Nature is Nonlinear

George Sugihara



We like to see the world as being constant in some sense, especially if the preservation of species (and the re-establishment of fixed baselines) is the goal. However, constancy and stability is a fiction, a dangerous approximation in many cases. It is just this kind of wishful thinking that causes us to be constantly surprised and which can lead to systemic mismanagement and collapse. Indeed, this was one of the problems contributing to the recent financial collapse, where measures of default risk based on the most recent decade of data were taken and applied as though the world markets were stable and unchanging fixtures in time.

Statistical methods that assume an unchanging universe can get us into trouble. Thus scientific tools and conceptions built to work well in controlled engineering contexts — involving simple and stable clockworkmechanical devices — do not work well in natural systems, where nonlinear instability is literally a fact of life. That is, real populations and ecosystems behave in an unstable, nonstationary way so that the rules for change depend on context (“state-dependent dynamics”). Nonetheless, this is not how most of our models work, and it is models that represent our formal conceptions of nature.

—To read entire article, download as PDF file (135KB)
Painting by Adam Wolpert

George Sugihara (Ph.D, Princeton) is a theoretician at the Scripps Institution of Oceanography who has worked in a variety of fields, including algebraic topology, quantitative finance, and more recently fisheries, neurobiology and atmospheric science.

return to Biodiversity contents