Abstract
It is commonplace to attribute memories to groups of individuals both large and small. Attributions of memories to groups are also found in social science research. This article proposes using philosophical accounts from the literature on social ontology to help clarify and deepen our understanding of how these terms are being employed in the social sciences. Two contrasting accounts of collective remembering are presented: the joint commitment account derived from the seminal work of Margaret Gilbert, and the participatory intentions account based on Christopher Kutz’s analysis of collective action. The implications of these accounts for clarifying notions of collective memory and remembering in the social sciences are explored through two case studies—one involving a social media site that promotes sharing of memories among users and the other concerning organizational remembering.
Keywords
Introduction
It is commonplace to attribute memories, or their absence, to groups of individuals both large and small, or to urge them to acquire or hold on to such memories. Attributions of memories to groups are also found in social science research. Sociologists, anthropologists, social psychologists, and historians appeal to the concept of a group’s collective memory or a group’s collective amnesia to explain current social, political, and cultural events and practices. However, common though these attributions are, it is not always obvious how, from a theoretical standpoint, they are to be explained and justified. Yet it is the very currency of these attributions combined with doubts about their meaning that compels us to take collective memory seriously as a topic for theoretical investigation.
In each case one might think that an adequate theoretical account would just be that there is the presence or absence of a common memory among a number of individuals (Margalit, 2002: 51). 1 However, this would be a mistake. To see why, consider some facts about memory understood in this way. Common memory, in the sense of several individual memories having a common content, can occur even when none of the individuals is aware of the others’ existence, let alone that the others share the same memory. It can also happen when there is no knowledge on anyone’s part that they are playing a part in a group activity of remembering and have no intention of doing so. Yet the notions of collective memory and remembering in common discourse and in the social sciences, while sometimes analyzable in these terms, frequently include elements that the concept of common memory leaves out, namely, elements of mutual awareness and mutual knowledge and much more.
One might go further than this and stipulate that a common memory is not a genuine collective memory at all, since for collective memory there is a minimum requirement of some sort of cognitive linkage between the members of the group, and common memory fails to satisfy it. However, this is not the approach I take here. One of my chief aims is to make some headway in constructing a typology of uses of the expression “collective memory” in the social sciences, and for this purpose it is best not to be so prescriptive. Instead of labeling certain memory-related phenomena as genuine instances of collective memory and distinguishing them from other such phenomena that do not warrant this title, it would be more profitable to allow a range of phenomena to qualify for that label, including but not limited to, common memory and to devote the bulk of one’s efforts to articulating their conceptual differences.
I have two main aims in this article. As I note in section “Group remembering: a preliminary foray,” the notion of group remembering is ambiguous, so I propose that philosophical accounts in the literature on social ontology and collective intentionality be mined to help clarify and deepen our understanding of the different ways that expression is used, particularly in the social sciences (for a similar effort, see Michaelian and Sutton, 2017). 2 Those accounts provide conceptual machinery and theoretical frameworks that can supplement the theoretical work on collective memory that is carried out within the social sciences. I also suggest that these accounts can be used by social scientists prospectively, to help guide their research so that the particular features that warrant attributing collective memory to a specific group are made the focus of attention. Conceptual analysis and theories of group agency articulate the conditions that must be satisfied to justify attributions of collective memory, and this can help shape the direction of empirical investigation.
Considerable work has been done on collective memory and remembering from psychological, sociological, anthropological, and historical perspectives. Psychologists have studied how basic psychological mechanisms can explain the dynamics of collective memory formation (Anastasio et al., 2012; Hirst et al., 2018; Middleton and Edwards, 1990); sociologists, many of whom have been inspired by the foundational work of Maurice Halbwachs, have explored the social aspects of individual memory as well as memory as a group-level phenomenon (Connerton, 1989; Halbwachs, 1980; Olick, 1999; Zerubavel, 1996); anthropologists have examined the transmission of memory in oral traditions and practices as well as symbols and myths (Bastide, 1978; Evans-Pritchard, 1940; Goody, 1986); and historians have reflected on the role of personal and collective memory in the construction of historical narratives and on the relationship between history and memory as modes of understanding the past (Burke, 1989; Yerushalmi, 1982; Zerubavel, 1995). Philosophy has not been as central in memory studies as have some of these other disciplines, but much philosophical work has been done on the nature of groups and group agency and this can be brought into productive contact with related issues in the social sciences. This is what I propose to do in this article, albeit in a preliminary way.
A second aim, more purely philosophical, is to explain how memory can ultimately reside in individual minds and yet be attributable to groups. Within the philosophical literature there are rival accounts of group agency, belief, intention, and cognition, and these can be transposed to the case of memory to yield different accounts of group remembering (see Bratman, 2014; Gilbert, 2014; Kutz, 2000; Miller, 1995; Searle, 2010; Theiner, 2013; Tuomela, 2013). I briefly discuss two of them from the philosophical literature in section “Two contrasting philosophical accounts,” the largest section of the article, and I illustrate them with two case studies in section “Two illustrative cases.” In my view, it would be a mistake to think that because groups are composed of individuals and remember via the actions of individuals, and so do not exist “over and above” them, it follows that there can be nothing more to group or collective memory than the same mechanism of personal memory that exists in the individual case, or that group memory can only be understood as an aggregate of individual memories (Pettit, 2014; Theiner, 2014). For arguably individuals can interact with one another in such a way that they can be said to remember as one or as a unit, and not just individually, or that they can share an end that justifies such an ascription. In other words, a group-level outcome can emerge from processes at the individual level. The two accounts explain how.
There is a general philosophical problem about how to explain attributions of a variety of mental states to groups, such as group belief and group intention, and group and collective memory raises many of the same issues as these other group phenomena (for an introduction to current debates on group belief and agency, see Tollefsen, 2015). Some have argued that notions of group remembering and group memories, to the extent that they are intended to refer to something other than aggregates of individual memories, are merely metaphorical, since only individuals are capable of memory (Gedi and Elam, 1996); or that they improperly hypostatize a group, treating it as a kind of thing (as in Brubaker and Cooper, 2000); or that they invoke an ontologically suspect kind of group mind (Roennegard, 2013; Vasquez, 1983, 2003). Were it indeed the case that notions of group remembering and group memories were not to be taken literally or had these problematic meanings, we would be well advised to steer clear of them or to put scare quotes around them. However, this is something that needs to be argued for, and there are good reasons for doubting that the effort will be successful.
