In a recent blog post, Dilbert writer Scott Adams wrote:

What’s is science’s biggest fail of all time?

I nominate everything about diet and fitness.

Maybe science has the diet and fitness stuff mostly right by now. I hope so. But I thought the same thing twenty years ago and I was wrong.

From there he goes on what can charitably be called a rant against science, arguing that the public is justified in not trusting the findings of science because science has been wrong before. Adams’ criticisms, however, are based largely in his own misunderstanding of science.

He makes two major errors in his analysis. The first is to confuse mainstream media reporting of science with the science itself. The second is to have an incorrect image of how science progresses over time.

Here is his list of examples where science allegedly got it wrong:

I used to think fatty food made you fat. Now it seems the opposite is true. Eating lots of peanuts, avocados, and cheese, for example, probably decreases your appetite and keeps you thin.

I used to think vitamins had been thoroughly studied for their health trade-offs. They haven’t. The reason you take one multivitamin pill a day is marketing, not science.

I used to think the U.S. food pyramid was good science. In the past it was not, and I assume it is not now.

I used to think drinking one glass of alcohol a day is good for health, but now I think that idea is probably just a correlation found in studies.

I used to think I needed to drink a crazy-large amount of water each day, because smart people said so, but that wasn’t science either.

Adams does have a point with his first example – in the 1970s and 80s preliminary evidence suggested that fat and cholesterol were bad, especially saturated fat. Dietary recommendations at the time included limiting total fat intake to 30% of total caloric intake. As is often true in science, the story is actually more complicated.

A recent review concludes that the recommendations at that time were not justified by the available evidence at that time. What we have learned since then is that the ratio of good fat to bad fat is more important than total fat intake. Dietary cholesterol is also not the problem we thought it was. Exercise and maintaining a lean body mass are probably more important than specific dietary factors. This does not mean, however, that you can eat as much fat as you want without consequences.

Further, Adams here is clearly listening to popular diet gurus, rather than following the more nuanced discussion happening in scientific circles. This is the core problem. If you look behind the curtain, scientists on such topics will have a range of opinions, the evidence is often complex and can be interpreted in multiple ways, and the current conclusions based on the evidence can be a moving target.

However, in certain areas, like diet and health, bottom line recommendations need to be made based on the best current evidence, even when incomplete. We can legitimately criticize government agencies for making premature recommendations based on the science. Also, popular diet fads and food industry marketing greatly exaggerate or exploit the science, turning preliminary findings into the low fat craze.

Adams is doing it again, countering the low fat craze with the more recent low carbohydrate craze. He is going to be disappointed again, because he is listening to self-help gurus rather than trying to understand what scientists are actually saying.

All his other points are far worse examples. There was never any scientific consensus on taking multivitamins. That was always supplement industry hype. He even says he knows it’s marketing, but does not seem to recognize the implications.

The US Food Pyramid is, at best, a representation of our current understanding of the science, even when we know the science is incomplete. In an applied science, we need to make decisions and recommendation based on incomplete science. When new evidence comes to light, the recommendations change. You might fault them for being too slow to adapt to changing evidence, but that is a criticism of a government agency, not the underlying science.

His point about alcohol is also valid. There was a debate for years about the possible health benefit of a small amount of regular alcohol intake. However, there was never a strong consensus around this conclusion, which was always controversial. This was based entirely on correlations in observational studies (you probably can’t ethically randomize subjects to drink alcohol). Observational studies are always tricky. In this case it’s possible that ex-alcoholics were contaminating the non-drinking group, dragging down their health outcomes and artificially making it seem that mild drinkers were healthier. When the ex-drinkers are eliminated, the effect tends to go away. Even still this is not entirely settled, but recent science has moved away from this hypothesis.

The problem here is that the media reports each small or preliminary study as if it is the consensus of scientific opinion. Maybe there is some token skepticism if you read deep enough into the article.

Drinking large amounts of water was always a myth, and never based on science nor officially recommended by any scientific organization.

In his list Adams gives a range of examples from outright myths, to premature hyping of preliminary findings, to government agencies making bad calls or even just properly reflecting the advance of our scientific understanding. All of this relates to how the media reports science, how the self-help industry exploits science and the public, how industries market pseudoscience or preliminary science, and the challenges of making recommendations based upon incomplete science. None of this is actually a criticism of science itself.

Adams gives a glimmer of understanding this point, writing:

So one could argue that the media and the government (schools in particular) are to blame for allowing so much non-science to taint the field of real science.

