Coevolving Innovations

… in Business Organizations and Information Technologies


Systems Thinking and Futures Studies (Systems Thinking Ontario, 2013-02-21)

Posted on March 21, 2013 by daviding

The pre-reading of Emery (1967), “The Next Thirty Years: Concepts, Methods and Anticipations” was introduced as a challenging article for the second meeting of Systems Thinking Ontario on Feb. 21, 2013.  The theme for the evening was “Systems Thinking and Future Studies”, so there was some irony in looking backwards to 1967 to have a discussion on looking forward.

In my role as reviewer in Systems Thinking Ontario sessions, I would prefer to try to stick to the text rather than adding editorializing.  However, since this Emery (1967) article is particular rich, I tried to provide some additional context to make the reading easier.

Fred Emery is especially known for his work with the Tavistock Institute for Human Relations, as one of the pioneers of the field we know today as organization science, including organization development and organization design.  In 1967, systems thinking was new:  The Systems Approach would be published by West Churchman in 1968, and On Purposeful Systems by Ackoff and Emery would follow years later in 1972.  The Club of Rome was founded in 1968.  In this article, Emery was thinking about how systems thinking frames viewing the future.

I. Prediction and Planning; II. Conceptual Bases for Predicting the Future

In I. Prediction and Planning, the systems of interest are in the social sciences.  The longer history of systems thinking prior to 1967 would have been a stronger cybernetics orientation.  The research from the Tavistock Institute on the socio-psychological, socio-technical and socio-ecological systems perspectives would have been fresh.  Emery wrote:

In this section we have sought to argue that:

  • (a) there is a need for developments in the social sciences that go beyond their present concerns;
  • (b) this development needs planning;
  • (c) the planning needs to be in a context of expected social developments for several decades ahead;
  • (d) the planning should be more than projection or forecasting;
  • (e) planning should actively seek to extend the choices men can make, not to dictate them.  [p. 199]

Essentially, the challenge is that human beings can shape their futures, and not just be passive participants in the changes.  Much of the influence that human beings have on the future, particularly when working collectively as a social group, is through planning.

From the article, the figures in II. Conceptual Bases for Predicting the Future were helpful towards deciphering the text.

Emery (1996), Figure 1

Figure 1: E == environmental conditions; R == (system) response

In Figure 1, the diagrams might be read bottom-up, from the past (at time t0) upwards to the future (to time t1 and t2).  In the left diamond, some environment conditional Et0 happens, and adaptation occurs either another change in environmental condition Et1, or system response Rt1 to get to the final goal at t2. In the right diamond, consider instead the possibility of reading from the top-down.  To get to the goal at t2, the starting point could be an environmental condition Et0, as before.  On the other hand, there’s also the option that the system could (pre-emptively or anticipatorily) respond with Rt0.

I’ve struggled with these diamonds before.  In the 1969 Penguin paperback edited by Emery, there’s a reference to Sommerhoff (1969), “The abstract characteristics of living systems”.  The story speaks about playing football (soccer) and wanting to score a goal.  Essentially, in the left diamond, the player can wait for the ball to come to him, and then kick it towards the goal.  In the right diamond, the player realizes that he could also move towards where the ball is currently positioned, and that might help him score the goal.

In later writing, Ackoff differentiates between adaptation (as passive) and learning (as potentially active).

Emery (1967) Figure 2

Figure 2: Two living processes, can better predict A at t+ than B and t+

Figure 2 brings in some systems language, in process, which can be described as an arrangement in time (as opposed to structure, which is an arrangement in space).  Three living processes are presented.

A has quite a lot of history, and thus provides some foundations from which to predict.

B has some history, giving some foundation for prediction, although probably with less validity than A.

C has no history, and thus would be difficult to predict in the future.  How should we handle that?

Framing a view into the future, in this example of three living processes, leads into questions that are familiar to systems thinkers:  what is the system of interest, and where do we draw the boundary?  Boundary can be be considered both in the structural sense (arrangement in space) and processual space (arrangement in time).

