International Encyclopedia of Systems and Cybernetics

"The theory that the behavior or processes of systems can be precisely and fully forecasted" (Adapted from J.Z. YOUNG, 1978, p.291).

YOUNG claims that the concept must be "considered to be meaningless because no such principle can be demonstrated".

The concept of absolute determinism was enunciated as follows by LAPLACE: "An intellect which at a given instant knew all the forces acting in nature, and the position of all things of which the world consists – supposing that said intellect were vast enough to subject these data to analysis – would embrace in the same formula the motions of the greatest bodies in the universe and those of the slightest atoms; nothing would be uncertain for it, and the future, like the past, would be present to its eyes".

Nobody can object such a statement, about a "theory of everything". But, obviously, no human being, nor any other living being possesses this Laplacian "intellect", as shown by relativity theory as well as by HEISENBERG 's indetermination principle.

While LAPLACE's statement was unequivocally hypothetical and quite carefully worded, subsequent 19th century generations of scientists and philosophers adhered to the concept as if it were possible, true and necessary as a rigorous causal (and even sometimes mono-causal) principle that should be applied strictly to every kind of scientific work.

Indeed, various aspects of such an absolute principle and its applications came under close critical scrutiny during the 20th century:

1. Physicists have been compelled to admit a basic indeterminacy at micro-physical level. No "human intellect' can know "all the forces" and "the position of all things" at a given moment. This situation turns elusive the quest for causal determinism and future becomes fuzzy and not anymore rigorously predictable. Observation perturbs and non-observation leaves us in the dark.

2. It is not anymore evident that the Universe is globally and perfectly coherent. As simultaneous local causes cannot propagate instantaneously, global determinism, if it exists, breaks up locally everywhere: local and more or less provisionally disconnected determinisms appear and provide for a more or less chaotic and unpredictable future.

3. Any universe (or system) is at least partly blind to itself (see Jumarie's relativistic theory of information – 1980 and 1982). The only perfect model of the universe (or the system) is itself. This could lead to a very curious reformulation of LAPLACE's statement.

4. Our laws (deterministic or otherwise) are constructs, or algorithms. We acquire them by consensus, but they are never completely immune to falseation "a la POPPER". We do read some measure of determinism in the universe, but we are safer epistemologically than ontologically.

As to systems, a consensus seems to be emerging: any system is governed by some basic determinisms at global or macro level, but its organizational closure provides it with a more or less broad measure of autonomy, i.e. the (determined!) capacity to determine itself, within limits.

It has ultimately become clear – through the theory and models of deterministic chaos – that determinism is necessarily relative due to the successive deconnections and reconnections of simultaneous events in complex systems. This introduces a relation of traditional determinism with the reversibility concept.

According to F. BONSACK: "When we look for an example of determinism, we readily recall two phenomena: the first, the collision of perfectly elastic balls, the other one, the rotation of the planets around the sun. One may make a remark that, at first sight, could look as anecdotal: both cases are related to almost reversible phenomena. Laplacian determinism is reversible: starting from time t₀, one may compute the state of the system at a future time t₁, as well as at a former time t_-1,. Moreover, in the balls model, one reaches a paradoxical situation: it is possible to go back earlier than the initial situation" (1983, p. 113).

The apparent paradox is a result of the ever recurrent confusion between the real situation and the model we use to describe it. Reversibility exists only within the conceptual model. The planet JUPITER has occupied, occupies and shall occupy (very probably) a certain position (related to an abstract frame of reference) on various moments of the past, the present (unique) and the future. But it cannot really backtrack in the time dimension. The various past and future positions are merely computable.

The real interest of classical determinism is of a practical nature: it permits to retrieve former situations and to predict future ones by computation, always within the limits – sometimes dubious – of the "…et ceteris paribus" principle.

F. BONSACK immediately makes the point:

"Laplacian determinism is reversible. But our familiar notion of causality is not: the cause is always clearly different from the effect and generally it is not possible to permute them. This means that there must be in our notion of causality something fundamentally irreversible" (Ibid).

For a very detailed discussion of this question, see F. BONSACK's paper.

V.V. NALIMOV makes a very interesting study of the origins and history of determinism in its pre-logical and logical development (1981 b, p. 90-95)