Group remembering: a preliminary foray
Consider the following attribution of a memory to a group: the memory of 9/11 that is said to be an indelible part of the American psyche. This may mean simply that all or most of the American people remember 9/11 (if suitably cued). In this case, the totality of individuals who remember this event may be said to constitute a group of a sort, but only in a quite limited sense. For while they are grouped together by outside observers on the basis of a property they all have in common, the memory of 9/11, they may not themselves be aware of participating in a shared memory. And if they cannot be aware of this because they are isolated from one another, they cannot work together as a group to pursue jointly held goals, such as commemorating that event, or have an understanding of themselves as jointly commemorating that event. They do not need to be aware of others and themselves having the same memories of 9/11 (indeed, they will not, since their memories are quite diverse) nor do they need to entertain the thought that all of them are remembering it together. But they can only jointly participate in commemorating that event if each of them, broadly speaking, remembers 9/11 and the fact that they do so is out in the open among them.
To secure that it is out in the open, we have to add a common-knowledge requirement. 3 It must be common knowledge among the members of the group that each remembers the events of 9/11. (In a simple two-person case, the common-knowledge requirement is met if A knows that B remembers 9/11; B knows that A remembers 9/11; A knows that B knows that A remembers 9/11; and B knows that A knows that B remembers 9/11; and so on.) In this way, it is publicly known among the participants that each is involved in remembering 9/11. Despite the fact that Americans are widely dispersed, come from a diversity of backgrounds, and have differing experiences of and relations to 9/11, the common-knowledge condition could be satisfied on those occasions when the events of 9/11 are brought home to them, as in the yearly commemorations at the former site of the World Trade Center. It is possible, in other words, that on these occasions Americans reasonably believe that all or most of them have a memory of 9/11 and that this belief is itself something they all or most of them share.
One obvious problem with this account is the percentage of the group that must remember 9/11 under the condition of common knowledge in order for the memory to count as experienced by the group as a whole. It is implausible to require that all of the members must do so. 4 Something near all, say 90% of the members, remembering the events under the condition of common knowledge seems sufficient to describe their memory as a group memory. But what about 51%? This indeterminacy of “all or most” creates an opening for considerable disagreement about whether a group remembers. We could of course stipulate what percentage is required for group memory, but the lack of specificity here is not the main problem with these accounts. The common knowledge requirement is a relatively weak condition to impose on group memory, indeed too weak to provide an explanatorily adequate account of many important cases of group memory. Though common knowledge connects parties together in a way that their merely having the same memories does not, there are putative instances of group memory in which parties are arguably connected even more tightly or closely than this. An adequate explanation of these instances of group memory would not be given by simply adding a common knowledge requirement to an aggregative account of group memory.
To bring this out, consider another example. The members of a book club start their regular monthly meeting by trying to recall the discussions they had at their previous meeting. Since none of them can recall everything they discussed, each is open to adjusting, revising, and augmenting his or her memories in light of the contributions of the others. In this way, the members’ fragmentary memories of the last meeting are pooled in a process of collaborative group recall, creating a group memory that they can at least provisionally agree on. 5 This collaboration creates the social context in which the individual members engage in the act of remembering, and each of them intends to contribute to the shared goal of reconstructing the events of the previous meeting. The memory that the members construct is not just his memory plus her memory plus the memories of the others, but their memory, jointly constructed. Once constructed, each member has his or her own memories, but there is a sense in which these memories are the same as others’ memories. This can be considered a kind of group remembering because individuals are remembering in groups, together with others, and following common usage I will call it collaborative remembering. Collaboration can take different forms and need not always be as interactive as in this example; it can be quite minimal or it can simply involve remembering in the presence of others (see Meade et al., 2018). But in every case, it is only because individuals are in a social and interpersonal context of some sort that they have the memories that they have, a context created by their collaboration.
There is, in addition, an importantly different kind of group remembering, according to which the members not only remember together, but in some way remember as one or as a body so that it is the group—not the individuals that compose it—that remembers. I will call this group-level remembering. This corresponds to the so-called “group-mind thesis” in the literature on distributed cognition, according to which groups are themselves the agents of memory. This differs from collaborative remembering because a pair of individuals can collaboratively remember without it being the case that the pair itself—as a kind of group—remembers. Collaborative remembering that occurs when individuals co-construct the memory of an event is an instance of what is referred to as “the social manifestation thesis”, because according to this thesis the group is the context in which individual remembering takes place, not what is itself irreducibly doing the remembering (see Barnier et al., 2008; Wilson, 2018).
Though I here count collaborative recall as a kind of collective memory, I acknowledge that they are sometimes distinguished, as in Abel et al. (2018): Most of the definitions of collective memory stress that collective memories encompass information that is of high importance to group identity, and that therefore has high salience for group members, and may even be related to high levels of emotionality . . . Studies on collaborative remembering . . . seldom involve recalling information that is emotional, clearly salient to participants, or even of relevance to individual or group identity. (p. 283)
In other words, what distinguishes collective memory from merely collaborative memory is that the former commonly refers to memories that are important for individual or group identity whereas the latter commonly does not. However, my interest in this article is in how conceptions of group agency in the philosophical literature can be brought to bear to sharpen our understanding of collective remembering in the social sciences and how group agency is involved in both collaborative and collective memory. From this perspective, it is not group identity itself that is critical but rather the fact that group identity is formed through or revealed in exercises of group agency.
