![Sample our Behavioral Sciences journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/110801/TOC-Subject-Sample-2021-Behavioral-Sciences.jpg"})
ABSTRACT
Within social psychology, group identification refers to a mental process that leads an individual to conceive of herself as a group member. This phenomenon has recently attracted a great deal of attention in the debate about shared agency. In this debate, group identification is appealing to many because it appears to explain important forms of intentionally shared actions in a cognitively unsophisticated way. This paper argues that, unless important issues about group identification are not illuminated, the heuristic function ascribed to this notion for an understanding of shared agency remains dubious at best and unfulfilled at worst. This paper offers such a clarification by distinguishing and describing two different mental processes that constitute group identification: adoption of the group perspective and transformation in self-understanding. It is claimed that the latter process consists in the production of what Ruth Millikan labels “Pushmi-Pullyu representations” and that it is developmentally prior with respect to the ability of adopting the group perspective.
Acknowledgments
This paper has a long pre-history. We sincerely thank Olle Blomberg who, back in 2013, first formulated the hypothesis that group identification might be related to the notion of Pushmi-Pullyu representations, thereby delivering the inspiration for this paper. Prolonged discussions with Olle on these topics substantially helped us shape the arguments of this paper. Our gratitude also goes to Danny Forde, Francesco Guala, Raul Hakli, John Michael, Don Ross, Glenda Satne, Deb Tollefsen, and Dan Zahavi, who read and commented on previous versions of this article. We presented this paper on too many occasions to be listed here and we are very grateful for the relevant feedback we received during all these events. Work on this paper has been facilitated by a Network Grant of the Suntory Foundation (2017), to which we express our gratitude. Finally, we would like to thank two referees for their important comments that helped us improve our paper.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. The socio-psychological characterization of group or social identification may serve as a preliminary or working definition of this concept. On this understanding, ‘group identification’ refers to “the process whereby an individual internalizes some form of social categorization so that it becomes a component of the self-concept, whether long-lasting or ephemeral” (Turner, Citation1982, p. 18). Integral to this definition is the idea that, when this specific transformation in self-understanding occurs – that is, when the subject acquires a social self (Brewer & Kramer, 1997) – a subjective form of group membership is established: “we are concerned here with group membership as a psychological and not a formal-institutional state, with the subjective sense of togetherness, we-ness, or belongingness” (Turner, Citation1982, p. 16).
2. This hypothesis is supported by the observation that children of 21–27 months of age encourage their partners to reengage when they disengage from joint activities. Importantly, the solicitation occurs equally often regardless of whether the goal of the activity, for its achievement, requires the causal contribution of the partner or not (in the latter case, the child would be able to reach the goal by herself). The observation is taken to show that the child understands the activity as a cooperative endeavor and the interactant as an intentional partner (Warneken, Gräfenhain, & Tomasello, Citation2012).
3. Other ToM tests (i.e., so-called “indirect” tests; see Low & Perner, Citation2012) predate the emergence of these abilities. For instance, Rubio-Fernández and Geurts (Citation2013) show that children of two and a half years of age have already acquired a ToM.
4. Accounts which include a common knowledge condition concur that mindreading abilities are required for shared intentions (see Bratman, Citation1993; Gilbert, Citation1997; Miller, Citation2001; Tuomela, Citation2007, among others). However, it is debated whether the notion of common knowledge should be understood in a deflationatory or non-deflationary sense and whether this condition is required at all. For instance, Blomberg (Citation2016) denies that shared intentions require common knowledge (see also Ludwig, Citation2016).
5. Butterfill (Citation2012) has explored another option to accommodate those two insights from developmental psychology: rather than invoking the notion of shared intentions to explain minimal joint actions, Butterfill develops a theory of these actions based on shared goals (where a goal is not the content of an intention but the outcome to which the action is directed). Our argument aligns with Butterfill’s theory insofar as we dismiss the notion of shared intention to explain minimal joint action, but it diverges from it insofar as we assign that explanatory role to PPRs.
6. These consequences are all premised on an internalistic understanding of group identification according to which group identification is an intra-mental process. Externalists about group identification need not subscribe to that idea, and hence, their position remains untouched by our arguments. For internalist versus externalist approaches in the philosophy of social sciences, see Ross (Citation2014, p. 236ff.).
