1 Introduction and Overview
In contemporary philosophy of collective intentionality, affective states such as emotions, feelings, moods, and sentiments, do not figure prominently in debates on the explanation and justification of joint action. Received philosophical theories analyze joint action in terms of common knowledge of cognitively complex, interconnected structures of intentions and action plans of the participants (e.g. Gilbert 1989, 1996, 2003, 2014b; Searle 1990, 1995, 2010; Bratman 1992, 1999, 2009, 2014; Tuomela 1995, 2002, 2007, 2013). These theories admit that collective emotions sometimes give rise to joint action or more typically, collective behavior that falls short of full-fledged joint intentional action. Such cases may involve the joy of football players who celebrate the goal scored by one of the team members, or the rage of protesters who destroy public property in their fury. On the one hand, insofar as emotionally motivated actions proceed without mediating joint intentions, they do not raise the interest of received philosophical theories that require higher-order intentional states for joint action. On the other hand, if collective emotions give rise to joint intentions that proximately give rise to joint action, then the analysis of the latter seems to suffice. In sum, the received theorists seem to bracket individuals’ mental states that precede the formation of their joint intentions as irrelevant for the necessary and sufficient conditions of joint intentions.
Yet even they seem to be aware of the affective dimension of typical joint actions. Thus Michael Bratman lists friendship, love, and “the joys of conversation”, along with singing and dancing together, as basic forms of sociality that he intends to study in his new book Shared Agency (2014). Although the affective phenomenology of these activities is familiar to anyone who has engaged in them, Bratman’s and the other received theorists’ sophisticated frameworks have meager resources for capturing the constitutive affective dimension of friendship and love, or the joys of conversation for that matter.
In contrast, the so-called minimalist theorists of joint action have paid some attention to the affective elements of joint action. They seeks to articulate a minimal architecture of representations and processes that is capable of accommodating joint action in its various forms, of which cognitively complex adult human cooperation is a special case (e.g. Sebanz and Knoblich 2009; Vesper et al. 2010; Knoblich et al. 2011). Several authors refer to empirical evidence on an association between lower-level synchrony and alignment in perceptual, motor, and behavioral processes in joint action on the one hand, and increased interpersonal liking, feelings of solidarity, and cooperativeness on the other hand (e.g. Tollefsen and Dale 2012; Pacherie 2014). Michael (2011) has notably distinguished several ways in which shared emotions contribute to coordination between participants in joint action. These include the functions of emotion detection, emotion/mood contagion, empathy, and rapport.1 His notion of shared emotion however cannot fully address the roles of emotions in joint action. In contrast, our account of shared emotions shows how these emotions function not only as coordination smoothers but as motivating and justifying reasons for joint actions. Moreover, our account enables us to see that collective emotions, by virtue of their intentionality and embodiment, range across the divide between higher and lower levels of representations and processes whose opposition has dominated the debate between received and minimalist theories of joint action.
In what follows, we discuss collective emotions in joint action in two contexts. We lay ground for the discussion by outlining a philosophical account of collective emotions (Section 2). Then we move on to joint actions, beginning from the role of collective emotions in their explanation and justification (Section 3). We argue that collective emotions can function as both motivating and justifying reasons for jointly intentional actions, in some cases even without prior joint intentions of the participants, as well as give rise to intentions-in-action envisaged by received theories. Second, we address the coordinating functions of collective emotions in joint action, supplementing minimalist theorists’ remarks – in particular Michael’s (2011) – on the basis of our more elaborate account of collective emotions (Section 4). In the concluding Section 5, we offer some suggestions on how existing philosophical accounts of joint action, both cognitively complex and minimalistic, should be recast in light of our account.
2 What are Collective Emotions?
The notion of “shared emotion” is ambiguous. On the one hand, sharing of emotion refers to a phenomenon in which one person’s expressed emotion is perceived by another person. This is the sense that Michael (2011) and some theorists of empathy (e.g. Rimé 2007) have in mind when they discuss shared emotions. Thus, Michael gives two necessary conditions for sharing an emotion between two persons, x and y:
x expresses his affective state (verbally or otherwise);
y perceives this expression.
On the other hand, the notion of sharing refers to several individuals experiencing an emotion of the same type and content, such as celebrating the success of their favorite team with other fans, with mutual awareness of their respective emotional state (joy). Most philosophical and empirical accounts of shared and collective emotions have been interested in the latter kind of phenomena, invoking the notion of sharing in this sense. Likewise, we suggest the following formulation: x and y share an emotion, or equivalently have a collective emotion if,
x and y experience an emotion of the same type with similar (1) evaluative content and (2) affective experience.2
x and y are mutually aware that (c).
Our main contribution in this paper is an elaboration of these necessary yet not sufficient conditions in terms of the degree of collectivity.
We first suggest that the collectivity of emotions should be understood as a continuum rather than as an on/off question.3, 4 This is because two main dimensions of emotions, evaluative content and affective experience, allow for a continuum in terms of their sharedness. Moreover, emotions involve action tendencies such as attack in anger and apologizing in guilt, as several emotion theorists have emphasized (Frijda 1986, 2002; Griffiths 2004; Scarantino 2010). In the collective context, these action tendencies come with more or less preparedness for joint action, and we suggest that this preparedness connects with the collectivity of the other dimensions of emotion (see Section 4 for discussion). Here we first highlight different ways in which individuals can share the evaluative content of emotion, and then discuss the sharing of affective experiences (see Salmela 2012, 2013 for more detailed accounts).
There is a wide interdisciplinary agreement among emotion researchers that emotions could not exist without underlying concerns (e.g. Frijda 1986; Ortony et al. 1988; Goldie 2000; Scherer 2001; Roberts 2003; Helm 2008; Schmid 2009). A concern in this context is a technical term which refers to a desire, goal, norm, or value – a representation with the world-to-mind direction of fit (see Roberts 2003 for a representative account) – which however need not have the form of a propositional attitude. For example, when Alex feels afraid while climbing a 30 meter wall, there is an underlying concern, which is his safety; when he feels a joy at seeing his children eating a wholesome meal, his underlying concern is their healthy growth. By the same token, when a group of people experience a collective emotion, they have some shared concern. Sharing the evaluative content of an emotion is thus a matter of appraising the particular object of emotion similarly with others on the basis of shared concerns, where the degree of sharedness can vary (see below). The appraisal process need not be collective though it can in some cases be, such as when an emotional appraisal is formed as a result of public discussion (see Halperin 2014). However, emotional appraisals are often so fast that it is impossible to make, let alone commit oneself to, them collectively. Although Gilbert (2002, 2014a) has defended a position of this kind, it is more reasonable to think that we can come to share the underlying concerns of collective emotions in importantly dissimilar ways, some of which involve commitment.5
In the weakest form of collectivity, people share a concern if they have overlapping private concerns. Insofar as people pursue their own survival, security, attachment, health, happiness, and attachment, these are private concerns. Tuomela (2007) calls concerns of this kind plain I-mode concerns. Individuals can establish groups whose members cooperate in promoting their convergent private concerns. Groups of this kind may include economic sharing groups and self-help groups, such as dieting groups, alcoholics anonymous, and so on. The fact that private concerns or goals are general or even universal among all humans does not amount to their collectivity but merely to their commonality, which is a different thing.
