Go to Chapter One Section Go to Book World's Review

Aging and Old Age
By Richard A. Posner

Chapter One: What is Aging, and Why?

The Process of Aging
As we get older, we "age." Or do we? Some gerontologists believe that the expression "normal aging" is an oxymoron; that "aging" does not denote a process at all, but merely describes a medley of unhappy outcomes. As we get older we are more susceptible to most diseases, so maybe all it means to "age" is to become increasingly afflicted by one or, more commonly, several diseases until finally we are overpowered. On this construal the only difference between an old person and a young one is that the former is likelier to be sicker. If he happens not to be sicker, then he will be identical to a young person.

One can acknowledge the fuzzy edges of the concept of "disease" without finding this conception of aging remotely persuasive. For one (little) thing, it ignores Aristotle's point--the changing balance over the life cycle between imagination and knowledge, a change that cannot be described, without great semantic violence, as a form of illness. For another, the idea that "aging" denotes merely an increasing frequency of disease ignores forms of physical and mental change that, while aptly characterized as marking a decline in capability, are again not aptly described as diseases. Most professional athletes in most sports, even if they escape significant injury, are "old" by their late twenties or early thirties, but they are not sick. Their muscles and nervous systems are not diseased in any useful sense of the word. It is simply that their reflexes and running speed have slowed slightly but critically. There are physiological causes of this slowing, of course, and they could if one wanted be called "disease" factors. But the "disease" of diminished athletic capabilities in one's twenties and thirties (or menopause in a woman's forties or fifties) is sufficiently different in the most socially relevant respects from such conditions as cancer, coronary artery disease, stroke, and diabetes to warrant--to demand--being called by a different name.

Aging is most usefully viewed as a process, one element of which is an inexorable decline across a broad range of bodily (including both physical and mental) capabilities: call this "bodily decline." Other elements of aging, or that are correlated with it--the nonsomatic elements, examined more closely in subsequent chapters--include the increasing proximity of death as we get older, which affects the balance between imagination and knowledge as intellectual resources and the incentive to invest in human capital; the effect of habit on adaptability to changed circumstances; the accrual of experience in working at specific jobs, as distinct from the accrual of general life experience that Aristotle associated with aging; and boredom as a consequence of long years of working at the same job. These are age-correlated changes, but they are not explicitly somatic and not all of them are declines.

The physical side of bodily decline (using "physical" narrowly, in the sense in which it is contrasted with "mental") involves diminution in such areas as athletic and related motor capabilities, reflexes, and muscle tone; physical strength, energy, and stamina; acuity of vision, hearing, and other senses; fertility and potency; scalp hair, hair color, and the smoothness of skin; the efficiency of the immune system; height and the percentage of weight accounted for by muscle. The mental side of the declivity includes loss of memory (especially short-term memory), diminution in reckless physical courage and in sexual desire, diminished willingness to take financial risks, impairment of puzzle- and problem-solving ability, and reduced willingness to adopt new ideas or reexamine one's old ideas. Some of the psychological changes may not be entirely somatic--we shall see that unwillingness to reexamine one's old ideas has a counterpart in the behavior of business firms, which do not age in a physical sense--but all have, I believe, a somatic component. Mind and emotions, at least in a scientific perspective, are dependent on bodily states; and the same, or at least the same kind of, cytological and other physiological changes that produce the symptoms of physical decline likewise produce those of mental decline.

Although the process of aging can usefully be distinguished from the age-related increase in susceptibility to specific diseases, that increase is a reality which must not be ignored. Aging would have significance for issues such as the financing of medical care even if there were no normal aging process but just an enhanced susceptibility to disease.

Resistance to the fact that there is such a thing as normal aging has become common in our culture of heightened sensitivity. To some, age stereotyping is every bit as vicious as racial stereotyping. The concern is that if everybody is believed to age, this might be thought to imply that every old person is less competent intellectually than an otherwise similar young person. That would indeed be false. Two distributions can have different means but still overlap considerably. That is certainly the case in comparing the capabilities of young and old people; and for two reasons--that people age at different rates, and that people start to age from different levels of capability. A 75-year-old who had outstanding capabilities when he was 30, and has aged slowly, not only may be immensely more capable than a 75-year-old who was mediocre at 30 and has aged rapidly; he may also--and this is what bothers people who complain about "ageism"--be more capable than a mediocre 30-year-old.