In accordance with the inclusive approach that I adopt in this article, I classify both the co-construction of memories within a group and group-level remembering as types of collective remembering, albeit with a significant difference. In the former, remembering is an individual-level process but also collective because of the irreducibly holistic end to which all contribute, as in my example jointly reconstructing the events of the prior book club meeting. In the latter, remembering is collective both because of the character of the end and because of the holistic character of the agent of remembering.
These remarks obviously need considerable elaboration, but we can draw on philosophical accounts of group agency from the social ontology and collective intentionality literature to underpin and help flesh out these and other conceptions of collective remembering. These are also a valuable resource for the social sciences of memory. Unfortunately, according to John Sutton (2008: 41), the latter have so far largely failed to take advantage of it. “It is curious,” he says, that the literature on memory in the social sciences has as yet made little contact with what should be the most helpful philosophical tradition here, the diverse defenses of the “plural subject” and related notions in social ontology. Assuming that “human size is not basic to agency,” such work in social ontology has developed robust analyses of notions like shared or joint action, mutual or collective belief, and shared intentions.
At the same time, the social ontology literature has also failed to engage with the social science literature: “In turn, even when that work [in social ontology] has addressed diachronic and historical aspects of groups, it has paid little attention to memory” (Sutton, 2008: 41). Sutton then calls for collaboration between these disciplines: “There is room here for natural alliances between social ontology, distributed cognition, and the interdisciplinary study of remembering” (Sutton, 2008: p. 41).
For those who participate in this collaboration, the aim of such an alliance would be two-fold: to draw on various accounts of group and collective agency in the philosophical literature to help bring greater clarity and specificity to notions of collective remembering in the social sciences and to enrich these accounts by drawing on interdisciplinary work on remembering. I will not attempt the second of these tasks in this article, however, and only the first to a very limited extent. My aim will simply be to illustrate by means of a few cases how work on group and collective action can be applied in the social sciences, where memory is construed as some kind of group phenomenon. It would be too much to suggest that we can align every case of collective memory discussed in the social sciences with notions in social ontology that can help sharpen its meaning. But even if we don’t have enough information to give a definitive answer, the effort would not be a waste of time: it would point to significant lacunae in the case description. Moreover, conceptions of group agency from the philosophical literature can be put to use prospectively in the construction of case studies and investigative tools so that it is clear from the outset how collective memory is being understood in a particular context and what empirical evidence is needed to confirm it.
Two contrasting philosophical accounts
I turn now to two influential accounts in the social ontology literature that can serve retrospectively as interpretive tools and prospectively as constructive aids in the social sciences. These accounts yield analyses of “collective remembering” (or possibly “collective remembrance”) and not in the first instance of collective memory. Collective remembering is an exercise of agency: it is something that people intentionally do in a group and/or as a group. (I understand this in such a way that one can intentionally remember in a group even if the group to which one belongs is merely a collection of individuals who remember the same thing.) Think here of family members collectively remembering the stories told by their grandparents; the members of a religious community collectively remembering the rituals that constitute the basis of their communion; or the citizens of a nation collectively remembering its victory over foreign adversaries or some historically significant traumatic event. In each case group members intentionally join with one another in the promotion of a shared goal, that of shining the spotlight on significant persons or events from the past. In the limit case, we can also speak of collective remembering when there is no intentional joining with others, as in common memory.
Collective memory, in contrast, is not an exercise of agency, so not of group agency in particular. It is, rather, a mental state that is ascribed to a social group, or a feature of the mentality of groups that has a bidirectional relationship to collective remembering. 6 In one direction, collective remembering employs mnemonic activities and practices of various sorts to invoke past events that have particular salience for a group of people. When these past events, stored as memories, become the object of broad social awareness, they provide content and focus for the process of collective remembering. In another direction, collective remembering produces new collective memories that, once formed, can in turn be used for a subsequent process of collective remembering. A collective trauma, for example, can be retained as a collective memory, and this can be revisited in a process of collective remembering.
The first account, alluded to by Sutton, is derived from the work of Margaret Gilbert. I call it the joint commitment account of collective remembering. According to this account, a group G can be said to remember event E just in case the members of G are jointly committed to remember E as a body. In being thus jointly committed, the members form what Gilbert calls a “plural subject,” in this case a plural subject of collective remembering. The second is derived from the work of Christopher Kutz, and I call it the participatory intentions account of collective remembering. According to this analysis, collective actions are combinations of “participatory intentions,” that is, intentions of individuals to participate in a shared end. Specifically, collective action occurs when there is sufficient overlap among individuals’ conceptions of the end to which they contribute.
These accounts are importantly different: what fails to be an instance of collective remembering according to the former account can nevertheless be an instance according to the latter. I choose these two because they are representative of divergent views on the extent to which collective memory should be understood in holistic terms and this is an issue that has significant implications for how to understand research on collective memory in the social sciences. I discuss the accounts of Gilbert and Kutz in some detail so that their potential usefulness in the social sciences can be fully appreciated.
Gilbert’s joint commitment account
The notion of a joint commitment is central to Gilbert’s account of unified social groups, and she explains it this way: “It is possible to describe any joint commitment according to the following formula: X, Y and whatever others are jointly committed to x as a body.” “Substitutes for ‘x’ are broadly speaking psychological terms such as ‘believe that such-and-such’, ‘endorse such-and-such a goal’, and so on” (Gilbert, 2014a: 348). We get another instantiation of this general formula by substituting “remember that such-and-such” for x: individuals remember some event (or person, etc.) as one when they are jointly committed to remember that event (or person, etc.) as a body, that is, as a unit or as one. The social unity that collective memory imparts to a number of co-remembering persons is rooted in this joint commitment, just as the social unity that shared beliefs impart to a number of cobelievers comes from their joint commitment to believe x as a body, and similarly for the social unity that comes from shared values.
As for what “as a body” means, Gilbert (2014a) explains it as follows: “To be jointly committed to x as a body is to be jointly committed to bring it about as far as is possible that the parties emulate a single x-er” (p. 348). Applied to memory, this yields a notion of group memory that is neither an aggregation of individual memories nor a state of affairs that consists in the memories of individuals and the interactions between them. Rather, a group remembers just in case participants in the group are jointly committed to bring it about as far as possible that they emulate a single rememberer. An indication that they have successfully emulated a single rememberer is that their collective memories play a role in the life of the group that is similar to the role that memories play in the life of a single individual.
