The China Study and Denise Minger

Posted by

Denise Minger and her critique of the China Study (which advocates vegetarianism) – “Correlation is not necessarily cause”.

Full disclosure:  I read the China study in 2007 and thought it was junk.

Amazon tells me that I originally purchased the China Study in Oct. 2007. Wikipedia tells us that the book has sold over a million copies. The China Study purports to make the case that animal protein and saturated fat cause disease, whereas a high carbohydrate, whole food, plant-based vegan diet does the opposite. It is the vegan’s bible and is endlessly trotted out, sometimes quite aggressively, by vegans to defend their own chosen lifestyle and proseletise. So it is an important book – but for the wrong reason in my opinion. Its importance lies in the fact that it is a best seller and has been hugely influential, not because Campbell gives convincing scientific evidence that animal protein causes disease.

Denise the self-confessed super nerd.

I was just going to give a link to Denise’s blog  . From what I remember, Denise (a reformed raw vegan) made herself an expert in epidemiology and looked at the  raw data in the China study  and found that Campbell’s conclusions didn’t stack up.  She also looked up the original paper where rats were tortured by being fed aflatoxins and then fed casein and gluten to see how quickly they developed liver cancer and exposed the flaws in this and Campbell’s own subsequent lab experiments with rats.

Was Denise’s criticism well founded?

I then realised that I needed to look again properly at Campbell’s book for my own education – I only skimmed it when I purchased it originally in 2007 – to see if Denise’s criticisms were justified. So this is what I am doing as I write this. I am going to write my own critical review before looking again at Denise’s blog and then, once I have written it I will look again at Denise’s critique and compare and contrast it with mine and try to resist the temptation to improve my own review after reading Denise’s ;-). I will also give a link to Chris Masterjohn’s and Loren Cordain’s critiques as well as Campbell’s reply to Denise’s critique.

Campbell: White hat or black hat?

Campbell would like to be seen as one of the good guys. (I’m not so sure – see my conclusion.) The book generates a good deal of venom on blogs and forums of the opposing camps (meat eaters vs vegan). As Denise Minger constantly reminds us in her blog entries, we should reserve our opprobium for those vested interests that put their own profits above our health. What we should all have in common is that we refuse to be brainwashed by the propaganda put out by agribusiness, big food and big pharma. We are all (Seignalists, vegans, paleos, raw foodists, wpf ‘ers, natural hygienists e tutti quanti) trying to uncover for ourselves the secrets of how diet affects health and put what we learn into practice in our own lives. So we should not get bogged down in rigid dogma because even post Seignalet, we all (including Campbell) have a lot left still to discover.

Campbell: Long on hyperbole, short on substance.

On page 103 we read: “What are the odds that all these associations (and many others) favoring (sic) a plant-based diet are due to pure chance? It is extremely unlikely, to say the least. Such consistency of evidence across a broad range of associations is rare in scientific research. It points to a new world view, a new paradigm. It defies the status quo, promises new health benefits and demands our attention.”

“Grains super healthy”.  Oh really?  Which ones?

That a diet high in plant based whole foods, ie. fruits and vegetables is healthy is a no brainer. (The paleo camp would agree I’m sure). However, Campbell includes grains in his ideal diet. There are hundreds of pub med articles proving the dangers of gluten for example and in the preceding pages, Campbell has produced mainly waffle on his assertion that animal protein promotes cancer while vegetable protein prevents it . He produces no real concrete evidence (apart from some very dodgy studies on rats) to support this sententious conclusion.

Campbell’s Generalisations without substantiation

On page 116 he writes: “….The cultures that have lower heart disease rates eat less saturated fat and animal protein and more whole grains fruits and vegetables. In other words they subsist mostly on plant foods while we subsist mostly on animal foods….” The book is full of generalisations like this. Where is Campbell’s proof for such assertions? According to a UN report: “The Atlas of Heart Disease and Stroke”, the Japanese have one of the lowest rates of coronary heart disease in the world yet they eat plenty of animal foods, particularly raw fish. The Okinawans have an even lower rate and they eat a lot of pork (like the Chinese) and they use pork fat for cooking.

In Chapter 5. on page 120, there is a chart which claims to show a strong correlation between the consumption of animal protein and heart disease: “Chart 5.3: Heart disease death rates for men aged 55 to 59 years and animal protein consumption across 20 countries”.

Seignalet:  not all proteins are created equal

1. The chart (if accurate) does indeed show a correlation. However, not all animal protein is the same. For Seignalet, animal protein cooked at high temperatures is the disease factor, not animal protein per se. Raw or very lightly cooked (ie. steamed or similar) animal protein should be eaten once per day according to Seignalet. Seignalet considers coronary disease to be a “Clogging disease” and in his clogging diseases table under the heading “Heart attacks prevented” he lists 5 heart attacks amongst 1,200 patients treated for other clogging diseases when statistics would predict 28. Interestingly, given Campbell’s insistence that animal protein leads to elevated cholesterol, Seignalet lists 70 patients with high cholesterol in his clogging diseases table. They showed an overall 35% reduction of cholesterol on his (non vegetarian) diet.

