gatherer research, for instance, can be completely missed, as it is in the perpetuation of this fallacy.
Clinical studies implicitly use those eating Western diets as the "omnivorous control" group. Typically, study after study or "reams of evidence" are cited as showing that veg*n diets are superior to "omnivorous" diets. But in fact, what the cited studies really indicate is simply that veg*n diets are superior to the deservedly maligned standard American/Western diet (SAD or SWD) with respect to certain degenerative diseases. That is, when not otherwise mentioned, by default those eating the SWD (or variants thereof) implicitly serve as the "control" group to represent "omnivores" in such studies.
Omnivorous hunter-gatherer diets differ from the SAD/SWD in two important respects. Yet just as typically, no mention is made of the fact that Western diets are considerably different than omnivorous hunter-gatherer diets, for example, in terms of the overall proportions and composition of foodstuffs besides meat. Equally as indiscriminate, there seems to be virtually zero awareness (since the subject is rarely if ever mentioned) of the considerable differences in the type of meat itself in such diets: that is, domesticated or feedlot meat in the typical Western diet vs. the wild game of hunter-gatherer diets.
Divergences in overall composition between Western vs. hunter-gatherer diets are covered in the section just below. The disparities between domesticated meats vs. wild game are discussed in the subsequent section, "Logical Fallacies and the Misinterpretation of Research," along with other issues that relate to the "Omnivorism = Western Diet" fallacy.
Hunter-Gatherer Diets vs. Western Diets:
Wild vs. Domesticated/Cultivated Foods
An important point to be noted concerning hunter-gatherer diets is that they are composed of wild rather than cultivated foods. Hunter-gatherer diets also generally exclude grains, legumes, and dairy, as they are the products of agriculture--something that hunter-gatherers do not practice.
The hunter-gatherer diet is generally composed of:
- Wild fruits, nuts, greens, tubers--raw and/or cooked.
- The lean meat and organs of wild animals.
- Insects (wild, though some tribes, e.g., the Ache of Paraguay, practice semi-cultivation of insects; see Clastres [1972]).
- Other foods--e.g., honey, as an example. (Note that wild honey usually includes the bee brood--bees in the form of larvae; reportedly it is the best-tasting part--at least that is what a friend who is a former instinctive eater relates.)
Most modern Western diets, in sharp contrast, consist of:
- Domesticated grains (wheat, corn, rice), tubers, and legumes (which present the problems of gluten, phytates, other antinutrients, and a high glycemic index for some tubers/refined grains).
- Meat from domesticated animals (high in saturated fat).
- Dairy (lactose intolerance, high in saturated fat).
- Cultivated fruits (bred for high sugar content; high glycemic index for some fruits).
- Hybrid vegetables.
- Manufactured and/or highly processed foods:
- Hydrogenated oils (synthetic fat, very difficult for the body to process).
- Heavily processed foods (additives, artificial ingredients).
- Large amounts of refined sugar (candy, cookies, other sweets).
Comparison of the above lists with the earlier table (herein) from O'Dea [1991] shows the dramatic differences between a potentially healthy omnivore/faunivore diet, i.e., the hunter-gatherer diet, and a diet generally regarded as unhealthy: the SAD/SWD.
The differences between wild fruit and cultivated fruits are discussed in the article, Wild vs. Cultivated Fruit, available on this site. The substantial differences between the meat of wild and domesticated animals is discussed in, "Disparities between Wild Game and Domesticated Meat in Fat Distribution and Composition," on this site (not yet available); additional information is provided in Speth and Spielmann [1983] and Naughton et al. [1986].
Eaton and Konner [1985] discuss the differences in protein and fat content for a Paleolithic diet of wild foods versus the SAD diet of cultivated and processed foods; see Eaton and Konner [1985, table 5, p. 288]. Their data estimate a Paleo diet as containing 34% protein and 21% fat by calories, while the SAD diet is 12% protein and 42% fat. The table also shows a ratio of polyunsaturated fat to saturated fat of 1.41 for a Paleolithic diet versus a ratio of 0.44 for the SAD. Hence, one might characterize a diet of feedlot meat (SAD) as high-fat, medium-protein, versus a diet of lean meat from wild animals (Paleo) as low-fat, high-protein.
| Logical Fallacies and the Misinterpretation of Research |
CLINICAL STUDIES, ANIMAL EXPERIMENT STUDIES, AND EPIDEMIOLOGICAL STUDIES
Introduction
Some of the more recent and more advanced comparative "proofs" of diet may cite as supportive evidence clinical studies, epidemiological studies such as the Cornell China Project, and/or animal experiment studies. A large number and variety of studies may sometimes be cited in comparative "proofs." However, sheer number of studies cited means little if the interpretations drawn from study results are based on fallacious logic and/or misinterpretation.
