This is a much expanded and updated version of an earlier post two years ago - I have now also added a resource section at the bottom, with commentaries on useful books. I've been doing this work in preparation for the CPD weekend “Working at the Edge of Chaos” on 29 & 30 April 2017 in Oxford.
As a therapist, have you ever struggled valiantly, but hopelessly against the entrenched resistance of a client who says they are desperate to change, but are actually digging their heels in at every turn, reacting with “yes, but …” to each and every suggestion?
These kinds of situations can often leave you with the sense that you are much more invested in their self-actualisation than they are, and that your therapeutic efforts on their behalf are somewhat like Sisyphos pushing the boulder uphill.
Here, a crucial feature of human psychology and a central conundrum of therapy become tangible: plainly, just as much as one part of the client’s psyche is indeed desperate to change, another part of their psyche seems equally invested in the status quo, in maintaining their habits and established comfort zones. It’s not uncommon for therapists to get seriously locked into this frustrating battle with what is effectively the client's internal conflict - and whilst we can argue that it's unwise for a therapist to identify principally with just one side of the clients conflict, what is the alternative when therapy is apparently committed to helping the client deliver their stated and contracted goals?
Is it any wonder that therapists who have sincerely engaged with the client’s agenda for therapy and been thwarted by the ‘resistance’ throw up their hands in despair and conclude that therapeutic effectiveness is not much to do with the therapist, but essentially dependent on the client's motivation?
This may remind us of well-known joke:“How many therapists does it to change a light bulb?” “Only one, but the light bulb really must want to change!”
Of course, once the client is seriously motivated, therapy becomes an eminently do-able job. Unfortunately, the percentage of clients who are unambiguously motivated is so small, there aren't enough of them to go round to make us all feel effective as therapists. My supervision experience suggests that few therapists can afford to refuse the kind of client who is - already in the initial interview - displaying clear signs of that internal deadlock. And how often do you then say to such clients: "I'm sorry, I cannot take you on - it seems to me you are not motivated enough, and it's going to be ineffective and frustrating for both of us. So let's not bother until you've managed to drum up some more motivation." Which also begs the question: what kind of pre-therapeutic procedure could possibly ever get the client ready for ‘motivated’ therapy? And who would deliver that pre-therapy and how would that work to generate client motivation?
Instead of requiring clients to make our job easier by turning into the right kind of lightbulb, we might become interested in non-linear change.
“What is that?” I hear you ask. In which case presumably you didn’t know that more than one kind of change was on the menu? Well, it depends what universe your restaurant is situated in. If you happen to be entertained in a Newtonian universe, all you get - for breakfast, lunch and dinner - is linear change. Chances are that is precisely the kind of change which the client was both demanding and resisting in the first place.
Linear change in a Newtonian universe
Linear change is what most of our culture - and received therapeutic wisdom – is oriented towards: “I want to go from pain to pleasure, from negative emotion to positive happiness, from pathological dysfunction to well-adapted fulfilment, from existential lostness to self-actualisation - please, in a straight line, ideally the shortest and cheapest distance!” (by-passing the less promising but more achievable route which leads, as Freud cautioned, from neurotic misery to common unhappiness).
Overcoming negative patterns and counteracting negative states is the quite legitimate agenda which many clients bring to therapy – what they don’t realise is that they have already brought an assumption of linear change as part of the package. Unfortunately (for them and for us), their objectives for therapy are formulated within a framework that already presumes and only computes linear change. Unfortunately, their defensive ego (who is doing their best trying to manage the various pressures arising between the reality principle and inner coherence, but has now cracked sufficiently to try therapy and is willing to pay for it) has been educated in a Western tradition in which effortful linear change is the best you can ask and hope for: the whole edifice runs on dualism down the ages in a straight line from Aristotelian logic to Newton and Descartes to CBT.
Because we are culturally steeped in a Newtonian mindset (where every action has a reaction, where for every force there is a counterforce, and the bigger force wins), clients and therapists operate in a universe where everything remains static unless we bring some force to bear. It is assumed that every change has to overcome some inherent inertia and out-manouevre some dastardly resistance. Change is therefore supposedly hard, and requires discipline and consumes energy.