Concerns can be shared in a somewhat stronger sense when individuals are privately committed to some concern [in part because of] believing that the others in the group have the same concern, and also believing that this is mutually believed in the group. Thus, for instance, if John as a Liverpool fan is concerned about the future of this prestigious football club, not only does he believe that the other Liverpool fans have the same concern, but also that the other fans believe the same about his and other fans’ having the concern in question. The bracketed clause refers to the fact that many of our shared concerns (as well as beliefs) are socially grounded. That is, we come to have concerns because we believe that other members of our group have them, where this belief is either a reason or a cause or both for my adopting the same concern. The commitment is still private, but the concern is shared with others, unlike in the first case where these too are private. Tuomela has characterized this type of collectivity as pro-group I-mode or, more recently (Tuomela 2013), weak we-mode collectivity. In our terminology, concerns of this kind are moderately collective. The main point is that the commitment is still up to the individual to revise and renounce for private reasons alone. Groups based on concerns of this kind may include loose associations, such as unorganized fan groups or social and religious movements.
The strongest mode of collectivity in sharing concerns is founded on the group members’ collective commitment, either explicit or implicit. In addition, there is a mutual belief among the group members that they share the concern to which they have collectively committed themselves. Through their collective commitment, the group members adopt the concern as theirs in a strong we-mode sense. Collective commitment provides the group members group reasons to think, want, feel, and act in ways that are in accordance with their shared concern. Moreover, the group members are allowed to revise their commitment to the shared concern only by reasons that are acceptable from the group’s point of view. Collective commitment implies a Collectivity Condition according to which the group members necessarily “stand or fall together” when acting as group members. For instance, individual players of a team win a match if and only if their team wins the match. No group is by definition an example of a strong we-mode group because that depends on its meeting the criteria of collective commitment, groups reasons, and Collectivity Condition. Candidates of such groups include organized fan groups, religious groups, workgroups, theater ensembles, bands, orchestras, and families.
The idea of emotions having reasons may strike one as strange. However, a closer look reveals that emotions have both justifying and motivating reasons; indeed this idea is accepted in contemporary philosophy of emotions and meta-ethics (e.g. Greenspan 1988; D’Arms and Jacobson 2000; Helm 2008; Brady 2011). Emotions have justifying reasons as they are liable to evaluation in terms of their appropriateness like other intentional attitudes. In general, an emotion is justified if its particular object has properties that render it fitting for an emotional response of a certain kind. For instance, my fear is appropriate if the object of my fear has properties that render it capable of inflicting significant harm on me (or some others I care about). Justifying reasons of collective emotions may refer to collectively accepted shared attitudes of a group, such as its values, goals, or intentions that were at stake in the emotion-eliciting situation. The same group attitudes may also serve as motivating reasons of collective emotions as group members evaluate situations in terms of their significance to the group’s values, goals, or intentions. Since emotions are not under voluntary control, there is no deliberation on reasons to feel in the same way as there is deliberation on reasons to act. We can nevertheless talk about reasons (rather than mere causes) of emotion because human emotions are not reflexes or fixed action patterns but flexible responses to cognitively processed situational meaning.
The typology of shared concerns explains how the the evaluative content of an emotion comes to be shared, but how is an affective experience shared? Ritual theories of collective emotions (Durkheim 2001; Collins 2004) highlight the alignment of bodily and behavioral aspects of individual emotional responses – physiological changes, facial expressions, action tendencies, and subjective feelings – as the central features of collective emotions. Causal processes that contribute to the synchronization of emotional responses and experiences between interacting subjects include attentional deployment (Collins 2004; Brosch 2014), emotional contagion (Hatfield et al. 1994), facial mimicry (Bourgois and Hess 2008), motor mimicry and imitation (Chartrand and Bargh 1999), and neural mirroring (Decety and Meyer 2008). Although these are all relevant mechanisms through which emotions come to be shared, as some minimalist theorists of joint action have recently observed, we argue that shared concerns provide a rational impetus to the causal processes of affective synchronization. A non-reflective absorption in shared affective experience may sometimes take the form of a phenomenological fusion of feelings into our feeling that Schmid (2009) highlights as the core of collective emotions. We do not believe though that a phenomenological fusion of feelings alone indicates strong collectivity because it seems possible to experience such a fusion with importantly dissimilar shared concerns, which makes it insufficient in distinguishing between collective emotions of different kinds.
Feelings of emotion face both inwards and outwards: they are felt as bodily sensations that emerge from the various organismic and behavioral changes that partially constitute emotion, but they also infuse our intentional representations of particular objects in the world in typical emotional experiences where the intentional object of emotion is also the object of attention (see Salmela 2005; Lambie and Marcel 2002). In these cases, the shared affective experience can be phenomenologically described as “feeling together that the object or occurrence in question matters” to the involved individuals, as Sánchez Guerrero (2013, p. 180) suggests. Phenomenologically, interpersonal affective synchrony gives rise to mutual feelings of togetherness and rapport (e.g. Hove & Risen 2009; van Baaren et al. 2009; Kirschner & Tomasello 2010; Reddish et al. 2013). We believe that mechanisms of affective synchronization provide an important enabling condition for the togetherness-aspect of shared intentional feelings as well.
Finally, beginning from the weakest type of collective emotion, we suggest that these emotions emerge when a group of individuals appraise the emotion-eliciting event convergently in relation to their overlapping private concerns, and such mechanisms as attentional deployment, emotional contagion, facial and motor mimicry, and behavioral entrainment synchronize the individuals’ emotional responses, producing a shared affective experience among individuals who are mutually aware that others are feeling the same.6 Moderately collective emotions have moderately shared concerns as the underlying ground of emotional appraisals. These concerns are constitutive of a social identity or a group in terms of which the individuals identify themselves. A shared affective experience emerges through similar mechanisms. Shared group membership reinforces the synchronization process, adding to the intensity of the shared affective experience. Moderately collective emotions are functionally and phenomenologically experienced in the role of a group member, but the group membership is normatively weak because it is self-appointed and maintained through a private identification or commitment. Finally, strongly collective emotions emerge when members of the group appraise the emotion-eliciting event convergently in relation to their strongly shared concern. The strong collectivity of an emotion is reflected both in the degree of synchronization of individual emotional responses into a shared affective experience and in the evaluative content of the group members’ emotions. The members of a winning team do not rejoice merely in winning the championship but instead in “our winning the championship” or in “our accomplishment” where the group membership is normative in a strong sense. In this way, collective content in the sense of indexicality and mode are built into collective emotions of the strongest kind.
Together, shared evaluative content and shared affective experience constitute the two dimensions of collective emotions. By virtue of these two dimensions, collective emotions are capable of (1) providing both motivating and justifying reasons for joint action and (2) functioning as coordinating smoothers in joint action. In the next two sections, we show how our account of collective emotions improves on existing accounts of joint action in both respects, thereby bridging the gap between the received and minimalist views.
