Quietly, the same week that most voters ignored last month's primary election, a court ruling threw onto the November ballot a local initiative far more important than any particular electoral race. Ignore this one at your peril. Two years ago, a former aide to Seattle City Council member Charlie Chong ran for city council against liberal cipher Richard Conlin. Jay Sauceda had all the right tools to run a formidable campaign - in addition to his work for Chong, he had originated the highly valuable, and popular, "911" guides to local government (this in an era before that number took on a far more ominous meaning).

But Sauceda didn't actually want to be on city council; instead, he was running solely to draw attention to a ballot initiative idea he was floating, to convert the Seattle City Council from its current system of nine members elected city-wide to a "mixed" one where three would be at large, and six would be elected geographically. By electing six of its members by neighborhood, Sauceda argued, councilpersons would become far more responsive to the needs and desires of their constituents - because they would actually have an identifiable constituency. And by confining elections to districts rather than the whole city, challengers would have more of a chance to beat incumbents. It would be possible to run a campaign, the theory went, by actually doorbelling a neighborhood - something that's physically impossible to do citywide - and it would require a lot less money.

All in all, it sounded like a great idea for a more accountable council. And it seemed to go nowhere, just like Sauceda's run. The week after 9/11 (the bad version), he lost in the primary. And his proposal seemed to have very little traction. It was bad timing for a good idea; the economy was good, the neighborhood revolt that spawned Chong was dead, and the prevailing attitude seemed to be not to fix what wasn't broke. And then, after 9/11, nobody was thinking much about city council anyway.

In politics, timing is everything. Fast forward two years, and now, rather than a good idea at a bad time, Sauceda's baby has become a bad idea at the perfect time.

Suddenly, with a lousy economy and a string of negative local stories, voters are strongly anti-incumbent. Because city council didn't think this initiative would be on the November ballot, it didn't bother to have a competing, better version on the ballot opposite it. And so the district initiative, which would put every incumbent on council up for re-election (or districted out of office) in two years, will be voted up or down in less than a month.

It is likely to pass. And it shouldn't. Because what we will vote on next month is not the same proposal Sauceda was hawking in 2001. The difference is critical - and how that difference came to pass shows exactly why the new version is more dangerous than welcome.

The version we'll vote up or down Nov. 4 has no at large seats. It would convert city council to nine districted seats.

How did the district initiative go from a 6-3 mix to a 9-0 system? Because that was the condition for Mayor Greg Nickels' support, and initiative supporters, lacking patrons in Seattle's political establishment, sold their souls to the mayoral devil.

Greg Nickels does not want a 9-0 district system because he thinks it will serve Seattleites better. He wants it because it will serve Greg Nickels better - because he will have a trump card to play in his thus-far contentious relations with council. He will be able to claim that only he represents the whole city, and that any council member opposing his (or Paul Allen's) initiatives is only speaking for the parochial interests of their district.

The presence of at large members in a council system is a critical check. It prevents the rise of ward politics. It makes it harder for lobbyists or business interests to pick off council members district by district. (A similar proposal last year in Portland turned out to be a stealth project of local business interests to be able to dominate that city's city council.)

It makes it harder for power-hungry mayors to dominate city politics.

There's one other critical difference between a mixed system and one that is solely elected by neighborhood. One of the biggest benefits of a district system is that it would also overturn the arcane committee structure of city council. Currently, each council member chairs a committee (parks, utilities, police, etc.), and in practice what this has meant is that all other council members defer to their colleague in matters involving that committee. Result: staff (or the police guild) lead the council member around by the nose, and if said council member doesn't care about constituent problems, you're out of luck.

Sauceda saw this first-hand; Chong was despised by his colleagues, in no small part because he and his aides would try to fix constituents' problems themselves rather than referring the complaints to the "appropriate" councilperson. Toes were stepped on, feathers ruffled.

But what happens to a district system without any at large members? Rather than, say, Jim Compton not returning your calls about police abuse, Jim Compton (or whoever) could be a roadblock to any matter you have with the city.

Look at Seattle's school board - and the current voter revolt against a dreadfully mismanaged system - for a prime example of how what should be a responsive system can become a collection of unaccountable hacks in real life.

That's what Greg Nickels wanted. It's a different type of politics than Seattle is used to, but that doesn't make it any less dangerous. And the removal of that critical check and balance has turned what was once a good idea into a very, very bad one.