Watch CBS News

The Lonely States Of America

This commentary was written by's Dick Meyer.

The American Sociological Review may have just published the social health equivalent of the 1964 Surgeon General's report that declared smoking causes cancer. The unpleasant but long suspected discovery in this case is that social isolation in America has grown dramatically in the past 20 years.

Some things are uncomfortable to know. We don't like knowing the earth is getting hotter; some people choose not to believe it. In 1964, about half of all adults smoked and they did not like knowing the habit caused cancer; some people chose not to believe it and some people still don't. The scientific evidence about smoking and cancer existed long before Jan. 11, 1964, but when the famous report was issued that day, people started believing it.

I expect something quieter and more eggheaded but quite similar will happen with an academic paper with the vanilla title, "Social Isolation in America: Changes in Core Discussion Networks over Two Decades." The authors, Miller McPherson, Lynn Smith-Lovin and Matthew Brashears, sociologists at Duke and the University of Arizona, have no such wild pretensions, but I think they've documented an enormous, stunning social change so clearly that it will alter the way we look at social and political life. It should.

And it should scare you.

The authors set out to empirically describe how socially connected Americans are by asking them questions like, "Who are the peoplewith whom you discussed matters important to you?" They did this as part of the General Social Survey, the Rolls Royce of face-to-face social surveys that has been conducted almost every year since 1972. In 2004, they precisely replicated questions about social networks that had not been asked since 1985.

Because the findings are so stark and clear, and come with no linguistic and philosophic adornment, I'll let the numbers speak for themselves in blunt bullet points:

  • From 1985 to 2004, "the number of people saying there is no one with whom they discuss important matters nearly tripled." Now, 24.6 percent report they have no confidants, family or non-family — that's one in four Americans. Another 19.6 percent say they have just one confidant. That means 43 percent of Americans have either no confidants or just one, a slice that has doubled since 1985.
  • More than half, 53.4 percent, do not have any confidants who aren't family. In 1985, 80 percent had at least one confidant who was not family; now only 57.2 percent do.
  • The average size of Americans' social networks decreased by a third between 1985 and 2004, from 2.94 to 2.08; basically this means the loss of one confidant.

  • The kinds of relationships that decreased the most in providing important contacts were neighbors and co-members of groups or voluntary associations (as opposed to spouse, sibling, parent, co-worker, etc.)
  • Women have more family in their networks than men, as they did in 1985. But then they had fewer non-kin close relationships than men did. Now women have about the same number of confidants outside family as do men. Unfortunately, that isn't because women have made more contacts outside kin, but because men have fewer.
  • More education correlates with having larger social networks. Non-whites and the elderly are populations with smaller networks.

    Don't let yourself be numbed by the numbers because they tell a dramatic story even though there are no victims, tears or sound bites.

    The bottom line: "The number of people who have someone to talk to about matters that are important to them has declined dramatically… we have gone from a quarter of the American population being isolated … to almost half of the populations falling into that category."

    Stop and think about that for a second. Almost half the people around you have at most one person they feel they can talk to about what is most important to them. Seems like a pretty lousy social system we've got going here, doesn't it?

    Does this cold statistical portrait comport with your own experience of the world and the people you are acquainted with? My first gut answer was "no." But when I thought about it harder, the answer changed. There are people who I think are frighteningly isolated even in my company, my small neighborhood, my extended family and the community based around my kids' school — and these are all social networks by definition. The most isolated, of course, I wouldn't even come across much.

    The authors were even more surprised at the findings and looked for every possible reason why the results could be wrong. They explored whether people have different notions of the word "discuss" or "important" than they did 20 years ago. They looked for technical problems in the survey. But the news stayed bad.

    So what explains this seismic social thud?

    The paper eliminates a couple suspects. It is not caused by great geographic mobility — the corporate nomad syndrome. It is not caused by employment rates. It does not correlate with increased television watching. Most importantly, it is not caused by the demographic facts that the population is aging and more ethnically diverse; if it were, those trends would have been countered by the increased educational levels since 1985, since education leads to larger networks.

    That means the answers will be deep and complicated.

    Though they are mostly into documenting not explaining, the authors do put out a couple of hypotheses. The main culprits are work time and commutes. Both have increased since 1985 and both take time away from families, friends and voluntary participation. As women entered the workforce in bulk, the total number of hours family members spent working outside the home went way up. As people fled the cities, suburbs and exurbs boomed and so did commute times.

    This especially affects "middle-aged, better-educated, higher-income families." As the paper points out, these are exactly the people who build neighborhoods and volunteer groups and those are the social structures that have most atrophied in the past 20 years.

    The more speculative hypothesis is that perhaps new communications technologies have led to people forming wider, but weaker social ties that are less dependent on geography. E-mail and cheap phone calling have made it easier to stay in frequent, sometimes constant touch with lots of people, no matter where they are.

    These weak ties are different than the confidant ties that this study measures, but the authors are open to the idea that a network of weaker ties can provide equally meaningful, but different, social support (a view supported by a quantitative study done by two university of Toronto sociologists for the Pew Internet & American Life Project). But they do point out the obvious: "some services and emotional support" do depend on proximity.

    Certainly, it's hard to escape complaints about the busy-ness and time-stress of life these days; it's an obvious, bad problem. For most people I know, it is exacerbated by the technology that is meant to make it easier for us to communicate and stay connected. Instead of feeling in touch, many feel on a leash. Portable, gadget driven communication doesn't count as soul-feeding bonding for many people I know, but is rather a cruel mockery.

    I do suspect that this study overlooks one simple contributing factor, the decline of real geographic communities — places where people grow up where their parents grew up, where non-nuclear relatives live near by, where friendships and acquaintances go across generations.

    Explaining social isolation will be controversial, but not as difficult as repairing it.

    In primitive and survival-dependent societies, social isolation was basically impossible. But modern societies have never been without chronic existential worries about isolation and loneliness; it is one of the defining marks of modernity. Literary and philosophic examinations of American souls and social life began with the very first American books, like Ben Franklin's autobiography.

    Looking at these issues empirically is a different matter. Social statistics aren't the stuff of teen angst, novels and high culture. But the story they tell is just as disturbing and just as hard for society to accept. Recent social science research, for example, about the decline of civic engagement and community participation has been exceedingly controversial and contested. There are even larger objections to the idea that "social science" can ever get a handle on these kinds of issues in a way that is at all scientific.

    It is hard to believe and accept that we live in a society where one person in four feels they don't have someone to confide in. It's depressing and even somewhat terrifying. We can, of course, ignore it all and choose to keep on smoking.

    Dick Meyer is the editorial director of

    E-mail questions, comments, complaints, arguments and ideas to
    Against the Grain. We will publish some of the interesting (and civil) ones, sometimes in edited form.

    By Dick Meyer

  • View CBS News In
    CBS News App Open
    Chrome Safari Continue
    Be the first to know
    Get browser notifications for breaking news, live events, and exclusive reporting.