An Eclectic Political Scientist: A Tribute to Robert Elgie

Abstract

Robert Elgie's contribution to comparative politics is considerable. His work covers a range of subfields, including the study of political institutions, French politics, political leadership, presidentialism and semi-presidentialism. He became the leading scholar of semi-presidentialism and indeed defined much of the research agenda in this subfield.

Robert Elgie's contribution to comparative politics is considerable. His work covers a range of areas, including the study of political institutions, French politics, political leadership, presidentialism, semi-presidentialism and democratic transition. He became the leading scholar on semi-presidentialism and indeed defined much of the research agenda in this subfield.

Last but not least, Elgie dealt with the definition, redefinition and measurement of new concepts in political science. He provided innovative and ground-breaking analyses of presidential power, as well as divided government, and presidentialization.

Semi-presidentialism

Elgie made seminal contributions to the definition of semi-presidentialism as well as conducting a wide range of empirical research on how this system operates in practice. I will consider his work on defining semi-presidentialism before reviewing his contributions to the empirical study of the concept.

‘Semi-presidentialism is where a constitution includes a popularly elected fixed-term president and a prime minister and cabinet who are collectively responsible to the legislature’ (Elgie 1999). This definition has become widely accepted among scholars as it sets the minimal features of a semi-presidential regime. Elgie's proposal has the virtue of leaving aside presidential features that, being difficult to measure, would have generated impressionistic or arbitrary approaches. Instead Elgie's more streamlined approach allowed for easier empirical testing, offering a contrast to Maurice Duverger's original definition of semi-presidentialism. For Duverger, a semi-presidential regime is one where:

1. 1. the president is elected by popular vote;

2. 2. the president possesses considerable powers; and

3. 3. there also exist a premier and a cabinet, subject to assembly confidence, who perform executive functions. (Duverger 1980: 161)

The indication of ‘considerable’ powers possessed by the president in Duverger's definition left room for different approaches, different ‘measurements’ and finally different classifications. Therefore, depending on the scholars and the research design they adopted, one country could be either included or excluded from the list of semi-presidential regimes. Duverger's definition was thus unable to cover all the variations in semi-presidential countries and was weak methodologically. This was the danger of not having mutually exclusive cases, where different presidents could be included in one category or another, simply depending on the scholar's attitude or approach.

Elgie's definition is more parsimonious and does not exclude internal variation in the wider semi-presidential category, but simply refers to specific studies measuring the oscillation in the balance of power, the weight of presidential powers, the role of the prime minister and that of the president. From this perspective, Elgie clearly shows his interest in the definition of semi-presidential subtypes, writing on his semipresidentialism.com website: ‘I like the distinction between premier-presidential and president-parliamentary forms of semi-presidentialism. This distinction was first made by Matthew Shugart and John Carey.’¹

Here he is referring to Shugart and Carey's seminal book (1992), which was important both in Elgie's intellectual training on the topic and for his further elaboration of the above-mentioned definition of a semi-presidential regime. Shugart and Carey felt that the ‘semi-presidential’ regime as proposed by Duverger needed a conceptual specification to underline the nuances that appear between different cases and over time. They proposed, therefore, two definitions for as many semi-presidential subtypes: ‘A premier-presidential regime is, as its name implies, one in which there is both a premier (prime minister), as in a parliamentary system, and a popularly elected president’ (Shugart and Carey 1992: 23). The focus is on the fact that the internal balance between the two heads of the executive power can shift, and in fact Shugart and Carey insist that ‘the president under premier-presidentialism is not necessarily the “chief” executive, but rather must coexist with a premier, who is head of the government. As we shall see, the relative status of each office within the executive may vary both across and within regimes’ (Shugart and Carey 1992: 23). On the other hand, in terms of the second subtype, Shugart and Carey report that the term ‘president-parliamentary’ captures a significant feature of the regime: the primacy of the president, plus the dependence of the cabinet on parliament. Such a regime is defined thus:

1. 1. The president is elected by popular vote.

