There is, of course, another "small screen" that is the subject of national discussion on social capital. The personal computer is the electronic doorway to a realm whose very name, internet, suggests a constituent social structure. The immediate question is what this new form of media technology means for other, more traditional forms of social structure. Or, as Putnam puts it,

Will the Internet in practice turn out to be a niftier telephone or a niftier television? In other words, will the Internet become predominantly a means of active, social communication or a means of private, passive entertainment? Will computer-mediated communication "crowd out" face-to-face ties? It is, in this domain especially, much too early to know.

Perhaps not entirely. Putnam's own SCCBS2000 dataset permits a parallel analysis to the one just performed for television. Figure 3 arrays responses to the question, "How many hours do you spend on the internet in a typical week?" in increasing order from none to more than five hours.

The contrast with television could hardly be more striking. WWW-time is positively associated with Voluntary Association, and the steep upward slopes indicate a lot more association. Intriguingly, Americans without high school degrees gain the most, almost doubling their group memberships across these extremes of internet time; even college graduates, who already tower above less-educated respondents in terms of overall membership, gain an average of half a group from left to right.

Informal Socializing is addressed in Figure 4, which tells a watered-down version of the same story. More internet time means more rather than less face-to-face time, and especially so for the least-educated Americans.

Of course, Putnam is suggesting that it is too early to see the long-term effects of the internet on social capital, but in advance of future analyses, the present data paint a pro-social capital picture.

The reference to future analyses elicits further thoughts about the limits of those presented here. In broad outline, the two forms of media technology spotlighted show contradictory effects. Although it is tempting to argue that positive internet effects will cancel out negative TV effects on social capital over the next decade, this would be too facile by far. For one thing, the causality question looms every bit as much above Figures 3 and 4 as it does over Figures 1 and 2. Does media technology drive out social capital, or do social capital choices drive the use of media?

Short of a million-dollar study to settle the media causality question, the way to seek answers is by delving into media content. Consider: all the figures in this analysis simply count hours onscreen. Clearly, what is on those screens matters. Watching TV news has been strongly linked to social involvement; watching The Simpsons has not. Organizing a neighborhood association online is a social capital creation; visiting porn websites is not.

More attention needs to be paid to which "ghosts" are emanating from which "machines," and what difference they make.

To return to the opening argument that sensationalized images and issues have convinced Americans that social capital has been lost, consider now a sociological surprise. According to Newman, author of Rampage,

Outside of Small-town USA, the decline of social capital and the corresponding longing for the resurrection of community has been a powerful narrative... But this decline and fall story, powerful as it is, cannot account for the tragedies in Heath and Westside. Putnam would find much to admire in both places, for they epitomize the kind of America he is searching for. Parents of Heath High School students are very involved in their children's lives and activities. Nobody locks their doors in the Westside area, because the neighbors have known each other for generations... Heath and Westside are the exceptions that prove Putnam's rule, or so they thought.

In fact, Newman's research revealed that "the very density of community structure" in small towns can contribute to school shootings of the Rampage variety. The point is not that high levels of social capital are actually a bad thing. The point is that the forms of social capital are intricately interconnected. Unraveling this complexity requires careful attention to the whole host of family, work, PTA, and neighborhood phantoms hovering over our heads whenever the TV is on or we go online.

Brian Jones is a professor of sociology at Villanova University.

Additional Reading