Friday, April 14, 2006

An Example of how Stephen Jay Gould was wrong

6 comments:

Anonymous said...

I think I've figured out the reasoning which led him to his wrong conclusions. Short version: if ecological change played an important role in the development of species, then the fossil record should show a correlation between ecological and evolutionary change. It does not. Thus, ecological change is not particularly central in the evolutionary process.

Once again, Gould took 'it cannot currently be demonstrated' to be a stronger statement than the facts warrant.

(I have a hunch that when body image capture technology is applied to old basketball tapes, Gould's 'myth' of the hot hand will also evaporate.)

Will Baird said...

Short version: if ecological change played an important role in the development of species, then the fossil record should show a correlation between ecological and evolutionary change. It does not. Thus, ecological change is not particularly central in the evolutionary process.

Isn't that a rather interesting thought in and of itself?

Anonymous said...

"Thus, ecological change is not particularly central in the evolutionary process."

I think I must have been indoctrinated by Wilson without even noticing it, because that sentence looks very counterintuitive.

btw, how solid is the data for that graph?

Will Baird said...

I think I must have been indoctrinated by Wilson without even noticing it, because that sentence looks very counterintuitive.

That statement looks hella counterintuitive, frankly. I'm hoping Carlos will either post a countercomment or a

btw, how solid is the data for that graph?

I think that graph comes from a pretty hefty bit of research that was covered in Catastrophes and Lesser Calamities. It seems that this graph is derived from two sources. The first is a database of families, genera, and orders compiled by JJ Sepkoski. Secondly is the work of MJ Benton of _When Life Nearly Died_ fame (Biodiversity on land and in the sea. Geological Journal 36, 211-230).

It's not perfect. It will be constantly updated, but I suspect that the trend will hold pretty true. It would be interesting if someone has correlated the availability of said fossil genera for a given time with the data in the graph. It might take away some of that curve. Or not.

To tie into things SFnal, James, it pokes its finger in Baxter's eye over comments he made in _Evolution_. Again. ;)

Anonymous said...

You will love the Baxter short story I just read then: humans lose higher technology but persist for another two billion years (I think) without changing significantly. At the same time, no species ever evolves that can work around our intelligence and tool use, so the world's ecosystems remain simple forever.

He explains the Fermi Paradox that way as well: the first human-level intelligence on a world destroys most of the complex life.

Anonymous said...

It is counterintuitive, and has to do with Gould's explanation for the slow equilibrium phases of punctuated equilibrium.

It ties into Gould's preference for statistical explanations for allometric trends -- e.g., increase in body size. Egression from the mean rather than ecologically driven evolutionary development.

A modern evo-devo type would likely say that most adaptations to ecological change occur at the molecular level -- think of changes in fur or skin color -- and take time to appear as distinct fossil taxons.

On the other hand, Vermeij, the other shell guy slash evolutionary theorist, thinks that there actually is more fossil evidence of ecological change driving evolutionary change than Gould gave credence to.

Either way, Gould is looking more and more tendentious in his interpretation.