Causal Texture Theory

Can systems purposively learn or coevolve? Natural systems may learn and coevolve either through laws of science (e.g. biochemistry) or instinct (e.g. with insect behaviour). Human systems have an additional feature of will, so that we can shape desired futures. Even when we espouse either incremental or transformation change, some systems resist efforts to learn and/or coevolve.

Purposive redesign occurs either by change(s) in the system, change(s) in the environment, or both

Living systems differ from non-living systems in behaviours that are goal-directed. From a systems perspective, goals can be achieved in one of two ways: (i) the system changes in response to its environment, or (ii) the system remains static while the environment changes. Directive correlation in the game of football (i.e. soccer to North Americans) is described with two situations: either (i) the player moves towards the ball, or (ii) the player positions so that the ball comes to him (Sommerhoff 1969, 174–186). These two situations are simplifications of the reality where both the player and the ball are changing position in real time.

In a summary describing directive correlation, learning and coevolving in living systems presents three features additional to those in non-living systems.

1. The distinctive organization of living systems manifests itself in the goal-directedness of their activities. [….]

4. Goal-seeking is not the same as equilibrium-seeking, nor is it co-extensive with feedback control.

5-6. Directive correlation enables many biological key concepts … [including] adaptation, regulation, co-ordination, learning, instinct and drive. (Sommerhoff 1969, 201–202)

Thus, living systems demonstrate goal-oriented behavior that non-living systems do not. The goals may or may not be coincident with an equilibrium or a feedback control. Goals enable living systems, as individuals and as groups, to adapt and learn.

Effective purposive change in a system can depend on the causal texture of the environments

If we define a system of interest to be a living system that is goal-directed, its environment can include both living and non-living parts. A systems and its environment does not represent the whole world. In causal texture theory, a system and its environment is described as field, which is a whole amongst other wholes. (Ramírez, Selsky, and van der Heijden 2008)

Causal texture is an emergent property of the whole field and concerns the behaviour of all systems within it. The causal texture of a field sets conditions on how these systems and their shared environments transact (Selsky et al, 2007, p74).

Figure 1 shows the simplest representation of a single field, with a system labelled as “1” and the environment labelled as “2”. Linkages between those parts are labelled as L11, L12, L21, and L22.

A system of interest that is goal-directed could alternately be encouraged or inhibited by its environment. As parts of the field, change simultaneously occur (i) with parts within the system interacting as L11; (ii) with the system acting on the environment as L12; (iii) from the environment influencing the systems as L21; and (iv) with parts of the environment interacting as L22.

Four possible links between system and environment exist:

1. L11 (read as 'El one, one', not as 'El eleven') denotes links that remain internal to a system.

2. L12 links the system to its environment – system outputs, related to the planning function.

3. L21 links the environment to the system – system inputs, related tot he learning function.

4. L22 denotes links between elements of the environment itself, and which occur independently of the system. (Ramírez, Selsky, and van der Heijden 2008, 19)

The four causal textures have been described in a metaphor of a surface with food and competitors.

• I: Goals (food) and noxiants (bads) are randomly distributed. Actors should know the system, with the ideal of homonomy (i.e. a sense of belonging). Learning involves conditioning, and planning is tactical (e.g. if you need food, move!)
• II. Goals and noxiants are lawfully distributed. Actors should know the system and effects of action, with the ideal of nurturance (i.e. caring for the field). Learning involves meaning, and planning is tactical and strategic (i.e. if you need food, move; there's lots of food, so encounter a competitor, just move somewhere else).
• III: Goals and noxiants are lawfully distributed as in Type II, but two or more systems are competing for the same resources. Actors should know the system, the effects of action, and the changes resulting from learning, with the ideal of humanity (in the broadest sense, with the context of limited resources). Learning is problem-solving and planning is tactical with operational strategies (i.e. if you need food, movements should take competitors into account).
• IV: The field is dynamic, as Type III plans leads to emergent and unexpected outcomes. In addition to knowing about the system, action and learning, the attention shifts to appreciating the environment. The ideal is beauty, as the multiple systems and environments should fit together naturally. Learning is puzzle-solving of non-linearities, and planning is active adaptive planning (i.e the field is in motion, and when you move, you may add to shaking the ground).

In circumstances where collective learning and coevolving are desirable, differences in the appreciation of the causal texture lead to conflict. If one party believes that resources are plentiful while others see resource as constrained, the nature of their plans will differ. If one party believes that changes in their system have no impact on their environment while others see the entire field as turbulent, the nature of their plans will differ. The perceived values of learning, and each party's inclination to coevolve cooperatively may lead to irreconcilable clashes.

The extended fields of directive correlations were categorized into four types: (I) random placid; (II) clustered placid; (III) disturbed reactive; and (IV) turbulent. Table 3 highlights some of the wisdom gained from the original research by Emery and Trist in 1965 continuing through to the causal textures research by 2008.

Ing, David. 2013. “Rethinking Systems Thinking: Learning and Coevolving with the World.” Systems Research and Behavioral Science 30 (5): 527–47. doi:10.1002/sres.2229 . preprint available at http://coevolving.com/commons/201310-rethinking-systems-thinking