Depoliticisation in smart cities : a case study of Amsterdam's smart city discourse & projects

(1)

DEPOLITICISATION IN SMART CITIES

A CASE STUDY OF AMSTERDAM’S SMART CITY DISCOURSE &

PROJECTS

JENNIFER SUMMERS 11378395 J_SUMMERS7@HOTMAIL.COM

HUMAN GEOGRAPHY (URBAN GEOGRAPHY) SUPERVISOR: CODY HOCHSTENBACH SECOND READER: WOUTER VAN GENT

SUBMISSION DATE: 26/06/2017 WORD COUNT: 18,113

(2)

ACKNOWLEDGEMENTS

I would like to thank my respondents for their participation in this research project. I would also like to thank Cody Hochstenbach for his guidance and supervision.

(3)

ABSTRACT

Smart cities are now a fundamental part of contemporary urban language, existing on the intersection between digital technology, sustainability and city planning (Söderström et al, 2000). In their inception, the potential of smart cities was celebrated somewhat uncritically. However recent years have seen a push from the academic world for more considered reflection of their implementation and impact. This research adds to this understanding through the use of qualitative research to explore the extent to which processes of depoliticisation are reflected in both the city and project level in Amsterdam. To do this, it focuses on how smart cities potentially reframe urban issues in terms of both morality and technology, thereby diluting the presence of the ‘political’. Whilst findings demonstrate that depoliticisation is indeed present to varying degrees across Amsterdam frameworks, the reality of this process differs from current academic theory in terms of intent and awareness. This presents a more nuanced understanding of how smart cities should be viewed and understood.

(4)

LIST OF FIGURES, TABLES & MAPS

FIGURES

Figure 1: Ranking the world’s ‘smartest’ cities Figure 2: SmartLight in Hoekenrodeplein Figure 3: NDSM Energie wind turbines

Figure 4: Graph of respondent awareness & favourability

Figure 5: The current smart city structure of discussions

TABLES

Table 1: Overview of key Amsterdam smart city projects Table 2: Methodology overview

Table 3: Operationalisation

Table 4: Respondent segmentation for analysis

MAPS

Map 1: Hoekenrodeplein in Amsterdam Map 2: NDSM in Amsterdam

(6)

INTRODUCTION

Smart cities are now a fundamental part of contemporary urban language, existing on the intersection between digital technology, sustainability and city planning (Söderström et al, 2000). It has been estimated that cities around the world will invest approximately $41 trillion over the next 20 years on smart city projects (CNBC News, 2016), making them one of the most important areas of urban planning today. Their rise to fame across both the public and private sector has been characterised by techno-optimistic celebrations from policy makers and tech corporates, who sell utopian visions of future cities of modernity, efficiency and sustainability.

Despite this financial and societal investment, there is still a distinct layer of ambiguity that surrounds the concept of smart cities regarding their true definition, purpose and role in society. Academics are increasingly making note of this contradiction, and have begun to counter balance the positive narrative that surrounds smart cities with a more skeptical approach, questioning their success, their real agenda and their universal appeal (see Hollands, 2008; Greenfield, 2013; Kitchin, 2014). This criticism predominately focuses on the disconnect between the urban imaginaries associated with smart cities, the reality of their implementation and their impact on urban communities.

However, it is also important to understand how this disconnect has been made possible. Much of what we see in smart city planning can also be linked to the characteristics of depoliticisation, and in turn, the post-political landscape that many academics argue is increasingly present in many societies today. Notions of depoliticisation and post-politics can best be understood through defining the ‘political’ as a concept that fundamentally allows for contrasting ideologies to be present. The dilution of the role of the political therefore leaves a landscape in which there is little room for the articulation of contrasting viewpoints (Mouffe, 2005). In the case of smart cities, this is driven by the ‘technology will save us’ narrative, combined with the moral obligation of sustainability, both of which appear to objectively justify the application of smart city ideas across the world with very little debate or opposition (Vanolo, 2014). Some academics also argue that these environments risk both an abuse of power from elites who are able to use smart city narratives to push other agendas, and the sidelining of alternative perspectives that may not hold the

(7)

same beliefs as those that dominate (Garsten and Jacobsson, 2011). This thesis will therefore specifically focus on the process of depoliticisation in smart city planning, to build understanding of both this process and of smart cities in general.

Current academic literature often examines smart cities on a conceptual, or at best, city level, with much less attention being paid to specific projects or case studies. This is potentially both limiting and misleading, as top-level discourses alone do not necessarily represent the real, on-the-ground work that is taking place within the smart city arena. To counter balance this, this thesis will use a multi-scalar approach; studying the city level discourse in Amsterdam as well as two contrasting project level case studies. I will answer the research question “How are smart city discourses mobilised to depoliticise overarching goals?”.

To answer this question, it is crucial to firstly understand why smart cities are potentially rich territories for processes of depoliticisation. This thesis will therefore begin by detailing how and why smart cities have become such a key component of urban planning in recent years, including global shifts in power and their perceived societal benefits (Chapter 1).

Chapter 2 will introduce one of the core criticisms of smart cities; that of their ambiguous definition, and the connotations of the terminology. This brings into question how it has been possible for smart cities to be adopted so uncritically, despite the concept having a lack of clear meaning.

Chapter 3 will explore this further by introducing the concept of depoliticisation. This chapter argues that depoliticisation poses many potential threats to society, and that signs of depoliticisation can be seen in the ways in which smart cities are understood, discussed and potentially used to justify alternative political actions. However, this chapter also makes note of the fact that much of this evidence focuses on smart cities on a conceptual or city level. This means there is a need for research that also looks at specific smart city projects. Chapter 4 therefore will lay out the research design used in this thesis, which utilises a multi-scalar, case study approach.

(8)

Chapters 5 and 6 will focus on the findings of this research; firstly detailing the empirical evidence collected and secondly discussing how these findings confirm, contradict and supplement current research. This thesis concludes in Chapter 7 by arguing that depoliticisation does appear to be present in Amsterdam’s smart city, but the process is not necessarily used intentionally as literature suggests. In addition, it poses questions regarding how academics research smart cities, and the role that social scientists can play in tackling smart city issues.