3 Collective Emotions as Motivating and Justifying Reasons of Joint Action
Our discussion in this section focuses on two issues: (1) collective emotions as motivating reasons of joint action with collectively intentional structure, and (2) collective emotions as justifying reasons of joint action. Regarding the first issue, our goal is to demonstrate that our account of collective emotions satisfies all the desiderata for collective intentionality specified by the received philosophical theories of joint action. Regarding the second, our goal is to show that collective emotions can provide justifying reasons for joint action that emerges from those emotions. This question has not been discussed in the received theories.
3.1 Collective Emotions as Motivating Reasons of Collectively Intentional Joint Action
Philosophical theories of joint action maintain that joint actions are more than aggregates of individual actions and therefore their intentionality is irreducible to that of individual actions. Moreover, these theories assume that the collectively intentional structure of joint actions can be explicated by defining the shared or joint intentions that both psychologically motivate and rationally justify instances of joint action.7 Thus an instance of collective behavior qualifies as joint intentional action if and only if it is founded on a shared intention of the participants. In spite of their differences, these theories have some common requirements for the intentional structure of joint action.8 Pacherie (2013) names these as (1) the common outcome requirement; (2) the individual intentional action requirement; (3) the common goal requirement; (4) the action coordination requirement; (5) the intentional action coordination requirement; and (6) the joint goal requirement. We call the theories with these characteristics as the received view of joint action.
The common goal requirement imposes that the common outcome requirement and the individual intentional action requirement are related in a particular way, namely that there is a certain outcome O to which several agents contribute and that the actions of each agent involved are intentional under the description of bringing about the outcome O. Yet actions that satisfy these three requirements are not joint as they allow actions in which the agents contribute to the same goal which they have in common only in a distributive sense, unbeknownst to each other and without coordination. Accordingly, the action coordination requirement and its modified form of intentional action coordination requirement demand coordination among the agents’ actions, which can be emergent and involuntary in the former case but must be voluntary and directed towards the agents’ shared goal in the latter case. Pacherie states that actions that meet the requirements (1–5) are jointly intentional in a weak sense in which agents intentionally coordinate the means they use to achieve a common goal. However, philosophical theories of joint action typically require that the goal of a joint action is collective in a stronger sense. The joint goal is achieved together or, as in some theories, there is also a collective commitment to the goal as our goal.9 Here the jointness of actions is not just a means to a goal but a part of the goal itself.
The received theories admit that individuals may have diverse motivations to participate in joint intentions and joint actions, some of which can be emotional. However, these theories pay no attention to emotions, either individual or collective, among the myriad of possible motivations of joint action. The idea seems to be that whatever other mental states agents have, joint intentions are the primary and sufficient motivating reasons for joint action. On this view, collective emotions can give rise to joint action by giving rise to joint intentions that mediate between collective emotions and joint action. Indeed, (Gilbert 2002, 2014a), whose theory of collective intentionality requires joint commitment as one of the central conditions for joint action, suggests that collective emotions involve a joint commitment to the emotion itself as well as to acting in accordance with the emotion. This means for example that when individuals are jointly committed to grieving as a body, they are committed to maintaining a somber outlook and to speaking in a silent voice. While Gilbert is right in allowing collective emotions to motivate and rationalize joint intentional action, this modeling of collective emotions on joint intention, or bracketing the former more generally, fails to do justice to the important motivating and justifying roles of collective emotions in joint action.
To make our case, we draw on Pacherie’s (2002) analysis of individual emotional action, and extend it to cases of joint emotional action. Pacherie’s typology of emotional actions is particularly useful in distinguishing between importantly dissimilar actions, some of which have an intentional structure even if they are motivated by emotions.
The category that best fits the accounts of received joint action theorists is what Pacherie calls semi-deliberate emotional actions. An emotional action of this kind emerges for example if a person gets angry at his or her oppressive treatment but there is none to punish directly to get the situation right, which leads the person to form a conscious intention to join an NGO that fights against the particular kind of oppression. Here the situation in which the emotion emerges and the situation in which the action tendency that emerges from the emotion finds its realization are different, or at least there is some means-end reasoning leading from the initial situation to its subsequent phase. Still, it is the emotional evaluation of the situation and the adjacent action tendency that provide the motivating reason for action.
It is easy to see why Pacherie’s semi-deliberate emotional actions fit nicely with received philosophical theories of joint action. There are mediating joint intentions between the collective emotions of individuals and their joint action. Collective emotion motivates individuals to join their forces in planning their joint action; joint intentions with meshing individual subplans are formed in this process; and those plans are executed with motivation that derives at least in part from the underlying collective emotion. As in the individual case, collective emotions provide motivating reasons for these jointly intentional actions. Even if joint intention and/or the participatory individual intentions of the agents are proximate reasons for joint action, the chain of reasons goes back to the collective emotion and – ultimately – to the shared concern of the group members. Moreover, even if the same joint action can in some cases be produced by joint intentions that are not based on a collective emotion, emotions may be necessary as motivating reasons in other cases.
One of these cases is collective guilt, perhaps the most studied kind of collective emotion so far (see e.g. Gilbert 2002; Branscombe and Doosje 2004; Tollefsen 2006; Konzelmann Ziv 2007; Ferguson and Branscombe 2014). This is an emotion that individuals feel upon some perceived wrong that the subjects themselves or members of their in-group have committed, typically, to an outgroup. Collective guilt motivates apologetic and restorative joint actions towards the wronged outgroup. Even if the same actions could be motivated by joint intentions formed without collective guilt, we would doubt the sincerity of those actions because the underlying emotion indicates that the actions spring from the ethically correct motives of remorse and regret. Stocker (1976) pointed out in a now classic essay in moral psychology that emotions are sometimes necessary as motives of morally virtuous action even if the same action such as visiting a friend in hospital could be performed from other motives. The example of collective guilt suggests the same: collective emotions are sometimes necessary as the right kind of motivating reasons of joint action.
More problematic cases are impulsive emotional actions, such as punching someone in a bar brawl, running away in fright, and hugging one’s fellow supporters in joy at the victory of one’s favorite football team (all Pacherie’s examples), that people often think of as paradigmatic examples of emotional actions. Instead of involving separate representations of goals and means, there is a collapse of evaluative and executive representations so that objects of emotions are represented in terms of properties such as horribleness, danger, oppressiveness, etc. that are also injunctions to act in accordance with the appraisal (see also Griffiths 2004; Scarantino 2010). To quote Pacherie (2002, p. 77), “there is no need for intermediate reflexive states, no need for the agent to first become consciously aware of his emotional state and to consciously formulate a plan for dealing with the situation”. The spontaneity of impulsive emotional actions is reflected also in their sense of passivity and experienced lack of voluntariness.10
Collective emotions can provide motivating reasons for impulsive joint actions insofar as those actions are done for reasons. One might however question the collective intentionality (i.e. the presence of an intentional structure specified above) of impulsive emotional actions that are spontaneous and done without regard for their consequences beyond the immediate situation. Our answer to such a skepticism is that it depends on the degree of collectivity of the emotion in question.
