The Greatest Host

Hi all, I re­cently wrote a blog post sum­ma­riz­ing some ar­ti­cles I found re­ally fas­ci­nat­ing by Stephen Hedrick that lev­er­age evolu­tion and viral ecol­ogy to ar­gue:

  1. While our im­mune sys­tem is in­deed so­phis­ti­cated, it doesn’t keep us any more pro­tected from par­a­sites than the more sim­ple im­mune sys­tem of an ant, oak tree or lob­ster.

  2. Be­cause of the agri­cul­tural rev­olu­tion, do­mes­ti­ca­tion of an­i­mals, and ur­ban­i­sa­tion, hu­mans are in fact likely to be the most dis­eased or­ganisms on the planet with en­tire classes of par­a­sites that few other or­ganisms face.

I thought this may be of in­ter­est to the LessWrong and if this is true, would love to get thoughts, cri­tiques, and en­gage in dis­cus­sion.

The origi­nal post is here: https://​​trent­​​Great­estHost/​​ and I have cross posted it be­low.

I re­cently finished the ex­cel­lent in­tro­duc­tion to im­munol­ogy: How the Im­mune Sys­tem Works, by Lau­ren Som­payrac. A ma­jor part of the hu­man im­mune sys­tem (and that of all ver­te­brates) is the adap­tive sys­tem, which un­like the in­nate sys­tem, is able to mount a defense against al­most any par­a­site[1]. While “How the Im­mune Sys­tem Works” does talk about some of the dis­ad­van­tages of hav­ing such a pow­er­ful sys­tem, the big one be­ing au­toim­mune di­s­or­ders, it paints an awe in­spiring pic­ture of a highly so­phis­ti­cated and ver­sa­tile sys­tem cru­cial to keep­ing us al­ive.

How­ever, a se­ries of ar­ti­cles[2] by Stephen Hedrick lev­er­age evolu­tion and viral ecol­ogy to make a fas­ci­nat­ing ar­gu­ment that, while our im­mune sys­tem is in­deed so­phis­ti­cated, it doesn’t keep us any more pro­tected from par­a­sites than the more sim­ple im­mune sys­tem of an ant, oak tree or lob­ster. More­over, be­cause of the agri­cul­tural rev­olu­tion, do­mes­ti­ca­tion of an­i­mals, and ur­ban­i­sa­tion, hu­mans are in fact likely to be the most dis­eased or­ganisms on the planet with en­tire classes of par­a­sites that few other or­ganisms face.

Th­ese ar­gu­ments, which I will be sum­ma­riz­ing, pre­sent im­por­tant con­sid­er­a­tions for im­munol­ogy, epi­demiolog­i­cal mod­el­ling, zoonotic spillovers, and viral ecol­ogy. More broadly, these pieces have re­minded me of the im­por­tance of first prin­ci­ples think­ing and how cru­cial it is to ac­count for evolu­tion when think­ing about sys­tems, be­fore get­ting lost in the minu­tiae. Even more broadly, these pieces are a re­minder that we as hu­mans re­main stuck on the evolu­tion­ary tread­mill. It is cru­cial that we de­liber­ately take ac­tions to hop off be­fore it is too late. If our cur­rent rate of tech­nolog­i­cal progress con­tinues and we lack cau­tion about what we de­velop, hu­man­ity looks en route to kill it­self and it is some­what of a mir­a­cle that we haven’t already.

The Red Queen

Dr. Hedrick in “The Ac­quired Im­mune Sys­tem: A Van­tage from Be­neath” ar­gues that be­cause of the sheer num­ber and di­ver­sity of par­a­sites, and the fact that no im­mune sys­tem is in­vin­cible to ev­ery threat[3], it is the par­a­sites, not the im­mune sys­tem that are in the driver’s seat. How­ever, while the power is with the par­a­sites, they lack the free­dom to de­cide how to wield it. This is be­cause of their fun­da­men­tal de­pen­dence on the host, mean­ing they can only be so viru­lent or else cause their own ex­tinc­tion. As a re­sult, the par­a­sitic bur­den a pop­u­la­tion faces is cal­ibrated to, and in­de­pen­dent of, the so­phis­ti­ca­tion of its im­mune sys­tem (on evolu­tion­ary timescales). The bur­den an in­di­vi­d­ual host faces is only de­pen­dent on how strong their im­mune sys­tem is in re­la­tion to the rest of its pop­u­la­tion: “A ze­bra doesn’t have to out­run the lion, just the slow­est mem­ber of the herd.” This is why im­muno­com­pro­mised peo­ple can suc­cumb to par­a­sites that those with a nor­mal im­mune sys­tem com­bat or tol­er­ate effortlessly. David Vet­ter, the Bub­ble Boy, died trag­i­cally be­cause he lacked an adap­tive im­mune sys­tem, but more fun­da­men­tally be­cause hu­man co-evolved par­a­sites ex­pected that he would have one.