(9)

1. CITIES GETTING SMARTER

“The 19th_{century was a century of empires, 20}th_{century was a century of nation} states, and the 21st_{century will be a century of cities”}

Wellington Webb, Former Denver Mayor Emphasis on urban environments as the focal point of global future development has become more pronounced in recent years; forecasts that the world’s urban population will increase by 50% by 2030 makes managing urban areas a top priority on the political agenda (United Nations Population Fund, 2007). This growth has gone hand in hand with changing economic reliance from agriculture to professional and service industries, and a social shift from local communities to dispersed anonymity (Short, 2006).

These tangible changes have also been accompanied by a subtler reconceptualisation of the power of cities relevant across local, national and global scales. Saskia Sassen’s notion of the ‘Global City’ (2001) makes note of this, arguing that the power of cities goes beyond rural to urban migration, as globalisation has not only been characterised by an increase in cross-border economic relations, but also by the systemic shift in ownership of these processes from inter-states to inter-cities; driven by privatisation and deregulation (Sassen, 2001). This has led to a renewed relationship between the city and the state, with the city often being used by the state as a symbol of a country’s progression and identity (Kavaratzis and Ashworth, 2005).

Such changes have paved the way for smart cities to flourish. The term ‘smart city’ was first used in 1990s, originating from the private sector and having previously referred exclusively to small, discrete projects from technology corporates such as IBM, Cisco and Siemens (Townsend, 2013). Throughout the 1990s and 2000s ‘smart city’ was often used somewhat interchangeably with terms such as the ‘wired’ or ‘digital’ city. These terms share similarities with smart cities but are fundamentally different, referring more specifically to the physicality of cables and to connectivity in general (Hollands, 2008). Smart cities instead are better defined by their usage of these components for data monitoring and regulating to create efficiencies and improve productivity (Hall, 2000; Kitchin, 2014).

(10)

However, whilst technology is often a key component of smart cities, many also argue that this should not define them. Giffinger and Haindl (2009) instead note that smart city ideas can encompass many aspects of city life, including a smart economy, smart mobility, a smart environment, smart people, smart living, and smart governance. Smart city narratives highlight clean energy usage, responsible resource management and innovative building solutions, as well as the integration of green spaces and wildlife into urban regions. In addition to this, a key component of a smart city is an economy that is “increasingly driven by technically inspired innovation, creativity and entrepreneurship, enacted by smart people” (Kitchin, 2014, p. 131). Finally, smart cities are also orientated to a particular organisational idea linked to both public-private partnerships and citizen participation (Hajer and Dassen, 2014).

Overall, the smart city has come to be understood as a concept that attempts to combine economic aspirations with environmental sustainability through both the utilisation of technology and the restructuring of current city planning processes (Herrschel, 2000). However, the ideological nature of these claims has been counter balanced by critiques that these promises lack real meaning and practical application. These ideas will be explored in the next chapter.

(11)

2. THE ISSUE OF DEFINITION

Despite an increase in usage of the term ‘smart city’ across the public and private sector, there is still not a clear and consistent definition associated with it. Although it is clearly possible to gauge some understanding of what a smart city is from literature today, this not does negate the inconsistency in definitions that exists both globally and locally. For example, Albino, Berardi and Dangelico’s (2015) analysis of 143 smart city developments from around the world indicated “the definitions posed by particular cities calling themselves “smart cities” lack universality” (p. 18). The authors argue that the far-reaching application of smart cities is one of the key drivers behind the lack of consistent definition. As smart cities are talked about in many contrasting environments – from corporate strategy to urban development - companies or individuals who may wish to emphasise different elements of smart cities will use the term in different ways. This means even within the same city different definitions are often present:

“It has, on the one hand, been applied to “hard” domains such as, buildings, energy grids, natural resources, water management, waste management, mobility, and logistics…. the term has also been applied to “soft domains” such as, education, culture, policy innovations, social inclusion, and government, where the application of ICT are not usually decisive”

Albino, Berardi and Dangelic, 2015, p. 10

Many academics argue that this vagueness paves the way for the term to be used predominately for promotional purposes, rather than referring to actual urban developments. Hollands (2008) suggests that the label of ‘smart’ plays a significant role in creating certain visions and expectations of smart cities that may not match reality, and in branding a city as progressive and innovative. Here he argues that the language of ‘smart’ indicates an uncritical approach to urban development, implying “positive urban-based technological innovation” (p. 304). For Hollands (2008) the word ‘smart’ feels so inherently appealing, it raises the question - which cities would not want to be considered smart? It has similarly been noted elsewhere that that the term ‘smart city’ manages to combine grand visions of future urban utopias with commonsensical language that feels practical and straight forward, and therefore all the more engaging (Gibbs et al, 2013).

(12)

With these debates and concerns in mind, it is useful to refer to the perspective of Vanolo (2014), who acknowledges that the term smart city is perhaps an “evocative slogan lacking a well defined conceptual core” (p. 884) rather than a term with a formal definition that can be applied objectively. However, this raises the question of how it is possible for smart cities to have gained such traction, despite not having a formal definition in place. It is possible that depoliticisation plays a role in this; therefore this will be explored further in the upcoming chapters.

(13)

3. DEPOLITICISATION & SMART CITIES

In recent years there has been increasing academic focus from the social sciences on the process of depoliticisation and related terms such as ‘post-political’. Although the exact meaning of these terms is debated, Wilson and Swyngedouw (2014) argue that “broadly speaking…they all refer to a situation in which the political – understood as a space of contestation and agonistic engagement – is increasingly colonised by politics – understood as technocratic mechanisms and consensual procedures that operate within an unquestioned framework of representative democracy, free market economics, and cosmopolitan liberalism” (p. 6).

It should be noted here that authors often define ‘politics’ and ‘the political’ in different ways, adding to the complexity of this terminology. In the most basic of terms however, depoliticisation refers to a form of statecraft that essentially involves moving an issue from the governmental sphere into the public or private sphere (Hay, 2013). Garsten and Jacobsson (2011) further argue that depoliticisation is “taking what is essentially a political problem, removing it from the realm of political discourse, and recasting it in neutral language” (p. 432). In more detail, depoliticisation is characterised by the usage of seemingly objective arguments to create a shared, consensual vision, with reduced space for the articulation of diverging interests. Whilst this does not necessarily remove ‘the political’ altogether, it fundamentally downplays it’s relevance (Sahlin-Andersson, 2004).

