My previous editorial was about the differences in views between Plato and Aristotle, and how this difference is still of importance today. Plato thought that the objects of true knowledge inhabit another world – an abstract realm independent of time and space, accessible only to the intellect. He thought that the world we experience was a shallow, misleading façade, behind which lay a more important, more fundamental world of perfect forms and souls. But Aristotle thought that there was only one world that we could philosophize about – the world we inhabit and on which we can perform experiments. He was dismissive of Plato’s Ideal Forms. He did not believe they exist. Where Plato believed that truth could be reached only through reason, Aristotle believed that by combining experiments with rational analysis and mathematics, we could reveal universal truths. Aristotle is clearly a founding thinker of the scientific approach. As a scientist, clearly, all my sympathies were with Aristotle and his approach. I hope to gain knowledge and use it successfully.
In real life then, the ‘sensible thing’ is to work out what the probable consequences of a possible action might be, based on previous experiments, and then decide on the one you think is ‘best’. Over many years I have been involved in building models on the basis that they might help the people involved to understand what factors are important for their situation, and what are the dimensions of potential interest resulting from a possible action or decision. This did help us (the modellers) greatly in surviving financially long enough to develop complex systems modelling and to learn about various chunks of ‘real’ – urban development and transportation, fisheries, economic development and planning, organizational decisions, innovations and many others. Eventually it has led to ‘agent based modelling’ where the different agents and types of agent involved in an issue have individual characteristics, motivations and beliefs, and the models can explore how such a system may change over time, and potentially how the constituent individuals may change as well. This clearly links the rational external effects of the agents in interaction to their own inner layers of values and needs, which may be hidden and might cause surprises.
The first success of the early complex systems models showed how the variables might, over the medium term, self-organize into different patterns either spatially or structurally, which could radically alter the performance of the system as a whole. This was due to the non-linearities in the interactions between individual elements. The models could explore the permanent dialogue between the existing structure and the local fluctuations and deviations from average, and thus can permanently ‘test’ the stability of the existing structures. This was captured in early ‘complex systems’ models, which could generate different possible trajectories into the future, with different emergent structures and organizations, giving rise to new regimes of operation. So such models allowed the exploration of potential emergent structures, for different historical deviations from average. In this way, a range of possible futures could be imagined for cities as different possible hierarchies might emerge. Similarly, within cities, neighborhoods could take on different characteristics and activities. Fishing fleets could adopt different spatial strategies of ‘learning’ how to locate new fish stocks, and economic organizations could evolve creatively their production systems and organizational structures. In some ways the models revealed something of the possible medium term reorganizations that might occur. Such changes would not have been predicted by ‘traditional’ systems models, whose spatial or organizational structure remained constant. All this was potentially very useful in clarifying what might happen over the medium term.
Looking back now I can see that, for the longer term, there remained a problem. All of these systems evolved qualitatively (new variables and mechanisms) over time, and for many of them the arrival of new technologies, new issues and problems, new possible solutions gradually overtook the models. The early urban models talked of white collar and blue collar workers, for example, and obviously the actual nature of the work involved in particular sectors changed radically. Such models therefore started off close to reality, but over time were increasingly describing changes in terms of the wrong variables and mechanisms. Evolution both in human systems and in ecosystems more generally, will evolve and change both the inputs, the variables and the outputs that are relevant. Models should be seen therefore as exploring the effects of current rationalities, allowing for deviations from average and local fluctuations that can lead to different possible collective structures and patterns emerging. Notably, though, they do not anticipate changes in the rationalities and internal beliefs of the individuals represented in the models, and the identities, values, ethics and morals of the participants. We find that our models represent a kind of scientific ‘Aristotelian process’ sitting on a ‘Platonic’ set of impenetrable inner beliefs and views which may change in unpredictable ways over time. In the end we are forced to use models to test whether the variables and mechanisms within them are still true. The real ‘use’ of such models is to monitor the system and detect when something is no longer changing as forecast – and to force us to seek out what the reason might be. We can then ‘re-normalize’ our models and ‘bring them up to date’ with new, unexpected realities.
It is important to realize that we can never have models that ‘predict’ correctly the future over the long term. This is because, if the agents within a model ever believed in its predictions, then some of them at least would change their behavior in order to ‘profit’ from the predictions – thus invalidating the model and its predictions! This is the reality of ‘reflexivity’ which tells us that the model, if believed, becomes part of reality itself, and affects what will actually happen. In 1987 George Soros wrote a remarkable book entitled “The Alchemy of Finance”, which made this point. But perhaps being written by a financier and not by an academic it never got the academic recognition it deserved. However, it turns out that these ideas are not just true for finance, but really are true for social systems in general. Models of economic markets, urban development and of flows of goods, services and information will all display this alchemy, whereby the ‘predicted outcome’ may affect the behavior of some individuals and thereby invalidate the assumptions of the model and its predictions. These assumptions are couched in terms of behavior that reflects the ‘preferences’ of the individuals or organizations concerned. These are most often reflected in economic terms so that prices or cost/benefits are used to drive the model. With this, and the well-known ‘It’s the economy stupid’, we had moved to the idea that it is really only economics that matters. And that ‘learning’ was all about rational economic arguments.
