Silvia Ivani

Should scientists value simple theories? Is fruitfulness an important criterion to assess scientific theories? What is the role of moral, social, and political values in science? In recent years, there has been an increasing interest among philosophers of science in studying how cognitive and non-cognitive values influence and should influence the assessment and comparison of scientific theories. While cognitive values (e.g. simplicity and fruitfulness) are features of scientific theories that are indicative of the truth or empirical adequacy of theories, non-cognitive values are moral, political, social, and economic values. Understanding the roles of values in science is a particularly urgent issue. Clarifying the importance of cognitive values is important in order to be able to make accurate comparisons of scientific theories. Understanding the influence of non-cognitive values on science is crucial because moral, social, and political values are involved in many stages of research, such as decisions on methodologies and allocation of funds. Since these decisions affect all the members of the society (scientists, non-experts, and political institutions), understanding the impact of non-cognitive values on these choices is of primary importance. Although a large body of literature has investigated the roles of cognitive and non-cognitive values in theory appraisal, several questions remain to be analysed.

In this dissertation, I address four questions by using two main methods, namely the analysis of case studies from Evolutionary Psychology and the experimental method. First, philosophers have clarified the importance and roles in theory appraisal of some values such as simplicity, whereas little or no attention has been paid to other values, such as fruitfulness. In the second chapter of my dissertation, I start fill this gap by formulating a clear explication and a strategy to be employed in theory appraisal for the value of fruitfulness. Moreover, I use my account to assess the fruitfulness of Evolutionary Psychology.

Second, some philosophers of science argue that cognitive values are desirable and relevant to the assessment of theories because they are indicative of the truth or empirical adequacy of theories. However, why is it so? In the third chapter of my dissertation, I develop a context-sensitive approach to values. I argue that in order to understand the ground for the desirability of cognitive values and make an accurate appraisal of theories, we have to consider specific factors of the context in which theories are assessed, such as the availability of methodologies and the way cognitive values are interrelated to each other in that context.

Third, philosophers have traditionally argued that the influence of non-cognitive values on scientific reasoning threatens the epistemic authority of science. In the third chapter [sic], I challenge this view and argue that some non-cognitive values can play a cognitive role in science, i.e., they can be epistemically beneficial to the assessment of scientific theories. On the basis of a case study (the account of human mating in Evolutionary Psychology), I argue that feminist values can positively contribute to theory appraisal in various ways, such as by raising sensitivity to evidence that was neglected because of gender bias.

Fourth, some philosophers claim that non-cognitive values play a legitimate role in cases of inductive risk, namely cases in which scientists may wrongly assess scientific hypotheses (e.g. accepting a hypothesis that should be rejected) because of uncertainty. Mistakes can have consequences that can be morally or economically undesirable and non-cognitive values, some philosophers argue, provide the standards to evaluate and compare these possible consequences. However, little is known on how - specifically - non-cognitive values influence this evaluation. To address this issue, I conduct an experimental study clarifying how personal features, political values, and specific aspects of a risk (e.g. the chance of incurring Type I vs. Type II errors) determine people’s reactions to possible cases of inductive risk. On the basis of these results, I discuss recent calls made by institutions to align research agendas to citizens’ values, needs, and expectations.