How to Understand the Smart City: A Critical Realist Approach to Urban Technology Acceptance
By its very definition, the smart city is an inherently technological phenomenon. Simply put, despite difference in context, all smart city projects share one common attribute — the utilisation of technology in the pursuit of increased efficiency and sustainability. The practice of which has created a US$1.56 Trillion industry. A more detailed definition, one which is widely accepted within academic circles to be the most accurate, comes from the University of Milan’s Andrea Caragliu, who states that a city is smart when “investments in human and social capital and traditional ICT infrastructure fuel a sustainable economic growth and a high quality of life, with wise management of natural resources”. For a city to attain these end goals, through the use of urban technologies, a transformation into smartness is required to take place — for example here is a link to a case study of my home city of Cambridge (U.K) concerning how the city’s transportation network is transitioning into becoming smart. However, a smart city transformation occurs under a cloak of relative invisibility.
Smart city technology comprises amongst other examples, sensors, sensor networks, 5G and full fibre internet as well as the internet of things. Yet, the implementation of these technologies and the transformation which takes place as a result, occurs out of sight, for sensors and the like are measured in millimeters and are designed to be embedded within existing and ubiquitous urban infrastructures such as lampposts, bins and busses. This element of the smart city is highlighted by the author of the quite brilliant Radical Technologies: The Design of Everyday Life, Adam Greenfield, noting that within the smart city, “perhaps mercifully, the pedestrian is at best only liminally aware of the presence or operation of these sensors”. So, as the adage goes, out of sight out of mind? Well, not exactly.
There is sizable opposition to the smart city, with narratives surrounding prevalent and important issues such as data accumulation, surveillance, creeping privatisation and the access to technology forming the backbone of critique. A component of this unease at the encroachment of the digital upon the urban lies in the lack of technological acceptance of these projects being rolled out, for it is often the case that the public only encounter these technologies through innuendo and second hand simplifications. But, how can one accept something they don’t encounter in their daily lives?
In 1989, Fred Davis published the ‘technology acceptance model’, as a “valid measurement scale for predicting user acceptance of computers”. The determining factors for technology acceptance, according to Davis, was the ‘perceived usefulness’ and ‘perceived ease of use’ of a technology in the eyes of the user — in short, if a user thinks the technology will be useful and easy to use, then they are more likely to use it. The model has subsequently been developed to include determining social factors alongside pre-established ‘anchor’ views towards technology and the user’s behavioral ‘adjustments’ when using it. However, despite the significant intellectual undertaking in developing a model for technology acceptance, there are three issues which render it somewhat insufficient for analysing the acceptance of the smart city and its technology:
- There is an oversimplification of the social factors presented as determining the acceptance of technology — see Richard Bagozzi’s work for a robust critique of this tendency.
- There is an overly hierarchical tendency within the model. It was initially developed for managers to ascertain whether workers would utilise and accept new technologies and practices in their place of work — a city is not organised in the same way as a place of employment.
- Finally, and perhaps most importantly of all, technology acceptance solely analyses technologies in which people come into contact with. This is an obvious feature of the model and is seldom an issue, but, as I have explained above, the smart city is practically invisible to those who engage with it, yet it remains controversial and the model cannot encapsulate this phenomenon.
So, how can the acceptance of the smart city, or at least the investigation into it, be carried out? For the question begs, how can someone accept something they never see or directly interact with? This question is of importance however, for the success of technical transition is contingent upon the actions of individuals that contribute to the system working or not, and the related themes of system success and failure. In other words, for a new system to work and be effective, those who interact with it largely determine whether it is successful.
Yet, this raises another question, in that if a resident doesn’t encounter a technology, why would it matter if they accept it or not? Unlike organisations, cities aren’t defined by performance indicators such as efficiency, profitability or output — at least they shouldn’t be. The benefits of increased efficiency promised by the smart city are a means to an end. So, if a city presses on with an unaccepted transition into ‘smartness’, there could be issues down the road concerning democratic participation in the city. The value of citizen participation is stated here by Norwegian academic Lasse Berntzen:
Active participation enhances democracy, especially on the local level. Participation is not only about taking part in the decision making processes, but also to build sustainable local communities, where citizens care for each other
Therefore the importance of the residents accepting a smart city transition appears in two senses — both operationally and in a broader democratic context. So, the importance of technological acceptance within the smart city has been established, as have the limitations of current theory. Therefore it stands to reason that a means of understanding the acceptance of smart city technology is required. To pursue this goal, what follows is an introductory discussion of critical realism and the benefits it would have upon the understanding, and indeed fostering of smart city acceptance.
