part 2: BETWEEN SCIENCE AND ACTION – WHY COMPLEXITY

2.1  Disrupting the Reductionist Universe

The aim of any action is to alter an existing reality. What we do, and how we collectively organise our resources are all based on our understandings of cause and effect relations in our environment and our own capacities to leverage them. These in turn, are based on our fundamental understandings of reality and how it unfolds. Throughout human history, such convictions have mostly been fashioned by culture, religion, and philosophy. Yet the last three centuries have seen the rise of scientific enquiry as the key vehicle for shaping our deepest perceptions of reality, thereby stipulating our core principles for action.

The Cartesian revolution of the 17th Century ushered in a new way of thinking about nature. Rather than one that is shaped by the whims of a divine being, to one that follows predictable, knowable, even if yet undiscovered, rules. Man, through the power of deduction and experimentation could finally uncover these rules freely, no longer dependent on the interpretations of an intermediary Church. The new philosophy of science has redefined what knowledge is. In the words of Descartes - “Science is certain, evident knowledge. We reject all knowledge which is merely probable and judge that only those things should be believed… about which there can be no doubts”[i]. The formulation and legitimation of this scientific method has unleashed a swarm of would be scientists, enduring extraordinary and many times horrific  adventures, in their quest to observe, measure, and classify everything in the world around them[ii].

By the end of the 17th century, Isaac Newton formulated the first ‘Laws of Nature’ - motion and gravity, and the idea of the ‘Clockwork Universe’ was born. This Newtonian metaphor embodied a mechanical view of nature reflecting the main technological fascinations of the time. Similarly to a giant clock, nature was perceived as constructed of numerous gears and cogs whose movements are governed by knowable universal laws of nature. To fully understand nature, all we needed was to study these constituting elements and uncover the general laws that govern them. As Newton’s Laws exemplified “find the mathematics that describes a system and you can then predict how that system will unfold”[iii], be it the movement of planets, ocean tides, or the trajectory of thrown objects[iv]. What followed was an explosion of theories and models all based on a reductionist worldview – i.e. the fundamental belief that nature could be reduced into ever smaller constituting parts whose relations could be individually analysed, and then reintegrated to form a sense of the whole.

As nature began to be observed and explained from a mechanical perspective, specific cause-effect relations could be identified to explain how the behaviour of the different cogs making up the Clockwork Universe affected one another. Most importantly, these cause-effect relations could be directly translated into practical action. New understandings into thermodynamics, the behaviour of molecules and gases, the properties of different elements, and electricity, were instantly converted into technological advances. These included new and more efficient power sources, transportation, and communication methods. The scientific revolution enabled the industrial age, ushering in a new phase in human evolution – one of ever intensifying exponential technological and economic growth and all we associate with modernity today.

Our whole transformation into modernity was enabled by Newtonian thinking, and with it the way we defined and assembled knowledge, just as the manners in which we operated as societies. Throughout the second half of the 19th Century, knowledge relating to the social spheres was also institutionalised accordingly, with economics, political science and sociology leading the scientific endeavour to uncover the universal laws for human societies. This was to be achieved by creating clear intellectual boundaries defining each discipline’s unique scope of human phenomena, and emulating reductionist reasoning from the natural sciences, including the systematic production of evidence and controlled observations based on ceteris paribus (all things being equal) [v].

The innovative knowledge provided by the new academic institutions directly served the state-centric needs of the time, as governments were quickly adapting to the new economic, social and political challenges of marrying industrial development with international balance of power. Helpfully, the Newtonian world view which clearly identified causes and effects, provided easy frameworks that could be translated into action - if A leads to B, increase in A will increase B. Governments and even military machines could thus be designed to best reflect the Cartesian breakdown of operational challenges into their constituting parts and the corresponding organisation of action. Such actions were guided by overall policies informed by ideological preferences for promoting certain social end-states over others.

Naturally, institutional structures, whether in business or government, have also evolved to embody the same reductionist scientific reasoning. Throughout the 20th Centuries corporate and government bureaucratic structures were also inspired by ideals of mechanical systems, with standardisation, division of labour, and measurable responsibilities as their leading organising rationales. These were of course a huge improvement on pre-bureaucratic structures and enabled the professionalization and proliferation of the modern state apparatus as well as the scaling up of industrial production and distribution.

To a large extent, this era exemplified a considerable alignment between our understanding of nature and our mode of operation, a synchronisation that drew on the kind of problems human societies were mostly concerned with. Whether in construction, transportation, or mass production, most challenges could be defined as ordered complicated problems, i.e. engineering-like problems in which clear cause and effect relations could be identified, and once translated into modes of action, could be replicated and disseminated as best practices[vi]. Where problems could not be simplified down to two-variable relations, new statistical tools were developed to deal with very large numbers of variables, allowing scientists to analyse average properties contributing to the management of early communication and financial systems[vii]. Thus for many years, the bulk of human action has been well served by a certain understanding of reality and the knowledge required to alter it. 

