Method

Scale-Free Abstractions

• 11 min read

We want to maximise scope and detail as well as cognitive efficiency and evolvability of sensemaking.

A good strategy for this is using shorthand abstractions like “gravity” or “market”: highly general concepts taken from scientific thinking that we use as shorthands for more detailed models. They enable us to build more complex models models by compressing the lower-level ones they stand in for.11Flynn (2007). Flynn introduces the concept of a shorthand abstraction to explain the continuous rise in average IQ results in Western societies over the last century – we can think, he argues, more complex thoughts because we encapsulate more and more complexity in these abstractions. ↩

Because of the level of abstraction they provide, they can be used in a wide variety of contexts to build such complex models, providing an effective general-purpose toolkit for sensemaking. And because they’re smaller units of information than conventional sensemaking frameworks, they’re better suited to collaborative use and improvement; they can evolve while providing enough stability to enable the accumulation of knowledge.

This also makes them useful for effective strategising because, in Richard Rumelt’s words, “the core of strategy work is […] discovering the critical factors in a situation and designing a way of coordinating and focusing actions to deal with those factors.”22Rumelt (2011) ↩ The easier it is for us to build and test models, the more critical factors will we be able to discover.

In essence, we can use shorthand abstractions to establish a shared theoretical vocabulary for collective sensemaking and strategising.

Criteria for Useful Shorthand Abstractions

As is quite obvious from an example like “market”, shorthand abstractions also introduce a considerable risk: to build, often without noticing, ideological distortions into our sensemaking.33Following Sally Haslanger, we understand ideology as a “network of social meanings, tools, scripts, schemas, heuristics, principles, and the like” (Haslanger 2017, 155), which is distorted in specific ways so that it hides aspects of the world the perception of which would question or threaten systems of power and the social order they impose. This conception goes back to Althusser and the Marxist understanding of ideology as false consciousness. ↩ If we understand everything from ideas to marriages in terms of markets, we implicitly commodify common goods and relationships and unwittingly reinforce a totalising capitalist logic.

It is therefore essential to identify and use shorthand abstractions that, to the best of our knowledge, don’t have ideological assumptions baked into them. On the contrary, we should seek out abstractions that inherently reflect the interconnected nature of our world, prioritise relations over relata, and nudge us to be humble and avoid totalising.

At the same time, they should be connected to active research programs so that the models they compress are continously scrutinised. This helps us not only avoid sticking to outdated and distorting ontologies, but also making sure we don’t overstretch them by slipping into merely metaphorical use. Science keeps us in check.

Equally, the approach is ultimately only useful if we can use the abstractions to generate tailored models of specific areas of interest that have enough detail to become actionable – if they actually increase our situational awareness. This means the approach wants to be applied and tested in actual political strategising to prove its usefulness.

We propose a specific type of shorthand abstraction that meets these criteria: scale-free abstractions.

Introducing Scale-free Abstractions

Scale-free abstractions are based on contemporary thinking about complexity, evolution, and cognition. Thus they avoid importing and imposing atomising or totalising ontologies and are directly connected to active research communities. Because they are adequately abstract, we can use them to reveal scale-free patterns – patterns that show up across different system scales, in an atom’s behaviour as well as in a society’s, and that help us identify and describe the concrete critical system properties we need to discover to develop effective strategies.44It is important to note that scale-free abstractions are not metaphors or analogies exported from one area of science into different ones, but refer to abstract models of properties or patterns that systems of different scale, from atoms to organisms to societies, actually exhibit. This corresponds to Yaneer Bar-Yam’s conception of universality in complex systems (Bar-Yam 2017) and to what Manuel DeLanda, in his reconstruction of Deleuze’s ontology, calls topological, intensive, and population thinking (DeLanda 2002). ↩

We specifically propose the following basic scale-free abstractions:55They are presented here very concisely and informally, but each of them can be spelled out in detailed and formalised, often mathematical models, involving tools from areas like statistical mechanics, theoretical biology, and mathematical topology. ↩

  1. One can consider literally everything to be a system: a group of regularly interacting or interdependent items forming a coherent whole defined by its boundary, i.e. by what is and is not part of it. Examples are a society, a community, a human being, or a cell in its body. Systems in this sense can consist of wildly different components – societies, e.g., are composed of people, organisations, places, ideas, institutions, power structures, and much more.

  2. Each part of a system can itself be understood as a system. This means that, described abstractly, the world is a set of intersecting and nested systems – for example, humans are part of different communities, which are part of a society, which is part of the Earth System (a.k.a. Gaia).

