Methodological Stuff:
1. Introduction
2: Patterns
3: Patterns, Objectivity and Truth
4: Patterns and Processes
It has been a year since I defended my Ph. D research "Into Complexity", and since then I have been invited to give guest lectures on my pet subject, which has forced me to bring it down to its bare essentials. As the responses of people have been quite positive, I have decided to share this "Quick Guide to Complexity Thinking" with the community, hoping that it may help some researchers in their own work. So here is the first of a series on complexity thinking, hope you enjoy!
The "Pattern-oriented Approach to Complexity" (PAC) hinges on two concepts, namely 'complexity' and 'patterns'. The theme of complexity has become quite popular in recent years, and usually has connotations with 'chaos theory', 'non-linearity' and the likes, but the stance that 'complexity thinkers' take, is a bit different. Complexity thinkers do not really form a school or so, but rather must be seen as a voice that individual researchers in various scientific disciplines are raising. most of them are coming to the conclusion that the scientific (and cognitive) tools that they have at their disposal are insufficient to address the complexity of the themes they take interest in, and are starting to reconsider the foundations of science itself. Ecologist Robert Ulanowicz ("A Third Window: Natural Life beyond Newton and Darwin") , quantum physicist Marcelo Gleiser (A Tear at the Edge of Creation: A Radical New Vision for Life in an Imperfect Universe) and sociologist Edgar Morin (On Complexity (Advances in Systems Theory, Complexity, and the Human Sciences) are only a few of a motley crew of scientist worldwide, who are starting to think that science's traditional quest for order may need to be updated to a version 2.0 (as this site suggests), in order to face the challenges of a new era. more specifically, these voice think that the role of ambiguity, indeterminacy, uncertainty, chance and other concepts that defy the traditional view of order need serious attention, and may actually be tremendously important to address age-old questions, such as what life, intelligence and consciousness is. I will return to these questions at a later stage, but for now the main point that I want to make, is that these researchers believe, in their own words, that we actually don't understand how we can deal with these concepts. At best we can approximate ambguity or uncertainty with statistical and probabilisitc tools, but that this doesn't mean that we can equate them. A simple thought experiment may make this more clear!
Suppose we ask ourself the age-old question how life was formed. Currently, most researchers who want to find an answer to this question will, implicitly or explicitly take a reductionistic approach by taking a living cell, tearing it apart and then looking what bits and pieces they can find inside. This is a reasonable approach in itself, and maybe it is the only way that we can pursue this subject, but to date it hasn't been a very successful one.
Now take an alternative hypothesis. Suppose life began with a chance event, where certain chemicals and other resources blended together under certain external conditions (say, bolts of lightning) which resulted in something that has no resemblance to current cells, but already carried a form of self-organisation that kept this blob together. Over time, this blob provided the stability and conditions for the formation of new chemicals, which eventually allowed the blob to divide itself. From then on, the blob evolved to host chemicals of ever increasing complexity -loosing others in the process- until something emerged that resembles cells as we currently know them. If this hypothesis is correct, we can draw a number of conclusions:
With this we also can see a distinction between complexity thinking and complex systems researchers (or complexity science), such as the complexity advocates of the Santa Fe Institute who did great deal of popularising complexity in the past thirty years. Many of these complexity scientists take a strong positivist stance to science, and think that by adding non-linearity, chaos , fractals and so on to the models they devise, will bring us closer to completing the 'book of science' that will allow us to describe anything. This recent book on complexity is exemplary for this positivist stance: Simply Complexity: A Clear Guide to Complexity Theory
Complexity thinkers, on the other hand, take a more or less epistemological stance to complexity, and think that a theme is complex when an observer is unable to completely understand it. If we then assume that our Universe contains no observers who know everything of everything at any given time, then complexity becomes a universal problem: complexity thinking in this sense is the final departure of a God in the language of science, while many positivists still hope that science can (eventually) take His place.
Some may think that complexity thinking may therefore take a postmodern stance, but I would strongly argue against this, at least with regard to the exponents of postmodernity who take an extremely relativistic view on the world, and consider science to be one socially constructed narrative amongst many others. The denial of absolute truths does not equate with the absence of truth, and some narratives prove to be -depending on certain (socially constructed) criteria- to be stronger than alternative interpretations, for instance with respect to predictability, usefulness, aesthetics, or simply popularity. Complexity thinking in that respect is post-postmodern; I have found it useful to realise that complexity almost always deals with the question how something is organised or formed, and this includes the way that we observe our world and organise our observations in order to make (mental) models of our lifeworld. These observations are not random, but are influenced by our life story, our environment, our capabilities and our limitations. And yes, also our culture and processes of socialisation. This implies that our narratives are not arbritrary, but are formed by meaningful selections, events and experiences. Complexity thinking then (ought to) ask(s) the question why some narratives are more persistent than others, and I would think that 'power relations' are not the sole reason for such differences. With this, and the realisation that these models and the modelling activities are also complex activities, one can only conclude that complexity must be able to describe its own metaphysics in a true self-referential fashion!
