February 2020 – Philosophical Ostrogoth

Since reading and reviewing Paul Cillers’ ‘Complexity and Postmodernism: Understanding Complex Systems’ an idea for a post has entered my mind. This post will fulfil that idea and act as thought experiment regarding complexity and its potential application to a historical era I am interested in.

Can we identify complex systems in the past? If so, what are the implications of this for the subject of history and its relation to other disciplines, especially considering the fact that ideas regarding complexity mainly emerge from the sciences and social sciences? As far as I know complexity theory has not been used much in the discipline of history and so this post will act as a novel attempt to see if it is possible to apply it to the past. This post will have two parts; the first will use Cilliers’ criteria for defining complex systems and see if the Late Antique ‘world’ can be called a complex system. I use ‘world’ here to denote Northern Europe, as well the Mediterranean region. The second part of the post will examine any questions that might arise as a result of this interdisciplinary exploration. I will start by copying down Cillier’s criteria for complex systems below, before seeing if they are applicable to Late Antiquity:

Complex systems consist of a large number of elements. When the number is relatively small, the behaviour of the elements can often be given a formal description in conventional terms. However, when the number becomes sufficiently large, conventional means [e.g a system of differential equations] not only become impractical, they also cease to assist in any understanding of the system.
A large number of elements are necessary, but not sufficient. The grains of a sand on a beach do not interest us as a complex system. In order to constitute a complex system, the elements have to interact, and this interaction must be dynamic. A complex system changes with time. The interactions do not have to be physical; they can also be thought of as the transference of information.
The interaction is fairly rich, i.e any element in the system influences, and is influenced by, quite a few other ones. The behaviour of the system, however, is not determined by the exact amount of interactions associated with specific elements. If there are enough elements in the system [of which some are redundant] a number of sparsely connected elements can perform the same function as that of one richly connected element.
The interactions themselves have a number of important characteristics. Firstly, the interactions are non-linear. A large system of linear elements can usually be collapsed into an equivalent system that is very much smaller. Non-linearity also guarantees that small causes can have large results, and vice versa. It is a preconditon for complexity.
The interactions usually have a fairly short range, i.e information is received primarily from immediate neigbours. Long-range interaction is not impossible, but practical constraints usually force this consideration. This does not preclude wide-ranging influence– since the interaction is rich, the route from one element to any other can usually be covered in a few steps. As a result, the influence gets modulated along the way. It can be enhanced, suppressed or altered in a number of ways.
There are loops in the interactions. The effect of any activity can feed back onto itself, sometimes directly, sometimes after a number of intervening stages. This feedback can be positive [enhancing, stimulating] or negative [detracting, inhibiting]. Both kinds are necessary. The technical term for this aspect of a complex system is recurrency.
Complex systems are usually open systems, i.e they interact with their environment. As a matter of fact, it is often difficult to define the border of a complex system. Instead of being a characteristic of the system itself, the scope of the system is usually determined by the purpose of the description of the system. and is thus often influenced by the position of the observer. This process is called framing. Closed systems are usually merely complicated.
Complex systems operate under conditions far from equilibrum. There has to be a constant flow of energy to maintain the organisation of the system and to ensure its survival. Equilibrum is another word for death.
Complex systems have a history. Not only do they evolve through time, but their past is co-responsible for their present behaviour. Any analysis of a complex system that ignores the dimension of time is incomplete, or at most a synchronic snapshotof a diachronic process.
Each element in the system is ignorant of the behaviour of the system as a whole, it responds only to information that is available to it locally. This point is vitally important. If each element ‘knew’ what was happening to the system as a whole, all of the complexity would have to be present in that element. This would either entail a physical impossibility in the sense that a single element does not have the necessary capacity, or constitute a metaphysical move in the sense that ‘conciousness’ of the whole is contained in one particular unit. Complexity is the result of a rich interaction of simple elements that only respond to the limited information each of them are presented with. When we look at the behaviour of a complex system as a whole, our focus shifts from the individual element in the system to the complex structure of the system. The complexity emerges as a result of the patterns of interaction between the elements.