By jointly committing themselves to do something as a body, individuals constitute what Gilbert (2014d) calls a “plural subject.” It is a single subject, “in a sense unitary and indivisible” (p. 56), but also composed of a plurality of participants. It is not, she emphasizes, a single subject composed of a plurality of personal commitments to join with others in xing (believing, valuing, remembering, etc.), or a plurality of personal intentions to do so. She characterizes a personal commitment as one that “is brought into existence by one person alone,” and that “that person can terminate or rescind . . . simply by changing his or her mind” (Gilbert, 2014b: 31). So, for example, if a group memory were solely constituted by a set of interlocking personal commitments of individual group members, then absent special circumstances, anyone so committed could simply decide that he or she no longer wants to share in the group memory and withdraws from it.
The difference between linked personal commitments and a joint commitment is crucial for Gilbert. Collective memory is constituted by a joint commitment among group members to remember as a body, not by their linked personal commitments to remember: When two people enter a joint commitment each party gains a special standing in relation to the other. If one violates the commitment, the other is in a special position to object. It was, the other may emphasize, our commitment that you violated, not a personal one. (Gilbert, 2014b: 34)
Unlike personal commitments, a joint commitment grounds “rights of the parties against each other, rights to conformity to the commitment” (Gilbert, 2014b: 34). The rights and correlative obligations at issue here are in her view inherent in any joint commitment, and typically they are the product of the parties’ voluntary actions. In this way they resemble the rights and obligations created by promise making or explicit agreement, although they can arise in other ways as well. They can arise when each party simply expresses his or her personal readiness to participate with others in a joint commitment. A promise or explicit agreement to enter into and abide by a joint commitment is not required. This is particularly important for large-scale groups, where promises and explicit agreements among the parties to a joint commitment are likely to be lacking.
Joint commitments create mutual expectations of mutual reliance among participants. If these expectations are disappointed by a participant’s failure to act in accordance with the joint commitment, this may impose additional burdens on the remaining participants or leave them vulnerable to additional risks, in some cases quite serious ones. Gilbert insists, however, that no reference to such factors, or to promises or agreements, all of which she calls “external,” that is, external to the joint commitment itself, is needed to explain how rights and their corrective obligations are generated out of joint commitments (Gilbert, 2014e: 111). To the contrary, she claims that the obligations of fidelity arising from a joint commitment can be explained solely by reference to the fact of joint commitment itself. She refers to these obligations as “rational requirement[s], such that one is not being appropriately responsive to the considerations that bear on the case if one fails to act as one ought” (Gilbert, 2014e: 114). By rejecting these ordinary explanations of the ground of rights and duties, she wants to distinguish the obligations that arise from joint commitments themselves from (at least ordinary) moral obligations.
To further clarify the notion of a joint commitment to remember as a body, contrast it with remembering in collaboration with others. As Gilbert would explain it, individuals enter into a joint commitment to remember x as a body when each of them openly expresses his or her readiness to commit with all of them to remember x; and the joint commitment is sustained by each in fact doing his or her part in that joint commitment (which is not to say that each creates part of the joint commitment). The readiness to commit can be signaled by entering into an agreement with others to participate in a joint commemorative activity, and one does one’s part by performing the role that the joint commitment has assigned one. This is group-level remembering. Now consider the book club example from section “Group remembering: a preliminary foray.” As I described it there, this was simply a case of each club member actively contributing details of the previous meeting in furtherance of the shared goal of remembering what was discussed at that meeting. Each member, we may suppose further, did not just do this for fun but was personally committed to helping construct a group memory of that meeting. This is a case of collaborative remembering. How does a joint commitment change this state of affairs?
We can explain it this way. An indication of one’s readiness to participate in the creation of a joint commitment is sharing one’s memories with the others. Sharing memories occurs in both types of collective remembering. However, given the nature of a joint commitment to remember x (or to remember x in a certain way, or to remember something about x), a joint commitment to remember x as a body brings with it a determined readiness to set aside one’s personal memories of x if this is what is required for the formation of a collective memory. The book club members may have very different personal memories of their experience, but their readiness to jointly commit to remember it as a body, if this is an apt description of their condition, is not a readiness to jointly commit only insofar as the collective memory agrees with their personal memories. Rather the readiness to jointly commit to remember as a body brings with it an additional sort of readiness, a readiness to set aside personal memories if this is necessary for the sake of the collective memory.
To further explain, consider this example. Suppose the adult children of the Henderson family have widely different memories of their recently deceased father. Some do not have any fond memories at all; some remember a few occasions when their father was supportive and caring, and many when he was not; some have largely pleasant memories of their father and their relations with him. What kind of memory might the group itself, composed of the Henderson children, have? One possibility is that they jointly commit to adopting the collective stance of remembering their father in positive terms. They might do this for various reasons—for example, because they believe personal memories should remain private or because they do not want to upset colleagues and associates, who have only good memories of their father—so they are prepared to set aside whatever personal memories they may have that conflict with this stance. It might be objected that the Henderson children are not really remembering their father in this way, only pretending to do so. Comparing memories to beliefs might help show why this is mistaken: just as we allow that beliefs can be ascribed to a group even if none of the individuals that compose the group personally has those beliefs, so should we do the same for memories. Gilbert (2014c) would no doubt accept this comparison since she is offering an account of group mental states that is intended to apply to the entire range of those states. 7
The readiness to set aside personal memories for the sake of the collective memory is a requirement of joint commitment since the existence of a joint commitment to remember places constraints on permissible appeals to personal memory. In contrast, a readiness to set aside one’s personal memories in the case of pooled personal commitments to remember is a contingent and not an essential feature of the process. Even if one gives up one’s personal memory because it is contradicted by the person with whom one shares it, it is not a normative requirement that in conjoining personal commitments to remember one should be ready to do this. Entering into a joint commitment to remember as a body is different in this respect. It is formed through appropriate expressions of a readiness to join with others in its creation, and this sort of readiness entails other individual (as distinguished from personal) commitments, including the commitment not to withdraw from it unilaterally because it disagrees with one’s personal memories.
