Abstract/Summary

This thesis explores unintended consequences resulting from the use of Internet of Things (IoT) technology systems. It does this through a Critical Realist (CR) lens, which assumes a stratified ontology that distinguishes the deep generative mechanisms formed in the interaction between powers and tendencies inherent in structures and agency. The observable empirical experiences of matter and meaning contribute to uncovering the complex effects that these generative mechanisms produce, which in turn allow inferences as to their causality. The thesis utilizes examples of IoT use as applied in car parking solutions in a Smart City that were collected as part of a case study. The first contribution progresses theoretical development through the creation of a Sociotechnical Behavioural Framework (SBF) that abductively infers the mechanisms causing unintended consequences by modelling the paths to events that arise through the systematic interaction between the three sociotechnical domains of Material Agency, Human Agency, and Interpretive Schemas. The SBF is laid out in a Venn diagram where each sociotechnical domain is depicted by a circle with interconnections and with all three circles overlapping at the centre. This central three-way intersection represents an emergent property beyond purely technological effects. The powers and tendencies of the three domains of human and technology interaction produce the deep constitutive entanglement that is described in the theory of Sociomateriality (Orlikowski and Scott, 2008). As a first step, the framework analyses the sequence of interaction loops of designed and intended pathways between the three sociotechnical domains. In a second step, the work draws upon Affordance Theory and Analytical Dualism (Archer, 2010) to separate, temporally and analytically, the contribution of individual components to the recursive relationship between the practice of Material Agency, Human Agency, and Interpretive Schemas. By modelling the variations between expected and actual inputs, it becomes possible to identify the unexpected outputs and unintended consequences that arise from the interplay of the three domains. The framework can be used to analyse actual recorded examples of unintended consequences and to potentially predict future problems. It advances relevant theory by connecting the underlying roots of Sociomateriality (Orlikowski, 2000) with Affordance Theory (Gibson, 2014) and Analytical Dualism (Archer, 2010) as a route to understanding morphogenesis (i.e., the way a system changes) over time. The second contribution, in line with the work’s onto-epistemology of Critical Realism, consists of the retroductive testing of the abductive framework. This provides practical learnings and improvement suggestions through a case study using data from an actual IoT-based car parking system within the Smart City of Santander, Spain. Data collected from system designers, operators, city officials, and parking system users facilitated the identification of expected outcomes through process flow charts. The comparison between intentions and observations resulted in the identification of 30 routes leading to unintended consequences. These were classified and studied using the SBF to identify their properties, and through reasoning for their occurrence, allowed the suggestion of their generative mechanisms. Testing the framework in this way adds to the knowledge of the sociotechnical design and suggests improvement opportunities in comparable contexts. The newness of the technology, its actual availability in cities, time, and complexity, limited this exploratory case study to one detailed example. The transferability question of the single case study is strengthened in two ways: by utilizing input from actual users of the technology to understand the generative mechanisms in more depth, and through a thought experiment that is included in the thesis. The final and third contribution concerns the application of findings as a series of recommendations and a methodology to identify how the new framework could be used in practice to limit the occurrence of unintended consequences in sociomaterial systems and/or to analyse their causality and hence improve IoT and other system designs. The value of the framework lies in the fact that it offers a logical and structured way for considering where disconnects in assumptions and intentions of policy makers, business analysts, project managers, developers, users, and other stakeholders could produce unintended consequences in the development of complex systems involving sociomaterial interaction. It can provide an opportunity for advanced consideration of the potential risks that may otherwise be overlooked in phased handovers between stakeholder communities. There is an opportunity for future research to explore the use of the SBF beyond the context of IoT in a Smart City, once IoT technology becomes more widely implemented and user experience data becomes available. This could be accomplished through a Critical Realist study in which the identified generative mechanisms are retrodictively assessed and either validated or adjusted. Such an attempt at extending the SBF’s generalizability could take place in a different sociotechnical setting within the realm of the IoT or in other emergent and disruptive technologies.