Life span, a characteristic of life history that is the product of evolution, refers to the duration of an organism's entire life course. Application of the concept is straightforward at both the individual and cohort levels. At the individual level, it is the period between birth and death; at the cohort level (including both real and synthetic cohorts), it is the average length of life or life expectancy at birth. Life span applied to a population or a species, however, requires a modifier to avoid ambiguity. Maximum observed life span is the highest verified age at death, possibly limited to a particular population or time period. The overall highest verified age for a species is also called its record life span. The theoretical highest attainable age is known as maximum potential life span, maximum theoretical life span, or species-specific life span. Depending on context, maximum life span can refer to either the observed or the potential maximum.
Maximum observed life spans (i.e., longevity records) are not synonymous with theoretical maximums for at least two reasons. First, maximum longevity is not an appropriate general concept because an animal dies before the age of infinity not because it cannot pass some boundary age but because the probability of its riding out the ever-present risk of death for that long is infinitesimally small. In other words, there is no identifiable age for each species to which some select individuals can survive but none can live beyond. Second, the number of individuals observed heavily influences the record age of a species. That is, the longevity records for species in which the life spans of large numbers of individuals have been observed will be significantly greater than the corresponding figure for a species that has the same longevity but is represented by a few dozen individuals. For the vast majority of longevity records by species, the population at risk, and therefore the denominator, is unknown.
Conceptual Aspects of Life Span
The life span concept is relevant only to species in which an individual exits–to entities circumscribed by distinct birth and death processes. Thus the concept does not apply to bacteria, which reproduce by binary fission, to plant species that reproduce by cloning, or to modular organisms with iterated growth such as coral or honeybee colonies. When a single reproductive event occurs at the end of the life course that results in the death of the individual, then life span is linked deterministically to the species' natural history. This occurs with the seed set of annual plants (e.g., grasses), in drone (male) honeybees as a consequence of the physical damage caused by mating, in many mayfly species when a female's abdomen ruptures to release her eggs after she drops into a lake or stream, and in anadromous (riverspawning) salmon that die shortly after spawning. Life span can be considered indeterminate for species (including humans) that are capable of repeated (iteroparous) reproduction. That life span is indeterminate in many species is consistent with what is known about the lack of cutoff points in biology–all evidence suggests that species do not have an internal clock for terminating life.
Changes that occur in organisms that enter resting states such as dormancy, hibernation, and estivation (a state of resting that occurs in summer) reduce mortality rates and thus increase longevity. This also occurs when individuals are subjected to caloric restriction or when their reproductive efforts are reduced. A species' life course may consist of many phases such as infant, juvenile, and pre-and postreproductive periods; therefore, a change in overall life span will correspond to a commensurate change in the duration of one or more of these stages. When environmental conditions are greatly improved, such as for animals kept in zoos or laboratories or under the conditions experienced by contemporary humans, mortality rates usually decrease and thus longevity increases. Whereas earlier stages, such as the prereproductive period, are evolved life history traits, the added segment(s) arising at the end of the life course are byproducts of selection for robustness or durability at earlier stages and are thus not evolved traits. Rather, these additional life segments are due to "ecological release" and are referred to as "post-Darwinian" age classes.
Life span can be thought of as the sum total of the duration of each phase of the life course, either potential or realized. Thus implicit in life span extension (shortening) is an increase (decrease) in one or more of the phases of the life course. Because it is not possible to change one segment of the life course without affecting all other segments, life span extension (and shortening) will affect either directly or indirectly the timing and rhythm of all life events, from maturation and parental care to reproduction and grandparenting.
