PART ONE
INTRODUCTION

One
Necessary Knowledge: Age and Aging in the Societies of the Past

Peter Laslett

For our purposes here, necessary knowledge, that is, necessary knowledge from the past, is a body of information that all persons must have to understand themselves as they are today. The historian, especially the historical sociologist, might claim that in theory at least the whole of history is relevant to present people and might demand a demonstration of how far age and aging should be accorded priority over other things that present people ought to be aware of. Such a demonstration is clearly required in the introduction to a book about aging in the past.

Why Do People Have to Know About Aging in the Past?

The populations of developed societies have grown old at an amazing pace. Within the last hundred years, and to a considerable degree within the last fifty years, the populations of Europe, North America, Australasia, and Japan have become far and away the oldest human populations of which we have knowledge. These populations are older—and still getting older—in two important senses: average individual lifetimes last for very much longer than they ever have before anywhere or at any time, and these populations have among them quite unprecedented numbers of elderly people. It is obvious that the situation could not always have been like this, and during the last century some European populations were getting younger. In England, for example, proportions of elderly people were decreasing slightly between 1800 and 1830. A knowledge of aging in the past, that is to say, the historical demography of aging, is evidently necessary in order to recognize how novel the situation now is in the advanced countries. There is no other way of grasping that fact.

I shall maintain that aging has been so sudden that there has not yet been time enough to take account of the transformation. Contemporary Europeans and citizens of societies of European origin must, along with the Japanese, be living in a state of cultural lag, even of false consciousness, at least to some degree. They are continuing to make assumptions about age and aging that, though they had always been true before the present century, have incontinently disappeared. They have to be brought up against the facts of aging as they always used to be and compare them with the very different facts as they now are.

The information produced by the historical demography of aging is therefore first and foremost knowledge with a view to ourselves. But the developed societies are not the only ones to be getting older in the last decade of the twentieth century. All contemporary populations are aging, including those classed as less developed or even "primitive." Young as they still are, sooner or later they will join the industrialized countries in their age-transformed condition. We shall illustrate this below for some Southeast Asian countries.

Nevertheless, the developed nations are much farther along the road. Except for Japan, they have spent a much longer time during the aging process than these other nations will, even though the change in the West has had to be described as so sudden and unexpected. It is the advanced countries that have to learn to adapt themselves, to modify their outlooks, to found the new institutions, to bring forth and develop the ideas and attitudes that are now for the first time required and that will be required into the indefinite future. At the present moment, they are in a situation not unlike that of Western Europe, especially of Britain, at the dawn of industrialization, the first to adventure into an entirely new world. The history of aging in the now developed countries has a significance for Brazil, shall we say, or India or China, analogous to that of the economic history of those developed countries, and the analogy is the more telling because of the fundamental similarity of demographic processes wherever and whenever they occur. The sooner the cultural lag or false consciousness, as I have called it, is made to disappear from the countries that already have age-transformed populations, the better for the world as a whole.^[1] The first requirement is that the facts should be known, the historical facts in particular.

There is another reason why the historical demography of aging is highly significant. Together with gender, ethnicity, and class, aging is one of the four dimensions of individual and social experience, though it has hitherto been given much less attention than the other three. It has to be known how the age composition of societies has changed over time, along with the longevity and the life course of individuals, if a properly historical social analysis, a durational analysis, is to be undertaken. Until this is done, we shall not fully comprehend the societies in which our ancestors lived, nor

shall we be able to grasp the contrast between their situation and our own. This means that we shall not be able to understand ourselves, to understand ourselves in time, to place ourselves in the procession, as I shall finally put it.

Remarkable as the aging of the populations of advanced societies has been, the phenomenon is rather more complex than it may seem and must be viewed in proper proportion. It is now frequently said that in a country like Britain, longevity has doubled and relative numbers of the old have trebled within the hundred years preceding our own day and that this has mostly happened within the lifetimes of its older citizens. As the figures in table 1.3 will demonstrate, however, this is somewhat of an overstatement as to changes in the two variables, though the historical period during which the change has occurred is correct.

In England the gain in expectation of life at birth over that period has been more like two-thirds for men and just slightly more for women, with a multiplication of some two and a half times for men and somewhat under three times for women in proportions over 60. ^[2] When the more revealing measures that are suggested below in the section "Experiential Measures in the Historical Demography of Aging" are applied, the change has been somewhat less than this in England with respect to longevity. Length of life after the fifteenth birthday, the suggested statistic for general comparison over periods of time, has not doubled but increased by more like a half, and at age 50, more appropriate for comparisons of this kind for those in later life, the rise is about one-half for men and two-thirds for women in Britain since the 1880s. If the share of those over age 60 of all those over age 25 is reckoned, the realistic experiential measure suggested for the weight in the population of senior people, there has been a twofold rather than a threefold gain, though rather less for men.

We need not exaggerate these historic changes, therefore, and we should be prepared to face the complications that arise when deciding on the most informative ways of describing them. In making out a case for the great importance of the history of aging, moreover, we must take due account of other drastic and unprecedented demographic changes that have recently taken place in Western countries and in Japan as well, though not to the same extent. The extraordinary increases in divorce, in births outside marriage, in numbers of persons living alone, and in consensual unions have gone forward over the last twenty or twenty-five years alongside the process of aging and have had their own individual and collective effects on social life. The combination of these developments with aging has made the position of the Western countries in the 1990s singular indeed. But these other alterations in behavior, though frequently dramatic, have been more recent and rather less general than aging, showing abrupt vicissitudes and varying from country to country.^[3] Their occurrence, in my view, does nothing to modify the claim that the historical demography of aging represents necessary knowledge.

This is the first volume, as far as we know, ever to be devoted to the historical demography of aging. The introductory chapter accordingly begins with a discussion of it as a subject itself and an appreciation of the extent and reliability of its results. We go on to describe and discuss the course of aging in developed societies, under the heading "The Secular Shift in Aging," a shift from a lower plateau to a higher plateau that has taken place, as has been seen, for the most part during the present century. The two temporal areas will alternatively be referred to as the Before and the After.^[4] We will insist that longevity, that is, expectation of life at birth and at later ages, is as important a constituent of the history of aging as proportions of elderly, though not generally recognized as such by demographers. New indicators will be suggested to supplement those in use at present, which are ineffective for historical purposes and removed from experience, experience in our own day as well as in the past.

No apology will be offered for giving so much space to these indicators and to the discussion of definitions, conventions, and measures for this newly emerging study. When in the early 1970s a book of this type was issued, opening up the then-novel field of the history of household and family (Laslett and Wall, Household and family in past time , with a comprehensive introduction by Laslett), much space had to be used in this way.

The Historical Demography of Aging

Historical demography proper began its career fairly recently, in the 1960s. Statistics for age and aging have always been among its results, but it is only during the last year or two that much attention has been paid to them, and the information so far available is rather restricted. Scarcity of data imposes stringent limits on what can be accomplished by historical demography in any field. Fertility and mortality, both essential to the reconstruction of the history, of aging, can only be worked out for societies that have recorded births, marriages, and deaths (or in Christian areas, baptisms, marriages, and burials), societies whose records of this kind are still extant and available for study. Migration, often critical in the study of age composition, generally requires information additional to the registration of vital events. The fact that these are preconditions of satisfactory analysis means that we shall never have anything like precise numerical knowledge of aging in wholly illiterate societies, past or present, or in literate communities that have failed to carry out the necessary recordings and to leave them to us for study.

Hence accurate estimates of life expectation and age composition even in highly civilized earlier populations like those of Greece or Rome, or in any European population before the end of the Middle Ages, will always elude us. Along with them go the peoples of the whole of the rest of the world in the premodern past, though here and there a tiny pool of uncer-

tain light glimmers in the darkness, nearly always fitfully illuminating groups of elite individuals.

It has to be said that this enforced ignorance makes it difficult for us to observe at all closely any population that could be supposed to have the aging characteristics of a wholly traditional society of the preindustrial kind existing on the lower aging plateau in a wholly traditional world. It might be thought that the developing societies, especially the "primitive" ones, surviving in our own day would provide just what we would like to have, instances of something like "natural" aging. This is scarcely the case, however, because these contemporary societies, if they carry out the required registrations at all, do so at the behest of literate minorities whose very presence may alter their aging characteristics to some degree. Moreover, such "backward" communities exist in a world so dominated by the highly industrialized countries that their longevity and age structure are hardly likely to be at the traditional level.^[5] We are in the same position as those who study birthrates, and our best chance of estimating a "natural" standard for aging lies in reconstructing the demographic history of the one or two countries that did maintain usable recordings for the whole of their populations during at least some stretches of time before the demographic transition and the secular shift. Since they have now become industrialized themselves, however, and since their material standards were probably already higher than those of the rest of humanity at the periods in the Before for which their situations with regard to aging can be recovered, the evidence of these countries is even less suited to show a "natural" aging condition than is the case with "natural" fertility.

Archaeodemography, the establishment of a general notion of vital rates through the examination of exhumed skeletal remains, has sometimes been used to prompt the record of the historical demography of aging (e.g., Laslett [1976] 1985: table 2a). Through this technique, indications of ages at death, of the life span, and of gender differentiation over the life course have been recovered; life tables have even been constructed. This evidence is all that is open to biologists of aging who have to interest themselves in the very distant past, distant enough to allow for natural selection. Historians concerned with much later periods have contrasted its outcomes with those derived from other sources with rather discouraging consequences. Although archaeodemography may serve for approximate aging estimates, the reckoning of limiting values for the most part, the sketchiness of its re-suits, and questions as to their accuracy and representativeness prevent it from being a source of much importance for the historical demography of aging as it will be expounded in this volume.^[6]

Studies of what could be termed the history of aging as distinct from the historical demography of aging are subject to somewhat the same judgment. Where particular written records have survived from the prestatistical era

which can be manipulated for the purposes of demography, though not created to that end, they can yield some estimates of age. Conspicuous examples are Ulpian's life table (3d century A.D. ) or the inquisitiones post-mortem in medieval England (inquiries at a landholder's death about his properties and claimants to them). The first appears in many works on mortality and life expectancy not concerned with history and the second, along with the evidence of poll taxes, in a comprehensive historical study such as J. C. Russell's British medieval population (1948) and in many more limited analyses.^[7]

When proper allowance is made for their limitations, biases, and inaccuracies, materials of this kind are of considerable use to the historical demography of aging. They could not be said, however, themselves to constitute that study over tracts of time in which nothing more systematic, interconnected, and informative is, or can be, forthcoming. Every detail pertaining to aging in past time is important to this book. But if nothing of greater value to our purposes could be recovered for former generations, our present project could never have been undertaken.

There are yet other studies of particular subjects in the history of aging that are of relevance to our purposes, and they will be touched on when we come to the aging of particular groups and restricted areas and aging at particular times. But we must glance here at the general, outline histories of the topic that its salience at the present moment has given rise to. They seem to be of much more limited value than the writings we have gone over. Composed, like some of the special studies, with the use of traditional historical sources, mainly literary evidence, and sometimes with scant respect for demographic analysis or even demographic reality, they seem to be preoccupied with past ideas about the division of the life course. They are to be classed as histories of aging attitudes and attitudes to the old rather than as histories of aging and will perhaps impede the progress of the study rather than forward it.^[8]

Nevertheless, no great stress will be laid on the contrast between the historical demography of aging and the history of aging. As we approach demographic analysis proper, however, it has to be pronounced that by and large the historical demography of aging is confined to the now developed countries and, with salient exceptions in the history of those developed countries themselves, to the period that is called the statistical era. The statistical era is the time during which states have carried out exact recording of vital events, analyzed and preserved those records, and made them accessible. For most of the developed countries, the statistical era begins some one hundred fifty years ago, in the middle of the nineteenth century, though a century earlier in Scandinavia, notably Sweden. For these Western nations a great deal of what the historical demographer wishes to know about age and aging can now be recovered. Census and local census-type materials, where and for what years they are open to examination,^[9] make it

possible also to investigate such things as the position in the household of persons of various ages, their kin relationships, and, by reference to their vital statistics, what has come to be called their life course transitions. Data of this kind have been extensively used in the chapters in this volume.

There are even some developed countries, notably those of Scandinavia, along with the Netherlands, Belgium, and Italy, where population registers were regularly maintained by local authorities in past periods and are now accessible. These are lists of all the inhabitants of a particular place distributed into households, giving ages and updated for every birth, marriage, death, entrance, or departure. In these very special cases, if the registers have survived, in principle virtually everything that the historical demographer wishes to know about aging is available.^[10]

Since the 1950s and because of the activities of international statistical bodies such as the demographic division of the United Nations, data are plentiful for populations at all stages of development and multiply rapidly. Here the task of the historical demographer is to disentangle the figures that can be used to produce an intelligible account within that excessively narrow time period. As he or she works backward toward the beginnings of the statistical era, the evidence thins out rapidly, but in comparative terms the temporal depth remains woefully shallow. For by far the largest part of history in its conventional definition—past time during which written records were maintained and can be consulted—all there is to go on is demographic theory and analogy. This makes the historical demography of aging a very different pursuit from previously established forms of historical study. Not only is nearly all the available material confined to the last two or two and a half centuries, compressed so to speak in a dense layer on the surface of time, but almost all the notable action is in fact confined to that time space. The temptation that has to be resisted is to follow the unfortunate example of social scientists generally and to read history backward from the present moment. Our enterprise as to aging in the past is undeniably demanding and difficult, replete with somewhat hazardous inferential argument.

This comes out in the fact that the statements already made about necessary knowledge apply to all the world's populations over the whole of their history, while the relevant evidence is so recent and so concentrated. We have to do everything we can to make use of the records of a handful of national populations, as opposed to individual communities, for which reliable numerical knowledge of aging can be pushed back for a hundred years earlier than the statistical era, to the middle of the eighteenth century. France and the Scandinavian countries and parts of Italy are in this position. In one country, England, using the same means but in what might be called heroic form, the time horizon has been stretched backward another two hundred years to the mid-sixteenth century for the entire national population.^[11] All these statistics from the time before the official statistical era

have been derived from registrations made by the Christian church. We shall see, however, that even the four and a half centuries of the English historical record on aging, from the 1540s to the 1990s, are rather short for our purposes.

We now turn to the actual process of aging in populations of some size, nearly always national populations, and begin with the demographic transition itself.

The Aging of National Populations

Since the demographic transition consists in a monotonic and irreversible fall in fertility and in mortality, from generally stable high levels to less stable low levels, it must always lengthen life and inevitably give rise, in due course, to an increase in the proportion of elderly people in a population. However, like fluctuations in total numbers, aging in this latter sense, that of our second aging variable, is a second-order demographic effect. The increase in proportions of elderly persons does not come about as a direct result of change in first-order variables such as fertility, mortality, and migration. It is an outcome of the interplay between those variables as they change over time, on the one hand, with the initial age composition of the population, on the other hand. It is the case, however, that in most situations, certainly those that characterize populations on the lower aging plateau, a fall in fertility, that is as large and continuous as it has to be during the demographic transition always initiates and maintains a proportionate increase in the numbers of older persons. The part played by falling mortality, which goes forward at the same time, is negligible at these earlier stages of the change in age composition, though it is of course wholly responsible for increase in longevity.

