Let's assume that a physical property is emergent, in the sense that it cannot be reduced to a function of the properties of its components. Can such a property be artificially recreated, or is that a contradiction?

We can look at the problem the other way: If there exists a reliable method that allows us to reproduce the emergent property every time, then that method is the bridge law that allows us to describe this property in terms of properties of its individual components. So the existence of a method for artificially recreating the emergent property makes the property reducible by definition, and an artificially re-creatable emergent property is a contradiction in terms.

Consider the text book candidate for an emergent phenomenon, consciousness. If emergence holds, then consciousness could never be reduced to the computational properties of individual neurons.

If we are ever able to artificially reproduce consciousness using "algorithm A" then the law for reducing consciousness to the computational properties of individual neurons is: "Consciousness is the result of applying algorithm A to X number of neurons in initial state S".

My questions:

  1. Does the existence of a method for artificially recreating an emergent property necessarily imply the existence of reduction of that property to that of its individual components?
  2. Are there any examples of emergent properties that can be artificially recreated?
  • A very related question is what does it mean to "recreate" something. As an example, chaotic flow is a very emergent phenomenon, which is very easy to recreate. However, each recreation is slightly different. However, each recreation will share some properties. Are these the "emergent" properties?
    – Cort Ammon
    Jul 11 '16 at 2:19
  • @CortAmmon not being sarcastic, but you say they're emergent. confusing expressions !
    – user6917
    Jul 11 '16 at 16:27
  • 1
    @MATHEMETICIAN It was really hard to write that. Chaotic flow is described as having emergent phenomena. Each time you create a new chaotic flow, many of the properties of the flow change. However, some of the properties of this flow will exist in each instance, just in a slightly different form. Whether these properties count as "emergent" or not affects the way the answer can go. We can clearly recreate this class of properties, but there are properties of each chaotic flow which have also "emerged," but are not easily recreatable. They make each flow unique.
    – Cort Ammon
    Jul 11 '16 at 16:32
  • Highly related is the phrase often given to kids: You're special and unique, just like everyone else.
    – Cort Ammon
    Jul 11 '16 at 16:32
  • No, (most) people know how to create conscious beings, but "the result of opposite sex individuals performing Kama-sutra moves" is not a reduction. Black box algorithms aren't, no matter how reliably they work, you need to show how theoretically. Supervenience Dupre calls "God's eye reductionism" and quips "there is presumably some set of facts that could be known that would permit the inference... Perhaps we could not, even in principle, know these facts. But God, I suppose, would need merely to exist in order to know them." classes.matthewjbrown.net/teaching-files/hps/dupre.pdf
    – Conifold
    Jul 14 '16 at 4:29

If one is looking at physically emergent properties, then it should be reasonable to assume that that property can be defined by looking at the state of an object. We may not be able to specify a formal string defining that property, but we can accept that there is one or else we have to revisit what a physical property means. This definition explicitly calls for path-invariance, which will be important.

When it comes to physically emergent properties, it is also generally assumed that there was a time evolution function to go from an initial state to the final object with that property (else we call it a "miracle"). However, we do not always know what that initial state is, nor do we necessarily have control over all of the factors that go into the time evolution of that object. In fact, it is generally assumed that if we choose to call a property "emergent," then there must have been features which we did not control.

To create an "artificial" object with the same property as the original, we must find a path which arrives at a state which has said property, but for which the time evolution only depends on features we can control. If such a path exists, then we can say that we have created an artificial object with the same properties as the original.

Whether this can be done or not is very dependent on which property you are interested in. As an example, vortexes in chaotic flows are typically considered emergent behavior. Some of these properties can be reproduced artificially. For example, if we define the emergent behavior based on the energies stored in the vortexes as they are shed, we can show statistical patterns of these very emergent structures which are extremely repeatable. However, if we want to say "the emergent property we are interested in is that this vortex is this vortex, and behaves as it does" then we cannot recreate it artificially because we cannot sufficiently measure the chaotic flow of that vortex to reproduce it. It is unique, but it follows statistical templates which are reproducible.

An excellent example to consider would be the creation of artificial heart valves. Heart valves, as grown by a young embryo, are decidedly emergent structures. They grow in a way which fits perfectly with that particular embryo's body size and shape. If you need a heart valve replacement, we cannot grow an exact replacement for your existing one in every way shape and form. However, we can artificially grow a replacement which is sufficiently similar to the old one that it's good enough to keep your blood pumping for years to come. The original structure is certainly emergent, even though we have found that we can artificially create a structure which is sufficient for our needs.


It might be possible to artificially create "emergence" and if it is created, it would not be a contradiction. This is based on the fact that some times "the total is greater than the sum of its parts." This is also known as "synergy." So, since in some processes there is synergy and in others there isn't, there is no guarantee that "emergence" will be recreated, even if you assemble the same parts again.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.