2. 2. The president appoints and dismisses cabinet ministers.

3. 3. Cabinet ministers are subject to parliamentary confidence.

4. 4. The president has the power to dissolve parliament or legislative powers, or both.

Elgie (1999, 2011) provided a clear, effective and parsimonious contribution to the concepts and definitions of the two semi-presidential subtypes. His definitions leave less room for interpretation, arbitrariness or scholars’ approach bias:

1. 1. President-parliamentarism is a form of semi-presidentialism where the prime minister and cabinet are collectively responsible to both the legislature and the president.

2. 2. Premier-presidentialism is a form of semi-presidentialism where the prime minister and cabinet are collectively responsible solely to the legislature.

More than a quarter of a century later, Elgie's article reviewing Shugart and Carey's section on semi-presidentialism (Elgie 2020) adds further specifications to his view on the regime type and on the subtypes. He agrees with the definition of the two semi-presidential subtypes but he also aims to foster comparative investigation within and between the two groups of countries.

In his contribution to a special issue of Political Studies Review dedicated to Shugart and Carey's 1992 Presidents and Assemblies, Elgie makes an ‘intellectual history’ of the concepts of premier-presidentialism and president-parliamentarism (Elgie 2020). Through a ‘deep reading’ of the works of Matthew Shugart, Elgie shows how the two concepts have been adopted over time and he indicates two main phases of interpretation. Initially, premier-presidentialism was understood as a synonym of semi-presidentialism. Since roughly 2005, however, it has been used as one of the two subtypes of semi-presidentialism. With intellectual parsimony, Elgie discusses the advantages and the criticisms of Shugart and Carey's conceptualization and specifications of semi-presidentialism subtypes. He highlights the virtues of the book, and at the same time he elegantly criticizes the two authors. An example of this is when Elgie addresses the most frequently discussed ‘Duverger problem’ (that is, the measurement of presidential powers), and he stresses that ‘it is moot as to whether they [Shugart and Carey] were ever guilty of the “Duverger problem” in Presidents and Assemblies. All the same, there was a certain conceptual clarification in this regard over time’ (Elgie 2020: 17).

Together with the theoretical assumptions and thoughts, Elgie also discusses the empirical evidence to support his arguments and compares the outcomes over time in studies of premier-presidentialism and president-parliamentarism (Elgie 2020: 22‒23). Elgie suggests what could be the future research agenda in dealing with the study of semi-presidential regimes. In particular, he stresses the importance of analysing cohabitation in the premier-presidential context, as under president-parliamentarism regimes there is less likely to be cohabitation. And it is precisely the role of cohabitation that brings us to Divided Government (2001), edited by Elgie, whose analysis and theoretical framework have become a standard for those undertaking research on the topic.

Elgie also underscores the importance of David Samuels and Matthew Shugart's book (2010) in rebooting the research agenda in comparative politics. In particular, he emphasizes – in agreement with Samuels and Shugart ‒ that together with the institutions that are able to shape political party organization, ‘party organization might be the “missing variable” in studies on the effect of regime types’ (Elgie 2011: 408; Passarelli 2015, 2020). In this way, Elgie connects the study of semi-presidentialism as a political and institutional regime with political parties. He does not neglect the relevance of the ontological structure of political parties, although he emphasizes the role of the institutions in affecting the role of political actors.

Elgie's standard definition of semi-presidentialism is a constitutional one. It stresses a taxonomic approach and can be considered a post-Duvergerian definition of semi-presidentialism as a country's president ‘does not have to have a certain amount of power … in order for that country to be classified as semi-presidential’ (Elgie 2011: 397). His research on the topic is strongly related to his early studies on the different aspects of the French political system, as noted by Ludger Helms (2019). The French case and his residence in Paris helped Elgie (who was fluent in French) to question whether the French Fifth Republic truly represented an exceptional case, an ‘archetype’ in his own words, or if there was something else under the surface that needed to be studied from a comparative perspective.