It is a power-over model, fitting to the age of Empire - the late 19th century when psychotherapy as we know it was initially conceived. The zeitgeist at Freud’s time construed a universe of random deterministic atoms where the bigger billiard ball pushes the smaller one. So we are entitled to wonder whether this a case of the well-known aphorism: “we do not see reality as it is, but the way we are”? Is the universe really organised on Victorian Empire principles, or is that a socially constructed projection into reality, through the lens of a particular point in history?
The limitations of a Newtonian paradigm
Chaos and complexity theory historically emerged out of the shortcomings of the Newtonian paradigm. It’s not so much that the Newtonian paradigm is wrong, it is just that it is quite partial: it doesn’t apply to everything under the sun (even though the paths of the planets revolving around it can be calculated fairly accurately based on Newton’s laws of gravity, which in itself is quite a feat; but even there the limitations become evident quite quickly: just introduce an additional moon and you have the unpredictable chaos of the three-body problem).
For all Newton’s undiminishable genius, there are two problems with his equations: they work pretty well in simple, inanimate controlled systems, with just a few variables. They do not work well in complex, self-organising systems, with many separate but interrelated and connected agents. And they do not work well in evolving, biological living systems, let alone human systems, let alone in psychology and intersubjectivity.
Therefore, linear change via Newton’s equations doesn’t work well in human relationships, especially helping and therapeutic relationships.
That’s one of the problems we are bound to encounter when we extrapolate from the physics of inanimate matter to the biology and psychology of human beings: it is easy to forget that these sciences operate by different epistemologies and different laws (which is the key point that postmodern social scientists and philosophers had to struggle to get recognised over the last 70-odd years, e.g. see the work of Jürgen Habermas). As human beings we are not just dealing with supposedly objective facts, but with subjectivities and interpretations.
However, to this day, the logic of physics is being imported into the logic of psyche, ignoring the fundamental point that psycho-logic is different from mental or mathematical logic, and not reducible to it (unless you turn the human being into an abstract idea first). The metaphors of technology – first mechanics and later of computation - shaped the thinking about psychology from Freud’s hydraulic analogies to modern neuroscience.
Non-linear change in dynamic, evolving, complex systems
Complexity theory, and its early precursor chaos theory, precipitated the recognition and slowly growing understanding of non-linear change – change which applies to dynamic, evolving, complex systems. They operate in a universe of impermanence, in a web of self-organising systems.
Contrary to our notion of chaos, these sciences define chaos not as random disorder, but as a deeper, non-linear order - a good image for this are fractals: complex, organic images which are not linear and repetitive, but each fractal does have its own inherent order (a fact which our right brain can visually recognise in an instant, as what’s called self-similarity).
The most famous catchphrase for the paradigm difference between linear change in a Newtonian universe and non-linear change in chaos is the ‘butterfly effect’: the idea that a butterfly flapping its wings at one end of the planet can be implicated in a hurricane at the other.
Whilst this is a catchy image, it is often misunderstood precisely through established linear cause-and-effect assumptions: if we can only track the more subtle and complex non-linear chain of cause and effect, then we might be able to control it. If only we can understand non-linear change within the terms and parameters we are accustomed to in handling linear change, we will be able to control the predicted outcome, avert catastrophes and make even the most resistant client change in a way that their defensive ego idealistically demands.
But that is to misunderstand the radical departure and sacrifice of our Newtonian position which non-linear change invites us into. When it comes to our role in the grand scheme of things, in contrast to the omniscient and omnipotent Newtonian position, chaos theory only offers much less grand and reassuring possibilities for influence: if we can divest ourselves of the ideas of control and predictability and get immersed in uncertainty and interested in the changes that are already always happening, we might be able to involve ourselves in the process and establish the possibility for a more humble contribution. This applies on a global, planetary level, and it also applies in the consulting room.
An established metaphor amongst organisational consultants who draw on complexity theory is the comparison between paddling a boat in a straight line across the pond versus white water rafting. That’s a good image for the paradigm clash between linear and non-linear change.
