Application Methods for Various Types of Social Glue

Whitehouse’s article on social cohesion provides a mix of research agenda and aspirational vision. The research agenda springs from the “Ritual, Community, and Conflict” project that he directs, but Whitehouse also aspires to employ an advanced understanding of social cohesion to “predict, prevent, and resolve civil wars,” and to “mobilize a global response to economic inequality and environmental threat.”  As a student of social group dynamics I would be excited to see Whitehouse and colleagues succeed in this regard, however fraught the task. The forces of social cohesion are critical in explaining social change and thereby provide stepping-stones to applying our understanding to higher social goals.  With these goals in mind, I would like to consider how the types of social cohesion Whitehouse tables might influence societal evolution, and if and how such forces might be employed for a common good.

Before I start I would like to dispense with an issue of which Whitehouse is certainly very aware yet did not address in the target article, namely that the forces of social cohesion and those of social diversification are merely opposite sides of the same psychological coin.  Theoretical and empirical research bears this out.  Evolutionary theory shows that when the benefits of cooperation can be isolated to the same group that pays its costs, cooperation is more likely to arise and persist.  Countless empirical studies demonstrate that human cooperation is obsessively group focused, suggesting that in-group favoritism, parochial altruism, conditional reciprocity and related behaviors are the sorts of adaptations that have made human cooperation stable and human societies successful over evolutionary time.  Cultural group selection provides a succinct and efficient way of tracking the evolutionary linkages between individuals and groups, and it is of direct relevance to the evolution of cooperation, ritual and institution.  I believe this broad region of group-centric cooperation, cultural evolution and adapted human psychology to be the relevant context for interpretation of Whitehouse’s article, and will return to the implications at the end.

On types of social glue

Whitehouse outlines two different mechanisms of social cohesion, identity fusion and social identification.  Identity fusion is a personal, emotional bond shared by small groups that have undergone intense and stressful shared experiences such as wars and initiation rituals.  Groups of individuals who have passed together through such events tend to have deep empathy for one another.  These groups are often willing to sacrifice to aid and protect their fellows.  This personal identity fusion differs in nearly every respect from symbolic “social identification,” which does not require intense personal experiences but is instead mediated by categorical relationships, rote ritual, symbolic markings and may extend across vast social groups much larger than could possibly undergo personal identity fusion.  While these symbolically marked social identities may include millions of individuals, they do so only weakly in comparison to the strength of commitment between people who have experienced intense shared events and become fused.  As a result, individuals may belong to great number of symbolic social identities, whereas, people only ever have one or a few personal identity fusions.

Personal identity fusion and symbolic social identification are, of course, categories of convenience created to simplify the empirical world.  Although they are extreme categories, with large grey zones and complex interactions, they are useful ones.  They are useful because they expose distinct social properties in a manner that makes it easier to understand how societies change over time.  Moreover, it seems that the two mechanisms of cohesion are not merely two extremes on a continuum, but, as can be seen in their links with euphoric vs. dysphoric ritual and imagistic vs. doctrinal social forms (Atkinson & Whitehouse, 2011; Whitehouse, 2002), the two mechanisms are also separable objects of empirical study.

To aid in considering the dynamical properties of these two social cohesion mechanisms (in preparation for building a theoretical model), it is useful to abstract or exaggerate their core differences.  The purpose here is to consider the properties of two pure types of glue, so that we may consider their application, interactions and dynamical properties more clearly.  I see five dimensions along which these two mechanisms of social cohesion differ, at least in theory.

Group size – Symbolic social identity may extend to millions of strangers, while personal identity fusion may only occur within relatively small groups with whom it is directly possible to share an experience.

Exclusivity Humans may simultaneously belong to many social groups, each symbolically marked in various ways, but seldom undergo identity fusion and when they do it is an exclusive formation, not usually open to further members.

Cooperative strength Both fused identities and social identities carry the capacity to elicit altruism and cooperation from group members, but the individuals whose identities have been ‘fused’ are usually willing to sacrifice more for their fellows than are individuals of the same large social group.

Symbolic membership – Social identities are often denoted with special symbols and language, or social markers.   Social markers and other outward cues of belonging may be imitated or used strategically for the benefit of individuals and groups.  Fused personal identities, on the other hand, derive membership from involvement in the defining event, and may not be imitated or manipulated as easily because members recognize each other personally.