And then came the BREXIT (the UK referendum on staying or leaving the EU) vote. In that economics is seen as the most important basis of politics and government, it was assumed that by presenting the economic calculations about what would happen with or without BREXIT, then people would vote to remain in the EU. Despite the overwhelming evidence given by every professional organization concerned with the economics of the UK there was nevertheless a majority for leaving. This shows that people are not necessarily swayed by reason and reasonable arguments, even about economics. David Hume in the early 1700s said “Reason is, and ought only to be the slave of the passions”. In other words, it is not ‘reason’ that dictates what we love and what we hate, but rather that we use ‘reason’ to better pursue our loves and hates. The feeling of some people was to simply to give the government a good kicking, and they calculated correctly that a vote to leave the EU would be a great protest.
Analysis of the voting showed that less well educated, older people tended to vote ‘leave’, and younger, educated people tended to vote to ‘stay’. The point being that the older, less educated people used the referendum to vent their feelings about successive UK governments that had allowed the gap between rich and poor to grow, had allowed in lots of immigrants, from both inside and outside the EU, who competed for jobs, and who also had seen the impact of ‘free trade’ international competition on manufacturing and many ordinary jobs, leading to increased pressure on their wages and their jobs. In addition, the UK has always had a rather strong nationalist streak that is massaged by the constant focus of a savage press on celebrities, and on the wonders of the UK and its amazing history. This has meant that English people on the whole do not speak foreign languages, are not very concerned by events outside the UK and simply revel in being British. All this has meant that instead of participating in European Democracy and its Parliament, the few people that bothered to vote in European elections, sent candidates like Nigel Farage there, and watched in amusement as he tried to cause mayhem and to disrupt proceedings.
This reveals how the rational/irrational decision to leave the EU was based on internal feelings and emotions about identity that were not based on rational economic assessments. Of course, the reality of the ‘leave’ vote will only be felt over time – several years at least and will depend on the kind of deal that we can manage with the EU. It is totally premature to suggest that things are okay because the stock markets have not crashed. Right now, through the currency devaluation of 10% that followed the vote, the people of the UK ‘lost’ – £183 Billion which dwarfs the (false) figures quoted by the ‘leave’ campaign for what we ‘paid’ into the EU. All this goes to show how the hidden, internal thoughts and identities are crucially important in what they do. These are shaped by history and how it is depicted, by individual histories, local events and how these are portrayed to them. Clearly, although young people were in favor of staying in the EU, older people still were not. The overhanging memories and heroic part played by the UK in successive World Wars clearly still made many older people see Europe as more a source of problems than of solutions. The young had already shifted their perspective to a more positive view of Europe.
Anyway, the point of this Editorial is not to argue for or against BREXIT, but to show that ‘reason’ is not as important as we think. Science can show many things through repeatable experiments, but in social systems it is not clear which experiments will be repeatable, and how to ‘capture’ the inner beast. The idea of representing human behavior by that of Homo Economicus or Rational Man is clearly erroneous when one thinks how people’s decisions will reflect their desired ‘image’, their hormones and their particular self-delusions. In a way the whole operation has highlighted the dangers of a government by plebiscite, as different individuals can use the occasion for a multitude of different purposes. Particularly when the actual consequences are not clear as nobody has detailed the different trajectories that might occur. It demonstrates the importance of a parliamentary ‘shield’ through which ideas and their possible consequences are debated thoroughly. It does seem that a yes/no referendum for such a complex question was really not a sensible idea.
Complexity shows us that in reality, evolved systems have multiple levels, which co-evolve, and hence which combine internal structures and beliefs with external effects and issues. In essence Emergence: Complexity and Organization is all about this amazing, creative evolutionary system which is the universe. This means that over time the actual elements and structures of a system will change – the current dictionary will have to evolve in pursuit of reality. As the first chapter of the Dao de Jing says “That of which you can speak is not the reality – The Way that can be walked is not the eternal Way”.
This shows us that what constitutes ‘reason’ itself will evolve and change, and any model will need to be revised over time. In Chapter 4 of the Dao de Jing we find – “It is the underdetermined nature of the world that makes it, like a bottomless goblet, inexhaustibly capacious. The Way is empty, yet inexhaustible, like an abyss!” This phrase is absolutely marvelous and uses (it seems) some Chinese equivalent of ‘underdetermined’!!! This means that it understands that although there is some ‘causality’ in the world, it is not all-powerful, and so some things will happen which were not explicit in what went before. The world, although often running along in an unsurprising way, is also free to do something unexpected and creative. Within a temporary framework of structure and organization, new elements, characteristics and functionalities can emerge! In other words, an inexhaustible evolution will occur.
In many ways complexity allows us to bring science to this ancient wisdom, and instead of proving it false, to see now we may recognize and enhance its brilliant thoughts. No model that we can build will be a fixed ‘reality’, we shall need to change the dictionary (of variables and mechanisms) over time. And for this we need to make sure that we look for changes occurring inside its structure and elements as well as within its complex evolved organized networks that constitutes its current structure. The example of BREXIT merely goes to show us that inner dimensions may well throw surprises in the way of seemingly clear rational conclusions. To misquote Hume “Rationality is not only the slave of the passions, but the passions are to an extent the slave of history”.