Figure 1: A plesionic system is an interval of space-time (an arena) containing, among other things, a group of one or more agents (self) working together to understand and possibly influence an interval of space-time that contains, among other things, some plesionic systems (other). A wolf feeding her cubs is a plesionic system, so too is the pack she runs with, the researchers who work on the wolves (at least when they are co-located), the herdspeople trying to keep wolves at bay (at least while they are co-located) and the committee that funded the research team. Some plesionic systems (wolves, say, or individual humans) are persistent, others are transient artefacts of purpose and role-playing. An institution like a bank or a religious community is incapable of coming together to act as a coherent unit. We humans have evolved the ability to interact with them as if they were 'virtual agents' and developed complex protocols and legal codes that regulate our interaction with them.
Figure 2: In the Modern Synthesis populations are presumed to have been organised into reproductively isolated units that evolve by divergence to create a strict Linnaean hierarchy of forms (a). The Venn diagram (b) represents the classification obtained by analysing the hierarchic structure.
Figure 3: The evidence for primates suggests a network or heterarchy (a) of divergent and re-crossing lineages. Note how the classification (b) contains complex, cross-cutting boundaries.
Figure 4: Gene flow and biological clocks. These two trees show a hypothetical system in which three species, named A, B and C, diverge from one shared ancestor (to the top of the figure). The actual genetic history is shown on the left, with A diverging from B and C first and B and C from one another later. If this were all that had happened, the assumption of hierarchy would be justified and molecular clock methods would work straightforwardly. However, as the left side tree shows, in actual fact our system includes some subsequent gene flow between species A and B (indicated by the arrow). This would cause A and B to share some genes to the exclusion of C, and affect the way we reconstructed their relationships. The inferred evolutionary tree (on the right) would be distorted because the molecular clock would appear to run faster on some branches than others. Specifically, the B-C divergence would be pushed back into the past and the A-B divergence brought forward. The topology of the inferred tree would be a poor guide to evolutionary pathways, not because the data were incorrect, but because the hierarchic assumption would be unjustified.