InterJournal Complex Systems, 1003
Status: Accepted
Manuscript Number: [1003]
Submission Date: 2004
Emergence and Entities
Author(s): Russ Abbott

Subject(s): CX.0



Emergence and Entities Emergence is the appearance of a macro entity or property from micro components. Attempts to characterize emergence have been only partly successful. To date, emergence has been treated primarily in three ways: (a) as a vague concept that seems to point to something (e.g., swarm behavior) that we can't quite characterize; (b) as a spooky phenomenon (e.g., vitalism) that is beyond scientific investigation; (c) as little more than a fancy new term for entailment. In this talk I will discuss emergence as a physically real phenomenon. In addition, since the notion of emergence depends fundamentally on the notion of an entity (after all, it is an entity of some sort that one thinks of as emerging), I also plan to talk about entities and how they may be characterized. I will distinguish among a number of types of emergence. · Multi-process emergence: the emergence of a pattern of activities from the interactive performances of concurrent processes, e.g., swarm emergence. · Single process, temporal, or sequential emergence: e.g., the emergence of a melody from a collection of notes by sounding a collection of notes in a particular sequence or the emergence of a (single threaded) computer process by executing a collection of computer instructions in a particular sequence. · Syntactic or structural emergence: e.g., the emergence of meaning (in the formal semantics sense) as a structured collection of symbols or the emergence of an algorithm as a structured collection of operations. · Static emergence: e.g., the two dimensionality of cloth arising from the one dimensionality of thread. · Definitional emergence: e.g., the emergence of prime numbers within the context of the natural numbers. I will discuss (a) the parallel between rules of composition in emergence and fundamental forces in physics: the fundamental forces are, in effect, rules of composition that allow micro elements to combine to form macro elements and (b) the parallel between emergence in general and the concept of object instantiation in object-oriented programming. Emergence presupposes the notion of entity. (When one talks about emergence, one talks about the emergence of some thing.) I will argue that we have grouped four different categories of concepts under the general notion of entity. · Physical entities: aggregations of mass with the property that the sum of the masses of the components, were they considered separately, exceeds the mass of the aggregation, e.g., an atomic nucleus. This applies to physical entities at all levels and essentially means that degree to which some physical aggregation may be considered an entity may be characterized by the energy required to break it apart. A handful of wet sand is less of an entity than an atomic nucleus. · Attractor entities: e.g., a lake, which consists of the water that has gathered at a low-point. It is really the attractor and the structure it creates as a container, i.e., the lake bed rather than the stuff that is contained, that defines the entity. · Process entities: an environment within which a process operates as long as resources are available, e.g., a fire. The environment is the area hot enough to support combustion. The resources are the combustible materials. Another example is a glider in the game of life. The environment is the collection of cells that effect or are affected by the glider cells at any time step. In the game of life, the resources are not relevant since they are supplied by the rules that define cellular automata. As two gliders meet, their environments merge, leading to a merged entity, which differs from the two original separate entities. These are the most interesting kinds of entities in that they are what we often give as examples of emergence, e.g., a swarm. The issue in discussing these entities is to characterize (a) the processes that define them, (b) the environment that supports the continued performance of those processes, and (c) the resources the processes consume, if any. Most entities in this category are not physically fixed: the physical materials that participate in one of these entities, e.g., a fire or a human being, change as the entity’s processes proceeds. The entity is the process and environment, not its materials. · Definitional entities: whatever we define to be entities, e.g., a pair of socks. These are the least interesting entities since they are completely arbitrary. They are useful for creating a structure for process entities to inhabit, e.g., a corporation—although for a corporation to persist usually requires the payment of an annual fee, i.e., a minimal process and the consumption of minimal resources. We also use them to make mental models—which may or may not turn out to be useful.

Retrieve Manuscript
Submit referee report/comment

Public Comments: