My learnings from the course Systems Thinking 2.


How to approach complexity?

What we’re dealing with

When dealing with issues of design, planning, and organizing, we have to face both complicated and complex problems. A complicated problem is something that might be difficult, but still understandable and predictable. For example, building a watch or a bridge might be very difficult, but still something that we can understand and deal with careful planning. However, many problems are complex, meaning that they are difficult to understand and explain thoroughly. These kinds of problems, also known as messes (Ackoff, 1974) or wicked problems (Rittel & Webber, 1973) require different kinds of approaches and thinking than what we’re normally using in our daily lives.

Where do these problems arise? In an earlier blog post I briefly discussed the concept of SOHO systems (Kay et al, 1999, Kay and Schneider, 1994). SOHO is an acronym for Self-organizing, Holarchic, Open systems. Let’s take a closer look:

  1. Self-organizing means that when the system is pushed from its equilibrium, it might exhibit spontaneous coherent behavior and organization.
  2. Holarchic means that the system is formed up of part-wholes, i.e. holons, and is itself a part-whole. These part-wholes have dynamic interactions both horizontally and vertically across different scales of space and time.
  3. Open means that the system exchanges matter and energy between its environment.

SOHO systems are therefore dynamically relating part-wholes where non-linear feedback loops result in self-organization at different scales in the holarchy. What’s more, when these SOHO systems receive energy from their environment, they develop new structures and processes that make them more effective at receiving energy from their environment.

Ecosystems and human activity systems are prime examples of SOHO systems, both of which exhibit spontaneous coherent behavior and organization, are formed up of part-wholes, and are open. The messes that we face are the result of these dynamic interactions: as we have changed the environment we live in, we have created new and unpredictable changes in ecosystems.

What to do then? When we can’t predict our environment, we need to be able to coevolve with it, and this is where the concept of resiliency comes in.

Design systems for resiliency and learning

First, what is resiliency? From what I’ve learned, there’s at least two views on what resiliency means. The first, engineering approach, says that a resilient system is one that can take on outside shocks and quickly return to equilibrium. The other view, social-ecological resilience, says that a resilient system doesn’t have only one equilibrium, but instead can shift between different states, and learn, change and adapt.

In a wonderful article from 2006, Carl Folke had this to say about social-ecological resilience:

“Adaptive processes that relate to the capacity to tolerate and deal with change emerge out of the system’s self-organization. Furthermore, the dynamics after a disturbance or even a regime shift is crucially dependent on the self-organizing capacity of the complex adaptive system and the self-organizing process draws on temporal and spatial scales above and below the system in focus.”

Without going too much into details, the way I understand Folke is that resilient systems have the capacity to self-organize and create new structures and processes after a disturbance. Moreover, this capacity to self-organize is not a characteristic of one scale in time and space, but is derived from scales both above and below the system, as well as from different scales in time.

I think this point about different temporal and spatial scales is absolutely key here. The way we usually design human systems is by taking a look at one scale or holon at a time. When doing organizational design, you don’t usually start designing the whole that contains the organization. However, based on what Folke said, the capacity to self-organize is not contained at one level of time and space, but draws from all the other scales as well. This is why it’s about social-ecological resiliency, not only human resiliency. Building resiliency in one level will not be enough, and is in fact an illusion.

This also poses a major challenge to planning and design. How to design for resiliency when resiliency is not dependent on any one system? How can we ever build the capacity to self-organize when we can only affect a small part of the whole system at a time?

It would sound that enabling resiliency in our systems requires a paradigm shift at all levels of design. To enable resiliency, we need to change the organizing principles at all scales of social-ecological organization. How exactly are we going to get that done will have to be left for another discussion.


Ackoff, R. (1974). Systems, messes, and interactive planning. Portions of chapters 1 and 2 of Redesigning the future. New York / London. Wiley.

Folke, C. (2006). Resilience: The emergence of a perspective for social–ecological systems analyses. Global Environmental Change. Vol. 16. Pages 253–267.

Kay, J., Regier, H., Boyle, M., Francis, G. (1999). An ecosystem approach for sustainability: addressing the challenge of complexity. Futures. Vol. 31. Pages 721-742.

Rittel, H., Webber, M. (1973). Dilemmas in a general theory of planning. Policy Sciences. Vol 4. Pages 155-169.

Schneider, ED. & Kay, JJ. (1994). Complexity and thermodynamics: towards a new ecology. Futures. Vol. 19. Pages 25-48.




