Pragmatic definitions in biology

Written by: Bjørn Østman

Primary Source: Pleiotropy

Biology is littered with concepts that biologists cannot always agree on how to define or where there are special cases where the common definition have to be amended. Species. Complexity. Modularity. Evolvability. Evolution. Genes. Community. Robustness. Open-ended evolution. Fitness. Life.

Take species. The current state of affairs is that there are many different definitions, and people can’t always agree which one is best. Each one of us may have a favorite. (Mine is the Ecological Species Concept by Van Valen (1976) Ecological Species, Multispecies, and Oaks. Also, best title ever.) This often leads to more or less antagonistic attitudes among people, and can have negative effects on the review process.

As far as I’m concerned John Wilkins is the man to go to for species definitions. He lists 26 of them.

In this case, “species” is the concept, and curse Ernst Mayr for being first to call a proposed definition a “concept” (i.e., the Biological Species Concept – which should also really have been the Reproductive Species definition).

What I propose is the stop calling the proposed definitions “definitions”, and instead call them “criteria”.

That would make it

  • The Reproductive Species criterion
  • The Ecological Species criterion
  • The Phylogenetic Species criterion
  • The Taxonomic Species criterion
  • … et cetera.

When determining if two groups of living organisms are different species, all you’d have to do is go down the list and check off those criteria that are met (easier said that done, I know). And then when talking about this, qualify the type of species by naming it according to the matching criteria. Your two closely related groups of organisms would then be ecological and phylogenetic species in the case brown bears and polar bears, and reproductive species in the case of horses and donkeys.

The reason I propose this is that this sort of pragmatic meta-definition has the potential to end unproductive arguments and replace them with clarity – a clarity that emphatically depends on people qualifying the type of species/complexity/modularity or whatever else they are talking about.

Same thing for the other difficult-to-define concepts. Open-ended evolution is the idea that evolution just keeps going, and new forms and features (species, traits, genotypes, etc.) keep appearing. But for how long? Forever? That’s longer than anything, so thats not very pragmatic, i.e. it is not a definition that can be applied, because we are too impatient to wait forever. Is natural evolution on Earth even open-ended? Would life on Earth ever reach a steady state after which evolution does not produce new things? Does co-evolution count if this produces new forms that have already existed in the past? Life on Earth is very much affected by the changing fitness landscapes when meteors arrive, volcanos erupt, and solar winds fry the planet, or whatever. Can we just create an evolving computational system in which huge disasters cause mass-extinctions at arbitrary intervals, and then say that the system exhibits open-ended evolution because it keeps evolving? How about these criteria:

  • System evolves in never-ending cycles (revolving open-ended evolution)
  • System reaches steady-state but is reset/interrupted by disasters (reboot open-ended evolution)
  • System continues to evolve new forms for as long as we could wait (temporal open-ended evolution)

The point is to move research forward, rather than letting it be mired in argument, and I think that can be done by simply being more explicit about what we mean when we say something.

Anybody feel like taking a crack at evolvability or life?

– See more at:

The following two tabs change content below.
Bjørn Østman
Bjørn Østman is an evolutionary biologist postdoc working in the BEACON Center for the Study of Evolution in Action.
I am interested in many aspects of evolution. I work in computational biology, using various approaches to learn about fundamental processes of evolution. Bioinformatics is good for learning about real genes (data generously supplied by other researchers), and simulations are good for testing the mechanisms of evolution. I am particularly interested in how populations and organisms adapt to changing environments, both at the genetic and phenotypic level. Lately my research has focused on the evolutionary dynamics of populations evolving in rugged fitness landscapes.
Bjørn Østman

Latest posts by Bjørn Østman (see all)