Most studies of social ontology rely on a common assumption about the relationship between mind and society. Since there cannot be a society without individuals, the consensus view holds that an explanation of the social reality entails an account of intentionality at its core. This assumption lies at the core of Standard Model of Social Ontology (Guala 2007), which has dominated the debate for the last 25 years. It is not difficult to see why social ontology must encompass an investigation of the intentional features of thought. Objects, kinds, and states of affairs acquire socially significant meanings through language and culture, when our attitudes and thoughts are “projected” onto the world in the course of interaction with other people (p. 50). From the point of view of the philosophy of mind, an analysis of intentionality (meaning) in the social context, or collective intentionality, is part and parcel of the task of understanding the nature and structure of social facts.
At least two options are available to carry out an analysis of collective intentionality (for a review, see Chant et al. 2014). Some philosophers opt for a bottom-up approach and focus on joint action as the paradigmatic case of shared understanding. A theory of collective, or shared, intentionality would then specify the conditions that must obtain for the mental attitudes and states of people to become shared in ways that sustain the formation of planned courses of action. Other theorists are motivated by the recognition that social facts display an intersubjective ‘force,’ which binds the thought and behavior of people beyond their own individuality. These authors take a top-down approach, arguing that some sort of “we-experience,” that is, a pre-reflective understanding of things as relevant for us all, the group of people we belong to at any given time, must be a requirement of analyses aimed to make the architecture of human society intelligible. There are many subtle formulations of these two approaches, however, both presuppose that society ultimately depends on some form of collective intentionality.
The Ant Trap is a powerful reminder that we ought to be cautious about the scope and significance of collective intentionality. For one thing, there is no agreement about what collective attitudes, or intentions more narrowly, are. Like other central terms of social ontology, Epstein notes that the term ‘collective attitudes’ suffers from a lack of clarity that is “endemic in all sides of the literature” (p. 59). The Ant Trap calls for greater clarity and serious philosophical scrutiny at a time when it seems that future progress in social ontology will depend on refining our conceptual toolkit by taking on board inputs and insights from other fields of research. Epstein does that superbly when he engages with work in contemporary metaphysics and the result is a better picture of social metaphysics. However, if this is the way forward, as it should be, we would also need to look at research across the cognitive sciences in search for more sophisticated theoretical resources to assess the role of collective mental attitudes in social ontology.
There is now a tendency among philosophers to take an exceedingly critical stance towards collective intentionality. Some of them claim that an appeal to collective attitudes would underlie a somewhat ‘mysterious’ and opaque view of social ontology. What is surprising about this stance is that in recent years cognitive research has vastly improved our understanding of the processes and mechanisms that sustain the formation of collective intentions. These advances should therefore contribute to an integrated theory of social ontology. This is yet to become common practice in social ontology though, and Epstein’s book is no exception. Although all major concepts of current debates in social ontology are discussed in The Ant Trap, and some in great detail, one is left with the impression that we should take the tendency to talk about collective attitudes loosely as evidence that collective intentionality is neither a necessary nor a sufficient condition for explaining the social reality. I shall resist this tendency by arguing that, if we aim at a good theory of social ontology, we need a good theory of collective intentionality, one that goes beyond the traditional and caricatured view of ‘we-attitudes’ presented in The Ant Trap. In the last section of this paper, I will suggest one possible way to improve our understanding of the underpinnings of collective attitudes in line with the pluralistic approach championed by Epstein.
2 Epstein on Collective Attitudes
The Standard Model of Social Ontology is the target of Epstein’s critique of the ‘anthropocentric’ view of the social world. This is the view that the “social world is made and maintained by us, by our mental attitudes” (p. 56). Epstein gives convincing reasons for believing that there is more to the social reality than the psychological make up of people. He does not deny that an account of social ontology may require some reference to the intentional features of thought and agency. In fact, his rebuttal of the anthropocentric view is, at bottom, a rejection of the assumption that the mental attitudes that contribute to the creation and persistence of the social world are collective. Yet, to say that there is more to the anchors and the grounds of social facts than the minds of people, is not the same as saying that collective attitudes should be ruled out from a theory of social ontology. There is an important difference between these two claims, which explains why one can be sympathetic with the former while questioning the latter. I shall illustrate this difference in the present section, before I discuss the role of collective intentionality in the next sections.