2. “Animal protein” across China would, I’m guessing, include a large proportion of cooked pork. The Weston Price Foundation has discovered that cooked pork causes blood cells to clump.

“Correlation is not alwåys cause”

Could the blood clumping problem with pork just on its own explain Campbell’s correlation between animal protein consumption and heart disease in China and the other diseases that Campbell discusses? Or could it at least be a factor? Remember: “correlation is not always cause”  Here are some Weston Price Foundation recipes which are touted to avoid the blood clumping problem by the way. Note that only the Adobo
recipe would be (just about) Seignalet legal. According to the WPF study, cured pork does not cause clumping. Seignalet allows charcuterie (French for salami and other cured meats) . I, personally, put the salami (saucisson) etc. in the freezer for a couple of weeks to kill any potential parasites. Another tip that I learned from reading Sally and Mary’s book.)

Campbell confusion

Campbell cites various studies and programs (without citations) where people with heart disease have reversed their condition eating a low fat, whole food, plant-based diet. Could this be due to the fact that they are avoiding two Seignalet disease factors: milk and dairy and animal protein cooked at high temperature? I’m sure they would have had even greater success had they also avoided grains, apart from rice, quinoa and buckwheat (grains allowed by Seignalet) and eaten raw or very lightly cooked (steamed or similar) animal protein.

Where are the citations?

Campbell continues in similar vein with chapters on Obesity, Diabetes and Common Cancers and cites studies where according to him, a plant based, high carbohydrate diet has success in preventing or treating these conditions. In “Chapter 7 Diabetes” (p. 151) he quotes the impressive results obtained with Diabetes 2 patients by James Anderson M.D. using a high fibre, near vegetarian diet with some “cold cuts” (my underline emphasis) allowed. (What cold cuts are these? Are we talking charcuterie?) Again, in Seignalet terms, could the results be partly explained by the avoidance of cooked meat and milk and dairy? For Seignalet, diabetes is another clogging disease and his disease results table for this shows out of 25 diabetes patients, 20 had complete remissions and 5 showed 50% improvement. Seignalet attributes diabetes to clogging of the pancreas and muscle tissue by extraneous molecules which are gradually eliminated from the cells once his diet starts to take effect and the gut heals.

A few good bits

here are some excellent sections in Campbell’s book. His chapter 9 for example on Autoimmune diseases with what appears to me (a non expert) to be a good vulgarisation and explanation for dummies of how the immune system works and this particular paragraph comes very close to Seignalet’s “Xenoimmune” theory of the pathogenesis of Autoimmune diseases. (Campbell,p. 186):

“………..It so happens that the antigens that trick our bodies into attacking our own cells may be in food. During the process of digestion, for example, some proteins slip into our bloodstream from the intestine without being fully broken down into their amino acid parts. The remnants of undigested proteins are treated as foreign invaders by our immune system, which sets about making moulds to destroy them and sets into motion the self-destructive autoimmune process.”

For Campbell, the main culprit in this process is milk. (For Seignalet of course, other foods are also implicated). Campbell quotes various studies which show a link between milk consumption and autoimmune disease.

So both Campbell and Seignalet ban the consumption of milk and dairy. Interestingly, neither one makes any distinction between raw milk and pasteurised milk. For the Weston Price Foundation of course it is mainly pasteurised milk which is the culprit with the high temperature treatment of pasteurisation denaturing the proteins, killing the enzymes and milk’s natural probiotics and generally turning raw milk into a milk coloured and flavoured analogue. I don’t think that Cordain or Wolf would find anything to argue about with Campbell on the subject of milk.

Campbell is also very good on how American nutrition and medicine are corrupted by the drug and various food industries, how academics are bought and sold by these interests and how the media diverts or distorts any message which might threaten big pharma and big food. Campbell uses the opportunity to settle quite a few old scores with rivals and I found his bitchiness quite amusing (which I don’t think was his intention).

conclusion the book is a curate’s egg (look it up), good in parts;

