Folded or interwoven
A review of Mensen zijn complex (People are complex), Paul van Geert's Dutch-language introduction to the theory of complex dynamic systems for psychological and educational practice.
Paul van Geert (2026). Mensen zijn complex: Een inleiding in de theorie van complexe dynamische systemen voor de psychologische en pedagogische praktijk. Groningen: University of Groningen Press. Open access: doi.org/10.21827/6a438e064459b
A crumpled piece of paper can be smoothed out again. A knitted jumper can also be unravelled, but what remains is a long thread, and nothing in that thread shows how it once formed a jumper. Paul van Geert opens his new book with this image. It points to the origins of two words we often use interchangeably: complicated comes from folding together, complex from weaving together. What has been folded can be unfolded; what has been interwoven loses something essential when it is taken apart.
It may seem a linguistic distinction, but it has consequences for the way we look at problems in mental health care. Much of our diagnostics, research methodology and protocols implicitly assumes that problems are folded: that they can be broken down into separate factors which can be measured and treated one by one. Over almost four hundred pages, Van Geert develops the idea that people are interwoven rather than folded, and explores what this means for observing, measuring, treating and evaluating.
A life's work, translated into practice
Van Geert, emeritus professor of developmental psychology in Groningen, was among the first to develop dynamic systems theory within psychology. This book is not a collection of earlier work but a carefully considered introduction for practising psychologists and educationalists. The first part (chapters 1 to 5) lays the foundations: a short history of complexity thinking, a discussion of what he calls the "standard approach", and the question of when something can be understood as a (coupled) dynamic system. The second part (chapters 6 to 10) deals with the core concepts: emergence, self-organisation, attractors and, finally, a model of causality for practice.
Throughout the book we follow a fictional family, the De Vries family from the village of Norg: Anne, the father, who is depressed; Baukje, the mother, who is worried; their son Casper, for whom the question of ADHD has been raised; and their daughter Dieuwke, who has dyslexia. The family gives the book coherence and makes abstract concepts recognisable. From chapter 3 onwards, each chapter closes with "the key points for practice", which also lend themselves well to teaching. The mathematics remains modest: simple equations, such as the logistic growth model, mostly placed in boxes that the reader can skip without loss.
Reification: what a disorder is not
Conceptually, we found chapter 2 the most valuable. There Van Geert looks for the philosophical roots of our practice: an Aristotelian way of thinking in categories that are supposed to capture the essence of a case, the Cartesian legacy of internal representations, and what he calls reification. We say that Casper "has" ADHD, as if ADHD were a thing located inside him. Our language invites this, just as we say that "it" is raining without there being an "it" that does anything.
This way of speaking has consequences. Reification can hide the personal, process-like character of a problem from view. It can reverse cause and effect, so that ADHD becomes the explanation of the behaviour rather than the name for a pattern that sustains itself. And it easily turns differences between people into random variation around a single core or, when that does not fit, into "comorbidity". Those who work in diagnostics will recognise this, and it is a pleasure to see it argued so carefully.
Why the average is not about this client
Chapter 4 contains one of the clearest Dutch-language explanations of non-ergodicity that we know of. Whether one tosses a hundred coins at once or one coin a hundred times, the average is the same: that is an ergodic process. With families visiting an amusement park, matters are different; the average of all visitors says little about the choices of a single family. Van Geert makes the translation to the clinic himself: the association between insomnia and low mood measured across five hundred people is something other than that association across five hundred days in the life of Mr De Vries. Depression, in his view, is non-ergodic, as are most psychological variables.
A second useful warning is that of "the Crazy Ant": a series of random steps that average out at zero, yet can drift a long way and then appear to form a trend. When reading a routine outcome monitoring (ROM) curve or a series of diary scores, it is worth bearing in mind that an apparent turning point may also arise by chance.
Resistance as stability
Clinically, we found the way Van Geert reformulates resistance, secondary gain and relapse particularly fruitful. A system actively maintains its own stability. On this reading, resistance to change is not a property of the client but an expression of the way a system reproduces itself. He puts it as follows: "A depression is a system that produces depression" (p. 87). A person who is depressed unintentionally shapes their surroundings so that the depression is confirmed, for instance by avoiding contact or by hearing selectively, and sometimes even because being in therapy underlines the reality of the depression.
In doing so he broadens the network perspective in a way that is still uncommon in the literature on symptom networks. His network includes not only symptoms but also the reactions of family members, secondary gain, the therapy and the diagnosis itself. Drawing on Ian Hacking, he shows that a diagnosis, once made and shared, becomes part of the network it describes: sometimes liberating, sometimes perpetuating. For those who make diagnoses, it is a thought worth considering.