Overall, the underlying belief shared by scientists and decision-makers alike, was that by improving humanity’s access to information we would be able to solve ever more complicated problems. Overall, this domineering Newtonian paradigm upheld a world view in which the “future could in principle be known. The more careful your measurements are today, the better you could predict what will happen tomorrow”[viii].  Yet over the second half of the 20th century some cracks began to form. Pushing them wide open were both new scientific discoveries that undermined old axioms, and shifting conditions in the socio-political environments that both set the stage for new philosophical and cultural views, and directed decision-makers’ attention towards new kinds of challenging problems. 

Perhaps the key scientific discovery to help break the Newtonian tradition was the discovery of the phenomenon of ‘chaos’. Ironically, this intellectually and culturally unsettling development emerged with the best of Newtonian intentions in mind. In the 1950s and 1960s, as scientists explored ever widening areas that could utilise new computing powers, it was naturally assumed that our global weather system is just the next frontier for theorizing and modelling. Theoretically, the weather system should be like any other mechanical system. Once mathematically modelled, and given the right data, it should become fully predictable. And yet, when Edward Lawrence, an American meteorologist tested out his new mathematical equations something strange happened. His models of air currents failed to work and produced no useful predictions whatsoever - “it was as if the lightest breath of wind one day, could make the difference a month later, between a snow storm and a perfectly sunny day”[ix].

In essence it seemed the tiniest immeasurable differences in certain starting conditions could mean a completely different systemic outcome. Lawrence brilliantly captured his discovery in his famous talk titled “Does the flap of a butterfly’s wings in Brazil set off a tornado in Texas?” The radical idea suggested by the ‘butterfly effect’ was that even a fully mathematically described system could become unpredictable, this crucially, without any outside interference[x]. In other words, reductionist reasoning and information gathering did not necessarily produce new knowledge.

As the butterfly effect began to be identified across many different areas, from fluctuating herd sizes to market prices, it seemed uncertainty and randomness had to be accepted as an inherent part of any system. The idea that systematically developing theories to explain general governing laws could help us predict the world around us was unhinged, and the scientific hope encapsulated by the “Newtonian crystal ball” was forever lost.    

The discovery of chaos helped legitimise other revolutionary discoveries previously side-lined by the scientific mainstream establishment. Leading amongst them was the concept of ‘self-organisation’ explaining phenomena such as autopoesis through which identical embryonic stem cells can simply transform and organise themselves into all other kinds of cells needed for the human body; or the Belousov reaction in which an undisturbed chemical soup spontaneously creates resonating emerging patterns. Within the previous paradigm such phenomena in which systems transformed themselves without any outside intervention, were simply deemed as unthinkable. Such observations could only imply a badly run experiment as Belousov himself was told, driving him to give up on science all together.

Yet suddenly the discovery of chaos opened completely new lines for scientific enquiry in which systems could be seen as harbouring self-generating, and so by definition, non-mechanical-like properties. Such new thinking also implied the need to reconsider processes of change in the world around us and therefore inevitably also reassess our existing modes of action. After all, if the aim of any action is to alter an existing reality, then any undermining of our understanding of nature around us will inevitably also undermine our confidence in the way we operate. Such renewed questioning has also coincided with a different set of practical concerns, this time contributing to a corresponding blurring of scientific disciplinary borders. After all as George Orwell once explained “a theory does not gain ground unless material conditions favour it”[xi].   

2.2  The Complexity Turn

The social sciences were institutionalised between 1850 and 1945 in a world dominated by five knowledge generating localities – Great Britain, France, the Germanies, the Italies and the United States. To a large extent their institutionalisation aimed to provide decision makers with the means to organise and rationalise the systemic transformation into the modern sovereignty-based order still recognisable today. Thus “social science was very much a creature, if not the creator of the states, taking their boundaries as crucial social containers”[xii].  However, key transformations following WWII challenged the previously upheld division of labour between economics, political science and sociology, as well as the trend towards ever more internal disciplinary specialisation of knowledge and therefore narrowing prescriptions for action. The two most important transformations within our wider global environment related to the huge expansion in the world’s population, trade, advanced productive capacity, and the practical challenges of disseminating modernisation and development efforts across the non-Western world.  