Emery (1967) Figure 3

Figure 3: a and b are phases of A, no basis for predicting B

Figure 3 deepens the challenge of time, when changes don’t occur linearly, but instead cyclically.  “While some prediction about the future part of a is theoretically possible, there is no basis for predicting the specific characteristics of phase b” [pp. 204-205].  Learning about A and appreciating that as a cycle may or may not give insight into B.  B might similarly be a cycle, or a cycle of different periodicity, or not a cycle at all.

Consider the problems that would have occurred when astronomers watched points of light in the sky move steadily one direction, and then reverse in direction.  Thinking about the future could lead the scientist to think about the prior patterns as recurring.  The system of interest could be more than just the object itself.

Emery (1967) Figure 4

Figure 4: A and B are coextensive in time, B is part process of A, so A easier to predict than B

Figure 4 complicates predicting the future by considering systems where there are whole-part relations.

A and B are coextensive in time but B is a part process of A.  One would expect that predictions about A would theoretically be easier than predictions about B.  The basis for this expectation is the general property of part-whole relations.  A sets some of the parameters of B and hence, whatever one knows of the values likely to be taken by B, one knows more if one knows how these parameters might change.  The future of B is dependent upon the future of A in a way that A is not dependent upon B.  At the same time, predictions about A will be less specific than could be predictions about B.  [p. 205]

Novices might make sense of futures with part-part interactions, but deeper thinking could uncover part-whole interactions.

Emery (1967) Figure 5

Figure 5: Two processes will interact at t+; A and B survive, or larger inclusive system?

Figure 5 covers the case where there might be part-part interactions, but then complicates the matter by opening up the possibility that there could be a new emergent whole from that interaction.

… two processes which are presumed to interact after some point t+ in the futue.  If A and B survive the interaction, some of their system properties may predictably survive.  What seems unpredictable are the processes set up by the interaction and the changes occuring in A and B if they become directively correlated to form a larger containing system.  [p. 205]

An appreciation of emergence is the true mark of a systems thinker.  No matter how much one studies the hydrogen and oxygen as parts of water, the property of wetness occurs in the whole (i.e. water) and not in the parts (i.e. hydrogen and oxygen).

Having covered these conceptual bases above, Emery turned to III. Methodologies for Predicting the Future.

III. Methodologies for Predicting the Future

At the session, the shortness of time led to a quick scan over this content:  essentially, the concepts and language of systems thinking can reshape the way that the future is considered.

  • 1. Two aspects of the general methodological problem:
    • (a) to identify the system in terms of its members and the dimension in which they are arranged;
      • (structural constraints, requisites, operational systems, codes –> persistence or decay)
    • (b) to identify the characteristic generation function of the system.
      • (phases; trends and forecasts as aspects of a system without relation to behaviour of the whole system)
  • 2. Social methodological difficulties that arise with predicting the future of a large complex social system:
    • (a) complexity; (b) early detection
  • 3. Methods that have been enveloped or proposed for overcoming these difficulties
    • (a) Complexity
      • 1. Ashby’s model for studying conditions for survival;
      • 2. method of identifying “the leading part”:
      • 3. models for studying subordinate goals (values) e.g. Ackoff-Churchman, Cantril;
      • 4. models for studying the starting conditions for change (coenetic variables).
    • (b) Early detection
      • 1. model derived from the properties of weakly competing systems (Angyal) (i) weak emerging system –> parasitic in symptoms of debility, (ii) stronger emerging system –> symptoms of intrusion, (iii) emerging and existing system in balance –> mutual invasion;
      • 2. sigmoid growth models;
      • 3. models based on analysis of symbols, values, and linguistic usage.

The reference to Angyal was worth expanding,  Angyal (1941), “A logic of systems” is also excerpted in the 1969 Penguin paperback edited by Emery.  The reading of Angyal had previously been found so overlooked that a session was convened at ISSS San Jose 2012 called “Human Systems are Different: Andras Angyal via Eric Trist” where we solicited the help of some former University of Pennsylvania Social Systems Science program graduates (David Hawk and Rafael Ramirez) to make sense of part-whole and whole-whole relations.

As a brief tangent from the Emery (1967) reading, I drew attention to the importance of environment — in particular, the causal texture of social environments published by Emery and Trist in 1965.