  • Good: Good vulgarisation and explanation of autoimmune diseases and the immune system, vitamin D metabolism, an insider/whistle blower’s view of corruption of government and academia by food and drug interests.
  • Mediocre: Anyone or anything associated with Campbell is showered with superlatives. His graduate students are always “brilliant”, doctors who use a dietary approach he approves of have “exceptionally impressive results” and “top rank” with “….the accolades, the titles and the awards” etc. Campbell “had the opportunity to study diet and disease at some of the great medical history libraries in the Western World”. The epidemiology study in China was a “massive human study”, “exceptionally diverse in scope”. His findings were “published in the highest quality scientific journals” and “as you will see, these findings are nothing short of spectacular”. It all starts to grate a bit.
  • Bad: The whole time I was reading the book, (and this reminded me of why I just skimmed it when I bought it in 2007), I was searching for these “findings (which) are nothing short of spectacular”. All we have are Campbell’s endless generalisations, liberally spiced with superlatives. Where’s the beef? The “exceptionally” thin sliver of vegetarian meat substitute consists of two things:

a) An obscure Indian study where rats were poisoned with aflatoxin (a toxin in mould which silently kills liver cells) and one group was fed gluten and the other was fed milk casein to see which group developed liver cancer the quickest. Apparently it was the casein group which succumbed first. On this, Campbell constructs his whole theory: plant protein = good, animal
protein =bad.

Have rats (contrary to humans) evolved to thrive on gluten?

Having read Dr. Jean Seignalet’s book, it occurs to me that from an evolutionary point of view, grass seeds would have been much more of a rat staple than for humans before the neolithic period and from the dawn of agriculture onwards, their population explosion would have exceeded our own as they raided our granaries. Rats only live between six months to a year but can have many litters so there have probably been hundreds of times more generations of rats than humans since the dawn of agriculture and thus a much larger pool of subjects for mutations to take place that selected for digestive enzymes that break down gluten properly. Compared to our own evolution then, much more opportunity for the rat metabolism to evolve to thrive on a diet of wheat grain, stolen from our stores. So gluten, I’m guessing, would be a very natural food for rats whereas cows’ milk would certainly not be.

Have rats have evolved with digestive enzymes specifically designed to deal with gluten and other grain components but certainly not to deal with cow’s milk? Rats are weaned at 3 weeks old, so not only is it unnatural for an adult rat to be eating a milk protein isolate (in the same way as it is unnatural for adult humans), it would be even more unnatural for a rat to be eating milk protein from a different species of mammal, many thousands of times its own weight. The problem with milk comes not from the fact that it is animal protein but that mammals have evolved to suckle from their own species and have not evolved to cope with drinking milk as adults, yet alone from another species. Throughout the China Study, Campbell rails against “Scientific Reductionism” and uses it to put down anyone who does not agree with him. (When he is not accusing them of having been bought!) But how reductive is it to draw the sweeping conclusion that animal protein causes cancer and disease and vegetable protein does not from this pointless torturing of rats?

This is the study in question:

Archives of Pathology 1968 Feb; 85(2):133-7

The effect of dietary protein on carcinogenesis of aflatoxin. Madlavan TV, Gopalan C

There is no abstract and no article available on the online pubmed database. To get hold of the article or any other article for that matter, your local University library should be able to give you a visitor’s card and once you have the card they will get hold of the article for you.

b) Where are the “spectacular findings” produced by this epidemiological study in China: “The Most Comprehensive Study of Nutrition Ever Conducted”? Nowhere to be found! There are just 7 pages at the end of the book under the heading “Appendix B” “Experimental Design of the China Study” filled with the usual Campbell superlatives and waffle. That’s it??? He produces absolutely nothing from the study to support his contention that animal protein = bad and vegetable protein = good
It was only through an obscure reference note in chapter 4 that I could find the actual reference for the study, published 16 years before Campbell’s best seller. The actual study was done 20 years before! Here it is:

Diet Life-Style and Mortality in China (

available second hand on for a measly $492.

Here it is on for £992 new and £586 second hand. (My tip: buy from .com)

Even better tip: Read it for free here (found on Denise’s blog):

Campbell talks contemptuously in the China Study about people like Atkins who founded an empire and became rich on the back of a fad diet book. Could it be that Campbell was setting out to write just such a book and become rich himself? His book has sold over a million copies according to Wikipedia so he has succeeded if that was his aim.

OK. That was my review. I promise not to touch anything after reading Denise. (Apart from the above link).

Here is Denise’s critique of the book on her blog and more importantly, her analysis of the raw data in the original study and surprise, surprise, there is nothing in the data to support Campbell’s assertion in his book that animal protein causes disease and cancer whereas vegetable protein does not.

Here is Campbell’s reply on (get the catty introduction by vegsource)

More waffle liberally sprinkled with superlatives from Campbell. I had to laugh when he accused Denise of Hyperbole.

and here is another of Campbell’s replies:

and here is Denise’s critique of Campbell’s critique of her critique:

and I had to include this article by Denise on the WPF website, with Denise at her witty and acerbic best:

Here is Chris Masterjohn’s critique.

Here is a written debate between Loren Cordain and T. Colin Campbell.

These opinions are the author’s own (Chris Parkinson’s) and are not necessarily shared by the Seignalet family.