Yesterday, an incumbent Democrat Representative was almost certainly ousted (McKinney), an incumbent Republican Representative in my home state of Michigan was beaten in his primary and his opponent is almost certain to win, and Democrat Senator Lieberman faces what promises to be an intense battle after losing his primary.

There's an interesting way to interpret this which I haven't seen anyone mention yet. The theory of gerrymandering districts is that if you draw the district to be 60%+ solidly Republican, that the incumbent will basically be assured of re-election. Upon this quite a lot of the informal goverment system is based, including a lot of cynicism on the part of the electorate about the difficulty of actually affecting the government. However, the danger of gaming a system is that it tends to play you right back; you usually win a reprieve, but then whatever forces you were trying to deny come back redoubled.

It's possible that yesterday's results represents just that; gerrymandering didn't save McKinney from losing her office due to terminal embarrassment. (On the part of her constituency, not her.) The Michigan Representative wasn't saved by gerrymandering; his constituency didn't like his direction, so they tossed him out. True, it is almost certainly with another Republican, but that is like small comfort to the loser. Lieberman's situation is more of a fluke, but it's still a situation where what conventional wisdom would have called an assured victory a year ago has become a stiff battle.

Could Congress have pressed the Gerrymandering too far? Is this the pushback that they didn't anticipate? If so, I think it will tend to ameliorate the negative effects of Gerrymandering, as it's not very "useful" if it doesn't have the effect of isolating the elected from accountability to the electorate. Our system is set up in such a way that your really can't force people to keep representatives that they do not want. (Become suspicious if anyone suggests giving incumbants an automatic pass in primaries...)

Only two more years will tell if these three two events are just isolated coincidences, or a trend. But if incumbency in even the most gerrymandered of districts is no longer a near-automatic ticket to re-election, the game changes quite a bit, in ways that will probably be difficult to predict.

This is also predicated on the idea that this is an unusual number primary upsets. People keep constantly pointing out that Lieberman is the fourth Senate incumbant loss since 1980, but I haven't yet seen any stats on Representatives. If this isn't an unusual number, then the only fluky aspect is that they got so much attention.

Oh, and step one is complete...

(Slightly tweaked the 4th paragraph on Aug. 10, 2006, then majorly tweaked after a former co-worked emailed me to point out that Senate districts can't be gerrymandered without quite a lot of effort, since they are the state lines. I can only call that flaming idiocy. If I'd realized that, I wouldn't have posted this at all.)