The process is not easy to explain in terms accessible to those unfamiliar with demographic analysis. It is important to recognize that because the rise in the numbers and proportions of the old is a delayed rather than an immediate outcome of the demographic transition, the momentum which causes that rise continues after the transition is over. Different countries therefore show different patterns over time in changes in their age composition, in accordance with the differing calendar years over which they experience the transition and with the varying course that the transition may take. They also age at different speeds. We shall only be able to refer here to the historically more important of these differences between the records of the countries we are considering.^[12]

Countries have also varied in the timing and pace of aging in our first sense, in longevity, though we shall not be able to linger on this matter either. But the recognition of longevity as of equal significance to change in age composition in the aging process and its analysis is an extremely important matter.

Life expectation has been regarded by demographers hitherto as a function of the mortality prevalent in any given population and as nothing more. Accordingly, it is classified by social scientists and by administrators as a measure of general health and well-being, not in itself as a manifestation of aging, individual or social. Since their interest stretches over the whole of the life course, it is expectation of life at birth that is the commanding statistic, worked out and cited almost to the exclusion of longevity at later ages, except of course when life expectation at subsequent birthdays is the particular interest in mind. From our point of view, that of historical sociology, this is wholly too narrow a view of longevity and has to be expanded. Expectation of life at every age and rises or falls in its duration, the probability at one birthday of reaching a particular later birthday, are all crucial aging phenomena on their own account, both for the individual and for society at large, in our own day and assuredly in the past. What people have always wanted and needed to know is how long persons like themselves are likely to live, their friends and their relatives included, and social scientists need to know this too, if they are to grasp the durational reality.

The historical aging process in national societies will therefore be discussed here in terms of two variables, not one, two variables treated as if they were discrete. The first is change in longevity, or life expectation, that is to say, not only average expectation at birth in a population but increasingly as we proceed expectations at later ages. The second is change in the proportions of the population of those in the higher age groups, generally those over 60 but increasingly the share that these older persons made up of all those in mid- and later life, that is, all those over age 25. Tables 1.1 and 1.2 set out the statistics of these two variables in their initial form, expectation of life at birth and proportions over 60, for the populations of a number of now-developed countries between the 1880s and the 1980s. Numbers in the final columns in both tables will give some idea of the extent of the rise in values between the last quarter of the nineteenth century and the present day.

It is evident from the figures that in the 1880s there were considerable differences between the selected countries with respect to the two chosen aging variables. It is also evident that low mortality (high expectation of life) was not always associated with large proportions of elderly people. This comes out clearly in the contrast in the tables between France, on the one hand, and England and Wales, on the other. Longevity was lower in France in the 1880s, yet France had a two-thirds larger proportion of persons over 60. This contrast is even more marked when it is recalled that in Britain there was in progress during the 1880s a considerable out-migration that had continued for a long time. This made the British population older than it would otherwise have been. What made Britain older in this sense, of course, made countries of immigration like the United States, Canada, and Australia younger, because many of the migrants were younger Britons. The

variation between these populations and other similarly placed populations as they were in the 1880s and the rates of fall in fertility and mortality. that ensued during the transition ensured that they would age at different paces. But the effect to which we have already referred, monotonic fall in fertility initiating and sustaining change in age composition, was a constant, and it could be shown that the accompanying fall in mortality had very little to do with it until the last decades of the secular shift.

We have now reached the point that has to be called the pons asinorum of demographic aging, a bridge that has to have been crossed by everyone undertaking historical analysis of the subject. It is universally assumed by those who have never had occasion to examine the issue that populations grow old invariably because deaths go down and life expectation goes up. This mistake is the more understandable in that the demographic transition itself starts in most cases with a fall in mortality. But it is a serious mistake, nevertheless, particularly often made by historians without demographic knowledge. They think it natural to expect that if there is evidence of rising longevity at a particular period, proportions of older persons must have been growing, and vice versa. It usually does not occur to them that a fall in births will work to expand the proportion of the elderly in the population, that this effect could be reinforced by the very young living longer, least of all that a fall in deaths might contract that proportion because, especially in the Before, such a high proportion of deaths happened to babies.

The Secular Shift in Aging

The secular shift in aging in England is set out schematically in figure 1.1 in ideal terms and on a scale that goes back for some eight centuries into the pretransitional past, in the Before, and is projected forward in the After some six centuries from the 1990s.

The general shape of the curves conveys the idea of the secular shift as an upward thrust from a lower aging plateau stretching backward in time from the last decade of the nineteenth century up to a much higher aging plateau sloping upwards and stretching forward from the first decades of the twenty-first century. But in this hypothetical model, the higher plateau is a particularly speculative construct, and its course over time after the first few decades of the twenty-first century is very uncertain. Projections made by demographers specify a range of alternative futures, some of which are quite extreme. These uncertainties make it necessary to represent the future courses of both longevity and proportions in later life as areas of possibility rather than as determinate lines. The earlier plateau and the rise from the lower level after the 1890s are, however, quite well documented for a number of populations, as will be shown. In England, its steepest climb in both graphs evidently took place between the 1920s and the 1950s. These

Fig. 1.1.
Secular shift in aging: England, 1000s-2500s.

Fig. 1.2.
Expectation of life at birth and proportion of 
population over age 60: England, 1540s-1990s.

effects are very clear in the detailed representation of data for England set out in figure 1.2 covering the four hundred fifty years over which the requisite data are available. It is not possible to divide the genders until the 1870s, but the widening gap between male and female as we approach the 1990s is unmistakable.^[13] In figure 1.3, the English curves are set out again, along with directly comparable curves from France and Sweden, two other countries that also have evidence for both longevity and age composition stretching some way back from the beginnings of the statistical era, distinguishing male from female at the outset. Figure 1.4 compares Tuscany, that area of Italy in which Florence is situated, with England in expectation of life at birth for three hundred years after the early seventeenth century. The stretch of years between the 1640s and the 1890s in this figure represents the longest period during the Before over which life expectation at birth can at present be compared for two substantial European areas.

Apart from these pieces of relevant evidence, the English record is all we have to go on before about the year 1740 to assess the experience of aging

Fig. 1.3.
Expectation of life at birth and proportion of population 
over age 60: England, France, and Sweden, 1740s-1990s.

in Western countries—and in large part, the experience of the world as a whole—in properly numerical terms. This makes the statistics from England (and in more recent years, from England and Wales) strategically so important that all the relevant numbers for that country have been set out as table 1.3.

With this spread of information about England before, during, and after the secular shift in aging before us, there are two issues that have to be confronted if we are to appreciate critically its significance for the course of aging over the centuries and for the general subject of the historical demography of aging. These two issues are whether the English curves exaggerate the steepness of the rise in the two chosen variables and so cannot be called typical of the movement in all the countries concerned and whether the Before and the After are in fact to be thought of as plateaus. The upper plateau can scarcely be our present concern. Only the analogy of the level trajectory of the lower plateau, which we are about to discuss, justifies the suggested

Fig. 1.4.
Expectation of life at birth: England, 1540s-1990s, and 
Tuscany, 1630s-1930s. Data for Tuscany from Breschi 1990.

stretching forward of the higher plateau into the indefinite future, though the narrowing of gaps between survival curves in figure 1.5 hints at such a future development.^[14] Its possible or probable existence, however, gives a striking appearance to the secular shift in relation to what came before. It is to the lower aging plateau that we have to direct our attention.

Relative Historical Constancy in Aging Before the Secular Shift

Since England is unique in the way we have discussed above and since it was from the English evidence^[15] that the model of the secular shift in aging was originally constructed (see Laslett 1984), the question of whether England has been typical is not without its importance. There would seem to be little doubt that this was so with respect to the increase in expectation of life at birth, the first of our aging variables. France and Sweden, whose curves

for the course of this statistic over time are also depicted in figure 1.3, show a steep or very steep rise similar to that for England, a rise that began at some point in the last decade of the nineteenth century or the first decade of the twentieth. In the case of Tuscany in figure 1.4, the resemblance to the curve for England is quite striking, rather surprising in view of very different conditions of health and welfare. There seems to have been a slow upward incline of expectation of life at birth in all Western countries during the nineteenth century before the secular shift, and the Scandinavian countries were well above England by the time the really steep ascent began. Such a circumstance might be taken as putting England into the median position. Longevity has been excellently assessed in all now-developed countries for this period, and it is possible to be confident that England can stand for the rest. And if this was so in the run-up to the secular shift and for the shift itself, why should it not have been so in earlier decades and centuries?

The same cannot be so easily pronounced for our second aging variable, traced by the lower graph in figure 1.3, increase in proportions over the age of 60. Here the English figures show a tendency to fall in the earlier nineteenth century, as is evident in table 1.3, putting the English graph for proportions of elderly well below those for France and Sweden, and this can be confirmed in other countries. The subsequent precipitate ascent in England may well exaggerate the abruptness of the aging transformation at the secular shift. Even in other English-speaking populations, the share of the old rose more slowly than in the "mother country," though their character as immigrant receivers may have something to do with this.

The contrast between English experience of growth in the weight of older people in the population at large is particularly marked in relation to France. But it is well known that the demographic transition itself was more diffuse in France than elsewhere, and it has to be expected that in this respect the secular shift would also be more diffuse in that country. The general allure , as the French would themselves say, of the French curves is quite similar to that of the English and the Scandinavian. With such a small sample and in so novel and uncertain an area of investigation, the correspondence between these graphs seems acceptable. I am prepared to regard the shape of the secular shift in England as an ideal type of that development, ideal type in the sense used by Max Weber. We shall return to this point when we come to discuss more revealing aging measures than those we have used so far.

As for the levelness of the lower plateau, which is a way of expressing the long-term constancy of these two aging variables in historic populations, the course taken by the graphs in figure 1.3 certainly suggests that in the long term relative stability can be assumed as well. It is here that the length of the record over time becomes highly significant, so that even the near half-millennium covered by the English data may appear somewhat inadequate.

But the line representing the English statistics of proportions over 60 is conspicuously flat for the complete run, and those for the other countries being compared are very much the same for their interludes of record. A survey of the quinquennial figures for England contained in table 1.3 shows that proportions varied between extremes of 10.08 percent (maximum) in the years 1716 to 1720 and 6.54 percent (minimum) in the 1820s and 1830s but that three-fourths of the values fell between 6.80 percent and 8.50 percent. In spite of differences in the later eighteenth century and earlier nineteenth century, the impression of constancy remains with respect to the weight of older persons in the population at large. But we shall find ourselves wondering whether this was quite so evident for the proportion of elderly persons in the adult as opposed to the whole population and whether the rise was as considerable at the secular shift in that respect.

The course of expectation of life at birth before the secular shift has a much less even appearance than the course of proportions of elderly: ups and downs succeed each other in the English figures in a way that recalls the heraldic description dancetty . Nevertheless, the claim for an underlying constancy seems strong in this matter too. Variation between extremes of 41.68 years (maximum) in the 1580s and 27.77 years (minimum) in the 1560s—the propinquity in date of the maximal peak and minimal trough should be noted—is accompanied by a concentration of two-thirds of the values between 35 and 40 years. When the graphs for expectation of life at birth are smoothed by the use of moving averages,^[16] they also have a decidedly even appearance. Work in progress at the Cambridge length of life project suggests a surprising stability over time in the longevity figures for elite groups, back to the later Middle Ages for British peers and members of Parliament, back to the beginning of the Christian era for Chinese mandarins. It seems safe to assume that the secular shift is properly represented by a sudden, precipitous rise from a lower plateau to a higher level in both dimensions.

The implications of these considerations for the historical demography of aging must be quite evident. Although they were subject to quite sharp fluctuations in life expectation at birth-and here the effects of epidemics, wars, and food shortages spring to mind—our ancestors never seem to have been subject to aging changes on anything like the scale that has been experienced by the populations going through the secular shift. Recovery in duration of life was rapid after episodes of disaster, though it must be remembered that this recovery consisted largely of better prospects for newly arrived infants and children. Proportions of elderly persons in the population at large remained fairly constant, showing the same tendency to revert to the average after rises and falls, for centuries on end as far as we can tell.

This conclusion is reinforced by demographic knowledge and demographic theory. Although we have had to recognize that no example of a population in what might be called a "natural" condition with respect to ag-

ing is or is likely to be available, demographers have had extremely wide experience of populations that had not yet entered the demographic transition, or are in its early stages and hence so far not much affected by the secular shift. Their aging characteristics are indeed plateaulike. Episodes of low fertility along with relatively low mortality cannot have lasted long enough among these populations for them to grow old in the dual fashion that has been described for the secular shift. This has to be the case, since pretransition populations had and still have high birth- and death rates by definition.

Duration of Life at Every Age and Age Composition Historically Considered: The Rectangular Survival Curve

The theme of the historical demography of aging has been treated so far as if it were concerned only with those who reach, who have reached, or who are likely to reach late life. This may seem appropriate because gerontology, the scientific study of aging, is itself pursued in this way and because the chapters that follow here deal with such topics, many of them belonging interestingly enough to the period of the secular shift when it occurred to the population of the United States.

But the elderly have only ever been the topmost of the age levels of any society, and aging at all the other, lower levels are alike of importance to the social scientist, the historical sociologist especially. The system of distributing individuals into age groups—childhood, youth, middle age, and so on—found, but in differing forms, in every, society and at all times together with the interrelationships between those groups and their members are all of significance. Generational relationships are rightly recognized as a subject for anthropological and sociological as well as for historical and gerontological research. Taken as a whole these studies go by the name of age structuring, until recently pursued mainly by anthropologists, though with some reference to historical instances.^[17] Analyses of how age structuring differs from society to society and how far it has changed historically certainly belong with the historical demography of aging, since the relative size of age groups will change with demographic change. What is more, all members of all age groups above infancy bear the marks of their experiences as members of younger age groups. This highly significant fact is of greatest importance for those who have gone through the largest number of age divisions, or life course stages, that is to say, those who are old at any one time.^[18]

Both expectation of life and the relative sizes of the membership of age groups are relevant here once again. For it must never be forgotten that everyone is always getting older at all points in the life course and that prospective length of life to spend in each life course division, or in divisions

to come, makes a considerable difference to individual attitudes and other features of age structuring. The question that presents itself is how far expectation of life at all ages, not simply at birth, is likely to change over time (or to differ between countries and cultures) and how far relative age group sizes do the same.

An interesting conspectus of capacity to go on living and its fluctuations over time is provided in figure 1.5, which traces not so much values of years still to live as they have changed over the centuries but survival itself. Survival curves plot for successive dates the numbers of an original 1,000 newborn babies still alive after 1 year, 5 years, 10 years, 20 years, and so on, up to 85, or to 100 for the most recent case. The record is again the English one, the first country for which such estimates have become available so far into the past. Up to that for 1691-1695, each curve is based on evidence from one five-year period; after that the curves relate to briefer periods or a single year.

Fig. 1.5.
Survival curves for cohorts of one thousand newborns, by age group:
 England, 1541-1991. Data from Cambridge Group back projection files
 and English Life tables up to no. 15. Work of James Oeppen.

The lower aging plateau is evident in the curves up to that for 1841-1845; here lines are separated by intervals of half a century each. These lines are not only very close together but they change positions with each other in an order that is certainly not chronological. The one for the five-year period 1691-1695, for example, traces a course only just above that for 1541-1545 and well below that for 1591-1595, which in its turn is closest to that for 1841-1845, nearly two hundred fifty years later. The run-up to the secular shift is visible in the wide gap between 1841-1845 and 1891, when mortality was evidently falling but at some ages may still not have been below the levels previously reached for an individual year or so on the lower aging plateau. The subsequent course of the secular shift itself shows up vividly in the more and more conspicuous spaces between successive curves, this time temporally successive curves and separated by ten-year and not by fifty-year intervals as is the case for the earlier lines. The approach to the higher aging plateau in our own day is strongly suggested by the marked narrowing of the spaces after 1951.

