On approaches to systems thinking

Including critical systems thinking

This earlier article introduces systems thinkers and their ideas and distinguishes physical and social entities from activity systems.

Of course, seeing a business as a social entity is important; and is a primary responsibility of business managers.

Management consultants continually generate approaches to identifying problems in human social entities and solving them.

The question here is not whether these approaches are useful.

It is whether classifying all these approaches as varieties of "systems thinking” has a useful meaning.

This article is not so much about systems as about approaches to systems thinking

Gerald Midgley (2000) presented three kinds of approach - as an evolutionary sequence.

This article challenges this classification and the notion that “critical systems thinking” is a major advance.

It proposes some other ways of thinking about systems thinking.

Contents

Midgely’s classification of approaches. 1

Class 1: “Hard” systems thinking approaches. 1

Class 2: “Soft” systems thinking approaches. 1

Class 3: “Critical” systems thinking approaches. 1

Design thinking and double loop learning. 1

Conclusions and remarks. 1

Further reading on Avancier’s system theory page. 1

APPENDIX: Five more ways of classifying systems and systems thinking. 1

Midgely’s classification of approaches

If every discernible thing, every nameable discrete entity, is a system, the term is useless.

What is the point of system theory?

As in Bertalanffy's general system theory, Ashby's cybernetics, Forrester's system dynamics and Checkland's soft systems method?

It is to identify the features that characterize a system in general.

All those gurus defined a system as a model, an abstraction, a perspective of a discrete entity.

The model defines only those features that the entity displays when it realizes the system that is modelled.

To paraphrase Meadows: "Is anything not a system?

Yes, a passive structure, like the Dewey decimal system, or any other hierarchical or tabular structure.

Also, an informal social entity, a group of actors who do not interact in the particular ways that characterize a system."

In short, a system is definable as a particular set of regular or repeatable interactions.

All the systems of interest here feature actors (or components) that interact in the performance of activities.

Actors are structures that exist in space and perform activities; and in social systems thinking they are almost always human.

Activities are behaviors that happen over time, and change the state of the system or something in its environment.

Systems thinkers like to classify systems: e.g. into organisms, animal societies and machines.

They also like to classify approaches to thinking about systems and solving problems.

Midgley wrote of three phases of inquiry:

1. Hard systems thinking approaches

2. Soft systems thinking approaches

3. Critical system thinking approaches

He presented each class as an advance on the previous class, bringing a new set of methods.

This article suggests classification is misleading, for reasons to be explained.

Class 1: “Hard” systems thinking approaches

Midgely wrote that hard systems thinking focused on solving concrete ‘problems’ where there was perceived “unity of purpose”.

He listed the following approaches under this heading.

1956 General Systems Theory (Bertalanffy)

1956 Classical cybernetics (Ashby)

1957 Operations research (Churchman et al.)

1962 Systems engineering (Hall)

1963 Socio-technical systems (Trist et al.)

1965 RAND-systems analysis (Optner)

1971-72 System Dynamics (Forrester; Meadows et al.)

Von Bertalanffy introduced the idea of a cross-science general system theory in the 1940s.

In 1968, he wrote that “All scientific constructs are models representing certain aspects or perspectives of reality.”

General system theory

System Models

Observers <observe and envisage> System Realities

Unfortunately, by referring to a biological entity as a system, he tended to conflate (at least in readers’ minds) the reality and the system.

By contrast, other systems thinkers urged us to separate the model and the reality.

W Ross Ashby popularised the usage of the term 'cybernetics' in the study of self-regulating (rather than self-organising) systems.

“Our starting point is the idea, much more than a century old, that a machine, in given conditions and at a given internal state, always goes to a particular state.”

But when Ashby spoke of a machine, he wasn’t thinking of a real-word entity (physical, biological or social)

He was thinking of some repeated or regular activity that is displayed by such an entity.

People do casually point to an entity (a motor car, a church, or IBM) in the world and call it a system.

Yet equating entities with systems is the most common mistake in systems thinking discussion.