This opens a box of Pandora, when one aims to develop a methodology to investigate complex themes, for basically the methodology itself constrains the means one has to understand that theme. I will not go into this particular issue here, but this is one of the reasons why my thesis became so volumnous...
As opposed to complexity positivists, there have been many researchers who use complexity as a sort of bumper; at some point they say "Things are very complex" and that's it! With this, complexity runs the risk of becoming a hollow term. Here we face the two boundaries in which complexity thinking must operate; on one side there is (positivist) risk of not being able to live up to the expectations that are made, and on the other there is a lethargic sense of being overwhelmed. But complexity itself suggests 'order in chaos' and complexity thinking therefore balances between these antagonists. To put this more concretely, this means that a methodology of complexity aims to assist a researcher in discovering how order is formed in a contingent embedding. This means that such a methodology aims to find robust truths, theories, models, phenomena and what not. And this is where a pattern-orientation comes in. PAC is basically a collection (a pattern library) of robust processes that are capable of withstanding the contingencies related to complexity. I will discuss patterns the next time.
1. Introduction
2: Patterns
3: Patterns, Objectivity and Truth
4: Patterns and Processes
The Pattern Library:
It has been a year since I defended my Ph. D research "Into Complexity", and since then I have been invited to give guest lectures on my pet subject, which has forced me to bring it down to its bare essentials. As the responses of people have been quite positive, I have decided to share this "Quick Guide to Complexity Thinking" with the community, hoping that it may help some researchers in their own work. So here is the first of a series on complexity thinking, hope you enjoy!
The "Pattern-oriented Approach to Complexity" (PAC) hinges on two concepts, namely 'complexity' and 'patterns'. The theme of complexity has become quite popular in recent years, and usually has connotations with 'chaos theory', 'non-linearity' and the likes, but the stance that 'complexity thinkers' take, is a bit different. Complexity thinkers do not really form a school or so, but rather must be seen as a voice that individual researchers in various scientific disciplines are raising. most of them are coming to the conclusion that the scientific (and cognitive) tools that they have at their disposal are insufficient to address the complexity of the themes they take interest in, and are starting to reconsider the foundations of science itself. Ecologist Robert Ulanowicz ("A Third Window: Natural Life beyond Newton and Darwin") , quantum physicist Marcelo Gleiser (A Tear at the Edge of Creation: A Radical New Vision for Life in an Imperfect Universe) and sociologist Edgar Morin (On Complexity (Advances in Systems Theory, Complexity, and the Human Sciences) are only a few of a motley crew of scientist worldwide, who are starting to think that science's traditional quest for order may need to be updated to a version 2.0 (as this site suggests), in order to face the challenges of a new era. more specifically, these voice think that the role of ambiguity, indeterminacy, uncertainty, chance and other concepts that defy the traditional view of order need serious attention, and may actually be tremendously important to address age-old questions, such as what life, intelligence and consciousness is. I will return to these questions at a later stage, but for now the main point that I want to make, is that these researchers believe, in their own words, that we actually don't understand how we can deal with these concepts. At best we can approximate ambguity or uncertainty with statistical and probabilisitc tools, but that this doesn't mean that we can equate them. A simple thought experiment may make this more clear!
Suppose we ask ourself the age-old question how life was formed. Currently, most researchers who want to find an answer to this question will, implicitly or explicitly take a reductionistic approach by taking a living cell, tearing it apart and then looking what bits and pieces they can find inside. This is a reasonable approach in itself, and maybe it is the only way that we can pursue this subject, but to date it hasn't been a very successful one.
Now take an alternative hypothesis. Suppose life began with a chance event, where certain chemicals and other resources blended together under certain external conditions (say, bolts of lightning) which resulted in something that has no resemblance to current cells, but already carried a form of self-organisation that kept this blob together. Over time, this blob provided the stability and conditions for the formation of new chemicals, which eventually allowed the blob to divide itself. From then on, the blob evolved to host chemicals of ever increasing complexity -loosing others in the process- until something emerged that resembles cells as we currently know them. If this hypothesis is correct, we can draw a number of conclusions:
- Life is not a matter of assembling current chemical compounds. Rather, we must see life as being 'passed on' from previous generations which, under favourable conditions, can kick-start a self-organising process that can also reproduce itself (what Maturana&Varela have called autopoiesis, or self-creation). This is exactly what we see happening: life is 'passed on' to next generations.