The rise of ‘area studies’ emerged out the operational needs of decision makers to affect the non-Western world. Being both multidisciplinary by definition and practice driven by organisation, they inevitably exposed the artificiality in the division of social scientific knowledge[xiii]. Modernisation efforts in less developed countries and later in the former Communist bloc required careful integration of economic, political, and cultural concerns thereby pushing towards more inclusive conceptual frameworks.

This can be clearly seen in the evolving rationales guiding the development doctrines of the World Bank – the world’s leading vehicle for development aid. Since the 1970s the Bank continuously expanded its focus to include ever more issue areas into the modernisation endeavour. Thus, if during the 1970s the Bank focused its main efforts on engineering and physical infrastructure, by the 1980s the focus expanded to economic policies and market liberalisation, while the 1990s added support for legal institutions, women empowerment, quality of governance and environmental sustainability, ultimately reflecting a rather holistic view of nation-building.

In essence, the global quest for development and integration highlighted the natural interdependence between different areas of human activity all existing within one social system. However, the manner in which these expanded development doctrines were being implemented still reflected an accumulative reductionist approach rather than a deeper systemic integration. In other words, action was still very much in line with reductive reasoning.

While more areas of knowledge were drawn upon, the organisation of development projects still adhered to Newtonian thinking based on distinguishable variable whose impact needed to be predicted and presented as achievable end-states. Whether promoting community development projects or legal reforms, useful scientific knowledge had to conform to strict planning processes in which the relations between cause and effect, even if multiple and complicated, could be worked out in sequence so as to aim for clearly defined and operationally justifiable goals.

Unfortunately, as the pre-defined end-states were seldom achieved as planned, a growing sense of apprehension emerged, with growing mistrust directed towards both the scientific community and the development sector by politicians, funders and local communities alike. Old modes of operation where proving frustratingly ineffective, at a time when new understandings of nature that undermined the old clockwork universe worldview began to take hold.       

As the awareness of uncertainty and interdependence was further heightened by globalisation, similar concerns returned to haunt the developed world as well. Everyday experience taught decision-makers how previously held leverages over their own sovereign ecologies – especially in terms of economic policies were gradually fading, a trend that could no longer go unnoticed by their anxious citizens. The credibility of modern public policy as the articulation of social scientific knowledge through organised action was being undermined. Whether focusing on healthcare, finance, the environment, or national security, our public discourse concerning emerging challenges increasingly highlights the multiplicity, interdependency, and ambiguity of forces that must be dealt with.

A new language has entered the public sphere, describing problems as “complex”, “entrenched”, and “full of uncertainty”. How can one address problems of obesity, when they involve so many interrelated factors such as our psycho-cultural attachment to food; the economics of food production and distribution, not to mention advertising and the impacts of social interaction? What makes for a good immigration policy when new skills are needed to support innovation and entrepreneurship, yet competition for low-skills jobs is pushing down the earnings of the working poor, and when urban cultural tensions are on the rise? What can realistically be done about Boko Haram’s insurgency in Nigeria without addressing issues of economic development, ethnic-political rivalries, afro-Islamic networks and foreign investments in oil and gas? What are the distinguishable causes and effects, in other words where does current scientific knowledge help us change reality through action?

Overall our public narrative as projected through both traditional and new media platforms, suggests the need for new knowledge that would better serve our quest for effective action. Primarily, it raises some fundamental questions about our previous understanding of reality and how it unfolds. What if the systems around us are not all governed by general universal laws? What if change is not induced by outside intervention? What if the cause and effect relations governing them cannot be deduced by reductionist methods? What if cause and effect relations are neither proportional nor directly linked?  What if the entirety is different than the sum of its parts? Enter what has been coined as “the complexity turn”[xiv].

Whether in public policy or business, a growing sense of disappointment with the application of old theories to new problems, both propelled by and converged with new scientific thought, to create a greater openness to fresh ideas. A key emerging focal point was a collection of theories relating to the behaviour of complex dynamic systems as explored in physics, biology, chemistry and ecology, generally termed ‘complexity science’. Such theories aimed to explain how patterns and structures emerge spontaneously, in other words, without central control or top-down design, but rather from the bottom-up - through the ongoing interactions amongst large numbers of elements.

With the search for new conceptual frameworks across the social sciences, a growing incursion of complexity-related concepts could be detected across many fields of research. This growing fascination, commencing in the mid-1990s, also signalled the gradual breakdown of the traditional division between the ‘natural’ and ‘social sciences’, a direction further advocated by the Gulbenkian Commission on the Restructuring of the Social Sciences[xv].  Dubbed the ‘complexity turn’, this new wave of thought moved away from a reductionist analysis of social phenomena, which as discussed above, aims to isolate variables and deduce direct cause and effect relations, and towards a systemic one which aims to explain how a social phenomenon behaves in its entirety.