Students of aging will notice that the figure as a whole is reminiscent of the theory of what is called "The rectangularization of the survival curve," which is discussed in chapter 6 of A fresh map of life (Laslett 1989; see especially the revised version in press). This predictive theory, propounded by the American researcher James Fries and based on the very much more accurate survival curves now available, might provide a seductive concluding act to the historical drama we are commenting upon here. The finale would be a dramatic compression of mortality, showing itself in the survival curve pattern as a conspicuous narrowing of the gaps between the lines in figure 1.5, tending toward their elimination, particularly at the extreme right-hand side. Simultaneously these overlapping curves would change their shape even further in the direction shown forth in all their predecessors, tending more and more closely toward the rectangular. These shifts would indicate that survival will get better and better until finally virtually all persons born would be alive in their later 80s or their 90s. At some point during those years, it is further claimed, the end will come for a larger and larger number as time goes by, and after briefer and briefer final illness a "natural" death will supervene, a "natural" death at the completion of the "natural" span of human life.^[19]

In Fries's model, then, morbidity, the tendency to fall ill, is to be compressed along with mortality. In favor of this dual hypothesis is that at all periods before the secular shift, on the lower aging plateau, that is to say, sickness and death were distributed over the whole life course: during and since the secular shift there has been an increasing tendency toward postponement to the later years. When the higher plateau succeeds after the secular shift is over, Fries's theory predicts that both sickness and death will be almost entirely concentrated within a very short period, most likely during the middle or late 80s of the life course. If the model were to be validated, its ef-

fect would indeed be that the survival curve will eventually become rectangular, fulfilling the tendency so conspicuous in our diagram of English development over half a millennium.

It is becoming evident, however, that survival curves in our own day are resistant to such an interpretation. Some of the number of competing theories of aging at the present time imply that they will finally approximate the rectangular shape. Otherwise the life span, known to be lengthening, will do so indefinitely. But the attractive thesis of a prospective compression of the incidence of final illness and death has to be regarded as not proven, a fact with evident implications for the uncertainties over the higher aging plateau. The issues it raises, however, decidedly intensify the interest of this topic in the historical demography of aging. For our present purposes, the significance of four and a half centuries of English survival curves, which as we have seen do not necessarily represent in detail what has gone on in other comparable populations, lies elsewhere. It is in the fact that their shape was indeed remarkably constant over time until the secular shift. Habits, attitudes, and institutional arrangements that were based on the assumption that most people would be alive in the same proportion of numbers at the same age as time went by and have roughly the same numbers of years to expect to live were apparently well founded.

It can be quite simply asserted, then, that over the three and a half centuries of English history before the secular shift for which the required knowledge has been established, age structuring was subject to very little change with respect to longevity. As has been hinted, the most interesting of the effects that finally supervened at the height of the shift itself is the emergence of the Third Age, which will be initially defined demographically in terms of survival by the establishment of a novel relationship between the number of those alive at age 25 and the number alive at age 70.

Exactly the same conclusion as to relative constancy during the Before can be drawn from figure 1.6, which traces the fluctuating boundaries between age groups composing the entire English population over the same long period. (Note that estimates before 1870 cover England and Wales; after 1870, the two entities can be distinguished in official figures.) All the age groups show continuity over time in their relative sizes until the secular shift sets in. It is true that the line bounding the elderly (those over 60, the lowest proportion of course until the end of the nineteenth century) is noticeably smoother than the lines bounding other age groups (0-4, 5-14, 15-24, 25-59). But all five boundaries are quite regular before the shift, regular, that is, within short-term fluctuation up and down. The only comparable reconstruction, that for Tuscany from 1810 to 1940, has a very similar pattern (Breschi 1990: 165). It seems to have been exceptional fertility falls in England during the 1550s and the 1640s to 1670s that led to the frontiers between shares of children (5-14) and adolescents (15-24) crossing each other for a little while. But otherwise age composition displays overall con-

Fig. 1.6.
Age composition of the population: England, 1540s-1990s.

stancy, and the statements made above about survival and life expectation apply here as well.

Stability in age composition did not survive the onset of the secular shift, and the area near the right margin of figure 1.6 demonstrates as much. Proportions over 60 rose abruptly until by the 1950s they became the second-largest age group and went on growing without interruption until the end. It is well known that this relative expansion was, is, and decidedly will be accompanied by a parallel change in composition within the open-ended group of those over 60. The very old—the 80s to 84s, 85s to 89s, the over 90s—will increase rapidly, along with their longevousness. This is a radical change from traditional age structure and one that creates an entirely novel and very formidable prospect for the present generation. Although we do not have the information to confirm the generalization in detail, it is very probable that at all earlier times the composition of the group of those very late in life tended to be constant in all larger populations.

Our ancestors, that is to say, those who were citizens of our countries before the secular shift in aging, could rely on a relatively stable age structure as well as a relatively static life expectation at every point in the age range.

All over their populations and at every period, the proportions of infants (0-4), children (5-14), adolescents (15-24), and those at working age stayed about the same, as did proportions of the elderly. The same statements could probably be made with some modification about the populations still on the lower aging plateau at the present time. The secular shift upsets this perennial pattern quite radically, and all age boundaries and traditional assumptions as to length of future life tend to lose their stability as the shift proceeds. The whole of the evidence we have so far sun, eyed suggests that this represents a unique occurrence in human history.

Aging by Locality and by Social Group

Up to this point, we have talked about duration of life and about age composition over long historical periods and always in terms of averages over whole populations. If we are to reconstruct the aging landscape and habits, attitudes, and institutional arrangements of the past, and if we are to get some insight into the results of the secular shift in aging, this procedure is satisfactory enough. Societywide beliefs and norms were based on perduring and very general experience, nationwide experience for the most part, and still are in whole tribes or whole societies, whatever is the proper unit to take. But it is of great importance to recognize that there could be very different situations with regard to aging in different localities and in different social groups before the shift, in spite of the continuities and the constraints on extreme variations that we have examined. Relative constancy did not necessarily obtain at the local level or among select samples of people during the Before.

Individual towns or villages, even fairly large areas with particular ecologies, could show noticeably irregular aging characteristics. At a place like Hartland in the English county of Devon, quite a sizable but isolated settlement at the end of a road leading up to higher land, expectation of life at birth could be maintained at something like fifty years for the three centuries from the mid-1500s to the mid-1800s. This was about ten years above the estimated peak level reached for the whole population of the country over that period and one not generally attained in England until after 1910. The operative cause was undoubtedly markedly lower infant and child mortality than elsewhere, though we shall see that this did not necessarily imply better survival in later life. Similar circumstances can be found in Oxford-shire villages at the same time.^[20]

As for differences by social class and social group, it is natural to expect that there would have existed during the Before some form of hierarchy in longevity, descending from top to bottom of the society, a hierarchy that would be concealed by the averages we have cited. The landowning nobility and gentry could be supposed to have lived longer than the substantial farm-

ers, the wealthy traders, the small landholders, and so on down through the modest tradesmen, the handicraft people, the laborers, and finally to the cottagers and paupers. Although this may have been the general tendency, and it can be confirmed for infants and children, there were surprising irregularities. In England during some of the earlier part of the period we can examine, the aristocracy actually died somewhat earlier on average than the rest of the population. Since their longevity was not higher at later ages, this may not have been due, as might be expected, so much to their having put their babies out to wet nurses rather than breast-feeding them themselves, like their social inferiors, and to the use of nourishment other than human milk (see Fildes 1986, 1988). There were evidently other causes such as a predilection of the nobility to lose their lives in warfare.^[21] Of one thing we can be confident, however. Everywhere in the traditional world death came sooner among the inhabitants, even the richer inhabitants, of cities and towns than among the rural majority. The explanation is the life-endangering conditions of urban existence, of which putting children out to nurse was one.

A study of a particularly interesting example of a privileged group, the advocates of Edinburgh, has recently been published showing that these prosperous lawyers could expect to live after age 25 to 30—the start of their professional lives—for longer than the peers of the realm, and by the later eighteenth century, for as long as any of the select European groups that have been investigated.^[22] Such bands of well-placed individuals could anticipate a considerably larger number of additional years after midlife than those in lowlier positions. But difference by locality was not confined to length of life: it existed in age composition. Wide variation is found from place to place in our second variable, proportions of elderly. In Arezzo, Italy, in 1427, 15.9 percent of the citizens were recorded as over 60 years old, half as many again as the maximum on the lower aging plateau for England and a figure well up to late-twentieth-century proportions in developed countries. The national census of Iceland in 1729 gives a figure of 16.7 percent.

Arezzo and Iceland were evidently exceptional cases, and we should be very cautious in trusting their accuracy. The other communities in the list of thirty or so from which these figures for proportions of elderly have been recovered for earlier times fall well within the limits given above for England before the shift, and some are as low as those for contemporary developing societies.^[23] These exceptional figures, nearly all of them for relatively small bands of persons, may perhaps require us to modify a little the account that has been given of the contrast between the Before and the After. Their greatest significance for the historical sociology of aging, however, lies in two other circumstances of prime importance.

One is the caution that has always to be exercised in generalizing from the statistics of a group or a locality, either as to length of life or proportions elderly, to larger populations and to general conditions. Apart from the vari-

ability always encountered when numbers are restricted, there is the question of migration, migration that was never absent from preindustrial European society and the extent of which consistently surprises those who first encounter it. The immigration of poor elderly widows to places where public and charitable support might be forthcoming or the opposite, the emigration of younger people in search of employment opportunities, could easily make the share of the elderly in the population of a small community much larger, and the contrary effect might reduce that share accordingly. The chapter by Andrejs Plakans and Charles Wetherell in this volume discusses some of these issues in an eastern European context in the eighteenth century. Likewise in the matter of life expectation. It is transparently clear that one should not argue from the characteristics of a select group of the better off, or of a particularly favorably situated community, to a whole region or country. Great ingenuity is necessary if any guidance as to general conditions as to the life duration is to be gained from such evidence.^[24]

The further point about arguing from figures from localities and from exceptional groups of people needs particular emphasis. In spite of the low or very low general expectation of life in the Before, particular individuals could live very long lives, even in places where mortality was high. Our early research into maximal length of life makes this evident. A survey at Cambridge of the information generously supplied by genealogists and local researchers, mostly studying their own forebears using parish registers and the official registration system, yielded the following results. Some 3,500 individuals were fairly reliably reported to have survived to age 80 or beyond, and a fair proportion of these died in their 90s. There were 89 (2.5 percent), 52 females and 37 males, who attained 100 years or more, 2 dying in the seventeenth century, 19 in the eighteenth century, 21 in the nineteenth century, and 47 in the twentieth century.^[25] This may seem an unimpressive outcome, especially in view of the very numerous references to the extremely old, particularly to centenarians, in the literature of former times and up to our own day. But it could be said to indicate that survival to the tenth decade of life or, very, very occasionally, even to the eleventh decade (centenarians) was just a possibility on the lower plateau.

This exercise also demonstrates a salient fact about the very old, which no one interested in such matters can fail to observe. Both in the past and the present, in the developed countries, and those yet to undergo the secular shift, exaggeration of the later ages of individuals was and is lamentably widespread, especially when it comes to reaching 100. It is so for the alleged centenarians themselves, for their relatives and even for officials, local and national. It is blatantly evident on tombstone inscriptions and other family records. The worldwide interest in this topic continues, and even grows, not unexpectedly in view of the known tendency for the survival of ancients to increase. It may be intensified by the distinguished recognition given to cen-

tenarians, as for example by the telegrams sent out by the Queen of England and by professional gerontologists who regularly include sessions on centenarians in their conferences, national and international, and who show little initial insight into the registration data. This encourages exaggeration and the disposition to point to numbers of centenarians as a mark of prestige, national and local. The historians of aging cannot be too heavily warned against these distortive influences.

A particularly exasperating and highly misleading fable about survival to a great age is the claim so often made for very special areas such as the Caucasus. Usually remote and mountainous, these are alleged not only to have exceptionally high levels of survival but to produce centenarians at an unwonted rate. Such localities were extremely unlikely to have maintained the highly reliable registration practices which alone would make such claims acceptable, and the stories about them have been described as being "as mythical as Methusaleh" (Laslett 1989: 109). It may be a long time before the wheat is sorted from the chaff in these matters.

Experiential Measures in the Historical Demography of Aging

We have yet to consider the actual impact of the secular shift in aging on the societies in which it has so far occurred. Before we go on to a preliminary assessment of these effects, however, we have to acquaint ourselves with the experiential measures that have been alluded to already. These are numerical measures that it is hoped come closer to the personal and social realities of aging change over time and of aging differences between societies. We shall then find it useful in relation to the secular shift to discuss an illustrative example of the application of the principles of the historical demography of aging and of the measures that will be suggested.

We begin with longevity and meet at the outset an obstacle to understanding that recalls the pons asinorum described earlier. It is natural for someone unacquainted with what the demographers call the life table to suppose that expectation of life—or average expectation of years still to come, to give it its full title—is always highest at the very beginning, at age 0. But this is not so, except under the very special demographic circumstances of the advanced countries at the moment and over the last few decades. Even in their case longevity is at its peak not at birth itself but a few weeks later. In all other populations average expectation of years still to come is higher at exact age 1 than at birth, higher again at age 2, and usually higher still at age 5, the expected peak value in all but contemporary developed countries. It can often be higher at age 15 or 20, or even sometimes 25 or 30. Let us take an example.

In the life table calculated for England and Wales as it was during the years 1989 to 1991, expectation of life for women was 78.67 years at age 0, 78.21 at 1, 75.29 at 5, and 64.41 at 15, all the later values being lower than the value at the beginning. For Canada in 1831, in contrast, it has been reckoned that expectation of life for women was 39.84 years at birth, 46.49 at age 1, 49.48 at 5, and 42.61 at 15, still over three years longer than at birth. The outstanding difference between the two populations, and such is always the case when comparing longevity between the Before and the After, was the very much higher infant and child mortality in Canada during the 1830s. The death rate of females in the first year of life in Canada in 1831 was 162.36 per thousand, whereas in England and Wales in 1989 to 1991, the corresponding figure was 6.78. In the second year in Canada, it was 135.76 and in England and Wales, 5.59.

If the rise in expectation of life in the Canadian figures at successive birthdays after the first seems paradoxical, because it appears to indicate that an individual person had longer to live at 2, 5, or 15 or later than at 0, this is due to the fact that the character of life expectation has not been recognized as an average, an average of the years still to be lived by a specific group of persons of a given age. Since such an average is the result of dividing the pooled total of years of life remaining to all members of the group by the number of members of that group, it must be possible for that average to go up rather than down after an interval during which an appreciable proportion have died.^[26] This follows from the fact that such a loss could well reduce the numbers of persons surviving by relatively more, even very much more than it reduced pooled years still to live. Such would be the case in spite of the fact that the comparatively very large number of pooled years still to come would have been lessened by the years lived by those surviving through the interval, or through any part of it. A suitable metaphor might be that those still present at the end of the interval say to each other with some satisfaction, "Now that those unfortunates who were due to die during the interval are out of the way, we can all go on to live longer." Here the word "all" must be taken to indicate the new average of years remaining to each of them, that is, their expectation of life.

