International Encyclopedia of Systems and Cybernetics

A technique in futurology (or prospective) consisting in describing a possible future based on existing data and selected assumptions.

M. MARUYAMA prescribes the following steps, in accordance with his concepts relative to interconnected positive and negative causal loops (see 2nd cybernetics):

"1. Make not one, but several scenarios.

"2. The purpose of the scenarios is to consider various possibilities, not to make a prediction. Therefore they are useful to the extent that they include novel, unusual elements. The scenarios made in the past tended to extrapolate present trends quantitatively. But we need scenarios which include novel qualitative changes.

"3. For each causal diagram, first put down a novel element, such as fertile arid land or outer space industry. Then put down variables which are affected by this element, as well as variables which will affect this element. Draw causal arrows with plus or minus signs, strength indicators and delay indicators.

"4. Draw causal arrows to and from these variables from and to still more variables. Repeat the process.

"5. Eventually, the diagram becomes too crowded, Eliminate the variables which have weak direct or indirect relations to the first element as well as the variables which have weak effects on the type of business category or the firm for which you are making the future scenario. A workable number of finally selected variables is between 10 and 15.

"6. Make a list of all the loops found in the diagram.

"7. Eliminate loops which have weak loop strength or too long a total time delay, say 50 years or more.

"8. Now you have a small number of loops to consider. Write down in sentences the dynamics of these loops.

"9. For each scenario, determine whether your firm is a loser or a winner in comparison with your competitors.

"10. Work out a strategy to make your firma winner, and examine its feasibility and obstacles.

"In this type of exercise, you discover that causal loops behave in a way very different from what you expect by looking at each causal arrow separately" (1994, p.87).

This way to construct scenarios is satisfactory in systemic and cybernetic terms. It does however still depend on the personal acumenand judgment (or, unfortunately, prejudices and ignorance) of the researcher. The case is quite similar to the construction of the basic model in FORRESTER's systems dynamics.