Examples of these processes can be seen across many facets of modern day life. Recent years have seen many uphold the view that the current stage of economic-political development represents progression that is one step closer to societies of peace and harmony. Globalisation has led to what many sociologists have termed a ‘second modernity’, characterised by the liberation of individuals from nation ties and a worldwide focus on the implementation of human rights (Mouffe, 2005). In unison with this however, the political landscape has seen many additional changes that counter balance claims of positive progression. According to Gill et al (2012), the 2000s were a decade of wars, scandals involving the political elite and the use of public money to bail out the privileged during the financial crisis (p. 509). These events were surprisingly met with very little political opposition, which Gill et al (2012)

(14)

argue is representative of the notion of ‘post-political’, as there was instead apparent political consensus on a global scale regarding actions to be taken (Gill et al, 2012). Non-compliance that did exist appeared to be restricted to radical movements or opposition outside formal political arenas.

The reality of this new era is a world in which ideological debates are often replaced with seemingly more objective management techniques; conversations are centered on the best approach to tackling agreed-upon problems, rather than broader discussions regarding the problems themselves (Swyngedouw, 2009; Gill et al, 2012). Post-political landscapes are therefore best understood as ones in which the scope of debate is narrowed significantly, thereby diluting the relevance of the ‘political’ when defined as a concept that allows for debate between contrasting ideologies. For Chantal Mouffe (2005), such a landscape is fundamentally centered on the role of conflict, which she argues is increasingly considered illegitimate in post-political environments. Swyngedouw (2005) also notes that these landscapes are intrinsically tied to wider processes of neoliberalisation and state restructuring, which have further eroded the political sphere, leading to increased reliance on market forces that set “the rules of the game” (p. 1993).

The concept of depoliticisation can be further explored through studying smart cities. Swyngedouw (2007) argues many urban concepts such as smart cities are becoming increasingly post-political in nature, as they reduce urban visioning down to a single, undisputed idea. As argued by Vanolo (2014); “the smart city may increasingly become a generic and easily agreed target, without proper critical discussions” (p. 891). There is already evidence of this on a global scale. Research by Black and Veatch (2017) shows the universal approval of smart cities; 94% of global municipalities view the smart city movement as “transformational and capable of bringing positive, long-term impacts to cities around the world” (p. 3). Whilst this alone is not necessarily indicative of depoliticisation, it is important to note that this faith in smart cities exists despite smart cities also exhibiting a series of clear downsides, including data and privacy concerns, links to neoliberalism, risks of increasing social inequality and the definitional issues already discussed (see Hollands, 2008; Greenfield, 2013; Kitchin, 2014). Therefore, it is important to question why and how urban planners and tech corporates are able to view smart cities in such a positive light.

(15)

Smart city’s rise to fame has been driven by a number of certain assumptions regarding both the power of technology and the moral responsibility of urban sustainability. As noted previously, smart cities often adopt a ‘commonsensical’ language approach, characteristic of post-political landscapes, that works towards presenting smart cities as the most obvious and sensible answer to urban problems. As argued by Garsten and Jacobsson (2011) depoliticised ideas are often seen as “rational, objective, and neutral, based on the sound principles of human solidarity and market dynamics working together to form a humane capitalism” (p. 432), which fits with descriptions used for smart cities.

With this in mind, it can be argued that smart city visions “reduce political conflict, insurgence, radicalism and resistance in the contemporary city in favour of ‘disciplined’ cities” (Vanolo, 2014, p. 884). On one hand, techno-optimistic narratives are presented as a catchall solution to all city problems, driven by the promise of innovation. Hollands (2008) argues that smart cities assume an “automatically positive impact of IT on the urban form” (p. 314), whilst Hajer and Dassen (2014) similarly state that innovation is seen as a purely technological matter, “glorifying new possibilities” (p. 16). Alongside this, the concept of sustainability through smart cities is established in moral terms; cities are increasingly imagined as being morally responsible and citizens are expected to act in a way that is environmentally conscious (Brand, 2007). Swyngedouw and Cook (2009) argue that this process is particularly apparent in issues that relate to sustainability and climate change; “environmental politics is a politics legitimated by a scientific consensus which, in turn, translates into a political consensus” (p. 602). These things act as a powerful argument in favour of smart cities, positioned as a question of ‘right or wrong’ rather than ‘left or right’ (Mouffe, 2005). This works not only to ensure that citizens adhere to the concept of a smart city, but also to present smart cities as the only vision for urban futures. Whilst this does not mean that opposition does not exist in some form, it often means that there is reduced contestation to the offered solution, and instead conversations are focused on the practicalities of implementation (Swyngedouw, 2009).

When considering these ideas, it is interesting to note that smart city advocates such as Townsend (2013) claim to resolve the questions of defining smart cities through a reframing of the question from ‘what is a smart city?’ to ‘what do you want a smart

(16)

city to be?’, placing emphasis on their flexibility. However, such approaches actually only serve to reiterate the idea of smart cities as the sole solution, as it rules out the possibility to rethink structures on a wider level, or invent completely new solutions (Vanolo, 2014).

3.1 THE ROLE OF RANKINGS

Depoliticisation can also be seen in the ways in which smart cities are discussed publically. Smart cities are frequently understood through global indices and rankings, which benchmark different cities throughout the world against one another in terms of their ‘smart city success rate’, as shown in Figure 1 below.

Figure 1: Ranking the world’s ‘smartest’ cities (2016)

Source: Forbes, 2016

These rankings appear to standardise complex data, allowing it to be easily used and understood. This is particularly apparent when results are presented in the highly stylised and therefore visually appealing manner seen in Figure 1. However, Uitermark et al (2017) argue that the use of statistics in this way serves to both

(17)

objectify and obfuscate key issues, and risk such data becoming ‘black boxes’ that actually conceal information. Giffinger and Haindl (2009) similarly argue that rankings in particular encourage a ‘beauty contest’ approach, where focus is on the overall ranking, rather than the complex interrelations and causalities that sit behind them. Such approaches are in line with ideas of depoliticisation, as they downplay the downsides of an idea through the simplification of information. Although politicians often favour these tools for their apparent objectivity (Vanolo, 2014), the calculations behind them are frequently over-looked or downplayed. The result is that they can be discussed with little regard to the criteria for which the rankings were created, and therefore used subjectively to mobilise different political goals (Uitermark et al, 2017). Closer examination into such usage of smart city ranking data reveals that different sources do indeed use different techniques to produce their calculations, although this is not immediately visible. For example, Forbe’s 2016 list of the world’s ‘smartest’ cities (Figure 1) is predominately favourable towards economic conditions, whilst the Natural Resources Defense Council’s Smarter Cities Ranking is characterised by a strong bias toward environmental-related criteria (Albino et al, 2015). This means that ranking scores could be incorrectly discussed and used if calculations are not considered properly.

