My undergraduate senior thesis was on realigning elections.  It's been a long, long time since it was written, but my research from that admittedly simpler time in American politics has always made me look closely when election results seem to show that something is changing.

The results of yesterday's special election in Louisiana, where the Democrat won in a district that not only has been solidly GOP for more than three decades and would have been easily retained by the incumbent Republican had he not resigned from Congress, are such a sign.

This is especially the case because it follows the Democratic takeover in March of another previously solid Republican House district that absolutely would have remained in GOP hands had the incumbent --former Speaker Dennis Hastert --run for reelection.

Yes, this can and will be spun by Republicans.  Poorly funded candidates, special local circumstances that don't translate nationally, bad campaigners, etc. will all be mentioned in the days to come.

But special elections are typically dominated by the party in power.  The incumbent party can usually, and easily, get its supporters to the polls.  Combined with the almost always low-turnout in a special, that gives the incumbent party a huge and frequently insurmoutable advantage.

So what can you say when an insurmoutable advantage turns out not to be enough?  What does it mean when the same thing has now happened in Illinois and Louisiana, two very different states in two exceptionally different parts of the country?

First it indicates that national rather than local issues could dominate the 2008 election campaign.   This should scare the hell out of Republicans given that two of the top issues -- the economy and Iraq/Afghanistan -- clearly favor Democrats right now.

Second, even though he may not actually be on the ballot this November, George W. Bush may also be one of the big campiagn issues.  This should also be scaring Republicans given that the president's job-approval rating is at the lowest point in his presidency in many polls. 

Third, 29 House Republicans have announced that they are not running for reelection or have already resigned from office.  Given the results in these two specials, defending these vacant seats has to be a huge concern for the GOP.  In fact, other seats that are classified as leaning Republican with the incumbent running for reelection now also have to be of some concern.

Fourth, the GOP campaign strategy in Louisiana was to tie the Democratic candidate to Barack Obama and Nancy Pelosi and paint the local candidiate as too liberal for the district.  If using something as tried and true as the "liberal" label doesn't work, what does that say about what will be effective in November?