Tuning In, Tuning Out – Study Notes

Context & Provenance

  • Source article: “Tuning In, Tuning Out: The Strange Disappearance of Social Capital in America” (Robert D. Putnam, PS: Political Science & Politics, Vol. 28, No. 4, Dec 1995, pp. 664-683)

  • Occasion: 1995 Ithiel de Sola Pool Distinguished Lecture (APSA)

  • Pool’s legacy: pioneer in political communication, network theory, “small-world” research; predicted decentralising effects of electronic media

Putnam’s Intellectual Starting Point

  • Builds on prior work “Making Democracy Work” (1993) – Italian regional institutions

  • Key construct: social capital

    • “Networks, norms, and trust that enable participants to act together more effectively”

    • Moral neutrality: can be used for good (PTA) or ill (youth gangs)

    • Bridging vs bonding capital (spanning cleavages vs reinforcing in-group)

  • Distinctions

    • Social capital ≠ political participation (relations with each other vs with institutions)

    • Civic engagement: any connection to community life (incl. informal visiting)

    • Social trust vs political trust (trust in people vs authorities)

  • Empirical premise: social trust and civic engagement are strongly, positively correlated across individuals, states, and countries (e.g. r≈0.65r\approx0.65 cross-national; r≈0.71r\approx0.71 across U.S. states)

Empirical Evidence of Decline (≈1960-1995)

  • Multiple independent data sources

    • Membership rolls: PTA, Elks, League of Women Voters, Red Cross, unions, bowling leagues – down 25%−50%25\%-50\%

    • National time-budget surveys (1965, 1975, 1985): informal visiting ↓ ≈25%25\%; club/organization time ↓ ≈50%50\%

    • Political participation that requires collective action: rally attendance ↓ 36%36\%; town-school meetings ↓ 39%39\%; working for a party ↓ 56%56\% (1973-1993)

    • General Social Survey (GSS) 1974-94: group membership ↓ ≈25%25\%; social trust ↓ ≈33%33\% at all education levels

    • Church attendance: Gallup – ↓ 15%15\% in 1960s then flat; NORC – continued fall to ↓ ≈30%30\% by early 1990s

  • Figure 1 (GSS 1974-94, education-controlled): downward trends across virtually every type of voluntary group except nationality, hobby/garden, “other”

  • Conclusion: compared to parents’ generation, Americans are “bowling alone” despite still ranking high internationally

Counter-Trends & Misleading Cases

  • Rise of DC-based interest groups, mailing-list “tertiary” associations (AARP, Sierra Club) – little face-to-face interaction → not true social capital

  • Growth of the nonprofit “third sector” largely administrative/bureaucratic, not associational

The Detective Story – Candidate Explanations & Testing Criteria

  1. Time pressure

  2. Economic hardship or affluence

  3. Residential mobility

  4. Suburbanisation & metropolitan scale

  5. Women entering paid labour

  6. Family disruption (divorce, single parenthood)

  7. Structural economic change (chains, services)

  8. 1960s cultural shifts (Vietnam, Watergate, authority revolt)

  9. Welfare-state growth

  10. Civil-rights revolution & “white flight”

  11. Technological change (TV & electronics)

Evaluation benchmarks

  • Must correlate with trust/engagement

  • Correlation must survive controls (non-spurious)

  • Trend in the factor must match timing of decline

  • Must be exogenous to civic decline (cause, not consequence)

Variables Examined in Depth

Education

  • Strongest individual-level predictor; curvilinear “increasing returns” (Figure 2): last 2 years of college have double the effect of first 2 years of HS; bachelor’s+ have ∼10×\sim10\times impact versus early elementary

  • Macro-paradox: educational attainment rose sharply (HS <12 yrs cohort shrank 40%→18%40\%\to18\%; >12 yrs grew 28%→50%28\%\to50\%). Should have boosted social capital by ≈15%−20%\approx15\%-20\%, yet overall trust & memberships fell equivalently. → Education change cannot explain decline; need to control for education in all further tests

Busyness / Time & Money

  • Time-budget data: average free time ↑ ≈5 hours/week (1965-85) due to less housework & earlier retirements

  • GSS: employed people (esp. longer hours) belong to more groups; heavy workers cut TV, not civic acts (“Nightline, not Kiwanis”)

  • Economic distress: low income correlates with lower engagement, but declines occurred at all income levels; richer cohorts sometimes fell faster

  • Verdict: pressures of time & money not principal culprits

Residential Mobility & Suburbanisation

  • Census: annual mobility actually ↓ (1950s 20 %; 1990s 17 %); long-distance moves ↓ 6.9 %→6.6 %

  • Place size effects modest: big-city residents ≈10 % less trusting; suburbs slightly more trusting; rural areas slightly higher engagement. Trends downward everywhere identically. → Mobility & suburbanisation exonerated

Women’s Labour-Force Participation

  • Absolute declines in joining/trusting larger among women, yet within-period cross-section: working women register slightly higher memberships than homemakers

  • Time-budget 1965-85: org time ↑ for employed women, ↓ for non-employed; informal socialising decline concentrated among non-employed women

  • Possible selection bias, but overall hypothesis not proven

Marriage & Family Structure

  • Family bonds loosening: one-person households ×2 since 1950; currently unmarried adults ↑ 28 %→48 %

  • Married individuals ≈+33%+33\% more trusting, +15!–!25%+15!\text{–}!25\% more memberships; widowed resemble married