Epstein’s privileged target is Searle’s well-known collective acceptance theory (Searle 2010). As Searle’s story goes, it is because we each collectively accept, or recognize, that something is the case according to the generic form of a constitutive rule – “X counts as Y in C” – that a physical fact is a social fact (e.g. that piece of paper is a dollar bill). The fact that a piece of paper must be issued by a given institution in accordance with certain rules is not the same as the fact that the mental attitudes of people must be of a certain type in order for those rules to be collectively accepted. Grounds are not the same thing as anchors: the grounding conditions of a dollar bill are ontologically distinct from the anchors that set them up. If, then, we zoom in on anchors, we will find out that it is highly unlikely that the mental attitudes of individual people alone would play a role in the construction of the social reality. The “contingency, practicality and multiplicity of anchoring schemas” dictates that we ought to pursue a social ontology that is “free of commitment to one secret sauce that makes the social world exist” (Epstein 2014, p. 54). Hence, anchor pluralism seems the natural way to go.
One would think that pluralism stands against the assumption that a one-size-fit-all schema is all we need to explain the architecture of society. However, what emerges from The Ant Trap, quite vividly, is that the issue is not so much that individualists invoke only one schema but rather that, according to the very same schema, all the anchoring is done by facts about individual people. A pluralist could then argue that anchors must involve at least one fact that is emphatically not a fact about individual people, whatever the facts about individuals are, as suggested in the following passage: “The theorist who denies anchor individualism will take anchors to include facts that are not facts about individual people” (p. 104). Perhaps Epstein is trying to get across the message that we would need to postulate several anchors to make sense of the complexity and diversity of the principles by which social facts are set up, some being facts about individual people while others are not. His wording and the tone of discussion suggest a more radical interpretation though. What Epstein takes as facts about individual people are the collective attitudes of people.
To see why, note that traditional accounts of collective intentionality not only presuppose that there is one anchoring schema. Most importantly, they construe the schema in individualistic terms, since the ‘collective’ aspect of collective attitudes boils down to individual-level features of the psychological states of individual people, no matter whether the relevant attitudes are underpinned by dedicated intentional features of individual minds (brains), or consist of cognitive properties of individual psychology interlocked in the appropriate manner. 1 If this is correct, the point about pluralism is not just to say that there are different anchors at work, some of which do not involve the intentional features of individual people’s minds; the point is to deny that the grounding conditions of social facts are anchored by the collective intentionality of individual people. So, in advocating some form of anchor pluralism, Epstein actually hints at the possibility that collective intentionality can, and should, be discarded as a pillar of social ontology.
Now, it is one thing to challenge anchor individualism by arguing that multiple schemas do the anchoring including, most notably, schemas which do not involve facts about individual people; quite another to challenge anchor individualism with the aim to rule out collective intentionality altogether. To be clear, it is perfectly legitimate to argue that collective intentionality may have no place in a well informed and conceptually careful picture of social ontology, but one needs to provide an equally careful and articulate argument to make that case. In the next section, I suggest that the first step in this direction would be to tackle the question as to what collective attitudes are.
3 Beyond the Caricature View of We-Attitudes
Collective attitudes can play different functional roles in social ontology. In light of Epstein’s useful distinction, they can operate at the level of anchors, or grounds, or both. To make a start on this task, we should distinguish between the two approaches to collective intentionality sketched in the first section. For some philosophers, the collective aspect of collective attitudes refers to features of the cognitive configuration of individual people. This approach is informed by top-down considerations about the sort of attitudes that people must entertain for the social phenomena to display their distinctive features (i.e. it is because we all take things to be in a certain way that social reality is what it is). These considerations lead to the conclusion that the mental attitudes of individual people must be of a certain type for them to be able to exchange and compute information collectively (e.g. that a piece of paper is money). Such an approach is strikingly different from that which focuses, not on the psychological anchors of interacting individuals, but on the conditions that must obtain for acts of joint intentionality to unfold successfully. The former is about the anchors while the latter is about the grounds of social reality.