A brief discussion about boundaries and perspectives

Boundary judgments and the Hierarchy theory

In the second session of Systems Thinking 2 course at Aalto University we had an interesting discussion about boundaries and perspectives. One particular remark was made about boundaries, power, and decision-making. In all decision-making settings it’s important to be able to identify who gets to set the boundaries for decision-making. Often, especially in politics, people who have the most power are the ones who get to frame or reframe issues. This is important to understand because the problem-definitions we choose inevitably limit the different alternatives we consider for solving the issue. All kinds of questions are being answered for us when we allow someone else to set our boundaries, including who, what, why, when, and how?

The importance of boundary-setting and framing becomes even more evident when dealing with complex systems, and I dare to claim that most political issues are concerned with complexity. In complex systems, the dynamic interactions between the parts of the system and between the system and its environment make it difficult to predict the outcomes of our actions. It’s anything but easy to decide which interactions to consider in our decisions and which to leave out. So, when we allow others to define the problem space, not only are we placing a lot of faith into their hands, but we also give up much of our decision-making power.

The link between boundary-setting and decision-making is explained well in Hierarchy theory. In Mario Giampietro’s (1994) words:

“It is commonplace to experience a discrepancy in values when assessing the same phenomenon or action from different perspectives. For example, what is good to our taste (assessment on a short timescale) may be harmful to our health (assessment on a longer timescale); the lifestyle of singles is changed after marriage when the same individual becomes part of a larger structure, the family. This pattern is repeated at each enlargement of perspective that brings another hierarchical level into the picture: what is profitable for the family – less taxes – may be harmful to the community to which the family belongs.”

Where are you setting your boundaries?

Where are you setting your boundaries?

This means that decisions at one point in time and scale may affect either negatively or positively the situations in another point in time and/or scale. When we set the boundaries for decision-making at one scale, e.g. the economy, other scales and time-frames might be left out. We could also be neglecting other values and worldviews if we’re not careful.

However, change of boundaries might not lead to changes in perspective. Even when we point out that short-term changes in our economy might in some cases lead to degrading environment (and degrading economy in the long-term), we might still get stuck on the short-term economic perspective. A person might still view that improving the economy creates enough benefits in the short-term to justify lack of consideration for the long-term. In the end, we also have boundaries to our rationality.

Complexity as a matter of perspective

We also had a brief discussion about the nature of complexity. One of our teachers, David Ing, brought up a thought by Timothy F Allen, who had stated that complexity isn’t necessarily an innate characteristic of a system, but rather a matter of perspective. For example, the Apple iPhone might seem complex (or complicated) on the inside, but is relatively simple to use on the outside. A system can therefore be perceived simultaneously as both complex and simple, depending on your perspective.

Let’s take a closer look at the Apple ecosystem. I don’t have an engineering background, but from what I’ve understood, all Apple products taken together basically form an integral system architecture as opposed to a modular one. In a modular system or product architecture the different components of the system have high independence from each other. Thanks to this high independence, components of the system can be relatively easily disassembled and recombined into new configurations, and new product variants can be realized without much difficulty because changes to one component doesn’t lead to changes in other components (Voss & Hsuan, 2011.)

Lego bricks allow modular design.

Lego bricks allow modular design.

The opposite of modular system or product design is an integral design, which I think Apple has chosen for its system. With an integral architecture, components of the system or product are tightly coupled, meaning that modifications to one component require redesigning or re-configuring other components. Apple’s iPhone, Macbooks, the iPod and all other products are all highly integrated to each other. To my understanding Steve Jobs felt very strongly about creating a seamless user experience for Apple products, which he ensured by retaining a strong control on any product modifications. You can’t even change the battery of your iPhone without expert help, let alone start making modifications to the design. Even the chargers to Apple products are different from other manufacturers who mostly use universal chargers.

However, although Apple’s products form an integral system architecture from the perspective of the underlying hardware and software, the Apple Appstore is far from integral. Millions of applications have been created by app-developers all around the world for the Apple iPhone and iPad devices. I think that is an extremely interesting design choice, whether consciously done or not! Ensure efficiency and reliability where it counts (hardware and basic software), but allow variety where people really want it (applications).

So, from one point of view the Apple system is complex, from another simple.


Giampietro, M. (1994). Using hierarchy theory to explore the concept of sustainable development. Futures. Vol. 26, No. 6. Pages 616-625.

Voss, C. & Hsuan, J. (2011). Service science: The opportunity to re-think what we know about service design. In Dermikan, H. (Editor), Spohrer, J. (Editor), Krishna, V. (Editor). The Science of service systems (Pages 231-243.  Springer. New York, USA

Creative Commons Lego Bricks by Benjamin Esham are licensed under CC BY-SA 2.0.

Sense-making in the systems movement – observations of a novice

During the course Systems Thinking 2 at Aalto University we have already had the opportunity to explore different views on how systems thinking can be used in organizational sense-making and design. So far we’ve read and discussed articles about Ackoff’s Interactive Planning, Vicker’s Appreciative Systems, and Haeckel’s Sense-and-Respond organisation. All these views try to bring light on how organizational sense-making can occur, and on how to design systems that can deal with complexity.

