Abstract
In some democratic contexts, there is a strong aversion to the directive, individualistic and masculine expressions of leadership that have come to dominate the study of political leadership. Such leadership is antithetical to consensus democracies in parts of continental Europe, where the antipathy to leadership has linguistic, institutional as well as cultural dimensions. Political-administrative and socio-cultural contexts in these countries provide little room for heroic expressions of leadership. Consequently, alternative forms of leadership and associated vocabularies have developed that carry profound practical relevance but that have remained underexplored. Based on an in-depth mixed-methods study, this article presents the Dutch mayoralty as an insightful and exemplary case of what can be called ‘bridging-and-bonding leadership’; it provides a clear illustration of how understandings of democratic leadership can deviate from the dominant paradigm and of how leading in a consensus context brings about unique practical challenges for office holders. The analysis shows that the important leadership task of democratic guardianship that is performed by Dutch mayors is in danger of being overlooked by scholars of political leadership, as are consensus-oriented leadership roles in other parts of the world. For that reason, a recalibration of the leadership concept is needed, developing an increased theoretical sensitivity towards the non-decisive and process-oriented aspects of the leadership phenomenon. This article specifies how the future study of leadership, as a part of the change that is advocated, can benefit from adopting additional languages of leadership.
A multidimensional aversion to leadership
In the Netherlands, ‘leadership’ is a tainted concept, which often leads the Dutch to approaching the language of leadership with particular hesitancy. Although the phenomenon of leadership does exist also in the Netherlands, it is preferably not referred to by use of the word ‘leadership’ (‘t Hart and Ten Hooven, 2004). Some leading Dutch scholars even advocate doing away with the use of the word altogether: We should do away with the term ‘leadership’ … In the same way as there are places that are guilty – Auschwitz, Sobibor, the Killing Fields of Cambodja –, there are also words that are guilty. … One could be inclined to suggest that these words are simply just that: words, but that would be very naïve. Words are not guttural sound or printed letters. Words are parts of sentences, and sentences, in turn, are parts of thoughts and of nutty ideas (Zijderveld, 2010).
For cases where politicians and scholars cannot avoid the use of the leadership concept altogether, they have developed an almost euphemistic vocabulary of leadership that uses phrases such as ‘small-l leadership’, ‘facilitative leadership’ or ‘connecting leadership’ to describe the nature of political leadership in the Netherlands (Noordergraaf, 2006; ‘t Hart and Ten Hooven, 2004). In fact, the Dutch word ‘leiderschap’ is often put into contrast with the English word ‘leadership’ to highlight the alleged differences between the two conceptions of the leadership phenomenon (Hendriks and Tops, 2000). The Dutch discourse of leadership, thus, is one of caution.
There is also an important cultural dimension to the aversion to leadership in the Netherlands. Language and culture being inseparable (Jepson, 2010: 428), the cultural and linguistic dimensions are, of course, closely related, but they point to subtly different phenomena. It is not only the vocabulary of leadership that is approached with caution in the Netherlands, it is also its expressions in practice (Andeweg, 2000; Lijphart, 1999). The Dutch tend to dislike strong personalities (Hendriks and Tops, 2000) and traditionally have an aversion to notions of headship and followership (De Beus, 2006). The country is not used to cherishing ‘great leaders’, and frontrunners usually are not granted much room to manoeuvre (Cerfontaine, 2005). Rather than as ‘messiahs’, leaders are seen as people that one needs to be suspicious of, or as passers-by at best (‘t Hart, 2000; see also Judd, 2000). The ‘coincidental’ leaders that happen to be in office, then, are strictly scrutinized (Frissen, 2009). The fact that leadership, and especially strong, directive leadership, is generally mistrusted in the Netherlands, shows in the country’s relatively low score on Hofstede’s (2001) ‘power distance’ dimension and especially also the ‘masculinity’ dimension of culture. It is this make-up of the Dutch culture that makes showing leadership very difficult (Cerfontaine, 2005; ‘t Hart, 2005).
The aversion to leadership in the Netherlands is also rooted in the county’s laws and institutions, which aim to avoid concentrations of power as much as possible (‘t Hart, 2005). Checks and balances have been institutionalized throughout the political system, as well as countervailing powers and barriers to institutional change. These have been very important in blocking the introduction of more powerful individuals into the Dutch consensus-oriented system (Geurtz, 2012). There is, thus, also an important institutional dimension to the aversion to leadership.
One of the main checks on empowered individualistic political leadership is the legal principle of collegiality. By law, the local and national executive boards are collegial bodies, which means that their members collectively share responsibility and accountability for the board’s decisions. Another powerful check on individualistic political leadership is that the Netherlands knows no directly elected executives at any level of government. Through such mechanisms, an intentional ‘leadership vacuum’ was created that has rendered strong leadership almost absent and even counteracts it (‘t Hart, 2000).
Taken together, these three aversions constrain executive decisiveness to the extent that in the Netherlands, as well as in other consensus democracies, leadership is not of the same kind as the phenomenon that is described by traditional Anglo notions of leadership. Rather than being of the strong and decisive type, leadership in the consensus democracies is of the bridging-and-bonding type. Drawing on work by Putnam (2000), this concept, as developed by Hendriks and Karsten (2014), emphasizes in particular the relational aspect of leadership in bringing people together across societal and institutional boundaries.
