Tuesday, May 17, 2011

Happy 10th birthday, neutral theory!

Rosindell, Hubbell, and Etienne. (2011). The unified neutral theory of biodiversity and biogeography at age ten.

I would argue that neutral theory is not only the most controversial idea, but also the most successful idea to permeate community ecology in the last ten years. A quick keyword search suggests that ~30 ecological papers related to the topic were published in the last year, including some with titles still reflecting the controversy; “Different but equal: the implausible assumption at the heart of neutral theory”. Neutral theory makes a seemingly unreasonable assumption—that species identity doesn’t matter—and yet seems to predict species-area relationships and species abundance distributions as well or better than niche theory does. This made it an infuriating challenge for many ecologists. The number and quality of papers that it inspired—both in support and opposition—are a reminder that disagreement is good for science.

It’s been a decade since the publication of “The Unified Neutral Theory of Biodiversity and Biogeography”, in which Steve Hubbell proposed a controversial model in which coexistence results from drift, dispersal and speciation, rather than ecological differences between species. To mark this anniversary, a review in TREE by James Rosindell, Stephen Hubbell, and Rampal Etienne reflects on neutral theory’s first ten years, and examine the influence neutral theory has had in many areas of community ecology. The authors also note that some of the limitations of neutral theory can be dealt with by extending the classic formulation of the model, so that unrealistic assumptions related to spatial structure, speciation rates, or the zero-sum assumption can be relaxed. The excessive interest in neutral theory’s species-abundance predictions left its other predictions unexamined, and there is still room for tests of how neutral theory informs species-time relationships, modes of speciation, and even conservation decisions.

Despite these accomplishments, the review is remarkably subdued, underlined by statements such as neutral theory is a “good starting point”, a “valuable null model”, and a “useful baseline”. However, it seems unnecessary to state, as some have, that "neutral theory is dead". Its legacy, captured in the final paragraphs, is still incredibly important: “…niches have dominated our attention and left less obvious, but still important processes forgotten… Perhaps the most important contribution of neutral theory has been to highlight the key roles of dispersal limitation, speciation and ecological drift, by showing how much can be explained by these processes alone...”

George Box said it best: “All models are wrong, but some are useful”.

5 comments:

Jeremy said...

So even Steve Hubbell doesn't think of his 1997 Coral Reefs paper as the birth of neutral theory in ecology?

penall said...

Of course, disagreement is good for science, but the neutral theory has not enough supporters.

Jeremy said...

Surely neutral theory should have to *earn* supporters, just like any other idea in science? Speaking of an idea as not having "enough" supporters makes it sound like a new idea is somehow entitled to some level of support.

Anonymous said...

A theory can be said to deserve more support than prejudice has permitted it to earn or garner. Hubbell's faces off against a mountain of it. Ever heard of niches? Of specialization? It argues against both.

Marc Cadotte said...

Thanks for the comment. I think the above comment hits on an intuitive response that ecologists have to neutral theory -namely that it can't be correct given all the specialization and clear niche differences. However, as with most things in ecology, it is not so black and white. If all species co-occurring at a single site had complementary niche use than we should see stable coexistence. However, given that this is often not the case, further compounded by the repeated observation that some subsets of species seem to do very similar things, neutral-type dynamics may be a legitimate way to characterize their dynamics/patterns. What is interesting now, is not whether the world is solely niche or neutral (its not either of these), but how do these types of processes come together to form the communities that we see in nature.