So what’s the conclusion these consultants draw if we consider the turbulence of the relationships (in an organisation or in our consulting room) as white water rafting? As a facilitator or therapist, I don’t entirely give up the responsibility of involvement in the face of overwhelming complexity; but neither do I fancy myself as in control of the whole process.
In contrast to the celebrity status of the ‘butterfly effect’, the less emotive formulation in chaos theory is ‘dependence on small variations in initial conditions’. In complex systems, a tiny difference in the initial state can lead quite quickly to completely different outcomes (the connection with hurricanes is not entirely fictitious, as this principle was discovered in computer simulations of global meteorology – see James Gleick’s “Chaos”, a very readable introduction to the development of chaos theory, where he introduces the various scientists and disciplines who originated the idea of non-linear change).
Therefore, the rate of change in complex systems is not proportional to the apparent forces exerted (because much bigger forces are always already present in some kind of implicate dynamic equilibrium). Rather than slow and incremental, if these dynamic forces evolve towards a tipping point, change can be sudden and radical, when many apparently innocuous variables conspire to unleash those otherwise balanced energies. A negative feedback loop in a system can escalate quite quickly into unexpected proportions.
Here we are in a non-Newtonian universe of impermanence, where the one certain thing is uncertainty, where things are always already in flux, and dynamic equilibrium is as much stability as we’re going to get - temporarily.
If things are on the move already, it only needs a tiny straw to break the camel’s back, to make the barrel overflow, for the system to go into a chaotic, disorganised state from where it will settle into a new equilibrium.
Non-linear chaotic change between different equilibrium states
Complexity theorists studying many different kinds of systems use the term ‘attractor’ to convey the recognition that each and every system can be attracted to and organise itself into a variety of radically different equilibrium states (because systems evolve through self-organisation, such equilibrium is always dynamic rather than homeostatic - a significant difference to earlier, biologically rooted conceptions of equilibrium, including Freud's).
An apparently stable system can reach a tipping point where the established structure breaks down – we might think of the client’s characterological defences as such an apparently stable system. Intensifying relationships or life crises may have conspired to challenge the established structure of the client’s personality and sense of identity. As in any system that tips into a disorganised chaotic state, this may engender a creative void where everything is up for grabs. In therapy, the breakdown of habitual patterns and identifications may engender a moment of neuro-plasticity.
But we want to be careful not to over-idealise the destructiveness inherent in this process: from the inside, it probably feels unstable, scary and possibly catastrophic: regressive states of abandonment and terror, or spiritual emergencies as the dark night of the soul may be evoked. A chaotic system far from equilibrium is liable to be experienced as overwhelming the resources of the ego, which may crack into psychosis or surrender into transformation. Jung used to say that every monster to the Ego is a God(dess) to the Self, but we only jovially recognise that with hindsight once we have survived the process.
Whether a chaotic state is a good thing or not depends on subtle variations in context and environment. In therapy, that context is significantly constituted by the therapeutic relationship itself. Whether a chaotic state leads to a destructive enactment or re-traumatisation (which destroys the therapeutic container itself), or whether it leads to a transformative enactment which engenders both a deeper sense of self and a deeper connection, is not something that can be guaranteed with certainty or taught via a manual.
In therapy, the outcome of such chaotic processes largely depends on how held the therapist feels ‘held’ within themselves, within their therapeutic position and their professional context (i.e. their own therapy, their supervision, their community of practitioners). That, in turn, is reflected in how attached the therapist is to an exclusively Newtonian universe, or how gracefully they are able to surrender beyond it.
The conundrum of desirable therapeutic outcomes in non-linear change
For our therapeutic purposes and from a human perspective, some equilibrium states seem more desirable than others - we might distinguish between more regressive versus more progressive processes. However, in line with the overall complexity perspective, these notions regarding the outcomes of chaotic re-organisation are fraught with our own habitual assumptions and need to be applied with caution. As therapists we tend to be – understandably - invested in our clients moving towards more integrated and wholesome equilibrium states (by whatever parameters of wholesomeness we conceptualise these).
