Strategy

Collective Agency

‱ 11 min read

In order to effect the change we seek, we need collective power. But collective power is exactly what’s missing on the political Left, while the Right seems to be gaining it in abundance. How can we conceptualise how collective power is built in a fresh way that gets us unstuck?

Collective Power after Marxism

Why we need collective power is quite clear:

[T]he powerful always have what Baruch Spinoza called potestas to ensure that, when push comes to shove, people will do their bidding: the police, the army, the press, the wage relation, the accumulated fear and passive consent of the majority, all sorts of things that could be described as ‘power over’ or grouped under the vague expression the powers that be. The weak, on the other hand, have nothing but their capacity to act – their power to do things, to affect and be affected by each other, which Spinoza called potentia. Yet each individual’s potentia on its own is not much, and certainly not enough to face down potestas. It is therefore imperative that individuals come together, the capacity to act of each multiplying the capacity of all others. That is why the subject of politics is always collective.11Nunes (2021) ↩

But collective power isn’t the mere accumulation of indiviudal power – it is the power a “collective subject or agent” wields.22Ibid. ↩ The subject that Marxism has sought to develop is the working class. Today, the working class is fragmented, and its boundaries are overlaid by other, more consequential distinctions based on culturally constructed identities, hegemonic narratives, and the resulting web of specific social roles and relationships, making it an unlikely candidate for this function.

At the same time, we can clearly see the contours of a coherent social structure emerge on the right, badly captured as it is by labels like MAGA, populism, far right, or fascism. It consists of widely different social groups; narratives and specific collective identities define their political consciousness just as much as (or even more than) economic position and material needs, transcending the traditional framing of social structures as classes.

This calls for a reconceptualisation of collective power that takes the central insights of Marxism, e.g. the importance of collective agency and the grounding in material reality, and develops them further in light of our current situation.33More generally, our approach diverges from Marxism – following thinkers like Alex Williams (Williams 2020) and Rodrigo Nunes (Nunes 2021) – in emphasising emergence over determinism, composition over transitivity, and collapse-proof radical alternatives over top-down revolution. ↩

To do this, we suggest using a theoretical framework for thinking about collective agency that helps us capture the complexity of the challenge (empirical adequacy) and generate new strategic insights (heuristic fecundity).44This framework is based on a set of scale-free abstractions, most centrally that of an agent. ↩

Our Theoretical Framework

Our framework builds on the work of researchers like Karl Friston in neuroscience, Chris Fields in physics, and Michael Levin in biology55See, e.g., Friston (2017) and Levin (2022). ↩ and is based on a very general understanding of agents as a goal-seeking autonomous systems with varying levels of cognition and agency. We think such a general framework has the advantage of being applicable to a wide range of circumstances and potential political subjects.

According to Levin, agents are systems that are “the owner of goals, preferences and memories that don’t belong to their parts”. An agent is an emergent, coherent self

that is implemented by an interlocked triad of three things: a problem space within which it operates, a cognitive light cone that demarcates the size of the goals that it can pursue, and a set of cognitive processes that allow the system to navigate the problem space with some degree of competency.66Levin & Metzinger (2024) ↩

From this perspective, agency shows up across system scales and domains – molecular networks involved in DNA transcription and our inner organs count as agents just as much as social movements and political ideologies. Each of them lives in a different world, i.e. works in a very specific problem space. Sometimes it’s hard for us to recognise an agent whose problem space is very different from ours.77This paradigm shifts our perspective from a positivist understanding of the world as fully determined and, in a sense, “dead” to a basically animistic view of the world as co-created by a multitude of autonomous agents. Although it is grounded in the natural sciences, the underlying assumptions transcend traditional disciplinary boundaries and apply to both natural and social systems. We propose this perspective here because we believe it will be a useful lens for social movements, but, as always, the proof is in the pudding – i.e. in using it successfully to develop effective strategies. ↩

Treating systems as agents allows us to interpret and interact with them on the level of system-wide goals and strategies – we can literally think of, e.g., QAnon, the emerging hegemonic order, or our global civilisation as agents pursuing goals (e.g. survival and reproduction) and following strategies (e.g. expansion and extraction).88It’s important to note that we understand agency ascriptions instrumentally, not realistically: As with all models and concepts, we use them because they are useful, not because they truthfully represent some underlying reality. We just call systems that fulfil these criteria agents because that hopefully allows us to develop better options for action. ↩

This understanding of agency also allows us to reframe the challenge of building collective political power as an instance of a more general phenomenon, the emergence of collective agents, and our role in it as catalysers of this process.

Emergence of Collective Agency and Consciousness

Using the framework, we can now restate our challenge in more technical terms: How do we catalyse the emergence of a sufficiently large politically conscious collective agent?

Using a model developed by philosopher Thomas Metzinger99Metzinger (2003). Michael Graziano’s “Attention Schema Theory of Consciousness” has a very similar structure, replacing the idea of a general self-model with that of a more specialised “attention schema” (Graziano 2019). Our analysis below can be adapted to this framework as well. ↩, one can understand collective consciousness as the self-model of a larger-scale or higher-level agent that emerges from the long-term interactions between smaller-scale or lower-level agents. In this sense, all consciousness is collective: our “individual” consciousness is the self-model of the agent emerging from the long-term interactions of our cells, tissues, organs, etc. – our organism.

There are several conditions for the development of such an agent and its self-model:

  • Implicit or explicit coordination between lower-level agents so a higher-level system can emerge from their interaction. This coordination is easier if the lower-level agents have shared goals and interests.