The creation of smart city lists or rankings also corresponds to depoliticisation in another way; they imply a linear path of development; with those who are further along being the winners, and those further behind being the losers (Vanolo, 2014). Not only does this downplay the need to adapt policies and ideas to local conditions (McCann, 2011), but also it suggests once more that each city should be working towards the same end goal or vision.

3.2 THE DANGERS OF DEPOLITICISATION

From this we can see that depoliticisation is a process in which the political arena is evacuated of conflict, critique and debate, and instead characterised by harmony, trust and shared values. On the surface, this sounds both attractive and beneficial. However, such a viewpoint fails to acknowledge the cracks in this type of system.

Critics argue that consensus does not equal peace (Rancière, 2005, p. 8), and in fact such landscapes run simultaneous risks of some views being ignored, and others being accepted without critical evaluation (Mouffe, 2005). Mouffe (2005) for example

(18)

states that “conflicts are a constitutive part of ‘the political’ and necessary for a functioning democracy” (in Garsten and Jacobsson, 2011, p. 422), and that contrasting identities, viewpoints and ideologies should play a fundamental role in decision making, aiding understanding and empathy for others and paving the way for compromise where needed. Without room for conflict or forms of expression, there is risk that certain points of view are disregarded, whether intentionally or not, potentially leading to “the emergence of antagonisms” on an even larger scale (Mouffe, 2005, p. 4). Hallward (2005) further argues that this is a fundamental threat to democracy, as “the concern of democracy is not with the formulation of agreement or the perseveration of order, but with the intervention of new and hitherto unauthorised modes of disaggregation, disagreement and disorder” (pp. 34 - 35). For some, this environment is even characterised by shifts in democratic governing, in which traditional models are transformed into a society of control (Swyngedouw, 2009).

Swyngedouw (2009) also argues that depoliticisation in reference to environmental politics often has ‘contradictory effects’; promoting ideas of morality whilst actually hurting the most vulnerable. An example is that of rising energy prices that hit the UK in 2008. For some environmentalists, these price increases were positive for society; a reduction in accessibility was seen as a step forward in reducing oil consumption and therefore greenhouse gases. However, whilst richer sections of society may have been willing and able to take a financial hit for the greater good, poorer individuals “saw food prices spiral out of reach, food crops replaced by bio-fuels, access to energy curtailed and the cost of moving around going up”(Swyngedouw, 2009, p. 602). This environmental injustice is indicative of the uneven way in which politics centered on sustainability can impact society, demonstrating the need for all ideas to be discussed in detail regardless of how morally good they appear in theory.

Alongside this, an absence of conflict reduces the barriers for new ideas or policies, making them easier to implement. Therefore, it is important to recognise that a post-political environment is perhaps more favourable to the urban or economic elite, and that those in power could use such conditions for their own advantage. The depoliticisation of smart cities is potentially a powerful tool, as it is possible that discourses of sustainability and technology could be used to meet other targets. In

(19)

addition to critiques on the definition of smart cities, this is a key focal point of academic enquiry and debate.

Firstly, some have begun to question how far smart cities really contribute towards urban sustainability. Here they argue that smart city strategies are more strongly driven by financial investment and entrepreneurialism (Haarsted, 2016), and make fertile grounds for the particular set of beliefs and positions of neoliberalism, identified by Greenfield (2013) as being characterised by privatisation, deregulation, frictionless global trade and reduced taxes (p. 891). Brand (2007) explains why this may be the case in his 2007 paper, as he neatly summarises the relationship between neoliberalisation and urban environmentalism. He firstly notes the misalignment between global and city level responses to environmental issues, questioning why exactly it is that urban environmentalism has “flourished” whilst non-urban environments continue to struggle (p. 616). He argues that cities actually have a different end-goal to the ones being promoted; “the construction of a clean, green and intelligent city image is in fact useful to attract investments, leading sector professional workers and tourists” (Brand, 2007, p. 618). Secondly, Brand (2007) argues that environmental focus acts as a justification for the adoption of policies that prioritise economic development at the expense of urban equality. These claims are frequently seen in smart city examples. Hollands (2008) for example found in his research that smart city websites such as Edmonton and San Diego actually indicate an emphasis on “business-led or business friendly” criteria above environmentally friendly ones (p. 308).

Perhaps it is important not to demonise business development itself. However, academics have further argued that the type of development that smart cities encourage, which is predominately focused on the creation of creative and knowledge economies, risks creating social issues and division in cities. Hollands (2008) argues that although smart cities claim to represent diversity, they are predominately built for ‘smart workers’, potentially polarising citizens economically beyond the levels of inequality that already exist. Whilst highly educated and highly skilled workers are likely to benefit from these policies, lesser skilled employment opportunities are deprioritised or even replaced with technology (Hollands, 2008).

(20)

In addition to this, smart city narratives often fail to adequately address issues of data usage, privacy and security, despite a huge dependency on citizen data to operate (Kitchin, 2014). It is arguable that processes of depoliticisation play a key role in allowing this to happen. Debates on these matters extend far beyond smart cities; they are not just focused on the physical protection of data, but the moral and philosophical arguments surrounding data ownership and citizen privacy rights. Adam Greenfield (2013) has argued that smart cities often assume ownership of citizen data, creating a city in which everyday life is “owned and monetised” (p. 615). He further suggests that it is not necessarily in a government’s interest to encourage citizens to increase their personal privacy, and privacy issues are therefore purposefully downplayed.

The breadth and diversity of smart cities means that a vast variety of projects, policies and initiatives could technically come under the smart city banner, even if it does not necessarily directly involve either sustainability or technology. In addition, the excitement that surrounds such ideas could mean that clear downsides or issues are not properly considered. Depoliticisation plays a strong role in this, as it paves the way for a range of actors and institutions to pursue their own interests with reduced conflict. In fact, Haarsted (2016) argues that one of the key aspects of smart cities that makes them so appealing is the ‘political opportunity’ they represent (p. 3). This idea will be explored in this thesis.

