Bowling Alone

Yesterday I finished reading Robert Putnam's book Bowling Alone. I could hardly put the book down. I'm a data junkie, and it's hard to find a page in this book that isn't packed with data. Moreover, Putnam smoothly presents the data; it goes down like a milkshake.

I learned some surprising things from Putnam's examination of the decline of American social capital since the its peak in the first half of the 1960s. (To be clear, social capital includes both close links between family and friends as well as those links that stretch out further, such as between neighbors, coworkers, coreligionists, fellow citizens, etc.)

First, timing to a large degree exculpates the divorce explosion from being a cause of the decline in social capital; it was about ten years too late. Moreover, the more stable families took part in the decline. Also, as someone who has lived in nine states in twenty-two years, I couldn't help but suspect mobility had something to do it. But Americans were slightly more mobile in the 1950s than they are now.

Second, "the pressures of time and money," including the massive increase of female participation in the workforce cannot account for more than a fraction of the decline of social capital; Putnam estimates roughly ten percent, though I'd emphasize the roughness of the estimate more. Putnam also estimates suburban sprawl accounts for about ten percent.

Third and unsurprisingly, the two biggest factors were television and generational differences. Unfortunately, the latter factor is a bit question-begging, but we can at least say that post-WWII generations didn't go through the unique bonding experience of the common sacrifice and triumph that was America's participation in that great clash.

Putnam is excellent in dealing with the what and how. He seems to have found statistics on just about everything, and he handles them carefully: he never confuses correlation and causality, and when he doesn't have the stats to answers a question, he's honest about his ignorance. On normative questions, his writing is more speculative and sometimes superficial, especially when it comes to the question of what we should do in the here and now. I don't blame Putnam for this; the question of how social capital should be increased, or even if it should be increased, is just too broad. One should break the problem up into managable pieces, looking for low-hanging fruit: instances where social capital can be increased at low costs, where costs are measured in terms of economics as well as things like liberty.

I'm also more wary than Putnam of using the government to help matters. For example, TV and suburban sprawl wouldn't have caused much decline in social capital if so many people didn't like them. I don't think a civic quest to increase social capital, however nobly intended, justifies things like "smart growth," which Putnam apparently approves of. Far more often than not, the prudent and decent course to better ourselves and our neighbors by moral suasion, not by legislation.

That said, there are surely non-coercive ways to reform government so as to facilitate community-building. One idea mentioned by Putnam that I heartily endorse is decentralizing government so as to make our democracy more participatory. Here's the idea presented in rather radical form:

The ideal of participatory democracy has deep roots in American political philosophy. With our experiment in democracy still in its infancy, Thomas Jefferson proposed amending the Constitution to facilitate grassroots democracy. In an 1816 letter he suggested that "counties be divided into wards of such size that every citizen can attend, when called on, and act in person." The ward governments would have been charged with everything from running schools to caring for the poor to operating police and military forces to maintaining public roads. Jefferson believed that "making every citizen an acting member of the government, and in the offices nearest and most interesting to him, will attach him by his strongest feelings to the independence of his country, and its republican constitution."

I definitely wouldn't go that far, but it's certainly an idea one for which one would like some experimental data. Perhaps some county might give it a try?