The readiness to set aside personal memories if the collective memory calls for it raises the possibility that a collective memory of x can emerge even if no member of the group personally remembers it this way. The discrepancy between a group collective memory and the personal memories of the individuals constituting the group is especially likely in large-scale, far-flung groups. If the members of a group are widely dispersed and are affected in very different ways by events that affect them all, then it is extremely unlikely that each member’s personal memory of those events will match up with what the whole group remembers. Nevertheless, intuitively it seems the group as a whole can still, under certain circumstances, remember those events. For Gilbert, the critical question is, can there be some kind of joint commitment at work here?
In small-scale groups whose members are highly interdependent and cooperative, such as families, the parties jointly committed to remember x as a body openly express to each other their readiness to jointly commit themselves to the group project of remembering some event(s) or person(s). Gilbert argues, however, that a joint commitment can be formed even when the parties do not openly express “to each other” their readiness. Perhaps they do not know each other as individuals. Even so, she claims, it may be common knowledge in the group that everyone or almost everyone in the group has expressed his or her readiness to be jointly committed to remember x, even if the members of the group do not know each other personally. Moreover, it follows from Gilbert’s conception of joint commitment that the participants have rights to the conformity of others to the commitment to remember, and this must be possible in large-scale groups as well. When these conditions are satisfied, a joint commitment to remember x as a body has been created and a true collective memory emerges. As she argues, It is implausible to assume that in a very large population every single member of the population has expressed his readiness for a particular joint commitment. Nonetheless, there may be good evidence that such readiness is sufficiently widespread for one to judge that there is a plural subject whose boundaries at least approximate that of the population in question. (Gilbert, 2014a: 351)
The extension of Gilbert’s core idea of a joint commitment to large distanced populations may not be as unproblematic as she suggests, however. If the members of these populations have openly expressed their readiness to participate in a joint commitment, and it is sufficiently out in the open for all to witness, and if each has a right to strongly object when others withdraw from the commitment, then the conditions for the existence of a joint commitment are satisfied. But it may be unlikely when people are far removed from each other that they will know of one another that they are ready to participate in a joint commitment, let alone that they will be in a position to object to others’ failure to keep up their end of the joint commitment. Size and distance are not the only determining factors, of course, although open knowledge of everyone’s readiness to participate in a joint commitment is more likely in small-scale, face-to-face, and highly interdependent groups than in large-scale, widely dispersed ones. Also necessary and to some extent related to size and distance is a unifying intentional structure of some sort. If no such structure is present, a distanced population would not be capable of collective remembering. Small subgroups of members could still remember together, but the entire group could not remember as one.
Moreover, this concern about the plausibility of attributing a joint commitment to a distanced population has a bearing on the usefulness of Gilbert’s account in the social sciences. The collective phenomena that social scientists such as political scientists, historians, sociologists, and anthropologists study include large groups composed of distanced populations, and the relevance of Gilbert’s account to their concerns therefore depends on its applicability to these sorts of groups.
To conclude, an account of collective memory based on Gilbert’s joint commitment account of group mental states and activities sets the bar quite high for what counts as a collective memory and collective remembering. It is joint commitment that explains collective remembering, and joint commitment is no ordinary kind of commitment. It is a commitment that constitutes a plural subject, and a plural subject is said to have mental states of various sorts analogous to the individual subject of mental states. Following Gilbert, collective memory and collective remembering can only be attributed to a group if, based on the evidence, a joint commitment can plausibly be attributed to the individuals involved. This is what social scientists need to look for and validate, according this conception of collective remembering. If a joint commitment cannot plausibly be attributed, then this cannot be a case of collective remembering as Gilbert understands it, and whatever results from that process cannot be a true collective memory.
It seems likely that there are cases in the social sciences in which attributions of collective memory involve the sort of joint commitment that Gilbert requires. But because I take a non-exclusionary approach to what counts as collective memory, I do not want to privilege this as the uniquely correct analysis for the social sciences. On the contrary, I consider this but one among a number of equally legitimate uses of the term collective memory, some of which require an analysis of a different sort. I turn now to an account of collective remembering that may be more promising as a general account of the phenomenon, in part because it eschews her idea of a joint commitment to explain it.
Kutz’s participatory intentions account
Kutz develops an account of collective action that does not employ the notion of joint commitment so does not carry with it the normative implications of Gilbert’s account. There may be some cases of collective action that do essentially involve obligations and entitlements of the sort Gilbert says are internal to joint commitment. But there is also an important kind of collective action that does not essentially involve them. Kutz’s account is in this sense more modest in its conceptual requirements than Gilbert’s.
To explain Kutz’s view, and draw out the contrast with Gilbert’s, consider again the following example. The American people are said to have a collective memory of the attacks on 9/11. Indeed, given that the United States was the particular target of these attacks, their collective memory of these events is said to have an emotional intensity for them that makes it quite different from the collective memories that exist elsewhere in the world. Of course this does not mean that 9/11 is constantly being remembered. The memory of 9/11 may be dormant for much of the time for most people, rising to conscious and focused attention only on special occasions, like the yearly commemorative ceremonies around the country on 9/11 and at the site of the former World Trade Center. Then 9/11 becomes the object of collective remembering. At other times we may still speak of collective memory, but the memory is only dispositional, that is, prone to becoming actual under certain circumstances, like the yearly organized commemorative activities. These special occasions of collective remembering in turn contribute new collective memories and affect how the events of 9/11 are remembered on future occasions. What makes the public occasions ones in which the participants remember the events of 9/11 collectively?