There are several implications of this explanation for the comparative study of longevity over time and between populations and for the selection of more realistic longevity measures. It demonstrates that expectation of life at birth has serious disadvantages as a measure, only appropriate for comparison of length of life between the populations of advanced societies in very recent times indeed. In the 1950s in the United Kingdom, longevity at age 1 was still nearly a year greater for females than it was at age 0 and over a year greater for males. The efficiency of expectation of life at age 0 as an indicator of the experience of longevity in other populations is obviously

woefully impaired because it is so heavily affected by high death rates in the earliest years.

If we were to contrast the life expectation of Canadian women in 1831 at its maximum, that is, at age 5, with that of English women in 1989-1991 at its maximum, that is, at birth (49.48 years vs. 78.67 years), this would seem to be a more realistic comparison than that produced by contrasting expectation of life at birth on both sides. The outcome would be that there was a difference of 29.19 years in favor of the contemporary English women rather than 38.83, a substantial reduction of well over a quarter in the superiority of the present day. If both the populations had been on the lower aging plateau, of course, comparisons of maximum longevity would have to be for values at age 5 on both sides. Comparisons of this kind would have to be made, however, in the knowledge that babies and young children who would die before the age of 5 had been left out of account.

Nevertheless, the aging experiences of such young and very young persons might not necessarily be supposed to be of much significance to the subject we are pursuing, though reflecting on their position brings home the important fact that aging is different at different parts of the life course. The experience of proceeding from the fifth to the tenth birthday is certainly not the same as proceeding from the tenth to the fifteenth, let alone from the sixtieth to the sixty-fifth, and these obvious differences are even greater when the historical periods and social circumstances of the populations being compared are allowed for. It has to be reckoned that there is a degree of indeterminacy in comparing the experience of longevity over time and between societies and that there seems to be no easy way of overcoming it. This may be one of the reasons why demographers have not included longevity in their analysis of aging.

The compromise solution recommended here is that life expectation at age 15, not life expectation at age 0, should be taken as the value for comparing longevity between societies at all historical epochs and at all stages of development. This simple expedient has the following advantages over using expectation of life at birth. Taking life expectation at age 15 as the measure of longevity makes unnecessary the reckoning of life expectation at different ages on the different sides of the comparison. Moreover, it is a much better guide to changes in longevity in later life.

Taking up the contrast between Canadian women in 1831 and English women in 1989-1991 once again, the percentage difference in life expectation at birth was 97 percent, that is, English women had 97 percent longer to live than Canadian women as reckoned in this way. But at age 15 the difference was 51 percent in favor of modern English women, while at ages 50, 60, and 70, it was 50 percent, 57 percent, and 63 percent respectively, each value quite evidently considerably closer to the differential figure for expectation of life at 15 than the differential figure for expectation of life at

birth. The value at age 5 would have similar advantages over expectation of life at birth as an indicator for this purpose, and it has the attraction that it is the highest longevity value for a very large proportion of all populations on the lower plateau. But expectation of life at 15 marks the point in the age range at which in many of them the average of years still to come ceases to be higher than expectation of life at birth. It could be said, moreover, that expectation of life at age 15 represents a closer approach to the median experience of longevity in the middle range of the population. But once again it has to be borne in mind that the aging experience of all immature persons is being omitted, and in populations on the lower aging plateau this means a quarter or even a third of the total.

If expectation of life at 15 is adopted as that measure of longevity that is least likely to misrepresent the aging experience of populations that are being compared, the question arises, what change might this imply for the secular shift in aging? Values for expectation of life at age 15 (e₁₅ ) are accordingly graphed in figure 1.7 for England, France, and Sweden, the countries shown in figure 1.3. Values for Canada, the United States, and Japan are also represented for as far back as records go.

Comparison of the curves in figure 1.7 for e₁₅ with those in figure 1.3 for expectation of life at birth (e₀ ) makes it immediately apparent that the suddenness and steepness of the secular shift is much more pronounced with respect to longevity when expectation of life is reckoned for age 15. It is also evident that the experience of all the Western countries that appear in the figure has been very close, that England could indeed be taken for each of them without misrepresentation. Interesting, too, is the fact that the two North American, European-descended populations should follow the domestic European populations so faithfully but that their statistics are somewhat higher during the nineteenth century. The graph for Japan, however, though it follows a parallel course from a markedly inferior initial level, climbs from the lower to the higher plateau with an amazing steepness after the 1950s, almost vertically in fact, and rises above all the others. The aging trajectory for that country could be said to be something of a caricature of the secular shift as it has been experienced by Western countries, the extraordinary drop in the 1940s being an outcome of the Second World War. We shall return to these differences later on.

The generalizations that have been made about the secular shift, then, are emphatically confirmed for longevity by the use of the more realistic measure of life expectation at age 15. It might perhaps be asked why this indicator was not adopted at the outset. The answer has to be that it has a grave disadvantage in practice. Because the only statistic for longevity usually published for any population is expectation of life at birth, any other measure but this is very seldom likely to be available. For further discussion of this and other circumstances in the measurement of longevity in experiential

Fig. 1.7.
Expectation of life at age 15: Canada, England, 
France, Japan, Sweden, and the United States.

terms, see the appendix to this chapter together with table 1.A1 and figure 1.A1. The outcome of our consideration of longevity to this point might have to be that we must continue to accept expectation of life at 0 as the standard measure but that this measure must be used bearing in mind all the drawbacks and obliquities that have been set out.

We are not quite at the end of the complications about the reckoning of longevity for comparative historical purposes. There is an important difference between what is called period or generation life expectation, which has been used so far, and cohort life expectation, which has not been mentioned.

Period life expectation is nearly always the one that is used. But it is a synthetic construct. In reckoning the total years still to live, which have to be averaged out among every member of the population in question, it assumes that the death rates being experienced at all ages during the year at issue re-

main the same over the whole time during which the people concerned will go on living. It is almost as if everyone had his or her whole life experience in the current year. In reality, of course, these mortality rates will undoubtedly change as the years go by, perhaps not by very much but by enough to make this synthetic calculation a hypothetical estimate. The enormous advantage of the period calculation, however, is that it can be done for any chosen year or, more usually, for any two or three years, provided only that mortality by age is completely known for the year or years concerned.

Cohort life expectation makes use of the actual number of deaths experienced by a cohort (persons born in the same year) from the time of its appearance until the last member of that cohort has died, that is, something like one hundred years later, or even more. Cohort life expectation, then, is about as realistic as such a statistic could be. But it has the enormous disadvantage that it can only be calculated after every member of the collection of people in which we are interested has ceased to exist. It is for this reason entirely impractical to recommend cohort life expectation as being more realistic and true to experience than period life expectation. Only for historians studying populations existing at least a century ago would cohort life expectation be of use. Though such a statistic could in theory be used for comparisons between past populations, all of which existed a century or more ago, it cannot be used for comparisons involving any more recently existing population. What is more, the necessary data for such calculations so far into the past only survive for a small number of historic populations. Nevertheless, cohort life expectation can usefully be calculated for some purposes on occasion, such as reckoning of the Third Age Indicator, which will concern us later on. It is available for English cohorts born between 1541 and 1781 and again for those born between 1841 and 1876.^[27]

This brings us almost to the end of the considerations necessary in measuring expectation of life for historical purposes, though in the appendix to this chapter we shall look at yet a further possible indicator. This is the age at which there are a given number of years still to live (say, two years more or five years more), an age that differs in a most interesting way between the Before and the After. It goes down rather than up during and after the secular shift (compare especially Bourdelais 1993). As with all measures, that which is the most satisfactory, or perhaps in this case the least unsatisfactory, for general purposes may not be of much use for particular purposes. In describing the emergence of the Third Age as an outcome of the secular shift, a somewhat different statistic from that of e₁₅ or of expected years still to live will be suggested in relation to longevity. This will be the chances of reaching age 70 from age 25, the Third Age Indicator, or 3AI. It would be possible to argue that from the point of view of the interests of most students of aging—and certainly of most of the contributors to this volume—the 3AI, reckoned either in the period or the cohort mode, might

be even more realistic as a general measure than expectation of life at age 15, and closer to the experience of past people. I shall suggest later on that a low level of the 3AI might have disposed those people to write off the possibility of ever becoming old. These points are illustrated and enlarged in the appendix to this chapter.

Interesting as this possibility, might be thought to be, the major concern of those who now study aging is in the numbers, experience, condition, and prospects of those already in late life and their relations with their juniors and their juniors with them. It is this concern that informs the suggestion of a revised measure of proportions in late life, our second aging variable, a measure more realistic than a simple fraction of older persons in the whole population. The realistic, experiential measure suggested here is the share of those over 60 of all adults, of all those over 25, in symbols 60+/25+. A more detailed impression of the significance of older persons with respect to their numerical size in relation to that of other age bands, together with changes over time in such relationships, can be gained of course from figure 1.6.

This realistic age-proportional measure, relative weight of the elderly and old among all adults, passes over an even greater number of the nonadult, the immature, than the realistic longevity measure just discussed, life expectation at age 15. The justification for doing so is just as strong in my view, or perhaps even stronger. When attempting to sense the presence of those in late life in a society, it is not easy to see why every individual member of that society should weigh as much as every other. The young and very young are undeniably of significance in the social structure. We in our own day are not the first body of people to be conscious of the immature and maturing members of our society and of the crucial character of our relationships with them and theirs with us.^[28] The young and the very young are also of evident importance because they are largely dependent and their numbers and proportions must be known to study support relationships and transfers between age groups. In this respect, the young are in a position very similar to that of the dependent old. These are highly significant examples of generational interchanges and of age structuring. Where changes in the proportions of age over time are accompanied by changes in the flow of support, with the result that some cohorts are privileged over other cohorts, a fascinating set of issues to do with intergenerational justice comes into view.^[29]

These issues can certainly be classed as consequentially related to the historical demography of aging, but relationships as to aging experience between the various age groups are scarcely affected. Figure 1.8 traces the course of the ratio recommended, proportion of all adults who are over 60, for the four hundred fifty years of known English aging experience and the two hundred fifty years of French and Swedish experience, along with the briefer periods observable in Canada, the United States, and Japan. In my

Fig. 1.8.
Proportion of adult population (i.e., all those over age 25) over 
age 60: England, France, Japan, Sweden, and the United States.

view, it presents the clearest and most revealing comparative numerical account of the weight and importance of the elderly over the stretches of time which it is possible to observe and complements for aging experience of this kind the results we have just surveyed for the experience of longevity. The figure includes, it will be noticed, a line for the threshold of the Third Age, which, along with the 3AI itself, will be discussed below.

The features that stand out in figure 1.8 are once more the similarity with regard to the abruptness and the shape of the secular shift to what is to be seen for all the populations included in figure 1.3 and the somewhat greater unevenness of the lines, none of which shows quite the same smooth progression that marks the long English graph for proportions over 60 in the whole population portrayed in figure 1.2. The pronounced peak in the English profile in figure 1.8 during the early eighteenth century is somewhat disconcerting, since it reaches a height not seen again until the secular shift was well under way in the 1930s. At 22.4 percent, the proportion that those over 60 made of all English adults in the five-year period 1706 to 1710 was over two-thirds of its level now (1991 = 31.2 percent). It might be thought that the share of the elderly in the population was not as constant as has been made out, or that the experience of the English was less than representative.

Closer examination shows, however, that more modest peaks of this kind are present in the graphs for the other populations represented and that the general resemblance to the lines in figure 1.2 is quite pronounced. Once again the relative shortness of the lines in the figures, the woeful lack of temporal depth in our data, makes judgment difficult. Inspection of figure 1.6 and of the quinquennial figures in table 1.3 reveals that, during the

early decades of the eighteenth century, proportions over 60 were at their highest in the whole English series before the 1930s and that the age band 25-69 fell into a trough at the same time. Though neither movement was particularly sharp, their coincidence seems to have produced the effect we are examining.

The adoption of a more realistic, experiential measure for proportions of elderly scarcely confirms the arguments that have been presented using the whole population as a divisor in quite the decisive way in which the substitution of expectation of life at 15 does for longevity. The rise during the secular shift in the proportion of adults who were over 60 was only 57 percent in England. But the dissonance cannot be said to affect in any great degree the theory of relative constancy on the lower plateau, the abruptness of the secular shift, or of the general typicality of the English data. The reexamination of the aging processes and particularly of the secular shift by the use of experiential measures is highly illuminating, nevertheless. We shall find this point confirmed as we take up an illustrative example in the aging history of two neighboring European countries whose comparative development has preoccupied us so much.

Aging in France and England in the Eighteenth and Nineteenth Centuries: an Illustrative Example

The idiosyncratic demographic behavior of France has come up several times, just as it tends to do in the discussion of the demographic transition itself. It is of significance for understanding the secular shift in aging and the emergence of the Third Age for the following reasons. If it could be shown that French longevity, during the period when it is accurately known for the later eighteenth century and nineteenth century, along with proportions of elderly rose to levels that have been claimed never to have been reached before the secular shift was well under way in the twentieth century, then the assertion that there was a sudden upthrust from plateau to plateau would be less persuasive. So also would be the case for assuming an overall constancy in aging in the Before and, by analogy, in the After.

Perhaps more important would be the effect on the claims made here about the uniqueness of our own contemporary, developed societies. Experience of aging in a very important European nation during the eighteenth century might have anticipated to a significant extent the experience that has been described as confined to the developed countries at the present day, though eventually to be shared by all societies. Since what happened in France might well have happened elsewhere in the illimitable time space of the Before, then it would be likely that all social structures along with their constituted mentalities would have a resilience in the face of the effects of pronounced aging built into them. The secular shift, though the very ex-

treme of change, could perhaps have been accommodated by the traditional social structure and so not have led to that incoherence which, as has been claimed, marks the very old, developed societies of today.

Alternatively, if it could be assumed that France in the eighteenth century was the first country ever to have a foretaste of what was to come two hundred years later, then the French attitudes and practices at that time would have a peculiar significance in the history of aging. They would have taken into account the aging experience not simply of a restricted locality, or that of a select group, but of a whole national society, at that period a national society preeminent in the life of the Western world. During its Grand Siècle, France was intellectually dominant in Europe, the center of the Enlightenment yet the seed bed of that revolutionism that was to shatter the political, cultural, and religious framework of the traditional European world.

Though no commentator on age and aging in France in the eighteenth century has made statements as positive as these, a recent authoritative work by a historian of aging, David Troyansky, Old age in the Old Régime: Image and experience in 18th-century France (1989), certainly hints at them. He insists, quoting Pierre Chaunu, that the lengthening of human life was "the only great event of the 18th century" (14) and sees this demonstrated in the literary and cultural life of France at the time. They reveal in themselves, he seems to be maintaining, that France was then already exceptionally old. The scene changes for the nineteenth century, when French supremacy in these directions had disappeared and British industrialism had supplanted the Enlightenment as a hallmark of national success. Then it was the French statesmen and wiseacres of the time rather than subsequent historians who interpreted what they saw in a very different sense. They tended to blame French decline after the revolutionary and Napoleonic interlude, the military reverses and economic sluggishness, on the drag of excessive numbers of the old in the population.^[30] The object of this discussion is to show how easily interpretations of this kind could have arisen when the historical demography of aging was of an entirely unsystematic and impressionistic character and the comparative historical evidence for the secular shift in aging had still to be worked out.