What interests me in particular are the different assumptions and broader worldviews that the different approaches hold. The systems movement is a rather complex phenomenon in itself, and for a novice like me, seeing how these approaches relate to each other and the larger context is difficult at first. However, in this blog post I will try to make some sense of the systems movement and explore two major world views that I have recently come across in my readings.

According to the first view, the world is systemic, meaning that it’s formed of interconnected systems. We can objectively observe and design these systems by applying systems thinking principles. This view is based on positivism, spectator theory of knowledge (Dewey, 1929), and functionalism (Zexian & Xuhui, 2010). The second view dismisses the notion that world is essentially systemic and that we can objectively observe it. Instead, this world view builds on social constructionism, the interpretive paradigm, and Dewey’s (1929) experimental theory of knowledge (ibid.) The advocates of the second view argue that the process of inquiry is systemic, and that systems should be viewed ‘as if’ they existed in the real world.

Below is a more detailed discussion of both views, as I have understood them.

The world is systemic – first order (hard) systems thinking

The shift from the doctrine of reductionism and the analytical mode of thought to the doctrine of expansionism and the synthetic mode of thought that took place in the early decades of the 20th century brought with it several lines of inquiry into systems. According to Russell Ackoff (1974), the shift itself began with different scholars in separate fields making a move away from reductionism towards more expansionist thinking. For example, Suzanne Langer discussed the meaning of symbols in the 1940s, with Charles Morris later building on her work to study languages in late 40s and early 50s. From languages the next step was communication by Claude Shannon in 1949 and control and cybernetics by Norbert Wiener in 1948. According to Ackoff (1974), the final “Aha” moment that launched the systems movement was Ludwig von Bertalanffy’s General Systems Theory in the 50s and 60s.

Russell Ackoff.

Russell Ackoff.

These developments lead to the incubation of three distinct, but related systems fields: General Systems Theory, Cybernetics, and Systems Dynamics. All three strands of systems thinking departed from reductionist approaches in that they emphasized the importance of dynamic interactions between the parts of the system and between systems and their environments (Stacey, 2010, 201.) Problem solving did no longer begin by isolating the problem from its environment, but by looking at how the problem is connected to the larger whole that it’s a part of. Synthesis would now precede analysis, instead of the other way around (Ackoff, 1981).

The strands of systems thinking that emerged in these early decades of the systems movement are today called first order systems thinking, or hard systems thinking (Stacey, 2010; Zexian & Xuhui, 2010). Although the proponents of hard systems thinking dismissed the reductionist mode of thought and analytical thinking that formed the basis of the scientific method, hard systems thinking still held many of the beliefs behind reductionist thinking. According to Ralph Stacey (2010), hard systems thinking assumes an objective reality that can be rationally observed by individuals. When it comes to social systems, the social world is essentially assumed to be formed up of systems that have a purpose, and that can be objectively observed and modelled (ibid). The assumptions behind hard systems thinking were conveniently summarized in a 2010 paper by Zexian and Xuhui:

  • System objectively exist in our world and it has a good structure and identified goal.
  • The parts of the system have the same worldviews, values and interests.
  • The system intervener is an outsider of system and is not influenced by the system.
  • Achieving the optimal results is the ultimate goal of problem-solving process (Zexian & Xuhui, 2010, 143.)

According to Zexian and Zuhui (2010), hard systems thinking conforms to positivism in natural science and largely ignores the diverse worldviews, values and interests existing in human organization. Furthermore, hard systems thinking complies with the tradition of epistemology that ignores the relationship between the subject and the object, which Dewey (1929) called the spectator theory of knowledge (ibid.)

In summary, while hard systems thinking moved away from reductionism, i.e., observing and designing parts of a system in isolation, the world view still held on to many of the assumptions behind natural sciences. General Systems Theory, Cybernetics, and Systems Dynamics all assumed that systems could be modelled and understood objectively. Design and sense-making were therefore only a matter of patience and use of rational decision-making tools. Although synthesis would precede analysis, the design of systems could be done using the same scientific rigor that natural scientists used when analyzing natural phenomena.

Emergent behavior. A starling flock near Athens.

Emergent behavior. A starling flock near Athens.


The process of inquiry is systemic – Second order (soft) systems thinking

To recap, General Systems Theory, Cybernetics, and Systems Dynamics all assumed that systems exist as objective phenomena, and that social systems have identifiable goals, structures, and behaviors that we can evaluate and design. In the 1970s and 80s, however, systems thinking scholars began to question these assumptions. Among the most notable critiques were West Churchman (boundaries and moral), Russel Ackoff (Interactive Planning), and Peter Checkland (Soft Systems Methodology), who developed alternative approaches to organizational sense-making and design that involved people. Later the Critical Systems Thinking approach was built on top of the critique from Chruchman, Ackoff, and Checkland (Stacey, 2010.)

