Evolving to Extinction

It is a very com­mon mis­con­cep­tion that an evolu­tion works for the good of its species. Can you re­mem­ber hear­ing some­one talk about two rab­bits breed­ing eight rab­bits and thereby “con­tribut­ing to the sur­vival of their species”? A mod­ern evolu­tion­ary biol­o­gist would never say such a thing; they’d sooner breed with a rab­bit.

It’s yet an­other case where you’ve got to si­mul­ta­neously con­sider mul­ti­ple ab­stract con­cepts and keep them dis­tinct. Evolu­tion doesn’t op­er­ate on par­tic­u­lar in­di­vi­d­u­als; in­di­vi­d­u­als keep what­ever genes they’re born with. Evolu­tion op­er­ates on a re­pro­duc­ing pop­u­la­tion, a species, over time. There’s a nat­u­ral ten­dency to think that if an Evolu­tion Fairy is op­er­at­ing on the species, she must be op­ti­miz­ing for the species. But what re­ally changes are the gene fre­quen­cies, and fre­quen­cies don’t in­crease or de­crease ac­cord­ing to how much the gene helps the species as a whole. As we shall later see, it’s quite pos­si­ble for a species to evolve to ex­tinc­tion.

Why are boys and girls born in roughly equal num­bers? (Leav­ing aside crazy coun­tries that use ar­tifi­cial gen­der se­lec­tion tech­nolo­gies.) To see why this is sur­pris­ing, con­sider that 1 male can im­preg­nate 2, 10, or 100 fe­males; it wouldn’t seem that you need the same num­ber of males as fe­males to en­sure the sur­vival of the species. This is even more sur­pris­ing in the vast ma­jor­ity of an­i­mal species where the male con­tributes very lit­tle to rais­ing the chil­dren—hu­mans are ex­traor­di­nary, even among pri­mates, for their level of pa­ter­nal in­vest­ment. Balanced gen­der ra­tios are found even in species where the male im­preg­nates the fe­male and van­ishes into the mist.

Con­sider two groups on differ­ent sides of a moun­tain; in group A, each mother gives birth to 2 males and 2 fe­males; in group B, each mother gives birth to 3 fe­males and 1 male. Group A and group B will have the same num­ber of chil­dren, but group B will have 50% more grand­chil­dren and 125% more great-grand­chil­dren. You might think this would be a sig­nifi­cant evolu­tion­ary ad­van­tage.

But con­sider: The rarer males be­come, the more re­pro­duc­tively valuable they be­come—not to the group, but to the in­di­vi­d­ual par­ent. Every child has one male and one fe­male par­ent. Then in ev­ery gen­er­a­tion, the to­tal ge­netic con­tri­bu­tion from all males equals the to­tal ge­netic con­tri­bu­tion from all fe­males. The fewer males, the greater the in­di­vi­d­ual ge­netic con­tri­bu­tion per male. If all the fe­males around you are do­ing what’s good for the group, what’s good for the species, and birthing 1 male per 10 fe­males, you can make a ge­netic kil­ling by birthing all males, each of whom will have (on av­er­age) ten times as many grand­chil­dren as their fe­male cous­ins.

So while group se­lec­tion ought to fa­vor more girls, in­di­vi­d­ual se­lec­tion fa­vors equal in­vest­ment in male and fe­male offspring. Look­ing at the statis­tics of a ma­ter­nity ward, you can see at a glance that the quan­ti­ta­tive bal­ance be­tween group se­lec­tion forces and in­di­vi­d­ual se­lec­tion forces is over­whelm­ingly tilted in fa­vor of in­di­vi­d­ual se­lec­tion in Homo sapi­ens.

(Tech­ni­cally, this isn’t quite a glance. In­di­vi­d­ual se­lec­tion fa­vors equal parental in­vest­ments in male and fe­male offspring. If males cost half as much to birth and/​or raise, twice as many males as fe­males will be born at the evolu­tion­ar­ily sta­ble equil­ibrium. If the same num­ber of males and fe­males were born in the pop­u­la­tion at large, but males were twice as cheap to birth, then you could again make a ge­netic kil­ling by birthing more males. So the ma­ter­nity ward should re­flect the bal­ance of parental op­por­tu­nity costs, in a hunter-gath­erer so­ciety, be­tween rais­ing boys and rais­ing girls; and you’d have to as­sess that some­how. But ya know, it doesn’t seem all that much more re­pro­duc­tive-op­por­tu­nity-costly for a hunter-gath­erer fam­ily to raise a girl, so it’s kinda sus­pi­cious that around the same num­ber of boys are born as girls.)

