REPLY Society Strategy Development (SD6878)

[SDMAIL Jack Harich]

Posted by  Jack Harich <register at thwink.org>

Kim, Jim, and others,

Nice idea, Kim, to visionize about what outstanding achievements system 
dynamics might have made 20 years from now. This supports Jim’s idea, 
that “We hope to have some concrete steps that the Society can take to 
support the growth of the field and the Society.”

But isn’t this putting the cart before the horse? Isn’t this essentially 
saying that we have this tool called system dynamics, and now we want to 
grow its use? To me, this comes uncomfortably close to slipping into the 
goal of organization self-perpetuation. This is one of the first signs 
of a bureaucracy, whose interest is no longer serving others, but itself.

Better, I suspect, would be to ask: What problem domain is it this 
nascent field is trying to solve? Every field attempts to address some 
problem area of importance. Over the years, fields like physics, 
chemistry, and electrical engineering have accumulated a mature toolset 
for practitioners to successfully solve problems.

So what kind of problem is SD trying to solve? Currently it seems to be 
mostly whatever can best be solved with stock and flow models. But this 
is too limiting, and may be a principle source of the limited success 
the field has seen.

Is the domain *problems whose solution requires deep and correct 
understanding of dynamic feedback structure?* That’s a much broader, 
more customer oriented domain. Call this domain A.

A diverse set of tools can already provide a sense of understanding of 
structure, like accounting, spreadsheets, stock and flow modeling, agent 
based modeling, causal flow modeling, economic modeling, etc. But in so 
many cases these tools fall short. Economists are still unable to 
predict recessions, growth spurts, or economic crises. Most businesses 
are unable to do the equivalent, except for the short term, even if they 
use SD. On large social problems, society continues to be unable to 
solve problems like corruption, systemic poverty, war, mal-distribution 
of wealth, environmental sustainability, and urban decay/slums, although 
some of these have been partially solved.

Given this state, it seems the real question is not how can the field of 
SD move forward, but: *How can the maturity of problem solving tools for 
domain A be increased to the point where they can solve the highest 
priority problems?* This would elevate our thinking to the proper level 
of abstraction, and we would be less likely to remain mired in the 
mental traps and ruts we may have slipped into. This can happen to any 
field. Indeed, it has happened to most. Over the centuries, countless 
new promising fields have emerged. Most have fallen from the screen, to 
be absorbed by others or to disappear altogether.

So what are the highest priority problems in domain A? I see two broad 
classes: business management and social problems. If we looked at the 
top ten problems of each and the benefits of solution to society, as 
well as the costs of no solution, the results would be a landslide 
towards the need to solve the world’s major social problems first.

Of course, better ability to do that would translate to better ability 
to solve business management problems. Since the average business 
problem is an order of magnitude easier to solve than the average social 
problem, the translation would be amplified.

By the way, I just don’t feel enthusiastic about helping a field become 
better at solving yet another market share problem. But I can get 
excited about being able to celebrate in 2028 the achievement of being 
able to reliably solve social problems like those listed above. I can 
see the headlines now: “After 10,000 Years of Searching, The Key to 
Eradication of Poverty Finally Found” and “Spaceship Earth Breathes Sigh 
of Relief as Sustainability Problem Solved at Last.”

*The Goal*

This line of reasoning leads to a clearly stated goal: *Develop the 
ability to reliably solve large, pressing social problems.*

Would it be possible to fine tune this goal even more? Yes, if there 
were certain social problems that, if unsolved, made all other problems 
(whether business or social) irrelevant. As far as I can tell, there are 
two problems that fit this class: war waged with weapons of mass 
destruction and global environmental sustainability. Either one, if not 
solved, will wipe out most or all human population. Let’s call these 
mega social problems. This leads to a refined goal: *Develop the ability 
to reliably solve mega social problems.*

This will not be easy. It may even look impossible. But long ago, so did 
solving the serfdom problem, the rule of kings problem, the slavery 
problem, the women’s suffrage problem, etc. Problem solvers have already 
solved an impressive collection of social problems. They have already 
proved the impossible is possible. All we are doing is tackling 
progressively more difficult problems, as well as ones that must be 
solved much faster. This is why better tools are needed.

*Strategy to Achieve the Goal*

Next, what would be the topmost strategy to achieve this goal? “The 
ability to reliably solve” holds a major clue. We would be developing 
the ability to reliably accomplish something. Taking a leap, my 
definition of process is a repeatable series of steps for achieving a 
goal. It follows that the goal can be restated as: *Develop a process 
for solving mega social problems.*

Some Society members may feel uncomfortable with this goal. It may seem 
too ambitious, too impractical, or too self-righteous. But the way I 
look at it is what are the world’s biggest needs at the highest level of 
abstraction? What would be the most incredible contribution to science 
we could make? Would not this advance be the biggest in all of science 
since the invention of the Scientific Method itself? Wouldn’t it cap the 
progression of science so well that we could at last proudly say that 
the (false) dichotomy of the hard versus the soft sciences has been 
shattered at last?

To allow those participating in this exercise to feel more comfortable 
with this stretch goal, I have available a paper on “The need for a 
standard process for solving difficult social problems using system 
dynamics.” While it focuses only on SD, I’m sure readers can see that a 
standard process would require far more than just stock and flow 
modeling to accomplish the goal. The paper explains what being process 
centric is all about, why modeling is a repeatable and improvable 
exercise regardless of the problem domain, and presents a sample process 
in order to illustrate the argument. The paper is designed to cast 
strong lights on the shadows of impossibility, and turn them into the 
possible.

However, it’s probably best to first discuss the goal, and then later 
take a look at the paper, so that we don’t put the cart before the horse.

*Avoiding Repeating Your Own Mistakes*

“Those who cannot remember the past are condemned to repeat it.” – 
George Santayana, Life of Reason, 1905.

Awhile back I prepared another paper that may be of some interest. It 
was an analysis of a similar organization, so similar it is somewhat SD 
centric. Like the Society, they too had failed and they knew it. Like 
the Society, they embarked on a self-examination and self-improvement 
project. So, as part of this I prepared a paper on “Can These Best 
Practices Make (name of organization) Effective Once Again?”

Sadly, the paper had little effect. It seems this organization was not 
ready for a fresh viewpoint. In fact, they even shot the messenger.

Just as sadly, the organization appears to be continuing its old 
methods, and is trying the same thing over and over, and hoping the 
results are going to be different….

But again, to help us focus, it’s probably best to first discuss the 
goal, and then take a look at this second paper.


Hope this helps us move forward,

Jack
Posted by  Jack Harich <register at thwink.org>
posting date  Mon, 07 Apr 2008 11:09:16 -0400