Common logical fallacies. Typically, such studies are presented as allegedly indicating that veg*n diets are healthier than omnivore diets. However, nearly all of such claims are actually logical fallacies. A short summary of such commonly encountered fallacies is as follows.
- Results from studies using domesticated or feedlot meat cannot be generalized to all omnivore diets. Results of experiments in which test subjects eat domesticated/feedlot meats are often interpreted--usually unknowingly, but at the least implicitly--as reflecting the results of all meats and/or all omnivore diets. However, the composition of domesticated/feedlot meat is very different from the composition of the lean wild animal meat (in terms of both overall fat level as well as type of fats contained) that was a part of humanity's evolutionary diet. (Not yet available: "Disparities between Wild Game and Domesticated Meat in Fat Distribution and Composition" explores these differences in detail.)
Eaton et al. [1996, p. 1735] note that the level of fat in domesticated meat compared to wild game is on average five times higher. Just as tellingly, the proportion of saturated fat as a percentage of total fat is five to six times higher in domesticated meat compared to wild game. And there are other significant differences as well.
Eaton and Konner [1985, p. 285] point out that EPA, an essential omega-3 fatty acid, constitutes approximately 4% of total fat in wild game, while in domesticated beef there is almost none. Eaton et al. [1996, p. 1736] further note that game meats are also high in other essential fatty acids such as arachidonic acid [AA] and docosahexaenoic acid [DHA], with the result that the character of meats consumed in hunter-gatherer diets results in high plasma levels of the aforementioned essential lipids compared to Westerners, and helps to maintain a dietary omega-6 to omega-3 fatty acid ratio ranging from 1:1 to 4:1--far below the 11:1 ratio typical of Westernized countries. The significance of the difference here as discussed by Eaton is that the high omega-6 to omega-3 ratio that prevails in Western countries may be linked with the promotion of cancer.
- Logically invalid extrapolation from SAD/SWD diets to all omnivore diets. Likewise, clinical studies comparing the SAD (standard American diet) or SWD (standard Western diet) to veg*n diets often assume that the poor results on the SAD/SWD imply all omnivore/faunivore diets would give similar results. That is, bad results on the SAD/SWD are cited as "proof" that virtually all omnivore/faunivore diets are "bad." There is no logical basis for this; it is an example of "unsupportable extrapolation."
- Results of epidemiological studies may be exaggerated. The importance and relevance of epidemiological data, especially the China Project, may be exaggerated by dietary advocates. This will be discussed in greater depth in a later section.
Let us now begin our summary exploration of these topics.
How reliable are animal studies that use domesticated/feedlot meat?
Due to the significant differences in composition of wild vs. domesticated meats, animal studies that use meat from feedlot animals cannot be used to reliably project the results of all possible omnivore diets. That is, the results of such studies are not representative of hunter-gatherer diets, or, for that matter, any other omnivore diet that excludes domesticated/feedlot meat. In other words, to project from animal studies using domesticated meats to all omnivore/faunivore diets is a logical fallacy--a fallacy one may find commonly practiced by raw/veg*n diet advocates. Some additional remarks on this issue are:
- Hunter-gatherers generally consume all (or nearly all) of the animals they harvest. (O'Dea [1991] discusses this point regarding the Aborigines of Australia.) Preference is given to internal fatty organs over the lean muscle meats. In contrast, test animals in experiments may be fed nothing but marbled muscle meats. (The nutritional composition of organs is itself different from the composition of muscle meat, not to mention marbled, domesticated/feedlot muscle meats.)
