The death of a theory

Lagt ut av

If I haven’t seen further than others it’s because giants were standing on my shoulders« – Richard Feinman quoting Hal Abelson quoting his roommate. 

I’ve been reading a fantastic book lately. Lee Smolins “The trouble with physics – The Rise of String Theory, the Fall of a Science, and What Comes Next.” The book is exactly as great as the title sounds, although admittedly, I only understand a fraction of it. Lee Smolin discusses string theory, how is came to be the leading theory in theoretical physics and how it might prove to be one of the greatest dead ends in the history of science. I’m by no means qualified to discuss theoretical physics, but as always when reading science my head draws parallels to my own area of research. String theory, as the leading theory in theoretical physics, does have a lot in common with the leading theory in obesity treatment, the low fat theory.
There is little doubt that the “low fat diet for weight loss” theory is a theory that should have been rejected a long time ago. There are good reasons it should be rejected. It only indirectly addresses the physiological causes of excess fat storage. And because it only indirectly addresses the real problem, it only works temporarily.

«They will devise numerous articulations and ad hoc modifications of their theory in order to eliminate any apparent conflict.» – T. Kuhn

In the near future the low fat theory might be considered one of the greatest blunders in health science. At least, it has the potential to be considered as such. Some theories have been shown to be near immortal and this is definitely one of them.

It is easy to draw parallels between the low fat theory and string theory. String theory is an attempt to unite different aspects of physics, general relativity and quantum mechanics, into one great unifying theory, a complete theory of nature. It is based on simplicity and beauty, but what seems intuitively logic is not necessarily logic at all.

Before Keppler, the planetary orbits were thought to be circular. A circle is beautiful and symmetric and it seemed logic that this was how the planets moved. And yet, observations showed the planetary orbits to be elliptical. The low fat theory is also intuitively logic. Fat contains much more energy than other nutrients of the same weight and excess energy is stored as fat. Thus eating fat in excess makes the body store fat in excess. Fat from foods makes fat in the body. It is simple, beautiful, logic and wrong.

String theory, like the low fat theory, does not have a good track record as far as theories go. It started as a simple theory, and many people think that a theory that incorporates all of physics should be simple. But there were fundamental problems with string theory right from the start. In order to get it to work on paper a lot of dimensions had to be added (the 3 dimensions we are used to, were by far enough), new and unobserved particles were invented and the theory had to be background dependent to work when the whole point was that it should be background independent. It also made few predictions and proved close to impossible to falsify by experiment. A theory has to be falsifiable, because only by opposing repeated attempts of falsification does a theory evolve into truth.

«A nice adaptation of conditions will make almost any hypothesis agree with the phenomena. This will please the imagination, but does not advance our knowledge.» – J. Black

Still, string theory survived. It survived because it was constantly added ad hoc additions and because conditions were constantly changed. Now, this is a normal scientific process. Few theories are perfect when they emerge. The question is, how many conditions can be changed and alterations be made before the theory should be replaced by a new one?

Like the string theory, the low fat theory needs to rest on several assumptions in order to survive. The low fat theory is based on the belief that energy intake and energy expenditure are independent factors. Little scientific data support this, and it is not possible to consider the body to be a closed system. It is also based on the assumption that we can all control our energy intake and expenditure by will. Low fat diets don’t work in the long run. Upon discovering this we can change the theory or consider it faulty. Because it didn’t work, modifications were made; People are lazy, have poor self control, exercise too little and so on. None of these assumptions are justified. They are often not given directly, but disguised by the fact that they are logical consequences of the theory.

Both the low fat and string theory were the leading theories in their field for a long time and in many ways still are. Other theories, even better theories, has constantly been placed in the shadow and were given little financial support. People who worked on alternative theories to string theory were automatically outsiders, low in the field’s hierarchy, just like the researchers working on low carb diets was given less attention than they deserved.

This is painful for many who have invested years and even decades of their working lives in string theory. If it is painful for me, imagine how some of my friends who have staked their whole careers on string theory must feel. Still, even if it hurts like hell, acknowledging the reduction ad absurdum seems a rational and honest response to the situation. It is a response that few people I know have chosen. But it is not one that most string theorists choose.”  – Lee Smolin

Although there’s still some hope for string theory, there is none for the low fat theory. We’ve tried it. It didn’t work. We are getting fatter than ever. Often the scientific standards themselves are reduced in order to keep these gargantuan theories alive. Intentionally or not, the result of this is sometimes the death of science itself and the people considered scientists are no longer scientists in the true definition of the word. The death of a theory is replaced by the death of science.

