BOWLING ALONE
Page 30
Something about age is clearly a key to our puzzle. However, this clue is fundamentally ambiguous, for it might strengthen either of two quite different interpretations. Do people of different ages behave differently because they are momentarily at different points in a common life cycle or because they enduringly belong to different generations? Age is an exceedingly valuable clue, but it is not so nearly infallible as fingerprints or DNA, so we need to explore the evidence with care.
AT THE END of the twentieth century American males in their sixties and seventies had much worse eyesight than their grandsons in their twenties and thirties, and the older men were also much more likely to have served in the military than were their grandsons. However, these two age-related patterns have quite different origins. The eyesight effect is due entirely to the life cycle: as we age, virtually everyone’s vision deteriorates. On the other hand, the different rates of military service are due to generational differences. About 80 percent of men born in the 1920s served in the military, as compared with about 10 percent of men born in the 1960s, a difference attributable entirely to differences in world affairs when each group reached eighteen. Eyesight reflects the life cycle, whereas military service reflects generations. When the grandsons reach their grandfathers’ age, their vision too will blur, but they will never share their grandfathers’ military service.
With evidence from a single point in time, we cannot distinguish between life cycle and generational effects, but if we follow a given cohort over the years, we can more readily distinguish the two. And the two effects have dramatically different social consequences. Life cycle effects mean that individuals change, but society as a whole does not. Generational effects mean that society changes, even though individuals do not. There is little reason to believe that average eyesight in America will deteriorate in the early decades of the twenty-first century, but it is virtually certain that veterans will become rarer.
So before we can tell whether the ubiquitous age-related differences in civic engagement are truly generational, and thus producing social change, we need to determine whether these differences are attributable to the normal life cycle. With comparable evidence across several decades, we can follow each cohort as its members move through various stages of life. If successive cohorts generally retrace the same ups and downs as they age, we can be reasonably sure that we are observing a life cycle pattern. If not, it is more likely that age-related differences are generational in origin.3
Life cycle patterns in social behavior are typically caused by one of three factors—the demands of family (that is, marriage and parenting), the slackening of energy (declining from adolescence to old age), and the shape of careers (that is, entering and leaving the labor force). Different forms of civic involvement peak at different stages of the life cycle. Sports clubs attract the energies of youth. Time with friends peaks in one’s early twenties, declines with marriage and kids, and rebounds in one’s sixties with retirement and widowhood. Child-related activities, like parent-teacher meetings, picnicking, and athletic events, are tied to the prime parenting years (twenties and thirties). Membership in civic organizations and professional societies crests among men and women in their forties and fifties. Donation of blood rises to a peak in one’s thirties and falls off sharply after fifty, whereas donation of money rises later in life. Church involvement spurts during one’s twenties (with the advent of marriage and children), plateaus, and then resumes rising gently among seniors. Volunteering used to have a single peak in one’s thirties, reflecting PTA bake sales and Little League coaching, but in recent decades (as we saw in chapter 7) a second, postretirement spurt in volunteering has appeared. Civic engagement in general typically traces a pattern like that in figure 70, rising from early adulthood toward a plateau in middle age, from which it gradually declines. This humpback pattern represents the natural arc of life’s engagements.4
If this normal cycle of life’s events entirely explained age-related differences in civic engagement, older Americans should be much less involved
Figure 70: Membership in Associations Rises and Falls with Age
civically than middle-aged people. Classic sociological studies in the 1950s and 1960s found exactly that. By the 1990s, however, middle-aged men and women were, unexpectedly, not much more engaged than their elders.