- In some tests, animals are put on a diet that may be far beyond their range of natural diets, e.g., putting a natural vegetarian (like a rabbit) on a diet of, say, cooked bacon. As the gut morphology of a rabbit is not adapted to a pure meat diet, the results of such tests must be interpreted with caution.
- In some tests, the animals used have been domesticated extensively, e.g., "nude" mice that are considered desirable because they readily show results to drugs. Obviously, the results of tests on animals that were bred to "yield results" must be interpreted carefully.
- The issues of reliability and projection in cross-species comparisons deserve some consideration. See Freedman [1988] for a discussion of the (technical) statistical issues involved in developing cancer models based on data from mice.
The above points suggest the following conclusions:
- Studies using domesticated or feedlot meat are not representative of all omnivore diets. Due to the many differences in the composition of wild vs. domesticated meats, the results of animal studies based on the latter do not necessarily reflect results one might get on a hunter-gatherer type of diet, or any other omnivore diet that excludes feedlot meats.
- Obviously, however, animal studies that use feedlot meat have relevance for the SAD/SWD diet, as such diets generally include feedlot meat.
Thus we observe that extrapolating from animal studies using domesticated/feedlot meats to all omnivore/faunivore diets is an unsupportable extrapolation and a logical fallacy.
Clinical studies based on the SAD/SWD diet
Narrow claims: Conventional veg*n vs. SAD/SWD. One does not have to look far to find raw/veg*n advocates citing clinical studies that show the negative health effects of the SAD/SWD diet. The more savvy/honest advocates are specific in their claims, and state something like, "Study X shows the (conventional) vegan diet promotes better health than does the SAD/SWD." Such precision in language is good; however, the fact that healthy omnivore/faunivore diets do in fact exist should also be mentioned, at least occasionally, for completeness and for honesty. (It is rare to find a raw or veg*n advocate who bothers to mention that healthy omnivore diets do exist.)
Fallacious claim: One type of veg*n diet vs. all omnivore/faunivore diets. The less savvy/honest raw/veg*n diet advocates point to clinical studies of veg*ns vs. SAD/SWD consumers, and then make the massive logical fallacy of saying that such evidence proves or indicates that the veg*n diet tested is better than all omnivore diets. It is a logical fallacy to assume that test results for one type of vegan diet versus one type of omnivore diet (the SAD/SWD, usually) indicate that the tested vegan diet (or, even worse, all vegan diets) are better than all omnivore diets.
The SAD/SWD is only one of a large variety of possible omnivore diets. For example, all of the diets in the China Project (discussed in the next section) are omnivore diets, and are generally different from the SAD/SWD diet. Obviously, omnivore diets vary considerably--just as vegan diets do.
Unconscious double standard. One criticism veg*n advocates make concerning some of the clinical studies that show negative health effects on veg*n diets is that such studies may use non-representative veg*n diets, like macrobiotics, as their sampling base. Along the same lines, note that the "conventional" vegan diet (one that makes use of grains, legumes, etc., usually cooked) is radically different from the 100% raw fruitarian diet. Further, 100% raw fruitarian diets have a dismal record of failure in the long run. The usual long-run result of 100% raw fruitarian diets (per anecdotal evidence) is ill health. Assume that a long-term clinical study of fruitarians existed, and it showed the negative health results so common in anecdotal reports. Would it be fair to extrapolate from the fruitarian diet and condemn all vegan diets? Not really--that would be a logical fallacy. In a similar manner, raw and/or conventional veg*n diet advocates who use clinical studies based on the SAD/SWD diet to condemn all omnivore/faunivore diets are engaging in a logical fallacy.
Examples of citing clinical studies based on the SAD/SWD diet
For examples, let's look at two of the studies cited by Ted Altar in his paper, What Might Be the "Natural" Diet for Human Beings. [Note: See reference list for link to paper on Internet]. Altar cites Friedman et al. [1964] as proof that red blood cells become "sticky and sludge" after a meal of animal flesh. Armstrong et al. [1981] is cited as evidence that anti-inflammatory hormones and sex hormones increase after eating animal flesh. Altar then claims these "immediate body reactions to meat" suggest that meat is not in accord with our evolutionary diet.