- The evolution of life passes a number of chance events that 'fly under the radar of statistics'. Ulanowicz speaks of 'rogue chance' and Jeff Wayne, in his musical version of 'War of the Worlds': "The chances of anything coming from Mars is a miilion to one...but still...they come".
- We will not find the answer to the question of life by disecting cells.
With this we also can see a distinction between complexity thinking and complex systems researchers (or complexity science), such as the complexity advocates of the Santa Fe Institute who did great deal of popularising complexity in the past thirty years. Many of these complexity scientists take a strong positivist stance to science, and think that by adding non-linearity, chaos , fractals and so on to the models they devise, will bring us closer to completing the 'book of science' that will allow us to describe anything. This recent book on complexity is exemplary for this positivist stance: Simply Complexity: A Clear Guide to Complexity Theory
Complexity thinkers, on the other hand, take a more or less epistemological stance to complexity, and think that a theme is complex when an observer is unable to completely understand it. If we then assume that our Universe contains no observers who know everything of everything at any given time, then complexity becomes a universal problem: complexity thinking in this sense is the final departure of a God in the language of science, while many positivists still hope that science can (eventually) take His place.
Some may think that complexity thinking may therefore take a postmodern stance, but I would strongly argue against this, at least with regard to the exponents of postmodernity who take an extremely relativistic view on the world, and consider science to be one socially constructed narrative amongst many others. The denial of absolute truths does not equate with the absence of truth, and some narratives prove to be -depending on certain (socially constructed) criteria- to be stronger than alternative interpretations, for instance with respect to predictability, usefulness, aesthetics, or simply popularity. Complexity thinking in that respect is post-postmodern; I have found it useful to realise that complexity almost always deals with the question how something is organised or formed, and this includes the way that we observe our world and organise our observations in order to make (mental) models of our lifeworld. These observations are not random, but are influenced by our life story, our environment, our capabilities and our limitations. And yes, also our culture and processes of socialisation. This implies that our narratives are not arbritrary, but are formed by meaningful selections, events and experiences. Complexity thinking then (ought to) ask(s) the question why some narratives are more persistent than others, and I would think that 'power relations' are not the sole reason for such differences. With this, and the realisation that these models and the modelling activities are also complex activities, one can only conclude that complexity must be able to describe its own metaphysics in a true self-referential fashion!
This opens a box of Pandora, when one aims to develop a methodology to investigate complex themes, for basically the methodology itself constrains the means one has to understand that theme. I will not go into this particular issue here, but this is one of the reasons why my thesis became so volumnous...
As opposed to complexity positivists, there have been many researchers who use complexity as a sort of bumper; at some point they say "Things are very complex" and that's it! With this, complexity runs the risk of becoming a hollow term. Here we face the two boundaries in which complexity thinking must operate; on one side there is (positivist) risk of not being able to live up to the expectations that are made, and on the other there is a lethargic sense of being overwhelmed. But complexity itself suggests 'order in chaos' and complexity thinking therefore balances between these antagonists. To put this more concretely, this means that a methodology of complexity aims to assist a researcher in discovering how order is formed in a contingent embedding. This means that such a methodology aims to find robust truths, theories, models, phenomena and what not. And this is where a pattern-orientation comes in. PAC is basically a collection (a pattern library) of robust processes that are capable of withstanding the contingencies related to complexity. I will discuss patterns the next time.
In the first, there is no reason to presume that all processes are necessarily serial. It is entirely possible that multiple processes occur (or develop) in parallel and at some later point those processes themselves become consolidated to form a more complex organization.
In other words, something like the evolution of the cell might depend more on the ability to "collect" processes, rather than requiring that they all develop within a specific context.
In the second, it isn't necessarily valid to consider that all processes develop towards a positive objective. An example is how neurons connect in the brain. While these cells wander during development looking for their connection partner, the complexity of this model rapidly becomes overwhelming when it is considered how difficult it is to locate such a "partner". However, the problem becomes eminently simpler when we consider that the neuron doesn't "look" for its "partner", but actually wanders by avoiding those that aren't its "partner". In other words, the complexity that we assumed existed was based on looking for a proactive, positive direction of development, instead of recognizing that the actual process was avoiding bad connections.