Key to understanding the difference between the two approaches are some fundamental questions regarding the creation of order. Order refers to any observable results of the social interplay, be it the price of milk, dinner etiquette, or the business models of organised crime. Within a reductionist framework such orders are assumed to be formed through certain identifiable and stable relations amongst autonomous variables, which together form the observable phenomenon. While these could be multiple and complicated, with enough careful research and analysis into their constituting parts the overall order could be scientifically explained. Of course, once explained in detail, this new knowledge could be used to prescribe actual policies. The rationale behind such policies is to manipulate some of the variables so as to influence and alter the observable order, thereby theoretically at least, directly contributing to effective action.

A systemic approach on the other hand, does not conflate order with constituting parts but rather with constituting dynamic processes. Relations between the different observable variables are not deemed as constant, but rather as dependent on all the simultaneous interactions amongst all the other variables. For example, raising government subsidies to milk farmers in relatively similar circumstances, could lead either to a rise, or to a fall in the price of cheese. All would depend on many other connected variables including the current state in the intensive competition amongst big supermarkets, levels of investment in global transportation technologies, as well as the new celebrity diet craze.

Each of these other observable orders are themselves shaped by a whole range of other trends and influences. In other words, the whole notion of “all else being equal” within social contexts is viewed by complexity theorists as an ontological impossibility. As John Urry suggests, it’s like “walking through a maze whose walls rearrange themselves as one walks through”[xvi]. What are perceived by Newtonian approaches as autonomous variables, are viewed by complexity approaches as further observable orders, similarly created through uncertain processes of interdependent interactions among many other variables, themselves observable orders, and so on, and so forth. We therefore can never deduce a universal description of the relations between all the variables that make up a certain order, and obviously cannot translate them into policy prescriptions.

2.3  The Operational Gap

Now this might be intellectually fascinating but the obvious question is – if we cannot say what the effect of a certain economic policy such as farming subsidy will do, what use is it? Ultimately, what does the complexity turn contribute to our ability to create effective new knowledge - what can complexity science ever do for us? In order to answer this question, we need to go back and consider some of the fundamentals of the scientific enquiry – the search for universal laws of nature, the sources of systemic change, and the nature of cause-effect relations.

As discussed earlier, the goal of Newtonian science was to uncover the general laws of nature that would help us model and predict the environment around us. To figure out these universal axioms one had to carefully unpack a given phenomenon, be it in the natural or the social world, research the relations between its individual variables, and put them back together again. The deduced laws, i.e. theories themselves were defined through their constituting variables or by their observable end states. For example, ‘information asymmetry theory’ explained how economic players with access to unequal information produced different market results; and the ‘democratic peace theory’ explained why democracies never fight one another.

Complexity science has not really diverged from the scientific aspiration to generally explain our environment, but has rather redefined the purpose of universal laws themselves. Instead of developing theories that explain set relations between autonomous variables or certain end-states, it developed general theories explaining how end-states come about in the first place. In so doing, it doesn’t so much as confront reductionist science as moves sideways to highlight a completely different perspective on nature and reality. In a nutshell[xvii], complexity suggests end-states are the product of three synergic processes – local interactions through agent networks; emergences of systemic structures; and an endless evolution of both local behaviours and structures. 

Human reality is created through individual human action. While seemingly obvious, this fact tends to be overlooked as so much of our attention revolves around linking different conditions, trends and forces. Economies only exist because individuals innovate, build, buy, and sell different products and services; cultures exist because people act, and react to the actions of others in certain repetitive ways; just as conflicts ultimately exist because people physically attack, impede, or prevent resources from reaching others. It is the everyday actions of billions of individuals across the world, not their mere thoughts or intentions, which continuously create the local and global realities we live in. Crucially though, these are not actions but interactions.

Humans have evolved as collaborative social creatures whose strategies for survival and reproduction requires them to interact and communicate with others. Our foraging ancestors have relied on team work to hunt, navigate, and care for their young, while the first agricultural settlements required the sharing of new information as specialisation in tool-making and farming techniques developed. Similarly today, people do not just make things; they give, take, trade and share them, while also depending on the resources and actions of others to create them in the first place. In other words, while reality is built on individual actions, these actions are taken within the settings of social networks.

Such networks are constructed through individual connections that overtime create vast webs of people, all connected directly and indirectly to one another. Some of complexity’s research has thus focused on exploring the behaviours of such networks, aiming to identify general laws of network expansion and evolution, a subject I will return to later on. Generally speaking though, from a complexity perspective, social life, like its biological counterpart, is networked based. Thus, at the heart all social phenomena lay the conceptual and physical manifestations of individual behaviour across networks. Through these webs of human interactions, individuals, groups, and whole societies disseminate ideas, knowledge, strategies and resources, thereby actively creating the world around them. Crucially though, these interactions do not take place within a vacuum, rather they are embedded in and channelled through a structured and dynamic social landscape.