It doesn’t take an expert
There are usually no sides in science. Sometimes people ask me; what do you believe in/ who do you believe to be right or what dietary method do you teach? The way I see it the only correct answer is that I don’t believe anything. I know what some of the facts are. I can make some calculated guesses on what is less certain, and there is a lot I don’t know. Of course, I cannot actually give this answer. People need substance, and as soon as I mention carbohydrates, I’m on the low carb side. But there are no sides. Science is not a battle between teams, it is a unified search for truth.

The reason that mathematics invented the idea of proof and made it the criterion for belief is that human intuition has so often proved faulty– Lee Smolin 
Basic scientific principles are not difficult to understand and they are independent of scientific subject. When scientific standards are reduced, any lay person can see the faulty logic. Tom Naughton is being accused of writing about things he does not have the authority to write about. But, he does know what he is talking about, much more so than many of the experts in the field. You do not have to be a nutritionist or even a scientist to address the core problems related to basic scientific principles. A beginners mind easily spots the obvious flaws. Diabetes means you can’t handle dietary carbohydrates, yet a doctor may easily tell you to cut fat from your diet after diagnosing you. The doctor’s mind is that of an expert, clouded by experience.

«…insist that we should change the rules of science just to save a theory that has failed to fulfill the expectations we originally had for it. – Lee Smolin

Why aren’t we all fat?

I have noticed that there is one argument used in the nutrition debate that is used by both the low carb and the low fat community (in this context, dividing it into two sides make perfect sense). The same argument keep flying back and forth but I feel it is rarely given the response it deserves.

It goes something like this; “If carbohydrates make us fat, why aren’t all the people of the world, like many Asian people, who live on high rice diets, fat?”

For the record, I am fully aware that the composition of a traditional Asian diets is debated and that a general shortage of food may protect against any harmful effects of a high carb diet. The above argument is a poor argument for another more important reason.

In the low carb world the same argument is used, often in a form resembling the following; “Eating fat cannot make us fat. Because, if fat made us fat, why aren’t all people who live on regular high fat diets, like the Inuit or the Masai fat?”

This argument (and all similar) is easily refuted. Part of the reason these arguments are fallacious is that they rely on an unstated assumption, an assumption we cannot make. A causal factor may affect us differently and may only end up in full blown disease in predisposed individuals. We are not genetic copies. If we were all clones, finding causal factors would be so much easier (there is a reason twins make for popular study subjects). But we’re not. We respond differently to the same stimuli and only by being clones would we all be able to react similarly.

Smokers have a greatly increased risk of lung cancer, yet many smokers smoke their way through long lives with no cancer.

Still, the causal factor is still the causal factor. Insulin and glucose drive fat storage in all of us. Some of us however have counter regulatory mechanisms that may overpower the fat storing effects causing us to remain lean. The above arguments rest on the assumption that we are genetic copies. We’re not.

Ok, I realize the clone analogy is stretching it a bit. After all, as members of the same species we are very similar and we may easily all respond similarly to the same stimuli. But the truth is that the arguments do rest on the assumption that we will respond in the same way and the danger is that the assumption will be used to discard a theory when responses differ.

In some studies of low carb diets the results have been less than expected. My experience is that many of these studies are also easily criticized, but that doesn’t really matter. It also doesn’t matter that studies using body weight may not detect large alterations in body composition. What matters is that the cause for a disease does not necessarily make everybody exposed to the causative agent, sick. Carbohydrates may easily be the main cause of excess fat accumulation in fat cells even if not everybody who eats great amounts of carbs gets fat. We cannot simply reject a theory that points to carbohydrates as the main cause of overweight just because there are people who don’t get fat in the face of plentiful carbohydrates.

«If any and every failure to fit were ground for theory rejection, all theories ought to be rejected at all times.» – T. Kuhn 

Imagine a disease spreading through society and everything points to a virus as a cause. Then some people argument that the virus cannot be the cause of the disease, simply because there are people being exposed that does not get sick. It makes no sense.

On a similar note, the fact that you can lose weight by starving yourself (eating less energy) does not in any way contradict a theory pointing to carbohydrate as the causative agent.

The simple and undeniable fact is that many high quality studies of low carb diets have illustrated the diets efficiency in reducing weight. It makes perfect physiological sense and even evolutionary perspectives support it. These results cannot be rejected simply because someone loses weight by eating less energy. The results still stand. Getting them to fit with results from energy restriction studies is a matter of physiology, not theoretical science.

So, although carbohydrate restriction does not always give the expected positive results, this is not ground for theory rejection. On the other hand, when low fat diets have never given the long term results the theory predicts, this is ground for rejection. Treating overweigh people with low fat diets is a prime example of treating a symptom, the fat stores, rather than treating the cause of the fat storage. Because this is a strategy focusing on a symptom is will not work long term. Science supports this. Low fat diets simply do not work long term. We must reject the theory and try something else.