  • The emergence of a coherent self with1010Levin (2022) ↩

    • a problem space within which it operates and which is different from the problem spaces of the lower-level systems,
    • a cognitive light cone that demarcates the size of the goals it can pursue,
    • a set of cognitive processes to pursue goals and navigate the problem space.

  • In order to pursue goals, the emerging agent needs a model of itself. This model describes its current state (e.g. hungry), its target state (e.g. satiated), and how they are connected (e.g. through eating). Such a self-model:

    • can be implemented in a variety of ways and doesn’t need to resemble explicit or symbolic models but can be implicit and embodied, e.g. implemented in a body plan or in intuitions;
    • can be highly simplified, e.g. by representing the complex biological system that we really are as a nonphysical, quality-less essence, the “self” we intuitively think of ourselves as.

Interventions to Catalyse this Emergence

Following this explication, there are several things we can concretely do to catalyse the emergence of a politically conscious collective agent:

  • Clarify which smaller-scale political agents share material and psychological interests, i.e. whose futures depend on similar conditions1111Standing (2019) proposes that there is an emerging social class, the “precariat”, which consists of people who share the precariousness of their job situation, from migrant service workers to academics without tenure, making them vulnerable to exploitation (of course to different degrees). Although they are different in lots of other respects, they share a central problem around which they could align. That this grouping cuts across many other social boundaries (industry, type of work, geographic region or urban quarter, social capital, existing collective identity) is opportunity and challenge: It potentially unlocks significant class size and thus power, but it also means the shared problem needs to be prioritised over other collective identities, which requires significant amounts of popular education and propaganda. ↩

  • Expose and undermine hegemonic narratives and systems of division that separate agents with aligned material interest

  • Discover or invent other shared interests, e.g. through identities and moral, social or aesthetic norms

  • Collectivise resources to create a shared material interest and necessitate coordination between smaller-scale agents

  • Create implicit coordination mechanisms for them, e.g. through spatial closeness, shared experiences or co-regulation of nervous systems

  • Support explicit coordination of collective action, e.g. through an organisation, union, party, alliance, or game

  • Understand the problem space the emerging collective agent is operating in and develop collective cognitive processes that enlarge its cognitive light cone, e.g. through research, training, or collective sensemaking

  • Offer a simple (abstract, vague) but sufficient self-model, e.g. a myth, national identity, ideology, or conspiracy theory

For an effective political strategy, every group, organisation, or movement needs to select the option(s) from this list that are the best match for their resources. This will maximise their contribution to the emergence of a larger collective agent with its own emergent strategy.

For example, matching the resources at our disposal, the Movement Ecology Collective will focus on

  • clarifying shared needs and interests in popular education,

  • supporting implicit and explicit coordination through convening encounters and facilitating collaboration and reconnection,

  • augmenting collective cognitive processes by offering training in, facilitation of, and tools for sensemaking and strategy development, and

  • developing, testing and seeding effective memes as evolving self-models for an emerging collective agent.

Overcoming Separation

Our ultimate problem is our globally dominant exploitative civilisation, and the idea of separation at its heart. Any new collective agent needs to work towards an alternative future based on interconnectedness. This means it needs to respect limitations, establish and maintain feedback loops with its environment, and refrain from exploitation.

This requires a self-model based on interconnectedness. We can foster its emergence in at least three ways:

  1. Frontloading its development with memes that bias it towards this idea.

  2. (Re-)producing a culture of connection: developing and spreading alternative models of production and coordination that are based on an assumption and experience of interconnectedness with each other and nature.1212See the work of Organising from Elsewhere for an example of this. ↩

  3. Staying in close connection with the feedback that we receive about the direction we’re going in.

This isn’t something we’re currently good at on the political left:1313A recent study (Hogdson & Tryl 2025) shows that “progressive activists” are more likely to refuse to work alongside people with opposing views compared to other groups pushing for change within the UK. They are also less willing to discuss topics they deem too harmful within the context of democratic debate, compared with other groups. ↩ Already notorious for infighting, the twin dynamics of cancel culture and purity politics have heightened our tendencies to reject those who don’t agree with us and use tactics like boycotting against potential allies.

This is not to say we shouldn’t try to influence each other, and of course there is reason to redirect energy away from those who drain it. But giving up on each other can be all too quick. It weakens the movement as a whole by reducing our collective power and doesn’t reflect the belief that our collective liberation is intertwined.

It’s not the fault of individuals who have been socialised in an exploitative system to reproduce these toxic behaviours. If we focus on individuals instead of the system dynamics that shape their behaviour, we have fallen prey to the habit of looking at the world through the lens of separation.

If we want to change behaviours, we need to change culture by developing new models of working together. Our socialisation is not our fault, but it is our responsibility to overcome it and learn new ways of relating to each other, and to support others to do the same and hold them lovingly accountable.1414This is not to absolve individuals who perpetuate this toxic system. If individuals knowingly enact harm and have the power to change that, it’s their responsibility to do it. It is to say that if we want to change them and their behaviour, we need to change the constraints and incentives that keep them in the grip of our extractive and exploitative culture. ↩

References

← Back to Articles

Stay Connected!

Get updates on new articles, workshops, and resources for building collective power in social movements.

Ready to Build Collective Power?

Get in touch to discuss how we can help you or your group navigate the complexity of movement building.

Drop us an email!