Primary driver – Fused groups emerge largely as a consequence of traumatic events (be they ritual, catastrophe or war), while symbolic social identities arise, persist and spread through a combination of social processes including cultural diffusion and institutional dynamics.

Before coming to rest on the dynamical aspects, I will make a few notes about these two mechanisms.  Personal identity fusion is by its nature inversely related to group size and physical distance, but positively related to catastrophe, conflict and death.  While dysphoric hazing and initiation rituals make personal identity fusion a force that can be employed by societies and institutions, it is significant that in peacetime most of the fusion events that a person experiences may be exogenous to symbolic social groupings.  Childhood, the physical challenge of survival and intra-group conflict likely provide the majority share of fusion events in peacetime, and many of these events occur frequently and perpetually in human life.  This background frequency provides an important baseline of regional social cohesion against which the effects of ritual and institution and fusion due to inter-group conflict may be compared.  Furthermore, it is possible that above a certain frequency of intense events even a large population may become “fused,” in a chainmail fashion.  Such a linkage could enable warfare or peaceful collective action such as the civil rights movement.

By contrast, the symbolic markers of social identity may be used strategically by group members and outsiders alike.  Symbolic markers are ripe material for cultural evolutionary processes.   Social identity markers may often arise endogenously (Efferson, Lalive, & Fehr, 2008), are prone to psychological mechanisms of imitation (Mesoudi, 2009), and are strengthened at social group boundaries (McElreath, Boyd, & Richerson, 2003).  Moreover, symbolic social identities such as those that designate large religious populations may have played a key role in the expansion of human cooperation in the last 10,000 years (Henrich et al., 2010).  Thus, symbolic social identity would seem to be more directly susceptible to cultural evolutionary forces than fused personal identities, even when dysphoric fusion rituals are accounted for.

On mixing glues

To explore the dynamic influences of these types of social cohesion, it may be most instructive to explore a few of their possible interactions.

First, the two cohesive forces may often align.  Alignment occurs when fusion events occur within a population delimited by its symbolic social identity.  The amount of personal fusion occurring within a symbolic group may in fact be a more useful way to conceptualize social resilience.  Ibn Khaldun called collective solidarity within a group asabiyya, and Turchin (2003) extended and enriched Khaldun’s theory of the growth and decline of empires by creating mathematical models of the interaction between asabiyya, populations, and political boundaries.  Khaldun’s verbal model and Turchin’s mathematical models suggest that asabiyya was strongest within small social groups and on social boundaries.  Successful groups grow in population and space until they create vast cultural heartlands within which social homogeneity allows asabiyya to wane due to a lack of contact with other social groups.  Eventually, an over-extended empire may collapse and be replaced by one that emerges from border areas.  Mapping our two glues onto this pattern, asabiyya becomes the frequency or strength of personal identity fusion occurring within a symbolic social group.  In small social groups and borderlands it is more likely that fusion events will involve interactions with other social identities, and thereby come to reinforce the strength of the existing symbolic boundaries.  In ethnic heartlands, symbolic differentiation is low and thus cannot drive fusion events.  Any fusion events that do occur will not occur across symbolic social boundaries, because there are none, and the personal fusion component of collective solidarity wanes.

Next, the two types of social glue may compete on various time scales.  In a related interview, Whitehead suggests that rituals employing the two types of social cohesion may be competing over evolutionary time as design features in social systems (Jones, 2013).  He states that low frequency intense rituals have been eliminated in modern social groups because they are powerful and destabilize larger social structures.  This may be because symbolic social identities grow to include very large populations, and may control a larger share of observed cohesion.  Symbolic social identities are also more prone to cultural evolutionary forces than are fusion events and rituals.  It may be then that the corpus of beliefs and customs that compose a large society tends to become dominated more by the types of rituals that can be easily transmitted and imitated, causing symbolic social identities to grow by contrast to personally fused identities.

Or, it may be that for the same reasons, practices exploiting symbolic cohesion spread fast while those exploiting identity fusion cohesion spread slowly.  This last possibility would predict a U-shaped relationship between the fraction of cohesion that derives from identity fusion and time as symbolic processes at first outstrip fusion processes, which then eventually regain.

