Supervenience

Supervenience is used of the relationship between two kinds of properties that things may have. It refers to the way in which one kind of property may only be present in virtue of the presence of some other kind: a thing can only possess a property of the first, supervening kind because it has properties of the underlying kind, but once the underlying kind is fixed, then the properties of the first kind are fixed as well. The supervening features exist only because of the underlying, or ‘subjacent’ properties, and these are sufficient to determine how the supervening features come out. For example, a person can only be good in virtue of being kind, or generous, or possessing some other personal qualities, and an animal can only be alive in virtue of possessing some kind of advanced physical organization. Equally, a painting can only represent a subject in virtue of the geometrical arrangement of light-reflecting surfaces, and its representational powers supervene on this arrangement. A melody supervenes on a sequence of notes, and the dispositions and powers of a thing may supervene on its physical constitution.

Although the word supervenience first appears in twentieth-century philosophy, the concept had previously appeared in discussion of the ‘emergence’ of life from underlying physical complexity. The central philosophical problem lies in understanding the relationship between the two levels. We do not want the relationship to be entirely mysterious, as if it is just a metaphysical accident that properties of the upper level arise when things are suitably organized at the lower level. On the other hand, if the relationship becomes too close so that, for instance, it is a logical truth that once the lower-level properties are in place the upper-level ones emerge, the idea that there are two genuinely distinct levels becomes problematic: perhaps the upper-level properties are really nothing but lower-level ones differently described.

If this problem is dealt with, there may still remain difficulties in thinking about the upper-level properties. For example, can they be said to cause things, or explain things, or must these notions be reserved for the lower-level properties? Supposing that only lower-level properties really do any work leads to epiphenomenalism – the idea that the upper-level properties really play no role in determining the course of events. This seems to clash with common-sense belief in the causal powers of various properties that undoubtedly supervene on others, and also leads to a difficult search for some conception of the final, basic or lowest level of fact on which all else supervenes.