But we need to be mindful, of course, that such investments on our part may become the very straw that tips the system more towards regression - for many of us as clients, our therapist’s investment in our supposed health replicates parental injunctions which we did not feel seen and met by. Therefore, any sense I might communicate to my client, that I fancy myself as knowing better than they do just what their healthy equilibrium needs to look like, may actually become a counterproductive imposition which propels them in the opposite direction.
I am using this as an example to illustrate the subtle systemic awareness which is engendered by inhabiting the complexity perspective within therapy. Whatever ideas and concepts we formulate and apply via such a perspective, I want to remain mindful that all my internal processes – all my assumptions, thoughts, feelings - are part of the system, feeding back into it, and that I am not operating on it from outside. A complexity perspective is not immune against fuelling destructive enactments, even whilst philosophising about them.
One such trap - which my formulation so far lends itself to - is, of course, to set up a dichotomy between the supposedly outdated dualistic Newtonian paradigm on the one hand, and the supposedly superior non-linear complexity paradigm on the other.
However, this would be to misconstrue and oversimplify the relationship between these paradigms: in the same way as straight lines are only a subcategory of all possible lines, linear change is only a specific instance of a much wider realm of non-linear change. That does not mean straight lines do not exist, or that we ignore the reality of Newtonian battles, socially, inter-personally or intra-psychically. What we want to do is to give dualistic reality its home within its wider, more complex, systemic, dynamic context in (relational) space as well as in time.
How are chaos and complexity theory helpful to therapists?
Owing to its origins in the late 19th century, and its further development throughout the 20th, counselling and psychotherapy are pervaded by dualistic Newtonian assumptions of linear change. Some therapeutic approaches and traditions may be more wedded to such dualisms than others, with some of the later humanistic developments clearly championing holistic and non-linear intuitions.
However, any deep understanding of the psyche and the therapeutic process tends to lead us beyond the dualistic trap. We can find non-linear intuitions in just about every depth-psychological approach. Across the various traditions, these intuitions have been enhanced in recent years by explicit applications of complexity principles (a selection and summary of the most useful and accessible books can be found in the references below).
But there is a limit to how far theory, reading and thinking can take us in this realm: when faced with the client’s defensive ego, its insistence on linear change and its addiction to dualism, many approaches and many therapists struggle to consequently translate their non-linear intuitions into their actual practice. So a crucial question in practice, addressed by none of the books that I have read on the subject, is: how can we engage and collaborate with the client’s ego in pursuing linear change, whilst maintaining an awareness of the paradoxes and complexities of non-linear change?
Too often the admittedly profound differences between linear and non-linear change are turned into another dualism (which rather defeats the object). So in experimenting with complexity the question that has been rather exercising me over the years (with very nitty-gritty implications in practice), is: how do we avoid falling into the trap of fighting dualism with more dualism?
Chaos implies ‘embodiment’ and bodymind process
One crucial limitation, of course, that traditional therapeutic approaches have in engaging with non-linearity is their bias towards language and the reflective mind. The talking therapies immediately run into a glass ceiling when it comes to the spontaneity of emergent processes. This notion - one of the most basic and immediately helpful distinctions which we derive from complexity theory: between 'established structures' and 'emergent processes' - is not exactly the same as top-down versus bottom-up change, but in many situations it overlaps: often emergent processes announce themselves first through somatic experience: spontaneous sensations, gestures or impulses can thus be the harbingers of an as yet unknown future self.
Following the therapeutic process at this level of attention to bodymind and systemic micro-detail, both internally and interpersonally, requires a therapeutic presence that is equally fluid and solid: anchored and stable as well as nimble and mercurial. We then recognise that on pre-reflexive levels of the interaction in the therapeutic relationship, the attachment – and the working alliance – is indeed a shifting, oscillating complex dance - there are many butterflies flapping their wings all the time, and it needs our own differentiated embodiment and flesh-and-blood presence to notice and pursue them. Left-brain reflection – as important as it is in the therapeutic position – usually happens after the event. In this territory, timing, responsiveness and spontaneity are crucial – therefore, learning about therapy at the edge of chaos cannot happen via a manual, not even a video: you need to be present and embodied in the room, engaging with your own emergent processes and the group around you, and participate.