A tenet of critical realism is that it asserts that reality is composed of what Roy Bhaskar — the founder of the theory — terms three ‘domains’ — the real, the actual and the empirical. How does this work? Well, the best way to envisage the three domains of reality is to think of an iceberg (below), a classic and oft-used metaphor for describing the notion that there is more to what we see than what we, well… see. The tip of the iceberg, the bit that sticks out of the frigid sea, embodies the empirical domain. The empirical is the realm of events as we experience them, where we interact and sensuously engage with phenomena. In the smart city, the empirical domain is the smartphone in your hand, the digital bus timetable, the council backed city car parking app — it is the realm of human computer interaction.
An example of this comes from again, Cambridge and the city’s Motion Map app, which provides a user with real-time information about the location of the next bus, amongst other travel related information. This information is obtained through the geo-locating of the bus vis-a-vis the use of a GPS, which is then uploaded to the app and, subsequently, to the user, who is in turn relying on 4G or public wifi to get the up to date information. This is the mechanics of the smart city and where the real value of its installation is revealed through the increased efficiencies obtained through the harvesting of more information and data. Yet, this whole process is invisible to the user. The commuter waiting for their bus, smartphone in hand, only sees their app and the icon for a bus trundling along the virtual road and is somewhat oblivious to the collection, processing and utilisation of data taking place. This is because the ‘mechanics’ of the smart city take place in a different domain of reality; the ‘actual’.
According to Amber Fletcher, the domain of the actual is where “events occur whether or not we experience or interpret them, and these true occurrences are often different from what is observed at the empirical level”, and is found just beneath the surface on our critical realist iceberg. From this quotation it is easy to see how the background mechanics of the smart city fit into this domain. For example, at home you may have a wireless internet connection. The only empirically visible elements of it are the black box that sits near your phone line and the icon on your device of choice to say if it’s working or not. Yet, the internet is delivered to your device through the actual domain, and the same can be said of the majority of smart city technologies.
So, to recap, the majority of smart city function takes place in the unobservable, ‘actual’ domain of reality, with the tip of the iceberg being the empirically observable elements of human computer interaction. However, the actual and the empirical don’t explain the source of what exists in these two domains. No, that is the domain of the real. Once again we turn to Dr Fletcher’s analysis of critical realism, to explain the ‘real’ domain:
At this level causal structures, or ‘causal mechanisms,’ exist. These are the inherent properties in an object or structure that act as causal forces to produce events (i.e. those appearing at the empirical level). It is the primary goal of CR to explain social events through reference to these causal mechanisms and the effects they can have throughout the three-layered ‘iceberg’ of reality
In the case of a smart city, the real is populated by the organisations who usher in the transition, as well as the politics and ideologies behind its implementation. To illustrate this, I have created a basic diagram of the structure of a critical realist approach to understanding the smart city (below). In the diagram I have used the EU Regional Development Fund (EURDF) as a financial backer which supports projects that work towards the EU’s Horizon 2020 sustainability goals. The EURDF money acts as a causal mechanism, in that it turns an idea into a reality, with a large number of smart city projects across the EU (including the U.K) receiving funding from it. Now, the notion of a causal mechanism is that which is “responsible for the overt behaviour of the substances” of a reality. In this imagined case, the EURDF provides the financial means for a project to be realised.
Say the city has a problem with bad traffic and wants to monitor and understand where the traffic is at its worst, at which times and which types of vehicles are most impacted by this. The city applies to the EURDF with a project to improve traffic flows and in turn improve air quality. The project is accepted and the city receives funding. The city subsequently installs a sensor network which tracks traffic flows and can identify types of vehicles through number plate recognition. The sensors can also analyse ambient air quality. It is revealed that the main cause of traffic is a major increase of private car use during 8–9 in the morning and 5–6 in the evening. At the same times of day, particulate matter count is also at its highest. From this the city can ascertain that the major cause of the traffic increase is commuters driving to and from work. Therefore, as a result, the city uses the data from the investigation into traffic and begins to track public transport with GPS. This is for the intention to develop an app for commuters to provide information into public transport and to make its use as efficient and attractive as possible, with the side effect of also improving air quality.
