“The Group Knobe Effect”: evidence that people intuitively attribute agency and responsibility to groups

Abstract

In the current paper, we present and discuss a series of experiments in which we investigated people's willingness to ascribe intentions, as well as blame and praise, to groups. The experiments draw upon the so-called "Knobe Effect". Knobe [2003. "Intentional action and side effects in ordinary language." Analysis 63: [190][191][192][193][194] found that the positiveness or negativeness of side-effects of actions influences people's assessment of whether those side-effects were brought about intentionally, and also that people are more willing to assign blame for negative side-effects of actions than they are to assign praise for positive side-effect of actions. Building upon this research, we found evidence that the positiveness or negativeness of side-effects of group actions influences people's willingness to attribute intentions to groups (Experiment 1a), and that people are more willing to assign blame to groups for negative side-effects of actions than they are to assign praise to groups for positive side-effects of actions (Experiment 1b). We also found evidence (Experiments 2a, 2b, 3 and 4) that the "Group Knobe Effect" persists even when intentions and blame/ praise are attributed to groups non-distributively, indicating that people tend not to think of group intentions and group blame/praise in distributive terms. We conclude that the folk are collectivist about group intentions, and also about the blameworthiness and praiseworthiness of groups.

The Group Knobe Effect revisited: epistemic and doxastic side-effect effects in intuitive judgments concerning group agents

“…Our results confirmed our expectations-we successfully replicated Experiment 4 originally conducted by Michael and Szigeti (2019) and extended their result by finding a group variant of the EKE and DKE. However, we think it is worth noticing that the sizes of the effects we observed were smaller than those obtained by M&S. 9 This, of course, does not mean that they are theoretically unimportant (especially that we cannot claim that our estimation of the real effect size is more accurate than what M&S reported).…”

“…What needs to be done is therefore obtaining the GKE in a way that is incompatible with the distributive interpretation. Michael and Szigeti (2019) conducted the first study of GKE (a set of experiments focused on folk intuitions about group agency in Knobe-like scenarios). The study consists of a series of experiments, of which we take Experiment 4 and Experiment 2b to be the most informative.…”

“…Our first study was a replication and extension of Experiment 4 conducted by Michael and Szigeti (2019). We chose to replicate this experiment first because, according to M&S, it was the most straightforward way of putting the hypothesis about the GKE to test among the studies they carried out.…”

“…We explain how the existence of the Group Knobe Effect and its epistemic and doxastic counterparts affects the philosophical debate on collective agency and intentionality and supports the intuitiveness of realism about collective agency among laypeople. We also critically assess the reasoning presented by Michael and Szigeti (2019) in favor of the realist-collectivist interpretation of their results (as opposed to the realist-distributivist interpretation). We argue that a thorough analysis of both their data and our new findings shows a rather wide range of differing intuitions among laypeople regarding the status of groups as agents.…”

A model-based approach to social ontology

“…Multiple experiments are typically required in order to arrive at unambiguous data (Hacking 1983), some generalizations are uninformative or trivially true (Cummins 2000), and the proximate results of many simple “toy models” and simulations only seem to teach us about possibilities, not about the way the world actually is (Fumagalli 2016; Grüne-Yanoff 2009). Given that social ontology rarely involves conducting experiments or carrying out simulations (however, see Michael & Szigeti 2019), I will not discuss these scientific practices in more detail in this paper. Rather, I will concentrate on the roles that models and theories play in scientific practice.…”