Ashby wrote:

“At this point we must be clear about how a "system" is to be defined.

Our first impulse is to point at [some real-world entity] and to say "the system is that thing there".

This method, however, has a fundamental disadvantage: every material object contains no less than an infinity of variables and therefore of possible systems.

Any suggestion that we should study "all" the facts is unrealistic, and actually the attempt is never made.

What is necessary is that we should pick out and study the facts that are relevant to some main interest that is already given.” (Ashby 1956)

In cybernetics, a system is an abstraction, a theory of how an entity behaves, or should behave.

Ashby’s system is a model of some regular activity; it represents any entity or “real machine” that performs as described in the model.

Ashby’s cybernetics

Systems

Observers <observe and envisage> Real machines

Jay Forrester (a professor at the MIT Sloan School of Management) was the founder of System Dynamics.

He defined a system as a set of quantifiable stocks (populations or resources) that interact and affect each other.

System Dynamics

Causal loop diagrams

System modellers <observe and envisage> Inter-related quantities of things

This site gives more detail http://systemdynamics.org/what-is-sd

Remember Midgely wrote that hard systems thinking focused on solving concrete ‘problems’ where there was perceived “unity of purpose”.

To the contrary, hard system approaches are used where the purposes of the system are not agreed.

(And some theorits duck the question by saying the Purpose of a System Is What It Does, POSIWID).

In a “hard” approach to business system planning, like TOGAF, people are expected to

· Look at the big picture, overarching business drivers, goals and principles

· Identify stakeholders, their differing goals, concerns and viewpoints

· Define value propositions for each stakeholder group.

· Continually asses and manage changes to requirements and constraints

· Analyse the current situation and consider/define changes.

Even mechanical engineers are taught identify stakeholders and their different perspectives of what a system is for.

Class 2: “Soft” systems thinking approaches

Midgely wrote that soft systems thinking approaches take a wider perspective of people and their perspectives.

He said they focus on problems where the purposes of the system are not agreed.

He listed the following approaches under this heading.

1971 Inquiring systems design (Churchman).

1972 Second order cybernetics (Bateson)

1972 Soft systems methodology (Checkland)

1981 Interactive management (Ackoff)

1981 Strategic assumption surface testing (Mason and Mitroff)

1988 Cognitive mapping for strategic options development and analysis (Eden).

The distinction between hard and soft systems is questionable.

For some, soft systems are ones in which human actors perform activities.

For others, soft systems are abstractions from real-world entities or situations.

And for others, soft systems are both of the above.

Peter Checkland promoted a “soft systems methodology”.

He saw a business system as an input-to-output transformation, describable in a “business activity model”.

He regarded the model as a perspective of a reality, a world view or “Weltenshauung”.

Meaning that different observers may perceive different systems, some in conflict, in any one human organization.

Checkland’s Soft systems methodology

World views

Observers <observe and envisage> Human organizations

Russel Ackoff, a writer on management science, spoke of abstract and concrete systems.

An abstract system is a description or model of how an entity behaves, or should behave.

E.g. the game of “poker” describes how some people behave, or should behave, when playing a game of cards.

A concrete system is any entity that conforms well enough to an abstract system.

E.g. a real-world game of poker) is the realisation by a particular social entity of an abstract system.

Ackoff’s system theory

Abstract systems

System thinkers <observe and envisage> Concrete systems

Not only is the distinction between hard and soft systems questionable.

Checkland observed also that the distinction between hard and soft system approaches is also slippery.

Most of the ideas in his soft systems methodology have counterparts in approaches used in designing deterministic human and computer activity systems.

Heinz von Foerster was an Austrian American scientist who combined physics with philosophy.

He is widely credited as the originator of “second-order cybernetics.”

A basic idea was to include the system describer in the system that is described.

And allow the system to be self-organizing, in the sense of creatively self-describing.

From the 1970s, second-order cybernetics was developed and pursued by thinkers including Heinz von Foerster, Gregory Bateson and Margaret Mead.

It is said to be the circular or recursive application of cybernetics to itself.