Contextualizing the aversion to leadership in the Netherlands
The explanation for the multidimensional aversion to leadership in the Netherlands is at least twofold. First, there is the historical background to approaching leadership with such caution. In further historicizing the institutional, cultural and linguistic dimensions just mentioned, we should emphasize the history of state formation in the Netherlands – the cherished (his)story of a freedom-loving, highly decentralized, republic of semi-autonomous provinces and towns, breaking away from power-centralizing Habsburgian rule in the 16th and 17th century (Schama, 1987). For centuries, the idea of monarchy, let alone absolute monarchy, was done away with, until its reappearance as a highly constrained and increasingly ceremonial monarchy in the 19th century. With the Netherlands being situated largely below sea level, the Dutch had to synchronise widely dispersed efforts in collaborative arrangements, which developed bottom up, gelling separate water boards in an integrated system of multi-level, cooperative governance (Hendriks, 2010: 160–163). As a trading country, hemmed in between physically more forceful countries, the Netherlands cultivated pragmatic skills of negotiation, mediation and diplomacy: elements of soft power rather than hard power (Nye, 2004). For many, however, the aversion to leadership is most prominently linked to the dubious role of strong leadership – the Leiter- or Führerprinzip – in the Second World War and the horrors that came out of that (De Beus, 2006). In an attempt to explain why there is a ‘lack of leadership’ in the Netherlands, some scholars add the questionable role that a number of leading men in the Netherlands played when they collaborated with the occupying forces during the war (e.g., Kellerman, 2014). Although we do not necessarily agree with this line of reasoning, it is certainly the case that leadership in the Netherlands just after the Second World War was particularly pragmatic and technocratic, rather than visionary and decisive. This not only facilitated the swift uptake of New Public Management-style managerialism in the 1990s (Hendriks and Tops, 2003), but also had a profound impact on how the Dutch understand leadership today (Hendriks, 2010; ‘t Hart and Ten Hooven, 2004).
Second, there are socio-political reasons that connect to the ‘consociational’ governance tradition of the Netherlands. Traditionally, Dutch society was characterized by high levels of social segmentation and by a situation of coinciding political, religious, economic, social, and cultural cleavages (Andeweg, 2000; Lijphart, 1999). These produced a number of strong subcultures, or ‘pillars’, that each had its own political, social and economic networks. The so-called ‘pillarization’ of society produced a high risk of political conflict, but the Dutch managed to minimize such risks and to achieve high levels of cooperation between the pillars. This result was arrived at through a process of seeking compromises between the elites representing each of the pillars (Daalder, 1964; Lijphart, 1999; Andeweg, 2000). The high level of cooperation was facilitated by a process of depoliticization, that is defining issues as technical problems rather than as ideological conflicts (Hendriks, 2010).
To the extent that there was room for leadership in this governance model, it was a function of collective entities (Gronn, 2009). It was a form of shared, non-political leadership that left very little room for the individualistic forms of leadership that are typical of the Anglo-centric tradition (Gladdish, 1991; Lijphart, 1968). In fact, whereas Margaret Thatcher famously claimed that consensus was the antithesis of leadership (‘t Hart, 2005), in the Netherlands reaching consensus was the ultimate expression of what passed as leadership. Although the pillars have nowadays disintegrated to a large extent, both culturally and institutionally, there is still a deep-seated tradition of power-sharing and consensual leadership preserved in the Dutch governance model that results from its consociational past.
At the same time, it should be recognized that more recently the Netherlands, like many other countries in Europe, has witnessed the rise of a call for stronger leadership (SCP, 2011; see also Hendriks and Karsten, 2014). Scholars have signalled a tendency in Europe to break away from more collective and consensual forms of leadership: more and more powers are being vested in the executive branch of government, also in the hands of individual executives (Steyvers et al., 2008).
Although this does raise questions as to whether there is still such a big difference between the Anglo-American and the continental traditions of leadership, there are good reasons to believe so. First, the studies showing that there is a call for leadership among citizens are not uncontroversial. Some of the survey questions that support the leadership-call claim are, for example being criticized for being double-barrelled, which makes it difficult to reliably interpret the data (Hendriks et al., 2012). Second, figures show that the call for stronger leadership was stronger in the 1950s, 1960s and 1970s (SCP, 2011). Third, to the extent that there is indeed a noticeable call for leadership, it is often qualified as being ‘ambiguous’ (Karsten, 2013) or ‘dubious’ (Frissen, 2009). Rather than as a genuine desire for decisive leaders, the call of leadership is seen as the unsolicited result of a feeling of confusion (‘t Hart, 2005). This is because citizens express a demand for administrative decisiveness, but at the same time want their leaders to be highly responsive to their interests. Citizens seem to call for stronger leadership, but whereas this is an essential part of strong leadership not for the kind of directive leadership that acts against their wishes (Karsten, 2013; see also Dunn, 1999; see also Kane and Patapan, 2012). It might thus be that, rather than a call for heroic forms of leadership, we are witnessing an inarticulate call for responsiveness and good governance. In any case, the consensual style of decision-making is still highly valued in consensus democracies, as is the type of leadership that accompanies it (Hendriks, 2010).