Nick Totton and Michael Soth are offering a CPD weekend in Oxford on 29 & 30 April 2017, entitled “Working at the Edge of Chaos”. Some places are still available.
They write: “We will turn the fact that complexity cannot be learnt or taught in a book, but needs your whole bodymind, left- and right-brain to be present to ‘get’ complexity, into a feature of the weekend: just as we do not have control over the process in therapy, we cannot and will not set a curriculum for this weekend, and you will become co-responsible for the unfolding of your own and the group’s learning process. The weekend is an opportunity to dance at your own growing edge as a person and a therapist, to deepen your own idiosyncratic therapeutic style and find your own way to inhabit the paradox of risk and stability.”
Resources on Chaos and Complexity Theory
If you want to follow up and get engrossed in the subject in more detail, here are some of the resources you can find, each with a brief commentary:
The most engaging and for psychotherapists probably best and most accessible book (that is also most likely to translate into practice), is:
By Terry Marks-Tarlow (2002) Routledge
If you don't want to buy the book, many of Terry’s writings can be found - freely available - on her website:
Terry uses images and poetry to make the topic come alive, not like some books on complexity that emphasise scientific language or even maths (although the maths is very interesting).
In the field of relational psychoanalysis, probably the foremost writer thinking in terms of field theory and complexity is Donnel Stern. He has now published several books, but I recommend this one:
By Donnel Stern (2015) Routledge
The other psychoanalyst who is well-known for his long-term interest in complexity since it first got going decades ago is Robert M. Galatzer-Levy. You can find his paper: "The edge of chaos: A nonlinear view of psychoanalytic technique" online (it's an accessible introduction to complexity from a psychoanalytic point of view), but he also published a book very recently (which I haven't read yet, so cannot comment):
By Robert M. Galatzer-Levy (2017) Routledge
Although it is not dedicated exactly to complexity, but to field theory (but many of the ideas overlap, and you can see that Donnel Stern mentioned above is involved), there is now the International Field Theory Association, that offers a blog (published mainly by Montana Katz) and some materials, establishing a rationale and frame for applying these theories to psychoanalysis.
Another fairly recent book I found, but don't know, is:
by William J. Coburn (2014) Routledge
My very first reading around chaos and complexity, probably in the early 90s was this book, which has stood the test of time (there is now an anniversary version of it):
James Gleick (1987) London: Abacus.
He is a science journalist who has a knack for making these subjects accessible to the lay person, in an engaging and personal style. The book is a whistlestop tour introducing the originators of chaos theory both as people and scientists, whilst getting the essential ideas across (and their paradigm-shifting significance).
For many reasons I read the following book when it first came out (I was a fan of Deleuze and Guattari in my early 20s and at the time I was trying to turn their 'desire machines' into experiential workshops around the body and its impulses), however all I can remember is feeling vaguely disappointed (and thus not remembering much about it):
By Joseph Dodds (2012) Routledge
Chaos and complexity theory have had more significant influence in the fields of organisations, leadership and management than in therapy, so you can find plenty of materials in that realm, and I include a few below. However, the question always is: what's the depth-psychological and facilitative presence and skill of the people applying these theories, in whatever human context? In my opinion, we need more therapists to work towards an integration of theory and intersubjective practice in this regard (otherwise these very promising and paradigm-shifting ideas never quite seem to have the promised effects).