Is there a chance that such observations have already been made but ignored because, if confirmed, they would be inconvenient for our theorizing?” – Lee Smolin

9 kommentarer

  1. Nice post. Interesting regarding Tom Naughton getting taken to task about not being 'qualified' to write what he writes. I find these arguments of authority tiresome. Even when you are a nutritionist, the authority card gets played. I am a nutritionist, not a dietitian, so I automatically don't know what I am talking about. I would imagine that if I was a dietitian, I wouldn't be a doctor. And if I were a doctor, I wouldn't be a researcher… where does it end.

    Such is human nature I guess… when I was coaching track cyclists differently from convention, I apparently had no authority to do so – I was only a sport scientist, not a cycling coach, nor had I competed in track cycling.

    The problem is not that I disseminate false information, but that I am doing things differently and in a way that might undermine somebody else's authority. And some (many) egos cannot handle that or admit they are wrong.


  2. And here's another one by Peter Woit: Not Even Wrong: The Failure of String Theory and the Search for Unity in Physical Law.
    A theory described by Wolfgang Pauli as «Not Even Wrong», meant its hypothesis was not falsifiable in Popper' s sense.

    And QED by Richard Feynmann, the physicist who worked on the atomic bomb – no relation to Feinman. Not much in in nutritional science can be said to be QED, as too many extend the meaning and strict conclusions of a study well beyond the conditions of the experiment. There is the principle of «ceteris paribus» – all other (things) being equal ( kept the same).


  3. Ah! So that is what was meant by the Woit title. Heard about the book and I think I'm going to have to read it next.
    True about the overextending of results in nutritional science. Although I think we're all guilty of it to some degree, I also think some overextending has do be done. The ceteris paribus principle breaks down in confronting a complex biological machine like the human body. Nothing is ever kept the same.


  4. There is a wordplay by Feynmann in his use of QED – Quantum ElectoDynamics vs Quod Erat Demonstandum.

    But then the guy played the bongos and also used his code breaking skills to break into his colleagues' wall-safes when he worked in Los Alamos during the war. A very clever iconoclast – he would detest string theory.

    Ol' Briffa is guilty of confirmation bias when commenting on a study – he uses the notion of «common sense» to add a no-study supported hypothesis re insulin BG levels and fasting insulin to accept the causation from the epidemiological correlation in the study.

    This is an example of over-extending.

    Hey, I agree we are all guilty of it to some degree, but school debates and their point scoring is not science.

    «…the more insulin one secretes over the course of one’s life, the more likely one is to become insulin resistant. Eating less blood sugar-disruptive carb could, therefore, lead to generally lower levels of insulin and less risk of insulin resistance. . . . . common sense dictates that a carb-controlled diet should help prevent type 2 diabetes.»

    He then suggests that risk ratios of 1.2 or 1.27 represent for the proffered conclusion.

    Unless I see risk ratios in excess of 10, I yawn and say, «1.27 ?, hmmm, these are the sorts of RRs quoted by statinators to justify taking statins». Need I say more?

    (B Sc Physics, Maths; M Sc Statistics – now in my pre-dotage!)


  5. Great post Pal! I've been always wondering if there are true genetical differences regarding carbs tolerance between populations. E.g. it seems to me that populations from South Europe are more tolerant than the North Europe Populations. That could be the reason why people in Italy, Spain, Greece or even France do not get as obese as others. What are your thoughts about that?


  6. Thank you Aitor!
    Genetics is not my strong side. I've been wondering about the same thing myself. But you can't say something about a persons carb tolerance by looking at his genes. You could perhaps say something about his risks or potential for having a good or bad carb tolerance. But the genetic potential is influenced by so many lifestyle factors.

    There are examples of populations living on high carb diets that does not seem to be negatively affected by this, but is doesn't mean their genes are to blame. There are so many other factors that could explain population differences.

    I would not be surprised, however if anyone showed me data that clearly illustrated population differences in carb tolerance.
    Genetic changes need not take as long as sometimes assumed:

    But as I said, this is not my strong side. There far more qualified people out there that might have an answer.

    By the way, do people in Spain and Greece not get as fat as others?


Legg igjen en kommentar

Fyll inn i feltene under, eller klikk på et ikon for å logge inn:

Du kommenterer med bruk av din konto. Logg ut /  Endre )


Du kommenterer med bruk av din Twitter konto. Logg ut /  Endre )


Du kommenterer med bruk av din Facebook konto. Logg ut /  Endre )

Kobler til %s