But that there is increasing variability within age cohorts and overlap between persons in different age cohorts does not refute the existence of normal aging; it assumes it. Even cognitive as distinct from purely physical aging--gradual until about the age of 65 and accelerating from then till death--is normal and, pending scientific breakthroughs at present unforeseen, inevitable for all of us. Further evidence is that good physical and even mental health appears not to retard cognitive aging significantly, as one would expect it to do if such aging were simply a by-product of illness. Nor does such aging appear to be, to a significant extent, an artifact of age-cohort effects due to changes over time in environmental conditions such as poverty and lack of education, or of sampling bias, or of disuse because of lack of intellectual challenge or stimulation ("use it or lose it"). Far from being "ageist," moreover, a refusal to acknowledge normal, and in particular normal cognitive, aging can create exaggerated doubts about the competence of old people, doubts that the conception of normal aging can allay. If cognitive decline is not a normal aspect of aging, but rather is always a symptom of disease, the implication is that the vast majority of old people are afflicted with senile dementia. Almost all elderly people experience a cognitive decline the symptoms of which are difficult to distinguish from the earliest manifestations of dementia; yet in most the condition does not progress to dementia. It is possible, given the steep age gradient of dementia (of which more presently), that anyone who lived long enough would become demented; but most people die before then.

To avoid confusing normal cognitive aging with dementia requires distinguishing carefully among the following terms: Alzheimer's disease, or, as it nowadays is often called, SDAT (Senile Dementia of the Alzheimer's Type), dementia, senile dementia, and normal age-related cognitive decline. Alzheimer's disease, though commonly used by lay people as a synonym for senile dementia, is actually a specific type of rapidly progressive dementia that produces distinctive changes in the brain tissues and that, though more common after age 65, can strike at earlier, sometimes much earlier, ages; sufferers from Down's syndrome are often hit by Alzheimer's disease in their teens. So renaming Alzheimer's disease "Senile Dementia of the Alzheimer's Type" has been a source of confusion; indeed originally the term Alzheimer's disease was limited to presenile dementia. Some students of the disease continue to distinguish between Alzheimer's (presenile) and SDAT (senile), although they appear to be a single disease which merely hits people at different ages, like many cancers.

Dementia is the most general term for disabling mental deterioration, and thus embraces a variety of specific disease states; and senile dementia denotes dementia in old people. SDAT appears to account for a majority of cases of senile dementia, perhaps as many as 80 percent, the rest being due to such diseases or conditions as stroke, alcohol abuse, Parkinson's disease, Vitamin B-12 deficiency, and hydrocephalus. For my purposes the differences between SDAT and senile dementia are unimportant. I shall use "SDAT," "Alzheimer's," and "senile dementia" interchangeably.

The number of old people afflicted with senile dementia is not known with precision but has been responsibly estimated at 11.3 percent of the entire 65-and-over population. The percentage rises rapidly with age. In the 65 through 74 group, it is only 3.9 percent; it rises to 16.4 percent for persons 75 through 84 and to 47.6 percent for those 85 and older. Thus, while almost all old people suffer from some cognitive decline, especially in fluid intelligence, which peaks earlier than crystallized, only a minority suffers from dementia, though it is a substantial minority.

We must be careful in interpreting these numbers. They do not differentiate between the mild early symptoms of dementia and the severe late ones, and a further complication is that mild dementia does not always progress to the severely demented state that is characteristic of SDAT. (This intermediate state between normal age-related cognitive decline and progressive dementia is called "Mild Cognitive Impairment," or MCI.) If attention is limited to cases of severe rather than merely mild or moderate cognitive impairment, the prevalence in the three age groups is said to fall to 0.3 percent, 5.6 percent, and 19.6 percent, respectively. But these may be underestimates. The population sample on which they are based excluded institutionalized persons, among whom the prevalence of severe cognitive impairment is higher. Even taken at face value, the figures show a very steep age gradient, implying that a continued rapid increase in the size of the very oldest age group will cause an even more rapid increase in the percentage of severely demented people in the elderly population.

