A 1995 System Dynamics Listserv Discussion on a Plan to Reformat our Journal, System Dynamics Review:

[The discussion extended from March 10,1995 to March 30,1995. It began with the following email sent to the Listserv by me, Arlen Wolpert. All of my editorial comments or clarifications are given in square brackets. I am an independent scholar.]

Arlen [March 10]:

Open Letter to the System Dynamics Community:

I believe the System Dynamics Review [SDR] should be reformatted in the following way:

  1. All main articles should include the following three items:
  2. In subsequent SDR issues Continuing Commentary of the various target articles should be published with the Reply (if possible) of the author of the target article.
This format has been used with great success in the journal, Behavioral and Brain Sciences. It has been very effective in clarifying and integrating issues during the coelesence [sic] of fields now going on among cognitive neuroscience, philosophy of the mind, linguistics, and psychology.

System Dynamics models almost always experience resistance to acceptance, particularly from outside the system dynamics community. For that reason invitations for Open Peer Commentary should include, not only relevant system dynamicists, but scientists from the field of application. It is important to choose a few key scientists and thinkers who are most opposed to the model and its implications. Such commentary will bring out into the open unstated, controversial, or negative opinions and ideas where they can be analyzed and discussed.

Arlen [March 20]:

On March 10 I submitted a plan to the SD Discussion Group for reformatting the System Dynamics Review. Let me make an additional comment on this reformatting plan:

I have been working on a system dynamics model full time for the past nine years. The first four years were spent in Phase I: the attempt to develop the first pass at a model. During the past five years 25% of my time has been spent in Phase II: trying to introduce the model and its implications. I am finding terrific resistance in Phase II. My first hand awareness of this resistance has made it clear that Phase II resistance is the common lot of all system dynamicists who attempt to apply or introduce their model. The most obvious case is the resistance experienced by Prof. Forrester and his unsung heroic followers associated with the World Dynamics Project.

The easy way is to publish the model produced in Phase I and go on to the next modelling or academic idea and avoid Phase II. Academia tends to condition the system dynamicist along these lines. The present format of SDR further conditions him or her in this way.

Let us be clear: Despite the fact that most system dynamicists are situated in relatively conservative business schools, academia, or corporate life, the truth is, at the very essence of our field, we are involved in a revolution. While I may be wrong, I am offering my view that we should now press forward and make SDR a little more bold and courageous.

Jay W. Forrester [March 23]:
[Forrester is the originator of system dynamics. He is the Germeshausen Professor Emeritus and Senior Lecturer at the Sloan School of Management at the Massachusetts Institute of Technology. There was a parallel Listserv discussion going on concerning the history of system dynamics. Forrester may have been trying to contribute to both discussions at the same time with this contribution.]

In his March 20 communication, Wolpert observes,

" During the past five years 25% of my time has been spent in Phase II: trying to introduce the model and its implications. I am finding terrific resistance in Phase II. My first hand awareness of this resistance has made it clear that Phase II resistance is the common lot of all system dynamicists who attempt to apply or introduce their model. The most obvious case is the resistance experienced by Prof. Forrester and his unsung heroic followers associated with the World Dynamics Project."
My experience has been very different from that which Wolpert reports. Perhaps two different situations should be identified.

First is when a system dynamicist works with a client to create understanding and to solve a problem. The team works together and acceptance is generated along the way. The challenge here is not to get acceptance after a model has been created, but rather to establish the initial working relationship that unites the system dynamicist with a client that realizes a need for modeling. The client comes before the modeling.

The second situation fits the setting described by Wolpert and also the World Dynamics model. Here, the modeler decides that a projected model will be accepted and found useful after the model is complete and presented. The system dynamicist takes the risk of judging what will be of interest to future potential recipients. Wolpert seems to be describing a model that he feels is important but it appears that others are not sharing his enthusiasm. That is very different from the World Dynamics model to which he refers.