Leadership in a consensus democracy: The typical case of the Dutch mayor
Although other examples are available throughout the world, one of the clearest illustrations of what leadership in a consensus democracy amounts to is provided by the Dutch mayoralty. It is also one of the clearest illustrations of how an understanding of leadership can deviate from the still-dominant Anglo-centric notion of leadership. Notwithstanding the welcome rise of non-traditional and more critical leadership studies, which place high value on the relational and collective nature of leadership, also in Anglo contexts (e.g., Collinson, 2014; Gronn, 2009; Schedlitzki and Edwards, 2014), there is still a tendency within mainstream leadership literature to rely on a model of political leaders as ‘visionary lone rangers’ (Van Esch, 2016; also Teles, 2015: 23). In contrast, the Dutch mayoralty provides an insightful and exemplary case of some of the less well-recognized aspects of political leadership in consensus democracies (Hendriks and Karsten, 2014). It is striking to see the extent to which local leadership in the Dutch context deviates from the heroic understanding of mayoral leadership that has recently come to predominate the worldwide debate on local and even global governance, most prominently in the work of Barber (2013).
Whereas Barber, in his seminal book If mayors ruled the world, describes a type of leadership that might, as he claims, be particularly authoritative and effective in many circumstances, it does not correspond very well to the type of mayoral leadership that can be observed in a consensus democracy like the Netherlands. Whereas Barber champions mayors for exhibiting a non-partisan and pragmatic styles of governance, the types of mayors that he discusses are typically institutionally strong and wilful mayors that have a personal-political, though not necessarily party-political, profile. They tend to be people of ambitious vision who are not afraid to get their hands dirty. On the view of Barber, it is their effectiveness in realizing certain policy goals that sets them apart from the allegedly dysfunctional nation state.
To illustrate the focus on results and output-legitimacy, Barber (2013: 87) quotes Mayor Nutter of Philadelphia: ‘You have to perform, you have to get things done … You either fill potholes, or don’t …’. According to Barber (2013: 88): the ‘can-do thinking’ of mayors worldwide demands a strong personality marked by both hubris and humour, a pragmatic approach to governing, personal engagement, and commitment to the city.
The current study is not claiming that Barber’s analysis of these executives fails to provide valuable insights into local leadership, but rather that it provides insufficient insight into alternative types of leadership that carry considerable relevance for the wider debate on what is political leadership in a democratic context (Kane and Patapan, 2014; Teles, 2015). Dutch mayors play a fundamentally different role in local governance than Barber’s mayors, the analysis of which requires a different understanding of what leadership is and also requires a different vocabulary. Mayors in consensus democracies do not show decide-and-accomplish leadership like, for example, former NYC-mayor Bloomberg (Barber, 2013: 25–28), but bridging-and-bonding leadership, and in the particular case of the Dutch mayor ‘democratic guardianship’. This leadership role is seen to be crucial for the authority and effectiveness of Dutch mayors, not only in the past but also nowadays (Karsten et al., 2014). A clear illustration is provided by former Amsterdam mayor Job Cohen after the assassination of prominent publicist Theo van Gogh in 2004. Through his bridging-and-bonding leadership Cohen famously managed to ‘keep things together’ (Hajer, 2009). The nature of Dutch mayoral leadership challenges some of the existing conceptions of what the role of local leaders is. In support of some of the more critical leadership scholars, we signal a need for a recalibrated leadership concept (e.g., Contractor et al., 2012; Gronn, 2009; Teles, 2015).
Methods
This article presents the results of an extensive mixed-methods study on the state of the Dutch mayoral office. Between September and November 2013, a web-based survey was conducted. The survey was sent out by e-mail to all 405 Dutch mayors on our behalf by the Dutch Association of Mayors. After three reminders, 243 Dutch mayors (60 percent of the population) had completed the questionnaire. An analysis of the responses shows that the sample is representative of the population of Dutch mayors in terms of gender, party-political membership, municipal size and the ratio of sitting mayors to acting mayors. 1 In addition, 12 semi-structured interviews with selected mayors were conducted for our study and the authors shadowed individual mayors for a total of four days and attended eight additional meetings for participatory observation. Further, the nature of mayoral leadership was discussed in four focus groups, in which a total of 24 mayors and seven other political-administrative actors participated. Also, using the lexical analysis software tools Textstatt and WordSmith, a computerized text analysis was conducted of 202 job vacancy texts for Dutch mayors that were published between 2008 and 2014. This represents 86 percent of all job vacancy texts that were published since the introduction of a new nationwide toolkit for drafting those. 2
Leadership and guardianship: The Dutch mayor
As we have argued, the Dutch mayoral office is an insightful case of non-Anglo leadership that reinforces existing doubts regarding mainstream leadership studies’ preoccupation with individualist and visionary approaches to leadership (Collinson, 2014; Gronn, 2009; Teles, 2015). There are at least two dimensions to the particularity of the Dutch mayoralty. The first is the statutory position of the Dutch mayor. The second is the leadership role that the Dutch mayor takes up in local government in practice. These two dimensions will be discussed separately in this section, followed by a characterization of the leadership role of the Dutch mayor in terms of ‘democratic guardianship’.
