International Encyclopedia of Systems and Cybernetics

HUME, as quoted by F. DAVID PEAT (1987, p.40), wrote: "We have no other notion of cause and effect, but that of certain objects which have been always conjoined together… We cannot penetrate into the reason of the conjunction". Elaborating on HUME, PEAT concludes: "The chain of linear causality, therefore, is a mixture of habit, belief, and common sense. But this common sense is based on a number of assumptions, namely:

- That two events are unambiguously separated from each other and have their own independent existence, as, for example, two bodies with well defined boundaries.

- That some contact, force, or influence flows from one body or event to the other.

- That there is a clear flow of time with the cause occuring in the past and the effect in the present" (p.41).

The spanish philosopher J MARIAS writes: "… causal explanation has to be schematic and universal, leaving out the concrete individuality, which is the most real part of the thing explained" (1975, p.145).

As stated by R. LILIENFELD: "… logical or pseudo-logical schemes… gain manipulability at the cost of reality" (1979, p.5).

In PEAT's terms, the supposed chain of causality is "in fact a complex network of causation". And "The more the limits of this network are extended, the more it is seen to stretch out over the entire earth and ultimately the universe itself" (p.45). We are thus faced with two practical problems when we want to construct causal models:

- Where to fix reasonably – i.e. on the base of a sufficiently global information – the limits of the network related to the system, situation or phenomenon we seek to modelize?. ("To talk of planetary influence on a game of tennis is obviously far-fetched" (p.45) – But possibly not so the effects of planetary temperature increase on investment at a seaside resort at a 30 years time lapse.

- How much should we extend the network into the past, admitting that we can obtain reliable information For example, using historical and geological information about the Mississippi flood plains.

E. JANTSCH equates causality with local determinism, "… often effective in describing microaspects of change in a nonsystemic and short range perspective. Newtonian mechanics belongs here as well as many predictive cause/effect models in the social domain" (1975, p.98).

(POINCARÉ's three bodies problem shows that Newtonian mechanics with all its achievements, is indeed a simplified view of celestial mechanics).

Complex systemic models are still models, but at least, they aim at a more complete description of real entities. To construct such models we are in need of cybernetic and systemic conceptual means, able to reflect a higher degree of complexity. New logical and mathematical formalisms are also required, as for example SPENCER BROWN's Laws of form; ZADEH's fuzzy sets, or LORENZ, SMALE's et al, deterministic chaos.

We thus end up with, in A. ARIELY's words: "… an album of maps to be studied not as an ultimate truth, but as way-points that the applicator can use provided that he verifies the model by his intuition, imagination and the conditions of the problem at hand" (1979, p.12).