COHORT PERSPECTIVEShistory, usually age in different ways. For example,
the enjoyment of ‘‘midlife’’ experienced at around
age 50 by cohort members studied in the 1990s
may not be felt by some future cohort until age 85.
Intercohort Perspectives. Broader than the
intracohort focus, is a focus on the lives of mem-
bers of two or more successive cohorts who are
growing older under differing historical or socio-
cultural conditions. Studies of intercohort differ-
ences in the late-twentieth century demonstrated
for other sciences what sociologists had learned
early: the central principle that the process of
aging is not immutable or fixed for all time, but
varies across and within cohorts as society changes
(Riley 1978). Such studies have shown that mem-
bers of cohorts already old differ markedly from
those in cohorts not yet old in such respects as
standard of living, education, work history, age of
menarche, experience with acute vs. chronic dis-
eases, and perhaps most importantly the number
of years they can expect to live. These cohort
differences cannot be explained by evolutionary
changes in the human genome, which remains
much the same from cohort to cohort; instead,
they result from a relatively unchanging genetic
background combined with a continually chang-
ing society (Riley and Abeles 1990, p.iii). Thus the
finding of cohort differences has pointed to possi-
ble linkages of lives with particular social or cultur-
al changes over historical time, or with particular
‘‘period’’ events such as epidemics, wars, or de-
pressions (e.g., Elder and Rockwell 1979). These
linkages are useful in postulating explanations for
changes—or absence of changes—in the process
of aging.
Studies of cohort differences focus on aging
processes at either the individual or the collective
level. At the individual level, cohort membership is
treated as a contextual characteristic of the individu-
al, and then analyzed together with education,
religion, and other personal characteristics to in-
vestigate how history and other factors affect the
heterogeneous ways individuals grow older (e.g.,
Messeri 1988; but see Riley 1998). At the collective
level, the lives of members are aggregated within
each cohort to examine alterations in average
patterns of aging. Striking advances have recently
been made in the data banks available for intercohort
comparisons. Archived data from many large-scale
studies now cover long periods of history, multiple
societies, and multidisciplinary aspects of the life
course; and repeated longitudinal studies are be-
ing launched, such as the National Institute on
Aging’s Health and Retirement Study (HRS) and
Asset and Health Dynamics Among the Oldest
Old (AHEAD) (cf. Campbell 1994; O’Rand and
Campbell 1999).Cohort perspectives are useful, not only in
explaining past changes in aging processes, but
also in improving forecasts of future changes.
Unlike the more usual straight projections of cross-
sectional information, forecasts based on cohorts
can be informed by established facts about the past
lives of people in each of the cohorts already alive
(e.g., Manton 1989). Thus, if cohorts of teenagers
today are on the average less healthy, less cared
for, or less prepared for life than their parents
were at the same age (National Association of State
Boards of Education 1990), the lives of both off-
spring and parents will predictably also differ in
the future when both have grown older.COMPOSITIONAL PERSPECTIVES.Complementing sociological work on cohort dif-
ferences (or similarities) in the aging process are
studies of how cohort succession contributes to
formation and change in the age composition of
the population. Thus in Figure 1, the perpendicu-
lar lines indicate how cohorts of people fit togeth-
er at given historical periods to form the cross-
sectional age strata of society; and how, as society
changes, new cohorts of people are continually
aging and entering these strata, replacing the pre-
vious incumbents.Single Period of Time. In Figure 1, as indicat-
ed above, a single vertical line at a given period (as
in 1990) is a cross-sectional slice through all the
coexisting cohorts, each with its unique size, com-
position, earlier life experiences, and historical
background. This familiar cross-sectional view of
all the age strata is often denigrated because its
misinterpretation is the source of the life-course
fallacy—that is, the erroneous assumption that
cross-sectional age differences refer directly to the
process of aging, hence disregarding the cohort
differences that may also be implicated (Riley
1973). That people who are differentially located
in the age composition of society differ not only in
age but also in cohort membership was drama-
tized early by Mannheim ([1928] 1952) and Ryder