Life Span as an Adaptation
In evolutionary biology an adaptation is a characteristic of organisms whose properties are the result of selection in a particular functional context. Different bird beaks are adaptations for exploiting different niches that have had to be balanced with other traits such as body size and flight propensity. In the same way, the longevity of an animal is an adaptation that has had to be balanced with other traits, particularly with reproduction. The variations in the relationship between reproduction and longevity can only make sense when placed within the context of such factors as duration of the infantile period, number of young, and the species' ecological niche–the organism's overall life history strategy. Indeed, the longevity potential of a species is not an arbitrary or random out-come of evolutionary forces but rather an adaptive one that must fit into the broader life history of the species. In as much as life spans differ by 5,000-fold in insects (2 days for mayflies to 30 years for termite queens), by 50-fold in mammals (2 years for mice to 122 years for humans), and by 15-fold in birds (4 years for songbirds to 60 years for the albatross), it is clear that life span is a life history adaptation that is part of the grand life history design for each species.
A Life Span Classification Scheme
The literature on aging and longevity contains descriptions of only a small number of life span correlates, including the well-known relationship between longevity and both body mass and relative brain size and the observation that animals that possess armor (e.g., beetles, turtles) or capability of flight (e.g., birds, bats) are often long-lived. But major inconsistencies exist within even this small set of correlates. For example, there are several exceptions to the relationship of extended longevity and large body size (e.g., bats are generally small but most bat species are long-lived), and this positive relationship may be either absent or reversed within certain orders–including a negative correlation within the Pinnipeds (seals and walruses) and no correlation within the Chiroptera (bats). Likewise, the observation that flight ability and extended longevity are correlated does not provide any insight into why within-group differences in life span (e.g., among birds) exist, nor does it account for the variation in longevity in insects where adults of the majority of species can fly.
A classification system for the life span determinants of species with extended longevity that applies to a wide range of invertebrate and vertebrate species consists of the following two categories: (1) environ-mentally selected life spans and (2) socially selected life spans (see Table 1). The first category includes animals whose life histories evolved under conditions in which food is scarce and where resource availability is uncertain or environmental conditions are predictably adverse part of the time. Some of the longest-lived small and medium-sized mammals (e.g., rodents, foxes, small equines, ungulates) live in deserts where rainfall and, thus reproduction, is episodic and unpredictable. Examples include gerbils, rock hyrax, and feral asses. The extended longevity of animals in this category evolved through natural selection. The second category, socially selected life spans, includes species that exhibit extensive parental investment, extensive parental care, and eusociality (the social strategy characteristic of ants, bees, wasps, and termites, featuring overlapping generations, cooperative care of young, and a reproductive division of labor). It includes all of the social primates including humans. The extended longevity of animals in this category results from natural, sexual, and kin selection.
This classification system places the relationship between life span and two conventional correlates, relative brain size and flight capability, in the context of life history. That is, brain size is related to the size of the social group and the degree of sociality, which is, in turn, linked to extended life span. And intensive parental care is linked to flight capability in birds and bats, which, in turn, is also linked to extended life span. For example, most bird species are monogamous, with both sexes helping in the rearing (e.g., one protecting the nest while the other collects food). The reproductive strategy of the majority of bat species is to produce only a single altricial (naked and helpless), relatively large offspring at a time–flight preempts the possibility of the female foraging for food while gestating multiple young. Thus bat parental investment in a single offspring is substantial.
Life Span Patterns: Humans as Primates
Estimates based on regressions of longevity against brain and body mass for anthropoid primate subfamilies or limited to extant (currently living) apes indicate a major increase in longevity between Homo habilis (52 to 56 years) and H. erectus (60 to 63 years), occurring roughly 1.7 to 2 million years ago (see Table 2). The predicted life span for small-bodied H. sapiens is 66 to 72 years. From a catarrhine (Old World monkeys and apes) comparison group, when contemporary human data are excluded from the predictive equation, a life span of 91 years for humans is predicted. For early hominids, to live as long as predicted was probably extremely rare; the important point is that the basic Old World primate design resulted in an organism with the potential to survive long beyond a contemporary mother's ability to give birth. This suggests that postmenopausal survival is not an artifact of modern lifestyle but may have originated between 1 and 2 million years ago, coincident with the radiation of hominids out of Africa.