To group all these accounts under the same umbrella, i.e. a “generalized operation of the human mind,” implies that there would be one criterion to assess whether collective attitudes are necessary and/or sufficient. In fact, to make some progress on this task we should tackle the question in different ways depending on which account of the functional role of collective intentionality is at stake. Consider, again, Searle’s view of we-intentions. What Searle refers to as the capacity of collective intentionality is an individual-level, mental mechanism responsible for bringing about the attitudes that must be in place for people to recognize things from the perspective of the collectivity they belong to (i.e. that money can only be issued by the relevant institution). In Epstein’s own terms, Searlean collective intentionality is an anchor of the social world: the conditions that must obtain for a fact to be a fact of a certain social kind must be accepted collectively, which requires people to have the appropriate attitudes in the first instance. My contention is that if one conceives of collective attitudes as a cognitive anchor of the principles that govern social reality, something ought to be said about the way people cognize in the course of social interactions. What, exactly, should be said is a matter of empirical investigation and not just conceptual analysis.
Suppose that this view turns out to provide a useful framework to interpret some of the evidence from empirical studies of the processes whereby people share mental states (Tollefsen and Dale 2012; Gallotti and Frith 2013). To make appeal to the capacity of collective intentionality will then be no more surprising than invoking any other social-cognitive mechanism among those which are standardly posited to interpret data. As long as the empirical question is addressed with the tools and methods of the best science of the day, it would make perfect sense to say that there is at least one level of description, i.e. the level of neural implementation, at which the capacity of collective intentionality might be spelled out. Neither would the claim that people can have all the collective intentionality of the world in their heads be particularly surprising in light of traditional claims of (social) cognitive neuroscience – if, once again, it turned out that this is the way the social cognition of individuals unfolds when they engage in forms of collective acceptance and recognition.
The time is certainly ripe to move past intuition and loose language in social ontology. Also, Epstein is right that a one-size-fit-all anchoring schema is likely to explain very little about the complexity of social phenomena. Whilst this argument is not new (Meijers 2003), it is traditionally premised on the caricature view that all that is needed to do the anchoring is that individual people have some sort of unspecified, intrinsically collective ‘we-attitudes’ in their heads. We have gone a long way from that view, and more sophisticated and robust views of the underpinnings of we-attitudes are now available. By the same token, Epstein would need to unpack the concept of collective attitudes in much more detail, on the background of the rich conceptual framework provided in The Ant Trap, and take recent developments in the collective-intentionality literature seriously into account. In the last section, I would like to press on this point and suggest the possibility that collective intentionality may play some role in a theory of anchoring.
4 Towards a Theory of Aligned Minds
The complex ontology of institutions like the US Supreme Court is such that the attitudes of people are only one among the factors that contribute to its life and activity. More generally, when social phenomena acquire a life of their own, they depend on several factors for their existence and maintenance through time, which are not limited to the psychological aspects of social interactions. Yet, however platitudinous it may sound, the attitudes and acts that we entertain and enact together still capture essential aspects of social phenomena at the appropriate level of description. So, it is only fair to say that a theory of social ontology will have to say something about the dispositions underlying interactions among people. In this section, I would like to suggest one way to problematize the concept of collective attitude, beyond the caricature view, and to integrate it into a pluralistic approach to social ontology.
In addition to being over-intellectualistic (p. 5–6), classic accounts of collective intentionality have delivered a static picture of what it means for individual people to entertain and display collective attitudes. ‘Static’ means that most accounts specify a list of conditions that must be in place for people to have shared mental states, no matter what functional role they play, while they neglect the fundamental question as to how the relevant sharing unfolds in space and time. One promising development in the literature comes from attempts to think about the processes by which people share mental states as processes whereby they align their thoughts and experiences, bodily postures and behavioral dispositions, to accomplish a task together (Dale et al. 2013). Interestingly, when they are faced with the challenge of spelling out the relevant processes in detail, these accounts give pride of place to factors, which, at some level of description, resist the traditional individual-social dichotomy.
In the same vein, in The Ant Trap Epstein remarks that we have often tried hard to classify things as either individual or social, but there are indeed things that are neither individual nor social. The question, then, arises as to how best to describe individual-level mechanisms, including cognitive states and behavioral dispositions, which are only triggered in group environments. Epstein correctly suggests that it is implausible to assume that the mere fact of having a we-attitude in one’s mind entails that one can explain the anchoring of any frame principle of the social world. The one-to-one correspondence between a specific anchor and the relevant constitutive rule (i.e. frame principle), suggested by Searle, is too restrictive (p. 105). “Instead” – Epstein argues – “he [the anchor individualist] may just make a claim about the relation of a whole set of anchors to a whole set of frame principles” (ibid.; emphasis mine). What is interesting about recent discussions of interactive processes of alignment is that they provide novel, more reasonable and empirically sound ways of articulating the causal role of reciprocal relations in the formation of shared mental states. For example, one way to spell out the concept of relation in the context of social-cognitive research is in terms of the exchange of socially relevant sensory information, between the participants in a social interaction, in inducing a synchrony between their internal brain states (Friston and Frith 2015).