It is an an­cient, an­ces­tral species of “jawed fish” that we have to blame for first evolv­ing a pri­mor­dial adap­tive im­mune sys­tem of the sort we now have, ap­prox­i­mately 500 mil­lion years ago (Fla­jnik and Kasa­hara, 2011). This novel in­ven­tion al­lowed it to en­joy a golden pe­riod likely free of par­a­sitic in­fec­tion, with the abil­ity to out­com­pete friends and foes, and pro­lifer­ate ex­ten­sively (such that all ver­te­brates “from sharks to aard­varks” have the same adap­tive im­mu­nity). How­ever, this pe­riod was short lived as the par­a­sites quickly evolved to catch up and the co-evolu­tion­ary race be­tween host and par­a­site was back on the tread­mill again. As Dr. Hedrick elo­quently states: “By se­lect­ing for ever-more-de­vi­ous par­a­sites, the im­mune sys­tem is the cause of its own ne­ces­sity”.

Un­for­tu­nately, this race is also not with­out its con­se­quences as the adap­tive im­mune sys­tem has im­mense en­ergy re­quire­ments and causes auto-im­mune di­s­or­ders and im­munopathol­ogy. For ex­am­ple, “more than 3% of peo­ple in the United States ex­pe­rience a form of au­toim­mune dis­ease” that can be de­bil­i­tat­ing or life threat­en­ing (Cooper and Stroehla, 2003).

It may be helpful to know that this form of evolu­tion­ary lock-in is of­ten referred to as the “Red Queen Hy­poth­e­sis”, a refer­ence to Lewis Car­roll’s Red Queen in Alice in Won­der­land: “Now, here, you see, it takes all the run­ning you can do, to keep in the same place. If you want to get some­where else, you must run at least twice as fast as that.” (Van Valen, 1973).

In­ver­te­brate Immunity

Dr. Hedrick sup­ports his ar­gu­ment with in­ter­est­ing ex­am­ples of in­ver­te­brates that don’t suffer a level of par­a­sitic in­fec­tion that would be ex­pected if the so­phis­ti­ca­tion of their im­mune sys­tem mat­tered. This is shown by ex­plor­ing in­ver­te­brate lifes­pan, causes of death and in­ver­te­brate par­a­sites.

Re­gard­ing lifes­pan: “red sea urch­ins and ocean qua­hogs can live to be more than 200 years of age.” (I thought ocean qua­hogs would be some­thing highly ex­otic… sadly it is just an­other name for a clam!). “Lob­sters can live to be at least 30 years of age.” Out­side of in­ver­te­brates, “the gi­ant se­quoia can live 2000 years and the bristle­cone pine can live past 4000 years.” Dr. Hedrick notes that these ex­am­ples of longevity in the wild are un­likely to be rare cases but rather the av­er­age longevity of the species. He lev­er­ages the power of evolu­tion again to briefly ex­plain why and cites in­ter­est­ing pa­pers on the topic[4].

Perform­ing di­rect com­par­i­sons be­tween ver­te­brate and in­ver­te­brate lifes­pans he notes:

A species of sub­so­cial dung bee­tles (Pas­sal­i­dae) has an av­er­age lifes­pan of greater than 2 years in the wild (hardly a clean en­vi­ron­ment!), and ap­prox­i­mately 5 years in cap­tivity (Cam­be­fort and Han­ski, 1991). This is not so differ­ent from our fa­vorite species for study­ing ac­quired im­mu­nity, the house mouse, mus mus­cu­lus, which has an av­er­age lifes­pan in the wild of ap­prox­i­mately 1 year and a lifes­pan in cap­tivity of 2–5 years. A sec­ond ex­am­ple is the lob­ster (Ho­marus amer­i­canus), which has been stud­ied ex­ten­sively due to its com­mer­cial im­por­tance. Lob­sters reach sex­ual ma­tu­rity at 5-8 years. In­clud­ing pre­da­tion, dis­ease, and storm dam­age, the nat­u­ral mor­tal­ity rate of ju­ve­niles and adults (ex­clud­ing hu­man har­vest­ing) is very low with es­ti­mates rang­ing from 2%–8% per year (Thomas, 1973; En­nis et al., 1986; Fog­a­rty, 1995).