To summarize the discussion thus far, age brings with it increased susceptibility to a number of diseases, somatic changes that are a consequence of normal aging, and nonsomatic changes that are a consequence of the same process. Somatic changes are of two kinds, physical and mental. Nonsomatic changes are of three kinds: the increasing proximity of death (a purely "external" change, a change in the person's environment rather than in himself), which is the emphasis in the literature of human capital; the increased ratio of knowledge to imagination in the cognitive balance--Aristotle's point; and changes due to time spent working (experience and its baleful obverses, inflexibility, boredom, and sometimes burnout) and therefore merely correlated with aging.

Not only are these changes correlated with each other, all being related to age; some interact. Of particular significance for later chapters is the interaction between the decline of fluid intelligence and the knowledge shift identified by Aristotle. The combined effect is a pronounced age-related shift from abstract to concrete reasoning, or in terms of another useful Aristotelian dichotomy, from exact (logical or scientific) to practical reasoning. This immediately helps us understand why, for example, adjudication is a more geriatric profession than theoretical physics.

We should keep in mind that not all physical and mental changes correlated with age are seriously negative and that some are even positive, depending on circumstances. The "redistribution" of hair from the scalp to the body, the wrinkling of the skin, and the thickening of nose and ears have only cosmetic significance, though that is, of course, important to many people. The pluses include escape from the diseases of the young and slowing in the rate of growth of cancer cells. And there are changes that are pluses for some people (perhaps for society as a whole), though not for others, such as the reduction in sexual drive, anger, and aggressiveness that accompanies diminished production of testosterone. Some middle-aged and even elderly people are better-looking than they were when young.

The symptoms of old age do not all appear at the same time or progress at the same rate. Despite my reference to "inexorable" decline, symptoms of aging sometimes appear suddenly, as with the onset of presbyopia in one's forties, or of tinnitus, or of pattern baldness, and may plateau rather than continue to grow worse. The rate of decline differs not only across capabilities, but also, as I have pointed out, across persons, making the classification of people in age groups an inescapably arbitrary method of identifying the elderly. No one escapes the aging process, however, so that even a "healthy" old person will be less capable along a variety of physical and mental dimensions than an otherwise identical young person--though the qualification "otherwise identical" is crucial. The percentage of unhealthy old persons is much greater than that of unhealthy young persons, since age-related changes such as the diminished efficiency of the immune system increase susceptibility to illness. The probability of death doubles every eight years or so after a person reaches 30. The incidence of serious illnesses, especially of degenerative (as distinct from infectious) illnesses such as cancer, stroke, and heart diseases, also rises at an increasing rate with age.

Anyone who doubts that there are palpable, substantial, systematic, universal, measurable, demoralizing, and in the present state of biological and medical knowledge inevitable declines in physical and mental functioning even for the "normal" or "healthy" aged in this the world's most medically pampered society--anyone who believes that these age-related "declines" are a product of mass delusion or of vicious, irrational prejudice--is out of touch with reality. But there is scope for rational debate over when decline sets in, how steep it is, how much variance there is among persons within particular age groups, and the degree to which the cognitive effects of aging may, up to a point anyway, be offset by experience of life, including work experience, and by compensatory strategies such as being more careful or taking more time to plan or accomplish tasks. The rate of aging, moreover, mental and especially physical, can be, and is being, retarded by improvements in diet, by increased exercise, and by advances in medical technology. We cannot eliminate old age, but we can postpone it; we have postponed it. We are much less likely to think of a healthy 60-year-old or even 70-year-old as being "old" than we were thirty years ago. So while there are more "old" people alive today than ever before, there are fewer than the shift in the age distribution might be thought to imply.

