Book Review: Moalem, Survival of the Sickest


Survival of the Sickest is a collection of eight pop-sci essays on medicine from an evolutionary perspective. It does not present any single cohesive line of argument, but the book's title refers to one of the main themes: the idea that common hereditary diseases would not have become widespread in the gene pool unless they once conferred an adaptive edge on individuals.

I read the book quite avidly and it is unlikely to disappoint anyone with an interest in the subject. Yet still I feel that it's a flawed piece of work in two important and interlinked respects: scientific credibility and writing style.

The subtitle of the book is "A medical maverick discovers why we need disease". The publisher clearly feels that Dr Sharon Moalem is a maverick, and I suppose he must agree at least to some extent or the title would have been different. Now, the book's target audience is clearly educated lay readers such as myself. I know very little about evolutionary aspects on medicine. But when I approach an interesting new field, I don't want to learn the controversial ideas of a fringe maverick. I want to know the current consensus among respected scholars, just like I don't like to turn to maverick plumbers, maverick dentists or maverick auto mechanics for a professional opinion.

Moalem's maverick status means that reading the book feels like walking on thin ice: whenever he says something that surprises me, I wonder, "is this accepted knowledge or a controversial hypothesis? What do non-mavericks in the field think?" To his credit, Moalem has provided the reader with an ample reasoned bibliography, documenting page by page where his ideas come from. But I don't want to read twenty books on the subject, interesting though it may be: I just want to know what the current consensus in the field is. Moalem could have made this easier for the reader by indicating clearly which of his beliefs are maverick ones.

The credibility issue is aggravated by a breathless, over-enthusiastic writing style similar to that of really bad TV commercial voiceovers. This is probably not Moalem's own fault, but that of his collaborator Jonathan Prince, "whose inspired writing helped to elevate this project immensely". Unfortunately, Mr Prince's writing appears to be inspired by the sales patter of his local used-car dealer. There's a good word for this in Swedish: flåshurtig, "so perky that he's panting".

I don't have the necessary knowledge to judge whether hereditary diseases spread in the gene pool because they were once adaptive. I suspect that they might spread because they are not maladaptive enough and happen to be linked fortuitously in development to some adaptive trait. To my mind, it seems highly unlikely that diabetes would have helped Mesolithic Scandinavians survive extreme cold thanks to the anti-freeze effects of high blood-sugar levels, as suggested by Moalem.

What I do know, however, is that an auxiliary pet theory of Moalem's is wrong. He believes that common and long-lived cultural traits, including folk medicine and regional cuisine, must be adaptive. This has been amply disproven. People will stick to the most insane customs and traditions for centuries and millennia though they may in fact be quite harmful. It's the same thing here: a trait may become widespread and long-lived through not being maladaptive enough while being linked to an adaptive trait. If a tribe believes that a man can only use a particularly efficient kind of bow for hunting if he sacrifices the end of his left pinkie to the Goddess of the Woods, then pinkie joints will be chopped off generation after generation despite the pain and risk of infection -- despite the fact that the custom has no adaptive value whatsoever in itself. But Moalem blithely deduces that as therapeutic bloodletting was practiced for thousands of years there must be something beneficial about it (p. 18). No, my good doctor, believe me: pre-scientific medicine was for the most part abject superstition. Left to their own devices, people will do pointless and stupid things.

Moving on from adaptive diseases, Moalem covers the evolutionary interplay between animals and plants (e.g. clover producing a sheep contraceptive to alleviate grazing pressure), parasitical host manipulation (e.g. the idea that common-cold viruses want us to be able to walk around sneezing, while the malaria parasite wants us to become too weak to swat mosquitoes), jumping genes and the possibility of genetically programmed mutation (i.e. Lamarckian evolution), DNA methylation and epigenetics (not all the building plans of an organism are in DNA, i.e. Lamarckian evolution again), and finally, death and aging as adaptive traits, humankind evolving as an aquatic ape, and the benefits of swimming-pool child delivery.

I don't know to whom I might recommend this book. Certainly not to a layperson like myself: too much uncertainty and speculation. And a professional scholar in the field is likely to know most of what Moalem says already. The ideal reader would be someone between these groups as to their level of expertise: perhaps a student of medicine or biology, as optional sweetener for a reading-list otherwise dominated by the stolid views of non-mavericks.