In the weakest sense of collectivity, where the underlying shared concerns of collective emotions are overlapping private concerns, the ensuing impulsive collective behavior does not qualify as jointly intentional action. There is no joint or even common goal, nor any coordination between individuals insofar as everyone seeks to protect his or her interests, as in the case of panicked shareholders who rush to sell their stocks in the fear of an economic meltdown. As the result of the shareholders’ impulsive emotional collective behavior, stock prices plummet and everyone loses. Moreover, the actions of individual shareholders that contribute to this outcome are not intentional under the description of bringing about that outcome together with other shareholders – quite the contrary: individual shareholders intended to save their own skins by selling their stocks. The example indicates that impulsive collective behavior motivated by collective emotions of the weakest type is not jointly intentional as it has no joint or common goal, nor coordination between the agents.11
Strongly collective emotions in contrast can motivate jointly intentional action. These emotions are founded on strongly shared concerns to which individuals have committed themselves collectively as group members. Strongly shared concerns provide the group members group reasons to feel emotions in situations where the members believe or perceive that their shared concern is affected favorably or adversely. Those collective emotions also provide motivating reasons for the group members to act in accordance with the action tendency of their collective emotion. Most importantly, it seems that emotions of this kind are capable of initiating collectively intentional joint action spontaneously, without prior deliberation on goals and means, and the group-level joint intentions and participatory intentions of individual group members to do their parts of the joint action are formed implicitly during rather than prior to acting together.
Received theorists have made room for this possibility of joint intentions emerging in action rather than prior to it. Searle (1983) introduced the distinction between a prior intention and an intention-in-action to accommodate the execution of prior intentions as well as the intentionality of spontaneous individual actions, and the same distinction can be applied to joint intentional actions (Searle 2010). Thus there are prior we-intentions to act, and we-intentions-in-action during the performance of a joint action. In a like manner, Tuomela (2007, p. 109) states that a joint intention to perform X jointly “can be understood liberally so as to allow that the participants need only have an intention-in-action to participate in the collective performance of X”. However, the notion of an intention-in-action merely accommodates such a possibility within the theoretical framework of the received view, telling us little about how we-intentions-in-action can actually arise.12 We claim that strongly collective emotions provide one such mechanism underlying spontaneous joint actions.
More specifically, we suggest that the action tendencies of strongly collective emotions provide a plausible mechanism to initiate and maintain joint intentions. The intentional object of emotion provides the target of an action and the emotion type specifies the action type – escape or avoidance of danger in fear; retaliating or punishing an offender in anger; apologizing and/or compensating to a victim in guilt, and so on. Strongly collective emotions give rise to impulsive yet collectively intentional action especially in situations that afford ways of satisfying the action tendency of the emotion on the spot. Through their intentional structure, these emotions also specify the group members an outcome to which the members contribute under the description of bringing about that outcome in emergent and/or voluntary coordination with each other. Moreover, a strongly collective emotion gives rise to a joint intention without an additional collective commitment to it. Instead, the collective commitment to the underlying shared concern of the emotion carries over to the relevant joint intention. One of Pacherie’s examples of impulsive emotional action, joyful celebration of the victory of one’s favorite team together with other fans of the team, may qualify as an example. The fans need not form a prior joint intention to celebrate the team’s victory even if they may anticipate it toward the end of the game if their favorite team is leading. Instead, celebration with songs, chants, and cheers breaks immediately after the game has ended in the team’s victory. Here we have a case of collectively intentional joint intention formed while acting on a strongly (or even moderately) collective emotion of the individuals. We discuss how our proposal relates to that of minimalist theorists in Section 4.
3.2 Collective Emotions as Justifying Reasons of Joint Action
The evaluative-cum-motivational structure of emotions allows them to provide reasons for action even if the agent is not aware of emotion as his or her reason at the time of acting. This lack of transparency undermines the status of emotions as reasons for action according to internalist theories of justification as Pacherie (2002) points out. However, she highlights adaptiveness as another independent and perhaps more important requirement for justification. “For an emotion not just to causally explain an action but also to rationalize it, it must be the case that: (1) the emotion is appropriate, that is, that the emotional assessment of the situation is correct, and (2) the action generated by the emotion is indeed adaptive, that it helps the end of the agent” (Pacherie 2002, p. 85; see also Greenspan 1988). The same criteria can be explicated for collective emotions as justifying reasons of joint action.
We suggest that a collective emotion justifies a joint action when (1) the emotion is appropriate, that is, the emotional assessment of the situation that individuals face as a group is correct from the group’s point of view, and (2) the emotion generates a joint action that promotes the shared concern of the group members. The key task of this proposal is to spell out the appropriateness of collective emotions. One of us (Salmela 2014) has argued that a collective emotion is appropriate if it is felt for a group reason that emerges from a set of internally coherent constitutive group attitudes whose aspects have not been adopted or maintained by ignoring evidence that is available to the group members. Two main aspects of this proposal are group reasons and the internal coherence of constitutive group attitudes.
Shared concerns provide justifying group reasons for collective emotions in situations in which the group members’ shared concerns are affected favorably or adversely. For instance, when a shared goal is in danger, the group members have group reasons to fear, or when the group members have reached their goal by their own efforts, they have group reasons to feel proud. However, the appropriateness of an emotional assessment from the group’s point view requires also that the shared concern is coherent with the group’s wider epistemic and normative perspective. The constitutive group attitudes must be in synchronic and diachronic coherence with each other, and these attitudes cannot be maintained by ignoring or neglecting evidence that is available to the group members.13
Criterion (2) requires joint actions generated by collective emotions to promote the shared concerns of the group members. Here the typology of collective emotions in terms of their collectivity is again important. In the weakest type of collective emotions, such as in the case of shareholders’ fear of a market meltdown, there are no robustly shared concerns but only overlapping private concerns. Accordingly, each shareholder is concerned about minimizing his or her losses. This concern justifies the shareholders’ action to sell their stocks qua individuals, but not qua group members because they all lose as the consequence of their aggregate action. Instead of protecting the shareholders from danger, acting on their weakly collective fear results in the realization of their fears. This indicates that collective behavior motivated by weakly collective emotions often cannot be justified by those emotions as it tends to harm the overlapping private concerns of the involved individuals. Contagious reactions to overlapping private concerns probably have evolutionary advantages, as in the case of a herd of impalas fleeing from a lion, but the analogous reactions by humans in modern contexts seem often harmful, though this ultimately depends on the context.
The situation is different with moderately and especially strongly collective emotions. Here we have groups with more robustly shared concerns of the members. Accordingly, joint actions motivated by collective emotions of these kinds promote the group members’ shared concerns by default rather than contingently. Semi-deliberate emotional joint actions may be more reliably adaptive than impulsive emotional joint actions from the group’s perspective as the former kind of joint actions allow for more planning, division of labor, and voluntary coordination in the action. Still, impulsive emotional joint actions can be adaptive from the group’s perspective as well. This is the case especially if the emotion-eliciting situation calls for immediate joint action. Here the collectively intentional structure of those actions – their having a joint outcome to which the group members contribute under the description of bringing about that outcome in coordination with each other – is central for those joint actions to be beneficial. These joint actions also benefit from more extensive interpersonal synchrony in comparison to semi-deliberate emotional joint actions. In the next section we relate our proposal to the minimalist theories, which highlight the process of this synchrony.
