SUBJECT: In defense of Ignorance
AUTHOR: William Silvert
DATE: 15th April 2003

Perhaps there is nothing that irritates politicians and managers more than a scientist who says that we don't know enough, we need more research. I like knowledge as much as anyone, but we have to accept that the amount by which our ability to make good predictions grows is a decreasing function of how much we know, and we can reach a point where no matter how much more we learn, we cannot describe the functioning of the system any better.

The butterfly effect is a good example. While we may be able to trace the trajectory back to determine which butterly "caused" a hurricane, we can never predict that a given butterfly will generate the next "big one" - and even if we eradicate all the butterflies in Australia, there will still be storms.

I doubt that we will ever be able to predict with certainty the consequences of environmental change, but we can assess the risks and advise on those. For example, we know that useful pharmaceuticals are most likely to come from toxic species, so the risk of missing the cure for cancer is greater when a cone shell goes extinct than when a guppy does. Of course this does not imply that we know which cone shell contains the secret ingredient!

Sure it would be nice to have a list of all the species in the world (a list that would have to be revised several times per minute if we include all the microbes), but I am not convinced that the fact that most of the world's species have not yet been scientifically identified is such a major
impediment to our ability to make sound recommendations on the management of
marine ecosystems.

A contribution by:

William Silvert
Instituto de Investigacao das Pescas e do Mar
Av. de Brasilia, s/n
1449-006 Lisboa
Portugal