[More blog entries about Technorati Tags: , , , ; , , , .]


More like this

To summarise: so far we have three general kinds of explanations of religion. There are sociological explanations in terms of the economic, societal and political conditions under which religions develop. There are psychological explanations in terms of experiences, existential dread, need for…
Below I mentioned the doyen of living population geneticists, James Crow, yeah, Jim Crow. Collaborater with Motoo Kimura of Neutral Theory fame, Crow is still an active member of the biological community. Recently he reviewed Genes in Conflict : The Biology of Selfish Genetic Elements by Bob…
Osborn's view of the progressive evolution of the brontotheres. Notice the references to the "geneplasm" at left, which Osborn interpreted as spontaneously giving rise to new adaptations. From Osborn (1935).Every now and then I like taking a break from the latest technical papers and conference…
Larry Moran has had a couple of articles up lately on Dr Sharon Moalem, a fellow who has a book out called Survival of the Sickest, and who also has a blog. Larry noticed a couple of things: he's writing utter tripe about junk DNA, he's editing and deleting comments about his science from his blog…

I don't know to whom I might recommend this book.

The Egnorant One of SUNY Stony Brook, perhaps...?

You hit on the big problem with Moalem's premise when you said "I suspect that they [hereditary diseases] might spread because they are not maladaptive enough and happen to be linked fortuitously in development to some adaptive trait." The persistence of hereditary diseases in populations was a topic we covered in my undergraduate genetics course in 1976, and Moalem doesn't bring anything novel to the discussion.

I heard Moalem on TV, and he seemed a self-promoting hack. I think the book is trolling for flames from creationists. Just what we need: bad science debating bad theology.

if you are interested in a really good book on this topic try " why we get sick" by Nesse and Williams. you will then understand how powerful this new application of evolutionary theory can be in our understanding of health. if you can wait, i am writing a book from a practicing physicians point of view on the same subject, tentatively called " point of origin".

By william meller (not verified) on 20 Mar 2007 #permalink

Please, i need help

Under "Good books read lately" you name one by Neil Gaiman (2002), with the illustrator Dave McKean:

For two years now i've been trying to find the picture "Moonspell" or "Moon spell" by McKean (a Sandman cover, i think) on Google images. It suddenly disppeared from google about two years ago, before i got the chance to print the picture on a nice laser printer. ;)

Do you by any chance have info about this art photo?

Those colours......


Thanks for the tip.
Apparently this is not an official web site, and the guy doesn't even know McKean's email address.

I think i just have to face reality, and buy the whole series once i can afford it.


Wow--thanks for the new vocabulary word. I can think of many situations where flåshurtig would express what I mean precisely and succinctly.

As a geneticist/molecular biologist, I can tell you that his main thesis is horsesh*t. He's not a maverick, he's an idiot (though to be fair, there are a couple examples where he's right, but mostly he's wrong).

There's an easy way to test his hypothesis. If a disease gene does in fact have an advantage to it, you'll see an increase in the frequency of heterozygotes in a population to above expected levels given the disadvantage that it confers. Let me give a common example: we see that the frequency of heterozygotes for the sickle-cell anemia allele (that is, people who have one regular copy and one messed up copy) are at a much higher frequency than expected in Africa given the dire situation that homozygotes find themselves in (people with both copies messed up get sickle-cell anemia). There's some evidence that heterozygotes have increased immunity to malaria. So there is a selective advantage to carrying that allele, but a pretty major selective disadvantage. This may seem to suggest his overall thesis is correct, but closer examination shows that this is an exception, not the rule.

This imbalance is seen with *very* few disease alleles. Most of them behave exactly how you would expect, indicating that there is no selective advantage to carrying them. The main reason you see changed frequencies is due to founder effects. If ten people colonize an island, and half of them are carrying a disease allele, it will take a long time to purge that allele from the population (particularly if it doesn't confer particularly strong reproductive disadvantages to the carriers).

Evolution doesn't require that all genes and alleles be advantageous. It requires that they *not* be disadvantageous. And all in all, examples like type II diabetes doesn't dramatically affect your ability to have offspring (after all, most people are done having kids long before type II diabetes shows up).