The statutory position of the Dutch mayor. Especially when compared to most of their European colleagues, seen from an institutional perspective, Dutch mayors are rather weak leaders (Bäck et al., 2006). Dutch mayors have few executive tasks and responsibilities. In the Netherlands, powers that in other countries usually rest with mayors individually are vested in the Board of Mayor and Aldermen. This executive board operates as a formal collective decision-making body, which means that mayors have very little individual decision-making power. In case of conflict, mayors have the deciding vote, but this right is almost never exercised because mayors are expected to follow the lead of the majority of the board or of the responsible alderman (Karsten et al., 2014). Further, for the execution of most of their tasks, Dutch mayors greatly depend on the political support of the popularly elected municipal council. Mayors stand fully accountable to the council, and the importance of this accountability relationship has substantially increased over the last decades. This has led to a significant weakening in the mayor’s statutory position (Karsten et al., 2013). Dutch mayors do have some executive capabilities that stem from their legal responsibility for the police and public safety, but also in this this respect they are fully accountable to the municipal council. The Dutch mayor is, thus, surrounded by a strong set of checks and balances that leave little room for directive and individualistic leadership.
In addition, what is highly unique about the statutory position of the Dutch mayor and what is also a strongly non-Anglo characteristic of the Dutch mayoral office is that Dutch mayors, unlike most other world mayors, are not directly or indirectly elected. In contrast, Dutch mayors are appointed by central government (Schaap et al., 2009). This ‘Crown appointment’ is one of the strongest expressions of the fact that Dutch mayors are expected to adopt a non-political stance. It expresses the fact that Dutch mayors are positioned to operate relatively autonomously and at a distance from the political goings-on. Although the influence of the municipal council on the selection of the mayor in the complex selection procedure is substantial, the Crown appointment mirrors the image of an independent and non-political mayor that exercises little individualistic leadership, particularly not political and decisive leadership.
Although Barber (2013: 98), in his discussion of mayoral leadership, does touch upon the unique selection procedure of Dutch mayors and also discusses some of its consequences for the role of the mayor, his account does not fully grasp the unique character of the Dutch mayoral subtype, which cannot justifiably be equated with Barber’s archetypal and typically Anglo-American mayor. Barber rightfully champions the pragmatism of Dutch mayor’s such as Van der Laan (Amsterdam) and Aboutaleb (Rotterdam), but at the same time overlooks important aspects of their weak statutory position and especially the fact that they are not decide-and-accomplish types of mayors. These mayors do not lead in the same way as many other world mayors do, at least in part as a consequence of how they are positioned in an institutional sense (Hajer, 2009). In addition, their pragmatism is not of the same kind as the ‘achieving’ kind that Barber and others applaud mayors for and that is typical for his Anglo-centric notion of leadership. Whereas Dutch mayors, from an Anglo-inspired perspective on leadership, might be viewed as non-leaders, they actually perform crucial leadership roles in local government in the way that we will elaborate on in the next section.
The role of the Dutch mayor in local government. The fact that heroic interpretations of the leadership concept do not fit with the leadership role of Dutch mayors becomes especially clear when we look at what the role of Dutch mayors is in practice. Whereas eminent leadership scholars see ‘developing strategic policy direction’ as one of the core leadership tasks for any kind of political leader (Blondel, 1987: 16; Bryson and Crosby, 1992; Leach and Wilson, 2002: 668), even for leaders of the facilitative kind (Greasley and Stoker, 2008: 723), it is exactly this task that may be the least accurate description of the role that mayors play in Dutch local government.
The importance of mayoral activities in relation to the municipal council.
Question: What is the importance of the following activities in your contact with the municipal council? (n = 243) – NA answers were coded as missing values.
The importance of mayoral activities in relation to the executive board.
Question: What is the importance of the following activities in your contact with the municipal board? (n = 243) – NA answers were coded as missing values.
Paired t-tests show that there are substantial differences in the perceived importance of mayors’ activities in the representative municipal council and in the executive board. Mayors’ roles in giving policy direction (t = 7.94; p < 0.001), overseeing the quality of the decision-making process (t = 3.92; p < 0.001), coaching (t = 8.34; p < 0.001) and mediating conflict (t = 2.82; p = 0.005) are statistically significantly more important in their contact with the executive board than in their contact with the representative council. These differences can be explained by the fact that the mayor is a non-member chair of the municipal council, but a full-member chair of the executive board. In contrast, these being legally defined responsibilities of the mayor, safeguarding integrity (t = −2.40; p = 0.017) and rendering account (t = −6.58; p < 0.001) are statistically more important in relation to the municipal council. Effect sizes range from small to medium – Cohen’s d ranges between 0.135 and 0.643. The activities of chairing (t = 1.05; p = 0.293) and building support (t = −1.35; p = 0.178) show no statistically significant differences. Although the leadership role of the Dutch mayor, thus, is more prominent in the executive board than in the representative council, it is more process-oriented than content-driven in both cases.
These survey results are corroborated by interviews, focus-group discussions and participatory observations. Dutch mayors report that one of the first things they learn after becoming mayor is to ‘sit on their hands’ and to not have an opinion. The following quote from a mayor gives a clear illustration: As a mayor I’ve learned not to act. I don’t do anything anymore. What I’ve come to understand in 16 years of being a mayor, is that it is not about what I think as a person, about my opinions, about my beliefs or about my expertise. I don’t do anything substantive anymore. Instead, it’s all about my social skills and about bringing people together.
We believe that such expectations are typical for the Dutch consensus-oriented context and are antithetical to the expectations that many local political leaders face in Anglo-American contexts. At the very least, they are contradictory to the decide-and-accomplish expectations that many Anglo-American scholars place on local political leaders (e.g., Barber, 2013; Giuliani, 2002). The recent introduction of stronger, directly elected executive mayors in many European counties other than the Netherlands has reinforced these expectations (Steyvers et al., 2008; Elcock, 2008). In that sense, the Dutch mayoralty may bear some resemblance to the older, largely ceremonial role of the English Lord Mayor or that of the Irish ‘Cathaoirleach’. At the same time, the Dutch mayor does have substantially more executive responsibilities than the latter.