This social landscape, full of valleys, hills, rivers and deserts, embodies the political, cultural, economic and technological patterns created by past network interactions. In other words, human exchanges themselves create the social environment within which further interactions can take place. While individuals always interact at a local level, their local contexts are based on their specific position within the network, and how the overall social landscape is uniquely manifested in their little neck of the woods. It is the unique typology of this environment that makes certain interactions, in certain locations within the network easy and efficient, while making other interactions in other locations inaccessible and costly.

The question is how does this landscape take shape? Here lies the core of the complexity conundrum – if the social landscape is created through human interaction how does it come about? And once created, how does it change?  How do the interactions among so many individual elements create what we recognise as the structures of the whole? And how do such structures shape the individual behaviours that drive them?  Complexity’s answer is – emergence and feedback.

Emergence describes the process through which new systemic properties, i.e. new orders, are created. While such properties only exist at the macro level, they emerge out of micro level behaviours of elements that do not necessarily exhibit those same properties. For example, a diamond might hold the macro property of extreme strength, however this property cannot be found in the carbon atoms that construct it. Similarly, a state of protracted conflict may not be reflected in the intentions and attitudes of the individual people fighting on both sides. While most of them might prefer peace over war, tragically, the conflict still endures. In other words, the property of the whole cannot be traced back to source, rather “it arises out of the combined agencies, but in a form which does not display the agents in action”.[xviii]

From a complexity perspective, all observable properties of human systems are emergent and self-organised in nature. From house prices to fashion trends, cultural norms to power structures, such properties only make sense when observed at the systemic or macro levels and cannot be deduced from observing the behaviour of the individual or micro levels that support it. However, the significance of emergence is not only down to how individual action is transformed into collective structures, but also to the manner in which these very structures come back to influence individual action, i.e. their systemic feedback effects.

Fashion trends for example, emerge out of the autonomous decisions by thousands of individuals to design, sell and wear certain shapes and colours. Take the recent spring - summer rage of culotte pants, a wide leg over the knee cut which has remarkably proven equally unflattering for all body types. Why did it become a systemic phenomenon? Overall, the world of fashion reflects the cultural, economic, and ideological zeitgeist of the time – this is the systemic landscape within which the fashion networks interact. However, predicting the rise and fall of specific trends is much more precarious, as the multi-million-dollar fashion forecasting industry would attest.

The new seasonal efforts by fashion creatives usually starts a year earlier with designers and scouters walking the streets and art galleries, or rambling through old markets in search for new inspiration. Out of the many different designs manufactured towards each season, ‘butterfly effect’ like circumstances will prompt certain shapes, colours and fabrics to gain attention over others. Once the adoption of a specific look reaches a certain threshold of fashionistas visibility, a fashion trend can be declared. Crucially though, once a new pattern emerges, other players will react and adapt to it. Other designers will bring out their own culottes versions, retailers will prominently stock and advertise them, and consumers will be culturally and economically incentivised to buy, and alas wear them. All these will further strengthen the initial systemic pattern in what is referred to as a “lock in” process, which for a while at least, like the extreme shoulder paddings of the eighties, will make complete sense to everyone.

A fashion trend, like all the other systemic orders, is not a fixed systemic state, but a self-sustaining pattern - a dynamic systemic interplay. Large numbers of individuals, in this case designers, retailers, fashion editors, and consumers, make certain individual choices. Out of all these accumulated actions certain systemic properties begin to emerge. These, can further incentivise and or constrain the next actions of those very individuals. The process is dynamic, ongoing, and uncontrollable by any individual players, be them the emerging winners or emerging losers of the new order.  

Here it is important to also differentiate between the inherent uncertainty as to which specific fashion trends will take off, and their high probability of relative correspondence with other existing conditions in the system – there is after all some method in the madness. Shapes, colours and fabrics might be uncertain but they are not random. Like evolutionary forces in nature, certain properties hold greater fitness with the existing landscape than others. So while it might be hard to predict if boyfriend or culottes shaped trousers will become the talk of the season, we can predict that Victorian corsets will not become a closet must have anytime soon. This is because the properties they possess are in vast contradiction to the culture and practicalities of women’s lives today. Moreover, the rise of one trend will prompt others that seem to complement it, thereby again providing clues to why certain shapes and colours prevail.

