Cuizhu Wang

This study explores the interaction between social norms and legal rules, inspired by operationalization of institutional beliefs and Hartian account of internal and external point of view as well as Bicchieri’s approach to social norms. Using an experimental design that combines vignettes and belief elicitation as distributions, choices about legal judgements and their different kinds of beliefs were studied. The vignettes were applied to collect choice data about participants’ legal judgements in terms of their willingness to punish (behavioral account); a Likert scale was applied to measure their perceptions of legal appropriateness (institutional beliefs), and four types of beliefs (first- and second- order descriptive and normative beliefs) were elicited as point and distribution estimates for norm elicitation. Results indicate significant differences across cases, despite the fact that scenarios relied on two types of potential conflict between social norms and legal rules interpreted literally, that is, threshold problems and phantom rules. Norm identification analysis revealed that no single, unanimously shared norm governed behavior across scenarios; rather, multiple competing norms may have influenced participants' responses. The study highlights the complexity of the interplay between legal systems and social norms and suggests directions for future research, particularly regarding more refined methods for belief elicitation and the study of professional legal participants.

There is a broad consensus in the leading general literature on norms and norm-change that norms are conditional, and based on descriptive and normative expectations. Expectations are a sub-set of beliefs. Hence some primary barriers to norm-change arise from dynamics among beliefs, and between beliefs and preferences. However, the literature has under-examined the distinction between two such barriers, preference falsification and pluralistic ignorance. We clarify the implications of the distinction for two leading conceptual frameworks (due to Kuran and Bicchieri, respectively), and ultimately for explicit models, of norms and norm change. We furthermore explain how, once this clarification is in place, the two models are naturally reconciled by interpreting the concepts of belief – and also preference – that they incorporate as based on attribution from Dennett’s intentional stance.

In this paper we argue that we have reason to use epistemic nudges to correct for biases influencing perceptions of normative expectations. There is a broad consensus in the literature on social norms that individuals’ normative behaviours are fundamentally conditional on normative (and empirical) expectations. Expectations are a sub-set of beliefs, hence normative expectations have an epistemic aspect. Our paper focuses on normative expectation which refers to an individual’s second-order belief about what most other people in her community believe she ought to do. Normative expectations can be influenced by various forms of bias which can lead individuals to overestimate or underestimate the prevalence or strength of certain norms in their community, or to conform to norms that do not provide social benefit. We propose using epistemic nudges to produce accurate normative expectations in those nudgees. We examine the sort of epistemic nudges that could do so by enabling better presentation and transparency of information.

This article expands Giuseppe Rocchè’s critique of Raz’s model of exclusionary reasons by addressing its limitations in real-world decision-making. By drawing from theories of bounded rationality and social norms, this commentary highlights how the interplay of cognitive biases and social constraints complicates the exclusionary reasoning process. Bounded rationality reveals cognitive limitations that prevent people from fully filtering biases, even when they intend to follow authority. Whereas research on social norms illustrates how normative pressures lead individuals to align their behaviour with perceived social expectations rather than act purely on rational considerations. These dual constraints underscore a fundamental tension in Raz’s framework: the interplay between individual cognitive limitations and contextual social factors. By integrating empirical and theoretical insights, this article strengthens Rocchè’s critique, offering a more comprehensive perspective on the challenges of human reasoning under authoritative guidance.

In this chapter, we integrate three distinct strands of thought to argue that the concept of “fairness” varies significantly across cultures. As a result, ensuring that human-AI interactions meet relevant fairness standards requires a deep understanding of the cultural contexts in which AI-enabled systems are deployed. Failure to do so will not only result in the generation of unfair outcomes by an AI-enabled system, but it will also degrade legitimacy of and trust in the system. The first strand concerns the dominant approach taken in the technology industry to ensure that AI-enabled systems are fair. This approach is to reduce fairness to some mathematical formalism that can be applied universally, a quintessentially Western conception of fairness. The second strand concerns alternative conceptions of fairness that have their roots in philosophical traditions of the East, namely Confucian virtue ethics. Understanding how individuals from diverse cultural backgrounds perceive fairness—particularly their beliefs about fairness in human-to-human interactions—is crucial for understanding how they will interpret fairness in human-AI interactions. Building on these philosophical and behavioral differences, as highlighted by empirical research, the third strand integrates insights from political science and interdisciplinary studies. This perspective offers valuable guidance on designing AI-enabled systems to align with contextually relevant standards of fairness. Examining existing beliefs about fairness within the context of institutional decision-making provides valuable insights into what people expect from AI-generated decisions. These expectations often include key elements such as a sufficient degree of transparency, clear lines of accountability, and mechanisms to contest decisions made by the system—all of which are essential components of procedural fairness. Rather than adopting a one-size-fits-all approach to ensuring fairness, the design and deployment of AI systems must carefully account for the operating environment, including the socio-political and cultural context, to ensure that the system aligns with the relevant standards of fairness.