However, it is important to note here that much of this analysis focuses on smart cities on a fairly top level - rather than on specific smart policies, projects or ideas - as academics predominately debate the terminology, or dissect smart city marketing. This is inherently problematic; smart cities are seen in academia as “a somewhat nebulous idea” (Shelton et al, 2014, p. 13) yet real money is being spent on bringing this idea to life all over the world. This means that even though the term may be flawed, we cannot simply dismiss the smart city as a meaningless concept. We must look at how smart cities actually manifest in order to move from high-level debates to more granular and practical understanding. Therefore, there is a need to examine the reality of smart cities through the use of case study and multi-scalar approaches. This will be detailed in the next chapters.

(21)

4. RESEARCH DESIGN

4.1 RESEARCH QUESTIONS

My overall research question for this thesis is:

“How are smart city discourses depoliticised to mobilise overarching goals?”

This question aims to explore how and why smart city discourses may be depoliticised in Amsterdam. In order to answer this question, I will also answer a series of sub-questions that will help me reach my final conclusion. These are as follows:

1. What are the overarching smart city discourses in Amsterdam? 2. What are the characteristics of two smart city projects in Amsterdam? 3. How do these two different smart city projects link to the overarching

smart city discourses in Amsterdam?

4. How are elements of depoliticisation reflected in smart cities across these two scales?

These questions have been designed to best utilise the multi-scalar approach of my research. Smart city discourses are not necessarily consistent even within the same city; therefore they should be examined from multiple perspectives. Questions 1 and 2 will identify the characteristics of smart city discourses at both the city and project level in Amsterdam, allowing the different scales to be viewed and understood independently from one another. Question 3 will compare and contrast these findings, highlighting any discrepancies or similarities that exist across scales. Question 4 will use the findings from questions 1,2 and 3 to identify elements of depoliticisation across the two scales and feed into my overall research question.

I hypothesise that my research will indicate that smart cities do indeed have elements of depoliticisation in them, and that such elements are likely to be used to promote a shared vision for the future despite any negative implications. I am also expecting to see some disconnect in the implementation of the smart city narrative across different scales, with the top down project perhaps being more driven by financial incentives due to the fact that it was created in conjunction with corporate companies.

(22)

4.2 CASE STUDY SELECTION

The multi-scalar approach of this thesis involved the selection of three different case studies; firstly the city itself, and secondly the smart city projects within the city to examine in more detail. City level narratives do not necessarily give the whole picture of smart cities; therefore it is important to look at specific projects too.

In order to maximise the potential impact and usability of my research findings, I selected a ‘regular’ smart city rather than a purpose built smart city such as Songdo or Masdar City, which are often seen as ‘special cases’. This is because they have been designed from scratch and are not necessarily reflective of other global cities (Greenfield, 2013). This means findings may be useful for cities beyond Amsterdam.

Amsterdam is a city that is considered to be at the forefront of smart city developments in Europe, described as a “futuristic tech hub” (The Kernal, 2015). It has received global recognition for its progress in the smart city arena, and has been awarded significant funding to further these developments in recent years. It is also a relatively well-established smart city; the Amsterdam Smart City (ASC) platform was founded in 2009 and had over 100 partners involved in more than 70 projects throughout the city in 2015. Table 1 below gives an overview of some of the key smart city projects in Amsterdam to date.

Table 1: Overview of key Amsterdam smart city projects

Smart Citizen Project Citizen participation project which allowed individuals to monitor air and noise pollution through a hardware sensor connected to an app.

Living Labs The bringing together of communities to

discuss and test ideas and initiatives that could be rolled out to the rest of the city.

City-Zen An energy saving program that stands for

for “city-zen carbon energy” and includes projects such as future proofed energy grids and sustainable buildings.

(23)

store energy generated during the day at peak hours in the evening to balance solar energy usage and consumption. Source: The Kernal, 2015

This means that there are enough smart city projects, companies and employees in Amsterdam to ensure doing research there is both feasible and worthwhile. In addition, studying a progressive city such as Amsterdam also positions my research as being at the forefront of these developments.

I also selected two smart city projects in Amsterdam to explore, the details of which can be found below. Given that smart cities are still a relatively new phenomenon, there are less clear boundaries regarding what a ‘typical’ smart city case study might look like compared to other research topics (Gerring, 2007). Instead, my case studies were chosen to be exploratory in nature; seeking to understand more about smart cities and depoliticisation rather than prove or disprove a pre-existing, established theory (Yin, 1994). They were also selected due to their contrasting structure, which created the opportunity to explore depoliticisation in smart cities from two different perspectives. As one is a top down initiative, and one is bottom up, projects were likely to differ in terms of goals, resources and actors involved, all of which could impact the extent to which processes of depoliticisation could occur.

1. The SmartLight project (the top down approach)

Amsterdam Economic Board ran SmartLight in conjunction with Amsterdam Smart City, with contributors including technology giants Cisco, Philips, Alliander and KPN. This project was focused on creating adaptive light, cameras and public Wi-Fi in the Hoekenrodeplein area (highlighted in the map below), with the aim of improving the environment in terms of both safety and atmosphere.

(24)

Map 1: Hoekenrodeplein area in South West Amsterdam

Source: Google Maps

More specifically, this development involved the implementation of adaptive lighting, controlled through remote controls and automatic sensors, which respond to various different needs impacted by things such as the type of event or weather. This not only creates a more pleasant and safer atmosphere, but ensures that lighting energy is only used when needed, thereby being more sustainable in nature (Amsterdam Economic Board, 2016).

Figure 2: SmartLight in Hoekenrodeplein

(25)

2. NDSM Energie (the bottom up approach)

Founded in May 2013, NDSM Energie is a co-operative association working to create a more sustainable NDSM (highlighted in the map below). Their aim is to self-generate all energy consumed in NDSM within the area and to raise awareness of sustainability.

Map 2: NDSM area in Amsterdam

Source: Google Maps

NDSM Energie focus on building wind turbines, helping local businesses and communities to gain access to solar panels, increasing awareness of sustainability and renewable energy, and the production of clean energy through recycling of organic waste. Although they are an independent, grassroots organisation, they are financed through membership participation from many businesses within the NDSM area, which include Hema, Viacom and Red Bull, amongst others (NDSM Energie, 2017).