Ex­am­ples of par­a­sites that change their viru­lence in re­sponse to their host are also abun­dant. For ex­am­ple, malaria (Nash, 2002) and try­panosomes(Turner and Barry, 1989).

As a fi­nal ex­am­ple of par­a­sitic co-evolu­tion, a study on “in­ver­te­brate iride­s­cent viruses” showed that in­fec­tions were non­lethal and that “the fre­quency of in­fec­tion was di­rectly cor­re­lated with species pro­por­tion, a hal­l­mark of fre­quency-de­pen­dent co­evolu­tion.” (Her­nan­dez et al., 2000).

In light of this ev­i­dence, Dr. Hedrick rightly raises the ques­tion:

“Per­haps the ques­tion is not why in­ver­te­brates man­age to suc­ceed in the ab­sence of an ac­quired im­mune sys­tem, but rather, why do we ver­te­brates have pathogens that ne­ces­si­tate ac­quired im­mu­nity?”

I be­lieve that the un­der­ly­ing evolu­tion­ary the­ory be­hind our adap­tive im­mu­nity not pro­tect­ing us any bet­ter from co-evolved par­a­sites is com­pel­ling and the above ev­i­dence is helpful. How­ever, since the writ­ing of “The Ac­quired Im­mune Sys­tem: A Van­tage from Be­neath” in 2004, I am cu­ri­ous what ad­di­tional ev­i­dence has been dis­cov­ered on both par­a­sitic in­fec­tion and ex­cess mor­tal­ity due to par­a­sites com­pared across or­ganisms with differ­ent im­mune sys­tems. Progress in metage­nomic se­quenc­ing should be helpful with this.

Par­a­sitic Optimization

What are the con­straints and con­sid­er­a­tions of the par­a­sites in their care­ful co-evolu­tion­ary bal­ance with their hosts? The par­a­sites are max­i­miz­ing for re­pro­duc­tion. This is of­ten called R0 (pro­nounced “R naught”), the num­ber of new in­fec­tions caused by an in­fected host. The fac­tors in­fluenc­ing re­pro­duc­tion are trans­mis­sion rate and viru­lence:

Trans­mis­sion rate is “sim­ply the rate at which a par­a­site is suc­cess­fully spread from host to host, and trans­mis­sion can oc­cur over the length of in­fec­tion that is de­ter­mined by a com­bi­na­tion of the host life span, the death rate due to in­fec­tion, and the rate of par­a­site clear­ance.” This is where is the num­ber of in­fec­tions and de­pends not only on the afore­men­tioned fea­tures like host life span but also how mo­bile the host re­mains dur­ing in­fec­tion and how of­ten an in­ter­ac­tion with a naive host leads to trans­mis­sion (how con­ta­gious).

Viru­lence is “the cost of in­fec­tion to the host … it is as­sumed to be as­so­ci­ated with the ra­pidity and ex­tent of in-host par­a­site repli­ca­tion.” I like to think of this in ab­stract terms as the raw amount of en­ergy the par­a­site is tak­ing from its host in or­der to repli­cate.

With a few ex­cep­tions, trans­mis­sion rate de­pends upon viru­lence. Without suffi­ciently high viru­lence, there can­not be trans­mis­sion. How­ever, if viru­lence is too high then it is likely the host will no longer have in­ter­ac­tions with naive hosts (be­cause of in­ca­pac­i­ta­tion or death) and the im­mune sys­tem will try to clear the par­a­site, both low­er­ing trans­mis­sion rate and, by proxy, par­a­sitic fit­ness.