Study cited does not compare vegetarian vs. non-vegetarian diets. However, the above suggestion does not follow from the studies cited by Altar. Both studies used the SWD diet (Armstrong et al. [1981] was done in Australia). Friedman et al. [1964] compared two groups of volunteers, all eating the SAD diet, one group consistenting of "personality type A," the other group personality type B." Both groups were fed a meal of ~1900 calories, 67% fat (bacon, eggs, butter, milk, cream). Despite the massive dose of fat, sludging was observed in 10 of the 12 members of group A, but only 3 of group B. Thus Friedman et al. [1964] is focused on comparing two personality groups, A and B, given identical non-vegetarian diets. If a conclusion is to be drawn, the results suggest personality may correlate with the body's ability to handle ingested fat.
More important, the Friedman et al. [1964] study is not a vegetarian vs. SAD comparison study; it did not attempt to compare animal vs. plant fats, and so on. Also, the study used large amounts of feedlot meat and dairy from domesticated animals. Hence the results cannot be projected onto hunter-gatherer diets, other omnivore diets, or even veg*n diets (as none of the subjects were veg*ns, and the foods tested were non-veg*n).
SAD used incorrectly to impute results for all omnivore diets. As for Armstrong et al. [1981], the study compared levels of reproductive hormones in vegetarian and non-vegetarian Seventh-Day Adventist women in Australia. It is appropriate to let Armstrong et al. speak for themselves ([1981, pp. 761, 765)]:
The dietary data were such as to permit only qualitative assessment of total dietary fat intake...
In accordance with our original hypothesis, the vegetarians had a lower excretion of estrogens than did the non-vegetarians. The difference was small but significant and largely due to a lower excretion of E3 [estriol, a type of estrogen]...
The E3 ratio was less in vegetarians than in non-vegetarians, which is the opposite of what might have been expected from the evidence of other studies relating E3 ratios to risk of breast cancer...
The total urinary estrogen values were lower in the vegetarians than in the nonvegetarians, the difference being mainly due to the lower E3 excretion.
Armstrong et al. then suggest that the lower estrogen excretion in vegetarians may be due to metabolic differences in estrogen pathways, or to lower estrogen production in vegetarians. Note the remark above that implies the E3 ratio in vegetarians might be associated with an increased risk of breast cancer. Suddenly, the lower estrogen levels of the vegetarian diet are not as attractive as before.
Inasmuch as the Armstrong et al. [1981] study uses the SWD--in this case, the standard Australian diet--as comparison, once again the results cannot be projected to all omnivore diets.
Claims are not supported by the studies. Thus we observe that Altar's claim that these "immediate body reactions to meat" are suggestive of meat not being in accord with our evolution is fallacious. The evidence presented by Altar is based on feedlot meats in the context of the SWD diet--foods that are far removed from our evolutionary diet.
Please note here that I have used Altar's paper as an example because it cites only a few studies, and it is possible to examine those in detail. With a large number of citations, it is difficult to check them all. It is not my intention here to "pick on" Ted Altar.
GO TO NEXT PART OF ARTICLE
(Drawbacks to Relying Exclusively on Clinical Studies of Diet)
Return to beginning of article
SEE REFERENCE LIST
SEE TABLE OF CONTENTS FOR:
PART 1 PART 2 PART 3 PART 4 PART 5 PART 6 PART 7 PART 8 PART 9
GO TO PART 1 - Brief Overview: What is the Relevance of Comparative Anatomical and Physiological "Proofs"?
GO TO PART 2 - Looking at Ape Diets: Myths, Realities, and Rationalizations
GO TO PART 3 - The Fossil-Record Evidence about Human Diet
GO TO PART 4 - Intelligence, Evolution of the Human Brain, and Diet
GO TO PART 5 - Limitations on Comparative Dietary Proofs
GO TO PART 6 - What Comparative Anatomy Does and Doesn't Tell Us about Human Diet
GO TO PART 7 - Insights about Human Nutrition & Digestion from Comparative Physiology
GO TO PART 8 - Further Issues in the Debate over Omnivorous vs. Vegetarian Diets
GO TO PART 9 - Conclusions: The End, or The Beginning of a New Approach to Your Diet?
Back to Research-Based Appraisals of Alternative Diet Lore
