At the First Global Conference on Research and Implementation, held in Canberra, Australia, in September 2013, Gerald Midgley gave an introductory talk on systems thinking. Midgley is a British organizational theorist, professor of systems thinking, director of the Centre for Systems Studies at the University of Hull, and past president (2013-14) of the International Society for the Systems Sciences. Other presidents of the ISSS that I have already written about, sometimes fleetingly, include Ashby (1962-64), Stafford Beer (1971), Mead (1972), Checkland (1986), Ackoff (1987), Churchman (1989), Mitroff (1992, pointed out to me by David Ing, himself president in 2011-12), Linstone (1993), Nelson (2000), and Jackson (2001). So now it’s the turn of Midgley, whom I know indirectly through my work with Bob William as a co-author of Wicked Solutions. Midgley’s talk is superbly concise (just 23 minutes if we leave out the introduction and the questions), so I decided to have a very close look at it, producing a roughly accurate transcript in the process. This post has been syndicated by The Systems Community of Inquiry at https://stream.syscoi.com, the global network of systems thinkers, scientists and practitioners.
Wicked problems Midgley indicates that the term ‘wicked problem’ continues to be popular among policy makers, which is good to know. Alternative terms are of course ‘mess’ (Ackoff), ‘complex problems’, and ‘ill-structured problems.’ Horst Rittel was the first to describe them in detail (see here or here), identifying 10 characteristics. About 5 years transpired between Rittel’s presentation of wicked problems at Berkeley in one of Churchman’s weekly seminars in 1967 and his first description of them in 1972. I take this long period to be indicative of the complexity of describing what wicked problems really are. Jeff Conklin, a student of Rittel I believe, reduced the 10 characteristics to 6. Midgley uses 5 bullets to describe them, which in his talk he summarizes in one sentence as: “many interlinked issues, cutting across silos, complexity, multiple agencies, multiple scales, multiple perspectives, conflict, power relations, uncertainty about the possible effects of action.”
Roles and perceptions I like Midgley’s short description so I turned it, with a number of minor changes, into a tiny concept map, which may require a brief explanation. The idea of entangled issues is expressed with the verb ‘mess up’, which is a reference to Ackoff’s term (‘messes’) for complex, wicked problems. I use actors in stead of agencies, because it is shorter and because there can be many different actors within agencies. It also refers to the categories of role distinguished by Churchman since actors play roles, e.g. those of client, decision-maker, planner and systems thinker. It can be very illuminating to think of individual actors as playing multiple roles. It is a well-known fact that many decision-makers do not overlook their own interests. Alternative terms for actors are stakeholders, agents etc. Churchman uses the role category of ‘client’ not only for beneficiaries, but also for victims, who are negative beneficiaries as it were. So the term ‘actors’ points at conflicts, that are played out by way of multiple perspectives. Different actors perceive differently the same options that may address the issues at hand, or claim to do so. The double meaning of ‘partially’ should not escape our attention. Churchman uses the term deception-perception for several basic principles of systems thinking. Deception makes any perception uncertain, subject to doubt, demanding judgment as to the credibility of evidence, while partiality makes it critically so. This type of judgment or critique is at the core of real critical thinking, without which actors are less likely to reach concordance on creative solutions or innovative improvements. A final note: silos and scales should be understood in the broadest possible sense, so for instance scales could be taken to refer to systems and sub-subsystems and so on. This is the embedding principle of Churchman (see this post). Silos here mean systems, processes, government departments, business units etc. that operate in relative isolation from others.
Systems thinking …, what is it? Midgley writes: “thinking uses language, language is socially shared, so actually dialogue helps us think. This is not some naïve notion of thinking that goes on in your mind. Thinking is both personal and collective.” This concurs with Churchman, who writes that systems thinking is a dialectical process (see also here) in the most general and fundamental sense. So any systems method that strengthens this process may be termed a dialectical systems heuristic. Midgley identifies four systems thinking skills, which he traces to Derek Cabrera, an evolutionary epistemologist born in 1970: system, boundary (in the vocabulary of Cabrera, 2008, ‘distinction’), relationship and perspective. All of these can be traced to Churchman’s work in the 1950s.