Thus, the uncertainty of specific emerging patterns is constrained within some emerging parameters. Complexity science refers to these as ‘attractor states’. These are spectrums of varieties held by some overarching emerging rationales that serve as gravitational pulls for the systemic interplay. They form a kind of mould into which alternative solutions adapt themselves[xix]. Yet even these underlying meta- structures are not permanent. Overtime, economic, political, cultural and technological shifts could gradually undermine these fundamental rationales, ultimately switching the gravitational pull from one spectrum to another. Hence, while Victorian corsets have no place in today’s world of fashion, it does not mean to say that within some future alternative social ecology they would not become all the rage again.      

Ultimately, both emerging systemic properties and attractor states are dynamic in nature, merely existing at different scales of the system. However, this still begs that question - what actually drives such macro transformations? Here a second key differentiation between Newtonian and Complexity thinking comes to the fore. As suggested earlier, Newtonian thinking assumed that for change to occur, some kind of intervention from outside the system is needed. However, complexity approaches suggest that changes are generated from within, in other words from within the interactions across the human networks that underlie them and the inevitable contradictions they create.

Complex systems are dynamic by nature. While we might be able to observe certain orders they are not at a state of equilibrium or rest, rather they are continuously being self-generated. Like the mesmerising shapes in water fountains, endlessly created through the pushing and pulling of countless water molecules, so are fashion trends, property prices, and power relations, continuously recreated everyday by the millions of actions that feed them. Ultimately, gradual shifts in individual actions will reach a critical mass at which the old order can no longer be sustained. At the same time, new observable structures will also begin to emerge.

But if individual action is shaped by the systemic order why would it change to begin with? Here the significance of connectivity, interdependency and the local context of human action comes back into play. While scientific research might choose to focus on explaining a specific social order – be it a fashion trend, a market price or a political belief, the boundaries of the system are methodologically rather than ontologically drawn. In other words, they emerge out of the scientist’s practical need to draw a line around the phenomenon she seeks to explore, rather than boundaries that can be objectively observed in reality. The problem is that complex systems are open systems. In the social realm this translates into a thick web of interwoven systemic interplays nested inside and across each other. Any individual action can therefor create multiple effects in our system, many of which tend to be unintentional and unpredictable.

Take for example a decision by Anna Wintour, Vogue’s legendary editor, to display a beautiful model wearing a fur coat on the cover. Such action would be part of the systemic interplay around the fur fashion trend – another macro order that has emerged over the past few years. However, this order is also part of several other macro formations including the economic globalisation of fur production; changes in the enforcement of poaching regulations; the rise of the new Asian middle classes; as well as the decline of previous fashion orders that have culturally delegitimised the use of fur in fashion, such as epitomized by the unforgettable “we’d rather go naked than wear fur” supermodel campaigns of the nineties. To complicate things further, Wintour’s daily actions are also part of many other processes of systemic emergence, including fashion industry structures, power struggles between elite fashion magazines and new fashion blogs; cultural trends in beauty and gender identities; and even political support for the Democrats for which she has been an ardent activist.

So one action by one individual can actually simultaneously impact many different systemic interplays. But as suggested earlier, individual actions themselves are incentivised and constrained by systemic formations, which means that they are also simultaneously influenced by all of those systemic dynamics. Crucially this makes them all connected and interdependent. Changes in any one of them will inevitably influence the rest. Such connectivity between both individual actions and multiple formations creates systemic effects that are completely separated from any intentional design. Wintour’s cover decision therefore influences and is influenced, consciously and unconsciously, intentionally and unintentionally, by a much wider systemic scope than we would have ever methodologically define. Moreover, all of its intended and unintended consequences will inevitably create new contradictions with the resonating outcomes of other people’s actions, as new opportunities are created for some, while pressures mount on others.  

For example, the trend to off shore production lines to China during the noughties, had significant influence on patterns of home burglaries in London. These seemingly unrelated macro orders, emerging from the actions of completely different sets of players are interconnected. One cascading effect explained how cheap manufacturing in China has completely undermined the average London burglar’s business model. While breaking into someone’s home is always a risky business, with the right expertise one could break in, grab a DVD player and sell it off for a good price in the stolen goods markets established during the nineties. However, Chinese manufacturing had significantly brought down the price of such goods, leading to the inevitable conclusion that “a DVD that costs 19.99 is just not worth stealing”[xx].

We can therefore better understand how London home burglary rates had significantly come down over the last decade. Unfortunately, this is not the end of the story, as further systemic shifts, for example in the technology and manufacturing of mobile devices, create other new patterns. The economic rationale for stealing phones and laptops off people on the street has gone up. This emerging macro order has different properties, the new business model requires acting in public and at times could involve physical violence. These could in turn affect many other economic, technological and political formations including public support for more intrusive public surveillance or the creation of new insurance markets. These, like all other systemic interplays will emerge out of the accumulated individual interaction amongst countless others, each reacting to new conditions created by these cascading effects in their own immediate local environment. Thus, the very actions taken by different players, across different fields of operation, will contribute to multiple emerging orders.        

