lack of available data means we cannot know whether these parties were governing in coalition and with whom. It may be, for example, that the role of New-Left/Green parties is larger than we found if we were able to consider coalition partners. Second, a constraining factor of the OECD definition of DMPs is the application of sortition. It may well be that, under New-Left/Green governments, deliberative procedures are commissioned, but not ones that recruit participants using democratic lottery; instead, they could opt for reaching more vulnerable or usually under-represented social groups through other strategies. Third, the data collection for the OECD database ends in 2021. That only captures the beginning of a significant wave of climate citizens’ assemblies across Europe (Smith 2024). It is a reasonable assumption that many of these more recent assemblies have been promoted by New-Left/Green governments, whether as the main party or as coalition partner.

Beyond adoption and subsequent diffusion, we find a strong ideological pattern relating to DMPs’ institutional design. Using the OECD (2020) categorisation of purpose, which helps us to differentiate between DMPs with different procedural characteristics, we find that Social-Democratic governments are statistically more likely than Christian-Democratic ones to be involved in the commissioning of procedures that aim to generate informed citizen recommendations. This finding is confirmed by our analysis of size and length of DMPs. Under Social-Democratic rule, procedures with a higher deliberative quality that produce citizen-authored recommendations are more likely. In contrast, Christian-Democratic governments are more involved in the organisation of small and short procedures of a lower deliberative quality that aim to elicit opinions and preferences rather than crafted recommendations.

Our results reaffirm the idea that ideological orientation influences parties’ conceptions of citizen participation (e.g., Kittilson & Scarrow 2003; Verge 2007). More than that, they help to show its influence on political practice and underline that the study of party ideology is fundamental to understand the different democratic capacities of DMPs in particular, and of democratic innovations in general. Indeed, this left-right split in deliberative practice again resonates with the experience of PB where more empowered and deliberative designs tend to be related to left-of-centre governments across Europe (Ganuza & Francés 2012; Sintomer et al. 2016), although here we only find evidence regarding the deliberative component of the procedures. This simple narrative, however, is disrupted by the Conservative pattern, which follows the Social-Democratic trend: small but longer procedures as compared to Christian-Democrats. Whether this unexpected finding holds over time (given the small number of cases related to Conservative ruling parties and their more extensive presence in the UK) will only come to light as further DMPs are commissioned across Europe.

Finally, our analysis highlights the need to address the variations in the institutional design of deliberative procedures. The sheer diversity of these procedures calls for large-N approaches to comprehend how and under what conditions their designs differ. The OECD’s (2020) distinction between DMPs that aim to elicit informed citizen recommendations and those that elicit citizen opinion represents a potentially valuable starting point for categorising the deliberative quality of DMP designs, as we have shown. Nevertheless, as PB scholarship cautions, it is crucial to approach the study of institutional design with care, avoiding reliance on self-ascription of the name of democratic innovations (Sintomer et al. 2016). This type of analysis needs to be complemented with more ‘objective’ criteria, such as our application of the number of participants and the duration of the procedure. Future research may introduce additional internal or even external design features (Boswell et al. 2023), thereby expanding the scope of analysis, helping to refine DMP classifications in terms of their institutional design and allowing for the detection of ideological patterns that may not be discernible relying on currently available categorisations.

Conclusion

Does party ideology matter in commissioning democratic innovations? Extensive scholarship on the diffusion of participatory budgeting tells us that it does – both in terms of the ideological persuasion of governments that are early adopters and the way that PB is enacted. The ideological characteristics of the European ‘deliberative wave’ resonate to a degree with the earlier diffusion of PB across Europe. We find a left-wing disposition towards DMPs. Under Social-Democratic governments we find more early adoption as well as a higher amount of overall deliberative procedures. Furthermore, those procedures of higher deliberative quality are more likely to be found under Social-Democratic compared to Christian-Democratic rule. As with PB, we also find national peculiarities in the diffusion of DMPs, especially when it comes to early adoption. For example, Christian-Democratic governments play a leading role in the adoption and diffusion of deliberative procedures in Germany and Austria. Our findings need to be treated with care. Compared to PB, less DMPs have been commissioned. And our data does not allow us to investigate the ideological complexion of governing coalitions. In both cases, this may lead to an underestimation of the impact New-Left/Green and Conservative governing parties. Even with these caveats, our analysis is an important first step in understanding the importance of the ideological character of the ‘deliberative wave’.

Funding Information

The author(s) are grateful to the Spanish Ministry of Science for funding (grant PID2019-106731GB-I00). Rodrigo Ramis-Moyano is beneficiary of the University Teacher Training Program (FPU2019) funded by the Spanish Ministry of Universities. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Author Informations

Rodrigo Ramis-Moyano is a PhD candidate at IESA-CSIC and the University of Cordoba (FPU), focusing on the study of citizen participation, particularly on the quality and functioning of democratic innovations. He holds a degree in Sociology (Universidad de Alicante) and master’s degree in Research Methodology in Social Sciences (Universidad Complutense de Madrid) and is an active member of the DEMOCRETS research group. His research interests include political sociology, economic sociology, and research methodology.

Graham Smith is Professor of Politics at the Centre for the Study of Democracy, University of Westminster. He is a specialist in democratic theory and practice and chair of the Foundation for Democracy and Sustainable Development. His publications include ‘Democratic innovations: designing institutions for citizen participation’ (Cambridge, 2009), ‘Can Democracy Safeguard the Future?’ (Polity, 2021) and ‘Democracy in a Pandemic: Participation in Response to Crisis’ (University of Westminster Press, 2021).

Ernesto Ganuza is a senior researcher at the IPP-CSIC. He has been visiting scholar in Berlin (Centre Marc Bloch), Paris (CNRS) and Providence (Brown University). His research areas are democratic innovations and political sociology. He works also as a consultant in the organization of participatory or deliberative processes. He has published several essays and books on these topics, such as ‘The paradox of participation’ with Gianpaolo Baiocchi or ‘Democracy is possible’ with Arantxa Mendiharat.

Thamy Pogrebinschi is a senior researcher at the WZB Berlin Social Science Center and faculty member of the Berlin Graduate School of Social Sciences at the Humboldt University in Berlin. Her research focuses on democratic innovations in Latin America and beyond, examining their impact on policy and democracy at the macro level. She has been the coordinator of the LATINNO project, and her most recent publication is ‘Innovating democracy? The Means and Ends of Citizen Participation in Latin America’ (Cambridge University Press, 2023).