The Causal Texture of Social Environments

This quick summary was based on some prior research on turbulent environments, supported by Chapter 2, “Historical and Conceptual Overview” by Ramirez, Selsky and van der Heijden (2008).

Two commenters provided their views on the prereading.

Jonathan Resnick reflected on the Emery (1967) in the larger context of the focus question “Where do systems thinking and futures studies (i) intersect, and (ii) diverge?”.

  • In comparing future studies with systems thinking …
    • futures studies deals more with concrete entities like STEEP (social, technological, economical, environmental, and political) trends as ways to imagine and anticipate what the future might look like, and less on intervention; whereas
    • systems thinking (at least by Emery (1967) deals with more abstract entities (e.g. turbulent fields) and potential interventions in the system to produce desired outcomes.
  • Despite these differences, there was some parallels …
    • between the adaptation <–> transformation continuum in Emery (1965); and
    • the empirical <–> constructivist continuum in futures studies, in that
    • the empirical school treats the environment as a given that must anticipated and adapted to, while
    • more constructivist futurists are inclined to treat the environment as a construction and hence transformable.

Jeremy Bowes offered the second commentary, in two questions:

(1) What have others said since the Emery (1967) was published?

  • Over 35 years later, the engagement of stakeholders to take part in the process of co-creation and designing of social systems has been recognized.
  • There’s been reinforcement by Peter Senge (i.e. The Fifth Discipline) with approaches to organizational management and co-creation, shared worldview and shared values.
  • There’s (still) a need for a systemic approach to discovery and design to complex problems, (of which the Bela H. Banathy social systems process is one).

(2) How does Emery (1967) fit with my thinking on the theme of systems thinking and futures?

  • Identifying the “leading part” and key element to reduce, and understand a complex system is important.
  • Symptoms of debility and intrusion provide characteristics of (emerging) dominant behaviour, so they should be watched.
  • The article shows early signals of some key aspects of methodology generalities: symbols analysis, value analysis, linguistic analysis.
  • The concept of “causal texture” is interesting, as it helps to establish relationships between parts, or at least the co-presence of elements.
  • In the idea of adaptation to turbulent environments, W. Ross Ashby had some ideas about downgrading complexity through segmentation, fractionalization or dissociation.  Is this compatible or different from Emery’s view?
  • For human beings in social systems, shared values are the most significant of self-regulating devices, enabling the overlap of structures and sharing of elements.
  • Design thinking can be a backbone of interrelated processes for systemic design, and for design in social systems.
  • Shared visions and values might be created through gigamapping, foresight techniques, and scenario building as methods for systems thinking in complex systems.

The group then broke out into three parallel discussion breakouts.

di_20130221_193548_st-on_group_al

Allenna Leonard reported that their group had several first-time participants, so the discussion was more on the overall theme than on academic details.

  • An early observation and continuing thread was the male-centric (some said sexist) language and assumptions in the article.
    • Since one (or two) had familiarity with the profile of feminism under communism, it was pointed out that women were encouraged to put the kitchen in second place in favour of serving the state and choosing mates on the basis of patriotism.
    • The munitions factory in Toronto and its record of high quality (portrayed in Bomb Girls) was put forward as an example of how necessity can and does drive ideology on these topics. And, of course, after the war, the women were fired and sent back to home and hearth, so that men could resume their role as peacetime breadwinners.
  • The question of different variations of unpredictability when a change occurred came up with Libya as an example of a predictable (removal of a lid from a boiling pot leading to explosive consequences), as compared to an unpredictable (where and when the system comes to rest afterwards).
    • The categories of change from static to turbulent were seen as somewhat mechanistic and also likely to be occurring simultaneously in different parts of the system.  (This discussion might have led to a comparison with the previous reading and the ‘skeleton of science’ categories but most of the group had not been at that meeting).
  • “Time Stop” was mentioned about the butterfly effect when someone travels in time and disobeys the order not to touch anything and brings back a flower. Much difficulty getting back and returning it.
  • Many examples of unintended consequences in Freakonomics.
  • The later half of the discussion focused on employers’ lack of long term investment in employees as they assume they will move on quickly.
  • Twenty or thirty years out seemed a long time to think ahead (Elliot Jacques notwithstanding) when results are measured quarter by quarter.
    • There are too many disincentives to long term investment in anything.

di_20130221_193616_st-on_group_ps

Peter Scott reported that his group started with Emery (1967), and wandered from there.