That question was a long-standing source of interest for Elgie. His work dealing with the concept dates back to the early 1990s, when his first article using the term ‘semi-presidentialism’ was published in 1991, shortly after the end of France's first experience with cohabitation, and over the next seven years he published three other pieces. The first two articles analyse the role and the power of the prime minister in France, as well as their resources, both political and constitutional. Both the presidential influence and the prime minister's activities depend on the personality involved and different areas of policy (Elgie and Machin 1991). In this regard, Elgie identified three different types of political leadership: pure presidential government, limited presidential government and prime ministerial government (Elgie 1992).

The third article in this series on semi-presidentialism (Elgie 1997b) proposes a theoretical framework for studying the executive power relations in both parliamentary and semi-presidential regimes. Here, Elgie offers his already defined scientific identification based on the comparison, the analysis of power, the political leadership and the institutions. Elgie establishes a framework for comparing executive branch power relations and identifies six models of executive politics, including a comprehensive set of ways in which power may be distributed among chief executives, cabinets, ministers and bureaucrats. It shows how Elgie's thoughts on semi-presidentialism were maturing: he is arguing that it is necessary to engage in empirical observation to determine which models of government occur and to identify the reasons why these models emerge. It is also worth mentioning that the article was nominated as one of the best articles published in Political Studies in the 1990s.

The fourth article (Elgie 1997a) represents in a sense a kind of unique contribution in Elgie's bibliography as it tackles the electoral system, a topic not central to his scientific production and intellectual speculation. However, even here Elgie shows his ability to analyse the context and to connect the electoral system to the role and impact of French institutions. In 1998, he analysed in depth the classification of regime types. The most important point that he raises is that the classification of regime types should be made with reference to dispositional properties alone (Elgie 1998).

These works were the prologue to the seminal book that Elgie edited and that represents the first scientific comparison of semi-presidential regimes. Covering 12 European cases presented by country experts (Elgie focused on France, of course), Semi-presidentialism in Europe (Elgie 1999) represents the first systematic and complete theoretical framework of semi-presidential regimes. Starting from this experience, he further develops and improves the analysis and the definitions to expand the comparison and the research.

Elgie's website, www.semipresidentialism.com, represents a rich source of information and analyses carried out over more than a decade. The blog not only provides reports on individual countries, but it has also become a point of reference for all those studying institutions, democratic transitions, democracy and presidents. Scholars can find detailed information there on the various forms of government across the countries of the world over time since the end of the Second World War – and even before, in some cases.

After that first book on semi-presidentialism, which basically covered Europe, Elgie spread his research interest to the countries of the recent wave of democratization and he went well beyond the Western European borders. In 2008's Semi-presidentialism in Central and Eastern Europe, co-edited with Sophia Moestrup, the focus was on the Central and Eastern European countries that adopted a semi-presidential system after the democratization process of the 1990s. This book – in line with Elgie's definition – demonstrates that semi-presidential regimes ‘can operate in quite different ways, some with very strong presidents, some with strong prime ministers and ceremonial presidents and some with a balance of presidential and prime ministerial powers’ (Elgie and Moestrup 2008). In light of the importance of the state-building process and the institutional and constitutional choice, Elgie and Moestrup discuss the advantages and disadvantages of semi-presidentialism and provide a theoretical framework for further comparisons.

In 2011 Elgie, Sophia Moestrup and Yu Shan Wu co-edited Semi-presidentialism and Democracy. In this book the editors and their collaborators systematically study semi-presidentialism in each region of the world where this type of regime is significantly present (Elgie et al. 2011). They compared the then more important countries that had a semi-presidential form of government, including all the cases from Europe to Africa, and Asia. Elgie's focus on variations among semi-presidential regimes – depending on the context, the personalities involved, and the extent of power exercised by the president and the prime minister – is evident again here.