In­ter­est­ing ex­cep­tions to this in­ter­de­pen­dence be­tween trans­mis­sion rate and viru­lence oc­cur when the par­a­site trans­mits via third party vec­tors. This in­cludes malaria (mosquitoes), cholera (wa­ter sup­ply) and an­thrax (in­cred­ibly hardy spores). For malaria, host in­ca­pac­i­ta­tion due to high viru­lence is in fact likely to be ad­van­ta­geous be­cause it makes them an eas­ier tar­get for mosquito bites.

This bal­ance be­tween trans­mis­si­bil­ity and viru­lence is referred to in the liter­a­ture as trade-off the­ory (An­der­son and May, 1982). Its pre­dic­tions have been repli­cated in lab­o­ra­tory and nat­u­ral ex­per­i­ments. In the lab, an in­cred­ible ex­per­i­ment started in the 1920s and last­ing over 15 years in­volved be­tween 100,000 and 200,000 mice, which were used to study the evolu­tion of viral and bac­te­rial viru­lence and mor­tal­ity rates as the rate at which naive hosts were in­tro­duced was al­tered (Green­wood et al., 1936; An­der­son and May, 1979). This evolu­tion of viru­lence means that when a par­a­site en­coun­ters many naive hosts, there is a re­duced fit­ness cost to high viru­lence and the most viru­lent, with the high­est trans­mis­sion rates, spreads the fastest. Mean­while, with fewer naive hosts available, the op­po­site is true (Lively, 2006). In a similar fash­ion, bac­te­ria that have ac­quired an­tibiotic re­sis­tance genes will lose them over time if they are not ex­posed to the an­tibiotics. This is be­cause these genes are ex­tra ge­netic bag­gage that re­quire en­ergy to repli­cate and var­i­ants free of them will out­com­pete those that re­main en­cum­bered (Bin­gle and Thomas, 2001). Ap­ply­ing this phe­nomenon to the on­go­ing SARS-CoV-2 pan­demic, be­fore global lock­down mea­sures were in­tro­duced there was se­lec­tion for more viru­lent strains of the virus. Now, with lock­down in place there is se­lec­tion for less viru­lent, longer lived strains.

The Most Gen­er­ous of Hosts

After Dr. Hedrick un­der­mined the ad­van­tages of our adap­tive im­mune sys­tem, he goes fur­ther in “Un­der­stand­ing Im­mu­nity through the Lens of Disease Ecol­ogy” by ar­gu­ing that since the do­mes­ti­ca­tion of an­i­mals and the agri­cul­tural rev­olu­tion, hu­mans have likely been the best par­a­sitic hosts on the planet.

Pre­vi­ously, hunter-gath­erer tribes were too small to sus­tain acute, in­fec­tious agents that have high viru­lence and trans­mis­sion rates but that the host even­tu­ally re­solves through ster­il­iz­ing im­mu­nity. Either the par­a­site would run out of naive hosts to in­fect, or kill the whole tribe, in ei­ther case go­ing ex­tinct. As a re­sult, these tribes could only sup­port par­a­sites with low viru­lence that per­sist over a life­time and could pass across gen­er­a­tions. Th­ese in­clude the Ep­stein-Barr and Hepatitis B viruses, sup­ported by stud­ies on in­dige­nous tribes in the Ama­zon who dis­played in­fec­tion by per­sis­tent par­a­sites, but not acute and tran­sient ones that come to mind for most peo­ple (Black, 1975).

Ur­ban­iza­tion, sup­ported by agri­cul­ture and the do­mes­ti­ca­tion of an­i­mals, led to pop­u­la­tions large and dense enough to cre­ate the class of acute, re­solv­ing in­fec­tious agents we are all too fa­mil­iar with[5]. You may already be fa­mil­iar with the rule that ev­ery­one is at most 6 peo­ple away from be­ing con­nected to each other (Mil­gram, 1967). Th­ese in­fec­tious agents such as flu, measles, and smal­l­pox, have kil­led on the or­der of billions over the cen­turies. In the 20th cen­tury alone, smal­l­pox is es­ti­mated to have kil­led 300 mil­lion, over dou­ble that of both World Wars com­bined (Fen­ner, 1993). Th­ese par­a­sites most of­ten spread through res­pi­ra­tory droplets and are un­der no se­lec­tive pres­sure to re­duce their viru­lence be­cause of the con­stant sup­ply of naive hosts in the form of im­mi­grants, ba­bies, and im­mune evad­ing mu­ta­tions. “As ev­i­dence of the suc­cess of this pathogen strat­egy, there are more than 200 differ­ent viruses from at least 6 differ­ent virus fam­i­lies (ade­n­ovirus, coro­n­aviruses, in­fluenza virus, parain­fluenza virus, res­pi­ra­tory syn­cy­tial virus, and rhinovirus) that cause “cold” symp­toms: sneez­ing, cough­ing, and runny nose.” (Hedrick, 2018) More­over, measles has been sug­gested to only have evolved within the last 6,000 years (Black, 1966) and “fades-out” in com­mu­ni­ties of less than 200,000 to 500,000 peo­ple (Bartlett, 1960).

