Diplomats argue over the shape of the table, politicians over the shape of the district.

Today, the Supreme Court will hear a case that could wipe out the map that Rep. Tom DeLay (R-Texas) and his allies crafted in Texas, which led to a six-seat Republican gain in the House. But even if the court fails to reach that result, the case illustrates many troubling aspects of redistricting that affect what the House will look like in the future.

In 2002, after the census, states redrew their congressional districts to account for population shifts. In those elections, Republicans gained control of the Texas Legislature for the first time in history. They threw out the old map and instituted a new one that would benefit Republicans.

No legislature had engaged in such a re-redistricting for over 100 years, although courts have sometimes ordered new maps middecade. The gains from the new map transformed what would have been a small overall Republican loss of House seats into a small gain in 2004.

In a series of cases grouped under LULAC v. Perry, today the petitioners will argue that the midcensus redistricting plan is unconstitutional. Their argument rests on a big principle and a cryptic comment.

The big principle is "one person, one vote," which has guided redistricting cases since the 1960s. Some states had allowed their congressional districts to get severely out of balance, with rural districts often having many fewer people than fast-growing urban districts. The court has since insisted that states draw districts with nearly absolutely equal numbers of people.

The petitioners are asking the court to invalidate the new map because Texas used the population figures from the 2000 census. By 2004, the petitioners argue, those figures were out of date, as population changes had occurred. Thus the districts created by Texas were not equal in population and violated the standard of one person, one vote.

The cryptic comment is one made by Justice Anthony Kennedy in the recent Vieth v. Jubilirer redistricting case. The court split 4-4, with the conservative justices arguing that the court could not invalidate a redistricting map just because it was partisan gerrymandering and the four liberal justices on the other side. Kennedy sided with the conservatives but made a comment that if some clear standard of partisan gerrymandering were developed he might accept it.

The petitioners' argument is unlikely to succeed for two reasons. First, all districts are subject to changes in population during the decade, and court-ordered redistricting plans have relied on census figures. Second, they are reading a lot into a vague comment by Kennedy.

However this case is decided, it does point out several serious problems with respect to redistricting. Recent years have seen a great polarization of the House and very few competitive districts. Political scientists are engaged in an argument as to how much redistricting causes polarization and uncompetitiveness. Other factors, such as the partisan realignment of the South, are important, and, of course, the Senate has no redistricting and suffers from the same trends. But with the advent of powerful computers and the development of databases with detailed information about individual voters, the parties in charge of redistricting have the information and ability to create ever more precisely gerrymandered districts.

So watch the court's decision, but also watch what states do in 2012 and whether in redistricting they care about anything other than partisan advantage or incumbent protection.

John C. Fortier is a research fellow at AEI.