Finally, in 2016, Elgie and Moestrup expanded the geographic pattern to the post-Soviet context and investigated the relationship between semi-presidentialism in the formal constitution and the verticality of power in reality. Their co-edited Semi-presidentialism in the Caucasus and Central Asia explores the extent to which semi-presidentialism has been responsible for the relative performance of democracy in each country. The main findings confirm ‘the importance of institutional variation within semi-presidentialism’ (Elgie and Moestrup 2016). In fact, the degree of democracy ‘has varied greatly both generally from one country to the next and over time within a number of countries’ (Elgie and Moestrup 2016: 2). Basically, by showing the variation based on differences in types of semi-presidentialism, they provide a theoretical foundation for the claim that ‘institutions can make a difference both in democracies and in authoritarian and competitive authoritarian regimes’ (Elgie and Moestrup 2016: 219).

By adopting a comparative approach, Elgie contributes enormously to our understanding of the process of regime change and democratic transition. In two books dealing with semi-presidential regimes, Elgie disentangles the research question of the impact of the institutions on democracy. This is an established research field in the literature but new in regard to presidentialism and parliamentarism. From a different theoretical perspective, Elgie asks whether semi-presidential institutions have a negative impact on democracy (Elgie 2011; Elgie et al. 2011). The book co-edited with Moestrup and Wu (Elgie et al. 2011) represents an attempt to test the influence of the semi-presidential regime in newly democratized countries. The contributors find that semi-presidentialism does represent a threat to democracy. This finding is important; the spread of semi-presidential constitutional arrangements has been considerable in recent decades. That spread has extended to new democracies in particular (currently more than 50 countries). In fact, ‘for newly democratizing countries, semi-presidentialism typically appears as a convenient constitutional compromise between proponents of a presidential system, on the one hand, and supporters of a parliamentary type of government, on the other’ (Elgie et al. 2011: 265). In particular, the contributors discuss the balance of power internal to semi-presidential regimes, and they expect greater institutional evolution within the universe of semi-presidential countries (Elgie et al. 2011: 272).

In this line of thinking, Elgie's books and articles contribute enormously to dismantling the idea that democracy is intrinsically related to the parliamentary regime. Beyond Juan Linz's (1990) approach, which identified the perils of presidentialism, Elgie demonstrates that – especially in new democracies – the major obstacle to full democratic transition is not the adoption of direct election of the president, and semi-presidential regimes (e.g. the African context) do not represent a problem per se. Rather, the problem with democratic transition can be represented by the internal imbalance of power if the president has vast power over the cabinet and the parliament alike (as is the case with president-parliamentarism).

In Semi-Presidentialism: Subtypes and Democratic Performance (2011) Elgie proves that the advantages of semi-presidentialism are outweighed by the disadvantages associated with cohabitation and divided minority government. The comparison covers 10 countries across all continents in order to include cases from both premier-presidential and president-parliamentary subtypes. This comparison sets the research agenda for the next generation of semi-presidential studies. In fact, he questions the relationship between presidential power and the form of semi-presidentialism. The theoretical circle was closed (Elgie 2011).

Elgie's name has become synonymous with semi-presidentialism for many scholars. His definition has become the standard for comparative research. Nevertheless, his intellectual contribution is much greater. Elgie published many important works on semi-presidentialism, but he also wrote extensively about political leadership – which is at the base of the relationship between the president, the prime minister and the parliament – and specifically, about presidential power.