Let us start with Gilbert. Her answer is that they remember collectively only if “there is good evidence that such readiness [for a particular joint commitment] is sufficiently widespread for one to judge” that the American people is a “plural subject” of the memory of events of 9/11. Is there good evidence of this in this case? In addition to the yearly commemoration at the site of the World Trade Center, numerous commemorative activities take place on 9/11 around the country. These activities do not just spring up spontaneously but are in response to a national call for a day of remembrance. Individuals and the various local communities of remembrance responding to the call are aware of other individuals and communities doing the same, and they coordinate their activities with them. They do not necessarily directly collaborate with one another in working out the details of their respective commemorations, of course, for that is chiefly a local and individual matter, but there is coordination nonetheless in the weaker sense that they all intend to do their part to realize the shared end of a national day of remembrance. Do the widely dispersed individuals and communities enter into a joint commitment to remember the events of 9/11?
There are two reasons to think that they may not. First, the kind of jointly intentional activity just described may not give rise to the rights and correlative obligations that Gilbert claims are inherent in joint commitments. Since directives for a national day of remembrance are of varying degrees of stringency, individuals and communities may be able to opt out of commemorating 9/11 without being answerable to individuals elsewhere and to other communities for their failure to participate. Second, individuals who participate in a joint commitment to remember as a body do not just intend to contribute to a shared end, such as remembering the events of 9/11. They also have intentions concerning what their group as a whole should do, namely, remember these events, and individuals can intend the former without intending the latter. Kutz (2000) calls intentions of the latter sort “group-intentions,” and he explains them this way: Group-intentions are ordinary, instrumental individual intentions whose subject is the individual agent and whose object is a collective act or outcome . . . In general, when agents act so as to realize the collective outcome, to the extent of aiding others in their contributions, we should attribute to them the group-intention to achieve that collective outcome. (pp. 96–97)
Group intentions differ from intentions to join with others in contributing to a group activity. The former sort of intention is an intention that we do something, and it is what jointly committing to doing something as a body involves. 8 The latter is an intention that I contribute to our jointly doing something, and it is an intention that everyone participating in the group activity has. Intending a group end and intending to do one’s part in achieving that end are easily confused, but conceptually distinct. A presidential proclamation that the third Monday of January of each year is designated a day to remember the life and work of Martin Luther King, Jr. involves a group intention, that is, an intention that we, the American people as a whole, make this a day of remembrance. This intention differs from the intentions of ordinary Americans who “have no views, let alone intentions, concerning what the group as a whole should do” (Kutz, 2000: 96). Thus, if we want to say that the coordinated activities of the various participants and communities of remembrance count as collectively remembering 9/11, as seems reasonable, it is not necessarily because they have openly expressed their readiness to participate in a joint commitment to remember it. Kutz has a different explanation.
There are three critical factors, in his view. First, the participants understand themselves to be acting as part of a group, specifically, a group engaged in remembering 9/11; second, they intend to participate in or contribute to the collective goal of remembering 9/11; and third, the goal is to be described in irreducibly holistic terms. Combining the second and third points, participants intend to do their part to help realize the end of collective remembering. Kutz (2000) calls these intentions “participatory intentions,” and he defines such an intention as “an intention to do my part of a collective act, where my part is defined as the task I ought to perform if we are to be successful in realizing a shared goal” (p. 81). Furthermore, two individuals share a goal “if there is at least one token activity or outcome involving the actions of the other whose performance or realization would satisfy the intentions of each” (Kutz, 2000: 76). Thus, a participatory intention has two satisfaction conditions: there is a shared end (“collective end”), and there is what an individual does to contribute to the shared end (“individual role”). An end is shared, and collective action occurs, only if there is sufficient overlap in the content of a number of participatory intentions of individual members of the relevant group. The “overlap” among the participants’ intentions is the “common ground in the states of affairs that satisfy the intentions of each” (Kutz, 2000: 94). In the case of remembering 9/11, there is sufficient overlap among the participatory intentions of those who participate in remembering the events of that day to constitute participation in collective action.
Participatory intentions directed toward the end of collective remembering commonly have different contents at a certain level of detail. Some individuals may conceive of attending a religious ceremony as their contribution to the collective memory of 9/11; others may contribute by participating in an artistic performance to mark the day; still others may decide to join in collective memory by visiting their local cemeteries where victims are buried; and so on. In general, participatory intentions concerning collective memory will often vary in these ways. Individuals approach the past from different perspectives with different histories, and they have their own understandings of how they should remember it. In societies that value and encourage freedom of expression, individuals are able to make contributions to the collective memory that embody those particular understandings.
Though quite varied, the extent of variation in participatory intentions is constrained, for without a common core, that is sufficient overlap, among the contents of these intentions, there is no participation in an activity of collective remembering but a collection of individual intentions to remember, each directed at a different object. There are two questions to ask about this: What is necessary for overlap to occur? And when is overlap sufficient?
First, for there to be overlap, mutual responsiveness between participants in collective action is not required. Mutual responsiveness, Kutz (2000) claims, is a feature of only “particular types of joint action” (p. 90), and not of joint action in general. Participants in public memorial activities such as those on 9/11 can join in the collective memory of 9/11 even if they are not mutually responsive to one another. Especially in a large group whose members are geographically dispersed, the opportunities for participant interaction are limited and participants may have only a general idea about the others’ intentions concerning commemoration.
Second, overlap of participatory intentions is a matter of degree, according to Kutz, and determining when it is sufficient for collective action does not straightforwardly follow from some context-independent rule. When group remembering involves many persons, for example, the overlap may be imperfect, as we can see as soon as we fill in the details of their individual intentions. Participants may have their own conceptions of the events or persons that are being memorialized, where these conceptions reflect their particular past experiences and social relations, and they may be more or less willing to compromise with the others on how these events or persons should be remembered. But even if perfect overlap is unlikely, collective remembering can still take place, if there is available a plausible description of the intentions of each under which they are all intending the same thing and they are willing to accept it, even if it means compromising. Kutz (2000) says this about the problem of overlap and some of the sources of disagreement that may bring it into question: Overlap is essentially a pragmatic concept and always a matter of degree, given inevitable differences in each agent’s expectations and conceptions of the group act. Agents will have more or less determinate conceptions of the group act, they may be more or less willing to compromise after bargaining on the character of that act, and they may have very different ideas about the scope of the group act, its duration and membership. As a result, a group act can be jointly intentional under one description and not jointly intentional under another. (p. 95)
So if the question is whether the overlap of participatory intentions is sufficient, the proper response is: sufficient for what? If overlap is a pragmatic concept, there will be practical interests at stake in making the determination one way rather than another. The question is not just factual, but normative as well.
