This is outside my area of expertise, but it seems that the author may not have presented Medawar's argument correctly. As I understand it, Medawar's theory of aging was not so much based in physics as based in evolution. The author presents it here as random DNA transcription errors building up mutations which eventually cause the animal to die.
But I think that Medawar's argument was that all animals are subject to random accidents, predation, disease, etc. This will produce an exponential distribution of lifetimes. As a consequence, beyond a few half-lifes, there will be exponentially few members of the species. So if there is some mutation in the DNA which causes the animal to die beyond that time, there will be so few members of the species around to exhibit it that there will be effectively no selection pressure to keep this mutation out of the population. As a consequence, even if you started out with a population that could in theory live forever, it will very quickly become riddled with mutations which start to kill the animals after a couple of half-lifes. From the outside, it would appear as though the animals were "programmed" to die at a certain time.
George Williams presented a similar argument, but also claimed that some mutations can increase an animal's fitness early in life at the expense of decreased fitness much later in life. If the animal will likely have died of some accident by that time anyway, this trait effectively has no downside and will be selected for.
But I think that Medawar's argument was that all animals are subject to random accidents, predation, disease, etc. This will produce an exponential distribution of lifetimes. As a consequence, beyond a few half-lifes, there will be exponentially few members of the species. So if there is some mutation in the DNA which causes the animal to die beyond that time, there will be so few members of the species around to exhibit it that there will be effectively no selection pressure to keep this mutation out of the population. As a consequence, even if you started out with a population that could in theory live forever, it will very quickly become riddled with mutations which start to kill the animals after a couple of half-lifes. From the outside, it would appear as though the animals were "programmed" to die at a certain time.
George Williams presented a similar argument, but also claimed that some mutations can increase an animal's fitness early in life at the expense of decreased fitness much later in life. If the animal will likely have died of some accident by that time anyway, this trait effectively has no downside and will be selected for.