It shifts attention from observed systems to the observers of systems, or rather, of human societies and businesses.

In human contexts, a system’s actors can also be system thinkers, who study and reorganise the system they play roles in.

Seeing the world as a duality of systems and observers is naive.

The several gurus mentioned above had a more sophisticated view, as shown in this triangle.

General system theory

System Models

Observers <observe and envisage> System Realities

A concrete entity is a system in reality only when and in so far as it realises a testable system description.

The observer(s) of a real word entity may abstract countless different (possibly conflicting) systems from its behaviour.

The actors in a social entity may act as observers of that entity – what it is and what it does.

They may be able to change the rules of activity systems realised by that entity.

The first trouble with second-order cybernetics is that it tends to conflate the concepts of:

· a social entity in which actors act as they choose

· an activity system in which actors play specific roles and follow specific rules.

The second is that it conflates a role in a system with a role in a higher process or meta system.

Since one human actor may play two roles, at different moments being.

· an actor in a system reality

· an observer of that same system reality.

The next section picks up this theme.

Class 3: “Critical” systems thinking approaches

The word “critical” implies what some call a dialectic.

That is, a logical investigation or discussion of the truth of propositions.

In the 19^th century, this was a feature of Hegel’s philosophy and Marxism (which tends to promote dialectic over evidence).

“Critical theory” emerged from the Frankfurt school of sociology, which came to prominence in the 1930s.

Midgley presented critical systems thinking as the latest development in a historical progression

He listed the following approaches under this heading.

1983 Critical system heuristics (Ulrich)

1990 System of systems methodologies (Jackson)

1990 Liberating systems theory (Flood)

1991 Interpretive systemology (Fuenmayor)

1991 Total systems intervention (Flood and Jackson)

2000 Systemic intervention (Midgley).

Midgely’s reference dates run suspiciously neatly in sequence from class 1 through class 2 to class 3.

Is it meaningful or useful to regard class 3 as an evolution of class 2, which is an evolution of class 1?

So, what is in critical systems thinking but not found in hard systems thinking approaches?

A logical investigation of perspectives?

Ulrich (1983) might have been thinking of critical theory when he added the word “critical” to “systems thinking”.

He defined three heuristics of the approach.

· Making sense of the situation: understanding assumptions and appreciating the bigger picture

· Unfolding multiple perspectives: promoting mutual understanding.

· Promoting reflective practice: analysing situations – and changing them.

OK, but that could be a description of Checkland’s soft system methodology.

Even in a “hard” approach to business system planning, like TOGAF, people are expected to

· Look at the big picture, overarching business drivers, goals and principles

· Identify stakeholders, their differing goals, concerns and viewpoints

· Define value propositions for each stakeholder group.

· Continually asses and manage changes to requirements and constraints

· Analyse the current situation and consider/define changes.

The concept of system transformation?

Midgley presented critical system thinking approaches as a post 1980 development of hard/soft systems thinking approaches.

The use of the terms “total” and “systemic” imply these approaches lead to a root and branch transformation.

The result, if not a revolution, is a major generational change from a current state of being to a new state of being.

OK, but this sounds like Michael Porter’s idea (1985) of radical business process reengineering

Or a “hard systems” approach like TOGAF in which business transformation involves

1. modelling a current system (N): analysing, discussing and criticising it, envisaging changes.

2. modelling a target system (N+1): discussing, reviewing and agreeing it.

3. planning the change from system N to system N+1.

4. changing system N to system N+1.

Agreeing the system boundary and encapsulating it?

Addressing boundary disputes and encapsulating systems may be a feature of critical systems thinking.

But agreeing the boundary of a system (expanding it, contracting it, or shifting it) has always been a feature of systems thinking;

A unified multi-method approach?

Some say critical systems thinking unifies different systems approaches, and advises managers how best to use them.

But then, so does any general architecture framework like TOGAF.

Every systems thinking approach recommends documenting mental models for discussion and agreement with others.

They suggest the kinds of model you can or should document.

They suggest using variety of techniques such as stakeholder management and risk management.

