Fred, thanks for the wonderful posting on Henry Gantt. It's useful to see
things in their historical perspective. Without the benefit of a business
education, I invented the Gantt chart on my own back in the 60s.
As a producer/director of films, I needed a method to plan multiple
projects which sometimes used the same resources. I "made up" the method
of setting down the delivery dates of the completed films and working
backward from there, showing all the steps visually on a matrix form.
When I showed this proudly to a colleague, he laughed. It was he who then
told me about Gantt charts.
Perhaps it was they way I came to Gantt charts that helped me understand
the disconnect between your posting and Karl-Henrik Robert's discussion of
"backcasting" at the Systems Thinking in Action conference.
Gantt-charting refers to a methodology for determining in what sequence
and in what time frames specified actions need to be taken to achieve a
specified outcome. My understanding of "backcasting" is that it is
As I understand it, "backcasting" is a methodology for determining *what
actions need to be taken* in order to arrive at a generalized state. The
desired state here is not a product or event fixed in time, but the
realization of a vision.
You would use a Gantt chart to plot (albeit backwards) what you will do
and when to detoxify Lake Erie by March, 2002. You would use
"backcasting" to identify what kinds of actions you must take, and what
kinds of actions you must abandon, in order to have toxin-free fresh water
in the Western Hemisphere.
Although I tend to agree with you that "backcasting" is jargon, it has the
merit of putting your attention on "forecasting" which - despite all
evidence to the contrary - is widely accepted as a possible human
activity, even by those who believe in backcasting.
>Working backward from a goal state is one of the problem solving
>strategies described by Allen Newell and Herbert Simon in their 1972 book,
>"Human Problem Solving." It is also a production planning technique of
>long, long standing (and I'll come back to that in a minute). But what
>prompts this posting is a combination of factors. First, other messages
>in the learning-org digest dealt with jargon. ("Backcasting" strikes me
>Second, I was reminded of Peter Drucker's comment that many people think
>they're using Gantt charts when they're not. Allow me to explain.
[...quote of Fred's msg trimmed by your host...]
-- Jack Hirschfeld With the clear undertanding that email@example.com this kind of thing can happen, shall we dance?