Presidential power

Probably because of the ‘Duverger problem’, Elgie also approached the study of presidential power. In doing so, he set the agenda for further research and proposed a new approach that has become one of the most widely quoted works due to its theoretical influence and intellectual clarity. Together with David Doyle, Elgie proposed a new measure of presidential power (Doyle and Elgie 2016). Published in the British Journal of Political Science, the article soon became a point of reference for those interested in the study of leadership and presidents. The idea behind this approach was to overcome methodological problems stemming from the arbitrariness of defining a president as strong, weak or merely ceremonial. Doyle and Elgie generated a new set of presidential power scores including several countries and for a longer time series than in other studies. From a methodological point of view, the authors maximized the reliability of the presidential power scores ‘by deriving them solely from measures based on constitutional indicators of presidential power’, in order to avoid the idiosyncrasies of country scores in existing measures. This latter approach has the clear advantage of incorporating new scores when the constitutional provisions change, as well as expanding the number of case studies. It was a clear challenge for the research agenda in the field.

Elgie, together with Cristina Bucur, Bernard Dolez and Annie Laurent (Elgie et al. 2014), examined the impact of direct presidential elections on legislative party systems and argued that presidential power shapes the effective number of presidential candidates in countries with direct presidential elections. They reported that ‘presidential power itself helps to determine the ENPC by encouraging parties to behave strategically but only in a way that we can clearly observe within such an intermediary range’ (Elgie et al. 2014: 475). It is worth mentioning that the authors concluded that ‘the effect of the [presidential power] only applies to countries with direct presidential elections. … [W]e suggest that the effects under consideration should be tested solely in countries with directly elected presidents’ (Elgie et al. 2014: 475). In contrast, the relationship between the coat-tail effect and presidential power seems to be less evident and needs further investigation.

After his study on the definition of semi-presidentialism, Elgie continued to emphasize measurable concepts, clear indicators and epistemologically grounded theories not only when writing an article, an essay or a book, but also in general in dealing with comparative politics. Otherwise we would simply have opinions, he said. Thus, Elgie investigates ‘presidential power’ as one of the main arguments of his intellectual speculation. Presidential powers were at the core of much of his research interest: leadership, semi-presidentialism, the democratic performance of different regimes and the concept of personalization and presidentialization. In dealing with presidential prerogatives, Elgie attempted to provide not only a methodological tool that is replicable and verifiable but also a conceptual contribution that does not lead to individual interpretations.

Elgie was always looking for more ambitious attempts to define new concepts and deal with political phenomena. One of his last contributions was a co-edited book with Alexander Baturo ‒ The Politics of Presidential Term Limits (Baturo and Elgie 2019) – in which he dealt with the formal limits to presidential powers. This comparative study aimed to cover the lack of a comprehensive understanding of the role and effects of presidential term limits because term lengths are one of the most fundamental institutions of democracy. Covering both theoretical and empirical perspectives, the book explains the factors behind the different characteristics of presidential term limits. In particular, the contributors focused on the consequences of changes to presidential term limits.

Political leadership

Elgie's study of leadership represents a compendium of philosophy, history and political psychology. Adopting a multidisciplinary approach, he followed the works of Richard Fenno and Jean Blondel, among others, in dealing with the study of political leadership. His first contribution was published in 1995 and proposes the analysis of political leadership in liberal democracies. There, Elgie underlines the opportunities and the constraints that political leaders face in political systems. Specifically, he compares six cases (Germany, France, Britain, the US, Italy and Japan), with an emphasis on the relationship between political leadership, the political environment and the institutional framework (Elgie 1995).

One of the most intriguing conclusions is that the role and the effectiveness of political leadership depend mainly on the interaction ‘between leaders and the leadership environment with which they are faced’. As Elgie adds, ‘the nature of this interaction determines the ways in which, and the extent to which, heads of state and heads of government are able to affect the outcome of the decision-making process in a country’ (Elgie 1995: 181). In this statement Elgie echoes Blondel's argument that ‘leadership cannot be divorced from the environment within which it occurs’ (Blondel 1987: 321).

