Supervenience, Emergence, Mind, and Magic

Peter Lupu has come out in favor of emergentism in the philosophy of mind.  Here is an argument he could use to defend the thesis that mental properties are emergent properties:

1. Materialistic Anti-Dualism: Human beings are nothing more than complex material systems.

2. Anti-Reductivism: Mental properties are not identical to physical properties, nor do the former logically imply the latter.

3. Anti-Eliminativism: Human beings do in fact instantiate mental properties.

4. Anti-Panpsychism:  The basic constituents of the physical world do not have mental properties.

Therefore

5. Mental properties are emergent properties, which implies that there are emergent properties. 

The cases for (2) are (3) are overwhelming, so I consider them 'off the table.'  Peter agrees.  Panpsychism ought to be investigated, but Peter finds it highly implausible, so let's assume it to be false for the sake of this discussion.  The crucial premise — the dialectical bone of contention if you will — between Peter and me is (1).  He accepts (1) while I reject it.  It is worth noting that there are at least three ways of rejecting (1): by being a substance dualist, or an idealist (see John Foster's work), or a Thomistic hylomorphic dualist.  So I would argue from ~(5) to ~(1).  But for now we assume that (1) is true.

For the above argument to work, a clear concept of emergence must be in play.  We should distinguish between synchronic and diachronic emergence, and between property and substance emergence.  For now we are concerned solely with synchronic property emergence.  James Van Cleve offers this definition:

If P is a property of w, then P is emergent iff P supervenes with nomological necessity, but not with logical necessity, on the properties of the parts of w. ("Mind-Dust or Magic? Panpsychism Versus Emergence," Phil. Perspectives 1990, p. 222.)

Synchronic property emergence is thus a species of supervenience.  Van Cleve refers to Jaegwon Kim for a definition of supervenience. Let A and B be families of properties closed under such Boolean operations as complementation, conjunction and disjunction. A strongly supervenes on B just in case:

(SS) Necessarily, for any property F in A, if any object x has F, then there exists a property G in B such that x has G, and necessarily anything having G has F.

Now how does emergence as defined by Van Cleve differ from strong supervenience as defined by Kim?  One difference is that (SS) is defined for a single domain of objects:  the supervenient  properties are properties of the same objects as are the subvenient properties.  But the extension to multiple domains seems easy enough.  Emergence is a kind of multiple domain supervenience inasmuch as the emergent-supervenient properties are properties of a whole the proper parts of which instantiate the subvenient properties.  A second difference is that the relation between the supervenient-emergent properties and the subvenient ones is not logical but nomological.

Some Questions About Synchronic Property Emergence

A.  True Emergence Versus Epistemological Emergence.  Talk of emergence might in the end be merely epistemological.  Suppose system S has an observable feature F that cannot be explained in terms of  the properties of the parts of S or in terms of the relations of these parts to each other or to things external to S.  It  could turn out that the parts of S  have properties that we have not yet discovered and that F derives from these properties.  Or it could be that there are parts of S that have yet to be discovered and that the properties of these parts imply F.  For these reasons, Thomas Nagel feels sure that "There are no truly emergent properties of complex systems." ("Panpsychism"  in Mortal Questions, CUP 1979, p. 182.)  On this approach, talk of emergent properties merely signals our ignorance.

So here is a challenge for Peter:  Are there some nice clear examples of complex systems that exemplify truly emergent properties?  No doubt there are plenty of wholes that have properties that are not properties of the parts of these wholes.  For example, the set of natural numbers is infnite, but no member of this set is infinite.  For a second example, a wall whose weight is x lbs is not such that each of its constituent stones weighs x lbs.   But a property of a whole that is not a property of its parts is not eo ipso an emergent property of the whole.  In the case of a wall made of stacked stones (and nothing else),  the weight of the wall is the sum of the weights of the constituent stones and is therefore logically determined by them.  Therefore, the weight of the wall is not an emergent property of the wall.  For a property of a whole to count as truly emergent it would seem that its connection to the properties of the parts of the whole would have to be logically contingent.  This desideratum is captured by Van Cleve's definition supra.

B. The 'Magic' or 'Poof'  Objection.  If mental properties are not logically implied by physical properties, then it might seem that they emerge from their physical base 'by magic.'  If this were the case, then emergentism would leave physicalism behind.  For an unconvincing and needlessly polemical presentation of the 'Poof' objection see here.  If the connection between base and emergent properties is not logical but nomological then perhaps the 'magic' objection can be defused.  Non-logical derivation need not be 'magic' derivation. This ought to be more carefully examined in a separate post.

C.  My Objection.  Consider the mental property of feeling anxious.  Assume it is truly emergent from certain base properties of the brain and central nervous system.  Surely a property cannot be anxious.  It is not the property, but that which has the property, that is anxious.  For a materialist, the subject of anxiety or of any mental state can only be a material entity, the brain or the brain-cum-CNS.  But for reasons given elsewhere, it is difficult to see how how the brain or any physical thing could be that which is conscious, how it could be the subject of conscious states.

I conclude that property emergentism is quite unavailing for the purposes of the naturalist who wishes to account for mind in wholly naturalist terms.  What the emergentist must do is to try to make sense of  substance or individual emergentism.  I am not a property but a thinking individual.  So even if mental properties are true emergents, it doesn't help.  If the thinking individual that I am can be said to be emergent then the naturalist may have a viable position.  At  this point we need to consult William Hasker's The Emergent Self.