For those who have been following the debate, there's not too much of pressing interest in the Nation's roundtable on the future of labor unions. Except, I think, for the very end, when the moderator asks AFL-CIO president John Sweeney, SEIU president Andy Stern, and UNITE HERE hospitality president John Wilhelm if they think that a split in the labor movement could be good for organizing:

Sweeney: You know, this is not the 1930s and '40s, when US industry was on the rise and we were shifting from an agricultural to an industrial economy. We are shifting to service, working families have no basic healthcare, the working poor is growing and people are barely getting by. Right-wingers in Congress are actively attacking worker protections at every turn. It should tell you something that those right-wingers are salivating at the prospect of a split. Look at the Republican websites--they are gleeful, because the truth is, we have maintained power out of proportion to our numbers because we have been organized. A split will help no one…

Stern: The labor movement is incredibly divided right now. The only way it is united is at a table in Washington, DC, or because it uses the same initials after its name. But when it comes to dealing with companies like United Airlines or national strategies about how to organize Wal-Mart or healthcare workers, there is no unity. The airline industry, the most heavily unionized industry in the country, is Exhibit A, B, C and D. If we don't believe our lack of ability to coordinate and cooperate within companies and across them is a factor, we are crazy. I don't think a split itself will create a whole new wellspring of growth and hope. I don't expect immediate results…

Wilhelm: There is no question that the historical context is radically different, but I would point out that the CIO did not begin as a split but as a group of unions who wanted to try doing some things differently. The AFL expelled them. The question is, Will the federation take advantage of the greatest opportunity since the Great Depression to respond to what the country needs the most, or won't we? I don't think there ought to be a split. … But the definition of insanity is doing the same thing over and over again and expecting a different result. …

Hm, interesting thoughts all around. One other thing to add, and this gets touched on very slightly earlier in the roundtable. My grasp of union history is somewhat fuzzy, but it seems, as Richard Freeman has demonstrated elsewhere, that labor density has historically grown in "spurts," during periods and under conditions that were very difficult to see and predict. So in World War I, for instance, developed countries needed the full cooperation of labor in order to mobilize and fight their splendid little war, and due to various resulting compromises, union density grew considerably. Likewise, union density grew during the Great Depression for obvious reasons, and during World War II because the government yet again needed cooperation. In the EU, union density grew considerably during the oil crisis inflation (but not, note, in the private sector in the United States during the same time). Likewise, experts have tolled the bell for various unions before—George Meany thought public sector unions were doomed in the 1950s—only to be horribly wrong.

So it's very often hard to foresee what's going to happen, and to some extent these productive spurts have to do with changes in labor law and legislation, but to some extent they also have to do with contingent historical events and the emergence of new unions and new modes of operating. One of the factors that were hard to foresee in the 1930s and 1940s was precisely the rise of the industrial union, which really fueled the surge in unionism during the Great Depression. (In conjunction with New Deal legislation, but not wholly dependent on it.) Likewise, during World War I, some of the main driving factors were the rise of war labor boards and mandatory arbitration. So it's very easy to rely too much on lessons from the past, which makes the question posed above something of a misleading one.