4 Collective Emotions as Coordination Smoothers in Joint Action
In this section, we focus on the role of collective emotions as coordination devices in joint action. Our discussion criticizes and expands the observations of minimalist theories of joint action, in particular John Michael’s. We begin with a brief general description of the minimalist approach, followed by Michael’s account of shared emotions as coordination smoothers in joint action. Our positive contribution here is founded on the point that the processes of alignment and synchrony studied by minimalist theories are constitutive rather than contingent aspects of collective emotions. Therefore collective emotions can not only initiate joint action “bottom up”, but also orchestrate it “top down”, facilitating coordination.
Minimalist theorists of joint action argue that the focus of received philosophical analyses of joint action “is on the coordination of agents’ intentions prior to acting, but they pay little heed to the perceptual, cognitive, and motor mechanisms enabling people to coordinate online” (Pacherie 2014, p. 30; see also Tollefsen 2014). Minimalists are eager to accuse philosophers of cooking up “heavily mentalistic stews” in their “solitary armchairs” (Sebanz and Knoblich 2009, p. 365). Criticisms of this kind are partially misdirected, though, since both high- and low-level mechanisms of coordination are typically involved in joint action, and should be accounted for in comprehensive explanations of joint action. Moreover, most mainstream philosophical theorists are aware of the limitations of their analytic method: their silence concerning low-level mechanisms of coordination does not reflect a dismissal of the importance of such mechanisms, but rather an acknowledgement that the study of these mechanisms is better left to competent empirical researchers. In sum, there is no reason why mainstream analytic approaches should not be complemented by empirical as well as theoretical research aiming to articulate the minimal architecture of representations and processes that is capable of accommodating joint action in its various forms, of which cognitively and normatively complex adult human joint action is only a special case. Knoblich et al. (2011) for example provide a comprehensive review of psychological research on joint action behavioral entrainment, action affordances, perception-action matching, and action simulation as sources of emergent coordination, and shared task representations and joint perceptions as processes that contribute to planned coordination. Their own proposal (with Cordula Vesper in Vesper et al. 2010) for a minimal architecture of joint action highlights, similarly, shared task or goal representations; monitoring and prediction processes that operate on these representations; and coordination smoothers that reliably simplify coordination. One type of coordination smoother is synchronization of movements that makes oneself or the other more predictable. Behavioral synchrony also constitutes the core of Tollefsen and Dale’s (2012) minimalistic account of joint action.
Tollefsen and Dale (2012) point out several problems in mainstream philosophical theories of joint action concerning the execution, implementation and multiple initiation of joint action, as well as the overly intellectual quality and verifiability of those theories. To address these problems of received theories, the authors present what they call the “alignment system”, “a heterogenous mix of diverse components entraining two or more people to specific patterns of behavior”. Instead of providing a theoretical account on this system, Tollefsen and Dale present bits and pieces of evidence on lower-level processes that contribute to behavioral entrainment during interaction. They refer, for example, to studies on alignment of body posture during naturalistic conversation (Shockley et al. 2009); coupling of visual attention during interaction (Richardson et al. 2007); the mirror neuron system (Rizzolatti and Craighero 2004); priming theories of alignment (Pickering and Garrod 2004); and emergent self-organizing produced through “coordinative structures” of two or more people, as suggested by dynamic system theories. Together, Tollefsen and Dale refer to these and other lower-level mechanisms of entrainment in joint action as “surface synchrony” that underlies and sustains “deep commitments” to higher-order goals and intentions but can also “kick-start” joint actions. Tollefsen and Dale mention collective emotions as an example of lower-level processes that can initiate joint actions, such as conversation. In contrast, our account of collective emotions (presented in Section 2) assigns them a non-reductive, functional-theoretical role of producing and maintaining wide-ranging synchrony and alignment between individuals. Since Michael (2011) is the only minimalist who offers a comparable functional account of shared emotions, we discuss first his account and then show ours extends his in a crucial way.
Michael discusses behavioral synchrony in shared emotions in the context of emotional contagion and rapport.14 Emotional contagion is defined by Hatfield et al. 1994, p. 5) as “the tendency to automatically mimic and synchronize facial expressions, vocalizations, postures, and movements with those of another person, and, consequently, to converge emotionally”. Michael notes that contagion contributes to cognitive and bodily alignment, and through its effects on attention and memory, renders particular objects and features of the environment more salient and some stored information more accessible, thereby facilitating prediction of each other’s behavior. Unconscious behavioral mimicry contributes also to rapport, interpersonal liking that emerges from reciprocal expressions of positive sentiments of individuals towards each other. Michael argues that rapport facilitates coordination of movements in joint action in several ways. It makes prediction of movements similar to one’s own easier and reinforces similarity through increased synchronization; contributes to prediction and monitoring by increased attention to each other’s movements; fosters a sense of commitment that influences behavior similarly to explicit commitments; and associates with normative pressure to conform to group norms and the tendency to punish dissenters.
Although Michael’s remarks on the emotional underpinnings of behavioral alignment and synchrony are more advanced than other minimalists his notion of sharing an emotion is rudimentary as it does not even require that the individuals have emotions of the same type; merely that I express my affective state (which can be an emotion, a feeling, or a mood) and you perceive it, and – possibly but not necessarily – vice versa. True enough, in emotional contagion, one person’s emotional expression causes another to enter into an affective state of the same type, such as fear, anger, joy, and so on. Importantly, however, in emotional contagion, the intentional objects of individuals’ affective states, insofar as they have any, are different individual tokens of the same type, as is the case with collective emotions of the weakest kind, such as shareholders’ fear for their investments. Yet without a shared intentional object, the synchronizing and coordinating effects of shared emotions remain weaker than when there is one.
To begin with, joint attention to the shared object of emotion is a constituent rather than a consequence of collective emotion when the individuals are co-present in the emotion-eliciting situation (see Collins 2004, 2014). Moreover, with moderate and strong forms of shared concerns comes group affiliation that influences processes of “surface synchrony”, surveyed by Tollefsen and Dale (2012). There is recent evidence that “emotional mimicry is not a ‘blind’ imitation of any given display but rather a social process that depends on the interactional context” (Hess et al. 2014, p. 96). Thus, we mimic facial, vocal, and postural expressions of emotion more with those with whom we affiliate by virtue of a shared group membership than with out-group members or, even less so, our enemies. Just as different aspects of emotional mimicry are connected, the same finding applies to behavioral mimicry and synchrony, insofar those behaviors are part of emotion such as spontaneous protest against injustice in anger rather than contingent behaviors like foot tapping. There is also evidence that the synchronizing effects of collective emotions extend from expressive behavior to cognitive processes. Epistemically, collective emotions lubricate creative processes, speeding imagination and recombination of ideas, thus enabling group flow, “a state of heightened consciousness, sharpened attention, and total immersion in the task at hand”, as Parker and Hackett (2012) point out. Indeed, collective emotions are unique in their capacity to produce and maintain an orchestrated set of synchronous processes that together function as coordination smoothers of joint action. This explains why collective emotions so often function as bottom-up elicitors of joint action.
