Understanding "Deontology" and "Utilitarianism" in Moral Dilemma Judgment - A Multinomial Modeling Approach

Abstract

Over the course of the last two decades, research on moral judgment has been heavily shaped by the application of moral dilemma research. Data obtained with this paradigm are commonly interpreted by assuming a hard split between two different kinds of processes – a “deontological” sensitivity to moral norms regardless of consequences, and a “utilitarian” sensitivity to consequences regardless of moral norms. Additionally, it is frequently assumed that these two processes arise from distinct cognitive systems, implicated in “emotional” and “rational” processing, respectively. Over the course of this thesis, I will address several methodological and conceptual assumptions of this conventional approach to understanding dilemma judgment with the application of multinomial modeling. Specifically, the current thesis investigates the impact of factors that are systematically confounded in the context of the conventional dilemma approach, and assess whether “deontological” response patterns are indeed insensitive to consequences, as the dominant conceptualization maintains (Chapter II). Results indicate the assumption of a hard split between “norms” and “consequences” (let alone “deontology” and “utilitarianism”) as determinants of dilemma judgment to be artificial and overly simplistic, and suggest that both response patterns may be understood in terms of expected consequences, and demonstrate the potential biasing impact of prominent confounds on individual response patterns. Subsequently, the current thesis assesses whether the findings of two of the presented experiments are consistent with the predictions of a model that avoids reliance on dual-process assumptions, and finds largely confirmatory evidence (Chapter III). Finally, application of the previously developed multinomial model in two additional studies (Chapter IV) further demonstrates the importance of controlling for prominent confounds such as response tendencies, as results suggest that spurious effects may otherwise arise and may be misinterpreted in the context of dual-process models. While integrating the work presented in these chapters, I discuss parallels between moral dilemma judgment and the phenomenon of “moral dumbfounding”, and subsequently integrate the empirical findings presented in this thesis with other models of moral judgments unrelated to the dilemma literature. In doing so, I suggest that the results of this thesis converge with other models developed outside the realm of dilemma research, which suggest moral judgment to be ultimately determined by the perception of harmful consequences, such that the perception of harm and immorality tend to go hand in hand. Thus, the current thesis argues against the view that moral dilemma judgments are best understood in terms of adherence to absolute norms versus impartial utilitarian calculations, and rejects associated dual-process assumptions. Instead, it proposes that moral dilemma judgments may be viewed through a consequentialist lens, a proposal which converges with evidence obtained outside the realm of moral dilemma research.