Having listed Cilliers’ criteria for defining complexity, I will now see if we can apply them to Late Antiquity.

This is perhaps the easiest to answer. Late Antiquity certainly consisted of a large number of elements, in terms of the number of people who were part of its ‘world’. However, source-wise we may only have access to a small proportion of the different ‘elements’, an implication I will return to later.
Individuals in Late Antiquity naturally interacted with each other, sometimes physically, other times through the transference of information [such as by letter]. The ‘world’ of Late Antiquity also evolved over time, events arose out of interactions between different elements. As you can see, many of the criteria for complex systems can be applied to human societies in general and not just Late Antiquity.
The interaction between components in Late Antiquity was rich. Individuals interacted with multiple other individuals. Gregory of Tours came into contact with a number of different elements during his career, such as Kings like Chilperic and a number of ecclesiastical figures he worked with. However, it is unlikely sparsely connected elements [such as those with few contacts or political influences] performed the same functions as those with lots of connections- raising doubts about whether we can precisely call Late Antiquity a complex system.
Interactions in Late Antiquity were certainly not predictable or linear, small causes could have large effects [and vice versa]. For example, it seems unlikely that Theoderic the Great’s diplomatic policy [a large cause] would have a had a smaller impact in the form of Hygelac’s raid on the Franks [as argued by Storms].
Interactions can defitnelty be said to be short range. Most of Cassiodorus’ Variae were directed towards people living in Italy or other Ostrogothic provinces. At the same time, rare long-range communication was possible. The Variae contain a few letters directed to the Eastern Roman Empire, as well as to foreign kings, like those of the Heruli and Thuringi. Influence in networks, could certainly be suppressed or enhanced by a number of factors, such as past friendliness and hostility.
This is one of the harder criteria to argue for. Outputs [effects] may have created inputs [causes], but it is hard to identify feedback loops in the sources, due to the diversity of events in Late Antqiuity [it is difficult to assign causal laws to the past].
The ‘world’ of Late Antiquity was certainly an open system. The Eastern Roman Empire, for example, interacted with Persia. It is hard to define the geographic boundaries of the Late Antique ‘world’ and it often seems to be more a historical tool employed by scholars, rather than a rigidly defined actuality.
Late Antiquity operated far from equilibrium due to the diversity and volume of forces within it. There was no state where opposing forces were balanced.
The Late Antique ‘world’ certainly had a history, from the influence of the Roman Empire to events that are recorded in histories or chronicles. Histories, like Gregory of Tours’, preserved memories [fictive or not], allowing the system’s past to affect its present.
Individual components in Late Antiquity did not realise they were part of a wider complex system, people did not recognise all the forces at work in the system and responded to the information they had at hand. However, at the same time, individuals may have had ideas of the a shared Roman past, this might undermine the idea that components were not aware of the whole complex system.

Overall, it seems that the ‘world’ of Late Antiquity fits many of Cillier’s criteria for being a complex system. There are some instances where we encounter difficulties, but it seems we can apply complex systems theory to the past with some sucess. What are the implications of this? It suggests we have reason to argue that the discipline of history should be more open to interdisciplinarity outside of the arts and humanities. Certain scientific ideas, particularly more philosophical ones, are applicable to the past. Another question is also raised, to what extent can the historical discipline add to our knowledge of complex systems? Its preoccupation with time, might be useful for illuminating how we should look at the history of different systems, especially with regards to the methodologies and source criticism involved.

Nevertheless, there are some issues that need to be raised about this potential new area of interdisciplinarity for history. The sciences deal with phenomena that can be found in the world, which can be tested rigorously in repeatable conditions. Whereas, history is reliant on the sources availble to try and reconstruct the past- which might limit what we can learn about past systems. Furthermore, the past cannot be examined in repeatable conditions. There are therefore some limitations that need to be considered when trying to break the boundaries between history and the sciences.