For those who may be unfamiliar with events following the publication of World Dynamics, a little history may be of interest. The public response was totally different from what I had expected. The book seemed to have everything necessary to guarantee no public notice:

  1. It has 34 pages of equations in the middle of the book.
  2. The interesting results are in the form of graphs vs. time, which many of the public cannot read easily.
  3. The book deals with issues up to a hundred years in the future and lie outside the presumed time of interest to the public.
  4. The book was brought out by a publisher that had published only one prior book, and I doubted that it had the commercial stature to be reviewed.
  5. The model in the book was the product of only two Saturdays of work. The book took another four months to write, edit, and get all the computer runs onto consistent scales. Even so, the total effort was only a small fraction of that devoted to my other books.
I believed the World Dynamics book was for perhaps a hundred people who would like to see how one can organize a mental model into an interesting simulation exercise. That was the worst prediction I have ever made.

The book came out the first week of June 1971. On the last week of June it was reviewed on the front page of the London Observer, which at that time circulated around the world. I received a letter from a faculty member in one of the State Universities of New York asking for more about what we were doing because he had been reading about it in the Singapore Times. In August the Christian Science Monitor devoted all of the first page of its second section to the book. In September it received a page and a half in Fortune, in October a column in the Wall Street Journal. World Dynamics was being debated in the editorial columns of mid-America newspapers, and in the environmental, zero-population growth, and underground anti-establishment student presses. In the middle of the political spectrum, World Dynamics was the subject of a full length article in Playboy. It was the subject of prime time television documentaries in Europe.

The public response was a good hundred fold over what could have been rationally expected. World Dynamics is still in print and available from the Productivity Press in Portland, Oregon.

'The Limits to Growth' by Donella Meadows, et. al, Universe Books, came out nine months later in March 1972. The message was essentially the same as in World Dynamics. Much work had been done to verify assumptions and to refine and extend the model. It was more popularly written and did not have equations. Even so, it seemed that The Limits to Growth might be an anti-climax after the vigorous public debate about World Dynamics. Wrong again in the same way. Public discussion went up by a factor of ten from where it had been. As one person in Washington told me, "No one is entitled to have no opinion about Limits to Growth." His estimate was, even though the book was initially politically non-discussable in some government quarters, that some four out of seven people agreed with it. A dinner meeting, attended by about 60 chief executives of major corporations and government agencies, was held at the Century Club in New York to discuss The Limits to Growth. Limits to Growth sold 400,000 copies in the first summer in the Dutch language alone. It has been translated into probably 40 languages. Beyond explanation, I received the Vietnamese translation of Limits to Growth during the US-Vietnam war.

Gerald Barney had been associated with the Urban Dynamics, World Dynamics, and Limits to Growth activities. Within the US Government, he carried on the thread of activity to write the 'Global 2000 Report to the PResident: Entering the Twenty-First Century,' (Penguin, 1982), which was sent by President Carter to heads of states of all countries. Barney is now director of the Millennium Institute in Arlington, VA, email millennium@igc.apc.org, which is providing modeling of growth and environmental issues for several countries.

Following the publication of World Dynamics and Limits to Growth, The International Institute for Systems Analysis in Austria held annual meetings on world modeling for ten years.

Many universities held debates on growth following the two books. There was considerable polarization of viewpoint. North American academic economists were generally very negative. Strong support came from students, the public, and corporate chief executive officers (in contrast to middle level executives). The books sharpened the debate about population growth and environmental issues.

Contrary to Wolpert's comment, it seems that the followup to World Dynamics was rather more than might reasonably be expected from two Saturdays of modeling.

John Sterman [March 24]:
[Prof. Sterman is Head of the System Dynamics Group at MIT.]

Those interested in the world modeling issues Jay raised in his last message should also read:

Dana Meadows, John Richardson, and Gerhart Bruckmann, 1982, Groping in the dark: the first decade of global modelling. Chichester: John Wiley.

It is a wonderful book with both hard technical detail on the many world models and their strengths and weaknesses, and also personal discussion about the conflicts, values, and ethical issues involved in modeling the planet as a whole.

