Of Parts and Wholes: International Relations beyond the Human

Complex Adaptive Systems

We live, as Mario Bunge has indicated, in a ‘world of systems’. ⁸ Everything above the level of the most elementary particles is a system of one form or another. This extends from atoms through molecules, cells, bodies and upwards into social systems, solar systems, galaxies and finally, perhaps, to the universe. From a complexity perspective these systems share certain common features, and all systems provide the environment for all other systems – in that sense, we exist in a totally (to whatever minimal extent) interconnected universe.

A system in this sense indicates the possibility of drawing a notional boundary around a certain group of elements that indicates a certain degree of autonomy. ⁹ Systems overlap and intersect with other systems, but can be distinguished from the remainder of the environment. One human body (itself made up of multiple systems) can be distinguished from its environment, but interacts with its environment in multiple ways – it takes on other systems in the form of food (plant and animal systems), it is constantly invaded by other systems (in the form of viruses), its actions contribute to and are constrained by social, political and economic systems, and ultimately it will be dissipated into a range of other systems. A global economic system intersects and overlaps with a system of states, local economic and social systems, corporations and individuals. In this sense, complex systems are said to be ‘open’ rather than closed.

A second feature of complex adaptive systems is that, rather than being fixed, they are inherently dynamic, and react to their environment. They are ‘adaptive’. Central to this dynamism is the notion of feedback. Much traditional systems analysis has focused on the idea of feedback as leading towards equilibrium – described as negative feedback. Complexity thinking suggests that in addition to negative feedback, we also need to consider positive feedback – forces which take a system further away from equilibrium and to greater instability. Systems can adapt to internal disturbance, for example the ways in which the human body deals with its internal deterioration, or external influences – the global economic systems will be affected by events in the political system.

Robert Ulanowicz argues that we can draw distinctions between a conventional or Newtonian notion of systems and complex or ecological systems. ¹⁰ Newtonian systems are causally closed in the sense that developments in the system can only be explained with reference to the elements of the system. They are deterministic, implying that it is possible to predict future events. Newtonian systems are also reversible in that the laws governing behaviour operate in both ways. In other words, there are no historical developments within such systems. These systems are atomistic in that there is a mechanical relationship between the parts and the whole. The system can be disassembled and then reconstructed.

By contrast, complex systems are open. Changes in the system can potentially be explained by internal or external features, or both. Events within a complex system are contingent rather than deterministic. They are irreversible, because the character of the system and the elements within in it are constantly changing. Complex systems are historical in that ‘irregularities often take the form of discontinuities, which degrade predictability into the future and obscure hindcasting’. ¹¹ Finally, complex systems are organic rather than atomistic and, as a result, are not easily decomposed and reassembled. Consider, for example, taking a rain forest apart and reassembling it.

Self-Organisation

Underlying the notion of self-organisation is the idea that there are patterns of behaviour in the interactions of matter. Rather than matter being solely involved in chaotic behaviour, there are regularities which emerge, even if they are contingent and temporary. For Morin, organisation is more than simple interactions and self-organisation is crucial in understanding the driving forces behind the emergence of systems. ¹² In other words, organisation is not just the process of interactions which create the system. It is an ongoing process which maintains the system’s autonomy within its environment. It is what maintains the system as a system and involves a certain degree of differentiating it from the systems around it. ¹³

An underlying assumption of complexity thinking in the study of the social world is that these processes are also a feature of human interaction. Self-organisation is seen as a spontaneous and non-directed process. This implies continuity between human and non-human nature. Our perspective is that features such as self-organisation do manifest themselves in human systems in addition to non-human systems. However, there are specific characteristics of human systems. Westley et al. point to some specific features of human systems, in particular, the human capacity for abstract thought, which actually increases self-organisation. ¹⁴

Emergence

If complex adaptive systems mark the cornerstone of complex thinking, and self-organisation indicates an inherent tendency for matter to interact in dynamic patterns, then the concept of emergence provides the link between the two.

The character of the system that emerges from the interaction at the unit level is unpredictable. For complexity thinkers, the form that any system will take is not predictable in advance. There is also debate as to whether the form that a complex system takes can be causally traced back to its original features with any degree of certainty. A classic example of this would be consciousness. From a complexity perspective, consciousness is an emergent feature of the interactions of the brain. Consciousness could not be predicted from an examination of the parts of the brain, or the way in which they function.