Ranking of the role orientations of Dutch mayors.
Question: Please rank the following roles according to their importance in being a mayor in relation to societal actors. Please assign the role you see as the most important the value 1, the second most important value 2, etc. The least important role receives the number 8 (n = 243).
This finding is supported by the fact that, in contrast, the visionary role is seen as being of low importance for Dutch mayors; only 1.2 percent of the office holders see this as their most important role out of the eight roles offered in the survey. In addition, contrary to what was expected, mayors rarely saw themselves in the role of ombudsperson, despite this role being traditionally linked to the Dutch mayoralty and also to the office of the mayor in other countries (Kjær, 2015; Klinkers et al., 1982). The participants in the focus groups explained this finding by arguing that an ombudsperson generally has to form a judgment in a conflict, whereas a mayor should remain impartial at all times, which also fits nicely with the non-decision-making role of the Dutch mayor. In focus-group discussions, mayors also expressed doubts about their role as a ‘moral leaders’, not because they did not see themselves as providing moral guidance, but mainly because of their aversion to the language of leadership. These discussions, again, underline the linguistic dimension there is to the aversion to the leadership principle in the Netherlands. Admittedly, standard deviations and interquartile ranges show that the uniformity in mayors’ opinions is lowest as regards the moral leadership role, indicating the existence of substantial differences in opinion. Some mayors, thus, are more inclined to see themselves as moral leaders. Several other roles included in our survey show considerable dispersion too, but the overall picture of a bridging-and-bonding rather than visionary leader is a quite clear one.
In our survey, when asked about the sources of authority Dutch mayors draw from when performing their roles as an office holder, the office holders indicated that they see their ‘position above and between political parties’ as the most important source of their political authority. More than two-thirds of our respondents included this particular authority source in their personal top five, out of the fourteen plausible sources offered, and more than 20 percent saw it as their most important source of authority. The phrase ‘above and between political parties’ is typically Dutch, and is one of the prime examples of the unique Dutch vocabulary of political leadership. It is hard to translate its full meaning to English. By approximation, what this phrase signals is mayors’ all important independence from party politics and their non-involvement in political decision-making.
Dutch mayors’ desire to adopt a non-political stance, that is act as politically independent liaisons rather than as strong local leaders, also because the socio-political context expects them to do so, produces the paradoxical situation where 99 percent of all Dutch mayors are members of political parties but many of them are very wary of being labeled as a representative of a particular political party (Karsten et al., 2014). Out of the fourteen plausible sources offered, political party membership is seen to be the least important authority source for Dutch mayors. It is included in their personal top five by only 4.53 percent of mayors, and then only as fourth of fifth in the ranking. The following quote of Amsterdam mayor Van der Laan gives a clear illustration of mayors’ wariness towards being seen as a party-political figure: ‘As long as I’m a mayor, I [will act as if I] have no affiliation with any political party’ (cited in Cohen, 2013: 23). This view expresses the widely shared conviction that Dutch mayors should try to avoid taking sides at all costs, as this is crucial to their ability to exercise bridging-and-bonding leadership.
The language of Dutch mayoral leadership. As discussed above, there is also an important linguistic dimension to mayors’ aversion to the notion of leadership. When describing their activities in relation to both the administration and society in focus-group discussions and interviews, Dutch mayors frequently used rather awkward alternatives for the word ‘leiderschap’ that sometimes not even exist in the Dutch language. One of these is ‘begeleiderschap’ (Karsten et al., 2014), which translates best to ‘guiding leadership’. In one of the focus groups, a mayor expressed his heart-felt discomfort with being called a ‘leading figure’ in local government and euphemistically described the role he played in local government using a sports metaphor: ‘being on the sidelines, but not off side’.
Our lexical analysis shows that when the verb ‘to lead’ occurs in job vacancy text for mayors, it is most often used in the formal and neutral meaning of chairing the municipal council and the executive board. The words ‘leadership’ in the meaning of being a frontrunner, or lexical variants thereof, occurs in only 9 percent of the job vacancy texts. Hence, some municipal councils express a desire for a powerful and decisive figure, but this is only the case in 18 out of 202 vacancies. By comparison, the word ‘begeleiderschap’, which counts as a softened alternative to the word ‘leadership’, occurs in 11 percent of the vacancy texts. Thus, when seeking for a new mayor, municipal councils are hesitant to use the word ‘leadership’ in the heroic meaning of the word and frequently use consensus-oriented alternatives that describe the guiding role that the Dutch mayor has.
The language of Dutch mayoral leadership is riddled with softened terms for different leadership roles that avoid the word ‘leadership’ itself, such as ‘burgervader’ (first citizen, ‘pater civitas’), ‘primus inter pares’ (first among equals), ‘oliemannetje’ (catalyst), and ‘procesbegeleider’ (process manager). Other terms associated with the Dutch mayoralty do include linguistic variants of the word ‘leadership’, but use softening adjectives such as ‘slow leadership’. Expressing the same idea, one of the official training courses for Dutch mayors is entitled ‘invisible leadership’ and educates mayors in their bridging-and-bonding leadership role as it is performed ‘behind the scenes’ (Dutch Association of Mayors, 2015).
