Sunday, August 19, 2012

"Experts Say"




  
An article titled "Is the era of oil nearing its end?" appeared in the Idaho Statesman a week ago (8/12/2012). It was from McClatchy Newspapers (the Statesman is a McClatchy paper) and written by Greg Gordon. It concerned the fear that despite the assurances of the energy companies and the bland lack of alarm on the part of the current administration and the one just past, we are speeding into very frightening times from a petroleum perspective. Mr. Gordon employed a phrase in the article that always irks me: "Experts say...."

Our era exemplifies of one of the fundamental errors in the proper use of reasoning: argumentum ad hominem. This principle identifies the tendency we have to consider any opinion false if it is articulated by someone we don't like. If we are Republicans, anything Barrack Obama says is a lie. If we are Democrats, anything Mitt Romney says is a lie. If Mitt Romney says the sky is blue and we are Democrats, it is most assuredly green. And vice-versa of course.

And now we have also created an interesting corollary to argumentum ad hominem: The Cult of the Expert. If we bestow the title of "Expert" on individuals, whatever they say regarding the subject on which they are an expert is true. If a person is not an "Expert," his or her opinion on the topic in question is of no value. In contrast to argumentum ad hominem, which is employed by everyone in just about every context, Cult of the Expert is observed almost exclusively in political debates and in journalism. It is used like a hammer to try to silence opposing views, but it does not reflect objective reality to the extent hoped for by those who brandish it.

To journalists the "Expert" is virtually always a tweed-jacketed university professor who has written a book about the issue in question. It's not too unreasonable an assumption. Professors are always intelligent, often brilliant, and you can get an idea about how accomplished they are from their peers. There are less well known people in the private sector, however, who may have as much or more expertise on a subject than an academic. The academic environment requires the professor to publish papers, but industry requires tangible results. Even if measurable results do not rise to the level of publishing a paper in a peer-review journal, and that would be a difficult argument, industry experts should not be summarily dismissed by journalists. And yet they are.

In the article I referenced above, Gordon implies that whatever the industry people say is probably false because of their strong financial interest in the ongoing use of fossil fuels. The assumption is that an academic would be free of that kind of bias in the formulation of their opinions. It is a dangerous assumption.

Academics, just like everyone else, have interests they have a need to protect. Among them are career, income and funding for their research. Biases they are not fully aware of can insinuate themselves into their theories and influence elements of their research. We once had a spirited discussion at Journal Club at the Medical University of South Carolina. The resident physicians and medical students were discussing an article written by a VA physician comparing new, expensive antipsychotics to the older, cheaper ones. The author of the paper described data demonstrating that the older antipsychotics were tolerated as well as the more expensive ones but they did not cause nearly as much weight gain, a serious problem for patients with mental illness. The resident physicians and medical students eagerly embraced the results of this article because the academic VA researcher would have less bias than researchers who worked for the pharmaceutical companies. "If this researcher can show that cheaper medications work as well as the newer ones with less side effects," I pointed out, "his career and reputation will skyrocket." Career is a powerful motivator in the world of academics, and every researcher seeks in his or her work to identify the need for yet more funding.

A Special Report in the September 2012 edition of Consumer Reports "On Health" provides advice to lay people on how to read news stories about medical science ("Should you trust that medical news?"). Among questions we are urged to ask include whether or not other sources are queried and who exactly is quoted and what are their credentials. These kinds of questions are ways to discern whether or not there is an undetermined bias in the opinion at the center of the reporting.

It is not reasonable to say that because a researcher works for a pharmaceutical company, his or her research demonstrating the value of the company's new drug is not useful. That would be argumentum ad hominem, would it not? We recognize however the possibility of bias and anticipate other studies confirming the findings. It would also be argumentum ad hominem to assume that the work of university professors is so influenced by their need to be promoted and tenured that their work is worthless as well. But to assume that because an academic is not directly supported by industry (a huge percentage of them in fact are) that there is no bias in their work is misinformed.

The use of the phrase "Experts say" is based on the assumption that virtually all of the Experts in a particular field agree. This assumption ignores one of the great truths of academic life: You don't get a Nobel Prize by agreeing with what everyone else in your field says. While university professors tend to speak with one voice on scientific issues that are heavily politicized, the laboratory and conference room witness constant argument and debate. When journalists cite the opinions of experts, they generally do not do a good job of telling us how much confidence we should have in an Expert's opinion and why. We do not get a sense of how much dissent there is among colleagues on the topic in question and what the stakes are for each camp. If you read science journalism critically, the phrase "Experts say" seems to have essentially one meaning: "In my opinion."

(c) Copyright 2012 Robert Albanese