Interpersonal synchrony seems to be strongest in impulsive emotional joint actions motivated by strongly collective emotions, which gives those emotional joint actions an advantage over semi-deliberate ones in terms of spontaneous coordination. On the other hand, deliberation and planning often compensate for the lesser degree of interpersonal synchrony in semi-deliberate emotional joint action. Success of joint action is thus facilitated by keeping the collective emotion active until the opportunity for a planned joint action in accordance with the emotion emerges, or if this is impossible, by reinvigorating the original collective emotion by conscious means at the time of acting. Indeed, this is a well-known emotion regulation strategy of political activists of all sorts (see e.g. Jasper 2014).
It is important to realize that strongly shared concerns that underlie strongly collective emotions may also contribute to the degree of interpersonal synchrony by providing a rational impetus to causal processes of synchronization. In weakly collective emotions with underlying private concerns, there is no rational impetus to synchronization; only synchrony produced by emotional contagion between the co-present individuals that may contingently contribute to an adaptive collective behavior of the individuals. However, a collective emotional response to perceiving or knowing that our shared concern is at stake may activate the joint action tendencies of the group members rather than the individual action tendencies that belong to emotions with underlying private concerns.15
Though we have emphasized the rational impetus for synchrony above, in typical cases of joint action where people are co-present, collective emotions proximately function as coordination smoothers by immediately and affectively rewarding joint action. Specifically, these rewards arise from sharing emotions with others during joint action. Note that this motivational role of shared emotions as affective rewards differs from their role as motivating reasons at the intentional level of explanation where shared emotions motivate specific joint actions such as celebration in joy or apologizing in guilt. Although Michael and other minimalist theorists have paid attention to the affective consequences of behavioral synchrony and alignment, synchrony is only one mechanism that produces affective rewards in joint action. With Marion Godman, we (in Godman et al. 2014) argue that acting together produces rewards that render all joint actions – not merely those motivated by collective emotions – prima facie intrinsically rewarding and thus motivating for the participants.
5 Conclusion: How Collective Emotions Complement Received and Minimalist Theories of Joint Action
The debate between received and minimalist theories of joint action is largely one between high-level and low-level accounts. Whereas received theories operate firmly on the level of complex cognitive, motivational, and normative representations, minimalist theories invoke low-level representations and processes as supplementary or, in some cases, alternative to the ones of high-level models. Insofar as shared or collective emotions are mentioned in this discussion, they are treated as states that belong either to the higher level, as in Gilbert’s theory, or to the lower level, as suggested by Tollefsen and Dale. Yet both treatments are incomplete insofar as emotions range across the divide between higher and lower levels. On the one hand, they associate with the higher level representations through their underlying concerns that in a group context involve shared goals, values, or norms of the group members. Emotions are also responsive to conceptual information and group reasons emerging from the group’s constitutive concerns that Tuomela (2007) calls “a group ethos”. At the same time, emotions are also embodied in lower-level states such as distinct neural, physiological, behavioral, and motor changes as well as feelings emerging from these changes.
Collective emotions bridge also another, temporal divide between received and minimalist theories. The former offer detailed analyses of the preconditions of joint action, whereas the latter focus on mechanisms of coordination during joint action. This temporal divide within the research topic has reinforced separation between the two approaches as a kind of “division of labor”. Yet the perspective of collective emotions shows the arbitrariness of this division as emotions are processes that extend from the preparation of joint action to its execution. Even if collective emotions precede joint action only contingently, they non-contingently emerge during such action, orchestrating synchronous processes that function as coordination smoothers in joint action, and with their rewards, motivating us to engage in further instances of joint action. Accordingly, collective emotions provide an excellent window to understanding how the planned and emergent aspects of coordination come together in joint action.
In this paper we have opened this window by showing how collective emotions can provide both motivating and justifying reasons for joint intentional actions, either with mediating prior joint intentions of individuals, or with joint intentions-in-action that emerge when the action elicited by collective emotion is already underway. Moreover, we have extended and amended Michael’s (2011) analysis of the ways in which collective emotions serve as coordination smoothers in joint action. The assimilation of shared intentional objects to collective emotions allows us to see how joint attention and joint action tendencies become coordinated as intrinsic elements rather than as contingent consequences of collective emotion. Finally, we highlighted the unique capacity of collective emotions to produce and maintain an orchestrated set of synchronous processes that together function as coordination smoothers of joint action.
Our results provide a compelling reason for the unification of received and minimal approaches to joint action, as they demonstrate that both remain incomplete without the other. It is not sufficient to pay lip service to the importance of different approaches while focusing on one’s own research. The received and minimalist approaches have to be put together in a complete analysis of joint action. Only then can we hope to understand such complex social phenomena as friendship and love, singing and dancing together, or the joy of conversation that involve coordination at many levels, both mental and bodily.
We wish to thank Elizabeth Pacherie, John Michael, Hans Bernhard Schmid, Michael Schmitz, and two anonymous reviewers of the Journal of Social Ontology as well as audiences at Collective Intentionality IX conference, Indiana University, Bloomington, Sep 10–13; Philosophy of Science seminar, University of Helsinki, Oct 20, 2014; and “The We and Its Emotions” workshop, Center for Subjectivity Research, University of Copenhagen, Nov 12, 2014 for their generous insightful and helpful comments to earlier versions of this paper.