The leadership dilemma for Dutch mayors. The linguistic aspect of the aversion to leadership is just one expression of the paradoxical leadership expectations that Dutch mayors are faced with and that may be typical of democratic leadership as such, and of bridging-and-bonding types of leadership in particular: Dutch mayors are expected to provide guidance, but not to lead from the front (see also Kane and Patapan, 2012). To repeat, they are expected to develop a political vision and to give policy direction but at the same time they are deemed to not take the substantive lead and to take a non-political stance. The following quote from a mayor also gives a clear illustration: Many municipal councils, like mine, have asked their mayors to develop a strategic vision for the municipality. But, when push comes to shove, they do not like the mayor taking the lead. Some people even say that the mayor is the best-paid civil servant of the municipality, but then one that cannot do anything. Some mayors feel deeply frustrated with this [paradox].
To honour these paradoxical leadership expectations, Dutch mayors can best operate behind the scenes. Rather than proposing solutions themselves, for example they are better off ‘continually asking the right questions’, as one mayor expressed his thoughts in an interview.
‘Democratic guardianship’: A characterisation of Dutch mayoral leadership
In reaction to the above, one possible response could be: ‘If Dutch mayors are not called leaders and cannot act as frontrunners, then maybe they are not leaders’. Such a rejoinder, however, does not do justice to the crucial role that Dutch mayors play in local government and that clearly is a practice of leadership, but maybe not of the decisive and individualistic kind (Karsten et al., 2014).
The practice of leadership as it is enacted by the Dutch mayor is one where the ‘leader’ plays a crucial role in producing good governance (Kane and Patapan, 2014; Rhodes and ‘t Hart, 2014). Dutch mayors increasingly take up a role of guardians of local democratic processes and of the quality of local decision-making (Karsten et al., 2014). In our survey, mayors report that their most important day-to-day activities in relation to both the municipal council and the executive board are ‘safeguarding the quality of local decision-making’ and ‘safeguarding the ethics and integrity of the local administration’ (Tables 1 and 2). One of the ways in which mayors execute this role is by organizing and institutionalizing reflective processes. They, for example organize reflective sessions with councilors on the culture of ethics within the local administration (Karsten et al., 2014). Scholars such as Heifetz (1994) recognize the crucial importance of such a leadership role in public as well as private organizations. National law has attributed to the Dutch mayor some formal responsibilities in this field, but this leadership role of the mayor is also deeply ingrained in the local administrative culture in the Netherlands. The mayoral office is traditionally associated with values such as integrity, impartiality and procedural justice, which is believed to be an important base for the authority of its holders (Karsten et al., 2014).
Typical for the role of Dutch mayors is that they have been assigned a more explicit formal responsibility to ‘advance the ethics and integrity of the local administration’ in 2016, while this is not accompanied by any formal powers. The successful fulfilment of this new responsibility is, therefore, fully dependent on the informal authority of Dutch mayors. Such a practice, on the one hand, mirrors the weakness of the statutory position of the Dutch mayor and, at the same time, shows the legislator’s trust in the mayors’ ability to act as informal leaders behind the scenes.
Because of their unique role in local government, Dutch mayors typically care more about ‘process values’ – procedural fairness, transparency and responsiveness – than about ‘substantive values’ (Bobrow and Dryzek, 1987). Their role is to bring people together and to unite them over fair democratic processes, with mayors being the politically independent guardians of those. This is also why Barber’s emphasis on the getting-things-done attitude of mayors does not very accurately represent the nature of mayoral leadership in the Dutch context.
There are, are of course, many alternative notions of leadership that can be very useful in analyzing leadership phenomena that are similar to the role that is executed by the Dutch mayor. We, however, would like to argue that most of these conceptions, some of which have become very popular among scholars when describing the transformation of contemporary local political leadership, miss out on important aspects of the role of the Dutch mayor too. The cherished facilitative-leadership concept, for example, ascribes an important direction-providing role to leaders that Dutch mayors lack notably (Greasley and Stoker, 2008). Likewise, neither the ‘laissez faire leadership’ concept (Bass and Bass, 2008) nor the ‘servant leadership’ concept (Barbuto and Wheeler, 2006) are very well applicable in the Dutch context since the Dutch mayoralty does have its own agenda. Mayoral leadership in the Netherlands is neither a follower-serving nor value-neutral type of leadership. ‘Being politically independent is not the same as being neutral’, a clerk illustrated this point in an observed meeting. Instead, Dutch mayors do actively strive for realizing valued outcomes, but these should be perceived in terms of process values rather than in terms of substantive values. Then again, the ethical-leadership concept is too narrow to capture all the different aspects of the role that Dutch mayors play in monitoring the democratic process.
Collective understandings of leadership, such as the shared-leadership concept, may be more suitable for developing our understanding of leadership in consensus democracies (for an overview, see Contractor et al., 2012). Some of these, however, focus primarily on the who-question of leadership (e.g. Gronn, 2009). The case of the Dutch mayor, however, draws particular attention to the what-question of leadership (what do leaders do?) since it highlights the importance of an alternative leadership task, that is guardianship, which is underexplored in the mainstream political leadership literature. We believe, therefore, that a more suitable alternative conception of leadership is required to accurately describe the leadership role of the Dutch mayor, for which we propose the term ‘democratic guardianship’.