Now that we have something like principled, if often inferential, knowledge of aging in Western countries over the two centuries with which we are here concerned, it is not difficult to see how ill-founded such judgments must be for France during the century of the Enlightenment. The simple juxtaposition in figure 1.3 of the French trajectories with those of England and Sweden shows such judgments to be improbable, and the graphs in figures 1.7 and 1.8 should dispose of them altogether. When the records of that country began in the 1740s, France was catching up to England and Sweden in longevity (expectation of life at age 15) but never clearly surpassed them, and the rise was not exceptional but of the same order as the fluctuations to be seen in longevity over the whole length of English record-

ings. Meanwhile, our more realistic indicator for the proportion of elderly among French adults behaved in rather the same way, except that it did go above the English level between 1800 and 1850 and was no doubt very high by European standards throughout the nineteenth century. But Swedish proportions were rising even faster and were well above the French by 1900, as were those of other Scandinavian countries. These circumstances for the French add particular significance to the possibility, suggested by recent work at Cambridge, that it was in the early eighteenth century that the longevity of adults amongst the elite began to differ from that of the masses in the Western populations.

If we were looking for an exceptional situation to justify an exceptional attitude arising from longevity and the relative weight of the elderly, it would not be in France over this period that we should be seeking it but rather in England. The population of that country was near or at the top of the scale of longevity in Europe from the 1740s to the 1940s and after the 1800s had an exceptionally low proportion of adults who were elderly. But this situation was certainly not to be classed as an anticipation of the present day, since while it continued the two aging indicators had opposite tendencies, high longevity and low proportions of elderly.

The imperfections of this illustrative example for the history of aging and its historical demography have to do with the methods and sources of historical sociology as well as the objective reality of the secular shift. Important and illuminating as may be the literature current in the past on the subject of aging and the elderly in the history of society and of culture, to seek in such sources indications of longevity in earlier times or of the numerical weight of the elderly in the adult population is a misdirected enterprise. It is particularly unfortunate, even potentially disastrous, when evidence of this kind alone survives to provide any indication of longevity or of proportions of elderly persons. The facts are, as we have seen, that such was the position as to knowledge of aging at almost all locations and points in time before the last two or three centuries of Western history.

With the rather uncertain exceptions cited earlier, such writings as the Old Testament, the poems of Homer, or the philosophical essays of medieval canonists and Renaissance philosophers contain practically all the evidence available. Under such circumstances, the proper policy for the historian is surely to make no attempt to use literary allusion as positive evidence for age and aging. The assumption has to be that the society in question must have been located at some level on the lower aging plateau, with the characteristics set out here for the Before and derived to a considerable extent from the theory and practice of demography in relation to primitive and developing societies. Textual allusions in such sources to growing old and to old people have to be taken as illustrative or complementary: their relationship to the demographic situation as it actually occurs can never be a reflection of reality.^[31]

The title with which this discussion began, Troyansky's Old age in the Old Régime , is an attempt not so much to recover from cultural sources the facts about aging as to confirm from all the indications the author can lay his hands on that old people actually existed in some numbers in eighteenth-century France, so as to drive home his claim for the importance of aging under the Enlightenment and the Ancien Régime generally. It might be supposed this is something that any knowledge of the historical demography of aging would make unnecessary. Troyansky goes to considerable lengths in this enterprise, citing a rise in the proportion dying after the age of 60 or 70 in nine villages whose families have been reconstituted by French demographers; an increase in the age at death of French bishops; an inaccuracy in the low figures worked out for adults at Caen in 1988; a growth in the economy and its general capacity for supporting the dependent elderly; and many other circumstances. He even cites figures for proportions of adults who were over 60, a ratio similar to that used here, but using age 20 rather than age 25 in the denominator. It has not been possible to cover all the bewildering array of arguments he presents, arguments that sometimes suggest he has not yet got across the pons asinorum of aging. It should be clear from figures 1.7 and 1.8 that France only became exceptional with respect to the relative weight of elderly persons during the nineteenth century and that it was never exceptional in longevity before the mid-twentieth century. The disposition of historians to advance a confusing medley of numerical and seminumerical arguments of this kind has to be described as an obstacle in the way of the understanding of age and aging in the past.^[32]

The Secular Shift in Aging: Its General Historical Position and its Outcomes

Industrialization and Modernization

The chapters in this volume can be read as a commentary on the results of the secular shift in aging, results that are shown to have been brought about during its historical course at particular places and for particular areas of activity. This is evident in Richard Wall's essay on the elderly in England and those with whom they resided, comparing the preindustrial past with the industrial present, the lower with the higher aging plateau in the language we have been using. Wall makes repeated references to the aging process, nearly all of them to rises in longevity, as does George Alter in analyzing railway pensions, though he refers to declining longevity as well. The other contributors make similar allusions, and it is evident that the secular shift is seldom far away from the changing situations that are being described, even when the subject is American home ownership or the elderly on the Texas frontier. The descriptions of facets of life as it was experienced by older persons at sites in traditional Europe—Hungarian experience in the case of An-

dorka, Latvian in the case of Plakans and Wetherell—seem deliberately aimed at conveying an impression of the position as it was in the Before. Some notion of what changed and what remained the same, a critical part of the task of historical sociology, can be gathered in this way.

Understandably, perhaps, considering the shortcomings of the available data that have had to concern us so much and the difficulties of the analysis, almost nothing is said in these chapters about the pace or extent of population aging change and their consequences. This introduction was not in the hands of the writers, and no compendium of the relevant demographic facts was available. None of our authors goes any way toward an estimation of the overall results of aging in Western societies, or of the historical situation and significance of the secular shift. Insofar as that estimation is yet feasible, it is very briefly attempted here, particular reference being made to such general outcomes as the transformation of the life course, the redirection of social transfers, and the swaying of the balance between the sexes leading to the feminization of the elderly population. Finally comes what in my view has to be called the most important of the outcomes for the present and future of developed societies and the future of societies yet to develop, that is, the emergence of the Third Age.

It cannot be expected that so profound a change in social relations and in the life of individuals should be adequately comprehended so soon after its crystallization as the present day, if indeed this change can yet be regarded as complete. This is particularly the case if, as has been claimed, the transformation itself has yet to be recognized and accepted and if even those who run the political and administrative systems of the advanced societies and determine their attitudes persist in a state of lagged awareness about the aging transformation. In the late 1980s, British Parliamentary candidates actually supposed that the way to get in touch with older voters was to go to the institutions where the infirm and dependent elderly lived "in spite of the fact that 96 percent of older people reside in ordinary housing."^[33] No better case could be made for the historical demography of aging being necessary knowledge.

Singling out the effects of the secular shift is made peculiarly difficult because of the other major movements taking place during the century or so when the shift occurred. In most but certainly not all the countries, this was also the time of industrialization, of social and political mobilization, and finally of procreative transformation, the time in fact of all those movements that usually go under the unsatisfactory rubric "modernization." The particular results of radical aging have furthermore to be distinguished from those of other demographic changes to which it is causally connected, continuing shrinking fertility, perpetually lowering mortality, and radical change in marriage, all supposedly embraced within the overall "modernization" model. This is not the occasion for dwelling on the difficulties and

the distortions, especially the historical distortions, that any such overarching concept as the modernization model is bound to bring about.^[34] But certain distinctions can be made here.

The secular shift in aging was by no means necessarily associated with the other changes being mentioned and is not so today. In some important countries, in Germany and Italy, for example, or the Scandinavian countries and Spain, the phase of intense industrialization did coincide with the early part of the secular shift. But these instances can be misleading, especially when they are seen as historically parallel with what is happening in the now-developing countries.

Industrialization in Britain occurred wholly on the lower aging plateau and in that country, modernization, insofar as the concept is applicable at all, was complete before industrialization began. In France, as we have seen, the secular shift was considerably more diffuse over time than elsewhere. In China as well as in other East Asian countries, as we shall see, the shift is already in progress. The risk in China and especially in the really laggardly countries is that the shift will proceed so quickly that economic development will be behindhand in providing means of support for the hugely expanding elderly population.

It may turn out indeed to be a fortunate historical chance for the Western countries that they have not experienced the secular shift under the conditions and at the pace that seem likely to characterize its trajectories in other areas of the world. In short, an analysis of the historical position of the secular shift leads to the conclusion that it has been independent of these other changes, though causally connected with some of those that are clearly demographic. The interconnections supposed by the modernization model and its theory of convergence seem not to be evident, in spite of the difficulty of disentangling aging from those other threads in the skein.

Age Structuring and the Life Course

Some of the outcomes are fairly straightforward nevertheless and have been hinted at as the shift has been described. The transition from a position of perduring constancy in life expectation and in age structuring to one in which family members live in company with each other for longer and longer and age groupings undergo radical disturbance has upset the relationships between age groups. It has changed the life course of individuals along with the developmental cycle of the family and of the domestic group as a whole. The immemorial association in family time between the exit of the parental generation and the beginnings of the independent family life of the child generation has gone forever. The consequent proliferation and prolongation of the situation known as the empty nest stage, in which the parental couple stays together in the home after the departure of their children, a situation so familiar in the contemporary West, are almost exclu-

sively the result of aging. The accompanying phenomenon, the rapid rise of the primary, individual, that is, single persons living alone, has causes other than aging, though this must be mainly responsible. Both existed in the Before, but it is probably the secular shift that has made them major characteristics of late-twentieth-century Western social structures.

As part of this development, age at inheritance has risen inordinately.

Long-expected one-and-twenty
Ling'ring year, at length is flown:
Pride and pleasure, pomp and plenty,
Great *** ****, are now your own.

Samuel Johnson's overfortunate youth, deliberately unnamed, be it noted, whose father had evidently died long before the great day, would be more like 58 or 60 years old at inheritance in contemporary Britain or America.

The accompanying elongation of vertical kin links, great-grandfather, grandfather, father, son, grandson, great-grandson, or even great-great-grandson, is entirely due to the secular shift. It reminds us that the populations of contemporary advanced societies, unlike any of their predecessors, cannot be envisaged in pyramidal form with respect to the distribution of age groups. They are more like rectangles posed on their shorter sides. The shift in flows of support between generations, particularly the appearance and spread of multiple dependency, in which a woman has at the same time to attend to the needs of her children or grandchildren and to those of her parents or even grandparents, is due to the same cause. This whole galaxy of consequences of aging has been brilliantly described by Michael Anderson but in a strictly preliminary fashion, nevertheless, for it is scarcely possible yet to know how many changes of this character have been, are being, or will be brought about.^[35]

The least specific but still extremely important possible outcome of the secular shift has still to be touched on. This is the possibility that the growth in the relative weight of the elderly in the population—the ever-increasing number of people with halting gaits, bent shoulders, lined faces, and shapeless figures—brought into being a hostility and even a hatred that was absent or less prominent in the Before, became strident during the shift itself, and persists in the After. This is a topic of considerable controversy, even of contradiction, and the interplay between demographic causes and cultural influence is especially hard to unravel. For it has to be insisted, as Daniel Scott Smith has repeatedly pointed out, that there was cultural change with respect to the elderly during the secular shift which went on separately from aging itself: mere demographic determinism is quite unacceptable in trying to understand what happened. There is a good case for supposing that there was a change for the worse for elderly British people in the later nineteenth and twentieth century, and historians of aging in the

United States are at odds as to when a similar but earlier change occurred in that country, although they seem convinced that it was an objective fact.^[36]

Uncertainties such as these can only be registered here and remain for historians of aging to investigate further. We glance now at two sets of changes much more securely attached to lengthening longevity and population aging: feminization and retirement.

Effects on the Sex Ratio and on Retirement.

As is evident from figure 1.9, comparing the growth of proportions of elderly in England and France during the secular shift in aging, the process altered numerical relationships between the sexes in later life in a conspicuous way. The widening of the gap between female and male proportions of elderly persons is evident in France from the 1820s but came rather later in England. This effect intensified as the secular shift proceeded and has continued until the present day. There is some indication in table 1.2 that in the southern and eastern European countries, the sex ratio of the elderly had been more equal on the lower aging plateau just as it often is in developing societies in the contemporary world. In 1880, there were more old men than old women in Italy,

Fig. 1.9.
Proportion of population over age 60: England
 and France, 1770s-1970s. Adapted from Laslett 1977.

Greece, and Bulgaria, as was the case in the eighteenth-century Hungarian village reported on by Rudolf Andorka in his contribution to this volume. The discrepancy in proportions lasted in Bulgaria into the 1920s, but at none of the dates in table 1.1 was expectation of life at birth higher for men than for women, close as the figures sometimes get. By the 1950s, the whole of Europe seems to have been fairly uniform as to the excess of women in the higher and highest age groups and as to their superior longevity.

We should not infer that this growing imbalance was caused by aging in a billiard ball sense; rather, the two accompanied each other, and living longer or being allowed to live longer gave women an opportunity, so to speak, to demonstrate more and more their superior capacity to persist in the face of natural hazards. However this may be, we have to recognize that in coming to terms with our age-transformed condition, we shall have to provide for a more and more feminine society of those in late life, with no tradition in our culture or in any culture to give us precedents or inspiration. We know little about the sex balance among senior people in the West before the secular shift, though in the small and unreliable samples analyzed for England (see Laslett 1977: 201-208), women already predominated in the seventeenth and eighteenth centuries, if not to the same degree that they do today. It would seem that neither in the Before nor in the After have social arrangements in this respect taken the facts of demography properly into account, though we are obviously less realistic than our forebears. The same indifference to, even defiance of, demographic "reality" is to be seen in the equally important matter of retirement.

Nothing whatever in the changes in age structure over the secular shift as portrayed in the figures could be said to justify the rapid and continuous withdrawal from the workforce at earlier and earlier ages that has marked recent Western social history. We have seen that the proportion in the working ages (15-59) was quite stable over the secular shift in England, though the mean age of the workforce has risen and will rise. As that change nears its end in the West, there is no surplus of older workers in the demographic sense in the way that there is a surplus of elderly women. Although account has to be taken of actions of government and employers in getting senior persons off the rolls of the employed, it is tempting to interpret retirement as a move by individual people in spite of demography, the assertion of a wish and a right to take advantage of lengthened life by changing the use of time. Above all this, of course, is the immense growth of wealth that was proceeding in all Western countries contemporaneously with the secular shift and that is doing so on an even greater scale in present-day Japan.^[37]

The growth of retirement in spite of the lengthening of later life, like the growth we shall observe in solitary living in spite of greater opportunities for elder people to reside with their married children, illustrates the fact that demography does not always have the expected effect on life-course transi-

tions, that fascinating topic handled in several of the chapters in this volume, particularly by Tamara Hareven and Peter Uhlenberg. We can go no farther in this direction here, but we must consider for a little while the familial effects of the shift, since the familial position of the elderly and old has been such a preoccupation of those who have pioneered the history of aging and who are well represented in our collection.

Familial Effects.