  • Divorce rise occurs after main civic drop begins; effect modest → accessory, not main cause

Welfare State Size

  • State-level & OECD cross-national correlations: social capital positively (or not) associated with public-spending share (state R² ≈ .52 with Elazar culture, not welfare); undermines “crowding-out” thesis

Race & Civil Rights

  • Controlling SES, blacks historically hold more memberships (esp. church/ethnic); trust much lower (≈17 % vs 45 % whites)

  • Post-1960s declines paralleled across races; “white segregationists” drop no faster than desegregationists (Figure 3). → Racism not key driver; civil-rights gains not culprit

Generational (Cohort) Dynamics – The Core Finding

  • Age strongly predicts engagement; but no evidence of increasing engagement as cohorts age → life-cycle explanation rejected

  • Figure 5 plots birth year vs engagement (trust, memberships, voting, newspaper). Pattern:

    • High plateau for cohorts born ≈1910!–!1940\approx1910!–!1940 (“long civic generation”)

    • Sharp, steady decline begins with cohorts born early-1930s; continues through 1960s births

    • Example: 1920s cohort vs 1960s cohort (education-controlled): memberships 1.9→1.1; trust 55 %→25 %; voting 75 %→≈42 %; newspaper daily 75 %→≈28 %

  • Period effect in 1980s (minor downward shift across all cohorts) but majority change = generational replacement

  • Demographic arithmetic (Figure 6): civic generation made up 62 % of 1960 electorate; only 31 % of 1992 electorate. Boomers & Gen X rise 24 %→60 % of adults (1974-94). → Future likely further decline without new positive period effect

Reformulated Mystery: What “X-Ray” Hit Post-War Cohorts?

  • Candidate variables exonerated by timing: divorce, working mothers, economic booms/slumps, Cold War, 1960s turmoil.

  • Strong circumstantial & correlational evidence points to television

Television Hypothesis

  • Diffusion: TV households 10 % (1950) → 90 % (1959) – fastest tech adoption on record; viewing hours ↑ 17-20 % (1960s), another 7-8 % (1970s); 1990s TV-household watching >50 % higher than 1950s

  • Multiple sets foster solitary/private viewing; TV absorbs ≈40%\approx40\% of discretionary time (1990s)

  • Media-engagement contrasts (Figures 7-9)

    • Newspaper reading positively correlated with memberships within every education band

    • TV viewing negatively correlated; “pure readers” hold 76 % more memberships and 55 % higher trust than “pure viewers” (education-controlled)

    • Each additional daily hour of TV plausibly explains ∼25%!–!50%\sim25\%!–!50\% of social-capital decline

  • Mechanisms

    1. Time displacement: TV substitutes for social visits, clubs, conversations (Canadian quasi-experiment – Williams 1986)

    2. Mean-world syndrome: heavy viewers overestimate crime → lower trust (Gerbner et al.)

    3. Cognitive/Social passivation: Postman (1985) – entertainment modality discourages interaction; Meyrowitz (1985) – alters social cues

    4. Childhood socialisation: kids (9-14 yrs) watch ≈40 hrs/week; evidence links heavy viewing to aggression, lower achievement, “psychosocial malfunctioning” (Condry 1993)

Ethical, Philosophical & Policy Implications

  • Pool’s “soft technological determinism”: tech trends shape but do not dictate social outcomes; societies can react

  • TV & emerging electronic media may promote individualism, decentralisation, but at cost of “community without contiguity” becoming community without community

  • Future questions (beyond lecture):

    • Can new media (internet) rebuild bridging capital or exacerbate fragmentation?

    • Deliberate interventions: media literacy, civic education, public-service programming, spaces for face-to-face interaction

    • Importance of recognising generational lag: rebuilding social capital today requires engaging younger cohorts

Key Numerical / Statistical References

  • Declines: memberships ↓ 25–50%25\text{–}50\%; trust ↓ 33%33\%; church attendance ↓ 15–30%15\text{–}30\%

  • Correlations: social trust vs memberships r<em>countries≈0.65r<em>{countries}\approx0.65; r</em>states≈0.71r</em>{states}\approx0.71

  • Education curves: graduate vs <HS difference ≈+1+1 group & +30%+30\% trust

  • TV vs newspapers: “pure reader” holds 76 % more memberships

  • Mobility: annual mover rate 20 % (1950s) → 17 % (1990s)

Connections to Other Literature & Real-World Context

  • Verba, Schlozman & Brady (1995) – education & participation; Rosenstone & Hansen (1993) – mobilization trends; Miller & Shanks (1995) generational turnout echo Putnam’s patterns

  • Contemporary relevance: ongoing digital revolution (streaming, social media) may replicate or amplify TV effects unless accompanied by intentional community-building strategies

  • Metaphor extended: “the ozone layer of civil society” – invisible erosion detected only decades later; urgent need for “social capital CFC ban” equivalents

Study Reminders & Potential Exam Prompts

  • Define social capital, distinguish from human & political capital; explain bridging vs bonding relevance

  • Reproduce Putnam’s argument structure: empirical decline → testing of hypotheses → generational finding → TV indictment

  • Be able to cite at least three independent data sources for decline and specific percentage figures

  • Explain criteria for evaluating causal candidates & apply to one variable (e.g., mobility)

  • Debate TV hypothesis: provide supporting mechanisms & possible counter-arguments (e.g., selective exposure, emergence of civic television content)

  • Discuss policy avenues for revitalising civic engagement in light of technological change