Complexity thinkers argue that the emergent features of a complex adaptive system always represent something more than the sum of the parts. The character of complex systems cannot be understood by simply combining the component parts. ¹⁵ There are new features which in some way constitute an added element which is only perceptible at the system level. These features, although a characteristic of the interactions of the component parts, cannot be understood simply as the interactions of the parts, but have to be studied at a systemic level. ¹⁶ The idea of emergence also implies that systems are less than the sum of their parts. For Morin, systems, in addition to the new features that they exhibit, also place constraints on the actions of the components. ¹⁷ In other words, an emergent feature of complex adaptive systems is the restrictions that complex systems place on the elements. Hence systems, as well as being more than the sum of the parts, are also ‘less’ in the sense that they remove some of the freedom of action of the component parts in the way of constraints.

Emergence is also a transformative process. As Morin puts it, ‘everything which forms transforms’. ¹⁸ This points to the dynamism of a complex adaptive system as changes at the unit level will also imply new forms of emergence at the systemic level. If we take the European Union as an example of a complex system, then we would expect the member states to have been transformed by their membership. They are not the same states that formed the organisation, and the organisation itself has also changed as a result of transformations at the unit level.

The concept of emergence is inherent, then, in understanding complexity thinking as an alternative perspective because it provides the explanation for why emergent features are ‘logically undeducible and physically irreducible’ to the component parts. ¹⁹ So, what does this imply for thinking about structure and agency in international relations?

From Evolution to Revolution

Now, comrades, what is the nature of this life of ours? Let us face it: our lives are miserable, laborious and short. We are born, we are given so much food as will keep the breath in our bodies, and those of us who are capable of it are forced to work to the last atom of our strength; and the very instant that our usefulness has come to an end we are slaughtered with hideous cruelty.…

Why do we then continue in our miserable condition? Because nearly the whole of the produce of our labour is stolen from us by human beings. There, comrades, is the answer to all our problems. It is summed up in a single word – Man. Man is the only real enemy we have.

Inspired by the ideas of Major, pigs lead farmed animals in the revolutionary overthrow of their human oppressors and institute a new regime for a new world – ‘Animal Farm’. Orwell’s book is read as a satire because it would be ludicrous to read it any other way – pigs are pigs and human beings remake their worlds. That is just ‘the way things are’, as the malevolent cat informs the pig who would be a sheepdog in Dick King-Smith’s children’s novel. The concept of agency has almost always been understood as applying to beings with wills, desires and intentions. It has been seen as a capacity for action which (usually) is a product of social relations, a relationally generated capacity. This capacity is what makes us human. Agency is concerned with the ways in which our world is reproduced and recast. Whilst non-human beings and things evolve, we humans, we political animals, go about planning and changing our world.

At least this is our common understanding. In the US sci-fi film The Rise of the Planet of the Apes (dir. Rupert Wyatt, 2011), however, we see non-human Great Apes become revolutionary subjects. The film’s main protagonist, Caesar, has enhanced intelligence passed on from his mother, a laboratory chimpanzee tested with gene therapy drugs intended to cure Alzheimer’s disease in humans. When the research companies’ laboratory chimps are culled after a false alarm about drug side effects, Caesar is taken home and raised by one of the scientists. In the house, Caesar occupies various roles – a very well-cared-for laboratory subject, a child and a pet. Outside the home, he is viewed by strangers as a dangerous addition to the suburban idyll. After an incident with a neighbour in which Caesar displays aggression, thereby behaving too much like a chimpanzee, he is handed over to a primate ‘sanctuary’ in which human keepers abuse the non-human Great Apes in their care and where Caesar finds himself isolated from, and bullied by, his own ‘kind’. Caesar’s advanced intelligence, however, enables him to make alliances and subdue enemies and incite other non-human Great Apes to rebellion. He orchestrates their exposure to Alzheimer treatment, their escape from the primate centre and the liberation of other Great Apes from a laboratory and a zoo to the sanctuary of a redwood forest nature reserve. Misanthropy is a key theme of the film, which ends with the hint that whilst the new Alzheimer drugs enhance intelligence in other Great Apes, in humans, they have terrible and contagious side effects.