Elgie puts a lot of effort into analysing the so-called ‘interactionism’. The first chapter of his book Studying Political Leadership (Elgie 2015) is devoted to research of the origins of the new paradigm and its limits. With his usual elegance, both in presenting concepts and defining the problems involved, Elgie reflects on more than a century of studies of political leadership. Through a systematic analysis and categorization of the major contributions, Elgie was able to disentangle the most important theoretical problems in studying political leadership. He was also able to unlock different and yet related areas of knowledge. Elgie reviews studies of leadership and then systematizes the main contributions, dividing them into three categories: positivism, constructivism and political psychology. He has tried to clarify the study of leadership, moving from ontological approaches to proposing specific – differentiated – foci on each aspect. Thus, he emphasizes that:

studying the political leadership from a philosophically informed heuristic, we have been able to demonstrate both the fundamental differences between particular sets of studies of political leadership, but also the fundamental similarities within each set of studies as well. Thus, we have imposed a degree of coherence on the study of political leadership. (Elgie 2015: 173)

Although he reports that studies of political leadership have discovered that the personalities involved were responsible for about 20% of the changes in attitudes at discussion meetings, Elgie claims that the philosophical assumptions made by scholars who study leadership must be identified.

In a second book on political leadership (Elgie 2018), Elgie builds upon the conclusion of his previous research. In Political Leadership, he provides exactly what he noted was required in the first book, a clear philosophically informed heuristic choice. The book provides a philosophically informed, institutionalist account of political leadership that is well rooted in the American philosophical tradition and privileges the study of institutions as a cause of leadership outcomes. After two chapters presenting the topic and ‘making sense of the world’, Elgie presents two important empirical chapters: the first – together with David Doyle – focusing on the psychological effects of presidential institutions, and the second on the factors increasing the likelihood of having accountable political leadership. The last two chapters go back to Elgie's first and most studied topics: cohabitation and France. Hence, in this book, the triptych of institutions, political leadership and political theory was restored as in Elgie's other works.

The studies of political leadership, as we have seen, date back to Elgie's first book on the comparison of institutions in several countries, and especially on the French Fifth Republic. Moreover, he also analysed the topic by publishing in outlets other than his own books. In a chapter in a volume edited by Ludger Helms (2012), Elgie examines the differences in political leadership between old and new democracies. Using an institutionalist approach, Elgie underlines that different institutions provide incentives for actors to behave in different ways (Elgie 2012). He operationalizes the concept of political leadership as the decision-making authority of the president and/or the prime minister in presidential and semi-presidential regimes. Although he includes evidence that patterns of decision-making are equally recognizable in both old and new democracies in countries that directly elect their president, Elgie emphasizes that there is ‘always room for particular patterns of decision-making authority in any particular country’ (Elgie 2012: 273).

Moreover, in his contribution to a special issue of Daedalus dedicated to political leadership, Elgie explores aspects of the relationship between political leadership and institutional power, comparing the different forms that presidential institutions have taken across the world and identifying the relationship between these structures and social, political and economic outcomes. And here he addresses the role – still overlooked – of personal leadership. Elgie notably concludes that ‘in the interaction between institutions, leaders, and context lies the eternal dilemma of the study of presidential leadership’ (Elgie 2016: 68).

To summarize, Elgie's works on political leadership represent a fruitful attempt to systematize the theoretical frameworks, as well as to introduce innovative methodological approaches to measure political leadership in different contexts. From a theoretical point of view Elgie's work on political leadership is innovative as he clarifies that the scholarship on leadership outcomes is founded on certain ontological and epistemological assumptions. Once he has defined the set of studies that can be compared, he moves to the institutional approach as he demonstrates that institutionalism provides us with the opportunity to generate well-settled beliefs about the effect of leadership institutions on various outcomes in the political systems. Consistently, the empirical analysis allowed Elgie to adopt a multimethod approach, varying from a laboratory experiment, to large-N cross-national observational study, as well as qualitative comparative analysis.