It might perhaps be expected that larger proportions of older people and the better survival of parents and grandparents would have led to an increase in multigenerational households all over Europe despite "modernization" as the secular shift proceeded, certainly in its early stages. It would seem, however, that this expectation is in general unjustifiable and that the effects were complex and multidirectional. Research has established (Wall 1989) that European countries vary very widely with respect to the living arrangements of their elderly citizens at the present day. There is no reason to suppose that this variation was less in the 1880s. It has been shown for England that while that society was on the lower aging plateau, independent living was quite a normal thing, though not very widespread for elderly people, even if they had married children with whom it was possible to co-reside. The tendency to live on their own is scarcely surprising in the European areas known to historical sociologists as maintaining neolocal household formation rules, that is, mostly the countries in the north and the west of Europe. In a nuclear family formed on such rules elderly persons have no right to join the families of their married children, any more than those married children have the right to rejoin the families of their parents. But it was always open to children to invite their parents into their households out of affection or because of failing health, or simply from a sense of duty. This evidently frequently happened. In southern and eastern Europe, where household formation rules were not the same, the secular shift may have progressively increased the proportion of multigenerational households, at least initially, but of this we have as yet little knowledge.^[38]

In all developed Western countries, however, there has been the movement we have mentioned several times, one which is distinct from the secular shift and opposite in its impact on family forms, a tidal drift toward solitary living at every adult age, especially at late ages. In Asia, where about three-fourths of elderly persons in the 1990s live with their married offspring, the position has been very different.^[39] It can be seen how intricate are the questions that have to be faced if the Western countries are to meet the challenge of providing precedents of use to those who will follow them on to the higher aging plateau.

There can be no doubt that the increase in the numbers of elderly, potentially dependent parents in European societies which came about as a re-

suit of the shift has led to an intensification of kin interchanges, particularly between mature children and their still-surviving parents. It is fascinating to see in a recent study of an industrializing town in the United Provinces where the population registers were kept how these interrelations grew more frequent and important in the later nineteenth and early twentieth century and how the domestic group itself frequently changed its structure in reaction to these developments (Janssens 1993). It can in general be said, however, that the final stage, the era of the older married couple living alone after the departure of their children and of the elderly solitary widow, widower, divorcée, divorcé, spinster, or bachelor had already begun by 1950.

Yet it is notorious that the image of the family group in the eyes of Western journalists, politicians, advertisers, and citizens themselves obstinately remains that of the married couple with two or more young children living with them. The great and growing numbers of elderly people living alone and the virtual absence of three-generational households have still to penetrate the consciousness of Western populations and to be recognized as a reality by Western politicians and intellectuals.

Effects on Social Transfers.

The secular shift, wherever and whenever it occurred, must have moved to some extent the burden of support for those at the working ages from maintaining large numbers of children to maintaining increasing numbers of the elderly. It seems unlikely that these changes would perforce have increased the total burden of dependency appreciably, since in most societies at most times dependent elderly persons are no more costly than dependent infants and children. We have just seen that support to the elderly in need was not necessarily given in the way of maintenance in the home, and it has now to be added that an appreciable proportion of that support, in England and perhaps other countries, took the form of transfers through the collectivity.

By collectivity is meant not simply the state but all agencies of social transfers that cannot be described as familial. The introduction of pensions in Bismarck's Germany and in Britain in 1908 marked an entirely new era in the extent and regularity of central state support for persons quitting the workforce. But the collectivity in European countries, evidently to varying degrees between country and country, had always taken rather more of a responsibility for demographic casualties such as failing elderly people than was realized until results of research on the point began to become available.^[40]

The traditional English Poor Law has been shown to have been comparatively generous to the elderly, providing something like pensions to widows as early as the seventeenth century, even continuing that support under the much more parsimonious New Poor Law of early Victorian times. In addition to these sources, transfers to the dependent elderly from towns and charitable institutions of all kinds, together with a system of elderly privileges, such as licenses to beg, to glean after the harvest, to ransack the hedgerows

for firewood, or even to pilfer, had been present in the traditional European social structure from time immemorial. There were multifarious arrangements for dealing with the incapacitated elderly, often agreed between the family and the collectivity, circulating them between households—of children, of kin, and of neighbors—and putting them into institutions where such existed. Nevertheless, transfers of this kind were unstable over time, plentiful at one period and scarce at others. There was evidently much ambivalence. But the truly decrepit, those entirely unable to care for themselves, could count more reliably on collective assistance. In England and probably elsewhere, it is an interesting fact that about the same proportions of old persons lived in institutions before the secular shift as do so today.

It seems clear, however, that the effect of the introduction of state pensions was to reduce the proportion of transfers reaching the elderly from these miscellaneous sources. After the middle of the twentieth century, very high proportions of the support of working-class elderly people began to come from the sole source of a state pension, a situation that seems to have become widespread elsewhere in Europe. It can certainly be observed among the workers of London in the 1930s.^[41] The point of importance is that the increase in the longevity and numbers of older people at the secular shift elicited collective support on a larger scale but certainly did not bring it into being.

Although usually assigned to the invention and expansion of the welfare state, more recent scholarly opinion is that universal pensioning at a steadily increasing level of those past work in twentieth-century Europe is perhaps better regarded as a continuation of traditional policies into a time when the enormous increase in disposable wealth made it possible to contemplate transfers on such an unprecedented scale, even to a rapidly expanding and increasingly older population. It would seem that given in this way the means to live independently, older people readily did so. There were other effects as well. The state pension has been shown to have led, for example, to the decline of the responsibility felt by factory and other employers for members of their workforce who had grown old in their employment, providing not simply occasional pensions but also a scale of diminishing demands in the way of effort, by assigning workers near the end of their time to such tasks as being night watchmen.^[42]

By the middle of the twentieth century, therefore, the secular shift had brought those retired from the workforce into direct contact with the nation state, in societies whose ever-increasing riches made it possible to support persons in later life at something like the level of workers in their active earning years. Except among the proletariat, savings over the life course, however, were surprisingly common even before the secular shift began, particularly in the United States, and were much enhanced in the decades of economic growth in the earlier twentieth century. Along with state support these resources gave to those in late life a prospect of per-

sonal independence on a mass scale quite unprecedented in the history of the old. Such were the financial prerequisites for the emergence of the Third Age in the fullest sense of that term, and they stand in exquisite contrast to the position as it was in all countries when they were on the lower aging plateau.

Nevertheless, dependent persons are always relatively poor, and dependent elderly proletarians in traditional societies like those of the preindustrial era in Europe are the poorest of all. The impression must not be given that the modestly placed 70-, 80-, and 90-year-olds of the past were ever anything but badly off in comparison to their juniors, and the historian is often puzzled that they managed to survive in the numbers that they did. In rich industrialized societies, as everyone should know, the working-class elderly are still wretchedly poor in comparison to the rest of the population. They are relatively poorer probably than the old persons of preindustrial times and getting relatively even poorer, at least in Britain. The outcome of this composite situation has been to create for contemporary citizens of developed nations a conviction that the now enormously expanded and expanding society of retired persons is an economic problem for themselves and a burden for posterity. They find it difficult to recognize that in the emergence of the Third Age their own personal future is being provided for in an entirely novel society.

The Emergence of the Third Age

The definition of the stage of life which in European countries has come to be called the Third Age is still uncertain. The term belongs to a division of the life course into four ages, put forward in Britain during the 1980s, which is in increasing use there and in Australia but not so much in the United States and Canada. Many circumstances have to be taken into account in determining the character of the Third Age and in maintaining its independence of birthday age. The time of retirement from a job marks its beginning for most people, and for them the Third Age will last until death, or the onset of the Fourth Age.^[43] Unless that interlude is substantial, the Third Age cannot come into being. The existence and development of retirement and the extent to which transfers and savings are adequate to make a life of self-realization possible for the retired are crucial circumstances. These, in turn, depend on the wealth of the society to which individuals of later life belong and on such things as the level of instruction, the numbers in education, and the availability of education over the whole life course. In this way, therefore, the emergence of the Third Age is inseparably bound up with national development and with political and social policy.

More fundamental perhaps than retirement from a job, especially for women, is the change in personal life that comes with the final departure

of children from the parental family and the arrival of what has already been termed the empty nest stage of the development cycle of the family. There has to be a sense of release from the responsibilities and trammels of the Second Age and a recognition of the opportunity to live a life of self-fulfillment thereafter. It goes without saying that the physical condition of the retired person has to be good enough for her or him to seize that opportunity. Judging this circumstance depends on the reconstruction of the history of health over the life course. The assertions on this topic made by several of our contributors, especially by George Alter, hint at one of the most important of the tasks of the historical sociologist of aging. Though more usually justified by the intimidating growth in the cost of invalidism in late life, a theme not pursued here but evident in practically all of the discussions of the subject, its relevance to the emergence of the Third Age of independence and creativity is clear.^[44]

The general theory of the emergence of the Third Age and of its relations with the Second and Fourth ages goes as follows. Though elements of such a life stage can be described during the Before, it could not properly exist then because its essential preconditions were absent, especially the demographic. The mass of individuals went without much in the way of an interlude between active life and old age, and old age was defined in terms corresponding largely to those we have used for the Fourth Age.^[45] In the absence of widespread and compulsory state-guaranteed retirement, the transition was not dictated by calendar age and would vary with family and generational position, with social class and occupation, with the economic situation, but above all with physical and mental condition. There is ample evidence of this in the essays published here in Part IV, Retirement and Mortality.

Prior to all these necessary features of the Third Age, therefore, are the demographic conditions that permit a large enough proportion of the population to live for long enough after the Second Age is over. It is the theory of the early discussions of the Third Age and its emergence (Laslett 1984, 1987a , 1989) that all the circumstances named, along with others that have to be passed over for lack of space, began to exist for the first time in most of the countries of Europe and in countries with populations of European origin by the middle decades (1940-1960) of the twentieth century, though the countries of eastern Europe may have to be excepted. The demographic qualifications have had to be decided in an entirely arbitrary fashion, and the suggestion is that the Third Age can only be present when at least half of a country's male population can expect to survive from age 25 to age 70 and when at least a quarter of adults, those over 25, are beyond the age of 60 years.

As for the survival condition, in life table terms, numbers alive at age 70 divided by numbers alive at age 25 has to be equal to one-half or more (1₇₀ /1₂₅> .5 for males). This is called the Third Age indicator, or 3AI.^[46] In table 1.4, values of the 3AI are given for England at fifty-year intervals from

TABLE 1.4 Demographic Qualifications for the Third Age, England and Britain, 1540s-1990s
		Number over Age 60b
Year		Number over Age 25
1541-1545	.292	19.19
1591-1595	.348	18.43
1641-1645	.325	18.67
1691-1695	.310	20.53
1741-1745	.297	17.87
1791-1795	.338	16.90
1841-1845	.391	15.25
1881-1885	.374	16.47
1891	.301	16.21
1901	.375	15.42
1911	.416	15.66
1921	.492	17.23
1931	.497	19.66
1941	.471	21.82
1951	.532	24.50
1961	.581	26.77
1971	.625	30.08
1981	.634	31.73
1991	.690	31.21
SOURCE : Calculated from data in the files of the Cambridge Group for the History of Population and Social Structure.
a Males only.
b Females and males.

the 1540s to the 1880s, along with proportions of adults over the age of 60, and at ten-year intervals thereafter until 1991. The fundamental differences between aging on the lower and higher plateaus are visible once again in the table, as is the complex relationship between the demographic transition and the secular shift. It will be noticed that the 3AI varied narrowly above and below .33 until 1901, a one-third chance for a male at age 25 ever attaining age 70. By 1951, however, the 3AI for men exceeded .5 for the first time, and proportions of all adults above the age of 60 exceeded a quarter for the first time by 1960. England (and Britain) had qualified demographically for the Third Age, on the criteria we are using, almost exactly at the midpoint of the twentieth century.

Climacteric in the Middle of the Twentieth Century

It would not be expected that all the developed countries should comport themselves in the same orderly fashion. It can be stated, nevertheless, that all the populations of the countries appearing in tables 1.1 and 1.2 had a

3AI for males of at least .5 by 1960 and some of them several decades earlier, but that none of them had a quarter of all adults over the age of 60 before the 1940s. Reference to figure 1.8, however, shows that within that decade or its successor, every population represented except Japan had risen decisively above the line denoted there as the qualifying level. In contrast, figure 1.10, with the further detail given in figure 1.11, shows that the dates by which countries qualified with respect to survival were spread over a much longer period. Sweden and New Zealand had a 3AI of .5 just after 1900; Denmark, by the 1920s; and the Netherlands and Italy (not in the figure), by the 1930s, still twenty years before the accession of the United Kingdom and thirty years before Japan.

It is only, therefore, when the course over time of our realistic, experiential indicator of the weight of older persons in a society is taken alongside the course over time of the 3AI that England, or the United Kingdom, can

Fig. 1.10.
3AIs (males only): Canada, Denmark, England, 
France, Japan, New Zealand, Sweden, and the 
United States. Data from National Life Tables. 
3AI = 1₇₀ /1₂₅ .

Fig. 1.11.
Detail of figure 1.10.

still be said to have had the standard, the paradigmatic, experience of the secular shift in aging. The accession of that country to the Third Age in the 1950s can, under these highly specific conditions, be said to have marked the movement of the whole body of the rich, highly developed countries of western and northwestern Europe, with their companions overseas, into the new demographic era where the Third Age can exist. The most significant of all the outcomes of the secular shift in aging had unmistakably manifested itself in the West.

Yet another tilt of circumstance occurred in England in the 1950s. It must have been in this decade that the almost vertical upthrust of the secular shift began to bend over toward the higher aging plateau, problematic as that plateau may turn out to be in the next century. The demographic history of the other Western countries appearing in the figures we have been at such pains to compare with the English would most probably show this same slight change of direction detectable in the drawing together of the decadal survival curves in figure 1.5 at corresponding points in their aging trajecto-

ries. Inconspicuous as this reorientation may appear to be, this presaging of the disappearance of the secular shift and the succession of the higher plateau were pregnant with the future of aging in the Western societies we have lingered over. It is surely correct to think of the 1950s as the time of the "Great Climacteric" for the West in the history of age and aging. It remains to be seen what will correspond to it in the history of populations that have yet to be caught up in the secular shift.

Age Trajectories Over the Secular Shift in the West and in Some Countries of East^[*]

We have made use of the demographic statistics of Japan in the course of the argument, as a supplement and a contrast to those of the Western developed countries. We have also repeatedly asserted that the historical de-mography of aging in every country, and this means, of course, a collection consisting as to a vast majority of less-developed countries, will sooner or later follow the same course. The higher will succeed to the lower plateau in all of them, and the secular shift in aging will do much to transform their social landscapes.

Here we will make use of the Japanese data as well as data from one or two other East Asian countries to give substance to these propositions and to justify the statement made at the outset that the Western aging experience has a transcendent importance because it has happened first. With these additional facts in front of us, we shall proceed to a conclusion.

Because our sources for this section are published demographic statements, it will not be possible in this very brief exposition of such facts as have become known to us to keep to the experiential aging measures that have here been declared necessary for the comparative pursuit of the historical demography of aging. We shall have to content ourselves with the cruder measures we began with and which have been subjected to criticism. We shall proceed, moreover, in a summary fashion, by way of commentary on the six figures that follow.

In figure 1.12, two features have to be noticed. The first is that the trajectories of the growth in proportions of elderly (here those over 65 rather than those over 60) are measured on a scale of years from starting points placed thirty years apart, that of Japan from 1950 and that of China from 1980. This circumstance makes the second feature even more remarkable, that the two trajectories should follow each other so closely. This is in spite of the gap between their starting points, which implies that the shifts were initiated in very different world situations and very different local, North-

[*] Written with Zhongwei Zhao using the data and figures that he prepared. For an extended version, see Zhao in press.