Certainly, in a context of abuse by medical testing and incarceration, and of vulnerability in a human-centred system of social relations, Caesar and his comrades have much to be angry about. But this is not a posthumanist film. Rather, it is a transhumanist one – non-humans are enhanced by the incorporation of human qualities. The Rise of the Planet of the Apes is an old-fashioned tale of revolution in which non-human Great Apes have agency to the extent that they can become human, or are subject to ‘uplift’. There are two indicators of the possibilities for ape rebellion: reflexivity and a particular kind of communication, language. Caesar demonstrates both these qualities and at this point it seems only polite to bring him into conversation with Margaret Archer. Luckily for Margaret, Caesar’s awakening as an uplifted subject does not mean that he fully embraces enlightened chimpdom. He keeps his trousers on.

Archer, Agency and the Uplifted Chimpanzee

In much posthumanist work, whether in science, technology and society (STS), political geography, animal studies or elsewhere, the term ‘agency’ is broadly defined and used in different ways. For example, agency is often used to refer to a being with the capacity for action. Action is usually defined as the ability to make decisions, choose options and thereby act. Agency is seen as a causal power invested in singular entities – persons, beings or ‘things’. In some cases, it is implied that agency is a universal property of sentient beings. Leslie Irvine, for example, identifies dogs as ‘agentic beings’, ³⁶ with agency defined as ‘the capacity for self-willed action’. ³⁷

For Archer, this commits the sin of conflation. What people like Irvine do is to muddle the capacity for action, which dogs or chimpanzees clearly have, with agency. Archer makes a distinction between ‘actors’ who are singular and ‘agents’ who are plural. ³⁸ Agents are collectivities formed by those who share the same life chances. This position is involuntary because the social relations within which we may find ourselves, and the advantages or disadvantages they imply, pre-date our arrival. So, agents come into being in historically and geographically specific and contingent locations in society’s distribution of resources. From birth, we are enmeshed with the structural relations of society and relations of social power. The extent to which we can exercise choice and act on the decisions we make is linked to these structural relations. Born into a captive life as a research tool for human gain, Caesar does not have the best start in life, nor would any other chimpanzee so placed. Non-human animal being and lives are shaped by their locations within a distribution of resources where their habitats and their bodies are subject to human interests and depredation. Their lives are subordinated to the desires and whims of humans and the material conditions for their survival are frequently disregarded. Archer calls this involuntary placement in social relations ‘primary agency’, and there is no reason, other than religious belief or humanist prejudice perhaps, why this cannot be extended beyond the boundaries of the human to other species. This placement, however, is insufficient for either Major the pig or Caesar the chimp to rebel against their oppression by social structures which privilege the human.

After spending hours free in the trees of a nature reserve, Caesar is called back ‘home’ and taken to the car on a leash. On seeing a dog similarly tethered, he is troubled by the multiple identities of pet and almost-human companion he is expected to occupy. Later, in his cage in the primate centre, he shares his experiences of frustration in sign language with a fellow-signing orang-utan. Here, Caesar the individual is reflecting upon the involuntary and oppressive situation in which he finds himself and the arbitrary and unjust nature of human power. He recognises what he shares in common with the other primates and orchestrates collective action to challenge human power and change the life situation of some of his fellows. For Archer, this would be an illustration of the move from primary agency to corporate agency, albeit Archer’s discussion of corporate agency is on the firmly humanist territory of the emergence of social movements.

Whilst Archer herself would have no truck with animals, Bob Carter and Nickie Charles use Archer to inject some rigour into discussions of agency in sociological animal studies. For them, animals can possess primary agency but are unable to move from this to corporate agency. This is because they lack: ‘The recognition of shared life chances, an assessment of their possible causes and judgements about possible political remedies requires the mobilisation of political, cultural and linguistic resources.’ ³⁹ Carter and Charles allow that the sociality and sociability of many animal species is well established, including recognition of hierarchies, socialisation of infants and struggles over resources. Indeed, they suggest that forms of sociality amongst human and non-human animals are perhaps best viewed as a continuum, extending from those animals for which sociality is meaningless through to those with extensive sociality, such as dogs, horses and higher primates. However, for them, the development of syntactical language amongst humans marks a qualitative break with this continuum.