French politics

Elgie also extensively investigated French politics and institutions. As regards Elgie's studies on French politics, it is worth mentioning here his role in 2003 in founding (together with Andrew Appleton) and leading for 17 years the journal French Politics, which he rooted in the community of political scientists. Thanks to his leadership, the journal continues to thrive. Moreover, the co-editing of the Oxford Handbook of French Politics – together with Emiliano Grossman and Amy Mazur (2016) ‒ confirmed the influence of Elgie in analysing French politics and institutions, thereby contributing to making the study of ‘the Hexagon’ – as France is often called – less parochial than it has been in some periods. The record of Elgie's publications on France is vast and lasted throughout his career. Between 1993 and 1996 he published three books covering the French case, either alone or in a comparative perspective. Beyond the book on the role of the French prime minister (Elgie 1993), Elgie included France in his seminal book comparing political leadership in liberal democracies (Elgie 1995) and then coordinated a research group focusing on the French presidential elections of 1995 (Elgie 1996).

In 2000, Elgie edited a volume investigating The Changing French Political System, which looks at the new challenges and the responses the (semi-presidential) institutions were providing, especially in terms of political parties and political mobilization (Elgie 2000). The same topics and themes were also investigated in a book co-authored with Stephen Griggs where the authors emphasize the challenges faced by the French political system, its institutions and leadership, focusing in particular on the relationship between the prime minister and the president (Elgie and Griggs 2000). The study of French politics is at the core of many of Elgie's articles and chapters, although with less frequency from 2000 onwards (Elgie 2000, 2003), as he intensively published more on semi-presidentialism, institutions and presidential politics.

Presidentialization

The presidentialization of politics was one of the last projects started by Elgie. The topic represented a perfect synthesis of Elgie's research interests; in fact, in studying presidentialization he could deal all at once with political leadership, the institutional regime, political concepts and the empirical and methodological approaches used to measure it. The first systematic attempt to deal with the phenomenon, it was, as usual in Elgie's approach, a theoretical review and analysis of it. He was interested in depicting all the possible biases, the main theoretical contributions, the schools of thought and the way to converge on a common definition. Only from that is it possible to move on to measurement, comparison and data analysis.

In his article on presidentialization co-authored with Gianluca Passarelli (Elgie and Passarelli 2019), Elgie focuses on the two main contributions to the contemporary academic debate about the term ‘presidentialization’, namely the book by Samuels and Shugart (2010) and the work of Thomas Poguntke and Paul Webb (1995). The article's authors try to disentangle the ways in which the same term is applied in the two studies. The intent is not only to evidence differences and similarities but also to consider how each approach operationalizes presidentialization in a different way. Only after having clarified that we are dealing with two (to some extent opposite) approaches is it possible to pave the way to detailed and independent research agendas. The authors suggest that even though Samuels and Shugart (2010) and Poguntke and Webb (1995) use the same term, they are engaged in quite distinct exercises. Samuels and Shugart are concerned with studying the specific impact of constitutional presidentialization on party presidentialization while Poguntke and Webb present a grand narrative arc that addresses the development of a somewhat ineffable political idea. Without claiming that one of the two is better than the other, Elgie and Passarelli argue that it is important to underline the theoretical and methodological differences between them to allow the debate about presidentialization to move forward on a firmer footing.

In the same theoretical line, Elgie (again with Passarelli) reports the main contributions to the debate on presidentialization, and the differences with the concept of personalization (Elgie and Passarelli 2020). The idea – as Elgie has done with the concept of political leadership – is to analyse in depth the theoretical pillars, as well as the weaknesses and the biases, of the political phenomenon under investigation. Depicting the most important contributions – both theoretical and empirical – to the debate on the differences between presidentialization and personalization, which is a flourishing field of investigation, is in fact crucial. Elgie and Passarelli (2020) identify the four main questions that are asked in the contemporary study of presidentialization: (1) what is meant by the term and how should it be defined? (2) what are the sources of presidentialization? (3) to what extent has there been a presidentialization of contemporary polities in practice? and (4) what are the consequences of presidentialization? The authors find that if we consider presidentialization to be a general process and personalization to be merely part of the electoral face of presidentialization, then there is the potential for systemic presidentialization to have an effect on voter expectations, perhaps in terms of attitudes towards what political leaders can be expected to achieve, or the electoral mandate that they are given.