From this perspective, Anna Wintour is but one node in a vast network of interconnected people that encompasses not only the fashion world but extends far beyond. In fact, the only natural boundary one could put around this system would be the world itself.  From a complexity science perspective, her active choice for the Vogue cover will ripple throughout the whole network, cascading to various levels of affect through a variety of cultural, economic, political and technological structures across the world. Now you might expect this to put unbearable pressures on her slender shoulders, but it is safe to assume that Wintour does not take all of these repercussions under daily consideration. This takes us back to the crucial role of context.

As suggested earlier, no one operates at the systemic level. Each player make their own individual choices based on conditions and circumstances deep inside their social network and within their local landscape. Bankers trading securitized mortgages did not operate with a global financial crisis in mind, just as politicians de-regulating the mortgage sector, or families buying homes at prices they could never afford. Each were operating within their own position in the network, reacting to the pressures and incentives that surround them. Bankers were solving a local network question – how could we take actions that would increase our organisation’s profits, given the long rise in the US property market, and specifically how could we show better results than our nearest competitors? Politicians were solving their own local question – how could we allow more people, and especially groups previously side-lined from the property ladder, to own their own homes, as well as win public support? While families were solving their own immediate question – what new opportunities can allow us to buy a home, as prices around us continue to rise and so many of our peers have already made the jump?  

The outcome of de-regulation, over-spending and spiral trading has ultimately caused systemic collapse. Yet no single player across the network, even if aware of the rising contradictions between emerging trends – and many were, had any real control over the systemic formation. Changing the emerging propensities would have required changing the individual behaviour of millions of individuals. While some change in actions by some players could have had real impact on the intensity or timing of the crisis, it could not have been averted all together.

Today’s problems are created by past solutions, just as new policies however seemingly sensible, will inevitably create new contradiction in our future social landscapes. Complexity approaches would thus argue for the need of continuous assessment not of the macro formations themselves but rather of the emerging tensions between macro formations. Off course these structural tensions should than be explored in terms of the individual behaviours that drive them and the immediate context that shapes them, if any effective action could ever be designed. 

Every action, whether taken by presidents or parking attendants, is a solution to a local problem as observed from the unique perspective of the person making it. Its form will be designed according to the individual personal understanding of their immediate network context, the constraints and opportunities they present. Ultimately, people trade, socialise, love or fight one another, not at a systemic but at a peer to peer level. Even seemingly systemic ambitions to change the world will always manifest through local actions, whether it’s drafting and voting for new laws, distributing money towards new causes, or vaccinating the world against epidemics. Implementation requires multitudes of human to human interactions. It is the unique “nano” local conditions that surround these very interactions that determine the actions taken, including the drivers, ambitions and cognitive attributes of human action.

At the “nano” local level, we are innovative creatures, constantly striving to better our personal material, emotional and physical wellbeing. Thus, as we try to find, or randomly stumble upon new means for answering our local needs, we change the system from within. Social systems are therefore autopoetic – they continuously change themselves from the inside. For a while, self-organisation will sustain the emerging order, but inevitably, as individuals continue to adapt and innovate within their personal and professional lives, the order will begin to unravel. The system will transform.

The universal laws of transformation produced by complexity science relate to this very process of unfolding, for example how human behaviour drives these bottom-up emergences, not their content or end-states. Our social fabric emerges through not one, but countless structures that are all interconnected. Most importantly, they are all in a continuous process of adapting to each other. As structures cascade across the social landscape they alter the local contexts of the individuals that support them, requiring them to adapt, take new actions, and thereby set off new dynamics. Thus cause and effect relations do exist – our social formations are not random, yet their relations are inherently dynamic, unpredictable, and in most cases indirectly mediated through many others systemic structures.

2.4  The Next Challenge

So to recap, a complexity lens to the social realm exposes the internal dynamics through which social realities emerge and transform. Social orders are dynamic, self-generating and interdependent. They transform through cascading events that alter the local and personal context in which individuals make decisions and take action, and it is these very actions that ultimately shape the systemic formations we all observe. So overall, while complexity science does not provide us with the predictive abilities theoretically associated with reductionist approaches, i.e. what will happen, it does provide us with a much deeper understanding of how things happen – how reality unfolds. Complexity theories thus help explain a vital dimension in the creation of any social phenomena.

Yet, the question still remains - is this non-predictive and non-prescriptive complexity-based understanding of social systems of any practical use? And if so, how does it inform new modes for action? One could claim complexity thinking is more descriptive than explanatory. Granted reductionist theories can only explain a very narrow spectrum of complicated rather than complex phenomena, however the knowledge they were able to scientifically substantiate have had enormous value for informing action.