The general regression equation expresses the relationship of longevity to body and brain mass when 20 Old World anthropoid primate genera are the comparison group. Ninety-one years is the predicted longevity for a 50-kilogram (110-pound) primate with a brain mass of 1,250 grams (44 ounces; conservative values for humans) when a case-deletion regression method is employed (that is, the prediction is generated from the equation excluding the species in question) and 72 when humans are included within the predictive equation. When six genera of apes are used as the comparison group, the regression equation is:
yielding a predicted human longevity of 82.3 years. Thus, a typical Old World primate with the body size and brain size of Homo sapiens can be expected to live between 72 and 91 years with good nutrition and protection from predation.
The contemporary maximum human life span of over 120 years based on the highest recorded age at death consists of two segments: (1) the Darwinian or "evolved" segment of 72 to 90 years; and (2) the post-Darwinian segment, which is the artifactual component that emerged because of the improved living conditions of modern society. Therefore the arguments that the maximum human life span has not changed in 100,000 years can be considered substantially correct when the "evolved" maximum life span is considered. It is clear, however, that this is not correct when the nonevolved segment of the human life span is considered: There is evidence from Swedish death records that the record age in humans (the maximum observed life span) has been increasing for well over a century.
Life Span Extension in Humans Is Self-Reinforcing
Improved health and increased longevity in societies may set in motion a self-perpetuating system of longevity extension. Increased survival from birth to sexual maturity reduces the number of children desired by parents. Because of the reduced drain of childbearing and rearing, parents with fewer children remain healthier longer and raise healthier children with higher survival rates, which, in turn, fosters yet further reductions in fertility. Greater longevity of parents also increases the likelihood that they can contribute as grandparents to the fitness of both their children and grandchildren. This self-reinforcing cycle, a positive feedback relationship, may be one reason why the average human life span has been continuing to increase.
The decline in mortality rates during the early stages of industrialization in countries such as the United States was probably one of the forces behind
the expansion of educational effort and the growing mobility of people across space and between occupations. Whereas previous conditions of high mortality and crippling morbidity (disease) effectively reduced the prospective rewards to investment in education during the preindustrial period, expectancy for a prolonged working life span must have made people more ready to accept the risks and costs of seeking their fortunes in distant places and in new occupations. The positive feedback of gains in longevity on future gains involves a complex interaction among the various stages of the life cycle, with long-term societal implications in terms of the investment in human capital, intergenerational relations, and the synergism between technological and physiological improvements. In other words, long-term investment in science and education provides the tools for extending longevity, which, in turn, makes more attractive further long-term investments in individual education. Thus humans gain progressively greater control over their environment, their health, and their overall quality of life.
The positive correlation between health and income per capita is well known in international development studies, usually interpreted with income as the determining factor. But the correlation is partly explained by a causal link running the other way–from health to income. In other words, productivity, education, investment in physical capital, and the "demographic dividend" (advantageous changes in birth and death rates) are all self-reinforcing–these factors contribute to health, and better health (and greater longevity) contributes to their improvement.
Carey, James R., and D. S. Judge. 2001. "Life Span Extension in Humans Is Self-Reinforcing: A General Theory of Longevity." Population and Development Review 27: 411–436.
Goldwasser, L. 2001. "The Biodemography of Life Span: Resources, Allocation, and Metabolism." Trends in Ecology and Evolution 16: 536–538.
Judge, D. S., and J. R. Carey. 2000. "Postreproductive Life Predicted by Primate Patterns." Journal of Gerontology: Biological Sciences 55A: B201–B209.
Wilmoth, J. R., L. J. Deegan, H. Lundstrom, and S. Horiuchi. 2000. "Increase of Maximum Life-Span in Sweden, 1861–1999." Science 289:2366–2368.
James R. Carey