Emancipation and self-organization?

Emancipation means giving rights and power to the actors in a “system”, as in second order cybernetics

This is a human sociological idea rather than a general system theory idea.

Which is fine; the idea is important in today’s “agile development methods” and “learning organizations”.

However, if actors can define their own actions, that undermines the concept of an activity system.

Ashby and Maturana rejected self-organization as undermining system theory.

In effect, they argued that second order cybernetics confuses:

· the acts of actors in an activity system with

· the acts of actors in social entity that can define or change the activity system

Bear in mind, the same actor can play a role in both.

Not only can one social entity can realise several (possibly conflicting) activity systems - at the same time and over time.

But also, there is a need to verify that entity does indeed behave in accord with a particular system description.

If we don't distinguish an ever-evolving social entity from the different activity systems it may realise, the concept of a system evaporates.

Design thinking and double loop learning

The Hasso-Plattner Institute of Design at Stanford promotes a design thinking approach with five phases.

These phases aren't strictly sequential process steps – they can occur in parallel and be repeated (exactly as TOGAF says of its processes).

· Empathise – with your users

· Define – your users’ needs, their problem, and your insights

· Ideate – by challenging assumptions and creating ideas for innovative solutions

· Prototype – to start creating solutions

· Test – solutions

Design thinking ideas have long and widely been used in conventional (hard and soft) system design methods.

TOGAF is neither a prescriptive process, nor entirely consistent and coherent.

It is a flawed but highly flexible assembly of processes and products people adapt as they choose for change initiatives that need big-up-front design and planning.

It embodies the "design thinking" principles of Herbert Simon in “The Sciences of the Artificial” (1969).

His core ideas being that

· designers spend time up front deciding the basic/fundamental/root issue(s) to be addressed

· don't search for a solution until they have determined the real problem; and

· consider a range of potential solutions before settling on one.

TOGAF embodies those and other design thinking ideas below.

1. Capture the inspiration, the vision (phase A).

2. Take a human-centric view of business processes (A and B).

3. Manage stakeholders and value propositions (throughout).

4. Treat all design as re-design, as a baseline to target transformation (throughout).

5. Make ideas tangible to facilitate communication. Use visual languages, sketch diagrams and technical drawings to show abstract requirements may be met by concrete systems (A to E).

6. Double loop learning (throughout).

The last idea is of particular interest here.

Remember the distinction made by W Ross Ashby in 1956 between system state change and system mutation?

Single loop learning is the everyday response to some condition.

E.g. A thermostat detects a room temperature less than the selected temperature, turns on the heating, and keeps it on till the room temperature equals the selected temperature.

Double loop learning analyses why the system exists, and considers ways to change it.

E.g. Why is the room heated at all? What is the best way to heat it? Does it need better insulation?

Conclusions and remarks

Systems thinking approaches are supposed to help us understand, design and solve problems in systems.

Social systems thinkers have a tendency to deprecate so-called hard systems approaches.

Yet Checkland observed the distinction between hard and soft system approaches is slippery.

And hard system approaches often include ideas also found in soft systems, critical systems and design thinking approaches.

The distinction between hard and soft systems approaches is questionable.

By contrast, the distinction between classical and second-order cybernetics is fundamental.

On the schism in systems thinking

This article assembles many and various definitions of cybernetics.

Since some definitions are misleading or naïve, the article confuses more than clarifies.

And it masks the deep schism between two branches of “cybernetics”.

Cybernetics (after Ashby) and system dynamics (after Forrester)

Both of these activity system theories are about regular interactions between actors.

They are about a set of quantified state variables, which are advanced by repeated activities.

The variables are updated by inputs/events (Ashby), or flows that may aggregate a time unit-worth of events (Forrester).

Activity systems can be modelled in causal loop diagrams.

Activity system theories address “emergence”, “non-linear behavior”, “chaos” and self-organisation”.

Ashby included the observer in the system, but only as an observer or describer of the system.

“Second-order cybernetics” (after von Foerster')

The basic idea may be expressed as allowing people to be both actors in and observers of a system.