These speculative interactions between the two types of social cohesion and between the associated types of ritual may be tested theoretically (with models) and empirically (as Whitehouse is doing currently).  Related research questions include:

  • What is the baseline frequency of personal fusion events in peaceful times?
  • What is the maximum group size that can hold a fusion event?
  • What frequency of smaller fusion events is necessary to create the chainmail-like effect, or asabiyya?
  • What is the difference in cooperative strength between identify fusion and symbolic identification when measured experimentally?
  • Does the average correlation between fusion and symbolic identity (i.e. asabiyya) tend to vary across social scales, and if so, which scale leads to most cohesion, and which scale minimizes the chances of war?
  • What measures of “social scale” are most relevant (e.g. population size, social network measures, etc)?
  • What factors control the long-term evolutionary interactions between the dysphoric rituals that generate identity fusion and the more frequent scripted doctrinal rituals that help secure symbolic identification?

On applying social glues

The aspirational content of Whitehouse’s article is, of course, hopeful in the extreme.  Even ignoring the ethical black hole it conjures, we do not know if social cohesion can be effectively manipulated at all, let alone to positive social outcomes. But Whitehouse asks the right question, and I believe his categories of glue can be used constructively.  So, how might one hypothetically apply the forces of social cohesion to constructive societal outcomes?  I have two comments in this regard.

First, how might we use this knowledge to avoid war?  If Whitehouse is correct that symbolic social cohesion has enabled the expansion of human society, then we should look to applying it for its more inclusive properties over the more exclusive and local process of identity fusion.  However, we must always expect that at least a baseline level of fusion cohesion will occur.  Also, the situation to be most avoided is when identity fusion events unfold across symbolic social identity boundaries, building social tension.  This condition is often a precursor to revolution or war, and always a consequence.  Therefore, societies should avoid situations where fusion events occur across symbolic social identity boundaries.  The problem is boiling this mechanism down to a tip – to stop war, stop people of different social groups from fighting each other – makes mighty weak tea.  Perhaps, as the research matures, Whitehouse will be in a better position to elaborate.

Finally, we return to the proper context – the evolution of cooperation, group functional rituals and cultural traits.  As I mentioned earlier, there is abundant evidence that the evolved human capacities for social cohesion are tightly tied with the human proclivity to mark, segregate and discriminate between people of different types.  When we think of glue, we think of applying it to hold things together.  The problem with social glue, however, is that it binds one sub-population at the expense of a rift in the larger population.  That is, neither type of social glue necessarily generates new, additional cohesion.  Rather, human social cohesion seems to act in more of a thermodynamic way, extracting cohesion from one source and accumulating it in another.  It seems better to me to approach application from a more fundamental level, when we ask the question, “what factors can retard the processes of social segregation or ethnogenesis?” or “which mechanisms add to total cohesion across social groups?”  Two such mechanisms may be migration (Richerson & Boyd, 2008) and economic equality (Baland, Bardhan, & Bowles, 2007; Wilkinson & Pickett, 2009).

Timothy M. Waring
School of Economics
University of Maine


Atkinson, Q. D., & Whitehouse, H. (2011). The cultural morphospace of ritual form: Examining modes of religiosity cross-culturally. Evolution and Human Behavior, 32(1), 50–62. doi:10.1016/j.evolhumbehav.2010.09.002

Baland, J.-M., Bardhan, P., & Bowles, S. (Eds.). (2007). Inequality, Cooperation, and Environmental Sustainability. Delhi: Oxford University Press.

Efferson, C., Lalive, R., & Fehr, E. (2008). The Coevolution of Cultural Groups and Ingroup Favoritism. Science, 321(5897), 1844–1849. doi:10.1126/science.1155805

Henrich, J., Ensminger, J., McElreath, R., Barr, A., Barrett, C., Bolyanatz, A., … others. (2010). Markets, Religion, Community Size and the Evolution of Fairness and Punishment. Science, 327, 1480–1484.

Jones, D. (2013). Social evolution: The ritual animal. Nature, 493(7433), 470–472. doi:10.1038/493470a

McElreath, R., Boyd, R., & Richerson, P. J. (2003). Shared norms and the evolution of ethnic markers. Current Anthropology, 44(1), 122–129.

Mesoudi, A. (2009). How cultural evolutionary theory can inform social psychology and vice versa. Psychological Review, 116(4), 929–952. doi:10.1037/a0017062

Richerson, P. J., & Boyd, R. (2008). Being human: Migration: An engine for social change. Nature, 456(7224), 877. doi:10.1038/456877a

Turchin, P. (2003). Historical Dynamics: Why States Rise and Fall. Princeton, NJ: Princeton University Press.