The at­ten­tive reader should note the dis­crep­ancy be­tween this ar­gu­ment that hu­mans are the most in­fected and the ear­lier ar­gu­ment that the hu­man im­mune sys­tem had no in­fluence on par­a­site in­fec­tion rates. Hedrick does not di­rectly ad­dress this con­flict but I have ideas for the an­swer. One is that we are still at dis­e­quil­ibrium in co-evolv­ing with our new par­a­sites since the agri­cul­tural rev­olu­tion hap­pened only around 10,000 BC the blink of an eye in evolu­tion­ary time. Another an­swer is that we have cre­ated en­tirely new niches for short tran­sient in­fec­tions which have lit­tle to no com­pe­ti­tion with our pre­vi­ous (and con­tinued) per­sis­tent in­fec­tions. This does raise an in­ter­est­ing ques­tion about what the up­per bound on the num­ber of in­fec­tions a host can main­tain is. It would be in­ter­est­ing to take an en­ergy-based im­munol­ogy ap­proach to this in a “Biol­ogy by the Num­bers” fash­ion.

An in­ter­est­ing con­se­quence of these tran­sient, acute in­fec­tions is that peo­ple think per­sis­tent in­fec­tions that can last a life­time are un­usual when from an evolu­tion­ary per­spec­tive hu­man­ity is much bet­ter ac­quainted with them. For ex­am­ple, most peo­ple don’t know that Ep­stein-Barr virus in­fects ~90% of peo­ple. Per­son­ally, I still re­mem­ber the hor­ror of learn­ing that an HIV in­fec­tion can’t be cured, which Dr. Hedrick ar­gues has been the biggest mo­ti­va­tor for more re­cent re­search into par­a­sitic per­sisters.

A para­graph that re­ally caught my at­ten­tion on just how sus­cep­ti­ble mod­ern so­ciety is to acute, re­solv­ing in­fec­tious agents be­cause of the way we are in­ter­con­nected is (em­pha­sis mine):

A sec­ond at­tribute of mod­ern so­ciety is that the num­ber of in­ter­ac­tions that char­ac­ter­ize each in­di­vi­d­ual (the de­gree dis­tri­bu­tion) does not fol­low a nor­mal dis­tri­bu­tion. Rather, the num­ber of ‘friends’ or ‘con­nec­tions’ pos­sessed by each per­son is ex­tremely het­ero­ge­neous and more closely fol­lows a (trun­cated) power law (an ex­am­ple of a ‘heavy-tailed’ dis­tri­bu­tion). That is, most peo­ple have small num­ber of con­nec­tions, whereas some di­rectly in­ter­act with many peo­ple [72,77]. A sim­plified way of look­ing at this is that for dis­ease spread there would ex­ist highly con­nected in­di­vi­d­u­als who would be sure to prop­a­gate an epi­demic [78–81]. Math­e­mat­i­cal mod­el­ing in­di­cates that no mat­ter how in­effi­cient the trans­mis­sion of a dis­ease, in a net­work with a heavy-tailed dis­tri­bu­tion, an epi­demic is likely to per­ma­nently take hold (Hedrick, 2017).

