There has recently been a spate of commentary holding forth the prospect that the 2008 election may prove to be a "realigning" election. One should treat such predictions with great caution.

There are a couple of approaches that analysts have taken to suggest the realignment possibility. Some have pointed to the fact that all three plausible candidates—but especially Obama and McCain—have the potential to shake up the electoral map. McCain’s appeal to independents and moderate Democrats might put into play a number of states that Democrats won in 2008, including Oregon, Wisconsin, Pennsylvania, Michigan, and New Hampshire, perhaps even New Jersey and (McCain strategists hope) California. Obama, based on polls and his primary record, could put into play some smaller, traditionally-Republican states in the Midwest and some Southern states with large black populations.

Others, looking at generic polls on party identification, issues, presidential party preference, and the current landscape of congressional races, have pointed more specifically to the prospect of a Democratic realignment ala 1932. These analysts emphasize the potential of a long-term Democratic breakout from the recent partisan stalemate.

In order to assess these claims, it is important to think carefully about what a realignment is: a long-term, enduring shift in party loyalties in the electorate that translates into a significantly altered balance in government, and consequently a long-term shift in national policy.

There has been great debate in recent years, led by Yale political scientist David Mayhew, about whether the concept of realignment is even a useful one. At the least, Mayhew has done a service by exploding a rigid concept of realignment that held only a handful of American elections to have been truly important. It is clear that there have been a larger number of really crucial elections (both presidential and midterm) than the original version of realignment conceded.

If one strips away some of the rigidities of the original theory, there still remains a useful concept that would include the following provisions (propounded by James Sundquist and others):

• Realignments are most likely to spring out of a sense of crisis that leaves voters open to reconsidering their long-term partisan loyalties.
• The issues that led to the previous major realignment (say, the Civil War or the memories of Herbert Hoover) have to have decayed—that is, their hold on voters has to have diminished—enough that newer issues can supersede them.
• The new issues that come to the fore must be cross-cutting, that is they must divide at least one if not both of the major parties.
• The parties must take distinct and opposing positions on the those new issues, otherwise voters will not know how to reaggregate themselves on different lines.