Donella Meadows: Thinking in Systems (2008)

A Systems Library, Vol. 17

Donella Meadows’ Thinking in Systems is for many the first book on systems thinking they read. This is probably the only title in the field that deserves the status of bestseller. Yes, it’s a good book. It deserves to be read. Still, I am also going to be critical about it. Read on.

Donella Meadows died in 2001 at the age of 59. Thinking in Systems was posthumously published based on a draft manuscript that had been circulating for years. Diana Wright edited and restructured the material to make it publishable. And successfully so, as the voice that narrates this story is unmistakably Donella’s, as we know it from her other publications.

The book is positioned as a primer. “A simple book for and about a complex world.” It’s a gentle introduction to specifically the subfield of system dynamics. This body of knowledge helps us to understand the dynamics of a system. Asking ourselves why a given phenomenon (take climate change), exhibits an escalation behaviour, we might fruitfully turn to the conceptual apparatus of system dynamics to answer that question. Similarly, when we are curious about the reasons why an organisation stagnates and is unable to lift itself out of a rut. Whenever we wonder about a system’s behaviour over time, the system dynamics lens may help us to deepen our understanding of the underlying structures.

System dynamics is a modelling discipline. It relies on a mathematical formalism to translate a system’s properties into a dynamic trajectory. This book is not about mathematics, however, but addresses the underlying ideas. It also offers a number of practical thinking tools to approach knotty problems.

The book is laid out as follows. In the first chapter Meadows introduces the basic vocabulary: stocks, flows and feedback loops. Chapter 2 surveys different types of systems and the dynamic behaviour that is associated to them. A very brief third chapter foregrounds desirable systems properties: resilience, self-organisation and hierarchy. Chapter 4 (‘Why systems surprise us’) addresses the consequences of humans’ bounded rationality. The way our brain works happens to be maladapted to the complexity and non-linearity that is characteristic for many real world systems. This leads to errors of judgement on our part which, in interaction with these systems, leads to all kinds of undesirable effects. Chapter 5 inventorises a number of these system traps, conceptualised as system archetypes. These are simple causal loop models that help us to understand and articulate the mechanics behind these pernicious dynamics. In Chapter 6 the focus is on the famous Meadows Ladder, a hierarchical framework of leverage points that offers a powerful heuristic to reflect on ways to intervene in a system to help it to function better. Alternatively the Ladder can also be used to diagnose why a system doesn’t work as well as it should be. Finally, the book closes with a short chapter of systems wisdom to help us to navigate ‘a world of systems’.

Thinking in Systems offers a lot of bang for the buck. It acquaints the reader with the basic vocabulary and offers a powerful toolbox. The final chapter is worth rereading time and again.

And yet one also has to be critical about this book. My misgivings are prompted by two elements, one internal to the book and one related to its effects in the outside world.

As a primer, it’s perfectly fine that a book skids over some of the finer points of the theory. But my feeling is that this informality hides a rather annoying conceptual incoherence.

Epistemologically the book oscillates between a naive realism (there is complexity out there and we can model it more or less faithfully) and a muddled constructivism (we can’t really know what is out there but models are a useful construct to structure our interaction with the world with the aim to progressively learn about how to deal with the friction and problems in that world).
Then the book is anchored in the normative perspective of an engineer who is interested in the dynamics of depletion of natural resources. It is inevitable that this professional perspective engenders a very distinctive (but always disputable) way of evaluating systems behavior. It is equally inevitable that it reflects a rather obvious position on the political spectrum.
Finally the lack of conceptual clarity extends to some of the pivotal notions in the book. It remains, for instance, unclear how desirable systems behavior, resilience and self-organization are conceptually linked. Also, readers may be surprised by the progressively narrowing focus to how social systems may suffer from actors’ bounded rationality.

My other qualms have to do with the way this book is at the root of some of the ‘systems traps’ that it wants to help defeat. Many people will read the book in the conviction that this is more or less what there is to say about systems thinking. Meadows gives her readers no reason to believe otherwise. For her the scope of systems thinking seems to coincide with system dynamics.

In actual fact this scope is quite narrow. There is vastly more to be said about ‘systems thinking and doing’ than the MIT-centered school of system dynamics leads us to believe. By omitting references to other, ‘competing’ (or complementary) approaches the book puts the bar for aspiring learners rather low, leading to a premature sense of gratification of readers’ curiosity for systemic insights (the ‘eroding goals’ trap). Furthermore and related to the previous point, if readers’ attention is the stock that authors and publishers are competing for, then the net effect of this book’s ever-increasing popularity is that it crowds out other, contrasting accounts of systems thinking (the ’success to the successful’ trap). In other words, the success of the MIT brand of system dynamics is obscuring other valuable contributions in the systems domain to such an extent that it is becoming a liability rather than an asset (considered against the background of potential gains in intellectual capital and system performance that could be realised by other systems approaches).

I would like to persuade Meadows’ readers to dig a bit deeper and go beyond what this book has to offer. Systems thinking is anything but a neatly structured, systematic field. It’s rather an archipelago of different schools and approaches, at times based on radically different philosophical conceptions. Feel free to browse the other titles in this Systems Library to acquaint yourself with this unbelievably rich intellectual legacy of the twentieth century, with deep roots in the wisdom traditions across cultures.

Thanks for reading! This short piece helps you to find your way in my expanding series of Medium writings.