This is widely interpreted as meaning actors in a system can interact as they like, and change the system “on the fly”.

This social entity thinking has little to do with Ashby’s or Forrester’s ideas.

Unfortunately, followers use system terms (notably the four above) with different meanings.

The resulting confusion has undermined cybernetics, and system theory in general.

And disproved the definition by Margaret Mead quoted in the article!

The trouble is that second-order cybernetics tends to conflate the concepts of:

· A social entity in which actors act as they choose, and can change the rules of activity systems they play roles in

· An activity system in which actors play specific roles and follow specific rules.

Naturally, gurus like to present their preferred approach to systems as the latest development in a historical progression.

That doesn’t necessarily mean their approach evolved from past ones, or even that it applies essential system theory ideas at all.

Reshaping Midgely’s classification

Of course, seeing a business as a social entity is important; and is a primary responsibility of business managers.

Management consultants continually generate approaches to identifying problems in social entities and solving them.

The question here is not whether these approaches are useful

It is whether classifying all these approaches as varieties of "systems thinking” has a useful meaning.

If every problem or situation is a system, if every entity we name or point to is a system, then the term “system” is meaningless.

This table expresses the schism between two kinds of “system thinking”.

Activity system thinking	Social entity thinking
Classical cybernetics	Second order cybernetics
General to all domains of knowledge	Specific to situations in which humans interact
About roles, rules and regular activities	About individual actors who are purposeful people
About describing testable systems	About solving a social or business problem in a consensual way
Apolitical	Promoting a “participative democracy”

This table reshapes Midgely’s 3-way classification into a 4-way classification.

A 4-way thinking classification	About	Midgely’s classification	Cybernetic variant
Activity system thinking	A network of actors performing regular activities	Class 1 hard systems thinking	Classical cybernetics
Human activity system thinking	A network of human actors performing regular activities	Class 2 soft systems thinking	Classical cybernetics
Social entity thinking	A network of human actors who choose their activities	Class 2 soft systems thinking	Second order cybernetics
Situation thinking	Problematic situations	Class 3 Critical system thinking	Second order cybernetics

APPENDIX: Five more ways of classifying systems and systems thinking

This final section proposes several other ways to classifying thinking about systems thinking.

· Stateful systems versus stateless systems

· Loosely-scripted versus tightly-scripted processes

· Activity-oriented versus aim-oriented approaches

· Strong systems versus weak systems

· System thinking versus situation thinking

The actors of a business interact, directly or indirectly.

Sometimes they interact to meet aims of the business, sometimes they act to meet their own aims.

The actors form a social entity that can realise many systems – even conflicting ones.

Each of those systems has a current state, which is advanced by processes.

Stateful systems versus stateless systems

Stateless systems, which maintain no persistent structure or memory, are of little interest here.

Our interest in systems that do maintain a persistent structure or memory.

Social systems are ones in which actors share and remember information about the world.

Their memory of past events influences how actors respond to future events.

A business system can be seen as a formalised activity system in which the memory is persisted.

Information is persisted, for future access, in the data structures of messages, documents and databases.

And once persisted, the data structures can be shared, can be accessed by many actors.

Loosely-scripted versus tightly-scripted processes

The processes of a business system advance the state of its persistent memory.

Those processes may be more or less tightly scripted.

Tightly-scripted (procedural) processes

One or more actors may play defined roles and perform defined activities in end-to-end procedures or workflows. E.g

· Booking a train seat

· Applying for a job

In the course of an end-to-end procedure, actors may access and update one or more shared data structures.

Often, different roles have different rights to update different parts of the data.

There is usually an end goal, and the procedure ends when it is reached.

Loosely-scripted processes

One or more actors may be invited to do what they judge necessary to progress the life history of a data structure.

E.g.

· Write a document, e.g. a plan for work to be done.

· Post a topic on a message board; e.g. a Linkedin discussion thread

In a loosely-scripted system, actors may be given considerable freedom of choice over their actions.

To some extent, they may choose which activities they perform (e.g. post message, reply to message, edit message, delete message).