(26)

Figure 3 NDSM Energie wind turbines

Source: NDSM Energie, 2015

4.3 METHODOLOGY

To answer my research questions, I used qualitative research in the form of document analysis, semi-structured interviews, and site visits to the locations of smart city projects. Given the exploratory nature of my thesis topic, qualitative research was the most appropriate approach, as it allowed for in-depth and investigatory analysis, as well as an emphasis on the language, tone and explanations used by respondents (Clifford and Valentine, 2008).

An overview of my approach can be seen in Table 2. It should be noted that each stage of research was not entirely discrete, and there was overlap between all three in terms of timings and analysis.

(27)

Table 2: Methodology overview

Stage 1 Secondary / desk

research.

Analysis of policy documents.

12 policy documents.

Stage 2 Primary research.

Interviews with city level stakeholders.

5 semi-structured interviews.

Stage 3 Primary research.

Interviews with project level stakeholders. Project site visits.

3 semi-structured interviews. 2 site visits.

Stage 1

The first stage of my research involved the examination and analysis of a series of key policy documents related to Amsterdam’s smart city developments. These were identified through my own desk research and assistance from my supervisor. They were analysed thematically; key findings, quotes and statistics were made note of and then put into themes to be used in later stages of analysis. The purpose of this was to enable me to develop a concrete understanding of the ‘on paper’ smart city discourse in Amsterdam, including goals, preferred approaches and key participants involved. These findings were then to be incorporated into my interviews in stages 2 and 3, allowing me to explore how this narrative manifests in reality.

Stage 2

Stage 2 involved interviewing individuals on the city level smart city discourse in Amsterdam. The purpose of this was to understand the dominant narrative across the city, which could then be compared to both the policy documents and project level interviews in stage 3. These interviews were also supplemented with watching the documentary ‘Smart City: In search of the Smart Citizen’ by Dorien Zandbergen and Sara Blom which focuses on various aspects of Amsterdam’s smart city discourse.

(28)

Stage 3

The final stage of my research was to interview participants of the two aforementioned projects, SmartLight and NDSM Energie, as well as site visits to see the developments in person. The purpose of this stage was firstly to understand how the Amsterdam smart city narrative played out in real life smart city projects, and secondly to compare top down and bottom up smart city approaches.

4.4 SAMPLE & DATA COLLECTION

Given that my thesis topic is focused on the goals, structure and narratives of smart cities, rather than citizen facing impacts, it was necessary for me to recruit participants with professional experience of smart cities in the Netherlands. This was so that all respondents had enough knowledge on the subject matter, and that they could provide insight that is not necessarily publically available. These individuals can be described as ‘agents of change’ as they play a key role in driving Amsterdam’s smart city narrative. Therefore, interviews were focused on exploring whether these individuals talked about smart cities in a way that indicated depoliticisation, or if this was a subject they were aware of to some degree, although it was not expected that they would explicitly talk about depoliticisation itself (see Chapter 4.5).

To ensure I was able to get a holistic view of smart city developments in Amsterdam, I recruited a diverse range of individuals from corporates, knowledge institutions and public / private institutions. I also spoke to both smart city advocates and more skeptical individuals to get a diversity of opinion. However it is important to note that their experience, education levels and job titles mean that they were not representative of other Dutch citizens.

Potential respondents were identified through a combination of my own desk research and recommendations from my thesis supervisor. I used email to contact them, an example of which can be found in Appendix 3. In most instances the interviews were face to face, however two were done via telephone conference to accommodate the respondent’s schedule.

(29)

As my respondents were selected specifically for their expertise in certain areas of smart cities, I ensured that my interview style allowed them to discuss the topics that they knew most about in order to gain the most from my interviews. However, alongside this, it was necessary to ensure some consistency across all interviews to allow for comparative analysis. To cater for this, I created an item list overview with several key themes and core, general questions that would be used on all respondents (see Appendix 4), and then for each interview added individual questions tailored to the knowledge and expertise of the respondent.

Another key factor in interviewing was the need to recognise that some respondents may wish to present their work in the smart city field in a predominately positive way. The extent to which they did or did not do this was a key finding for my research. However it was also important for me to introduce questions regarding the negative aspects of smart cities to get a holistic view. Balancing these two interests meant ensuring I was not too leading in my questioning which could bias the research.

In total I secured eight in-depth interviews that lasted between 40 and 60 minutes, a full breakdown of which can be found in Appendix 5. There were two main issues in my sample. Firstly, Amsterdam Smart City – who would have been a key respondent - were unable to commit to an interview due to the high number of requests they get. However, they sent me a list of FAQs that answered many of my key questions, which was used in my analysis. Secondly, I was unable to interview anybody from the municipality to get a government perspective, which would have been highly useful. However, this was counter balanced to some extent through the policy document analysis. In addition, the high caliber of my respondents both in terms of their expertise and seniority meant each of my interviews was extremely informative and generated a huge amount of insight. This also meant I was able to have a relatively modest sample size without compromising on the quality of my research findings.

Recruitment and therefore data collection was slightly slower than anticipated, predominately due to the Easter and Kings Day national holidays which meant many individuals were unavailable in early to mid April. To accommodate for this, I changed my own schedule to allow for many interviews to take place at the end of April and beginning of May, which was later then initially planned.

(30)

A second issue in my data collection was the language barrier, as I do not speak Dutch. Whilst this was generally not a problem during interviews due to the high level of English spoken by my respondents, it was potentially limiting in terms of access to policy documents and other resources. Chapter 4.2 describes my reasons for choosing Amsterdam in terms of its smart city developments. However, researching Amsterdam as a non-native also allowed me to study the city in contrast to my own experiences and understanding of living, studying and working in the UK, which brought extra light to my findings. It also meant that I was able to view Amsterdam with a certain degree of distance both in terms of experience in the city and emotional attachment. This meant that this small downside of the language barrier was outweighed by the benefits of studying abroad.

4.5 OPERATIONALISATION

One of the most important aspects of my research design was to be clear in how I would determine if depoliticisation was present. Depoliticisation is predominately an academic term that is rarely used in daily life by the average individual; therefore it was unlikely that any respondent would use this terminology explicitly. My research was therefore focused on determining whether depoliticisation appeared to be present from the conversations I had with interviewees, rather than asking them directly about it.

