Problems with creating useful scientifically valid futures scenarios

Abstract

Attempting to do “Futures” research is beset with many scientific and mathematical problems. One major question is what does it really mean to develop useful scenarios for the future to help one understand the likely impact of key policy issues that a researcher wants to address? And if it is hard to create a believable projection for the relevant system to be studied in a “business-as-usual” scenario, is creating other types of Futures scenarios any more likely to be scientifically valid for doing policy research? Furthermore, what kinds of alternative Futures scenarios make sense to create, and how should the methodologies used to create them depend on the policy issues being addressed? This paper will attempt to address these questions in the context of what doing good science, the key ingredient, implies for doing useful futures research.

Another more technical issue which besets the creation of Futures scenarios is the question of to what extent can statistical methodologies be scientifically valid in their creation, and how are statistical methodologies often misused in doing futures research? Finally, the key issue of how one can make each scenario of the future internally consistent and plausible, if one can, is addressed.