1. Questionable sustainability

Related concerns are echoed within the Sante Fe Institute in the sharp differences between Murray Gell-Mann (head of the science board) and George Cowan (the former director). For Gell-Mann, as co-initiator of the World Resources Institute (Washington DC), human society must undergo six "fundamental transitions" within the next decades to achieve sustainability and avoid global catastrophes. In his view this requires widespread agreement on principles (Waldrop, p. 351). One might ask what this "consensus" could possibly constitute within the nonlinear dynamics of a social system subject to paradigm shifts.

For Cowan however: "Somehow the agenda has been put into the form of talking about a set of transitions from state A, the present, to a state B that's sustainable. The problem is that there is no such state. You have to assume that the transitions are going to continue forever and ever...You have to talk about systems that remain continuously dynamic, and that are embedded in environments that themselves are continuously dynamic....A term like 'sustainable' does not really capture that." (Waldrop, pp. 350-356). Genuine sustainability in a living system is maintained by the nonlinear dynamics of continuing instability. As currently conceived as a desirable stable state, "sustainability" would remove that instability and thus render the global system unsustainable. Such stability is death; somehow the world has to adapt itself to a condition of perpetual novelty.

3. Need for value instability

To what extent does this confusion not reflect that in dealing with human values? The "equilibrium-centred view" in this case pleads for a universal set of fundamental unchanging values. The "dynamic view" argues for slowly evolving values. And the "evolutionary view" awaits crises which would evoke a reconfiguration of values. But a genuinely "sustainable" pattern of living values would seemingly depend on a form of value instability -- beyond any simple equilibrium. Efforts to "stabilize" that instability may then be precisely what would render the global system of values unsustainable.

4. Demonizing the inappropriate

As noted earlier, the world of value-making and value-defining is above all characterized by much demonizing of that which is rejected as inappropriate. It is easy for young tigers and self-selected people of sensitivity and wisdom to define boundaries within which their values shine brightly in comparison to those in the outer darkness of the unenlightened. Many can now create such boundaries, but unfortunately the numbers of the "unenlightened" as seen from within any boundary do not appear to diminish. Regrettably, through its systematic exclusion of the softer social sciences, the Sante Fe Institute, for example, has proved insensitive to the way in which its own interpersonal dynamics and tensions provide an interesting metaphor for the substantive issues by which they are currently challenged.

5. A framework for incompatible perspectives

The challenge would appear to be to find ways of integrating those holding incompatible perspectives within a framework for which a qualifier like "common" might obscure a necessary degree of complexity.

A valuable lead is provided by the work of Christopher Langton (at the institute) based on insights of Stephen Wolfram. They identified four classes of dynamical systems:

• Class I: Governed by a single point attractor. Doomsday scenario in which agents or patterns of behaviour rapidly die out.
• Class II: Set of periodic attractors. Relatively predictable pattern of behaviour, but essentially static.
• Class III: Governed by strange attractors. Chaotic behaviour which never settles into a predictable pattern. No stability. Structures breakup as soon as they form.
• Class IV: Emergence of coherent structures that continue to grow, propagate, split, and combine in a complex way.

Langton explored the nature of Class IV behaviour as a second-order phase transition between Classes I/II and Class III, namely a condition under which chaos and order are combined. In these terms Classes I/II corresponded to "order", whereas Class III corresponded to "chaos". At the interface or transition between them, at the "edge of chaos", the "complexity" of Class IV behaviour could emerge (Waldrop, p. 234). Class IV is: "a class of behaviours in which the components of the system never quite lock into place, yet never quite dissolve into turbulence, either. These are the systems that are both stable enough to store information, and yet evanescent enough to transmit it. These are the systems that can be organized to perform complex computations, to react to the world, to be spontaneous, adaptive and alive" (p. 293).

6. Value-governed behaviours

These classes could be represented together on Diagram I. That juxtaposition of classes can be used as a way of looking at the status of different value-governed behaviours in society. Thus:

• Class I: Conceptions of the ultimate attractor to which decaying patterns of order are pulled. Thus for Ralph Abraham: "We need a unifying principle and attractor" (Abraham et al, p. 126). For his colleague Rupert Sheldrake: "Why the apocalypse is such a strong attractor is an interesting question. The attractor beyond all the doom may be another state of being that is extraordinarily blissful compared with anything we know here..." (Abraham et al, p. 160). A guru figure may also focus the experience of such an attractor. Negative counterparts may be seen in: progressive marginalization; contextually determined behaviour and absence of choice; victims of society destined for a "psychic sink"; traditional values (of ancestors) perceived as quaint and of little current relevance. Possibly the black hole of individual or collective self-righteousness.
• Class II: Static established patterns of order (the "Establishment"). Patterns of rights, obligations and constraints. Top-down advocacy/imposition of behavioural guidelines (UN Universal Declaration of Human Rights, etc). Organized liberty. Emphasis on collective consensus. Time-tested cultural, parental (family) values.
• Class III: Chaotic behavioural patterns. Impermanence of social structures. Unconstrained freedom of choice. Emphasis on individualism. Postmodernist value relativism.
• Class IV: Emergent behavioural patterns of higher orders of complexity. Surprise. Innovation. Higher orders of consensus. Grounding of new values.