Boundary …. is a special case. Churchman (1979, p. 32) claims that “it’s quite likely that the tradition of the systems approach goes back to primitive man. […] He was on his way to an early beginning of the systems approach, because he had done his very best to think about the “whole system” and specifically about its components, its boundaries, and his decision making relative to it.” Churchman goes on to see parallels in the I Ching, Buddhism, the pre-Socratics etc. What later became the boundary critique is what Churchman originally called the unfolding of the categories of the components (with resources) and the environment (aspects of reality that the decision-maker cannot – or will not – change). The boundary is the distinction (ah, Cabrera’s term) between the two. Churchman (1979, p. 91) concludes that “from the point of view of ideal planning, the question of the proper boundaries has no plausible, common-sense answer.” Hence the notion of boundary critique by Ulrich and Midgley. What is clear from this little discussion is how systems thinkers manage to confuse matters by emphasizing aspects of the same thing: environment (Churchman, 1967/68), boundaries (Ulrich, 1980s), distinction (Cabrera, 2008) all refer to the same thing. And they all suffer of the same problem fundamental to systems thinking, namely confusion and enlightenment, deception-perception (Churchman 1968, p. 231). The real problem is that by taking one meaning or interpretation on board, we throw a whole load of good other insights overboard. This, in turn, suggests that we may need different systems approaches to different problems. Midgley’s talk is about that, too.
Boundaries and values There is one neat diagram that Midgely uses to explain the idea of boundary. And “we have a partial view [of the situation] in two senses: a. we can’t see the whole, never. And b. values are key in helping us draw boundaries. So it’s partial in the sense that it is value-driven. The ellipse in the diagram is the boundary between included and excluded stakeholders and issues. The peak represents values and there is a two-way relationship between values and boundaries. The values that you bring into an arena of action will help drawing meaningful boundaries. Values are not general principles, they are linked to our personal goals (read that again!). But some boundaries are already given by institutions, and those boundaries constrain the type of values that may be expressed. Hence the two-way relationship. Through exploring the value judgments as well as the boundary judgments and looking at different possible boundaries that you get a more comprehensive understanding of the situation.”
Systems methods Midgley gives four examples of the application of different systems methods to real situations. Together they show what systems thinking is. I summarized this again in a tiny concept map. What is perhaps not clear from this picture is that it is the boundary critique and the appreciation (and possibly realignment) of conflicting or contrasting perspectives that power the rearrangement of complex relationships with a view to create or maintain viable, sustainable, responsive, innovative systems. Three systems methods are dealt with in some detail: causal loop diagrams (system dynamics?), viable systems model and soft systems methodology. Midgley didn’t mention critical heuristics, presumably because it is much better for inquiry than for design. In Midgley’s magnum opus, Systemic Intervention, which sits on my bookshelves.
Questions There was room for three questions at the end of the talk, the last two of which I would like to highlight. Question 2 was about the problem of training ministerial staff and members of the professional class in systems thinking. The first reason why they are not trained is, clearly, because they have been trained in another profession and never got acquainted with wicked problems, only solvable, professional problems. The second is that in order to appreciate systems thinking, one has to do it. It is not something one can read about, and think: oh, that’s good, and forget about it. Besides many will be convinced that they do things systemically already, which is true in a way, but the results are totally different when the right systems approaches are applied more or less professionally. It’s only after doing it, that they realize there is whole lot more to it than they realized. Question 3 is about the question why decision-makers don’t use system methods, even when they are convinced that it would be useful? The first reason is that “people cling on to their boundaries because actually it is useful for them in some way. So when you are asking them to challenge those, you are really threatening the possibility that they might lose the ability to control what they’ve got.” Some people are very naïve about how they think they can apply systems methods, e.g. in the form of stakeholder engagement to critique boundaries. Certain assumptions or pre-framings prevent them from achieving an appreciable level of effectiveness. For more information, click here.