Fig. 1.12.
Change in proportion of population age 65 and older: 
China, from 1950, and Japan, from 1980. Data from
 Keyfitz and Flieger 1986; Wu and Du 1992.

east Asian, North Pacific situations. The enormous dissimilarities between the two countries—in size, in culture, in social structure, in history (especially recent history), in wealth, in degree of development—seem not to have mattered.

This parallel effect is certainly surprising and suggests a definiteness and autonomy in the secular shift that could scarcely have been inferred from the evidence of the other countries we have surveyed. It is true that the comparison is largely based on projected trends of population aging (from the fourth decade in the case of Japan and the second in the case of China). It is also true that such projected results are essentially determined by the demographic parameters that have been used, and they could vary considerably according to the assumptions being made. But many other projected results indicate that in China, the aging trajectory is likely to be very similar to that which has been and will continue to be observed in Japan unless unexpected changes in fertility or mortality occur to alter them.^[47] Short of an entirely improbable set of circumstances, or of borrowings from one set of projections to form the other, the trajectories and their close rapprochement must represent a genuine phenomenon.

In figure 1.13, four trajectories have been added to those represented in figure 1.12. The South Korean case is based mainly on forecasts, while the European cases rely to a far smaller extent on such projections. It is difficult to decide precisely the time when each population embarks on the secular shift, but for the purpose of comparing its speed in various countries, a starting point is arbitrarily chosen from the trajectory of the growth in propor-

Fig. 1.13.
Change in proportion of population age 65 and
 older: China, England and Wales, France, Japan, 
South Korea, and Sweden.

tions elderly. Here the point selected is that at which the proportion age 65 and over is close to 5 percent of the total population and thereafter rises rapidly and monotonically. Figure 1.13 shows that a rise in the proportions of those over 65 from a level of 5 percent to a level of 15 percent occupied about 150 years in France, about 115 years in Sweden, and about 90 years in England and Wales but about 60 years in China and Japan. In South Korea, if the projected trend continues after the year 2020, the increase in the proportion of the elderly may be even more rapid.

This conspicuous foreshortening of the time taken for drastic aging to occur has been noted before. In 1988, Naohiro Ogawa published figures showing that the years required for the proportion of the population over 65 to rise from 10 to 20 percent were 24 in Japan, 48 in Finland, the fastest European case, and ranged from 54 to 85 (Sweden) in the eight other European countries he selected (Ogawa 1988: table 8). All this implies that the secular shift has been very much briefer in Japan—about half of that in most northwestern European countries—and will no doubt be somewhat the same in China. The speeding up of the secular shift and the indication that it may intensify as country succeeds to country undergoing the shift in East Asia must be classed as singular phenomena.

We have dwelt a great deal upon the suddenness of the aging process in the West, on the extent to which it has been overlooked, and on consequent "false consciousness" about it. It seems that East Asia and perhaps other countries embarking or about to embark on the secular shift will undergo it with even greater rapidity. The Western precedent seems to be less help-

ful here. No one can yet tell what may happen in the way of social discontinuity and disorientation with regard to aging and age relationships after a shift of such rapidity.

A very similar reflection is suggested when figure 1.14 is added to the series, delineating comparisons in longevity. Here the time scale is a real one and the statistics observational. Once more the Southeast Asian populations have increased and are increasing their longevity far more quickly than ever the Europeans did, and they will outstrip them in short order if Japan is to be taken as precedent.

Figure 1.14 suggests something further (it suggests only, because the temporal depth is so shallow for the Asian populations). It may be—and the little historical work that has been done and that is known to us does nothing to contradict this—that expectation of life on the lower plateau in Asia, or in parts of Asia, was below what it was in Europe. There is a hint of this in the trajectory for Japan in figure 1.7. Early and near universal marriage for women, which we suppose to have been a widespread characteristic of

Fig. 1.14.
Expectation of life at birth (females only): China, 
England and Wales, France, Japan, South
 Korea, and Sweden.

these countries, would certainly have made it possible for their populations to survive in the long term with lower expectation of life than has been found in the West in the past. We must be a little wary in comparison here, of course, because expectation of life at birth, which is all we have, may interfere a little with an objective contrast. But our knowledge of the plateau-like character of longevity over long periods certainly seems to imply that the rises in Japan, China, and South Korea are very unlikely to be a short-term variation in an otherwise generally even trajectory.

If this conjecture is correct, the lower longevity plateau in these countries may have been somewhat beneath that for Western populations but probably within its range of variation. We have as yet nothing to indicate the extent to which it was horizontal and flat. A further inference, therefore, that can be made from the figures is that the secular shift may start from a more modest level than that which obtained on the lower plateau in the West and so may have a longer slope upward. It could therefore take less time, travel farther, and be markedly steeper in these East Asian countries than it has been in Europe, America, and Australia. This would apply to both of our aging parameters, expectation of life and proportion in later life, and it has to be said that the first East Asian country to proceed through the shift, Japan, has displayed all these characteristics. The still more remote possibility that Western countries were perpetually somewhat longer lived than others on the lower plateau has no support known to us other than the indications of these figures. It is an intriguing possibility nevertheless.

Figures 1.15, 1.16, and 1.17 modulate the inferences we have made but certainly do not overset them. They are all in real time, over the period spec-

Fig. 1.15.
Expectation of life at birth (females only): Asia,
 China, Europe, Hong Kong, Japan, Singapore, 
South Korea, and Taiwan, 1950-2020.

Fig. 1.16.
Proportion of population age 65 and older: China, France, 
Japan, South Korea, Sweden, and the United Kingdom,
 1950-2050. Data from Keyfitz and Flieger 1986; Wu and Du 1992.

Fig. 1.17.
Proportion of population age 65 and older: China, Hong 
Kong, Japan, Malaysia, Singapore, South Korea, Taiwan,
 and Thailand, 1950-2050.

ified. The trajectory shown in figure 1.15 for expectation of life in Asia as a whole during this interlude (taken from Keyfitz and Flieger 1990) does not suggest that the entire continent will have a longer-lived population than the West. The pattern of the lines in that figure points to East Asian exceptions, however, and hints once again that longevity, could have been less before the secular shift began than in England (see, e.g., the figures in table 1.3, above). It is important to note that all the individual countries and areas shown in the figure have by now stepped over the threshold of the 3AI.

Figure 1.16 shows Japan overtaking European countries in proportions of elderly persons and rapidly acquiring the oldest age composition yet contemplated. The conceivable long-term difference between the Asian countries and the West with respect to both parameters on the lower plateau— Asia being the younger—is difficult to appreciate in view of the difference in the measures in figures 1.16 and 1.17 for proportions elderly from those used earlier, where proportions over 60 were used rather than proportions over 65.

But if it is remembered that proportions over 60 are likely to be about a third as high again as proportions over 65, a small margin in favor of the hypothesis is apparent. Finally, figure 1.17 makes it clear that in spite of the probability that all the Southeast Asian populations represented will show a similar pattern of change during the secular shift, the present temporal relationship between their trajectories is somewhat complex. The closeness of the observed and projected trajectories for Thailand and Malaysia, South Korea and China, Taiwan, Hong Kong, and Singapore brings us back to what was said about figure 1.12 as to the near-identity in the course of aging in Japan and China.

This exercise illustrates many of the issues about the Before and the After in aging, the two aging plateaus, and the secular shift in rather unexpected ways. It puts the historical aging processes that the Western developed countries have experienced into a comparative context. And it adds emphasis to the contention that the historical demography of aging has to be seen in the widest possible context over time—past, present, and future—and over all the world's political and cultural areas.

Conclusion: Processional Knowledge

A lengthy exposition has been required to demonstrate that the historical demography of aging is indeed necessary knowledge from the past. Some of this knowledge has had to be technical, even more of it numerical, not a little of it to consist in invoking visual judgment as to whether one line of a figure does or does not resemble other lines. These are not the usual qualities of historical discourse. And in the end, the knowledge and the inferences expounded turned out not to be from the past alone but from the past and the projected future, with the historically insubstantial present taking up a great deal of space.

For the title of a recent analysis of the grave issues of justice over time, the phrase "processional justice" had to be adopted.^[48] It was found, and this will not surprise a demographer, that the realities to be understood escaped the analyst if they were confined to a time point. A cross-sectional view was only useful so far as it could indicate not only what is or was at a point in time but also what had been and would be. There is a similar logic in the use made of projections in the section above. We do not understand our-

selves in time, and this understanding is the peculiar duty of the historical sociologist, unless we see ourselves as placed in a procession. Hence the description of necessary knowledge from the past as processional.

There are two further examples from the historical demography of aging that help us to grasp what we should be after. One was touched on in the discussion of the Third Age Indicator, the 3AI. If at the beginning of your productive life the expectation of ever growing old is so low that the prospect can be sensibly disregarded, then your attitude toward your late and very late life is likely to be quite different from that of a person who can confidently expect to be old. We do not know how far this did and does mark the attitude of people placed on the lower aging plateau. There, as table 1.4 and figure 1.10 show, the value of the 3AI was often below .33 and was likely to have been lower for a fair proportion of those "decreasing the wealth of the kingdom." We do know one relevant thing about their behavior as to age and aging, however. As late as the 1930s in England, working-class people failed to save for late life, but they did save penny by penny for their funerals (Johnson 1985), which could of course come at any age for them as it could for all their predecessors.

On the lower plateau, between 15 and 25 percent of an original male cohort were alive at the age of 70, whereas 65 to 75 percent are alive at that age today. If the proletarians did write off their old age in the Before, they were acting rationally, especially if public money was there to support them in old age as it certainly was in London during the 1930s. But they were behaving as if prompted by processional knowledge, knowledge of a kind that must inform the writing of the history of aging and especially its demography.

Our further example is the same set of circumstances looked at from another point of view. In discussing transfers of resources, social as well as familial, to those in late life in the West, it is properly supposed that the recipients receive these resources as of right, as a part of the normative structure of the society they live in. These rights have been of particular importance in the West because Western rules of household formation could be said to impose hardship on older people, especially widowed and decrepit older people. It does so by removing their children from their families at the marriage of those children, "nuclear family hardship" as it has come to be called.^[49] On the lower aging plateau, these rights, although they can be supposed to have attached to every individual, were only ever exercised by very few. This is a point on which the strongest stress has to be placed. The others died too soon. However important their right to support at the end of their lives may have been to them when living, seen processionally, those rights were a nullity.

One way of looking at the much discussed crisis of support for the old in Western countries today is that the secular shift in aging makes it more and more difficult, perhaps in some estimates finally impossible, to sustain the

inherited, perduring Western social structure in this regard. To recognize this possibility is also to practice processional thinking. Such thinking will be even more urgent in developing countries, as we have seen. It should be evident that responding to the acquisition of necessary knowledge from the past about aging, even in the preliminary fashion in which it has been done here, is to widen the intellectual horizon of historians, sociologists, and social scientists at large, as well as to explore a hitherto almost unknown area of the human reality. It does much to consolidate the concept of historical sociology and to demonstrate its overarching importance.

Appendix: Indicators for Comparison of Longevity

Measures of longevity are extensively discussed in the text above. The section, Experiential Measures in the Historical Demography of Aging, proposes that values for expectation of life at age 15 (e₁₅ ) be substituted for expectation of life at birth (e₀ ) as the most useful and revealing measure for comparing longevity between populations, particularly between those on the lower and the higher aging plateaus. It was pointed out, however, that this measure would only very seldom be available because expectation of life at birth alone is usually published in the relevant documents.

One of the objects of this appendix is to make it possible to proceed in a rough and ready fashion from e₀ to e₁₅ . In columns 1 and 2 of table 1.A1, values for twenty-five "levels" of life expectation are set out for e₀ and e₁₅ , and the trajectories of both are depicted in figure 1.A1, along with four other sets of statistics having to do with longevity. The source of all the information is the regional model life tables and stable populations published by Ansley Coale and Paul Demeny (2d ed., 1983, with the collaboration of Barbara Vaughan). All the statistics, except those for the 3AI (column 3), are for females.

The sets of tables ranged under the title "North" in that volume have been chosen for citation because the figures there seemed to correspond more closely than those in "East," "West," or "South" to the experience of England, at least up to the early nineteenth century, that is, on the lower aging plateau (see Wrigley and Schofield 1989: 110, 198, and passim). If, as is claimed in the text, English experience was in general typical of other Western countries, then the values in table 1.A1 can be regarded as indicative of Western demographic development as a whole. After the early nineteenth century, however, the English figures (and so perhaps those of contiguous European societies and their descendant societies overseas) resemble Coale and Demeny's "West" rather than "North," and this tendency must be allowed for when using the table.^[50]

What follows is a discussion of the columns of the table and the corresponding lines in the figure, together with the character and values of the longevity indicators that the numbers in the columns represent.

The numbers in column 1 (e₀ ) necessarily draw a straight line in the figure, since Coale and Demeny's tables were set up so as to make an e₀ of 20 for females the basis of level 1 and an e₀ of 22.5 as the basis for level 2, and so on, increasing by increments of 2.25 years up to an e₀ of 80. The relationship between columns 1 and 2 (e₀ and e₁₅ ) illustrates the points made in the text. Expectation of life at age 15 is as much as 12.8 years higher than that at e₀ when e₀ is 20 (level 1) years (e₀ /e₁₅ = 1.64), and this superiority diminishes gradually to 8.6 years when e₀ = 30 (level 5, multiplier 1.28) and to 2.8 (level 10, 1.06) when e₀ = 42.5. Longevity at age 15 does not fall be-

low that at age 0 until a point between level 13 and level 14. Extrapolation will have to be made by users of the table: the figure for e₀ in question here is presumably about 51.5 years. Thereafter the value of e₁₅ steadily falls below that for e₀ , until at the highest level the difference is 14.2 years, proportionally about as much above e₀ as it is below it at the age of 20. This interesting relationship comes out clearly in figure 1.A1.

The course of the 3AI, column 3, the only set of figures for males in the table, was surveyed in the section, "The Emergence of the Third Age," above. This indicator shows a very considerable range of values and must be regarded as a highly sensitive and realistic measure of longevity experience. It can be seen that its course is parallel over most of its range to that of e₀ itself. It might be permissible to infer from these circumstances that when at level 16 female expectation of life at birth reaches 57.5 (male 53.8), the 3AI (male) reaches 0.5, the qualifying level in respect to longevity, it will be remembered, for the emergence of the Third Age.^[51] It is interesting that at the highest levels, 22-25, the trajectory of the male 3AI ceases to be roughly linear and curves upward.

Mean age at death, column 4 in the table,^[52] is sometimes used as an approximate indicator of longevity, which is the more understandable since details of the age of death of individuals on the lower plateau is the relevant evidence most likely to be available. Inspection of the table and the figure shows, however, that such a practice is almost entirely misleading, since the course of mean age at death is curvilinear and cannot be used as a guide to e₀ or e₁₅ unless complicated calculations are made. It so happens that mean age at death is within a year or two of e₀ between levels 4 and 6 (e₀ 27.5 to e₀ 32.5), but the two curves diverge rapidly both above and below that narrow band. The distance is greatest at levels 17 to 19 (e₀ 60 to e₀ 65), when mean age at death is 20 years less than e₀ and 13 to 15 years less than e₁₅ . The differences certainly narrow toward the highest levels. But it should be borne in mind that age at death is a function of population growth as well as of longevity, which makes matters even more intricate.