We recognize that the concept of ‘democratic guardianship’ is in some ways paradoxical. Drawing from the seminal work of Dahl (1989) one might argue that the role of the guardian, with supposedly superior knowledge and virtue, is antithetical to democracy. We believe, though, that the case of the Dutch mayor shows that democratic guardianship is a real and valued possibility in practice (see also Kane and Patapan, 2014; Teles, 2015).
It can even be argued that some of the essentially non-democratic characteristics of the Dutch mayoral office, the role of the Crown in the selection of mayors in particular, make an important contribution to making democratic guardianship possible in practice. This is because the growing influence of the directly elected municipal council on the selection of the Dutch mayor has politicized the mayoral office and has made mayors more vulnerable to political processes. These developments has caused tensions within in the mayoral office that hamper office holders’ abilities to act as democratic guardians because they harm their political independence (Karsten et al., 2013). In contrast, at least in a symbolic way, the Crown appointment reaffirms the independence of Dutch mayors from party politics, which is an important source of authority for them in acting as a democratic guardian (Karsten et al., 2014). This is not to say, though, that the necessary political authority and institutional independence that support the democratic guardianship cannot be achieved by other means.
Other expressions of bridging-and-bonding leadership
In support of other more critical leadership scholars (e.g., Collinson, 2014; Schedlitzki and Edwards, 2014), the aim of this article is to give in-depth insight into an alternative to the heroic – wilful, decisive, rather masculine – perspective on political leadership. We have highlighted the Dutch mayoralty as a clearly contrasting case of bridging-and-bonding leadership that is consonant with consensus democracy. We dubbed its specific type of leadership ‘democratic guardianship’. There are, however, also other empirical expressions of bridging-and-bonding leadership in the context of consensus democracy, albeit of a somewhat different form. In the scope of this article, there is no room for additional in-depth treatment, but we will briefly present two other empirical variants, as these are insightful national and supra-national examples that complement the sub-national case of the mayoralty: the presidents of Switzerland and the European Council.
The President of the Swiss Confederation operates in an institutional context which Lijphart (1999), in his seminal book Patterns of democracy, categorized as a strong expression of consensus democracy. The logic of power-sharing and power-dispersal dominates the entire system; it even incites and supports the well-developed referendum culture in Switzerland (Linder, 2010; Papadopoulos, 2001). In the Swiss system, compromising and coalition building among political rivals has been institutionally fixed. The desire to maintain solidarity in diversity has led to the creation of an intimate complex of consultation, accommodation and integration. The distribution of seats in the federal executive between the four dominant political parties, for example follows a standard formula – 2:2:2:1.
The Swiss federal executive works on a strictly collegial, power-sharing and equalizing basis. The seven-member Federal Council is viewed as the collective head of state. Formally, there is a Swiss President who chairs the Federal Council, but (s)he is typically the primus inter pares, not vested with any special powers or prerogatives. The Presidency – and even the Vice-Presidency – is kept in check by a system of annual rotation. Every year another councillor serves as the acting President. As such, the President of the Swiss Confederation has a rather weak position (Kriesi and Trechsel, 2008: 69–80). Its institutional set-up is functional in bridging and bonding the linguistic, social and political sub communities that make up the divided Alpine country, not in producing executive leadership of the decide-and-accomplish type which is deemed ‘strong’ in the traditional Anglo-centric interpretation.
The European Union is also an institutional system in which power is distributed and must be shared. There are a great number of mechanisms aiming to prevent smaller countries from being overpowered by bigger countries or substantial minorities from being drowned out by simple numerical majorities. Compared to the early days of the European Union, the unanimity requirements have been toned down, but in the present-day European Union governing power remains a fragmented commodity. Therefore, in Lijphart’s (1999) distinction between majoritarian and consensus democracy, the European Union is clearly positioned on the consensual side. Owing to its modus operandi, the EU is considered by many as a distant and expertocratic system of governance, which tends to be multi-actor and multi-level in a way that obscures the distribution of responsibility and accountability (Scharpf, 1999). In a famous quip, former US Secretary of State Henry Kissinger is believed to have said: ‘Who do I call if I want to speak to Europe?’, signalling a perceived lack of EU leadership.
The current Presidency of the European Council must be understood against this backdrop. In 2014, Donald Tusk, former prime-minister of Poland, succeeded Herman van Rompuy, who had served as the European Council’s first President between 2009 and 2014. From the beginning, the Anglo-American press has derisively pictured the EC President as a sorry figure: not popularly elected, but appointed by, and thus dependent on, the chiefs of the national governments of the member state, the German chancellor and the French president in particular; not forceful or charismatic – while Tony Blair was available, the first President appointed to set the tone was Herman van Rompuy, well-respected in Belgium (itself a consensus democracy in the extreme) but little-known beyond. Reality is more nuanced. The EC Presidency is not as powerful as its American or French counterparts, but it is not as nominal as the Swiss version either. It lies somewhere in between, at best playing a role as bridge-builder, connecting actor or ‘political plumber’ – to repeat a phrase used by Van Rompuy. In terms of leadership, such a role is more facilitative than decisive, but certainly not irrelevant or ineffective. In the fragmented reality of European politics, the EC Presidency plays a crucial role in the getting together of the acts and tendencies of various powerful others (Van Esch, 2016).