Nat­u­ral se­lec­tion isn’t about groups, or species, or even in­di­vi­d­u­als. In a sex­ual species, an in­di­vi­d­ual or­ganism doesn’t evolve; it keeps what­ever genes it’s born with. An in­di­vi­d­ual is a once-off col­lec­tion of genes that will never reap­pear; how can you se­lect on that? When you con­sider that nearly all of your an­ces­tors are dead, it’s clear that “sur­vival of the fittest” is a tremen­dous mis­nomer. “Repli­ca­tion of the fit­ter” would be more ac­cu­rate, al­though tech­ni­cally, fit­ness is defined only in terms of repli­ca­tion.

Nat­u­ral se­lec­tion is re­ally about gene fre­quen­cies. To get a com­plex adap­ta­tion, a ma­chine with mul­ti­ple de­pen­dent parts, each new gene as it evolves de­pends on the other genes be­ing re­li­ably pre­sent in its ge­netic en­vi­ron­ment. They must have high fre­quen­cies. The more com­plex the ma­chine, the higher the fre­quen­cies must be. The sig­na­ture of nat­u­ral se­lec­tion oc­cur­ring is a gene ris­ing from 0.00001% of the gene pool to 99% of the gene pool. This is the in­for­ma­tion, in an in­for­ma­tion-the­o­retic sense; and this is what must hap­pen for large com­plex adap­ta­tions to evolve.

The real strug­gle in nat­u­ral se­lec­tion is not the com­pe­ti­tion of or­ganisms for re­sources; this is an ephemeral thing when all the par­ti­ci­pants will van­ish in an­other gen­er­a­tion. The real strug­gle is the com­pe­ti­tion of alle­les for fre­quency in the gene pool. This is the last­ing con­se­quence that cre­ates last­ing in­for­ma­tion. The two rams bel­low­ing and lock­ing horns are only pass­ing shad­ows.

It’s perfectly pos­si­ble for an allele to spread to fix­a­tion by out­com­pet­ing an al­ter­na­tive allele which was “bet­ter for the species”. If the Fly­ing Spaghetti Mon­ster mag­i­cally cre­ated a species whose gen­der mix was perfectly op­ti­mized to en­sure the sur­vival of the species—the op­ti­mal gen­der mix to bounce back re­li­ably from near-ex­tinc­tion events, adapt to new niches, etcetera—then the evolu­tion would rapidly de­grade this species op­ti­mum back into the in­di­vi­d­ual-se­lec­tion op­ti­mum of equal parental in­vest­ment in males and fe­males.

Imag­ine a “Frodo gene” that sac­ri­fices its ve­hi­cle to save its en­tire species from an ex­tinc­tion event. What hap­pens to the allele fre­quency as a re­sult? It goes down. Kthxbye.

If species-level ex­tinc­tion threats oc­cur reg­u­larly (call this a “Buffy en­vi­ron­ment”) then the Frodo gene will sys­tem­at­i­cally de­crease in fre­quency and van­ish, and soon there­after, so will the species. A hy­po­thet­i­cal ex­am­ple? Maybe. If the hu­man species was go­ing to stay biolog­i­cal for an­other cen­tury, it would be a good idea to start clon­ing Gandhi.

In viruses, there’s the ten­sion be­tween in­di­vi­d­ual viruses repli­cat­ing as fast as pos­si­ble, ver­sus the benefit of leav­ing the host al­ive long enough to trans­mit the ill­ness. This is a good real-world ex­am­ple of group se­lec­tion, and if the virus evolves to a point on the fit­ness land­scape where the group se­lec­tion pres­sures fail to over­come in­di­vi­d­ual pres­sures, the virus could van­ish shortly there­after. I don’t know if a dis­ease has ever been caught in the act of evolv­ing to ex­tinc­tion, but it’s prob­a­bly hap­pened any num­ber of times.

Se­gre­ga­tion-dis­torters sub­vert the mechanisms that usu­ally guaran­tee fair­ness of sex­ual re­pro­duc­tion. For ex­am­ple, there is a seg­re­ga­tion-dis­torter on the male sex chro­mo­some of some mice which causes only male chil­dren to be born, all car­ry­ing the seg­re­ga­tion-dis­torter. Then these males im­preg­nate fe­males, who give birth to only male chil­dren, and so on. You might cry “This is cheat­ing!” but that’s a hu­man per­spec­tive; the re­pro­duc­tive fit­ness of this allele is ex­tremely high, since it pro­duces twice as many copies of it­self in the suc­ceed­ing gen­er­a­tion as its non­mu­tant al­ter­na­tive. Even as fe­males be­come rarer and rarer, males car­ry­ing this gene are no less likely to mate than any other male, and so the seg­re­ga­tion-dis­torter re­mains twice as fit as its al­ter­na­tive allele. It’s spec­u­lated that real-world group se­lec­tion may have played a role in keep­ing the fre­quency of this gene as low as it seems to be. In which case, if mice were to evolve the abil­ity to fly and mi­grate for the win­ter, they would prob­a­bly form a sin­gle re­pro­duc­tive pop­u­la­tion, and would evolve to ex­tinc­tion as the seg­re­ga­tion-dis­torter evolved to fix­a­tion.