To some extent, they may choose which sequence they perform those activities in.

And they may be given roles that enable them to collaborate in an informal way.

Note that even in what appears to be a non-procedural system there are some defined roles and rules.

On a message board, the author role is distinguished from other contributors

The process flow may divide into parallel threads, each with its own constraints: e.g. (post > reply* > delete) and (post > edit* > delete).

If there are no roles for actors and rules for activities, if actors’ activities are wholly unconstrained, then there is no system, only a social entity.

A loosely-scripted process may have a goal and terminate when the data structure reaches a desired state – e.g. a document is approved.

Or else, there may be no end state - other than deletion of the data structure.

Activity-oriented system versus aim-oriented social entity

The table below contrasts two management styles which correspond to the schism in systems thinking

	Activity system	Social enttity
Processes are	tightly scripted	loosely or not at all scripted
Activities are	tightly constrained by rules	loosely or not at all constrained by rules
People are given	roles and rules follow	targets only
E.g.	call centre operators	door-to-door sales people
	Classical cybernetics	Second order cybernetics

Some approaches focus on defining aims (purposes, goals or targets) and motivating a group of people to meet them.

You may do this with little or no attention to defining roles for actors and rules for activities.

At the extreme, aim-oriented management means simply asking a group of people to meet some goals by doing whatever they choose.

You rely on the individuals’ abilities to interpret your directions and choose activities that lead to the given aims

If the nature and nurture-given abilities of the actors are up to it, they may succeed better than you expected.

Most people management involves some process-orientation and some target-orientation.

The question here is not which is better or worse, more or less advanced, it is whether it target-oriented management is well-called “systems thinking”.

At the extreme, there is a social entity, but no activity system - as the term is defined here.

Because a system is a set of roles and rules that determine who can do what and when.

Strong systems versus weak systems

The table below is one of several system classifications I have toyed with.

I don’t mean to present this as a scale of complexity, or a progression of any kind.

System kind	Description to reality relationship		Examples
Strong or Involuntary System Weak or Voluntary System	The script is embedded in the mechanisms of the concrete reality	Designed	Cuckoo Clock
		Natural	Solar system
	The script (DNA, code) is followed rigidly	Designed	Software system
	The script (DNA, code) is followed rigidly	Natural	Earthworm
	The script is followed loosely, as closely as actors choose	Designed	Orchestra, Church
	The script is written by the actors, more a social entity than a system	Natural	Marriage, Small business

System thinking versus situation thinking

Classical cybernetics is scientific in so far as its deals with activities that are regular, or deterministic, or reproducible.

Actual (empirical) performances of activities are tested for conformance to abstract (theoretical) descriptions of those activities.

As, for example, the actual orbits of planets are tested for conformance to astronomers’ descriptions of those orbits.

Social system thinking can be seen as a kit bag of ideas and techniques for “situation thinking”.

Some approaches are scientistic, meaning there is little or no evidence to verify or falsify them.

It easy to find problems in human organizations.

From the 1970s onwards, systems thinkers have claimed institutions are in crisis, and something must be done.

Surely, institutions will always have problems, and system thinking will never provide a final answer to those problems.

It will always be necessary to intervene now and then.

It is more difficult to propose viable changes, and more difficult again to make those changes.

Where an intervention involves describing regular activities, realizing them and testing the outcomes, it is an application of classical system theory.

But consultants may make interventions of others kinds – with little or nothing by way of process definition or testing of outcomes.

Even after a change has been made, it can be difficult to prove whether the change was for the better or not.

It isn’t always clear that the situation addressed is a “system” beyond being a social entity - a named organization or named group of actors.

If every problem or situation or social entity is called a “system”, then the word system adds nothing to our understanding.

A problem/situation might be a system, or it might not.

The solution might be the description, testing and implementation of a new of changed system, or it might not.

All free-to-read materials at http://avancier.website are paid for out of income from Avancier’s training courses and methods licences.

If you find the web site helpful, please spread the word and link to avancier.website in whichever social media you use..