For those of us involved with groups and organisations, Ralph Stacey (from the Business School of the University of Hertfordshire) has been one of the most influential people in the UK applying complexity to businesses and management, by catching onto complexity very early on and seeing its relevance for social organisations (I can remember being on a conference with him in 1996, when his approach was considered fresh and radical, but already well-established):
by Ralph D. Stacey (2014) Routledge
Some of Stacey’s colleagues are providing materials online. One of his early collaborators, Patricia Shaw (who was also involved with Metanoia at some point), wrote a book many years ago which became quite influential:
By Patricia Shaw (2002) Routledge
In the realm of social organisations, the most well-known complexity celebrity is Meg Wheatley, a management consultant from the US who has turned more into a spiritual teacher now (as you can see on our website):
Somebody gave me her book as a birthday present years ago, and it makes the case for complexity in quite a poetic way - it's more of a coffee table book, but very engaging:
by Margaret J. Wheatley and Myron Kellner-Rogers (2012)
IARPP is announcing the publication of a new book coming out soon by an Italian relational psychoanalyst:
Gabriele Lenti (Italy)
And if you search on the Internet, I'm sure you can find lots more writing, doctoral dissertations and the like, for example:
From Theory to Clinical Practice: Psychoanalytic Complexity Theory and the Lived Experience of Complexity. In: International Journal of Psychoanalytic Self Psychology Volume 11, 2016 - Issue 4. By Margy Sperry, The Institute of Contemporary Psychoanalysis, Los Angeles, CA
 I'm pretty confident you must have come across it, as it is such a common mechanism and dynamic - it was the first 'game' identified by Eric Berne in what eventually became his first book on Transactional Analysis "Games People Play"
 after all, as Fritz Perls said: the underdog always wins
 "Contemporary research indicates that ‘client factors’ are the principal drivers of therapeutic change: e.g. client engagement, participation, hope." – see Mick Cooper's video based on his 2008 book ‘Essential Research Findings in Counselling and Psychotherapy: The Facts are Friendly’, where he quotes Mike Lambert's research with the conclusion that about 70% of what happens in therapy is to do with client factors. Although unarguably true that the quality of the therapeutic relationship is co-created and therefore significantly influenced by the client, this has always seemed a bit of an evasive and disingenuous argument to me - is it not the therapist’s responsibility to gauge the client's lack of engagement by reflecting on their own presence and input by wondering about the unconscious dynamics and implicit ‘therapist factors’ that are putting the client off? It's unfortunate that most outcome research doesn't even consider, let alone factor in unconscious dynamics (understandably in one way because it would really complicate matters - how do you design a questionnaire about somebody's unconscious?). But that then leaves a huge paradigm gulf between research and the (unscientific) principles that most of us need to use as therapists in order to make therapy work in practice every day.
 “An example of a simple linear system that exhibits non-linear feedback effect is the classic ‘three body problem’ of gravitation. Consider a moon orbiting a planet. The path that the moon takes is well known - it was fully described by Sir Isaac Newton's mathematical laws of gravity. But suppose we introduce a second moon of the same size as the first. Would the moons' orbits now be only slightly more difficult to calculate? It turns out that the simple deterministic equations which govern the three-body system are unsolvable. They cannot predict the long-term path of the orbiting moons. The reason why the three-body problem cannot be solved is that gravity is a non-linear force and in a three-body system each body exerts its force on the other two. This produces non-linear feedback and results in chaotic motion of the moon orbits.” from: Introducing Chaos: A Graphic Guide by Ziauddin Sardar, Iwona Abrams p.22
 which hasn't stopped swathes of the psychological field propagating linear change throughout the 20th century, come hell or high water (another little chaotic inconvenience). But a large section of the psychological therapies remains unmoved by these little details, which is why I don’t see much evidence that as a profession we have quite arrived in the 21st century.
 although, admittedly, in nature they are few and far between, as linearity is mainly at home in the abstractions of the human mind (where arguably it has its uses): the discovery that complex curves could be approximated by calculating large numbers of tiny straight lines via differential equations has contributed to the vast technological progress over the last few hundred years. However, has it got us closer to reality? "Clouds are not spheres, mountains are not cones, coastlines are not circles, and bark is not smooth, nor does lightning travel in a straight line."(Mandelbrot, 1983) – Mandelbrot was one of the originators of chaos theory - see his coastline paradox
 partly because this can become just another enactment of a power struggle regarding the terms of engagement and the construction of the therapeutic space
 or also through spontaneous embodied imagery, as suggested by Arnie Mindell and Process-oriented Psychology which has given us many useful theoretical and practical tools for working at the edge of chaos