Evolution Actually Selects for Aging — And That Changes Everything

Why we age is one of biology's most contested questions. Michael Ringel's review in Biological Reviews provides perhaps the most rigorous logical framework to date for categorising and empirically testing competing theories of aging. The stakes are high: the correct answer fundamentally shapes how we design interventions and what outcomes we should expect.

Ringel organises all aging theories into three mutually exclusive, collectively exhaustive categories. Mechanistic theories (e.g., rate-of-living, oxidative damage, hyperfunction) explain the 'how' of aging but implicitly or explicitly claim that evolution fails to act — an extraordinary claim given the overwhelming evidence that natural selection operates broadly across biology. Weakening-force-of-selection theories (e.g., Medawar's mutation accumulation, Williams's antagonistic pleiotropy as classically framed) argue that declining numbers of older individuals reduce selection pressure, allowing late-acting deleterious mutations to accumulate. Optimisation theories, anchored by Kirkwood's disposable soma theory, argue that evolution actively selects for a finite investment in somatic maintenance because finite resources must be optimally allocated across survival, reproduction, generation time, and offspring quality.

To adjudicate between these categories, Ringel tests them against a rich empirical dataset. Caloric restriction extends lifespan across diverse species — consistent with optimisation (resource reallocation) but difficult to explain under mechanistic or weakening-selection frameworks. Long-lived organisms such as bats, naked mole rats, and eusocial insect queens show dramatically extended lifespans tied to specific ecological or social contexts, again fitting optimisation's prediction that lifespan is tuned to environmental conditions. Mortality rate plateaus in late life, where the rate of death paradoxically levels off, challenge mechanistic theories but are naturally accommodated by optimisation models. The malleability of lifespan — that single-gene mutations or environmental changes can substantially extend or shorten lifespan — is consistent with optimisation's prediction that existing biological machinery already encodes adjustable set points.

A critical and differentiating prediction of optimisation theories is that mutations extending lifespan should generally reduce fitness under natural conditions. Ringel reviews the available evidence and concludes that, contrary to some interpretations in the literature, this prediction is supported: longevity-extending mutations consistently show fitness costs in naturalistic settings. This is the key empirical dividing line between optimisation and other theories.

The review draws an illuminating parallel with allometry: mechanistic models failed to explain why metabolic scaling follows a ~3/4 power law, but an evolutionary optimisation model naturally derived the observed exponent and also predicted novel nuances confirmed by data. Aging theory appears to follow the same logic — evolutionary optimisation explains not just the central phenomenon but also its 'wild and wonderful' diversity across species. Implications for medicine are direct: the geroscience hypothesis (that slowing aging reduces age-related disease broadly) is strongly supported, and the existence of pre-evolved biological set points for lifespan means that pharmacological or genetic manipulation of aging rate is biologically tractable.