Moreover, as baby boomers passed through the normal civic life cycle, like a pig through a python, America should have experienced waves of increasing civic involvement, as the boomers ascended the normal life cycle of rising community involvement. We should have seen a boom in PTA membership in the 1970s and 1980s, along with rapidly rising church membership, and a profusion of civic involvement in the 1990s. (By this same logic, we should look forward to a boom in volunteering and philanthropy as the boomers begin to retire in the 2010s.) So far, however, none of those past waves of civic engagement has materialized—quite the contrary, as we have seen throughout this book: the boomers and their successors have not trod the same ascending civic path traced by previous generations. This civic “dog that didn’t bark” is an important clue to America’s civic decline in the past several decades, for the expected life cycle upswings must have been swamped by unexpected generational downswings. Political interest and participation, church attendance, community projects, charitable giving, organizational involve-ment—as we have seen, all these forms of civic involvement and more besides have declined largely, if not exclusively, because of the inexorable replacement of a highly civic generation by others that are much less so.5
We can see this fact most clearly by examining the civic engagement of successive generations as they pass by fixed milestones in the life cycle. Table 3 presents patterns of change among four different age groups over the last quarter of the twentieth century.6 This table, though packed with numbers, is worth poring over, for it portrays a striking picture of social change in America over the last quarter century. The table, in effect, holds life cycle differences constant in order to focus on generational differences. The first row in the table, for example, shows the extent of newspaper readership in four different age brackets at the beginning of the 1970s. In that era slightly less than half of all young adults (49 percent) read a newspaper daily, as compared with roughly three-quarters of each of the other three age groups in the population. Among people over sixty, for example, 76 percent were newspaper readers. The second row shows the level of newspaper readership in each of these same age groups in the mid-1990s. In that more recent period readership among young adults had fallen to 21 percent, less than half the figure for young adults two decades earlier, for a relative decline of 57 percent. At the other end of the age hierarchy, newspaper readership had also slipped a bit, but only by 10 percent. The third row in the table shows that the rate of decline in newspaper readership was much faster among younger cohorts than among older cohorts. People over sixty in the 1990s (that is, people born in the 1930s or earlier) were almost as likely to read newspapers as people that age had been in the 1970s. In short, the falloff in newspaper readership in America from the 1970s to the 1990s was heavily concentrated in the younger generations—the younger the cohort, the more rapid the decline over these two decades.
Now scanning down the table, one can see that this same pattern applies to practically every form of civic engagement. In virtually every case, disengagement was concentrated among the younger cohorts and is slightest among men and women born and raised before World War II. Among people over sixty, reading newspapers, signing petitions, and writing letters to the editor and to Congress were almost as common in the 1990s as in the 1970s, but among the youngest category these activities were half as common as they had been. Among the oldest cohort church attendance was essentially unchanged between 1973–74 and 1997–98, whereas among people under thirty it fell by nearly one-third. Even in cases such as union membership and work for a political party, in which every age group shows a falloff in participation, the rate of decline was significantly faster amo
ng younger cohorts. As the last three rows in the table show, participation in any of the twelve civic activities measured in the Roper Social and Political Trends surveys fell by 11 percent among those over sixty, by 22 percent among those aged forty-five to fifty-nine, by 32 percent among those aged thirty to forty-four, and by 44 percent among those under thirty. Reading across the row for the 1970s, we see the familiar life cycle humpback, with the oldest cohort significantly less engaged than the younger ones. By the 1990s, however, the life cycle hump was much flatter, as the younger cohorts were now only slightly more engaged than their elders. The more recent the cohort, the more dramatic its disengagement from community life. This is a strong clue that the overall decline in civic engagement in America over the last several decades had its roots in generational differences.7
The key question to ask about generational differences is not how old are people now, but when were they young.8 Figure 71 addresses this question, displaying various measures of civic engagement according to the respondents’ year of birth.9 In effect, figure 71 lines up Americans from left to right according to their date of birth, beginning with those born in the first third of the twentieth century and continuing across to the generation of their grandchildren, born in the last third of the century. To each successive birth cohort, we pose a series of tests of social capital and civic engagement: Did you vote in the last presidential election? How often do you read a newspaper? What voluntary associations do you belong to, if any? How often do you attend church? How many times last year did you attend a club meeting? Are you interested in politics? Did you work on any community projects last year? Do you think most people can be trusted or that you can’t be too careful?