Here, Carter and Charles effectively reproduce the distinction between ‘first’ and ‘second’ nature found in Jonathan Balcome’s work. Balcome warns us that science is rapidly revealing evidence that all kinds of animal species ‘are more aware and more sophisticated than we thought’. ⁴⁰ However, language enables further cultural emergent properties, setting humans on a qualitatively different evolutionary course from that of other animals. The distinction between first and second nature underpins earlier work by the social ecologist Murry Bookchin, who describes human exceptionalism in terms of community diversity, language and use of technology. ⁴¹ A further distinction is institutionalised hierarchy, an ‘exclusive characteristic of second nature’. ⁴² Carter and Charles similarly suggest that the symbol-laden nature of human social life means that the struggles over resources are political for humans, whereas for animals, ‘the struggles over who gets what, when and how take a different form’. ⁴³ To substantiate this claim, Carter and Charles use Frans de Waal’s work on chimpanzees, although it is uncertain that de Waal’s research supports their argument. De Waal explicitly argues that chimpanzees do have politics and that ‘the social organization of chimpanzees is almost too human to be true’. ⁴⁴ De Waal argues chimpanzee politics operates at two levels: the formal hierarchical system of rank (which is negotiated and socially formalised) and the networks of influence through which chimpanzees build coalitions, engage in exchange and maintain the balance of power. De Waal’s chimpanzees may be rather feudal in their politicking for contemporary democratic tastes, but political animals they are.

We would concur with Ted Benton that syntactical language is part of the ‘mode of life’ which is ‘peculiar’ to human beings and ‘the outcome of [a] unique evolutionary history’. ⁴⁵ We would not accept, however, that without this specific kind of communication, all kinds of non-human species are excluded from what Archer calls corporate agency. Benton argues that many species have overlapping forms of ‘species life’ with humans and that we should think about ‘differentiations’ rather than differences. ⁴⁶ Differentiations of species, and particular social, economic and ecological contexts, give rise to different categories of human–animal relationship. In addition, there are non-lingual forms of communication which species may share. ⁴⁷ So, Caesar can plot and plan, scheme and manipulate, without the uplifting effects of intelligence-enhancing drugs. De Waal’s chimpanzees indeed managed to orchestrate a group escape from their enclosure. But, unlike Caesar, they did not challenge their domination by humans. There are two questions to be asked here. Firstly, is it because, not having been subjected to ‘uplift’, they cannot know that they are dominated? Or, secondly, is it because they chose not so to act in the circumstances in which they found themselves?

For Archer, Carter and Charles the answer is straightforward – chimpanzees cannot know they are dominated. Language is necessary for reflexivity and this is ‘critically dependent on the linguistic ability to hold an “internal conversation” with oneself’. ⁴⁸ For Archer, the ‘internal conversation’ is the ability to order our concerns, to consider judgements and make decisions on the basis of them, leading to increased self-knowledge. ⁴⁹ Carter and Charles argue that language capacity is what enables the capacity to have awareness of one’s own consciousness. This is why, they might argue, Caesar needed the drugs. The work of animal behaviouralists, however, increasingly suggests that some non-humans, despite being without language, seem able to think about the future and engage in forward planning. Those like Bekoff would err on the side of caution when it comes to claims about what animals might or might not be able to do. We would concur. Is it possible for some non-human species to be reflexive? It would not surprise us if it was but we do not know. We would suggest that critical realists like Archer, Carter and Charles do not ‘know’ that this is impossible, either. In response to the second question above, we would pose another. What might it mean to be reflexive when you are a chimpanzee born into captivity in a society in which species relations overwhelmingly privilege the human and non-humans are exploited and depredated? As Carol Adams and Marjorie Proctor have remarked, a lack of contestation must not be interpreted as a sign of complicity when it comes to those species over which we exercise overwhelming power and control. ⁵⁰ What animals may or may not reflect on, and what their possible actions might be in a given situation, are extraordinarily difficult to ascertain. To assume that animals are unaware of, are complicit in or benefit from the oppressive institutions and practices into which we place them is an assumption easily made and conveniently undisruptive for human interests.