Arlen [March 24]:

On March 10 I submitted a plan to the SD Discussion Group for reformatting the System Dynamics Review. On March 20 I made some additional comments on why a reformatted SDR would make our journal into a more effective communication tool. Now let me give one example of how this reformatting plan would work:

Khalid Saeed has been developing an important body of work on the dynamics of economic growth and political instability in developing countries over the past 15 years. This work is very little known, yet it is perhaps one of the most important areas of study.

The editor of SDR would invite Prof. Saeed to write a target article on his most recent model in this area with full modeling detail together with the insights that he and his associates are drawing from their work. When this target article has been received, the editor would send requests for a critique from other system dynamicists who have published in this area, such as Wolstenholme, Mandal, Paraymo, Thirumurthy, Yamamura, J.M.Richardson, Mashayekhi, Bala, Tasrif. (I have probably left out important modelers in this list.) Requests for critique would also be sent to some high officials in governments and agencies associated with economic development, including the IMF and the World Bank. Requests would also be sent to responsible key political commentators from both North and South, both left and right.

All correspondance would be carried on through email to cut down on editorial work. Money would be supplied for additional editorial staff.

The key questions before us are: Are we ready to increase communication, dynamism, and activity in system dynamics and at the same time bear the pressure of the ensuing controversy resulting from our 'politically incorrect' journal? Do we have to wait patiently 20 to 50 years, while people are being educated in system dynamicists and while the world disintegrates before our eyes, before taking action?

George Richardson [March 24]:
[Prof. Richardson is at Rockefeller College at SUNY Albany and was the editor of the System Dynamics Review from 1984 to around 1993]

I have read with interest the various suggestions made about improving our scholarly communication in the System Dynamics Review.

We have tried in the past to generate discussions in the pages of the Review (see, e.g., the article and commentaries on modeling scientific revolutions, and also Maloney's challenging note on 'essential tension' in the field). Such discussions and commentaries are not easy to generate -- probably because academics do not get tenure and promotions for writing commentaries, and consultants and other practitioners don't have the time or rewards either.

I believe that restricting the way articles appear in our journal is not wise -- I would favor a very rich mix of articles in the System Dynamics Review, including insightful reports of applications and implementation efforts, research reports, methodological advances, and scholarly debate. Restricting publication to only those articles that fit a certain mold could further shrink the flow of high quality submissions, without a clear indication of the net benefit to the field.

There is nothing that currently prevents Wolpert's suggestions from being done for a given article -- assemble the team of authors and commentators and produce high quality manuscripts and send them into the review process. I'm sure Graham Winch and the editorial board would welcome such stimulating fare.

What many have said before about the growth and acceptance of work in the field still seems true to me: Pick significant problems; work on them in full knowledge and acknowledgement of the good work of others (in and out of our field); produce insightful work; write it up well; publish and/or consult widely; and try hard to learn from responsible criticism. I'm absolutely convinced that good work will be our best force for growth and acceptance.

And we must realize that there is very good work going on -- consultancies are growing, our best academic work is publishable in very high quality outlets (e.g., Management Science, Interfaces, and the like), and our best practitioners are in demand.

I'd suggest the best way to improve our scholarly communication in the System Dynamics Review and elsewhere is simply to do great work and communicate it well -- with or without commentary. Let's keep the restraint only on the quality dimension, not on an abstract form of discourse.

Jay Forrester [March 24]:

I very much agree with the March 24 communication from George Richardson.

The future of system dynamics rests on high quality work that serves a useful, practical purpose. Dangers lie in proliferation of models that are trivial, and in modeling that addresses only academic audiences and not the important concerns of society.

Acceptance and growth of system dynamics is probably developing as rapidly as we have a right to expect for a field that represents a paradigm that conflicts with many others that are long established in academia, business, and politics.

More important than any benefit that can come from reformatting the Review would be to have more books aimed at the public that use sound, rather simple, and powerful models to shed light on the great controversies of the day--rising government debt, welfare, population growth, retirement programs, increasing overhead in society, balance of trade, drug addition, and financial stability. These must not be the kind of presentations that are designed to support a prior political ideology, but rather they should show how the present problems arose, and what are the short-term versus long-term tradeoffs in the future.