And there is a danger of exaggerating the economic and social significance of the characteristic age-related declines in physical and mental performance. Declines in mental functioning tend to be measured by pen-and-paper tests and other laboratory-type experimental procedures that exaggerate the decline in useful capabilities over the life span. A related but more fundamental point, which I explore in chapter 4, is that the physical and mental capabilities of the young are often in excess of the economic and social demands placed upon them, so that up to a point--the point at which the excess has been aged away--the aging process may not cause a socially relevant diminution in capabilities. Another source of an exaggerated impression of the effects of aging is failure to grasp a point that I shall make in chapter 5--that elderly people rationally substitute time (which is cheap for them) for other inputs into activity and as a result move and speak more slowly, more hesitantly, than they are physically and mentally capable of doing.

In view of the large preponderance of women in the elderly population, an important question is whether aging affects men and women differently. If so, old men and old women would be on average more different from each other than young men and young women even after correction for transient features of the social environment, for example the fact that today's old women had less education relative to men than today's young women. This issue has been studied extensively, and as yet inconclusively. But it appears that, if there are sex differences in the rate or character of aging, they are small. There are, of course, more elderly women than elderly men. But it appears that, in a comparison of survivors, men and women of the same age are not at different points in the process of aging--do not differ in "agedness"--though they do differ, on average, in their proximity to death.

Aging and Evolution
We have not yet considered why the body (and hence mind) ages. The best explanation is genetic. Maintenance of an animal's body, like maintenance of an automobile, is costly. The more resources that are devoted to maintenance, the fewer that are available for the evolutionarily critical attribute of reproduction; a highly complex animal built to live a really long time would require a very long, and hence to parents very costly, period of gestation and infant development. So fewer of these built-to-last animals would be produced. If accidental destruction of such an animal, which maintenance would not prevent, was a significant risk, the added longevity might not offset (whether through the provision by this exceptionally long-lived animal of additional protection to its descendants, or by its having a longer reproductive life) the reproductive cost to its parents. Less complex animals, such as the turtle, that are at reduced risk of accidental destruction tend to be long-lived because the costs of "designing" the animal for long life are lower. But if reproductive fitness is sacrificed to durability unnecessarily, because the durability does not translate into commensurate survival, the parents of the more durable creature will have fewer descendants than their less durable competitors and their line will eventually become extinct.

This analysis, which is supported by evidence that animals age, mentally as well as physically, much as human beings do, explains why we wear out and die, and more specifically why the death rate increases rapidly after our prime reproductive period. But it does not provide much insight into the psychology or behavior of old people. For, as our sketch of the genetic theory of aging will have prepared us to see, it is unlikely that there is a genetic program for extended human survival, although this depends on how extended. It is relatively easy to see why in the evolutionary era--the prehistoric era in which, through the operation of natural selection, human beings evolved to approximately their present biological state--it might have been adaptive for men to live for several years beyond their prime, or women to live several years after menopause terminated their reproductive capacity. The older man could still reproduce, and his accumulated knowledge (particularly valuable in a preliterate culture) might compensate for physical decline in enabling him to furnish valuable protective services to his children and grandchildren. The older woman could better protect her younger children, who would not yet be fully grown, as well as assist in the care of her grandchildren. If these older (not elderly by our standards) people were valuable to their younger relatives during the evolutionary era, and hence valued by them--otherwise the older people would not have had good prospects for survival--this may explain why most people even today, even in the United States, feel some respect and protectiveness for older people, or at least for their own elderly relatives; these feelings may be instinctual.

The idea that nonreproducing relatives can promote inclusive fitness (the number of copies of their genes in their descendants), and therefore that there may be a genetic program for their survival and for their protection by their other relatives, is no longer a novelty. It is, for example, the key to the genetic theory of homosexuality. Equation 1.1 formalizes the idea. The optimal life expectancy of individual i at a particular age ([L.sub.i]) is shown as a function of i's remaining reproductive potential ([p.sub.i]) and of k's remaining reproductive potential ([P.sub.k]), where k is some relative (kin) of i, discounted by a measure of the closeness of the kinship ([r.sub.k]) and by the benefits that i's continued existence confers on k ([b.sub.k]), all summed over all of i's kin. So:

(1.1) [L.sub.i] = [p.sub.i] + [sigma] ([b.sub.k][r.sub.k][p.sub.k])

The equation shows that a reduction in an individual's personal reproductive fitness, say because of menopause, can be offset by an increase in the reproductive fitness of kin whom the individual assists.

