I'm trying to understanding supervenience emergentism, but the Stanford Encyclopedia of Philosophy entry is unclear.

My understanding of supervenience was that the lower level properties of the system determine the higher level properties of the system. So, once the lower level laws are known, we can determine the higher level laws.

However, the SEP entry says that ontological emergence (of which supervenience emergence is a type) claims that configurations can have "new primitive causal powers". I don't see how this is not a contradiction if the system's behavior is already determined by the lower level powers.

  • 1
    It doesn't determine all of them; heat is a property of matter in bulk, not of a particle. Apr 25, 2015 at 8:46
  • @MoziburUllah - But "heat" is not a "new primitive causal power". It is a bulk property that is used to simplify our description of the distribution of kinetic energies of groups of particles, and to explain regularities in their behavior in bulk.
    – Rex Kerr
    Apr 25, 2015 at 21:42
  • @kerr: sure; I don't disagree; it's in that sense it's called an emergent property; after all there is no sense of heat in a single particle - or is there? Apr 26, 2015 at 7:24
  • @MoziburUllah In a cute sort of way the uncertainty of position caused by the wave nature of matter, the lack of absolute rest discussed in your other post, can be thought of as entropic force, and so heat. But I would agree that we can only meaningfully observe heat statistically, and that we attribute cause to temperature in a way that does not reduce to the molecular level. The kettle boils because it reaches 100 degrees. The statistical phenomenon of temperature overcomes the statistical phenomenon of 'latent heat of condensation' which is keeping the water liquid.
    – user9166
    Apr 26, 2015 at 13:14
  • In a loose kind of way this is all related to the Sorites Paradox. Apr 26, 2015 at 18:00

1 Answer 1


If I get the notion of 'primitive' the SEP has in mind. This makes sense to me.

There are measures that can only be taken of the composite, and there are effects that can only be observed upon those measures by other measures that can only be taken at a similarly high level. So these are causal relations between things that cannot be stated in terms of the lower level attributes.

For instance, it seems clear that the decisions of corporations supervene from the psychological and physical needs of their individual employees, customers and stakeholders. But there are also causal forces that act upon corporations only as whole entities. For instance, we can see that monopoly has certain advantages, and a corporation may work hard to enter a market where it can secure and leverage a monopoly, perhaps by collecting patents.

But what is a monopoly in terms of individuals' needs? Monopoly, and more generally the concept of market-share, is a new driver of corporate action that arises only at the level of the whole corporation and relates mainly to other corporation-level and market-level drivers. Although they clearly proceed in detail from lower-level psychological or physical needs, its effects cannot be analyzed in terms of those alone. It is in that sense 'primitive' at this level.

Similarly, the notion of heat flowing through a space, or of osmotic pressure driving diffusion or of a certain pH causing corrosion can only be observed at a macroscopic level. We can prove how these effects are actually determined by statistical trends in molecular behaviors, so they supervene upon that lower-level behavior. But we cannot measure them or actually see their effects from that level. They are 'primitively chemical' as opposed to 'molecular' forces, and in that sense grossly chemical behavior is emergent from molecular behavior.

Likewise, mind may be emergent at the level of decision-making although it supervenes upon biological activity. We may observe mental processes that proceed from brain activity, but decisions may exhibit patterns that are not clearly discernable or describable at any lower level of organization.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .