Dynamics in Action: Intentional Behavior as a Complex System

Rating:
Summary: "A radically new perspective on causation"
Review: Choice (November 00) calls Dynamics in Action (together with Judea Pearl's Causality, which Choice reviews together with Pearl) "a radically new perspective on causation and the explanation of human behavior. Juarrero proposes a new framework for explaining human action. She probes deeply into the springs of human action to explicate the secret link between mental intention and physical behavior. She begins with a critique of Aristotle's legacy in attempts to understand human agency, arguing that modern philosophy has largely lost the insight of his distinction of four causes countenancicing only mechanistic efficient causes, while perpetuating Aristotle's erroneous principle that nothing moves itself. Juarrero's work is a paradigm of the integration of philosophical analysis with neuropsychological research, evolutionary theory, complex systems theory, and the physics of nonlinear systems. Causes of human action appear as dynamic constraints on complex adaptive systems. She draws implications for the practical understanding of human freedom and responsibility, even proposing bridges between the literary world and science... Both works are highly ambitious in rejecting traditional views. Both are written clearly and enthusiastically... Juarrero's and Pearl's books will greatly interest philosophers and scientists who are concerned with causality and the explanation of human behavior."

Rating:
Summary: What is all the fuss about?
Review: I wrote a formal review of this book for the June issue of _Philosophical Psychology_....

Basically, I think this book is a good idea, but poorly executed. Juarrero makes an interesting conncetion between problems in action theory, the branch of philosophy having to do with human action and its place in the world, and information theory. And it is an interesting project to solve some of these traditional problems using modern neuroscience and dynamical systems theory. So I laud the attempt.

But no matter how interesting the project, a book has to sink or swim with the details, and Juarrero gets many of them wrong. She misinterprets Donald Davidson's theory of actions as causes, uses mathematical terms such as bifurcation in non-standard ways, and gets the laws of thermodynamics plain wrong. The casual reader may be impressed with her expansive technical vocabulary, but ultimately it detracts from the interesting ideas in the book. Read through the first 200 pages, and you'll realize that the most contentious issues in dynamical systems theory are not even discussed; indeed, Juarrero takes too much of the science for granted.

And so, while it's an interesting topic, the book could have used a lot more research and done a lot more "connecting the dots" for readers. If you're after some other books on causality, I would suggest instead Judea Pearl's new book _Causality_. If you are interested in dynamical systems theory, I would read the later works of Andy Clark (such as _Being There_) or some of the papers published by Chris Eliasmith (available on the web).

Hope this helps :) --BNT

Rating:
Summary: "A radically new perspective on causation"
Review: Philosophers are always in danger of mistaking an obsolete scientific model for what they consider to be reality. Nowhere has this been more of a problem than in the philosophy of action and intention, which has relied on a billiard ball notion of causation inherited from Newtonian mechanics. In this remarkable book, Alicia Juarrero applies the new scientific paradigm of complexity theory to these issues. In doing so, she creates a philosophy of action and intention that actually works with complex biological systems and with the intricacies of human motivation; in other words, a philosophy that finally confirms and supports what has been learned about human beings by psychologists and psychoanalysts. "Dynamics in Action" is another very successful demonstration of the general applicability of complexity theory to the understanding of structure, process and organization everywhere. It is a must read for specialists in these fields and for anyone interested in the revolutionary changes taking place in the contemporary world view.

Rating:
Summary: A Review of Dynamics In Action
Review: While this book is worthwhile reading for cognitive scientists it is too speculative and fails to support most of its claims with technical detail. The only "detailed" example of a "dynamic system" referred to by Juarrero is a neural network with feedback and experiments with just those systems have failed to scale to human levels of performance (see Judd, Neural Network Design and the Complexity of Learning). Many cognitive scientists hope that this problem can be overcome with a modular network design. This is not considered by Juarrero. I reject Juarrero's claim that constraints constitute a new sort of cause. In fact actual experiments with constraint-based reasoning and with neural networks are performed on von Neumann computers composed of AND, OR, and NOT gates, that is relays/transistors. When looked at on a fine scale of granularity constraints are composed of many small causes/forces acting in the conventional way. In fact, I would argue that granularity and abstraction are the real issues Juarrero should have been addressing.

Rating:
Summary: A Review of Dynamics In Action
Review: While this book is worthwhile reading for cognitive scientists it is too speculative and fails to support most of its claims with technical detail. The only "detailed" example of a "dynamic system" referred to by Juarrero is a neural network with feedback and experiments with just those systems have failed to scale to human levels of performance (see Judd, Neural Network Design and the Complexity of Learning). Many cognitive scientists hope that this problem can be overcome with a modular network design. This is not considered by Juarrero. I reject Juarrero's claim that constraints constitute a new sort of cause. In fact actual experiments with constraint-based reasoning and with neural networks are performed on von Neumann computers composed of AND, OR, and NOT gates, that is relays/transistors. When looked at on a fine scale of granularity constraints are composed of many small causes/forces acting in the conventional way. In fact, I would argue that granularity and abstraction are the real issues Juarrero should have been addressing.