The full description of columns 5, 6, and 7 is mean age at which there are 5, 10, or 12.5 years still to live. Statistics of this kind have sometimes been recommended as measures of longevity and have an obvious usefulness from the point of view of individual experience.^[53] It is instructive for an individual to know when she (or he) is on average within 5, 10, or 12.5 years of the end of life and particularly instructive when Third Age living has become possible from the point of view of longevity, that is (as we have seen), at levels above number 16 in our table.

What may seem astonishing is that the graphs representing values in columns 6 and 7, particularly even in their courses or consistently parallel to each other, rise so very gradually throughout the twenty-five levels of e₀ . These two sets of values increase by only 35 percent and 40 percent while

Fig. 1.A1.
Longevity indicators. Females only, except males only for 3AI. Scale for 
3AI is .01 that for age in years. Numbers in square brackets after the
 titles of the variables on the lines correspond to the columns of table 1.A1.

the 3AI rises by 800 percent, and in columns 4 and 8 the values are 266 percent and 2,025 percent. These circumstances make transparently clear once again a point already heavily stressed in the text: in spite of the fact that very few could expect to live to the later decades on the lower plateau, those women who did might continue for an appreciable number of additional years. If the general expectation of life was as low as the early 20s (and a dozen years higher at 15), a woman could expect to go on for another 12.5 years if she was one of those lucky enough to reach age 60. As for column 5, it makes very clear how much later in life that limit to expectation comes, even in situations where longevity is low.

The remaining column, 8, headed "half of cohort dead," should explain itself as the age by which one-half of a set of persons born in the same year have disappeared. It is even more sensitive as an indicator than the 3AI and so of considerable potential usefulness in longevity studies. This seems to be recognized, however, more by demographers studying animals than those studying people. Unfortunately, this final indicator, like those in all the columns except 5 (whose disadvantages have been set out), requires life tables if it is to be calculated. For the really detailed historical study of longevity, therefore, a series of time-separated life tables is necessary.

It can be tentatively suggested from this discussion and from an examination of the table and the figure that during the Before in the West, that is, on the lower aging plateau, for women in the West expectation of life at birth was on average at a value corresponding to a point between level 7 and level 8, that is, in the middle and late 30s, and expectation of life at age 15, in the early 40s. Something over a quarter of their male counterparts could expect to live until age 70 if they reached age 25, only about one-half of the proportion required to satisfy the longevity criterion for the Third Age. Average age at death for women was late 20s or early 30s, but women who reached the early 60s could expect to go on for 12.5 years more, for 10 years more if they attained the mid-60s, and for 5 years more in their early 80s. One-half of those born in the same year would, on average, be dead by their early 40s.

During the After, on the higher aging plateau as it is beginning to unfold, expectation of life for women at birth is likely to be in the 80s, probably inching upward: at age 15, in the mid-60s. Mean age at death is likely to be in the late 60s or early 70s, and the prospect of men surviving from 25 to 70—from the beginning of the Second Age to the middle of the Third Age—is over 80 percent. From the point of view of length of life, therefore, the Third Age will be absolutely securely established. In her early 70s on this higher plateau, a woman could probably expect to live for 12.5 years more, for 10 years more in her later 70s, and for 5 years more once over age 85. At an average age of over 80, probably slowly rising, one-half of all women born at the same date will still be alive.

It can be concluded with confidence that we inhabit an entirely different world with respect to length of life and experience of living it than our ancestors did, both as individuals and as a collective society. Necessary knowledge, indeed.

References

Acsádi, G., and J. Nemeskéri. 1970. History of human life span and mortality . Budapest: Akademiai Kaido.

Anderson, Michael. 1985. "The emergence of the modern life cycle in Britain." Social History 10(3): 69-87.

Applebaum, Andrew. 1978. Old age in the new land . Baltimore: Johns Hopkins University Press.

Bonfield, Lloyd. 1989. "Was there a 'third age' in the preindustrial English past?" In An aging world , ed. John M. Eekelar and D. Pearl, 37-53. Oxford: Oxford University Press.

Bongaarts, John, T. K. Burch, and Kenneth W. Wachter, eds. 1987. Family demography: Methods and their application . Oxford: Oxford University Press.

Bourbeau, R., and J. Legaré. 1982. Evolution de la mortalité au Canada et au Québec, 1831-1931 [Evolution of mortality in Canada and Québec, 1831-1931]. Montréal: Presses de l'Université de Montréal.

Bourdelais, Patrice. 1993. L'âge de la vieillesse [The age of aging]. Paris: Odile Jacob.

Breschi, Marco. 1990. La popolazione della Toscana dal 1640 al 1940 [The population of Tuscany from 1640 to 1940]. Florence, Italy: University of Florence, Department of Statistics.

Bulder, Elles. 1993. The social economics of old age: Strategies to maintain income in the Netherlands 1880-1940 . Tinbergen Institute Research Series, no. 50 (P.O. Box 14791, 101LG, Amsterdam).

Chesnais, J.-C. 1990. "Demographic transition patterns and their impact on the age structure." Population and Development Review 16: 327-336.

———. [1986] 1992. The demographic transition: Stages, patterns and economic implications . Oxford: Oxford University Press. (Translation by E. and D. Kreager of La transition démographique: Etapes, formes, implications économiques . Paris: INED.)

Coale, Ansley J., Paul Demeny, and Barbara Vaughan. 1983. Regional model life tables and stable populations . 2d ed. New York and London: Academic Press.

Conrad, Christoph. 1991. "The emergence of modern retirement: Germany in international comparison." Population (Paris: INED), English selection no. 3.

Conrad, Christoph, and H.-J. Kondratowicz, eds. 1993. Toward a cultural history of aging . Berlin: Deutsche Centrum für Altersfrage.

Dublin, Louis I., A. J. Lotka, and M. Spiegelman. [1936] 1949. Length of life: A study of the life table . New York and London: Academic Press.

Dyer, Colin. 1977. Population and society in twentieth-century France . Sevenoaks: Hodder and Stoughton.

Elder, Glen. 1974. Children of the Great Depression . Chicago: University of Chicago Press.

Fildes, Valerie A. 1986. Breasts, bottles and babies . Edinburgh: Edinburgh University Press.

———. 1988. Wet nursing: A history from antiquity to the present . Oxford: Blackwell.

Fischer, David H. 1978. Growing old in America . New York: Oxford University Press.

Fries, James F. 1980. "Aging, natural death and the compression of morbidity." New England Journal of Medicine 303: 130-135.

Fries, James F., and L. M. Crapo. 1981. Vitality and aging: Implications of the rectangular survival curve . San Francisco: W. H. Freeman.

Gordon, Chris. 1988. "Familial support for the elderly in the past: The case of London's working class in the early 1930s." Ageing and Society 8(3): 287-320.

Hareven, Tamara. 1978. Transitions: The family and the life course in historical perspective . New York and London: Academic Press.

Hollingsworth, Thomas H. 1964. The demography of the British peerage . Supplement to Population Studies 18(2).

———. 1977. "Mortality in British peerage families since 1600." Population (Paris: INED), numéro spéciale.

Houston, Rab. 1992. "Mortality in early modern Scotland: The life expectancy of advocates." Continuity and Change 7(1): 47-69.

Howell, Nancy. 1979. Demography of the Dobe !Kung . New York and London: Academic Press.

Janssens, Angélique. 1993. Family and social change: The household as a process in an industrializing community . Cambridge: Cambridge University Press.

Johnson, Paul. 1985. Saving and spending: The working class economy in Britain 1870-1939 . Oxford: Oxford University Press.

Kannisto, Väinö. 1994. Development of oldest old mortality: Evidence from twenty-eight developed countries . Monographs on Population Aging. Odense, Denmark: Odense University Press.

Kertzer, David I., and D. P. Hogan. 1989. Family, political economy, and demographic change: The transformation of life in Casalecchio, Italy, 1861-1921 . Madison: University of Wisconsin Press.

Kertzer, David I., and K. W. Schaie, eds. 1989. Age structuring in comparative perspective . Hillsdale, N.J.: Lawrence Erlbaum.

Keyfitz, Nathan, and W. Flieger. 1990. World population: An analysis of vital data . Chicago: University of Chicago Press.

Kobayashi, Kazumasa, and Zenji Nanjo. 1988. Generational life tables for Japan, based on period life tables covering the years 1891-1986 . Tokyo: Nihon University Press.

Kobrin, Frances E. 1976. "The primary individual and the family: Changes in living arrangements in the United States since 1940." Journal of Marriage and the Family 38(2).

Laslett, Peter, 1966. "The numerical study of English society." In An introduction to English historical demography , ed. E. A. Wrigley, 1-13. London: Weidenfeld.

———. 1972. "The history of the family." Introduction to Household and family in past time , ed. Peter Laslett, with Richard Wall, 1-89. Cambridge: Cambridge University Press.

———. 1976. "The wrong way through the telescope: A note on literary evidence in sociology and historical sociology." British Journal of Sociology 26: 319-342.

———. 1977. "The history of aging and the aged." In his Family life and illicit love in earlier generations , 174-213. Cambridge: Cambridge University Press.

———. [1965] 1983. The world we have lost: English society before and after the coming of industry . New York: Macmillan; London: Methuen.

———. 1984. "The significance of the past in the study of aging." Introduction to the special issue "History and Ageing." Ageing and Society 4(4): 379-389.

———. [1976] 1985. "Societal development and aging." In Handbook of aging and the social sciences , ed. R. H. Binstock and E. Shanas, 87-116. New York: Van Nostrand.

———. 1987a . "The emergence of the Third Age." Ageing and Society 7(2): 133-159.

———. 1987b . "The character of family history, its limitations and the conditions for its proper pursuit." Journal of Family History 12: 263-284.

———. 1988a . "Family, kinship and collectivity as systems of support in preindustrial Europe: A consideration of 'The nuclear hardship hypothesis.'" Continuity and Change 3(2): 153-175.

———. 1988b . "La parenté en chiffres" [Kinship in figures]. Annales ESC: Economies, Societés, Civilizations (Paris): 1-23.

———. 1989. A fresh map of life: The emergence of the Third Age . London: Weidenfeld; U.S. paperback ed., Cambridge: Harvard University Press, 1991; 2d rev. ed., London: Macmillan, 1995.

———. 1990. Review of G. Minois, History of old age. Journal of the Social History of Medicine 3(3): 461.

———. 1991. "Child, family and society." Address to the European Community, Luxembourg, May.

———. 1992. "The family in the industrializing East and the industrial West." Address to the International Conference on Family Formation and Dissolution, Perspectives from East and West, Taipei, Taiwan.

———. 1993. "What is old age?" Paper for the IUSSP Conference on Elderly Populations, Sendai, Japan.

———. 1994. Review article forthcoming in Ageing and Society on the final report and research papers of the Carnegie Enquiry into the Third Age.

———. In press. "The aging of European populations and its social effects." Part 1 of chap. 7 of vol. 3 of L'histoire de la population Européenne , Paris.

Laslett, Peter, and James S. Fishkin, eds. 1992. Justice between age groups and generations . New Haven: Yale University Press.

Lesthaeghe, Ronald. 1991-1992. "The second demographic transition in Western countries: An interpretation." IPD working paper, Free University, Brussels.

Minois, Georges. 1989. History of old age . Oxford: Polity Press, Blackwell.

Mitterauer, Michael. [1986] 1992. A history of youth . Trans. Graeme Dunphy. Oxford: Blackwell.

Ogawa, Naohiro. 1988. "Aging in China—Demographic alternatives." Asia-Pacific Population Journal 3(3): 21-64.

Pelling, Margaret, and Richard M. Smith, eds. 1991. Life, death and the elderly: Historical perspectives . London: Routledge.

Petersen, William. 1975. "A demographer's view of prehistoric demography." Current Anthropology 15(2): 134-156.

Plakans, Andrejs. 1989. "Stepping down in former times: A comparative assessment of 'retirement' in traditional Europe." In Age structuring in comparative perspective , ed. David I. Kertzer and D. W. Schaie, 175-195. Hillsdale, N.J.: Lawrence Erlbaum.

Riddle, S. M. 1992. Doctoral thesis on the elderly worker in England during the later 19th century and earlier 20th century, Cambridge Group.

Riley, James C. 1989. Sickness, recovery and death . London: Macmillan.

Riley, Matilda W. 1987. "On the significance of age in sociology." American Sociological Review 52: 1-14.

Ruggles, Steven. 1994. "The transformation of American family structure." American Historical Review 99(1): 103-128.

Russell, Josiah C. 1948. British medieval population . Albuquerque: University of New Mexico Press.

Schofield, Roger, David Reher, and Alain Bideau, eds. 1991. The decline of mortality in Europe . Oxford: Oxford University Press.

Smith, Daniel S. 1981. "Historical change in the household structure of the elderly in economically developed societies." In Aging, stability and change in the family , ed. R. W. Fogel, E. Hatfield, S. B. Kiesler, and E. Shanas, 91-114. New York and London: Academic Press.

Smith, James E. 1987. "The microsimulation of kin sets and kin counts." In Family demography: Methods and their applications , ed. John T. Bongaarts, T. K. Burch, and Kenneth W. Wachter, 249-260. Oxford: Oxford University Press.

———. 1991. "Aging together, aging alone." In Life span extension: Consequences and open questions , ed. Frederic C. Ludwig, 81-92. New York: Springer.

Szreter, Simon. 1993. "The idea of demographic transition and the study of fertility change." Population and Development Review 19(4): 659-701.

Thane, Pat. 1993. "Old age in English history." In Toward a cultural history of aging , ed. Christoph Conrad and H.-J. Kondratowicz, 17-37. Berlin: Deutsche Centrum für Altersfrage.

Thomas, Keith. 1976. "Age and authority in early modern England." Proceedings of the British Academy 62: 205-248.

Troyansky, David. 1989. Old age in the Old Régime: Image and experience in 18th-century France . Ithaca: Cornell University Press.

U.S. Bureau of the Census. 1937. Vital statistics of the United States of America, Part 1 . Washington, D.C.: Government Printing Office.

Vallin, Jacques. 1991. "Mortality in Europe from 1720 to 1914: Long-term trends and changes in pattern by sex." In The decline of mortality in Europe , ed. Roger Schofield, D. Reher, and A. Bideau, 38-67. Oxford: Oxford University Press.

van de Kaa, D. J. 1987. "Europe's second demographic transition." Population Bulletin 42(1): 3-5.

Vinovskis, Maris A. 1989. "Stepping down in former times: The view from colonial and 19th-century America." In Age structuring in comparative perspective , ed. David I. Kertzer and K. W. Schaie, 215-225. Hillsdale, N.J.: Lawrence Erlbaum.

Wall, Richard. 1989. "The living arrangements of the elderly in Europe in the 1980s." In Becoming and being old , ed. Bill Bytheway, T. Keil, et al., 121-142. London: Sage.

Wrigley, E. A., and R. S. Schofield. [1981] 1989. The population history of England 1541-1871: A reconstruction . Cambridge: Cambridge University Press.

Wu, Cangping, and Peng Du. 1992. "Rethinking population aging in China." Population Science of China 3: 1-5. (In Chinese.)

Young, Michael, and Tom Schuller. 1991. Life after work: The arrival of the ageless society . London: Harper Collins.

Zhao, Zhongwei. In press. "Rapid demographic transition and its influence on kinship networks in East and South Asia, with particular reference to China." Population Studies .