Whitehouse, H. (2002). Modes of Religiosity: Towards a Cognitive Explanation of the Sociopolitical Dynamics of Religion. Method & Theory in the Study of Religion, 14(3-4), 293–315. doi:10.1163/157006802320909738

Wilkinson, R., & Pickett, K. (2009). The Spirit Level: Why More Equal Societies Almost Always Do Better. New York: Bloomsbury Press.

Published On: March 20, 2013

Timothy M. Waring

Timothy M. Waring

I study human culture and cooperation in relationship to the environment. In particular, I study how social norms, institutions and societies evolve, and how they are influenced by ecological and social forces. I strive to use this research to learn how to better build durable, sustainable and just institutions and societies. I use experimental economics and agent-based modeling to explore these connections, and work with lot of wonderful people.

Cooperation is a key component of fair and sustainable resource management. We therefore need to understand the forces that guide human cooperation.  My research has explored the dynamics of human cooperation in traditional irrigation systems in southern India, in university-citizen relationships, and in the laboratory.

One Comment

  • karlfrost says:

    I like your offerings to the conversation, Tim.
    I have a few topics of comment:
    1. The use of the word ‘’trauma”
    2. Speed of transmission as the deciding factor in cultural evolution
    3. Retarding social segregation: rituals for diffusing conflict.

    First, with regards the words “trauma” or “traumatic experience”, you and others in this conversation have been using the term, trauma, and I want to be clear of how it is being used. It seems like “traumatic” is being used as somehow a synonym for “dysphoric” or “stressful”. Is this how you are using it, or do you have another technical definition? I ask this because one of the other hats that I wear is that of a practitioner of somatic psychotherapy, where we have a very specific notion of what that word means. Trauma, as I use the term and understand it from this perspective, has not to my knowledge been shown to lead to bonding, but instead to disassociation (Levine, 1996). This is in fact axiomatic of what defines ‘trauma’ in psychotherapeutic practice. One can have wildly stressful experiences and not be traumatized, and one can also have moderately stressful events in one’s life and exhibit symptoms of trauma. In this use of the word, dysphoria and trauma are related but different phenomena.
    Second, with regards to speed of transmission, I think it is quite likely the case that doctrinal mode practices will generally be able to be socially transmitted more quickly than imagistic mode practices, though it had not occurred to me before you mentioned it as a possible driver of cultural evolution with respect to the practices. It would make sense that when identity is unstable that doctrinal mode religions would have the advantage of being able to be transmitted more quickly. I question whether this is essentially what has driven the success of doctrinal mode practices, however. I suspect that the answer is more likely to be found in cultural group selection through doctrinal mode practices being able to bond together larger groups, more effective in warfare and other forms of competition. It seems like a fun prospective modeling project, to look at these two theorized facets of doctrinal mode practices in synthesis with each other. My suspicion is that the group size effects are the more vital component of the evolutionary success of doctrinal mode practices, but that this success is helped substantially through speed of transmission effects, which seem most relevant when groups have been disturbed and lines of identity disrupted as in large scale warfare. Here, I think of the effectiveness of conquering groups at assimilating the conquered. Where I suspect imagistic mode societies were able to accomplish assimilation through a combination of syncretism and tolerance of ambiguity, I suspect that the potentially more rapid conversion process of doctrinal mode practices may have facilitated the effective assimilation of the conquered.
    Third, with regards to the idea of “retarding social segregation”, I am reflecting on some of the writings of Roy Rappaport on reconciliation rituals (Rappaport, 1999). He observed that rituals in Papua New Guinea, often euphoric and celebratory, were specifically used to bring together groups in conflict, to create a sense of unity across group lines and reconciliation of group differences. This also seems related to the disruption of identification with dominant, large scale religion which Barbara Ehrenreich suggests occurs with ecstatic ritual practices (Ehrenriech, 2007). To continue the glue metaphor, perhaps instead of looking to glue larger groups together, we should be looking for solvents to unglue clusters when those clusters can no longer stick to their surrounding society.

    Ehrenriech, B. (2007). Dancing in the Streets: A History of Collective Joy. Metropolitan Books.
    Levine, P. (1996). Memory, Trauma & Healing. Journal of Bio-Synthesis.
    Rappaport, R. (1999). Ritual and Religion in the Making of Humanity. cambridge: Cambridge University Press.

Leave a Reply

<textarea name="ak_hp_textarea" cols="45" rows="8" maxlength="100" style="display: none !important;">

This site uses Akismet to reduce spam. Learn how your comment data is processed.