Zexian and Xuhui (2010) view Checkland’s Soft Systems Thinking in particular as a major milestone in the systems thinking movement. Checkland critiqued the positivist nature of the earlier systems thinking approaches as well as noting that they don’t consider different human values and worldviews in their analyses. He also dismissed the word ‘system’ altogether and instead employed the term ‘purposeful holon’ to discuss human systems. Below is Checkland’s systems thinking summarized in seven points:

  • System thinking takes seriously the idea of a whole entity which may exhibit properties as a single whole (‘emergence properties’), properties which have no meaning in terms of the parts of the whole
  • To do systems thinking is to set some constructed abstract wholes against the perceived real world in order to learn about it
  • Within system thinking there are two complementary traditions. The ‘hard’ tradition takes the world to be systemic; the ‘soft’ tradition creates the process of enquiry as a system.
  • SSM is a systemic process of enquiry which happens to make use of system models. It thus subsumes the hard approach, which is a special case of it.
  • To make the above clear it would be better to use the word ‘holon’ for the constructed abstract wholes, conceding the word ‘system’ to everyday language and not trying to use it as a technical term
  • SSM uses a particular kind of holon, namely the so-called ‘human activity system’. This is a set of activities so connected as to make a purposeful whole, constructed to meet the requirement of the core system image (emergence properties, layered structure, process of communication and control)
  • In examining real-world situations characterized by purposeful action, there will never be only one relevant holon. It is necessary to create several models of human activity systems and to debate and so learn their relevance to real life (Checkland, 1990, 27).

Checkland therefore states that there is no objective reality that can be observed from the outside, and neither is there only one optimal system (holon) for any situation. If I understood correctly, this is strongly against the first order systems thinking tradition of the 50s and 60s.

In short, the second order systems thinking approaches that were developed in the later decades of the 20th century dismissed the notion of rational observers objectively assessing reality and designing ideal systems based on objective goals. The idea of a system was questioned altogether and replaced by the word ‘holon’. Later, during the 80s and 90s the theories of catastrophe, chaos, and complexity would be added to the already broad spectrum of systems theories and sciences. Exploring the contributions of complexity theories is widely beyond the scope of this blog post, so I will leave the realm of complexity for another discussion.

So, coming back to Vickers’ Appreciative System, Ackoff’s Interactive Planning, and Hacekel’s Sense-and-Respond organisation, I feel it’s already a bit more clear where they stand in the bigger picture. To my knowledge, Ackoff’s critique towards hard systems thinking acted as one of the foundations towards soft systems thinking. Vickers’ Appreciative System method came about as a critique towards the rational decision making models that he saw to have little bearing on how real world works (Burt & Van der Heijden, 2008, 1111). It would therefore seem like a safe bet to say that Vickers also represents second-order systems thinking. I would dare to say that Haeckel too represents second-order systems thinking. In his book ‘Adaptive Enterprise: Creating and Leading Sense-and-Respond Organizations’ (1999), Haeckel builds his idea of a Sense-and-Respond organisation on Learning Organisation theory and Complex Adaptive System (CAS) theory. Learning Organisation theory emerged along with other second-order systems thinking theories, while CAS theory is part of the complexity sciences family.


Ackoff, R. (1974). Systems, messes and interactive planning. Portions of chapters 1 and 2 of Redesigning the Future. New York/London. Wiley, 1974.

Ackoff RL. (1981). Creating the Corporate Future: Plan or Be Planned For. John Wiley and Sons, New York. Pages 16-17.

Burt, G. & van der Heijden, K. (2008). Towards a framework to understand purpose in futures studies: the role of Vickers’ appreciative system. Technological Froecasting & Social Change. Vol 75. Pages 1109-1127.

Checkland, P. (1990). Soft systems methodology in action. Wiley. Chichester, UK.

Dewey, J. (1929). The Quest for uncertainty: a study of the relation of the knowledge and action. Balch & Company. New York, USA.

Haeckel, S. (1999). Adaptive enterprise: creating and leading sense-and-respond organizations. Harvard Business School Press. Boston Massachusetts.

Stacey, R. (2010). Strategic management and organisational dynamics: The challenge of complexity. Pearson Education Limited. Edinburgh Gate, England. Pages 54-55, 201.

Zexian, Y. & Xuhui, Y. (2010). A Revolution in the field of systems thinking – a review of Checkland’s systems thinking. Systems Research and Behavioral Science. Vol 27. Pages 140-155.

Creative Commons Starlings near Athens Nov 2008 by muffin is licensed under CC BY 2.0.