In ad­di­tion to these new in­fec­tious agents, close con­tinual con­tact with wildlife has led to in­creased rates of zoonotic spillovers. Th­ese are the X-Fac­tor in that “rarely do differ­ent species ex­pe­rience the same in­fec­tious agent with ex­actly the same pathol­ogy”. The par­a­site is cal­ibrated to a differ­ent host and as a re­sult the new in­fec­tion can be in­cred­ibly deadly or in­con­se­quen­tial. Zoonoses can be deadly not only from a par­a­site that is too viru­lent and kills quickly, but also from im­munopathol­ogy where, like in an aller­gic re­ac­tion, the im­mune sys­tem be­comes over stim­u­lated. This can cause death, for ex­am­ple, via cy­tokine storms or en­cephal­itis. An in­ter­est­ing ex­am­ple of this var­i­ance in the effects of differ­ent zoonoses even within the same viral strain is:

Of the 31 species of the genus Mam­marena virus, most cause mild pathol­ogy in their nat­u­ral murid hosts and do not cause an ap­par­ent in­fec­tion in hu­man be­ings. How­ever, seven of the 31 species are known to cause hem­or­rhagic fever in hu­man be­ings with mor­tal­ity rates be­tween 15% and 30%. The other Mam­marena viruses ei­ther do not repli­cate in hu­man cells or, like lym­pho­cytic chori­omenin­gitis virus, cause mod­er­ate pathol­ogy and are even­tu­ally cleared (Za­p­ata and Sal­vato, 2013).

There are a few con­se­quences of these ar­gu­ments. Firstly, in or­der to un­der­stand the viru­lence of a par­a­site, know­ing about the im­mune sys­tem is in­suffi­cient. Host pop­u­la­tion den­sity, size, and spa­tial struc­ture in ad­di­tion to viral fea­tures like its ori­gins, trans­mis­sion mechanism, and mu­ta­tion rate are far more im­por­tant for pre­dict­ing its viru­lence and in­fec­tion time. Se­condly, zoonotic spillovers are in dis-equil­ibrium with hu­man im­mu­nity and fore­cast­ing their effects will con­tinue to be the most difficult challenge. Fi­nally, we need to have a much deeper un­der­stand­ing of, and re­spect for, per­sis­tent in­fec­tions. I think this re­spect has grown over time with our con­tin­u­ing failed at­tempts to vac­ci­nate against long co-evolved par­a­sites such as malaria, tu­ber­cu­lo­sis and HIV (co-evolved in our pri­mate cous­ins).

If an in­crease in tech­nol­ogy is to blame for all of our in­fec­tions, it must also be our solu­tion. With vac­cines and other treat­ments, our tech­nol­ogy is now de facto part of the hu­man im­mune sys­tem and we need to con­tinue lev­er­ag­ing it to catch up in our par­a­sitic dis­e­quil­ibrium and stay ahead. How­ever, I be­lieve it is helpful to re­mem­ber that our tech­nol­ogy is both the cause and solu­tion to this prob­lem. Yes, san­i­ta­tion has been won­der­ful for re­duc­ing in­fec­tion rates dra­mat­i­cally, but it was only nec­es­sary in the first place be­cause of ur­ban­iza­tion and the in­fec­tious agents that it cre­ated.


I be­lieve there is much to learn from an evolu­tion­ary and par­a­site ori­ented ap­proach to im­munol­ogy. Our adap­tive im­mune sys­tem is no op­ti­mal solu­tion, we should not ex­pect to ever be truly free of par­a­sites, some par­a­sites will be far more difficult to erad­i­cate than smal­l­pox, and tech­nolog­i­cal ad­vance­ments have put us in a very pre­car­i­ous po­si­tion with en­tire new classes of in­fec­tion both known, and yet to spillover.

More re­search should be done on per­sis­tent in­fec­tions and im­mune tol­er­ance. We must also ac­knowl­edge that a perfect mechanis­tic un­der­stand­ing of the hu­man im­mune sys­tem is in­suffi­cient to pre­dict and com­bat par­a­sites. A great deal of pre­dic­tive power seems to come from par­a­site ecol­ogy and epi­demiol­ogy. More of a pri­or­ity should also be made to­wards the study of in­fec­tious agents that have co-evolved with their hosts. Dr. Hedrick notes that we spend a great deal of time study­ing hu­man pathogens in mice, but why not study mice pathogens in mice? Look­ing at host pathogen co-evolu­tion may even help an­swer ques­tions about why our biol­ogy is of­ten so Byzan­tine.