Comparing the accounts
Kutz (2000) characterizes his conception of collective action as combining both individualistic and holistic elements: it is “individualistic with respect to agency and irreducibly holistic with respect to the content of agents’ intentions” (p. 112). The region of overlap among individuals’ participatory intentions, the shared goal to which they intend to contribute, cannot in general be explained in purely individualistic terms, that is, solely in terms of individuals, their actions, beliefs, desires, and so on. Here a holistic explanation is required. For Gilbert, the problem with this account is that its holism is only located in one place, in the content of participants’ commitments (or intentions) rather than in the character of the commitment that individuals engage in, the features that make it joint, in her sense. Kutz’s account of collective action therefore does not attach any particular significance to the standing that participants have to make demands of one another, and to rebuke one another for failure to comply with those demands, and this is something that in her view an account of genuinely collective action should be able to do. Participants “might reasonably express annoyance, disappointment, and so on” 9 at another’s failure to do his or her part in joint activity, but this is a weaker moral response than the response Gilbert thinks is appropriate in cases of genuinely collective action.
These two accounts of collective action offer contrasting conceptualizations of collective remembering. I do not claim that these are the only accounts in the literature on social ontology that we can draw on for this purpose, but they present compelling and instructively contrasting views to explain how participants are bound together in activities of collective remembering. One difference between them is that a participatory intentions account is less restrictive than a joint commitment account since it allows a wider variety of types of group action to count as collective. In fact, Kutz (2000) claims that “all collective action, hierarchical and non-hierarchical, preprogrammed and dynamic, planned and spontaneous, admits of a common analysis” (p. 103), the one that he provides. Collective remembering in organizations and institutions of various sorts as well as in groups formed by individuals coming together around the achievement of a particular goal is given a uniform analysis. On this view, joint commitment to do something as a body is only a special if important case of collective action. Another related difference is that a participatory intentions account makes fewer demands on participants than a joint commitment account, since according to the former, participants do not have to express their readiness to participate in a joint commitment with the other members of the group. It is sufficient if they simply intend to do their part to bring about a collective end, and this is a considerably weaker condition. 10
Given the diversity of ways in which notions of collective remembering and collective memory are used in the social sciences, only some of them will be illuminated by Gilbert’s joint commitment account. Kutz’s account of the basis of joint action stands a better chance of explaining how all the group phenomena which social scientists refer to in these ways can be genuine cases of collective remembering, despite significant differences in the size and structure of the groups engaged in the activity. The more inclusive account of collective action that Kutz provides might be attractive to social scientists for this reason, because it does not have the implications of Gilbert’s narrower view that only some of the putative instances of collective remembering count as genuine. Prospectively, social scientists adopting Kutz’s account will not be confined in their investigation of group phenomena to producing evidence of a joint commitment to remember as a body before being entitled to refer to them as instances of collective remembering.
Two illustrative cases
The terms “collective remembering” and (secondarily) “collective memory” have multiple meanings in both everyday contexts and contexts of social science research. I have suggested that philosophical analyses of the nature of group agency can be helpful in the social sciences in two ways: by clarifying the sense in which a particular group is engaging in collective remembering; and by orienting social scientists doing research to those features of a particular case that would justify attributing collective remembering to it. What those features are will vary according to the account of group agency on offer, and it will direct the researcher to accumulate evidence that substantiates the account that he or she accepts, if such evidence is available. Different accounts can be accepted provisionally in order to see which best fits the case at hand. Questions that might otherwise be of little interest will become salient in light of what the philosophical conception demands, and this can determine not only what evidence is needed but also how the researcher goes about gathering it.
In the spirit of furthering the “natural alliance” between social ontology and the social science study of remembering that Sutton advocates, consider how accounts in the philosophical literature can provide explanatory tools for conceptualizing collective remembering in the following two examples.
Remembering on the Internet as collective remembering
Platforms for content sharing on the Internet allow, indeed encourage, the participation of users in the production of content, and this sharing can alter how individual members of a group, namely the group of Internet users, remember past events. How should we understand collective remembering on social media sites that promote such sharing?
Take as an example the Egyptian revolution of January 2011 that overthrew the regime of Hosni Mubarak. Suppose there is a site, call it “Egyptian Memorial 2011,” set up on the web devoted to commemorating this event. Built on the occasion of the anniversary of the uprising, the memorial has the stated mission of “honoring the fallen martyrs in the revolution against the Mubarak dictatorship” and collects the names and photographs of the victims. Every visitor to the memorial can leave a token of their visit, a personal message or reminiscence, and make a donation to help defray the cost of site maintenance. The building of this site online in virtual space allows users to participate in a commemorative ritual just as they would in a ceremony in Cairo or elsewhere in Egypt. This participation can take place at any moment and from any place by people who are completely unacquainted with each other and would not be able to visit the actual places where the protests occurred.
Is this a case of collective remembering and if so according to what account? To qualify as such according to the account taken from Gilbert, the participants in the online memorial must jointly commit to remember the events of January 2011 as a body. A joint commitment in her sense is not just a psychological but a normative notion. That is, it generates rights and obligations, obligations on the part of persons who opt out of the commitment to render an account of why to those who remain, and rights on the part of those who remain to demand that account from them. One is not free to withdraw from a joint commitment the way one is free to rescind a personal commitment. It seems doubtful, however, that persons who are regular visitors to the online site stand in any such normative relationship to those who stop visiting it. The visitors are too widely dispersed, too loosely connected, and too little known to each other for there to be the kind of mutually binding commitment that Gilbert thinks is constitutive of collective action.