In 2018, Elgie and Passarelli had launched an expert survey to account systematically for variation in the change of level of presidentialization in different countries. This was (and still is) Elgie's attempt to set the new research agenda in the field of comparative politics by focusing on the relatively new concept of presidentialization.

The scientific legacy of an eclectic intellectual: Robert Elgie, researcher and gentleman

Complementing his academic activity, Robert played a role as an intellectual contributing to many fields. On the scientific side, he contributed greatly to the ECPR standing group on presidential politics. Without Robert's support, the group would have remained simply an idea of mine, valid perhaps, but it would have stayed in the world of good intentions. He was always keen to help senior and junior colleagues who had a humble attitude and were trying to learn from others.

Robert also established a blog on ‘presidential politics’, https://presidential-power.net. This was not conceived as a simple blog, although he never presented it as anything more pretentious. Nevertheless, thanks to his hard work and determination, it became, and remains, a rich source of information, using scientific language while also being accessible to the general public.

Robert contributed to other ventures as well. He joined the board of Government and Opposition in 2013 as Reviews Editor, innovating the way of conceiving the ‘book reviews’ and, as mentioned, was co-editor of French Politics. Last but not least, Robert co-founded and launched the Palgrave series in ‘Presidential Politics’. When I submitted my idea to him, he had already started working on it with the publisher, so he kindly offered me to co-edit it as we had the ‘same’ idea. Another example of genuine kindness and uncommon academic cooperation.

Robert would have certainly gently contested the title of this article and of this last section. He was a terrific scholar, capable of producing hundreds of articles, books and chapters that were always of a high standard without ever being pedantic or banal. Although he was modest, Robert was capable of exercising leadership (yes, the very thing he loved studying) and charisma to coordinate small and large groups of colleagues without being arrogant, and all the while focusing on the success of the common enterprise.

I have always thought of Robert as a man beyond the ordinary, a gentleman of yesteryear. In a context in which arrogance and mediocrity often flow, I was impressed by his calm, his patience, his elegance in dealing with harsh criticisms, negative comments or reviews, or even simply rude behaviour. He never lost his smile, even at moments when a harsh word was almost called for. Robert was a very professional scholar, attentive to all aspects of research and publication, from deontology to the research question, the methodology and even aesthetic aspects. He took great care over his choice of covers for books, and only while I was writing this piece did I realize how meaningful the cover of his book on leadership was (Elgie 2015). That cover presents a collage of different question marks distributed randomly; well, for Robert, the questions on the meaning of concepts, variables, ideas and the possibility of measuring them were at the core of his scientific investigation.

Robert's research interests were vast, and he was able to find an intellectual stimulus in all of his activities. He was expert in different fields and he even took his hobbies extremely seriously. He never made it obvious that he knew a lot of things, but his interlocutors used to learn quickly that he had in-depth knowledge of field X or Y. Moreover, Robert was able to speak good Italian, even better than many for whom it is their mother tongue. He knew art and sport: he was a keen and skilled cricketer and was fascinated by Italian cooking, always offering a philosophical interpretation of it. In 2016 Robert even measured (in an only semi-joking way) the performance of each regime type in the occasion of … the Olympics! The parliamentary regimes finished first, although the semi-presidential ones performed pretty well too.²

In 2017, Professor Robert Elgie was elected a member of the Royal Irish Academy. Even then, he never bragged about his success but remained humble. Even the President of Ireland appreciated his scientific works, and Robert also became involved with other political scientists, standing out as a true eclectic intellectual with a deep knowledge of the topics in which he specialized.

I have personally lost a dear friend and a great mentor. Ciao.