As we begin moving along this line from the oldest generation toward younger generations—from those born around the turn of the last century to those born during the Roaring Twenties—we at first find high and relatively stable levels of civic engagement and social capital. Then rather abruptly, how
Table 3: All Forms of Civic Disengagement Are Concentrated in Younger Cohorts
Age Brackets
18-29
30–44
45-59
60+
Read newspaper daily 1972-75
49%
72%
78%
76%
1996-98
21%
34%
53%
69%
Relative change
—577c
—52%
—317c
—107c
Attend church weekly 1973-74
36%
43%
47%
48%
1997-98
25%
32%
37%
47%
Relative change
—307c
—25%
—22%
—3%
Signed petition 1973-74
42%
42%
34%
22%
1993-94
23%
30%
31%
22%
Relative change
—46%
—277c
—87c
07c
Union member 1973-74
15%
18%
19%
10%
1993-94
5%
10%
13%
6%
Relative change
—647c
—417c
—32%
—42%
Attended public meeting 1973-74
19%
34%
23%
10%
1993-94
8%
17%
15%
8%
Relative change
—57%
—507c
—34%
—217c
Wrote congressman 1973-74
13%
19%
19%
14%
1993-94
7%
12%
14%
12%
Relative change
—47%
—34%
—277c
—157c
Officer or committee member of local organization 1973-74
13%
21%
17%
10%
1993-94
6%
10%
10%
8%
Relative change
—53%
—53%
—417c
—24%
Wrote letter to newspaper 1973-74
6%
6%
5%
4%
1993-94
3%
5%
5%
4%
Relative change
—497c
—187c
—9%
—4%
Worked for political party 1973-74
5%
7%
7%
5%
1993-94
2%
3%
4%
3%
Relative change
—647c
—597c
—497c
—36%
Ran for or held public office 1973-74
0.6%
1.5%
0.9%
0.6%
1993-94
0.3%
0.8%
0.8%
0.5%
Relative change
—43%
—497c
—87c
—22%
Took part in anyof twelve different forms of civic life* 1973-74
56%
61%
54%
37%
1993-94
31%
42%
42%
33%
Relative change
—44%
—317c
—22%
—117c
* Wrote Congress, wrote letter to editor, wrote magazine article, gave speech, attended rally, attended public meeting, worked for political party, served as officer or as committee member of local organization, signed petition, ran for office, and/or belonged to good-government organization.
* * *
Figure 71: Generational Trends in Civic Engagement (Education Held Constant)
ever, starting with men and women born sometime in the 1930s, we encounter signs of reduced community involvement. These preboomers are still relatively civic, in absolute terms, but they are somewhat less so than their older brothers and sisters. As we continue along the line to the boomers and then to the X’ers, this downward trend in joining, trusting, voting, newspaper reading, church attending, volunteering, and being interested in politics continues almost uninterruptedly for nearly forty years. (Attendance rates at churches and clubs decline across all the cohorts, rather than showing the distinctive break in the 1930s.) Figure 71, in sum, shows that each generation that has reached adulthood since the 1950s has been less engaged in community affairs than its immediate predecessor.
By any standard, these intergenerational differences are extraordinary. Controlling for educational disparities, members of the generation born in the 1920s belong to almost twice as many civic associations as do members of the generation of their grandchildren born in the late 1960s (roughly 1.9 memberships per capita, compared with roughly 1.1 memberships per capita). The grandparents are more than twice as likely to trust other people as the grandchildren are (50 percent vs. 20 percent). They vote at nearly double the rate of the most recent cohorts (80–85 percent vs. 45–50 percent). The grandparents are nearly twice as interested in politics (55 percent vs. 30–35 percent) and nearly twice as likely to attend church regularly (45 percent vs. 25 percent). They are twice as likely to work on a commu
nity project (35 percent did so in the previous year, compared with 15–20 percent of the younger generation). The grandparents are the last of the rabid newshounds: they are almost three times as likely to read a daily newspaper (75 percent vs. 25 percent) as the youngest cohort, and they provide the lion’s share of the audience for television news.10 And well-established life cycle patterns give little reason to expect that the youngest generation ever will come to match their grandparents’ levels of civic engagement.
Deciphered with this key, figure 71 depicts a long civic generation, born roughly between 1910 and 1940, a broad group of people substantially more engaged in community affairs and more trusting than those younger than they.11 The core of this civic generation is the cohort born in 1925–1930, who attended grade school during the Great Depression, spent World War II in high school (or on the battlefield), first voted in 1948 or 1952, set up housekeeping in the 1950s, and saw their first television when they were in the late twenties. Since national polling began, this cohort has been exceptionally civic—voting more, joining more, reading more, trusting more, giving more.
What is more, this group has played this leading civic role despite the fact that it received substantially less formal education than its children and grandchildren. Only one-quarter of Americans born between 1900 and 1940 went beyond high school, as compared with more than half of Americans born after that date. As far as formal education is concerned, the members of the long civic generation were “self-made” citizens. As the distinguished sociologist Charles Tilly (born in 1928) has said on behalf of his generation, “We are the last suckers.”12