Khalid Saeed [March 25]:
[In 1995 at the time of this conversation Prof. Saeed was the leading system dynamics thinker at the Asian Institute of Technology in Bangkok, Thailand. At the present time he is the Head of the Dept. of Social Science at Worcester Polytechnic Institute.]

I wish to add a word to a very fine summarizing done by George on the subject. I would perhaps be repeating what I have said earlier whenever the question of stepping up the effort to sell SD has been brought up.

Good products sell themselves and organizations creating sound production prosper. Jay Forrester pointed this out convincingly in his market growth model. Sales effort without sound production might only increase expectations which cannot be met. Let us put our resources into quality production.

Arlen [March 29]: Reply to Comments:

  1. To stimulate Commentary on Target articles by those on tenure track, such Commentary would be indexed or abstracted like all SDR articles. This is how the journal, Behavior and Brain Sciences, handles this problem.
  2. System dynamicists in the developed countries should be made aware of the fact that their counterparts in the undeveloped and developing countries are often [financially] supported by repressive governments, feudal systems, or dictators. One 'politically incorrect' move could lead to disaster for the system dynamicist and his or her family. Therefore, compared to system dynamicists in the developed countries, they usually have far less room to manueuver and to innovate or to initiate high leverage policies.
  3. I take Prof. Saeed's comment of March 25 to indicate that the work over the past 15 years on the dynamics of economic growth and political instability in developing countries has not yet resulted in a high quality model. Nevertheless, I believe that if an all out attack were mounted by the SD community on this important problem, a high quality model could be realized within a few years.

Khalid Saeed [March 30]:

This is to respond to Arlen Wolpert's comment on the implications of my comment on the quality of my own models.

Firstly, it is not for me to judge the quality of my own work; my academic peers would do this. Given that the models have been published in a number of learned journals and the two books I have written are well received, I have no reasons to complain.

As for the implementation of the policy guidelines, it would not be prudent to press for this for any new piece of knowledge, which after all is only another view of the complex reality from a different vantage. I think a bit of humility for the knowledge artifacts we create would facilitate the search for truth and this will make the world a much better place to live since it would provide room for reflection without any pressures. Given enough time in such an environment, there is a good chance that we come closer to truth.

Let me add that I greatly appreciate Arlen's kind words about my work and that I would continue the work in the academic niche I have selected for its intrinsic value to me, rather than for any extrinsic rewards. But, this attitude might very much be a part of my Eastern origin.

Although I would not have time to continue this discussion much longer, I wish to add a comments on points 2 and 3 of Wolpert's message:

The fact is that the north-south intellectual devide evident from the logic of Wolpert's conclusions is far more dangerous than the perceived repression in the developing countries. The learned designs of development policy design originating from the north have often had disasterous implications for the developing countries in which they were implemented. This is because the models they were based on lacked experiencial information, which incidentally is the cornerstone of SD modeling. I am glad the subject of north-south intellectual divide is now being openly discussed in the academic cirlces and a respectable body of literature is developing on it. This should also moderate the kind of initiatives proposed by Wolpert.

I view our SD work as a humble endeavor in science. Our primary responsibility is to add value and not push ahead for a revolution. It is practice with this attitude by our colleagues that has won acceptance of SD as a serious discipline. I believe our practice must be disciplined by science rather than an endeavor to reinvent science and I view all kinds of pushes to sell to be overbearing and harmful for our field.

George Richardson [March 30]:

Much admire Khalid's statement:

"I view our SD work as a humble endeavor in science. Our primary responsibility is to add value and not push ahead for a revolution. It is practice with this attitude by our colleagues that has won acceptance of SD as a serious discipline. I believe our practice must be disciplined by science rather than an endeavor to reinvent science and I view all kinds of pushes to sell to be overbearing and harmful for our field."

Bob Eberlein [March 30]:
[Dr. Eberlein is with Ventana Systems, Inc., and is the moderator or host of the system dynamics Listserv.]

I share George's admiration and I do think this is worth posting twice.

End of discussion.

Arlen Wolpert
Cambridge, MA
February 18,2000
http://world.std.com/~awolpert/gtr362.html

Return to homepage.