Finally, what might complexity theory tell us about Late Antiquity? It tells us to view it as a nuanced world. Old [and now not very prevalent] ideas about this era as ‘dark’ do not seem reasonable, in light of a complex systems approach. The Late Antique world can not be considered as ‘simple’, when it is viewed as a complex and dynamic system. It also teaches to not apply simple monocausal explanations to Late Antiquity, by allowing us to view its ‘world’ as full of rich and varied connections that affect the wider system. Again, we can no longer view Late Antiquity as simple, when in fact it was full of diverse interactions with multiple causes and impacts.

To conclude, this post has examined whether complex systems theory is applicable to the past by using the case study of Late Antiquity. The overall answer is that it can be, as long as we still take into consideration a number of limitations. I then examined a number of questions that might arise as a result of this applicability and how this might affect interdisciplinarity. I suggested that using complex systems’ ideas might be beneficial for both the historical and scientific disciplines, even if some questions are raised by doing so. It therefore appears that the prospect of further interdisciplinary dialogue may be achievable.

Bibliography:

Primary Sources

Cassiodorus, Variae in The Letters of Cassiodorus: Being A Condensed Translation Of The Variae Epistolae Of Magnus Aurelius Cassiodorus Senator translated by Thomas Hodgkin. London: Henry Frowde, 1886.

Gregory of Tours, Ten Books of Histories in The History of the Franks translated by Lewis Thorpe. London: Penguin, 1974.

Secondary Sources

Cilliers, Paul. Complexity and Postmodernism: Understanding Complex Systems. London: Routledge, 1998.

Storms, G. ‘The Significance of Hygelac’s Raid.’ Nottingham Medieval Studies 14 [1970]: 3-26.

This post will review Paul Cilliers 1998 book ‘Complexity and Postmodernism: Understanding Complex Systems’. Integrating theory from the sciences and social sciences with postructural thought it aims to encourage interdisciplinarity discussion by looking at how they might mutually benefit each other.

The first issue raised by Cilliers is how do we define complexity? Complexity is a framework used in the sciences to understand systems with a distinct set of features. Cilliers provides ten criteria for thinking about them. These include containing a large amount of components, having non-linear interactions, being open to the environment and operating far from equilibrium, as well as having components that communicate locally. There is not space to list them all attributes of complex systems here, as the definition of them is somewhat debated, but it suffices to say that Cilliers’ definition is broad and incorporates a number of features systems may have. The examples he uses to exemplify complexity, the human brain and language, are introduced here and used throughout the book. Meanwhile, Cilliers also makes an effective comparison to the complexity of a country’s economy as a useful introductory example. A country’s economy can be composed of millions of people, have non-linear interactions like interest and be influenced by outside factors. It also requires constant flow to exist and economic agents who usually operate with those closest to their proximity.

Cilliers also describes connectionism in the opening chapter before providing a more in-depth explanation in the second chapter. Connectionism is a form of information processing inspired by the our understanding of the human brain. It is made up of neurons or nodes that are connected to each other. These connections have a number of weights that determine the characteristics of the network. In a ‘training period’, in which the network learns its function, the weights adjust based on what it inputs they receive. Connectionism, Cilliers argues, should also be defined as a distributed rather than rule-based form of representation, there is no algorithm with central control, a theme which Cilliers returns to later in the book.

This section also introduces the idea that connectionism may be similar to Ferdinand de Saussure’s model of language, which is addressed more thoroughly in the next chapter. In Saussure’s model, language is a system of relationships between different signs. ‘Brown’ gets its meaning from how it differs to ‘black’, ‘blue’, ‘grey’ and ‘train’. It does not get its meaning because it is tied to a particularly concept of brownness. As Cilliers puts it ‘The sign is a node in a network of relationships. The relationships are not determined by the sign; rather, the sign is the result of interacting relationships’. It therefore shares a similarity with a distributed connectionist approach in that relationships take precedence over the individual nodes or signs in this instance.