Bacharach, M. (2006): “Beyond Individual Choice”. In: N. Gold and R. Sugden (Eds.). Princeton: Princeton University Press.Google Scholar
Brady, M. (2011): “Emotions, Perceptions, and Reasons”. In: C. Bagnoli, (Ed.): Morality and the Emotions. Oxford: Oxford University Press, p. 135–149.Google Scholar
Branscombe, N. R. and B. Doosje (Eds): (2004): Collective Guilt. An International Perspective. Cambridge: Cambridge University Press.Google Scholar
Bratman, M. (1992): “Shared Cooperative Activity”. In: The Philosophical Review 101, p. 327–341.Google Scholar
Bratman, M. (1999): Faces of Intention. Cambridge: Cambridge University Press.Google Scholar
Bratman, M. (2009): “Shared Agency”. In: C. Mantzavinos (Ed.): Philosophy of the Social Sciences: Philosophical Theory and Scientific Practice. Cambridge: Cambridge University Press, p. 41–59.Google Scholar
Bratman, M. (2014): Shared Agency: A Planning Theory of Acting Together. New York: Oxford University Press.Google Scholar
Brosch, T. (2014): “Neurocognitive Mechanisms of Attentional Prioritization in Social Interactions”. In: C. von Scheve and M. Salmela (Eds.): Collective Emotions. Oxford: Oxford University Press, p. 78–93.Google Scholar
Chartrand, T. L. and J. A. Bargh (1999): “The Chameleon Effect: The Perception-Behavior Link and Social Interaction”. In: Journal of Personality and Social Psychology 76, p. 893–910.Google Scholar
Collins, R. (2004): Interaction Ritual Chains. Princeton and Oxford: Princeton University Press.Google Scholar
Collins, R. (2014): “Interaction Ritual Chains and Collective Effervescence”. In: C. von Scheve and M. Salmela (Eds.): Collective Emotions. Oxford: Oxford University Press, p. 299–311.Google Scholar
D’Arms, J. and D. Jacobson (2000): “Sentiment and Value”. In: Ethics 110, p. 722–748.Google Scholar
Decety, J. and M. Meyer (2008): “From Emotion Resonance to Empathic Understanding: A Social Developmental Neuroscience Account”. In: Development and Psychopathology 20, p. 1053–1080.CrossrefGoogle Scholar
Durkheim, E. (2001): The Elementary Forms of Religious Life. A new translation by C. Cosman. Abridged with an introduction and notes by M.S. Cladis. Oxford: Oxford University Press.Google Scholar
Ferguson, M. and N. R. Branscombe (2014): “The Social Psychology of Collective Guilt”. In: C. von Scheve and M. Salmela (Eds.): Collective Emotions. Oxford: Oxford University Press, p. 251–265.Google Scholar
Frijda, N. H. (1986): The Emotions. Oxford: Oxford University Press.Google Scholar
Frijda, N. H. (2002): “Emotions as Motivational States”. In: European Review of Philosophy 5: Emotion and Action (ed. by E. Pacherie), p. 11–32.Google Scholar
Gilbert, M. (1989): On Social Facts. Princeton: Princeton University Press.Google Scholar
Gilbert, M. (1996): Living Together: Rationality, Sociality, and Obligation. Lanham: Rowman & Littlefield.Google Scholar
Gilbert, M. (2003): “The Structure of the Social Atom: Joint Commitment as the Foundation of Human Social Behavior”. In: F. Schmitt (Ed.): Socializing Metaphysics. Lanham: Rowman & Littlefield, p. 39–64.Google Scholar
Gilbert, M. (2014a): “How we Feel: Understanding Everyday Collective Emotion Ascription”. In: C. von Scheve and M. Salmela (Eds.): Collective Emotions. Oxford: Oxford University Press, p. 17–31.Google Scholar
Gilbert, M. (2014b): Joint Commitment: How we make the Social World. New York: Oxford University Press.Google Scholar
Godman, M., M. Nagatsu, and M. Salmela (2014): “The Social Motivation Hypothesis of Prosocial Behavior”. In: Philosophy of the Social Sciences 44, p. 563–587.Google Scholar
Goldie, P. (2000): The Emotions. Oxford: Oxford University Press.Google Scholar
Greenspan, P. S. (1988): Emotions and Reasons. London & New York: Routledge.Google Scholar
Griffiths, P. (2004): “Is Emotion a Natural Kind”? In: R. C. Solomon (Ed.): Thinking about Feeling. Oxford & New York: Oxford University Press, p. 233–249.Google Scholar
Halperin, E. (2014): “Collective Emotions and Emotion Regulation in Intractable Conflicts”. In: C. von Scheve and M. Salmela (Eds.): Collective Emotions. Oxford: Oxford University Press, p. 281–295.Google Scholar
Hatfield, E., J. Cacioppo, and R. Rapson (1994): Emotional Contagion. New York: Cambridge University Press.Google Scholar
Hess, U., S. Houde, and A. Fischer (2014): “Do We Mimic What We See or What We Know”? In: C. von Scheve and M. Salmela (Eds.): Collective Emotions. Oxford: Oxford University Press, p. 94–107.Google Scholar
Huebner, B. (2011): “Genuinely Collective Emotions”. In: European Journal for Philosophy of Science 1, p. 89–118.Google Scholar
Jasper, J. J. (2014): “Emotions, Sociology, and Protest”. In: C. von Scheve and M. Salmela (Eds.): Collective Emotions. Oxford: Oxford University Press, p. 341–355.Google Scholar
Kirschner, S. and M. Tomasello (2010): “Joint music Making Promotes Prosocial Behavior in 4-Year-Old Children”. In: Evolution & Human Behavior 31, p. 354–364.Google Scholar
Knoblich, G., S. Butterfill, and N. Sebanz (2011): “Psychological Research on Joint Action: Theory and Data”. In: Psychology of Learning and Motivation 54, p. 59–101.Google Scholar
Michael, J. (2011): “Shared Emotion and Joint Action”. In: Review of Philosophy and Psychology 2, p. 355–373.Google Scholar
Michael, J. and E. Pacherie (2014): “On Commitments and Other Uncertainty Reduction Tools in Joint Action”. In: Journal of Social Ontology 1, p. 89–120.Google Scholar
Ortony, A., G. Clore, and A. Collins (1988): The Cognitive Structure of Emotions. Cambridge: Cambridge University Press.Google Scholar
Pacherie, E. (2002): “The Role of Emotions in the Explanation of Action”. In: European Review of Philosophy 5, Emotion and Action (ed. by E. Pacherie), 53–92.Google Scholar
Pacherie, E. (2011): “Framing Joint Action”. In: Review of Philosophy and Psychology 2, p. 173–192.Google Scholar
Pacherie, E. (2013): “Intentional Joint Agency: Shared Intention Lite”. In: Synthese 190, p. 1813–1839.Google Scholar
Pacherie, E. (2014): “How does it Feel to Act Together”? In: Phenomenology and the Cognitive Sciences 13, p. 25–46.Google Scholar
Parkinson, B., A. Fischer, and A. Manstead (2005): Emotion in Social Relations. New York: The Psychology Press.Google Scholar
Pickering, M. J. and S. Garrod (2004): “Toward a Mechanistic Psychology of Dialogue”. In: Behavioral and Brain Sciences 27, p. 169–190.Google Scholar
Reddish, P., R. Fischer, and J. Bulbulia (2013): “Let’s Dance Together: Synchrony, Shared Intentionality, and Cooperation”. In: PLoS ONE 8. No. 8, p. e71182.Google Scholar
Richardson, D.C., R. Dale, and N. Z. Kirkham (2007): “The Art of Conversation is Coordination: Common Ground and the Coupling of Eye Movements during Dialogue”. Psychological Science 18, p. 407–413.Google Scholar
Rimé, B. (2007): “Interpersonal Emotion Regulation”. In: J. J. Gross (Ed.): Handbook of Emotion Regulation. New York: Guilford, p. 466–485.Google Scholar
Roberts, R. (2003): Emotions. Cambridge: Cambridge University Press.Google Scholar
Salmela, M. (2005): “Emotional Feelings as Twofold Representations”. In: Å. Carlson (Ed.): Philosophical Aspects of Emotions. Stockholm: Thales, p. 274–295.Google Scholar
Salmela, M. (2013): “The Functions of Collective Emotions in Social Groups”. In: H.B. Schmid and A. Konzelmann Ziv (Eds.): Institutions, Emotions, and Group Agents. Springer Studies in the Philosophy of Sociality, Vol. 2. Berlin: Springer, p. 159–176.Google Scholar
Salmela, M. (2014): “The Rational Appropriateness of Collective Emotions”. In: G.B. Sullivan (Ed.): Understanding Collective Pride and Group Identity London and New York: Routledge, p. 21–33.Google Scholar
Sanchez Guerrero, H. A. (2013): “Feelings of Being-Together and Caring-With”. In: H.B. Schmid and A. Konzelmann Ziv (Eds.): Institutions, Emotions, and Group Agents. Springer Studies in the Philosophy of Sociality, Vol. 2. Berlin: Springer, p. 177–193.Google Scholar