As suggested, complexity approaches do not help in predicting or manipulating end-states but rather explain how such end-states come about. While the last decade has seen many scientific journals fill with mathematic models and simulations that have provided ever deepening understanding of how complex systems behave, by definition, they could never provide the treasury with policy guidelines as to what to do to about inflation. In this sense, we could ask whether the ‘complexity turn’ implies the de-coupling of science and action, at least in the social realm? 

Two inescapable conclusions seem to be apparent. The first is that complexity science does seem to provide a world view that is much more coherent with reality as we all experience it, whether in our professional or personal life. Once we start to make sense of any phenomena around us as a multi-dimensional and dynamic open structure which evolves through systemic feedbacks between individual interactions and emerging patterns, it is hard to go back to a reductionist mode of operation, even if the end product is seemingly much more intelligible.  Nevertheless, a second conclusion is that there is an obvious operational gap between the better understanding of social reality that complexity science provides, and it’s very weak support for designing action. If knowledge is usable information, from the decision-makers point of view reductionism still wins big time.  

In the ‘Systems view of life’, Capra and Luisi proclaimed that we are living through “a crisis of perception”, in which decision makers are simply not aware enough of the interdependencies within the problems facing them, preferring to analyse and address them in isolation[xxi]. However, this project suggests that the current crisis is not down to oblivious decision makers, most of whom are the first to realise and lament the growing interdependencies and uncertainties that constrain their actions. Rather the problem lies in the difficulty of finding new tools and methods that will allow for systemic action. In other words, this is an operational rather than a conceptual crisis. Overcoming this crisis, will take much more than the mere realisation of the complexity around us. We must innovate and develop new modes of operation for decision makers, whether sitting in government, business, or the wider civil society, to design and carry out more effective “complexity-friendly” policies.

It is here where complexity science’s contribution to action reaches its limit. Creating new frameworks will require us to develop new conceptual bridges between our ability to analyse the systemic interplays underlying a social phenomenon, and our ability to identify potential leverages for action within the unique local contexts of decision-makers. In other words, we need to develop new approaches for operating within complex environments, thereby taking a player’s perspective, rather than analysing them as if observing a petri dish – the traditional scientist perspective. Developing such conceptual bridges requires us to roam outside the academic sanctuary and explore other sources of experiential knowledge that could help us develop new operational frameworks for effective action in complex environments.

As human history has always been complex, this exploration need not restrict us to the here and now, rather much can be learned from historic examples in which societies both perceived reality in complex terms, and developed corresponding modes of operation. The next Building Block will explore what can be learned from one such example – the ancient Chinese strategic world-view and its corresponding operational framework of Acupuncture. 

[i] Daniel Garber, “Science and Certainty in Descartes” in Michael Hooker ed. Descartes, (Baltimore: John Hopkins University Press, 1978) Sited in Between Heaven and Earth, Loc 567.

[ii] For some of these fascinating tales see Bill Bryson’s amazing book – A Short History of Nearly Everything

[iii] Halili, The secret life of chaos, http://www.dailymotion.com/video/xpxj1b_the-secret-life-of-chaos_tech

[iv] Bill Brysen, A short History of Nearly Everything

[v] Open the social sciences – repot of the Gulbenkian Commission, p31

[vi] Snowden Chynfin framework

[vii] Weaver Warren, 1948, Science and Complexity, Classical Papers – Science and Complexity Vol 6 No 3, 2004 pp65-74

[viii] Halili, The secret life of chaos, http://www.dailymotion.com/video/xpxj1b_the-secret-life-of-chaos_tech

[ix] Ibid

[x] Ibid

[xi] Quoted in Freedman Lawrence, Strategy: a history, Kindle edition, loc 191.

[xii] Open the social sciences – repot of the Gulbenkian Commission, p27

[xiii] Ibid  p38

[xiv] John Urry – “the complexity turn”, Theory, Culture and Society, 2005, Vol 22:1.

[xv] Open the social sciences – repot of the Gulbenkian Commission

[xvi] John Urry – “the complexity turn”, Theory, Culture and Society, 2005, Vol 22:1. P3

[xvii] For further discussions of complexity theories and concepts see ‘complexity bits’

[xviii] Sawyer quoted in Morcol Goktu, A Complexity Theory for Public Policy, (Routledge: 2014)

[xix] Simon 1996

[xx] History of the Now, BBC

[xxi] Fritjof Capra and Pier Luigi Luisi, The System’s View of Life – a Unifying Vision, (Cambridge, Cambridge University Press: 2014)