On a broader level, just like the jawed fish that de­vel­oped the pri­mor­dial im­mune sys­tem, the virus that gains greater viru­lence, and our agri­cul­ture-in­vent­ing an­ces­tors, hu­man­ity re­mains on the evolu­tion­ary tread­mill. The tread­mill acts not only on our biol­ogy but also through­out so­ciety wher­ever there is com­pe­ti­tion, fore­most in cap­i­tal­ism which con­tinues to in­cen­tivize tech­nolog­i­cal de­vel­op­ments. At a fun­da­men­tal level, tech­nol­ogy is power and a dou­ble edged sword. Tyler Cowen has said that he does not think a large num­ber of hu­mans will ever leave Earth to travel the galaxy. This is be­cause the amount of tech­nol­ogy and raw power that would be re­quired is so much that any in­di­vi­d­ual would be able to ac­quire suffi­cient power to de­stroy Earth and ev­ery­one on it upon a whim. The mor­bid and amus­ing thought ex­per­i­ment he pre­sents is, and I para­phrase: “if ev­ery­one on earth had their own big red but­ton and if press­ing it would kill ev­ery­one al­ive, how much time would pass be­fore some­one pressed their but­ton?” I will leave you to an­swer this your­self be­cause I don’t want to write my an­swer...

As a so­ciety, we must ac­knowl­edge that we are on the evolu­tion­ary tread­mill and face many co­or­di­na­tion prob­lems in or­der to all get off it at the same time. The on­go­ing SARS-CoV-2 pan­demic should make it clear just how ten­u­ous hu­man ex­is­tence is. We need to think hard about and sup­port ways to both di­rectly re­duce ex­is­ten­tial risks like through nu­clear disar­ma­ment and bet­ter biose­cu­rity, and in­di­rectly, through boost­ing our co­op­er­a­tion, in­tel­li­gence, and de­ci­sion mak­ing.

If my sum­mary of Dr. Hedrick’s ideas has res­onated, I en­courage you to read his work which is far more elo­quent and pro­vides many in­ter­est­ing cita­tions. If the gen­eral en­slave­ment to evolu­tion I have spo­ken about in­ter­ests you then Med­i­ta­tions on Moloch is one of my favourite pieces from my favourite blog. If you are into sci-fi, The Mote in God’s Eye is the most ac­cu­rate de­pic­tion of evolu­tion and aliens in any book I’ve read.

Open Ques­tions:

  • Do ver­te­brates have pathogens that ne­ces­si­tate ac­quired im­mu­nity?

  • What ad­di­tional ev­i­dence since Dr. Hedrick’s pieces has been dis­cov­ered on both par­a­sitic in­fec­tion and ex­cess mor­tal­ity due to par­a­sites com­pared across or­ganisms with differ­ent im­mune sys­tems?

  • Where do non-par­a­sitic, com­men­sal viruses and bac­te­ria fit into trade-off the­ory? How do they trans­mit well enough with­out be­ing viru­lent?

  • How are per­sis­tent par­a­sites able to evade and in­duce tol­er­ance in the im­mune sys­tem? How do they main­tain enough viru­lence to re­main a high enough trans­mis­sion rate?

  • What are vi­o­la­tions to trade-off the­ory and how has it de­vel­oped since its in­tro­duc­tion in 1982?

  • What is the up­per bound on the num­ber of in­fec­tions a host can main­tain? It would be in­ter­est­ing to take an en­ergy-based im­munol­ogy ap­proach to this in a “Biol­ogy by the Num­bers” fash­ion.

  • What are the cel­lu­lar/​molec­u­lar biol­ogy cor­re­lates of viru­lence and trans­mis­si­bil­ity? For ex­am­ple, if a par­a­site repli­cates in the res­pi­ra­tory tract, which seems like it would sig­nifi­cantly in­crease its trans­mis­sion rate, does this have a higher viru­lence than in­fect­ing a toe? If so, by how much?

  • Re­lat­ing the above ques­tion to SARS-CoV-2, We know from bio­chem­istry that the re­cep­tor bind­ing do­main of Spike pro­tein (nec­es­sary for viral en­try) binds ~10 to 20x tighter to hu­man ACE2 (Wrapp et al., 2020). How does this af­fect viru­lence vs trans­mis­si­bil­ity? Why does this still make SARS ~10x more lethal and MERS ~30x more lethal (Ruan, 2020)? Espe­cially as these coro­n­aviruses are all from the same fam­ily with the same trade­off the­ory? Surely they also aren’t too far away on the fit­ness land­scape, clearly SARS-CoV-2 is much closer to the lo­cal max­ima for R0 but why?