Scarantino, A. (2010): “Insights and Blindspots of the Cognitivist Theory of Emotions”. In: British Journal for the Philosophy of Science 61, p. 729–768.Google Scholar
Scherer, K. R. (2001): “Appraisal Considered as a Process of Multilevel Sequential Checking”. In: K. R. Scherer, A. Schorr and T. Johnstone (Eds.): Appraisal Processes in Emotion. Oxford and New York: Oxford University Press, p. 92–120.Google Scholar
Schmid, H. B. (2009): Plural Action. Essays in Philosophy and Social Science. Contributions to Phenomenology Vol. 58. Berlin: Springer.Google Scholar
Searle, J. R. (1983): Intentionality. Cambridge: Cambridge University Press.Google Scholar
Searle, J. R. (1990): “Collective Intentions and Action”. In: P. Cohen, M. Morgan and M. Pollack (Eds.): Intentions in Communication. Cambridge, MA: MIT Press, p. 401–415.Google Scholar
Searle, J. R. (1995): The Construction of Social Reality. New York: The Free Press.Google Scholar
Searle, J. R. (2010): Making the Social World: The Structure of Human Civilization. Oxford: Oxford University Press.Google Scholar
Sebanz, N. and G. Knoblich (2009): “Prediction in Joint Action: What, When, and Where”. In: Topics in Cognitive Science 1, p. 353–367.Google Scholar
Shockley, K., D. C. Richardson and R. Dale (2009): “Conversation and Coordinative Structures”. In: Topics in Cognitive Science 1, p. 305–319.Google Scholar
Tollefsen, D. (2006): “The Rationality of Collective Guilt”. In: Midwest Studies in Philosophy 30, p. 222–239.Google Scholar
Tollefsen, D. (2014): “A Dynamic Theory of Shared Intention and the Phenomenology of Joint Action”. In: S. R. Chant, F. Hindriks and G. Preyer (Eds.): From Individual to Collective Intentionality. New York: Oxford University Press, p. 13–33.Google Scholar
Tuomela, R. (1995): The Importance of US: A Philosophical Study of Basic Social Notions. Stanford: Stanford University Press.Google Scholar
Tuomela, R. (2002): The Philosophy of Social Practices: A Collective Acceptance View. Cambridge: Cambridge University Press.Google Scholar
Tuomela, R. (2007): The Philosophy of Sociality. New York: Oxford University Press.Google Scholar
Tuomela, R. (2013): Social Ontology. New York: Oxford University Press.Google Scholar
van Baaren, R., L. Janssen, T. L. Chartrand and A. Dijksterhuis (2009): “Where is Love? The Social Aspects of Mimicry”. In: Philosophical Transactions of the Royal Society B 364, p. 2381–2389.Google Scholar
von Scheve, C. and S. Ismer (2013): “Towards a Theory of Collective Emotions”. Emotion Review 5, p. 406–413.Google Scholar
Vesper, S., S. Butterfill, G. Knoblich and N. Sebanz (2010): “A Minimal Architecture for Joint Action”. Neural Networks 23, p. 998–1003.Google Scholar
In a recent paper, Michael and Pacherie (2014) analyze the role of moral emotions such as shame and guilt in reinforcing commitment to joint action. Important as they are, these emotions are social rather than collective.
Phenomenologically, the evaluative content and affective experience of an emotion are typically intertwined and intentionally directed at the particular object of emotion. Still, while the evaluative content of an emotion is necessarily intentional, the affective experience is only contingently so. For elaboration, see below.
There are other philosophical theories of collective emotions, most notably those of Gilbert (2002), Schmid (2009), and Huebner (2011). For a discussion and critique of the former two, see Salmela (2012).
The collectivity of emotions must also be distinguished from sociality even if most collective emotions occur in social contexts, because social emotions such as anger, envy, and guilt are collective only if they are shared with others.
While shared concerns are the most important background condition for the elicitation of shared emotional appraisals, they are embedded within a more comprehensive set of attitudes that subjects of these emotions share. People who share emotions often have a history of common experiences in the context of shared social practices as well as representations thereof, as Parkinson et al. (2005), Schmid (2009), Konzelmann Ziv (2009), and – most comprehensively – von Scheve and Ismer (2013) point out. Therefore, in addition to concerns, group members typically also share other cognitive, conative, and evaluative attitudes in a more or less collective sense, analogously to sharing concerns. Together such shared attitudes constitute the intentional background from which collective emotions can emerge in situations that impinge on some shared concern or concerns of individuals. Convergent emotional appraisals depend then on sharing at least some other attitudes besides concerns, of which the latter are still the most important because without them, collective emotions would not emerge in the first place.
Collective emotions of this kind may not be collectively intentional as they lack a shared intentional object and an underlying robustly shared concern. However, we would not go so far as to deny that these emotions are collective in any sense as one reviewer of this paper suggested. For the participants still have emotions of the same type with a similar (if not the same) evaluative content which, together with mutual awareness, meet our conditions (c) and (d) above.
One main disagreement between the received theories concerns the “collectivity” of intentions. Bratman (1999, 2014) defends an individualist view according to which shared intentions are interrelated individual intentions with a collective content (“that we J”) together with other conditions such as meshing individual subplans, disposition to help others to fill their roles in the relevant joint action if needed, interdependence in the persistence of each participant’s relevant intention, common knowledge of the previous conditions, and mutual responsiveness to the other’s intentions and actions. In contrast, Searle (1990, 1995, 2010), Gilbert (1989, 1996, 2003, 2014b), and Tuomela (2002, 2005, 2007, 2013) are collectivists in the sense that they require irreducibly collective we-intentions for joint intentional action.
Pacherie (2002) distinguishes a third type of emotional action, those of fully deliberate emotional actions. These are actions motivated by an anticipation of an emotion, either positive or negative, that the agent wants to either experience or avoid, using this anticipation as motivation for his or her action. Emotional joint actions of this kind are obviously possible, for instance, when a group of people use an anticipation of collective fear to avoid a possible danger to the group. However, we do not discuss these cases because they operate similarly through the agents’ joint intentions as semi-deliberate emotional joint actions.
Interestingly, for instance, a herd of impalas’ fleeing caused by a similar emotional contagion of one impala’s fear of a lion does not create a comparable collective action problem. For this reason it may be described as serving the goal of the herd, although the action is not collectively intentional in the received theorists’ sense.
Besides emotional contagion and rapport, Michael discusses the contribution of emotion detection and empathy for coordination in joint action. Here we focus on those forms of shared emotions that Michael associates with behavioral synchrony.
Bacharach (2006) proposes an analogous hypothesis called the interdependence hypothesis for team reasoning, which he formulates in game theoretic terms rather than in terms of concerns and emotions (see also Pacherie 2011).
About the article
Published Online: 2016-03-04
Citation Information: Journal of Social Ontology, Volume 2, Issue 1, Pages 33–57, ISSN (Online) 2196-9663, ISSN (Print) 2196-9655, DOI: https://doi.org/10.1515/jso-2015-0020.
©2016, Mikko Salmela and Michiru Nagatsu, published by De Gruyter. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License. BY-NC-ND 3.0