So, what might we draw thus far from placing Caesar in the room with Margaret? Margaret Archer, Caesar and the chimpanzees in de Waals’s study are all agential beings and we would suggest that, in addition, they all have agency. It is of different kinds, however, and this is the important intervention that Archer makes. Archer’s differentiation between ‘primary’ and ‘corporate’ agency is useful. The idea that agential beings emerge into a pre-existent web of social relations is vital to understanding the possibilities of agency. We prefer to talk about this placement in a situation of unequally distributed power and resources as reproductive agency. Herein, agential beings are placed in a context of social institutions and social relations and their practices over time reproduce those institutions and relations, with relatively minor alterations. Placement in contexts of inequality means that relations are not always apparent and the behaviour of agential beings often reproduces such patterns. Both humans and non-human agential beings and things have this reproductive agency. We also suggest a secondary form of agency which draws on Archer’s notion of corporate agency. Transformative agency is a term we would use to describe the cultural struggle over resources and social organisation to effect differences in that distribution. Whilst this is very much associated with humans, some species of non-humans have this capacity for social action and may even in Archer’s terms have a relationally generated capacity for such action. We have focused thus far on humans and non-humans who are relatively close to them. What of the agential powers of distant beings and things? What of the effects of natural systems – these have causal powers and emergent properties, but might they also have agency? In order to discuss these questions, we need to look at some of the perspectives on agency more commonly associated with ‘posthumanist’ approaches, and here we develop a third concept of agency, affective agency.

Posthumanism and Accounts of Agency

The history of political theory has ascribed political will and agency to humans alone. Even those contemporary thinkers with challenging and radical notions of the nature of decision-making in democratic political institutions consider that (human) speech is the medium of politics. ⁵¹ Those political theorists with environmental credentials have also cast both technologies and ‘natures’ as outside of political life – as the objects of political attention rather than political agents with affect and causality. ⁵² Those experimenting with radical notions of ‘environmental democracy’ have only experimented so far and tend to conceive of politics as a human-exclusive sphere of activity and democratic decision-making that is separate from the objects over which democratic deliberation takes place. ⁵³ Posthumanism can be broadly understood to be the decentring of human beings and human interests, and raises fundamental questions for our human-exclusive conceptions of the political, including political actors and the constitution of agency. ⁵⁴

Posthumanism has been influential in disrupting the binary categories of Western philosophy such as nature/culture or human/animal and there are various strands and projects. Particularly influential have been STS approaches. ⁵⁵ Enter stage left Bruno Latour, whose work has consistently problematised our humanocentric ways of seeing the ‘social’ world, albeit that he does not approve of the term posthumanism itself. Latour’s influential Actor Network Theory (ANT) holds that agency may be attributed to any object or ‘actant’ temporarily constituted by the emergent web of ‘materially heterogeneous relations’. ⁵⁶ Here, agency is inflated conceptually (so that it becomes simply a capacity for action) and extensively (so that anything that has an effect on something else is seen as an actor or actant, from fishermen to scallops). ⁵⁷ In the networks of ANT, any material entity can have agency, including all non-human animal species and inanimate objects. Other posthumanist positions lend themselves to progressive political projects such as those characterising critical animal studies, political ecologism, feminism and others. ⁵⁸ It has been suggested that exploitative and oppressive relations exist and must be taken seriously, and that their challenge has seen the emergence of a posthumanist, non-anthropocentric way of understanding and relating to other animals and the natural world more generally. ⁵⁹ These developments are more congruent with our own position because, in such accounts, the binaries can also be understood in terms of systemic relations of social power that we would see challenged.

John Law’s version of ANT focuses on the practices and performances which create realities as material. ⁶⁰ Different realities are created by different kinds of practices. This, however, is problematic when applied to Law’s example of cattle, which exist together with farmers in a web of relations which ‘produces them as different’. ⁶¹ This is problematic in two ways. Firstly, as Ted Benton points out, cattle are creatures with a specific material existence and powers and properties which are not context-dependent. ⁶² Secondly, how we understand the web of relations which brings cattle into being depends on how we conceptualise the web – whether cattle and farmers are co-producers of human food and of a specific spatial configuration by their use of land, or whether cattle are living commodities embedded in relations of exploitation in the local, regional and global networks of the industrial production of the human resources of milk and meat. Law’s web of productive relations is both simple (it is understood in a singular, local or regional dimension) and flat (non-hierarchical). Cattle breeds have certainly been brought into existence by a web of relations, literally and materially, for they would not exist without the dairy and meat industries, but this is not Law’s main concern. The flat, non-hierarchical networks for ANT cannot deal with power because ANT cannot make distinctions between nature and society or, for that matter, between humans, other animals, plants and objects. In theorising power, we need such distinctions. As Benton argues, the categories of species, however problematic, retain some saliency because there are distinct features of species life which some ‘species’ possess in relation to others. This means that our ideas of agency must vary with species. In addition, we need to acknowledge the systemic relations shaping collective agency and agential power. We argued above for a conception of differentiated agency in which the agential being of non-human animals, particularly mammals, is countenanced, and the possibilities for agency very much depend on the relational systems which produce such being. What, then, of far more distant beings and even ‘things’?