Although evolutionary theory may explain the survival of persons to middle age, there probably were very few old people in the human evolutionary era. People lived in hunter-gatherer societies. Life in such societies is physically challenging, because, among other reasons, it is nomadic--people move around a lot. There is little surplus food with which to maintain people who are not directly productive (see chapter 9) and little surplus energy for carrying helpless old people from camp to camp. Only 8 percent of Yanomama Indians, a primitive South American tribe, survive to the age of 65, compared to 85 percent of modern Americans. In true neolithic cultures, as little as 2 percent of the population may have survived to the age of 50. With few people surviving to old age, there could not have been much natural selection among old people having different qualities--a process that would result eventually in the accentuation of those qualities that enabled a person to have more descendants--because there could not have been much variance. We know there were plenty of young women in the evolutionary era, just as today, so it is plausible to imagine that selection took place in favor of those having qualities--such as fertility and affection for children and attractiveness (and being attracted) to males likely to be good protectors of children--that would tend to increase the number of their descendants. But with so few old people for selection to work on, a comparable process is not easily envisaged for them. The social and material progress of mankind has brought about a stage of life that the genes have not choreographed.

Of course there is an element of circularity in arguing that the fittest elderly were not selected for because not enough people survived to what we would regard as old age to enable natural selection to work on them. If old age conferred a substantial benefit on one's descendants, perhaps by enabling the transmission to them of valuable information that would increase their reproductive fitness, we might expect more old people to survive, even if the "design" of their bodies that enabled such survival sacrificed some reproductive fitness to their greater durability. But the informational value of older people to society may not grow much after they are middle-aged; and it may be that only a tiny fraction of each age cohort need survive even to middle age in order to pass on to the succeeding cohort essential information about food, predators, and social structure. If so, there would be little evolutionary value to "engineering" human beings to survive into old age, and there would be an inevitable cost in diminished reproductive fitness. The underlying point is that our genetic endowment, including our biological "clock," is adaptive to a different environment from that of today. Recall my earlier point about excess capability. If life was physically and perhaps even mentally more challenging in the hunter-gatherer era than it is in the modern era, the young may have become programmed with physical and mental capabilities that are not required for most activities of modern life, while the old with their diminished capabilities may nevertheless be capable of coping in the modern era to a degree impossible in a hunter-gatherer society. The less that is demanded of human beings, the more likely they are to be able to meet the demand despite diminished capacity. Aging is not an accident of evolution, but survival to old age may be.

This discussion illustrates how evolutionary biology can cast light on social issues relating to old age even if survival to old age lacks survivorship properties in a Darwinian sense: even if, like birth-control pills and sperm banks, old age as we understand it did not exist in the environment from which we derive our genetic legacy. We shall consider in subsequent chapters, particularly chapter 5, other examples of the paradoxical fruitfulness of genetics in explaining the genetically unprogrammed stage of life that we call old age.

If natural selection implies that we are "designed" for limited durability, the conquest of old age by medicine may seem a quixotic endeavor, and the proper focus of geriatric research the alleviation of the disabilities of old age. Biologists who accept the genetic account of aging that I have been sketching do tend to believe that we simply are not programmed by our genes for indefinite life. They may be right. There may be a biological limit to the number of times human cells can divide and thus replace themselves as they wear out. But if so, the limit is not known and, in any event, may, for all one can know today, be extendable indefinitely by the medical science of the future. Just more effective control of known risk factors that are controllable with existing techniques would increase life expectancy substantially. Consistent with this suggestion, recent data from Sweden confirm the likelihood that life expectancy will continue increasing even in populations where it already is very long. Even without major research breakthroughs, it is entirely possible that a life expectancy at birth of 85 years is achievable. That would imply a large expansion in what is already a very large elderly population.

© 1995 University of Chicago

Back to the top