These additional examples show that bridging-and-bonding leadership - as opposed to decide-and-accomplish leadership - is not confined to the particular case of the Dutch mayor. It can be found at (supra)national levels of governing beyond the particular Dutch context.
Conclusion and discussion
We have argued in this article that the traditional Anglo-centric notions of decisive leadership, which, notwithstanding the welcome rise of more critical conceptions, continue to dominate the study of political leadership, do not adequately capture crucial aspects of the bridging-and-bonding type of leadership that is more common in consensus democracies in significant parts of continental Europe. Democratic theory suggests that the supportive context of consensualism is even more widely dispersed, with the European Rhineland as perhaps the best-known but certainly not the only breading ground: important parts of Asia and Africa are also sensitive to an alternative, bridging-and-bonding understanding of democratic leadership (Hendriks, 2010; Lijphart, 1999).
The case of the Dutch mayor provides us with a theoretically and empirically salient alternative to decide-and-accomplish leadership in the form of ‘democratic guardianship’. Dutch mayors show how deliberately not giving substantive policy direction and adopting a non-political stance can be crucial aspects of effective leadership. In addition, the Dutch mayoralty shows that monitoring the quality of local decision-making and safeguarding the ethics and integrity of the local administration can be important leadership roles in a democratic context. Such roles cannot be captured accurately when using the English language of strong leadership (Jepson, 2010; Teles, 2015). This language is not particularly sensitive to the process-oriented aspects of important leadership roles in consensus democracies.
This decide-and-accomplish bias in studying leadership sometimes leads to undesirable misunderstandings about the role of leaders in a consensus context, for example, them being described as ‘non-leaders’ (e.g., Hayward, 2008; Kellerman, 2014), whereas they perform valuable but underexplored leadership tasks (Van Esch, 2016). Further, some of the pivotal guardianship roles that leaders play in consensus democracies have not yet gained a prominent place in the established literature on leadership tasks. We believe that there is plenty of room to increase the ‘theoretical sensitivity’ towards such phenomena (Tummers and Karsten, 2012). This, in turn, can improve leadership studies’ abilities to help leaders in consensus democracies, and indeed elsewhere, deal with some of the unique challenges that they face on a day-to-day basis.
In our understanding of Dutch mayoral leaders, we join a growing body of literature now emerging from Anglo contexts that challenges the heroic notion of leadership and that highlights the collective, non-decisive aspect of leadership (e.g., Collinson, 2014; Gronn, 2009; Schedlitzki and Edwards, 2014). At the same time, we believe that this strand of research can be further developed by studying the specific democratic leadership functions that frontrunners fulfil in consensus democracies because these tend to remain undertheorized (Kane and Patapan, 2014; Teles, 2015).
One of the reasons for the existing mismatch between leadership theory and practice is that the international language of leadership encounters active resistance when it is used in a consensus context. Again, the Dutch mayoralty provides an insightful case because it is has developed a language of leadership that bypasses the word ‘leadership’ without necessarily rejecting the leadership phenomenon altogether. Rather, it has redefined some of the core leadership tasks, changing the focus from substantive values to process values.
What is required, therefore, is more than rhetorical innovation alone. It is also a recalibration of the leadership concept itself, as to give more room to bridging-and-bonding and process-oriented forms of leadership in addition to its already well-covered decide-and-accomplish expressions. In this article, we have suggested that in the case of the Dutch mayor the term ‘democratic guardianship’ provides a viable and acceptable alternative. This concept adequately captures Dutch mayors’ practical ability to usefully unite guardianship and the democratic ethos in the function of good local governance. In one dominant strand of democratic theory, this might seem a paradoxical concept, as it essentially holds guardianship to be at odds with democracy (Dahl, 1989). The study of leadership in consensus democracies, thus, also requires a recalibration of some of the theory of democratic leadership (Hendriks and Karsten, 2014).
While emphasizing theoretical relevance of the case, we fully recognize that the Dutch mayor is only one particular example of a bridging-and-bonding leader out of the many that are available. Within the confines of this article, we could provide brief additional discussions of just a few. Further exploration is required of how the different expressions of bridging-and-bonding leadership compare in different contexts (see also Contractor et al., 2012). As it is, the Dutch mayoralty already poses considerable challenges to the current Anglo-centric debate on global governance, in which mayors are ascribed a decisive role. Additional research could also usefully answer remaining questions about (a) the proliferation of bridging-and-bonding types of leadership in different democratic contexts, (b) how bridging-and-bonding leadership interacts with the ambiguous call for stronger, individualistic, decide-and-accomplish leadership it is facing, and (c) what good governance practices thrive under the guidance of bridging-and-bonding leadership in different cultural and institutional contexts and (d) what democratic guardianship implies in various democratic contexts. Although our analysis of the Dutch mayoral office provides only a starting point for this line of research, it does already clearly underline the need for broadening our conceptions of leadership beyond the traditionally heroic notions and for developing additional languages of leadership, some of which may even want to avoid of the word ‘leadership’ itself.
Footnotes
Acknowledgements
The authors would like to thank the two anonymous reviewers for their constructive comments on an earlier version of this article, as well as Karin Lasthuizen (VU University Amsterdam). The authors would like to thank Sabine van Zuydam (Tilburg University) and Frank van Kooten (Necker van Naem) for their valuable contributions to the collection of the data.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The empirical research for this article was funded by the Dutch Ministry of the Interior.