It is easy to fall into the trap of distinguishing "better" from "worse" amongst such classes -- especially in the effort to gain recognition for Class IV. Class IV then becomes the focus of attention as the new frontier onto which all illusions of progress can be projected. Class I can be disparaged as the "losers" in the evolutionary process. But a coevolving system requires the presence of all such classes and the real challenge is to recognize their interweaving functions more clearly.

There would therefore be merit in portraying such annotations to Diagram I in two parallel forms: one reflecting the image that each class of behaviours has for the others (to reinforce their identities); one reflecting the self-image of each as the most significant class of behaviour.

7. Demystification and self-mapping

It is quite possible therefore that there are useful ways of understanding some of the subtleties of the relationship between the classes of Diagram I. These may suggest metaphors throughwhich the "established" and the "emergent" can be more effectively related.

One devastatingly simple approach is to interpret the classes through the metaphor of the characteristic behaviours at different stages in the human lifecycle: Class II is readily associated with those of maturity and adulthood; Class III with the exploratory behaviours of youth and immaturity; Class I with those of the elderly. Remember however that such correspondences refer to behavioural patterns. A physically young person may exhibit Class I behaviour, just as a physically old person may exhibit Class III behaviour. Class IV behaviour is that associated with social innovation and the emergence of new patterns of relationship (partnerships, coalitions, teams, groups, etc) but especially that which ensures renewal (including reproduction).

The merit of this approach is to highlight the need for a healthy balance between the different classes of behaviour, despite the well-known challenges of relating youth and elderly within adult patterns. Some of the pathologies are clear in the "dump granny" approach of Class I behaviour, and the obsession of whole societies with maintaining a youthful image of Class III behaviour. The confusions, typical of New Agers, are also seen in failure to distinguish between Class III and Class IV behaviour. And the positive function of Class I behaviour is better seen in that which is evoked by attitudes towards the elderly in Latin and Oriental cultures, and the respect in which the wisdom of the elders can be held.

Much operational understanding of complexity is mapped into current understanding of such social relationships. The question might well be asked as to whether it is only through such a form of self-mapping that complexity can most effectively be understood (Judge, 1984).

Other approaches to recognizing the balance between such classes of behaviour would include those of analytical psychology inspired by Jung, for which such a four-fold division is quite fundamental. Buddhists, as noted above, have explored such configurations through mandalas. And Taoists would see a relation to their four-fold logic.

Michael Kirby (Aida et al, 1985), in discussing democracy and governance, notes Elgin's suggestion that there are four possible responses to the crisis of system complexity: successful muddling through; descent into chaos; authoritarianism; and system transformation into a higher level of efficiency and simplicity. These could be seen as corresponding to Classes II, III, I and IV respectively.

Such views should encourage a re-evaluation of the class distinctions (as originally derived from behaviour of cellular automata with insights from phase transitions). It is likely that there may be subtler features through which such behaviours should be distinguished to be of relevance to the social realm. The mathematics may be obscuring insights which have been well articulated through more accessible metaphors.

8. Functional relations between classes

The above exercise helps to clarify understanding of the nature of the functional relationships through which the classes are integrated into a coevolving pattern of behaviours. These are represented inDiagram I.

9. Phases, transitions and world views

Whilst there is undoubtedly merit in pursuing the four-fold set of classes, there is also value in recognizing the richness of the phase transition metaphor through which Langton identified Class IV. He was particular struck by Class IV as a phase transition between "solid" (Class I and II) and "liquid" (Class III). Other enthusiasts of four-fold symbolism would point to the traditional schema of Earth, Air, Water and Fire much scorned by modern science. They would undoubtedly respond to Langton's view that here "solid" and "fluid" are not just two fundamental states of matter (as in water or ice), they are two fundamental classes of dynamical behaviour, equally detectable in such utterly nonmaterial realms as the space of cellular automaton rules or the space of abstract algorithms (Waldrop, p. 234).

Given the attention by traditionalists to the subtleties of these distinctions, they might argue for the correspondences to be as follows: Earth (Class I), Water (Class II), Air (Class III) and Fire (Class IV).

In practice the relationship between such states is described in phase diagrams. Within these there are subtleties and variants that go beyond the ideal four-fold presentation. These include such notions as critical points and various solid states (7 of ice). These are healthy reminders that much remains to be explored.

Figure 1: Complementary relationships of value classes (tentative)