Search

Corruption has been heavily researched in the past decade, garnering attention from a number of different fields, such as political science, organizations, sociology, and information systems. As a result, focus on anti-corruption strategies has also increased. Motivated by the notion that individuals are mindfully rationalist (Palmer 2012), such policies fall under two categories: the manipulation of the incentive structure, thus increasing the cost of corruption, and enhancing transparency so that individuals will feel less inclined to engage in corruption under the watchful eye of citizens. In order to support such investments, e-government has been deemed the platform for it. However, studies in this field have been limited in their generation of theoretical frameworks. In addition, it has been observed that most of the studies have adopted a technological deterministic view to technology, overlooking any social factors which might affect the deployment of transparency. As governments are highly complex institutions, this has meant that many of the e-government initiatives have failed to yield the expected results. In regards to transparency specifically, much of the research has been normative in nature. Part of the issue lies in how transparency has been conceptualized. As identified, there are over fifteen working definitions for transparency, most of them elusive in nature, making it difficult to measure and study it. As a result, and based on Taylor and Lips's (2008) definition, it was proposed that transparency be conceptualized as an information flow, comprising of three distinct stages: creation of information, publication, and public access to information. When viewed in such a manner, it is possible to identify that most of the research conducted on transparency has been to address the publication and access phases, leaving the creational aspect of transparency vastly underresearched. Based on this, this thesis thereby proposed to adopt a socio-technical approach to technology and study a cognitive system (based on the distributed cognition theoretical - Dcog - framework - adopted) situated within a governmental agency in the Brazilian federal government (referred to as Agency X). Brazil, as a focus for this research was chosen due to its documented efforts in investing in transparency which thus far have failed to result in reduced levels of corruption. Coupled with distributed cognition, moral disengagement theory (MD) was also adopted due to the presence of ongoing acts of corruption at Agency X which had been previously identified. This thus made it the perfect setting under which to understand how transparency is created and why it has failed to deter corruption. The methodological approach was therefore a case study and, in order to analyze data effectively, the DiCoT (distributed cognition for teamwork) methodological framework was adopted. Drawing from DCog literature, DiCoT proposes a structured way in which to analyze findings, generating five different models (physical layout, artefact, information flow, social structure, and evolutionary), which together generate a powerful overview of how cognitive processes are distributed between human agents and artefacts, and what role the environment plays in such a setting. Coupled with MD, findings indicated that this cognitive system is unevenly distributed between the human and the technical components, with individuals leading the information flow and transformation, largely dictating how information is stored and processed. This has created opportunities for "breakdowns" whereby the quality of information has been compromised, thereby affecting the overall state of the transparency system. In addition, findings indicated that corruption has persisted due to three factors: (1) technical systems' failings to address local needs; (2) hierarchical structures, with unethical leaders leading the unethical decision-making at Agency X; (3) the high levels of informality. This dynamic, as findings suggested, was facilitated by moral disengagement mechanisms. The result has been a unique study that has provided an in-depth account of how transparency is created and how unethical individuals have dealt with the technical changes that resulted from transparency deployment. In doing so, this study filled a gap in the literature and provided a framework for future studies.