Around 50% of the to­tal genome of maize con­sists of trans­posons, DNA el­e­ments whose pri­mary func­tion is to copy them­selves into other lo­ca­tions of DNA. A class of trans­posons called “P el­e­ments” seem to have first ap­peared in Drosophila only in the mid­dle of the 20th cen­tury, and spread to ev­ery pop­u­la­tion of the species within 50 years. The “Alu se­quence” in hu­mans, a 300-base trans­po­son, is re­peated be­tween 300,000 and a mil­lion times in the hu­man genome. This may not ex­tin­guish a species, but it doesn’t help it; trans­posons cause more mu­ta­tions which are as always mostly harm­ful, de­crease the effec­tive copy­ing fidelity of DNA. Yet such cheaters are ex­tremely fit.

Sup­pose that in some sex­u­ally re­pro­duc­ing species, a perfect DNA-copy­ing mechanism is in­vented. Since most mu­ta­tions are detri­men­tal, this gene com­plex is an ad­van­tage to its hold­ers. Now you might won­der about benefi­cial mu­ta­tions—they do hap­pen oc­ca­sion­ally, so wouldn’t the un­mutable be at a dis­ad­van­tage? But in a sex­ual species, a benefi­cial mu­ta­tion that be­gan in a muta­ble can spread to the de­scen­dants of un­muta­bles as well. The muta­bles suffer from de­gen­er­ate mu­ta­tions in each gen­er­a­tion; and the un­muta­bles can sex­u­ally ac­quire, and thereby benefit from, any benefi­cial mu­ta­tions that oc­cur in the muta­bles. Thus the muta­bles have a pure dis­ad­van­tage. The perfect DNA-copy­ing mechanism rises in fre­quency to fix­a­tion. Ten thou­sand years later there’s an ice age and the species goes out of busi­ness. It evolved to ex­tinc­tion.

The “by­stan­der effect” is that, when some­one is in trou­ble, soli­tary in­di­vi­d­u­als are more likely to in­ter­vene than groups. A col­lege stu­dent ap­par­ently hav­ing an epilep­tic seizure was helped 85% of the time by a sin­gle by­stan­der, and 31% of the time by five by­stan­ders. I spec­u­late that even if the kin­ship re­la­tion in a hunter-gath­erer tribe was strong enough to cre­ate a se­lec­tion pres­sure for helping in­di­vi­d­u­als not di­rectly re­lated, when sev­eral po­ten­tial helpers were pre­sent, a ge­netic arms race might oc­cur to be the last one to step for­ward. Every­one de­lays, hop­ing that some­one else will do it. Hu­man­ity is fac­ing mul­ti­ple species-level ex­tinc­tion threats right now, and I gotta tell ya, there ain’t a lot of peo­ple step­pin’ for­ward. If we lose this fight be­cause vir­tu­ally no one showed up on the bat­tlefield, then—like a prob­a­bly-large num­ber of species which we don’t see around to­day—we will have evolved to ex­tinc­tion.

Cancer­ous cells do pretty well in the body, pros­per­ing and amass­ing more re­sources, far out­com­pet­ing their more obe­di­ent coun­ter­parts. For a while.

Mul­ti­cel­lu­lar or­ganisms can only ex­ist be­cause they’ve evolved pow­er­ful in­ter­nal mechanisms to out­law evolu­tion. If the cells start evolv­ing, they rapidly evolve to ex­tinc­tion: the or­ganism dies.

So praise not evolu­tion for the so­lic­i­tous con­cern it shows for the in­di­vi­d­ual; nearly all of your an­ces­tors are dead. Praise not evolu­tion for the so­lic­i­tous con­cern it shows for a species; no one has ever found a com­plex adap­ta­tion which can only be in­ter­preted as op­er­at­ing to pre­serve a species, and the math­e­mat­ics would seem to in­di­cate that this is vir­tu­ally im­pos­si­ble. In­deed, it’s perfectly pos­si­ble for a species to evolve to ex­tinc­tion. Hu­man­ity may be finish­ing up the pro­cess right now. You can’t even praise evolu­tion for the so­lic­i­tous con­cern it shows for genes; the bat­tle be­tween two al­ter­na­tive alle­les at the same lo­ca­tion is a zero-sum game for fre­quency.

Fit­ness is not always your friend.