« The bird in the mirror | Main | Bias in science vs. honest errors »

Ask the right experts

A book review titled "Redefining 'natural' in agriculture" makes some interesting points. I haven't read the book, which is about organic farming and transgenic crops, although I know both authors slightly from my years as a professor at UC Davis. The review notes that many people have strong opinions about agricultural issues even though they lack relevant expertise. Anthony Trewavas, the author of the review, suggests that even "being a scientist doesn't qualify you to advise on any subject except your specialty."

So what is his own specialty?

A quick check of Web of Science finds he has published substantial scientific papers on "Dynamic localization of calmodulin domain protein kinase (CDPK) and its relationship to calcium signaling in growing pollen tubes" and so on, so if I ever want advice on those topics, I'll know where to turn.

When he wanders from his field of expertise things get interesting, but not necessarily credible. For example, his recent paper on "Green plants as intelligent organisms" claims that

When provided with water only once a year, young trees learn to predict when water will be provided in the future and synchronize their growth and metabolism with this period only

How exciting! Learning in plants! Except that I looked up the cited book chapter and it doesn't show that. All of the trees in the experiment described were
watered to field capacity (27% vb/v) once per year, at the beginning of the growing season, to simulate natural patterns of water availability

In other words, the trees got water at the same time of year as their ancestors, so they didn't have to learn anything new. Instead, they just executed their DNA-based programs, shaped by past natural selection.

What is missing from Trewavas impressive publication list is original field research on environmental risks or benefits of biotechnology, pesticides, etc. If we want to know which pesticides cause cancer, we need to consult cancer researchers, not chemical engineers. Similarly, if we want to know about ecological and evolutionary risks of biotechnology, we need to consult ecologists and evolutionary biologists, not genetic engineers.

Fortunately, the Ecological Society of America, a stellar group of 8000 scientists who collectively publish a significant fraction of ecological research papers worldwide, has issued a public statement on risks and benefits of transgenic crops. Individual ecologists and evolutionary biologists have published on particular biotechnology related topics, but this statement is a balanced expert overview of issues that should be considered. They make a good case for a "cautious approach" while supporting "judicious use of biotechnology."

Trewavas suggests that risks of pesticides should be compared to risks from the natural chemicals plants make to defend themselves against insects, that wider use of organic methods could worsen food shortages, and that reduced tillage methods using herbicides can reduce erosion. I agree with each of these points, to some extent, but in each case there are other factors to consider. Watch for my book on Darwinian Agriculture, sometime in 2009. Go ahead, ask whether I have published papers with original data on both evolutionary biology and agricultural field experiments.


Good post. I’m just a graduate student in biology, but often get asked my opinion on matters biological, sometimes on subjects about which I consider myself an ignoramus. So I try and couch any answer with qualifiers reflecting my expertise, or lack thereof. Interestingly, even if I don’t know much about a particular subject, I have found my overall education has been enough to give me far more background than many of the laypeople asking the questions. It’s easy to forget just how poor basic science education is in the general population. So I make it incumbent upon myself to make sure that any answer I give is well-thought out, with full recognition of my limitations, because for all I know, my answer may actually influence a layperson’s vote on a crucial issue.

Sadly, all too often those who ought to be experts are willfully ignorant, and it falls to outsiders to do what they should have done.

Barbara McClintock met a chilly reception from geneticists with her evidence of "jumping genes". Evidence for regeneration in mammalian nervous systems was suppressed for forty years by the field's leading lights. Lord Kelvin held back quantitative electromagnetics as long as he was able. Social aggression in echinoderms was considered impossible ("they have no brains!") until someone made time-lapse films of it.

Astronomers still insist (in press releases) that comets originate in "the Oort Cloud" despite every scrap of actual evidence detecting no difference from any other asteroid, besides the eccentric orbit.

One could go on with examples, without pause, for decades.

Putting Bt toxin in plants might not harm most people, and might reduce pesticide inputs for a while, but it will certainly breed insects resistant to the principle pest control means used by organic farmers. Some people, including many experts, would consider that a Good Thing.

Expertise is always used in service of goals, which too often turn out to be evil or simply badly thought through.

Dave, I knew a civil engineer who always got asked about car repair. Nathan, "never attribute to malice what can be adequately explained by incompetence."

Ford: Incompetence doesn't prove benevolence.

Anyway, Exxon's profits are the highest, this year, that any any company has taken, ever. Is that evidence of somebody's competence? Incompetence? Malice? Benevolance? Certainly the voters were incompetent, and the Cheney is almost a cartoon of malevolence, but oil executives made sure that what they wanted to occur did, and the stockholders benefitted.

Once is happenstance, twice is coincidence, three times is enemy action.

Ford: Good eye man. Trewavas is a long time flack for big, like really big, ag. He is the rough equivalent to the ersatz scientists who front for tobacco and the petrochemical industry.

I agree that where there's a clear conflict of interest we should suspect "errors" may be deliberate. So, for example, risks of biotechnology need to be assessed by people without links to the industry. But is there really a sinister conspiracy to mislead us about echinoderms? The nice thing about science is that the truth comes out eventually. In the short term, of course, every grant or job that goes to a fraud deprives someone else of that benefit.

Don, I think you qualify as an expert on this.

Ford: Neither was there a conspiracy to keep Dr. McClintock's work down. Despite that, the experts of the time did, successfully, until her own students achieved positions of influence. Were the experts of the time incompetent? Far from it. Malevolent? Certainly not. However, if her work turned out to be important, they would likely be obliged to work on eukaryotes themselves, or others with more of such experience would do the work instead. None of them wanted that.

It's far from clear that the truth does always come out eventually. We know it does sometimes, but we don't know about the cases where it was successfully buried.

My conclusion is that the experts cannot be counted on for good advice either. Besides conflict of interest, there are very often institutional biases none of the experts can overcome, or care to. For fun, sometime, try to get an astronomer to say the word "current".

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Type the characters you see in the picture above.