  • Also can bats sus­tain the “acute, re­solv­ing in­fec­tious agents” that we have made our­selves so sus­cep­ti­ble to be­cause of their high and dense pop­u­la­tions that are mostly sta­tion­ary in­side caves? If so, then what other species also have mod­ern hu­man like pop­u­la­tion dy­nam­ics? Aside from be­ing fel­low ver­te­brates and mam­mals, what is it about bats that make them such zoonotic threats?

Thanks to An­drew Liu, Alexan­der Bricken, Joe Choo-Choy and Michael McLaren, for the helpful ed­its, com­ments and sug­ges­tions.


  1. To be pre­cise, par­a­site here refers to: “in­fec­tious bac­te­ria, fungi, par­a­sitic in­ver­te­brates, and viruses”. ↩︎

  2. Thanks to a re­cent Gw­ern newslet­ter for bring­ing the first of these pieces, The Ac­quired Im­mune Sys­tem: A Van­tage from Be­neath to my at­ten­tion. ↩︎

  3. Liv­ing or­ganisms are such en­ergy rich tar­gets for par­a­sites that even if ev­ery par­a­site al­ive to­day went ex­tinct, novel ones would spring into ex­is­tence just as they have in the past. As far as mu­ta­tion rates go, “bac­te­ria can un­dergo 100,000 gen­er­a­tions for each one of ours” (source) with even higher gen­er­a­tion and mu­ta­tion rates for many viruses like HIV. ↩︎

  4. Why do or­ganisms age at differ­ent rates and live to a cer­tain length? It seems like a ma­jor in­fluence is the age at which a ma­jor­ity of a species dies from ex­ter­nal fac­tors (Kirk­wood and Aus­tad, 2000). Beyond this age, be­cause the ma­jor­ity of the pop­u­la­tion has died and re­pro­duced (which is of course also de­pen­dent on death), there is in­suffi­cient se­lec­tion pres­sure for longevity. This man­i­fests it­self in a few pos­si­ble ways: (i) dele­te­ri­ous mu­ta­tions that cause death af­ter a cer­tain age are al­lowed to ac­cu­mu­late be­cause they aren’t se­lected for (called “se­lec­tion shadow”); (ii) genes that sup­port en­ergy in­vest­ment into cel­lu­lar re­pair and main­te­nance aren’t se­lected for (“dis­pos­able soma” the­ory); (iii) genes that benefit or­ganisms early in life are favoured, even if they have nega­tive effects at later ages (“pleiotropy the­ory”). Kirk­wood and Aus­tad, 2000 re­view a num­ber of fas­ci­nat­ing ex­per­i­ments where the ex­ter­nal death rate in fruit flies (Drosophila melanogaster) and ne­ma­todes ( Caenorhab­di­tis el­e­gans) is al­tered and causes sub­se­quent al­ter­a­tions in longevity. An im­por­tant con­se­quence of this the­ory is that “Spe­cific genes se­lected to pro­mote age­ing are un­likely to ex­ist.” ↩︎

  5. Dr. Hedrick notes in “Un­der­stand­ing Im­mu­nity through the Lens of Disease Ecol­ogy” that the Amer­i­cas, even with their ur­ban­iz­ing civ­i­liza­tions such as the Mayan em­pire, did not see the same acute but short lived in­fec­tious agents emerge and gives one pos­si­ble rea­son as the lack of large herd an­i­mals suit­able for do­mes­ti­ca­tion. Else­where he ac­knowl­edges the stochas­tic na­ture of evolu­tion in gen­eral. How­ever, this serves as a tragic and ex­cel­lent ex­am­ple of the Red Queen Hy­poth­e­sis where the par­a­sites evolved for more co-evolved Euro­pean im­mune sys­tems could wreak mass havoc: “With ex­po­sure, all the rav­ages of mil­len­nia rained down at once upon dis­ease-naïve in­hab­itants, caus­ing pop­u­la­tion losses be­tween 50% and 90%. Euro­pean con­tact re­sulted in mul­ti­ple vir­gin soil epi­demics hor­rifi­cally played out over a great seg­ment of the world’s pop­u­la­tion.” ↩︎