According to Kutz’s conception of collective action, however, the joint memorial activities of the Internet users can count as collective remembering, since his account is less demanding than Gilbert’s. As Kutz would describe it, the visitors to the site overlap in the conception of a collective end to which they all intentionally contribute, and therefore they act collectively. The collective end is shared by all the participants, although this does not mean that they all have to memorialize the events of January 2011 in exactly the same way. It is an end that cannot be decomposed into the individual contributions of visitors to the site, for according to Kutz, the end is an activity of the group as a whole, something that must be described in holistic terms. Moreover, unlike joint commitments, overlapping participatory intentions do not by themselves generate mutual obligations and rights among the participants.
Thus, Kutz would have a different answer than Gilbert to the question, “Are the participants in Egyptian Memorial 2011 engaged in collective remembering?” For her this would not be a case of genuinely collective action because the participants are not really jointly committed to remember the events of the Egyptian revolution. In one sense of the word, the participants are not genuinely bound together as one (see Gilbert, 2002: 173). For Kutz, however, this is no less a case of collective remembering than one in which there is a joint commitment among and between the members of a group to remember as a body.
Organizational remembering as collective remembering
In organizational contexts there is a well-defined sense in which collective memory is distinct from collective remembering. In these contexts, the former often refers to organizational archives that are the repositories of an organization’s memories, that is, to a type of cultural memory, in Jan Assman’s sense of the term (Assmann, 2010). Thomas Eberle (2016) characterizes organizational archives this way: They document some aspects of an organization’s life that may have been forgotten by the actually living generations, and constitute a social memory in the mode of potentiality that allows an interested researcher to rediscover the past of an organization and reconstruct, to some degree, its structures and social processes. (p. 96)
It follows that organizational memories, like other manifestations of cultural memory, can exist even if no one actually uses the archives to reconstruct the organization’s past or engages in acts of remembering. But while there can be collective memory without collective remembering, these cases are not my primary concern. In line with my focus on group agency, I concentrate instead on organizational remembering, a sort of remembering that may rely on and be facilitated by organizational archives, but that is not tantamount to or reducible to them. In what follows I use organizational remembering as an example of collective remembering, that is, the collective remembering of an organization’s members, and I turn to Gilbert and Kutz to see what light they can shed on this.
The individual memories of an organization’s members, Eberle (2016) notes, “are socially distributed and represent many different perspectives on an organization” (p. 102), but the collective memory of the members seems to involve something more than their mere aggregation. Consider an organization with a well-defined identity forged over the course of its substantial history. What makes for this organization’s collective remembering? It depends, I suggest, on the members of the organization jointly committing themselves to give authority to certain individuals within the organizational hierarchy to speak on their behalf with respect to how the organization’s past will be remembered. Collective remembering may emerge from the sharing of memories among those who have jointly committed themselves in this way, but the sharing itself is not what makes it collective. Rather the collectivity is grounded in a joint commitment to remember as a body, a commitment that is actualized through the granting of authority to specific individuals within the organization to represent its members in this matter. The joint commitment authorizes certain people to speak for everyone else with respect to how the group’s past will be remembered and it explains how it is possible for the group as a whole to remember. Gilbert (2014f) comments on how her core notion of joint commitment can be extended to cases of this sort: The members of a population can jointly commit to stipulate, as a body, that a certain person or body of persons has the standing or authority to do various things in the name of the jointly committed persons: to issue commands to members of the population, to make decisions for them, and so on. (p. 68)
Among the various things the members might jointly commit to stipulate, as a body, is that a certain person or group of persons has the standing to promote and coordinate appropriate public acts of commemoration. Moreover, along with the granting of authority, members commit themselves to various courses of action, for example, to refrain from publicly challenging the official narrative and to remonstrate with an organizational member who publicly denies that this is how the organization should remember its past. (Of course if those who speak for the organization abuse their authority, individual members may no longer be bound by their commitment.)
A Gilbert-derived joint commitment account of collective remembering thus seems to provide a plausible explanation of the collective nature of organizational remembering. Can the participatory intentions account also explain what makes organizational remembering collective, or explain it as well as the joint commitment account? The two would not necessarily deliver different verdicts. If the conditions that are necessary for collective action on Kutz’s account are satisfied in this case, as I suppose they may be, then there is no disagreement as to whether this is a case of collective remembering. But what is missing from Kutz’s account, and is critical in Gilbert’s, is the normative character of joint action, and it is this that collective remembering in an organizational context seems to have. Being part of an organization engaged in remembering is not like being part of a country engaged in commemorating the events of 9/11. The former places demands on individual members that the latter does not. One can simply decide not to participate in commemorating the events of 9/11 without owing others an account of why: the commitments to remember are personal, not joint. But this is not so for organizationsl remembering: it is constitutive of membership in at least a just organization that one endorses the reasonable representations of organizational leaders on one’s behalf, among which are those that express and help shape the collective memory of the group, and this endorsement has the normative implications Gilbert describes. Because her account brings out what is distinctive about the organizational context, I conclude that with respect to organizational collective remembering, the joint commitment account fits the phenomenon better than the participatory intentions account.
Conclusion
This article argues that philosophical work on the nature of group agency can be profitably integrated with theoretical and empirical work in the social sciences of collective memory. By using these philosophical accounts to illuminate cases of memory-related phenomena of interest to social scientists, we can gain a deeper understanding of the variety of ways in which the terms “collective memory” and “collective remembering” are employed in those fields. These accounts can also have practical value for social scientists, since they can help direct the conduct of empirical research aimed at exploring the nature of those phenomena. But a philosophical account of group agency can only elucidate collective memory at a fairly high level of abstraction, so to fully understand the nature of collective memory as a social phenomenon we have to look to other disciplines that study remembering. This is why what is called for is an alliance, one that unites, on one hand, the different social science disciplines that contribute to memory studies and, on the other, the branch of philosophy that studies the nature of groups and group agency, extended to include its exercise in group remembering. Each participant in this alliance has much to learn from the other.