This thesis is an experimental study to investigate the operationalisability of the theory of social norms provided by Cristina Bicchieri. In Chapter 1 I critically summarise a main theme from recent literature and distinguish the accounts of norms based on social preferences from accounts based on social structure. I also summarise different theorists’ accounts of social norms as a social construct, in addition to surveying some issues scholars have raised empirically. Chapter 2 reviews the conceptual analysis of social norms by Bicchieri as a social structure based account. Bicchieri’s conceptual analysis introduces three kinds of condition for norm identification. I review these in detail, and suggest hypothesis testing corresponding to each kind of condition. Chapter 2 briefly analyses a critical problem for Bicchieri’s theory. Chapter 3 provides philosophical background that supports the intentional concepts applied in Bicchieri’s analysis of social norms. I suggest that the Dennettian account of Intentional Stance is the best philosophical framework for Bicchieri’s account of social norms. I also argue that Revealed Preference Theory from economics is an application of the Intentional Stance. I conjecture that adopting the Intentional Stance and applying Revealed Preference Theory to empirical data can allow for improved operationalisation of Bicchieri’s conceptual analysis. In Chapter 4 I provide critical review of some key experimental work by Bicchieri and co-authors applying her conceptual analysis of social norms. I then provide a critical review of a widely used toolbox from the current economic literature for norm elicitation. Then I introduce a more rigorous experimental protocol for investigating social norms understood following Bicchieri’s analysis. The toolbox suggested in my thesis addresses limitations identified in Bicchieri’s empirical work. Chapter 5 presents design of the experiment administered as the core element of the thesis. Chapter 6 shows and analyses the results from the experiments described in Chapter 5. It also introduces the statistical models used in my thesis to assess the extent to which Bicchieri’s analyses successfully guides experimental identification of social norms. Chapter 7 offers concluding theoretical reflections, and discusses possible extensions of the research presented in this thesis.

This chapter has three parts. Section 1 reviews the concept of social norms developed by Bicchieri (2006, 2017). I argue that this theory offers an operational definition of social norms which allows for the possibility of their empirical investigation. Section 2 proposes a Dennettian account of the Intentional Stance (IS) as the best philosophical framework for the intentional concepts applied in CB’s conceptual analysis of social norms. I also argue that Revealed Preference Theory (RPT) from economics is an application of the IS. Hence, I synthesize the IS and RPT, and argue that adopting the IS and applying RPT to empirical data can allow for operationalization of Bicchieri’s conceptual analysis of social norms. To align with the understanding provided in section 1 and section 2, section 3, on the basis of concerns about limitations arising from experimental work operating Bicchieri’s theory, it explains the Quadratic Scoring Rule (QSR) for eliciting subjective belief distributions. Three reasons will be provided to explain why the QSR is the best tool supporting the operationalization of Bicchieri’s conceptual analysis of social norms.

Some of our beliefs are institutional; that is, beliefs whose content is to a large extent shaped by institutions, such as beliefs about intellectual property, trade policy, or traffic rules. In this chapter, we propose a novel account of institutional beliefs, as we call them. In particular, we argue that institutional beliefs are primarily attributable to social entities, such as groups or collectives, and only secondarily to individual agents. This is because institutional beliefs respond to specific problems that, in principle, only collectives face. However, existing methodology in behavioural science primarily relies on eliciting beliefs through individual choice data, which presents challenges for empirically modelling our proposed concept of institutional beliefs. To address this challenge, we propose to speak of institutional beliefs attributable to individual agents as ‘qua beliefs’. That is, institutional beliefs attributable to individual agents acting qua members of certain institutionally shaped groups. Within this framework, individuals’ qua actions occur in a shared realm of problems that can only be addressed through collective efforts. The relationship between individuals and their groups can be rendered tractable under formalised conditions. Under these conditions, choice data collected from individuals need not obstruct the elicitation of institutional beliefs for empirical study at both the collective and the individual level. We refer to these conditions as “IS conditions”, which are developed based on the synthesis between an interpretationist understanding of mental states and standard models of revealed preference theory as used in experimental economics. For an illustration of how this approach can be operationalised experimentally, we direct our readers to chapter 10 of this volume, where an experimental study applied our account of institutional beliefs.