  • The article postulated on two key concepts that relates to intersection and divergence.
    • (a) Adaptation Approach:  adapting the environment, in a “passive” way
    • (b) Directive Correlation:  a proactive approach, in a “active” way
  • These led to many questions within the groups around them
    • What are future studies?
    • How do we use the present to forecast into the future?
    • Should everything we do, be look at through system thinking?
  • Systems are constantly changing.
    • Modeling and mapping tools can be key ways to aid the studies of future thinking for a better understanding.
    • There are limitations with empirical approaches to understanding the boundaries of complex systems.
    • The term “systems” can often be a confusing one, easily misused and misunderstood.
      • Examples:  healthcare systems; a car has a system of its own; “ecosystems”.
      • Different things all together, but still related?
      • How do we distinguish boundaries?
  • Systems can be viewed independently, though.
    • The game of tennis (e.g. Serena Williams) has its own ecosystem.
      • Locations for play
      • Temperature
      • Athlete (strength)
      • Racket technology
      • Techniques / skills
    • Walkerton Water Crisis (confluence of different forces)
      • Management (poor skill)
      • Surrounding farm land impact
      • Non chlorinated water
      • High rain season
      • Governmental silo department
  • Why are we studying systems thinking? Should we take a step back and think about this?
    • This implies a philosophical reflection, perhaps on the fact that man is still preoccupied with the notion of controlling the environment.
    • Systems thinking could be important for not just deep understanding, but also for appropriation (i.e. business management, profit).
  • Jamshid Gharajedaghi (engineer/pioneer systems thinker) also did work in business management and consulting focusing soft skills. What does this tells us?
    • It tells us the social systems are crucial systems.
    • Social systems are the only systems that allow humans to plan our future.
    • Humans are unique in the manner.
  • Certainly, there are mixtures of many systems. Gharajedaghi approach is interesting because he uses “purpose” to focus roles:
  • One aspect of our group discussion that might worth ending on is that perspective is important to the understanding of systems thinking.
    • This may help us to understand the difference between “natural world ecosystems” and independent (physical is you will) systems.
  • One insightful comment, and there were many, is that “the more we actually intervene with the system is the more unpredictable the system becomes.”
  • Controlling behavior within systems” was also noted as being impossible to predict.
  • Another group member drew a circle and demonstrated that “when on the outside of a system, we see one thing, verses when we are on the inside participating within the system”
  • Internal and external view point changes our perspective on what goes on within systems.
  • The impact we have on systems may be passive or active.
    • Active may be advantageous because we can plan ahead and not wait for disaster to happen.
    • One useful tool is scenario planning, where extreme situation can be build up to look how parts of a whole can behavior. This can be use glean information about the future.
  • Changing systems also include looking at organizational values, which underpins the functioning of systems.

di_20130221_193710_st-on_group_chx

The formal discussion adjourned, as usual, with many of the group continuing their talk over food and drink.

The continuing monthly meetings of Systems Thinking Ontario can be found at http://wiki.st-on.org/.

References

Emery, Fred E., and Eric L. Trist. 1965. “The Causal Texture of Organizational Environments.” Human Relations 18 (1) (February): 21–32. doi:10.1177/001872676501800103. http://dx.doi.org/10.1177/001872676501800103.

Emery, Fred E. 1967. “The Next Thirty Years: Concepts, Methods and Anticipations.” Human Relations 20 (3): 199–237. dx.doi.org/10.1177/001872679705000802 , or see the manuscrript in the Tavistock Anthology at Modern Times Workplace;

Emery, Fred E. 1997. “Postscript: The Next Thirty Years-A Short Reflection.” Human Relations 50 (8): 931–935. dx.doi.org/10.1177/001872679705000803.

Ramírez, Rafael, John W. Selsky, and Kees van der Heijden. 2008. Business Planning for Turbulent Times: New Methods for Applying ScenariosEarthscan, 2008, preview available on Google Books.

Leave a Reply

 



↑ Top