Dunbar's number

"In a society where the individual is not recognized, what counts above all is the tribe and the clan".

In a world where collaborative devices and horizontal organizations are multiplying, how can we think about human relational thresholds without sacrificing either the quality of links or collective power?

The notion of Dunbar's number, that cognitive ceiling estimated at around 150 stable relationships that a human being can maintain, raises profound questions about facilitation and collective intelligence practices. Is it a biological constraint, a cultural fact, or an opportunity to rethink the design of collaborative formats?

We will explore this tension in three parts: first, by questioning the real scope of Dunbar's number, then by examining its effects on facilitation and collectives, and finally by opening up an ethical and organizational reflection on its extensions.

Dunbar's number: a biological limit to be contextualized

Dunbar's number, proposed by anthropologist Robin Dunbar (1992), is based on a correlation between the size of primates' neocortex and the number of individuals with whom they can maintain stable social relationships. In humans, this limit would be around 150 people. However, this figure is not an inalterable biological truth: it reflects more an average derived from historical and sociological observations than the actual, situated capacities of human groups.

Researchers such as Lindenfors et al (2021) have shown that this limit varies from one context to another, depending in particular on social structuring, cultural norms and the technologies used. Dominique Cardon (2010), analyzing digital networks, has shown that the web does not indefinitely multiply our relational capacities: it reconfigures them according to circles of varying intensity. Not all "friends" on networks are equivalent; only a minority of digital links are endowed with real commitment.

In this sense, Dunbar's thesis can be understood as a contextual benchmark: a way of observing the density of human connection that our attention and emotions can actually sustain. This reading is reinforced by Yves Citton's (2014) work on the ecology of attention. In it, he defends the idea that our attentional capital is limited and constitutes a resource to be preserved. More than the number of people, it's the quality of the attention we can give them that's the real issue.

Designing from the threshold

Far from being a hindrance, the relational threshold proposed by Dunbar can become a strategic lever in facilitating collective intelligence. Indeed, in groups of more than a hundred participants, we often observe the effects of relational dilution: commitment wanes, tensions increase, and coordination becomes more fragile. These empirical findings are in line with Philippe Carré's (2018) analysis, which reminds us that all learning dynamics depend on the presence of meaningful relationships, mutual recognition and shared trust.

Consequently, facilitation has everything to gain by drawing on this cognitive threshold to design appropriate formats. Numerous practices - talking circles, open forums, mirrored sub-groups, reflective tandems - can be used to structure large groups into collaborative clusters capable of rediscovering this relational density conducive to co-construction. The creation of these "attentional niches" (Citton, 2014) helps avoid affective saturation and promotes a quality of presence.

In well-facilitated systems, quantity gives way to precision. It's no longer a question of amassing connections, but of enabling the emergence of ephemeral but intense communities, where everyone can be seen, heard and recognized. Joëlle Zask (2011), in her reflection on participation, insists on the importance of a commitment where attention, contribution and recognition are articulated. This triptych offers a particularly relevant framework for facilitation, in that it lays the foundations for embodied, non-instrumentalized participation.

Link ethics and organic architectures

Thinking about Dunbar's number also raises the question of social architecture. In learning organizations or self-managed collectives, the desire to maintain egalitarian relationships on a large scale can lead to relational exhaustion. So why not take on more fractal structures: autonomous cells, interdependent archipelagos, porous networks articulated around human-scale relational foci?

This is what authors like Wenger-Trayner & Wenger-Trayner (2020) implicitly propose with "social learning spaces", where each micro-group creates value in its own way, in resonance with others.

Digital technology makes this situation even more complex. On the one hand, it seems capable of expanding our communities of belonging. On the other, it challenges the depth of our commitments. Cardon (2010) notes that digital technology often reproduces logics of crushing attention, rather than broadening mutual care. Technical horizontality does not equal human reciprocity.

This is where the need for ethical facilitation comes in: it's no longer just a matter of managing a collective, but of caring for it. This means

• setting limits on the number of interactions,
• protecting spaces of slowness and silence,
• and honoring the finitude of our relational capacity.

The challenge is to prevent saturation, to avoid collective intelligence dissolving into a superficial multitude.

Finally, Dunbar's number, often presented as universal, deserves a decolonized reading. In non-Western cultures, forms of bonding are based less on the bilateral maintenance of a stable relationship than on diffuse, contextual and sometimes ritualistic affiliations. It's not certain that the Dunbar model applies to these variable-geometry collectives. This observation invites us not to make this measure absolute, but to consider it as one prism among others, to be handled with tact and discernment.

Significant relationships

Far from being a restrictive limit, Dunbar's number acts as a revealer. It highlights the tensions between density and breadth in our collectives, between the desire for openness and the need for care. For facilitation, this is not a dogma, but a benchmark: an incentive to think of the human as a finite, situated, precious relationship. This is the price we pay to keep collective intelligence alive, embodied and sustainable.

Sources

Carré, P. (2018). L'apprenance: vers un nouveau rapport au savoir. Dunod.

Cardon, D. (2010). Internet democracy. Promises and limits. Seuil.

Citton, Y. (2014). Pour une écologie de l'attention. Seuil.

Dunbar, R. I. M. (1992). Neocortex size as a constraint on group size in primates. Journal of Human Evolution, 22(6), 469-493.

Edmondson, A. (2019). The Fearless Organization: Creating Psychological Safety in the Workplace for Learning, Innovation, and Growth. Wiley.

Lindenfors, P., Wartel, A., & Lind, J. (2021). The limits of friendship: Dunbar's number deconstructed. Biology Letters, 17(3), 20200748.

Wenger-Trayner, B., & Wenger-Trayner, E. (2020). Learning to make a difference: Value creation in social learning spaces. Cambridge University Press.

Zask, J. (2011). Participate: Essay on democratic forms of participation. Le Bord de l'eau.