[Back to Reason, Science and Faith]

Introduction

At its most basic, an emergent property is a property of any system which is not present in the components of that system.

Beyond this, things start to get complicated.  People define terms - such as 'strong emergence' and 'weak emergence' - in various ways, generally with the intention of using the definition to support whatever position they are arguing for at the time.

I would like to suggest that this basic description of an emergent property is not particularly interesting or useful.  A roof may leak, or not, but leaking is not a property of any of the atoms which the roof is comprised of, so leaking is an emergent property.  But I suspect that it is only occasionally helpful to think about it in these terms: identifying such emergent properties is trivial, and adds nothing to your understanding of the system.

I would like to suggest a more useful definition.

• An emergent property as a property which is implicitly present in the components, but which can only be observed when it emerges in the larger system.

Once you see the emergent property, you can understand (possibly after some study) how the interaction of the components produces it.  The reactions studied by Chemists are emergent properties of the relevant elements and their physical properties: given what they are, they have to interact as they do; it is convenient to describe what happens in terms of Chemistry, but it is not essential - the chemical reactions are completely determined by the physical properties.

We also see emergent properties in living organisms: a termite mound is incredibly complex, but it is formed purely by the activity and interactions of simple termites; it looks planned, but none of the termites have a plan.  Shoals of fish and flocks of birds behave in complex ways, but they can be understood as the result of fairly simple behaviour on the part of the individual fish or birds.

Emergent Properties - Useful or Confusing

A property of a system which is implicitly present in the components, which naturally emerges from the properties and interactions of the system's components, can reasonably be described as an emergent property.  Understanding how the system behaves leads you to a deeper and clearer understanding of the components, and a better understanding of the components can potentially lead you to discover unexpected properties of the system.

When used this way, emergent properties can assist us in understanding the world around us.  Understanding the world, and the connections between its various parts, is useful.

When not used this way, emergent properties can be used to confuse, distract and obscure rather than enlighten.

Or, to put it another way, when you take a term which refers to something useful, something good and helpful, and you apply it to something quite different, this is usually because you want to make it sound like they are basically the same.

When we do not understand how a property of the system naturally emerges from the properties of the components, you explain nothing by calling it an 'emergent property' - you are simply giving a name to something you don't understand.  This is the scientific version of the 'God of the gaps' which scientists (and others) rightfully criticise.  You can wave your hands all you like, but giving it a name is not the same as understanding it.

Of course, something may be an emergent property, but we don't yet understand the connection.  You may believe it to be an emergent property, but science is not supposed to rely on what people believe.

To give two obvious examples of this problem: people sometimes claim that the mind is an emergent property of the brain, or that intelligence is an emergent property of matter, and think they have explained something.  This is not science; it's the scientific equivalent of the Wizard of Oz telling Dorothy, "Pay no attention to that man behind the curtain!"  The only reason to claim something as an emergent property, when you don't understand how it emerges, is to hide the fact that you don't understand.

I can't say how often this trick is used to confuse people, but it is a standard part of the scientific toolkit.  To take just one example, we are told that emergent properties "are often unpredictable and cannot be deduced solely from the characteristics of the individual parts."  As I say, this is not science.

In short, when something is described as an emergent property, take a careful look: is this a useful connection which aids understanding, or is someone trying to pull the wool over your eyes?