Basic HTML Version

which themselves generate the higher-order property.
This kind of emergent behavior cannot be detected by merely taking a snapshot of the
system and seeing what properties it has and how they stand with respect to each other. To
notice the emergent behavior you have to look at the dynamics of the system, how the higher-
order property becomes established and maintained. One of the things that happens in
emergence is that in order to explain the behavior of
E. coli
or the individual honey bee, you
have to appeal to things at a higher level of organization. There is a robustness of behavior in
the feedback mechanisms that tune the behaviors of the individual bees to the needs of the
colony. The whole exploration of the way in which these emergent structures are understood
and explained requires a fundamental shift in our mode of explanation. We have to give up a
certain kind of static view, in which we can explain higher-level properties by simply looking
at the components that constitute the system in an instant of time. It is not possible to explain
these higher-order properties unless we consider the dynamics of self-organization, feedback,
feedforward, and chaotic interactions. It is the dynamic perspective demanded by these
explanations that some frameworks for explanation have not been able to accommodate. That
is where the philosophical challenge comes from: explaining dynamic emergence.
Dynamic emergence satisfies the letter of the emergentist law. It has unpredictable
properties. The formation of these higher-level structures has been associated with chaotic
dynamics that are, by definition, unpredictable, as in bird flocking and in other mass
aggregate behaviors. Dynamic emergence displays novelty. The structures of the higher order
are not possessed by any of the individual components; they are not even possessed by an
aggregation of the individual components. They exist as a result of the dynamic interactions
between the components, the external context, and the internal context. Dynamic emergence
displays the gold standard of emergence: downward causation, when properties of the whole
change the behavior of the constituent parts, by both positive and negative feedback.
Meanwhile, the properties of the whole become stabilized by the behavior of the components,
permitting this kind of multi-level, inner causal dependence. There is a reductive character to
this phenomenon. We are not saying that this higher-level property emerges independently of
the components; however, to understand its properties and its stability, as well as the
properties of the components, you need to appeal both to higher and lower levels and to
downward causation.
Emergent structures and properties identify an important class of phenomenon that
require a revision in how explanations are generated. Philosophically, we need to represent
not just the presence or absence of properties but their dynamic stabilization, the unfolding of
new structures, new laws governing their behavior, and feedback and feedforward processes
that engage entities at multiple levels of organization. The need to appeal to multiple levels
requires non-reductive explanations and the appeal to the dynamics of behaviors requires
non-static explanations. We have to have new modes of explanation, because both reductive
and static approaches are inadequate to accommodate these new developments.