Non Experts: Which Ones Would Trust You?

ABSTRACT

Following Goldman’s seminal work, most contemporary philosophical contributions on the novice-expert relation have adopted a normative, expert-focused approach. In this paper, we aim to shift the focus of the philosophical analysis towards the characteristics of the novices, and how they might determine the choices that experts make. On the bases of recent empirical evidence from social psychology, we discuss how novices evaluate the messages that they receive and distinguish diverse kinds of novices according to their competence in message assessment. Building on that analysis, we discuss the difficulties of approaches to expertise that focus only on the standpoint of novices or assume novices are homogeneous. In our analysis, we introduce the standpoint of experts, and we pay special attention to the heterogeneity of novices. This approach allows us to identify and address the difficulties faced by experts in the context of science communication. In the last part of the paper, we characterise and discuss the problem of experts when choosing a strategy in the issuing of a public campaign to advise or inform certain populations of novices.

KEYWORDS:

• Expertise
• novice-expert problem
• science communication
• Goldman

Acknowledgments

We would like to thank Javier González de Prado, the participants in EPSA21 at the University of Turin, and two anonymous reviewers for their helpful comments.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

1. As it came up in the reviewing process, this is but one possible interpretation. A different interpretation of the aggregation of these criteria can be found, for example, in Martini (2019) and Martini and Andreoletti (2021). Under this interpretation, those criteria identify ‘proxies’ for expertise and provide the basis for estimating the expertise of a putative expert. It is considered that, under normal conditions, high levels on these (and others that might be missing) criteria combined correlate with a high degree of expertise.

2. It should be noted that, although much of the current discussion on the matter revolves around Goldman’s timely contribution, many relevant philosophical analyses of expertise preceded Goldman’s work (see, for instance, Crease and Selinger 2006).

3. In this paper, we mainly focus on the epistemic characteristics of novices. Nonetheless, as a reviewer brought to our attention, novices can also be classified along sociocultural traits (e.g. social values), which may significantly influence message assessment. An exhaustive taxonomy of novices should also take sociocultural factors into account.

4. As pointed out by one of the reviewers, in the limit, a competent novice that has sufficient knowledge could become an ‘interactional expert’ (Evans and Collins 2010). An interactional expert is a person that, although lacking contributory expertise in certain domains, is able to interact interestingly with the participants in the relevant scientific debates.

5. Fabrication of fake controversies is just one of the strategies by means of which science denialism might undermine message assessment (Hansson 2017). Other relevant strategies are cherry-picking of data, misrepresentation of denied claims, and deviant criteria of assent.

6. Martini and Andreoletti (2021), who address scientific controversies, propose a heuristic strategy for identifying bogus versus genuine scientific debates. This strategy, which relies on secondary evidence about discordant sources (e.g. authors’ field of expertise), could mitigate the impact of bogus controversies on message assessment.

7. In the present discussion of the expert/2-message problem, for the sake of argument, we will assume that the expert has the best available knowledge on the relevant matter and is diligent. Those cases in which the putative expert is not competent or trustworthy are beyond the scope of this first exploration. Nonetheless, we suspect that the previous work following the Goldman’s approach, which pays special attention to the pathologies of experts, will provide valuable resources for addressing those complex scenarios.

8. For the purposes of this model, we assume that complex messages involve elaborate reasoning about causes (for example, mechanistic reasoning), and usually require some background knowledge on basic scientific facts. An analysis of the transmission of a respiratory disease via aerosols, for example, could constitute an instance of a complex message. Simple messages, in turn, just refer to general causal links (without specifying the underlying causal processes) and require no prior background. Unspecific advice regarding mask wearing at all times or washing hands to help prevent a disease can be both seen as examples of simple messages. We also assume that generally, there will be a trade-off between the simplicity of a message and its accuracy. Complex messages usually include qualifications about the conditions under which a causal relationship holds. Nonetheless, those qualifications are usually omitted in simple messages, which tend to be sound unduly deterministic.

9. This result can also be (equivalently) expressed in terms of reasoning competence. When the number of novices with high reasoning competence – i.e. (HR,HS) plus (HR,LS) – is greater than the number of novices with low reasoning competence – i.e. (LR,HS) plus (LR,LS) –, the complex and accurate message is the most effective strategy. Nonetheless, the simple and inaccurate message is more effective when the number of novices with low reasoning competence is greater than the number of novices with high reasoning competence. When the number of novices with high reasoning competence is equal to the number of novices with low reasoning competence, both strategies are equally effective.

Additional information

Funding

This work was supported by the Italian Ministry of University and Research under Grant n. 201743F9YE (PRIN 2017 project ‘From models to decisions’) and the Spanish Ministry of Science and Innovation under Grant n. PID2021-125936NB-I00 (project ‘Evidence and mechanisms in the social sciences’).

Notes on contributors

Saúl Pérez-González

Saúl Pérez-González is assistant professor at the Department of Philosophy of the University of Valencia. Previously, he held a post-doctoral position at the Center for Logic, Language, and Cognition (LLC) of the University of Turin. His main areas of interest are philosophy of science, philosophy of the social sciences, and philosophy of the biomedical sciences.

María Jiménez-Buedo

María Jiménez-Buedo is a lecturer at the Department of Logic, History and Philosophy of Science, UNED. She works in the philosophy of the social sciences, with an emphasis on methodological issues. Her recent work focuses on evidence-based policy and on the experimental methods in the social sciences.