But he raises this objection, it seems, only to refute or ignore it. He dances around the issues, but then just comes to what seems like an emotional conclusion:

Perhaps my expectations were too high. I expected science to tell me the best ways to eat and to exercise. Science did the opposite, sometimes because of misleading studies and sometimes by being silent when bad science morphed into popular misconceptions. And science was pretty damned cocky about being right during this period in which it was so wrong.

The only legitimate point here I will grant him is that “science” (which I will interpret as scientists and scientific institutions) were “silent” in the face of popular misconceptions. I agree that scientists need to be much more proactive in communicating science to the public, countering pseudoscience, countering marketing exploitation of bad science, and correcting popular misconceptions. I do think the situation has improved in the last 40 years, but we are not there yet.

I don’t think he can justify his conclusion that science was “cocky.” If you read the actual literature a very different picture emerges. Again he is just confusing popular misconceptions with the actual science.

It gets worse:

The pattern science serves up, thanks to its winged monkeys in the media, is something like this:

Step One: We are totally sure the answer is X.

Step Two: Oops. X is wrong. But Y is totally right. Trust us this time.

The media is hardly the “winged monkeys” of scientists. The media is a problem that scientists have to manage. I will say that scientists and the press offices of universities also play their negative role in hyping research findings. I am not trying to blame all problems on the media. This is simply not a problem with the science itself, or the conclusions of scientists, but with how science is communicated to the public, at every level from scientist, to press office, to mainstream media.

Throughout Adams’ sloppy reasoning he is also consistently misunderstanding how science progresses. He argues that the average citizen is justified in not trusting science, because:

So how is a common citizen supposed to know when science is “done” and when it is halfway to done which is the same as being wrong?

This is wrong, and you can see this error wind its way through his entire argument. In his argument above, step 1 is not “we are totally sure X is right,” but rather, “the current evidence supports X to this degree.” Step 2 is not, “X is wrong and Y is right,” but, “X is incomplete, the story is more complex, and we have to modify X with Y.”

Science is not wrong until it is completely right. Rather, science makes ever more detailed approximations of reality. Science gets deeper, more nuanced, and more complex. It does not simply change one dogma for another separate dogma (as Adams’ “X” and “Y” examples imply).

With regard to diet, for example, Adams believes that science failed his generation with wrong recommendations, even going so far as to blame science for the increase in obesity and diabetes. Even in the 1970s, however, the science of nutrition had the basics correct. Most of the major nutrients were identified, we knew to a fairly accurate degree how much was needed, even leading to successful public fortification programs. We understood the macronutrients and their metabolism fairly well. The basic recommendations were to eat a varied diet, exercise regularly, and maintain a lean body mass. Those basic recommendations were true and still are true.

Things got more complicated when we tried to drill down deeper, adjusting, for example, the ratio of macronutrients to tweak vascular risk, for example. Scientists were moving beyond basic health to adjusting subtle risk factors. They correctly identified the role of fat and cholesterol, but the story was more complex. The type of fat was important, genetics is very important, and dietary cholesterol turned out to be unimportant. Still, their recommendations were not unhealthy, they were just not optimal.

Following the bottom line advice of the 1970s would not make you a fat diabetic. Adams is simply wrong in this accusation. Of course, there is disagreement over what caused the obesity epidemic, but no one is blaming the scientific recommendations of the time. The self-help diet industry is partly to blame, peddling fad diets rather the a sustainable healthy lifestyle. There is still debate about the relative contributions of increased sedentary lifestyles and increased caloric intake. The food industry has been feeding us more calories, while proclaiming their products “low fat” or “low carb.” It’s a complex public health issue, but the science is not to blame.

Adams goes on to make a third general error in his article. He says we need to ask why the public distrusts science, and then offering as the answer that science has betrayed the public trust. Actually, scientists have asked that question, and the answer is not what Adams claims. People distrust science when it conflicts with their valued beliefs, or when science suggests a solution or intervention that conflicts with their beliefs. People are happy to trust science when it does not conflict with their ideology or narrative.

In short – Adams gets it entirely wrong in his article. He misunderstands science, the history of science, and the relative roles of media and various industries in communicating and distorting science. There is a legitimate criticism to be made in terms of scientific institutions putting more resources and value into communicating science to the public, but Adams barely gives this point a glancing blow.

Ironically, Adams is contributing to the public misunderstanding of science in his article. He is contributing to the very problem about which he is complaining.