7. Recent research has suggested that forms of in-group favoritism may even be antecedent. For instance, Kitzler, Corriveau, and Harris (Citation2011) have found that pre-schoolers, when learning new information, trust speakers with native accents more than speakers with different accents. Similarly, Buttelmann, Daum, Zmyj, and Carpenter (Citation2013) have shown that, from the age of 14 months, children imitate those speaking in their native language (in-group members) more faithfully than those speaking a foreign language (out-group members). However, the jury is still out on whether these findings illustrate genuine understanding of group membership (Buttelmann et al., Citation2013, p. 427).
8. List and Pettit (Citation2011) rely on this point to show that mentality and thus, for example, moral responsibility can be assigned to groups. This is a controversial point (for criticisms, see Makela & Miller, Citation2005; Szigeti, Citation2013). We will remain entirely neutral on whether groups can have intentional attitudes or bear moral responsibility, as nothing in what follows hinges on this idea. Nevertheless, we will continue to use locutions that attribute mentality (especially intentions) to groups for the sake of linguistic simplicity.
9. “We … achieve an alignment between what the group’s attitudes require of us and our own preferences and then act without further reflection whenever our response is required. The alignment may be achieved … through explicitly adopting the group’s viewpoint – that is, adopting the group’s attitudes as our own” (List & Pettit, Citation2011, p. 192).
10. It is unlikely that the recognition of the rationality of (1) will be sufficient to motivate action – for (1) to be effective, it must go together with some of the other normative or cognitive factors mentioned above, possibly including a concern for the group. We will not explore this issue further; here we are merely concerned with how an individual comes to an understanding that (1) is a rational attitude from another agent’s point of view.
11. One could conjecture that this class of actions at least partly overlaps with the class of so-called “group actions” (see List & Pettit, Citation2011). Since we want to remain neutral on whether psychological or agentive predicates can be genuinely applied to groups, we will not explore this conjecture further.
12. Millikan (Citation2004) goes on to argue that all representational states in non-human animals are PPRs. This claim is controversial because it rules out the possibility of non-human animals with DERs such as beliefs. We do not necessarily endorse this part of her theory.
13. These features are shared by what Tamar Gendler calls “aliefs” (Gendler, Citation2008, Citation2012). Aliefs are said to be more primitive than beliefs: “as a class, aliefs are states that we share with non-human animals; they are developmentally and conceptually antecedent to other cognitive attitudes that the creature may go on to develop” (Gendler, Citation2008, p. 641). They are also not combinatoric: “[alief] is molar: it has, I have suggested, a characteristic R–A–B structure, whereby a particular representation (R) comes typically to be associated with a characteristic valence or evaluation or affective state (A) and with the activation of a behavioral repertoire (B)” (Gendler, Citation2012, p. 800). We speculate that the characteristics of aliefs are so similar to those of PPRs that “aliefs” and “PPRs” are best viewed as different labels for the same states. However, the label “PPR” would be less problematic than the label “alief” from a theoretical point of view. For the theoretical problems about Gendler’s characterization of aliefs, see, for example, Bayne and Hattiangadi (Citation2013) and Currie and Ichino (Citation2012).
14. The social self is only one motivating factor to engage in a joint action, but there can be other factors that counteract or block the motivational force of the social self, meaning that even when a social self is activated, the individual may not pursue all goals that she perceives as being “ours” (for instance, if the goal is morally blameworthy).
15. Self-transformation and its role in motivating joint actions is described as if it were a finite process which, once in motion, made people operate in a completely different mode. Obviously, however, any step of this process will be influenced by circumstantial factors, which we are not in a position to determine in this paper.
16. This is not supposed to exclude the possibility for the individual to deliberately create situations that increase the likelihood of group identification. Accordingly, there is a sense in which group identification could be under the indirect control of the individual – at least to a certain extent.
Additional information
Notes on contributors
Alessandro Salice
Alessandro Salice is Lecturer at the Department of Philosophy of University College Cork. He has published on a variety of topics related to phenomenology, philosophy of mind and action, and moral psychology. His current work mainly addresses issues concerning human sociality.
Kengo Miyazono
Kengo Miyazono is Associate Professor of Philosophy at Hiroshima University. His main research areas are philosophy of mind, philosophy of psychology, and philosophy of psychiatry. He is the author of Delusions and Beliefs: A Philosophical Inquiry (2018 Routledge).