This fictitious example demonstrates the benefit of viewing the smart city through a critical realist lens. We can see that the existing issue of bad traffic has acted as a causal mechanism for the city to apply to the EURDF which finances the ‘actual’ realm of technology installation, resulting in an empirical output, i.e. the app and hopefully less traffic and better air quality. We can now observe that the smart city is more than just the installation of sensors or technology in a city, but a whole process which embodies three different domains of reality. By understanding the different domains in which the smart city exists, projects, governments, academics and the like can better understand how and why the smart city is accepted or conversely, rejected.
To illustrate this point, I carried out a basic, but interesting study. I read and analysed an article on the Wharton School of the University of Pennsylvania’s website, appropriately titled What’s Fueling the Smart City Backlash? As the title suggests, the article delves into the public backlash against a number of different projects, but mainly focuses upon Toronto’s now abandoned Sidewalk Labs project. Within the article I counted seventeen sources of controversy and assigned them to either the real, actual or empirical domains of critical realist reality.
Out of the seventeen controversies, only one existed in the empirical domain — when the Sidewalk Labs project released its Master Innovation and Development Plan which was met in some quarters with disdain as it was seen as “frustratingly abstract,” and viewed some of its proposed innovations as “irrelevant or unnecessary”. Another six controversies were concerns about creeping privatisation, issues surrounding democratic participation, poor resource allocation and the politics of the projects — these exist in the ‘real’ domain. Finally, the other ten concerned the ‘actual; domain, and were bound up in concerns over data security and rights, increased surveillance, potential health impacts, tracking of citizens, biases in the technologies used and privacy. These are all concerns with the unobservable elements of the smart city, i.e. the data collection and networking that one seldom encounters. In this one simple example, we can see that the domain in which technology acceptance models operate is the least populated out of the three, and therefore misses the causes of potential smart city un-acceptance, as it were.
In the example supplied by the Wharton School, the fact that the clear majority of issues with the smart city exist and emulate from the unobservable realms of reality. This is revealing. By and large a city authority isn’t seeking to track and exploit its citizenry. These concerns about the smart city may be born out of a void between the city authorities and the citizenry. Europe’s most successful smart cities, such as Barcelona and Amsterdam have citizen engagement and participation as a central component in the development of their emerging cities. In turn, these cities flourish and their smart projects face less resistance than cities which pursue a top-down imposition of technology and keep the citizens at arms length.
Therefore as a tentative first step, I’d argue for a far more citizen centric approach to the smart city and its installation, with the expressed purpose of closing the gap between authority and citizen, and in turn the abolition of half-truths and myths which in part fuel these controversies. Thankfully, many cities already have or are set to pursue this approach, with examples emanating from the relatively mature smart cities of Barcelona, Amsterdam, Helsinki and Vienna, to emerging examples such as Cardiff and Peterborough in the U.K. However, there is still a long way to go before the smart city is accepted. Yet, through this approach a city can be clear about their intentions and demonstrate that they are pure by bringing the elements of the smart city which exist in the ‘actual’ domain into the ‘empirical’ — where they can be interrogated and discussed. Conversely, if they cannot be clear, then quite rightly the project should be viewed with suspicion and remoulded or resisted.
So, to conclude. The smart city is such a diverse and vast phenomenon, with technologies that operate beyond what we perceive, a multi-domain approach, like the one presented here, is seen to be essential going forward. Technology acceptance models still have a place in the smart city, for individuals still do interact with and encounter empirical technologies such as apps and public screens. Yet, they cannot explain the whole picture of smart city acceptance. The genie is certainly out of the smart city bottle and is set to alter our cities as we know them. Therefore an understanding of smart urbanism which encapsulates it in its entirety across the different domains of reality, is becoming somewhat of a necessity. By viewing the smart city through the lens of critical realism we can begin to take a step towards a better understanding of the emerging cities in which we live.
By Will Brown
Loughborough University, U.K
1 Greenfield, A, 2017., Radical Technologies: The Design of Everyday Life, Verso, London
2. Allen, D, Brown, A, Karanasios, S and Norman, A, 2013., How Should Technology-Mediated Organizational Change Be Explained? A Comparison of the Contributions of Critical Realism and Activity Theory, MIS Quarterly , September 2013, Vol. 37, №3 (September 2013), pp. 835–854
3. Bhaskar, R, 2008., A Realist Theory of Science, Routledge, Abingdon