Flinders and Buller (2006) state that assessing depoliticisation can be problematic, particularly in regards to the identification of causation, as shifts may be extremely subtle, and somewhat subjective. Therefore to help guide me, I used the ideas of Garsten and Jacobsson (2011) to identify signs of depoliticisation across three main pillars:

Table 3: Operationalisation

Organisational structure • Decentralised operations / network-based forms of co-ordination.

• Harmonious relationships between different actors involved.

(31)

• Projects that build upon the principles of voluntarism and partnership.

Goals / aims • Shared or consensual visions / ideals.

• Lack of conflict in deciding end goals. • Lack of clarity on timeframe and

practicalities of end goal.

Language / tone of voice • Emphasis on co-ordination and co-operation.

• The use of ranking / indices statistics or data.

• Framing of Smart Cities as a benefit for all. • Framing of Smart Cities as a moral, legal or

technical issue (see below).

The presence of a single signifier from the above list was not enough to argue that depoliticisation was present. Instead analysis focused on the presence of multiple corresponding signifiers and the strength of the signifiers.

Following this, I used the ideas of Mouffe (2005), to classify the type of depoliticisation. This was necessary as the purpose of my thesis was not just to see if depoliticisation is present, but also to comment on it’s role and purpose. Mouffe (2005) argues that depoliticisation can come in three key forms:

• Juridification: The transference of issues into legal issues, resolved by legal experts.

• Technocratisation: Decision making on the basis of technical expertise or evidence.

• Moralisation: The framing of a topic as a moral or ethical issue that benefits the whole of society.

(32)

4.6 POSITIONALITY

Understanding my own positionality is key in qualitative research, as argued by DiAngelo and Sensoy (2009); “knowledge is dependent upon a complex web of cultural values, beliefs, experiences, and ascribed social position” (p. 446). I am a young, British female, who has experience in working with the technology sector in a consultancy role through previous jobs in the UK. Issues of being a non-native have been discussed in Chapter 4.4. I would also describe myself as somebody who is simultaneously interested in both environmental issues and technology in general. Theoretically, this could make me more likely to view smart city developments favourably. It was therefore important for me to be aware of the potential biases that this may cause throughout my research and counter balance them where needed. I did this through in-depth analysis that considered multiple different perspectives and thorough interrogation of results.

4.7 DATA ANALYSIS

Transcription and early data analysis took place immediately after the data had been collected, with further, more in-depth analysis taking place again once the whole data collection process was finished. Initial analysis was more focused on making note of key findings, whilst the second stage of analysis focused on determining the overall story of the data and comparing results with academic literature. Data was analysed without the use of a data analysis tool such as atlas.ti. This was possible due to the small sample sizes in stages 2 and 3, which meant the amount of data collected was manageable and could be analysed manually. In addition, as both my personal and professional research experience has included the analysis of data in this way, this was my own preferred method. Two key goals drove the data analysis process; the first was to gather a clear and well-rounded view of smart cities in Amsterdam across scales, and the second was to identify elements of depoliticisation.

I received signed or verbal permission from all of my respondents to use their name in my research (an example of this document can be found in Appendix 6). However, some expressed concerns in having their name associated with quotes. Therefore, to protect the identity of my respondents I refrained from attributing their name and job title to quotes in the findings section of this thesis. However, in presenting results in this manner it was important not to ignore the contrasting viewpoints of different

(33)

respondents, impacted by their role, experience and their own personal outlooks. I therefore segmented respondents into three core groups, shown in Table 4.

Table 4: Respondent segmentation of interviewees for analysis Typology Definition

Commentator Those that research or study smart cities or associated issues / topics.

Implementer Those that work directly on smart city projects.

Collaborator Those that work with smart city projects / organisations as partners or collaborators.

In order to understand how these roles impact respondent perceptions, an overview of their attitudes can be seen in Figure 4, along with a brief description of each segment. This graph was created based on qualitative perceptions rather than through quantitative measures. These perceptions came from the interviews with respondents rather than the respondent’s job title or experience alone, although the two are of course inextricably linked.

In discussing the empirical findings of this research differences between respondent groups will be referenced. Where there is no mention of differences between groups, there was no significant difference in perceptions.

(34)

Figure 4: Graph of respondent awareness & favourability

Collaborators

Collaborators are those who do work in the smart city field, although not necessarily directly. They often have a consultancy role, or are more involved in connecting different smart city actors rather than implementing smart city projects themselves.

The positioning of those within the Collaborator category does not mean to imply that respondents were not highly knowledgeable on smart cities. Rather it is being used to demonstrate that the Implementers and the Commentators were simply more aware of the specific issues related to smart cities, and depoliticisation.

Implementers

Implementers differ from Collaborators as they work directly on designing and implementing smart city projects. They also work solely in the smart city field, whereas Collaborators tended to have a more general interest in smart cities amongst other topics.

Favourable towards smart cities

Unfavourable towards smart cities Aware of negatives smart cities & depoliticisation Less aware of issues of smart cities & depoliticisation Commentators Implementers Collaborators

(35)

As shown in this graph, my research showed that working within the smart city field does not necessarily equate to complete uncritical adoption of the concept as assumed in academic literature (Hollands, 2008; Vanolo, 2013). To the contrary, this research found that those who are advocates of smart cities had some of the greatest awareness of the wider questions and debates that smart city projects influence. Within this segment, it was also clear that those who had worked in the field longer were even happier to debate the downsides of smart cities, due to both their expertise and their confidence in acknowledging negatives whilst still promoting benefits.

Commentators

Commentators are those who research and analyse smart cities or associated concepts, but do not work directly in the smart city field.

Commentators were, as expected, extremely knowledgeable in terms of theoretical and conceptual understanding of smart cities. The key difference between the Commentators and the Implementers was their favourability towards smart cities as a whole, with Commentators being the most skeptical.