  3. A system can be fully characterised by describing the evolution of its states. This means that ontologically speaking, systems live in state spaces, where every possible system state corresponds to a point in the space. The state space of a pendulum, a very simple system, has just two dimensions (position and momentum), and the evolution of the system is a simple spiral that ends in the pendulum’s resting point. State spaces of complex systems like humans or societies have a myriad of dimensions, and their trajectories in these spaces look chaotic, but have an intricate order to them.66See Palmer (2022) for a detailed yet accessible explanation of this. ↩

  1. Which parts of its state space a system can reach, i.e. which states it can be in, is determined by the constraints acting on it. Constraints are relational properties that systems acquire by virtue of being embedded in a higher-level system.77Juarrero (1998), 234 ↩ For example, a cell that is part of a larger organism can’t move around freely, but it also doesn’t need to because it’s being nurtured by the organism; a human living a family is constrained, amongst other things, by its relative wealth, structure, rules and generational traumas.
  1. All systems are embedded in networks of constraints that shape what they are and can be – nothing is what it is in isolation. Another way of expressing this is to say that everything is interconnected – which means, as Kriti Sharma puts it, “considering things as mutually constituted, that is, viewing things as existing at all only due to their dependence on other things”.88Quoted in Escobar (2018), 101 (author’s emphasis). This insight is at the heart of complexity theory as well as the Buddhist doctrine of the “co-arising” of everything. ↩ This subverts the idea that there are essences that make systems what they are, independently of their relations to other systems.

  1. An attractor is the set of states a system tends to be in, a region in its state space it’s likely to be found in. When a system is in an attractor region, it can’t leave that region without some exogenous change. For example, the attractor state of a pendulum is its resting point, and once it’s at rest, it can’t leave that state without a push. Living systems like organisms or societies constantly reproduce the conditions under which they are stable, i.e. they work to stay in their attractor states. (That’s one way to understand what life is.) A system’s possible attractors can be more abstractly represented in an attractor landscape, with valleys standing for stable, attracting states, and hills for unstable, repelling ones. Within this analogy, the system state is usually represented as a ball rolling across (repelling) hills and into (attracting) valleys.

  2. Any system that consistently pursues goals and uses strategies to attain them in a changing environment can be described as an agent. Understood in these terms, the molecular networks transcribing DNA are agents just as much as persons, social movements, or economic systems. They all have cognition and agency, just on varying levels.99Michael Levin captures this in the metaphor of a cognitive light cone: A larger light cone means an agent can conceptualise and pursue goals that are larger in scope and further in the future because it has sufficiently complex and effective cognitive processes. These can be orders of magnitude simpler (and the goals smaller) than what we usually think of as cognition. Levin calls this “basal cognition”. (Levin 2022) ↩ Every agent that is made up of smaller agents is a collective agent – an organism, made up of organs, tissue and cells, is one, as is a social movement made of organisations, people, and ideas.

  1. Agents can copy behaviours which can be described as memes – units of information that are transmitted between carriers and undergo cultural evolution, from words and ideas to ideologies and complex social institutions. They are the building blocks of culture as a form of group-level adaptation. By taking a “meme’s eye view”1010Blackmore (1999), 37 ↩, we can treat it as an agent itself and the systems it is transmitted between as the environment in which it survives and which it uses to do so.

  1. Systems, agents and memes adapt to their environments in order to survive. They do this by changing individually during their lifetime (development, evolution of system states) and in populations over generations (evolution of systems).

  2. In the latter case, systems, agents and memes are selected if they are more successful, i.e. better adapted to their environment than their competitors for resources. In other words, what works (better) stays – from (more) stable molecules to (more) expansive cultures.

As with every conceptual tool, scale-free abstractions will be used if they are useful. This can be spelled out in three criteria:

  • Heuristic fecundity: How much new insight do they generate?
  • Explanatory depth: How deep is the knowledge they represent, i.e. how much do the scientific models they stand in for explain?
  • Action support: How well do they support effective action, i.e. help achieve political goals?

As concepts, scale-free abstractions are memes – which means they are themselves agents that adapt and get selected according to their usefulness. When we develop models using scale-free abstractions, we also contribute to their evolution and improvement, changing what abstractions we use and how.

Using Scale-Free Abstractions

Using the outlined abstractions, we can outline our current political situation in the following way:

  • “The System”, the set of social, political and economic institutions we4 live in, is a specific configuration of society as a social system. In other words, it is one of the attractor states a society can be in.
  • A key part of this configuration is ideology – the complex meme that constrains the models agents can have of society, i.e. frames their perceptions, expectations and actions.
  • What we are witnessing right now is that as this configuration starts to fray and crumble, our society’s attractor starts losing its grip, and we start moving around the attractor landscape. In practical terms, that means a breakdown of conventional norms and patterns, and the rise of novelty which may be perceived as chaos.
  • This process is both the emergent outcome of and a constraint on the interactions of the lower-level agents societies are composed of, e.g. social groups, movements, or ideologies.
  • The process also has effects on and is constrained by the higher-level systems societies are part of, e.g. ecosystems, the climate and more generally the Earth System a.k.a. Gaia.

Alex Williams’s update of Hegemony Theory1111Williams (2020) ↩ and Rodrigo Nunes’s theory of political organising1212Nunes (2021) ↩ can be seen as examples for using scale-free abstractions in political theory and analysis. They develop models that help us discover large-scale patterns in political dynamics while providing enough detail to give guidance for strategic action, proving the usefulness of the approach in concrete contexts.

We use scale-free abstractions in our analyses of the rise of the Right, the demise of the Western Empire, and the threat of civilisational collapse, as well as our strategic proposals for creating a collective agent, building alternative models, and acting under uncertainty.

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!