In the third chapter, Cilliers also engages with the ideas of the continental philosopher Jacques Derrida. This is because Saussure’s’ ideas can have some limitations when arguing for language as a complex system. For example, his theories present language as a closed, rather than open, system. Cilliers engages with the Derridean ideas of trace and différance. Trace, in this instance, refers to how a sign has no component that fully belongs to itself, ‘it is merely a collection of the traces of every other sign running through it.’ Whereas, différance refers in one case to the system of language consisting of differences and in another sense it refers to how meaning is continuously deferred. Cilliers then states ‘the characteristics of the system emerge as a result of the différance of traces, not as a result of essential characteristics of specific components of the system.’ Again, this draws comparisons with connectionist/neural networks. Because a weight only gains significance through its patterns of interaction, it might be possible to suggest, as Cilliers does, that ‘weight’ in a connectionist network and Derrida’s concept of trace are somewhat comparable. Likewise, différance can be used to explain how complex systems always contain loops and feedbacks. Because complex systems have delayed self-altering they are similar to différance, by the fact that they are also suspended between the active and the passive.

The next chapter deals with John Searle who would explicitly reject a postmodern way of understanding complex systems. Instead, he would take an analytical and rule-based approach to the brain and language. Cilliers engages with Searle’s ‘Chinese Room’ argument which is against strong AI and then criticises it. He also discusses the Searle/Derrida debate regarding speech acts. While short, the chapter effectively refutes some of the criticisms that could be launched be against Cilliers’ connectionist approach.

After this, Cilliers moves on to talking about the problem of representation. ‘Models of complex systems’ he states ‘will have to be as complex as the systems themselves’. Based on this Cilliers suggests classical theories of representation are inadequate to describe complex systems. In fact, connectionism puts the whole idea of representation at risk. This chapter consists of four main sections. The first critiques classical approaches, which establish a rule-based approach to understanding systems. It uses Hilary Putnam’s ideas to do this. The second section looks at connectionism and how it is better suited for modelling complex systems rather than algorithmic approaches. Firstly, connectionism does not require a theory to be developed before a solving a problem. Secondly, they can generalise solutions. Thirdly, connectionist networks are robust- they degrade slower than other forms of networks. Cilliers, in this section, also deals with some of the criticisms of connectionism. In the third part of the chapter, Cilliers looks at the practical benefit of connectionism for modelling, before he looks at some of the philosohical implications of connectionism in the final section.

After discussing representation, Cilliers moves on to look at self-organisation. Self-organisation is a feature of complex systems and can be defined as the ability of a system to develop or change their internal structure spontaneously and adaptively cope with its environment. As Cilliers describes, ‘self organisation in complex systems works in the following way. Clusters of information from the external world flow into the system. This information will influence the interaction of some of the components in the system- it will alter the values of the weights in the system.’ The chapter also applies selection- from theory of evolution- to how self-organising systems form, as well as looking again at some of the philosophical implications of the arguments in the chapter.

The final chapter of Cilliers’ book looks at the intersections between postmodernism and complexity. It does this with a very nuanced approach to postmodernism. Postmodernism is presented not as saying ‘anything goes’, but instead as having a sensitivity to complexity. Cilliers uses Lyotard’s The Postmodern Condition to support his arguments in this section. Cilliers then moves on to argue that postmodern society can be compared to some of the features of complex systems he outlined earlier. For example, it is composed of a high number of components and is characterised by a series of local ‘narratives’ over a metanarrative. The author then examines language as a complex system, before arguing for a postmodern approach to science, which values openness and flexibility regarding different narratives, rather than a more constrained and conservative approach. ‘Descriptions’ of the world ‘cannot be reduced to simple, coherent and universally valid discourses’ due to its complexity. Finally, the chapter also looks at the connections between postmodernism and the philosophy of ethics.

Overall, ‘Postmodernism and Complexity’ successfully integrates a number of interdisciplinary themes in a well-structured fashion. The book introduces the idea that postmodernism and complexity may be compared to each other, opening the door to a range of potential research directions. To summarise, Cillier’s contribution here is important as well as an exciting prospect to read.

Month: February 2020

Was the Late Antique World A Complex System?: A Thought Experiment

Review: Complexity and Postmodernism, Understanding Complex Systens