Variety Dynamics is concerned with the dynamic distribution of variety in any situation as the fundamental basis for the use of power, control and management. This is an especially valuable understanding for managing complex and hyper-complex situations. Variety is the number of possible options available to different elements in a situation.

Origins and relationship to Ashby's Law of Requisite Variety

Variety Dynamics originated in efforts to apply Ashby's Law of Requisite Variety to real-world complex contexts. This proved problematic because Ashby's Law is limited in four ways: it assumes the usual assumptions of systems and causal analysis, which include stability of boundaries, purposes, relationships, causal pathways and processes. Also, strictly, Ashby's Law of Requisite Variety applies only to two entities, a manager and something being managed. It assumes the variety of both is constant; and it assumes that, where humans are involved, their cognitive ability is unlimited.

Most important real-world situations do not fit these assumptions. Important real-world situations  have multiple elements managing and being managed. Additionally, entities, boundaries, purposes, relationships, causal pathways, processes and feedback loops in real-world situations are not stable and the above assumptions are not satisfied, This limits the valid application of  Ashby's Law of requisite Variety and also, perhaps more importantly, it bounds the valid applicability of systems analysis and causal analysis more generally.

In parallel, it became clear that human cognition is limited in predicting the behaviour and outcomes of real-world systems. Forrester and others identified as early as 1975 that human reasoning fails when addressing situations shaped by many feedback loops. This work's contribution was to identify, first, that this boundary of human cognition sits at a single feedback loop. Hence, modelling or some other non-cognitive method is needed for situations shaped by two or more feedback loops. Secondly, that research identified a widespread and strongly subjectively held human delusion that individuals believe they can reason through such situations regardless.

The above research developments led to changing and extending the Law of Requisite Variety into real-world practical situations. An early and significant step was developing a body of theory able to explain the whole of Machiavelli's advice in The Prince. This work led to lectures at universities internationally, including in Australia, in the early 2000s. These, and other practical extensions of Ashby's Law of Requisite Variety eventually produced the Foundational Axiom of Variety Dynamics. This extended the underlying concepts of Ashby's Law of Requisite Variety  into multidimensional, multi-agent space, beyond the reach of causal analysis.

During this time, it became clear that any adequate approach to providing guidance to managing important  real-world situations had to be intrinsically non-causal, and had to operate beyond systems thinking and dependence on human cognition, whether individual or collective,  to understand behaviours and guide the decisions and strategies that would result in preferable outcomes. Variety Dynamics axioms had to act at a higher level of abstraction than causal or systems thinking.

Since then, it has become clear that Variety Dynamics is a distinct field addressing real-world situations beyond the reach of methods from other domains. It is domain-independent: it applies wherever there is choice, and wherever situations do not conform to the assumptions of causal analysis or systems analysis  and this includes complexity analyses based on agent-based modelling or other evolutionary methods.

Over 65 individual Variety Dynamics axioms have been developed to date. These are practically applicable to guiding decision-making and effective strategy in complex and hyper-complex situations. To make them usable, the axioms are stated in natural language. However, each is defined so as to also be representable mathematically. Variety Dynamics has a complete parallel existence as a realm of mathematics defined in higher category and higher topos theories.

Current research strands

Variety Dynamics research currently spans eleven strands:

• The development and use of Variety Dynamics axioms to manage power and control in a wide variety of contested or complex situations
• Modifying the dynamics of distribution of variety to influence ownership and direction of power and control
• Using the Two-Feedback Loop Limitation Axiom to manage human agent behaviours and identify failure areas in decision-making
• The roles of time dynamics in variety dynamics
• Identifying the differing roles of transaction costs in modifying variety distributions to change the locus and ownership of power
• Identifying where Variety Dynamics is more effective than conventional force and related approaches to change the locus of power and its ownership
• Identifying the benefits of Variety Dynamics analysis over conventional causal analyses
• Mapping the benefits of the reduced information needs of Variety Dynamics guidance on decision-making, compared to conventional causal analyses
• Exploring the benefits and potential of the intrinsic covert nature of Variety Dynamics interventions
• Practical application of Variety Dynamics in improved guidance for addressing hyper-complex wicked problem situations that do not conform to standard system analysis assumptions
• Development of the new mathematical field of Variety Dynamics — the dynamic V distribution, its dynamics, and its mathematical relationship to power and control

A worked example: the Two-Feedback Loop Limitation Axiom

The Two-Feedback Loop Limitation Axiom holds that humans cannot mentally predict the outcomes or behaviour of any situation shaped by two or more feedback loops. This is a cognitive biological boundary as firm, in its own way, as other physical limits of the human body (like not being able to unaided jump 4m into the air). Six corollaries follow from this. They call into question much research and decision-making that relies on individuals' mental predictions about such situations, and point instead to mathematical or physical modelling as a valid alternative for complex systems that fulfil the assumptions of causally-based modelling or systems analysis. For situations that do not fulfil those assumptions, modelling does not work either. Instead it is necessary to use Variety Dynamics or other approach that operates independently of, and at a higher level of abstraction than, causality.