Dialectical Systems Learning

During the past month or so I have been ruminating over my post of May 10, 2019, which was about my latest effort to come to a better understanding of the workings of a systems approach described in a workbook that I am co-writer of. Wicked Solutions, as it is called, uses three operable systems concepts to explain systems thinking in a nutshell and encourages learners to apply them directly on a ‘wicked’ problem of their own so as to gain a direct, hands-on experience of their usefulness. The three concepts are: inter-relationships, perspectives, and boundaries. Last week I had a discussion with two members of staff of Australia’s Southern Cross University, Ken Doust and Andrew Swan, who has used Wicked Solutions in one of his courses. They had several critical observations that set me thinking. One idea was to focus the dialectical systems approach of Wicked Solutions around problem/solution trees to get a handle on engineering cases. When I told a close friend about it, he was so readily and overwhelmingly enthusiastic that I set to work integrating the idea into the insights gained in my previous post, all the while trying to keep things simple (as opposed to the spaghetti dragon of last time). So here it is:

(This post has been syndicated by The Systems Community of Inquiry to https://stream.syscoi.com, the global network of systems thinkers, scientists and practitioners)

Principles and methods     Churchman used to insist on the proper use of principles, rather than on the use of some method. This is probable one of the reasons why his dialectical systems approach, although brilliant, never caught on. It could be argued (as I did in my last post) that the learning cycle of Checkland’s Soft Systems Methodology is in fact a method that applies a good many of Churchman’s principles. This would imply that Churchman’s principles can and perhaps even must be used to fully understand Checkland’s method. (In fact, the idea of the principles-method dichotomy emerged while reading about the Darwin-Wallace controversy in the book ‘Krakatoa’ of Simon Winchester, a reading suggestion by Ken: thanks again!). This is not unimportant, because Wicked Solutions is to some extent based on both, either directly or indirectly.

Human activity models        … is the term used by Checkland to distinguish interventions from social systems, which he calls ‘human activity systems’. It occurred to me that human activity models can take many forms. One of the most common forms is that of linear management (see my post of 2013). The problem/solution tree approach is an adaptation of that form. The good thing about that approach, from a training perspective, is that it runs the risk of imposing early limits on the problem/solution space (see another post of 2013). Both Churchman’s principle of non-separability and his environmental fallacy refer to this risk. This means that the problem/solution tree approach can be used as an excellent starting point to show the benefits of the systems approach, to explain the epistemological nature rich picturing, to emphasize the need for stakeholder engagement as a way to apply the principles of deception-perception and make sure that assumptions that restrain the scope of a more systemic solution are brought to light.

Five steps       … are all that is needed for dialectical learning: rich picturing, framing of stakes, boundary debate, model selection, and feasibility analysis. Framing of stakes and boundary debate (or critique) are missing in Checkland’s SSM activity pattern that reflect the steps in his SSM inquiring/learning cycle. When I say they are missing, that does not mean that they must be necessarily included. It is just that for learners it is easier to have them included. It has the additional advantage that the learners can be assessed for following these steps. In Wicked Solutions the last steps of model selection and feasibility analysis are lacking. There, too, it does not mean that they had to be included. Most of the systemic process can be carried out in the first three steps. But some learners will think it unsatisfactory that these steps are lacking (as I did myself). The framing step is Bob William’s idea, whereas the boundary debate in the form of critical heuristics is Werner Ulrich’s contribution. Critical heuristics is often carried out on its own or prior to SSM.

Inquiry and design       It may be useful to look at the 5-step process from the angles of inquiry and design. Churchman and his principles are at the inquiry end of the process, while Checkland and his method is more on the design end. The actual conceptual design of novel solutions takes place somewhere in the middle. The last steps are just a clarification of what has been found in the earlier steps. This clarification is more apt to a methodological approach, whereas the development of novel concepts is a much more suited to an open-minded, principles-driven approach. The marriage of Churchman’s principles with Checkland’s method seems rather obvious. Checkland circumvented this ‘problem’ by insisting on ‘rich picturing’ as the first step, a step that was sorely missing with Churchman.

Principles and concepts       Churchman’s principles are scattered among his books and papers. He attempted to summarize them in his categorical framework, which is applying the principle of categorical assurance (where categories are about the  lack of inter- and intra-categorical assurance) as described in Edgar Singer’s ‘Experience and Reflection’, which was posthumously edited by Churchman in the period between his years devoted to the development of operations research and his last 16 or 17 years before his retirement devoted to the development of his systems approach. Bob Williams was the instigator of a meeting in 2004 in Berkeley to simplify systems concepts in a way that non-specialists could handle them more readily. These concepts were inter-relationships, perspectives and boundaries. He has successfully applied them in a book that provides an overview of a wide range of systems methods, for which he received the American Evaluation Association’s Paul F Lazarsfeld Evaluation Theory Award (for Bob’s books, click here). It is fairly easy to explain Churchman’s principles in terms of the three basic Berkeley concepts. This will prepare them for Bob’s introductions to other sections of the systems field, including that of systemic evaluation design, thus turning the various writings in a neat whole. And the development of assuring, synergetic ‘wholes’ is what systems thinking is all about.

Problem/solution trees    The end result of learner’s following the dialectical systems learning approach (still to be fully described) will be one or more alternative problem/solution trees. There are various ways of processing these trees into more refined models, using other approaches. One of the other criticisms of the two Australians was that Wicked Solutions lacked guidelines for stakeholder selection and engagement. There is probably no short answer. Two things spring to mind: 1. SSM has quite a body of practice that has been described in articles and books, some of them indicating how stakeholders were organized and encouraged to co-operate. 2. the Royal Tropical Institute has had considerable success applying a method called RAAKS, which is to some extent based on SSM principles, but which provides a considerable range of tools to enable stakeholder collaboration.

About csl4d

https://www.linkedin.com/in/sjonvanthof/
This entry was posted in General. Bookmark the permalink.

1 Response to Dialectical Systems Learning

  1. antlerboy - Benjamin P Taylor says:

    Reblogged this on Systems Community of Inquiry.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s