After the 1980 census, when California Democrats needed to redraw their state legislative districts, they turned to their map-making guru, Congressman Phillip Burton, one of the shrewdest political minds of his generation. Burton gave them a map so blatantly partisan that it sank the state's Republicans into a minority status from which they haven't recovered.

Needless to say, Republican legislators were livid. "What's the matter?" Burton said. "Can't they take a joke?"

Gerrymanders are no joke, as the congressman well knew. But they always look a lot more amusing to the party that draws them than to the party that finds itself frozen out. Shortly after Burton's death in 1983, Democrats in the Legislature drew a congressional district for his widow, Sala, that all but precluded a serious GOP challenge. Democrats gleefully called it a Salamander. Republicans didn't think that was so funny either.

We all know that gerrymanders are nothing to laugh about, but we haven't paid enough attention to the enormous political consequences that can follow. Some of them are so grossly unfair, however, that it's impossible to sweep them under the rug.

Wisconsin is a good example. It has been living for nearly a decade now with a set of Republican legislative districts so skewed that they have tainted the state's entire political system. In 2019, Republicans won 46 percent of the state House and Senate vote and came away with 63 percent of the seats. That year the state elected a Democratic governor, Tony Evers, and the GOP began using its spurious legislative majorities to thwart him at every turn — curtailing his administrative powers, rejecting his cabinet appointees and fighting his efforts to establish strict controls on social activity to slow the spread of the coronavirus. There was very little that Evers, elected fair and square by the state's voters, could do about this.

Even political leaders who foster gerrymanders frequently go on to experience a kind of remorse. Martin O'Malley, who as governor of Maryland was the key figure behind that state's egregious Democratic gerrymander in 2011, has become a born-again reformer. John Kasich, the Republican governor of Ohio who signed that state's partisan gerrymander the same year, went on to admit that "partisan gerrymanders are unconstitutional, are harming our republican government and readily can be identified and addressed by courts."

BASICALLY, EVERYBODY ADMITS that gerrymanders are a bad thing, even the politicians who dirty their hands with them. The question is whether there's anything meaningful we can do to get rid of them.

It's easy to argue that the answer is no. The U.S. Supreme Court, after decades of hinting that it might rule against partisan gerrymanders someday if it got the right case, ruled the opposite way in 2019. The justices signaled that they are willing to throw out gerrymanders if they are racially biased or involve population disparities, but not if their sole purpose is political mischief.

If the curse of gerrymandering is to end, or even come under some form of reasonable regulation, it will have to happen at the state level. Is it happening? Well, it's very hard to say. Voting in two states this week delivered a decidedly mixed result.

Virginia voters cast a clear majority of their ballots on Tuesday in favor of keeping a 2019 reform that takes state legislators out of the direct mapmaking process and gives it to a bipartisan commission, with eight members picked by legislators and eight chosen by a panel of retired judges. The Legislature can't amend the commission's product, although it can reject it. But if a map is rejected twice, it goes to the state Supreme Court for final resolution.

Much of the state's Democratic establishment opposed this reform, and for understandable reasons: For the past decade, Democrats have been living with a gerrymander engineered by Republicans when they were in control in 2011. Despite that gerrymander, Democrats managed to take the Legislature back in 2019, and after the new census they would have a chance to get even using the old system. I understand why they feel that way. But the reform is the right thing to do, and the voters thought so decisively.

On Tuesday, however, voters in Missouri narrowly threw out the Clean Missouri plan, passed in 2018, that would have turned redistricting over to a nonpartisan demographer. Republicans, who hold a majority in both the state House and Senate, challenged that reform with Amendment 3, a plan to substitute a commission whose members they and the governor would choose. In 2018, using a politically driven plan from the previous census, the GOP had taken more than 70 percent of the legislative seats with 57 percent of the statewide vote.

IT MAY SEEM that turning the redistricting process over to a bipartisan commission is the simplest answer to the problem. It's an answer, but not always a simple one. At the moment, some 21 states use a commission system of one sort or another. But there are commissions, and then there are commissions.

Placing the process in the hands of a truly nonpartisan panel or neutral expert, with no incentive or weakness for favoring one party or the other, is about the best you can hope for. But it's not always easy to implement.

Simply making the job bipartisan, with each side getting a presumably equal shot at the result it wants, isn't a guarantee of anything. Currently, eight states use a commission system that operates on what's described as a bipartisan basis. In virtually all these states, and in some of the nominally nonpartisan ones as well, it's still the politicians who hold the important cards. Arkansas uses a three-member commission whose members are the governor, the attorney general and the secretary of state. That's not exactly blind justice. In Pennsylvania, two legislative leaders from each party pick a commission member, and then try to agree on a fifth one. If they fail to agree, the tie-breaking member is chosen by the state Supreme Court, which in recent years has been about as apolitical as Tammany Hall in 1920.

Still, even the most flawed redistricting commission probably does reduce the prospects of a truly awful one-party gerrymander at least a little. But it is highly vulnerable to another defect: It does not prevent the legislators who select the commission members from conniving to create a system that protects incumbents of both parties by giving them safe districts and making it very difficult for genuine competition to break through. That what Missouri's Amendment 3 may well end up doing. A map drawn up that way may be highly satisfactory to both political parties. That doesn't make it fair.

THERE'S NO QUESTION that we have a problem of noncompetitive districts as well as a problem of partisan malfeasance. There's some dispute over just why we have it. Gerrymandering is obviously a culprit, but so too, it would appear, is clustering. That is, Americans more than in past generations live in communities and neighborhoods filled with people who think the way they do. This means many fewer 50-50 two-party constituencies and more 70-30 Democratic or Republican ones. It means more districts in which the general election is a foregone conclusion and the primary, which determines the ultimate result, skews toward the ideological extremes of each party.

This phenomenon was documented by the journalist Bill Bishop in his 2008 book, The Big Sort, and by all indications, it has deepened in the years since. The best example may be Pennsylvania, studied recently by the Stanford political scientist Jonathan Rodden. Rodden's research found that between 2012 and 2018, Democrats won 15 of 18 statewide elections in Pennsylvania, but never came close to a majority in the Legislature. This he attributes not so much to gerrymandering as to the fact that Democratic voters are more than ever clustered in the state's cities, where they give their candidates huge majorities that the candidates don't need. Republicans, scattered more evenly in suburban and exurban constituencies, win most of the seats by smaller but still decisive margins. "The Democrats' long-term problem," Rodden wrote in a book published last year, "is with seats, not votes."

There's a technical solution to this problem, based on the concept of "efficiency gaps." Under the ultimate logic of this idea, a state electoral system where one party consistently wins 60 percent of the vote but ends up with 40 percent of the legislative seats is unfair by definition, whether the cause is gerrymandering, clustering or anything else.

It's an argument that can be made effectively, but so far it has failed to win meaningful judicial endorsement. A federal district court in Wisconsin accepted it in the Gill v. Whitford case in 2016, but in 2018 the U.S. Supreme Court refused to consider it. And if the court had entertained it, myriad questions would have surfaced over how the rebalancing was to be done, who was authorized to do it, and what it would take for a set of electoral districts to be deemed politically fair. The whole process remains, as Justice Felix Frankfurter lamented in 1963, "a political thicket."

The bottom line is that tainted districting is a many-headed monster that cannot be slain by any single reform, or package of reforms. It has to be confronted one small step at a time, in a variety of places and situations. That is what we may be starting to do.