This thesis focuses on a topic that has received considerable attention in recent years, both in philosophy and in cognitive science: causal selection. Experimental studies have shown that various factors—such as norms, agent knowledge, and the causal structure of the situation —affect people's attribution of different degrees of causality to events with the same dependency relationship to the outcome. But the significance of this phenomenon goes beyond empirical research. Traditional attempts to provide a metaphysical theory of causation, which mainly focus on establishing the conditions under which an event is a cause of an effect rather than a non-cause, are called into question by evidence of systematic distinctions within the very set of events that these theories label as causes, raising the issue of whether and how these distinctions should be incorporated into their normative frameworks. Empirical studies on causal cognition have also fueled the philosophical debate about technical rather than ordinary notions of causality, such as those employed in legal contexts. Indeed, disclosing the mechanisms behind causal selection opens up new perspectives for exploring the features of the concept of legal causation and assessing whether they track that of everyday life. However, the aim of my work is not to provide a detailed and comprehensive summary of these discussions, but to attempt to address three questions that might contribute to them. The first question is: why do we select causes? To answer this, I outline a hypothesis about the psychological function of this practice—the Optimal Control Hypothesis. This not only offers a qualitative account of the effects of norms on our causal judgments, but also explains them in a way that is consistent with an interventionist theory of causality, thereby providing a justification for a normative analysis that encompasses causal selection. According to this hypothesis, the role of the various norms that influence our judgement—namely, both prescriptive and descriptive norms—is to shape our expectations. Controlling these expectations is precisely the purpose of causal selection: to identify the event that, more than any other, can make a difference to the outcome we expect in a given situation. However, more research is needed to fully shed light on our expectations and how norms and possibly other factors influence them. For example, although there is agreement that each of the different types of norm influences causal judgments in a similar way, their influence is less clear in a situation where more than one norm is at play. This leads to the second question: when both prescriptive and statistical norms inform a situation, how do they interact in influencing causal judgments? To contribute to an answer, I will present two new studies. Their results will be used to test two hypotheses: Bear and Knobe's (2017) hypothesis that the average of prescriptive and statistical considerations contributes to an undifferentiated notion of normality, and the predictions that Icard et al.'s (2017) model would make about our causal judgments under this assumption. Finally, I will focus on how theorizing about the ordinary function of causal reasoning— in particular the Optimal Control Hypothesis—can clarify more specialized causal concepts. I will therefore try to explicate the connection between the folk and legal conceptions of causality by arguing that their features can be mapped onto each other, as both follow the same rationale of maximizing control over events. By addressing these questions in three distinct papers, each corresponding to a different chapter, this thesis aims to take a step towards a more comprehensive understanding of causal selection, bridging the gap between empirical and normative investigations, while opening up new avenues for exploring the complex interplay of factors that shape our causal reasoning.

INVESTIGATING CAUSAL SELECTION: CAUSAL JUDGMENTS BETWEEN NORMS AND EXPECTATIONS

D'AMICO, MARINA ANTONIA
2025

Abstract

This thesis focuses on a topic that has received considerable attention in recent years, both in philosophy and in cognitive science: causal selection. Experimental studies have shown that various factors—such as norms, agent knowledge, and the causal structure of the situation —affect people's attribution of different degrees of causality to events with the same dependency relationship to the outcome. But the significance of this phenomenon goes beyond empirical research. Traditional attempts to provide a metaphysical theory of causation, which mainly focus on establishing the conditions under which an event is a cause of an effect rather than a non-cause, are called into question by evidence of systematic distinctions within the very set of events that these theories label as causes, raising the issue of whether and how these distinctions should be incorporated into their normative frameworks. Empirical studies on causal cognition have also fueled the philosophical debate about technical rather than ordinary notions of causality, such as those employed in legal contexts. Indeed, disclosing the mechanisms behind causal selection opens up new perspectives for exploring the features of the concept of legal causation and assessing whether they track that of everyday life. However, the aim of my work is not to provide a detailed and comprehensive summary of these discussions, but to attempt to address three questions that might contribute to them. The first question is: why do we select causes? To answer this, I outline a hypothesis about the psychological function of this practice—the Optimal Control Hypothesis. This not only offers a qualitative account of the effects of norms on our causal judgments, but also explains them in a way that is consistent with an interventionist theory of causality, thereby providing a justification for a normative analysis that encompasses causal selection. According to this hypothesis, the role of the various norms that influence our judgement—namely, both prescriptive and descriptive norms—is to shape our expectations. Controlling these expectations is precisely the purpose of causal selection: to identify the event that, more than any other, can make a difference to the outcome we expect in a given situation. However, more research is needed to fully shed light on our expectations and how norms and possibly other factors influence them. For example, although there is agreement that each of the different types of norm influences causal judgments in a similar way, their influence is less clear in a situation where more than one norm is at play. This leads to the second question: when both prescriptive and statistical norms inform a situation, how do they interact in influencing causal judgments? To contribute to an answer, I will present two new studies. Their results will be used to test two hypotheses: Bear and Knobe's (2017) hypothesis that the average of prescriptive and statistical considerations contributes to an undifferentiated notion of normality, and the predictions that Icard et al.'s (2017) model would make about our causal judgments under this assumption. Finally, I will focus on how theorizing about the ordinary function of causal reasoning— in particular the Optimal Control Hypothesis—can clarify more specialized causal concepts. I will therefore try to explicate the connection between the folk and legal conceptions of causality by arguing that their features can be mapped onto each other, as both follow the same rationale of maximizing control over events. By addressing these questions in three distinct papers, each corresponding to a different chapter, this thesis aims to take a step towards a more comprehensive understanding of causal selection, bridging the gap between empirical and normative investigations, while opening up new avenues for exploring the complex interplay of factors that shape our causal reasoning.
14-mag-2025
Inglese
GUALA, FRANCESCO
Università degli Studi di Milano
114
File in questo prodotto:
File Dimensione Formato  
phd_unimi_R13269.pdf

embargo fino al 20/11/2026

Dimensione 1.03 MB
Formato Adobe PDF
1.03 MB Adobe PDF

I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.14242/210044
Il codice NBN di questa tesi è URN:NBN:IT:UNIMI-210044