Jane Bennett argues for a ‘vital materialism’ in order to recognise the role of apparently inanimate matter in affecting and configuring situations and events. ⁶³ Bennett’s argument about the animism of all things resonates in different ways with the complexity understandings of Capra and of neuroscientists Humberto Maturana and Francisco Varela. ⁶⁴ However, in Spinozist vital materialism there is a tendency to minimise the differences between subjects and objects with this notion of a vitality which runs through both human and non-human matter. The end in view is the development of a more environmentally aware and cautious politics, but the elevation of the ‘shared materiality of all things’ is a rather blunt instrument in securing this end. ⁶⁵ Bennett’s notion of ‘thing power’ understands agentic capacity as distributed ‘across a range of ontological types’. ⁶⁶ In describing an electric grid as an assemblage of vital material and in her use of the term emergent properties, Bennett is actually close to the notion of a complex system. ⁶⁷ For Bennett, non-human assemblages can act. However, what she actually means is that assemblages can have an impact or effect on humans and non-humans. Herein lies a version of the problem which we found with both Latour and Law – a conflation between the idea of the properties and powers of beings and things, and the notion of action and the idea of agency. The retort of those such as Bennett, or Latour or Haraway, might be that unless we allow for the agency of things, and unless we decentre human language as an enabler of agency, we cannot decentre the human in political analysis. Bennett has rightly argued that the human/non-human dichotomy in political thought has not helped many of the political demands of humanism. Conversely, however, there are serious questions to be raised about Bennett’s assumption that a distributed concept of agency will be effective in unsettling humanocentric politics. ⁶⁸

Latour, however, is more careful than Bennett. He wants non-human agency as a counter to humanocentric prejudice and as a reflection of our reality as one of the multitude of species situated in a range of ‘attachments’ on planet earth. ⁶⁹ This is necessary to accurately capture the immense causal powers of the non-human in our current political world. Complexity biologists with concepts of species co-evolution and fitness landscapes and earth systems scientists encouraging humility by humans in the face of the powerful forces of self-regulation exerted by complex systems of the natural world would agree. ⁷⁰ However, the difficulty with Latour is that, in his broad sweep, all agency is understood as being of the same quality. In addition, it is a property of ‘things’ rather than, as complexity thinking suggests, of systems in relation. Like Latour, we want to be able to discuss the ways that all kinds of creatures, beings and things, bound up in relations of complex systems and relations with their system environments, are able to, in the words of Antony Giddens, ‘make a difference in the world’. ⁷¹ We would use a notion of agency to discuss the significant effects of natural systems and the beings and things caught up in them in their relations with other systems. We consider this to be a third kind of agency, ‘affective agency’. This is not simply the causal powers of a being or thing, but a systemic impact that is collective and significant. By significant, we mean that it ‘makes a difference in the world’, that is, it alters the systemic conditions, the agential landscape, for other beings and things. The impact of global warming or the effects of a viral pandemic would be examples here.

We need a posthumanist account of agency that problematises, rather than rejects outright, dualist ontologies in the hope of shifting those binaries and challenging their effects on humans, non-human creatures and ‘the environment’ more broadly. Whereas ANT considers agency simply as a quality of material existence, we need a conception of agency which incorporates the idea that non-human life and non-human animals are social actors able to exercise agency without seeing agency simply as a capacity that material beings can exercise. We argue therefore for a situated and differentiated notion of agency that understands the ability of creatures and things to ‘make a difference in the world’ as a question of situated relations rather than intrinsic capacity alone.