What becomes of collective intentionality on a relational view of the mechanisms and processes whereby mental attitudes become shared? For one thing, studies of low-level processes point to a variety of cognitive phenomena at the level of perceptual processing and motor representation, synchronization and coupling, co-representation and entrainment (Knoblich et al. 2011). The resulting picture of the formation of shared mental states is more modest in scope and less ontologically ambitious than traditional explanations. Current research on collective intentionality is also more diverse and remarkably less individualistic than first-wave theorists could possibly envision. It includes refined analyses of shared intention sensitive to developmental issues, discussions of group agency and macro-cognition, accounts of modes of cognition that diverge from the first-person singular – to mention a few (Vesper et al. 2012; Schilbach et al. 2013; Tomasello 2014; Tollefsen 2015; Zahavi and Rochat 2015). 2 One important lesson of these research projects is that the capacity to share mental attitudes plays a significant and ineliminable role in shaping individual cognition, yet shared cognitive activity also relies on the exercise of skills and abilities that are biologically and culturally situated in social interactions (Huebner 2014), perhaps in the sense hinted by Epstein’s remark as to how to construe the relationship between anchors and grounds.
The extent to which all relevant processes of sharing and alignment ought to be assessed, and possibly ruled out from a theory of social ontology, depends on what sort of factors collective attitudes are. In the absence of an informed analysis of the functional role of those processes, any conclusion that collective attitudes are not necessary (sufficient) in social ontology would be no less hasty than the much-criticized claim that collective intentionality is the building block of the social reality.
Chant, Sara Rachel, Frank Hindriks, and Gerhard Preyer (eds.) (2014): From Individual to Collective Intentionality. Oxford: Oxford University Press.
Dale, Rick, Riccardo Fusaroli, Nicholas D. Duran, and Daniel C. Richardson (2013): “The Self-organization of Human Interaction”. In: Psychology of Learning and Motivation 59, p. 43–95. [Web of Science] [Crossref]
Epstein, Brian (2014): “How Many Kinds of Glues Hold the Social World Together?” In: Mattia Gallotti and John Michael (Eds.): Perspectives on Social Ontology and Social Cognition. Dordrecht: Springer, p. 41–55.
Epstein, Brian (2015): The Ant Trap. New York: Oxford University Press.
Friston, Karl and Christopher D. Frith (2015): “A Duet for One”. In: Consciousness and Cognition 26, p. 290–405. [Web of Science]
Guala, Francesco (2007): “The Philosophy of Social Science: Metaphysical and Empirical”. In: Philosophy Compass 2, p. 954–980. [Crossref]
Huebner, Bryce (2014): Macrocognition. New York: Oxford University Press.
Knoblich, Gunther, Stephen A. Butterfill, and Natalie Sebanz (2011): “Psychological Research on Joint Action: Theory and Data”. In: H. Ross Brian (Ed.): The Psychology of Learning and Motivation. London: Academic Press, p. 59–101.
Meijers, Anthonie W.M. (2003): “Can Collective Intentionality Be Individualized?” In: The American Journal of Economics and Sociology 62, p. 167–183. [Crossref]
Schilbach, Leonhard, Bert Timmermans, Vasudevi Reddy, Alan Costall, Gary Bente, Tobias Schlicht, and Kai Vogeley (2013): “Toward a Second-Person Neuroscience”. In: The Behavioral and Brain Sciences 36. No. 4, p. 393–414. [Web of Science] [Crossref]
Searle, John (2010): Making the Social World. New York: Oxford University Press.
Tollefsen, Deborah (2015): Groups as Agents. Cambridge, UK: Polity Press.
Tomasello, Michael (2014): A Natural History of Human Thinking. Cambridge, MA: Harvard University Press.
Vesper, Cordula, Robrecht van der Wel, Gunther Knoblich, and Natalie Sebanz (2012): “Are You Ready to Jump? Predictive Mechanisms in Interpersonal Coordination”. In: Journal of Experimental Psychology: Human Perception and Performance 39. No. 1, p. 48–61. [Web of Science]