Hollows, hills and relapse
Van Geert proposes the attractor landscape, the image of a ball in a landscape of hollows and hills, as an overarching working model for practice. Chapter 9 is well structured from a didactic point of view. The depth of a hollow stands for the resistance to change that can be expected: the deeper the hollow, the more sustained the effort needed to get the ball out of it. The width stands for the number of starting situations that eventually end up in the same hollow. As long as there is only one hollow, the ball rolls back after every disturbance. An intervention only takes hold once a new, desired hollow emerges that maintains itself.
In chapter 10 he develops this into a simple model of relapse. When an intervention acts directly on the target variable, such as practising a realistic estimate of danger in anxiety, the effect gradually fades once it ends. When the intervention increases the carrying capacity of the system, for instance through self-reflection or social support, the effect is better maintained. On this reading, relapse is not primarily a failing of client or therapist but often "an intrinsic feature of the dynamics" (p. 363). Van Geert therefore argues for ending a treatment not as soon as the symptom score falls below a threshold, but only when the new pattern appears sufficiently anchored. We find this argument persuasive.
Implications for clinical practice
Van Geert summarises his approach in a single principle: "diagnostics describes dynamics, intervention perturbs self-sustaining patterns, evaluation follows reorganisation over time" (p. 154). From this follow proposals that are relevant to mental health care: a diagnosis as an idiographic dynamic profile alongside or instead of a label; treating variability as information, and sometimes as a therapeutic goal, rather than merely as noise; seeing the therapist as part of the system rather than as an external administrator of an intervention; idiographic system models drawn up together with the client; and person-centred process monitoring as a complement to routine outcome monitoring at a few measurement points.
Readers who follow the conversations with the De Vries family will also recognise a reading of psychotherapy through Vygotsky's zone of proximal development. A therapist who moves too quickly or leaves little room for autonomy can make the collaboration less effective, and therapist and client may, without noticing, hold each other in a less favourable pattern.
Critical notes
A book with a clear argument also invites critical notes. The "standard approach" is portrayed rather uniformly, whereas case conceptualisation, functional analysis and systemic therapy have been part of clinical practice for some time. Van Geert himself remarks that his philosophical summary is necessarily brief. The educational centre of gravity can also be felt: arithmetic teaching and the school playground receive a good deal of attention, mental health care somewhat less.
More importantly, some of the practical advice goes further than the current evidence. Many expectations are derived from models, and Van Geert acknowledges that their application to psychotherapy still largely awaits empirical investigation. Advice such as "do not aim directly for symptom reduction" will not always be compatible with clinical guidelines, or with situations in which symptom reduction should take priority. There is also a certain tension in his appreciation of HiTOP: he is critical of latent variables and population models, yet aligns himself with a dimensional model that was itself derived at population level. The proposal to use the area of a radar chart as a measure of complexity is one we would like to see better grounded, not least because that area depends on the order of the axes.
From the perspective of our platform, one more thing stands out. The book makes clear why process monitoring is meaningful, but says less about how: which variables, at what frequency, and how to discuss a time series together with the client. For the precursors of change, Van Geert draws on the literature on early warning signals, and he formulates this with appropriate caution. The synergetic tradition of Haken and Schiepek is discussed with respect, although critical fluctuations and the Synergetic Navigation System receive only brief attention. Finally, there are some editorial imperfections, such as typing errors, duplicated section numbers and references to sections that do not exist. In an open access publication these can fortunately be corrected with little effort.
Who is this book for?
The book seems to us valuable for clinicians, diagnosticians and trainers who have noticed that a client does not fit neatly into a category, that a treatment unfolds differently from what the protocol would lead one to expect, or that a relapse could not be predicted. For the Dutch postgraduate training programmes in clinical psychology (the GZ and KP programmes), it offers a Dutch-language foundation that has been lacking until now, with practice points that lend themselves well to teaching. For students and researchers it provides an accessible way into an international field of research. It does not offer concrete treatment protocols; above all, it invites a different way of looking.
In closing
Chapter 5 contains a sentence that has stayed with us: "This is not relapse, this is exploration" (p. 221). Whether that holds in a particular case will have to be established anew each time, but the question itself opens a different conversation with the client. Mensen zijn complex is a personal and committed book by someone who has worked on this subject for a lifetime. Because it has been published open access, it is available to everyone. For teams and training programmes wishing to explore this way of thinking in more depth, it seems to us a valuable starting point.