Along with differences between respondents, it is important to make note of the differences between my two data sources – policy documents and interviews. It was clear in my research that the policy documents adopt a less critical, more favourable approach when discussing smart cities than seen in interviews. In addition, they often have less depth, instead often discussing issues through a surface level approach. There are several reasons why this may be the case. Firstly, policy documents normally fulfill the purpose of representing a whole organisation, or even a whole city’s point of view or strategy. This means that it is to be expected that they will be more reserved in terms of expressing an opinion or acknowledging flaws in their own approach. Conversely, an interview setting creates opportunities for respondents to discuss their personal opinions as well as those of the company they work for in a more nuanced manner. Secondly, interviewees were often responding to questions that specifically enquired about the challenges, concerns and considerations of smart cities, again presenting more opportunities for them to discuss these ideas. Perhaps most importantly however, policy documents are shared widely and publicly, often acting as a form of city promotion in some way. They not only highlight strategies for

(36)

the future, but celebrate achievements of the present too, and fundamentally aim to present Amsterdam in a positive light. In doing so, negatives and potential downsides are often overlooked. As a result, there is a bias in my research findings towards the interviews, as this data source provided far richer insight with regards to my specific topic.

(37)

5. EMPIRICAL FINDINGS

This research showed that that the term ‘smart city’ is both widely recognised and used across Amsterdam at the city-level and to varying degrees at the project-level too. However, the city does not have one clear, consistent top-down smart city strategy, which created some differences in how smart cities are understood across different scales, projects and respondent segments. Instead, it was acknowledged that there is still a sense of the unknown in regards to smart cities in Amsterdam: “Somebody told me a joke about smart cities. They said smart cities are a bit like teenagers talking about sex - they think they know what they are talking about, but nobody has done it in real life and nobody really knows what it is. That's not truly the case anymore but it says a lot about it!”

Collaborator, 2017 Despite this, all respondents adhered to using the term, even if somewhat reluctantly. This was primarily due to an instinctive understanding of what a smart city is for those working directly in the field; “there is a very immediate understanding on a very high level as to what you mean by it” (Implementer, 2017). This means that individuals are easily able to identify or categorise a project or solution when it is associated with being a ‘smart city’. However, respondents also acknowledged that smart cities carry a sense of excitement around them that further gives reason for usage of the term; “there is some sort of energy, vibe going on about this topic” (Collaborator, 2017). This meant that respondents could use the term to give projects or ideas a boost or wider recognition; “what they need to be called smart for is to get funding, recognition, and be in the loop” (Commentator, 2017). Similarly, policy documents give the sense that the energy of smart cities is used to differentiate new projects from ‘old’ ways of working. Within policy documents, such as those from Amsterdam Economic Board and Amsterdam Smart City the phrase, “getting things done” was frequently used in conjunction with smart city ideas. This language indicates an action-orientated, agile and efficient approach that feels significantly different from the bureaucratic, process-driven and slow ways of working often associated with government policies.

(38)

Amsterdam was seen to be a concept that was brought to life by individual projects and people, rather than defined by governmental policies. However, even with no clear over-arching strategy, and clear differences in how respondents used the term, it was possible to determine some of the core elements of Amsterdam’s smart city discourse from this research. This was most notably demonstrated in the frequent use of the phrase ‘smart solution(s)’ in discussions at both the city and project level. Collaborators and Implementers both championed using smart tools as a way to solve pre-existing problems in the city; “we always want to think, well which problem are you going to solve?” (Collaborator, 2017). Therefore, there is a rejection of

innovation for the sake of innovation. Instead smart solutions are mobilised in response to an identified problem in the city, with a clear goal in terms of improvement or change. This means that much of the smart city process in Amsterdam involves city research and the identification of problems in the city prior to undertaking smart city advancements. This was noted to be in response to previous mistakes in terms of approach:

“A very common mistake that has been made in the recent years by cities, the municipalities and the companies involved in smart cities is that they didn’t look into the specific needs and values of the cities, they just wanted to roll out the technology and test the technology. And then often the link was missing between problems within the society, the people living in the city and the technology”

Implementer, 2017

In addition to this, smart cities were described by the majority of respondents to also

be multifunctional in nature; preferably saving money, being sustainable, improving

efficiencies, and / or doing social good simultaneously; “we can't really call something smart if it's not social or sustainable” (Collaborator, 2017). There was also a significant emphasis on collaboration, particularly in terms of encouraging different actors to work together in different ways; “I think a lot of things that we can achieve are in co-operation. What can you do, and what can we do together to help find a better solution” (Collaborator, 2017). Beyond these three elements, respondents viewed further traits of smart cities in different ways.

As in the academic world, the role of technology in smart cities was debated in Amsterdam. A broad view was taken by the Amsterdam Smart City platform, who

(39)

stated in their FAQs (2016) that smart city projects could include those without a strong technology component, although no justification of why this should be the case was provided. A policy document by The Waag Society and AMS Institute entitled ‘Amsterdam Smart Citizens Lab’ (2015) offered more insight, as it argued that smart cities should be rebranded to ‘the illuminated city’, as this term is a “metaphor that

recognizes the possibilities of technology, but asserts that there is no such thing as technological panaceas to social ills” (Waag Society & AMS institute, 2015, p. 17).

This places emphasis on how governments, companies and citizens use technology,

not the technology itself. Others instead saw technology as the key component that

differentiated a smart city project from simply another interesting or innovative idea; “I am a bit more of the school that I say there has to be some new technology and data component in it. Then we are talking about smart cities” (Collaborator, 2017). Unsurprisingly, technology companies such as Cisco were keen to ensure that technology remained center stage in smart city discussions, particularly in their own publications; “One thing remains constant: part of being “smart” is utilising information and communications technology (ICT) and the Internet to address urban challenges” (Cisco, 2013, p 1). However, it was interesting to note that even the most techno-optimistic of respondents were keen to distance Amsterdam from smart cities such as Songdo, which were seen to only use technology, without enough emphasis on citizens and communities.

Regardless of whether or not smart city projects should now include technology, it was clear across both policy documents and interviews that technology was at least the starting point in smart cities. Whilst not to understate the important role that the environment plays in smart city discussions, it was in many ways seen to be one of many domains that smart ideas can be applied to. Conversations were often more focused on what technology and connectivity could be applied to, rather than using sustainability goals as the starting point; “after the technology came the societal questions - we have this brand new fantastic technology, what will that mean for us?” (Implementer, 2017).

Beyond these discussions, true differences in points of view could be seen in terms of sentiment towards the usage and meaning of the term smart city. Written documents in general appeared somewhat uncritical of the term, seen for example in Amsterdam Smart City’s definition of a smart city;

Depoliticisation in smart cities : a case study of Amsterdam's smart city discourse & projects