American democracy faces alarming risks from extremist forces that have rapidly gained ground in our politics. The most urgent focus of political reform must be to marginalize, to the extent possible, these destabilizing forces.

Every reform proposal must be judged through this lens: Is it likely to fuel or to weaken the power of extremist politics and candidates?

In healthy democracies, they are rewarded for appealing to the broadest forces in politics, not the narrowest. This is precisely why American elections take place in a “first past the post” system rather than the proportional representation system many other democracies use.

What structural changes would reward politicians whose appeal is broadest? We should start with a focus on four areas.

Reform the presidential nomination process

Until the 1970s, presidential nominees were selected through a convention-based system, which means that a candidate had to obtain a broad consensus among the various interests and factions in the party. “Brokered conventions” — which required several rounds of balloting to choose a nominee — offered a vivid demonstration of how the sausage of consensus was made. In 1952, for example, the Republican Party convention selected the more moderate Dwight D. Eisenhower over Robert A. Taft, the popular leader of the more extreme wing of the party, who opposed the creation of NATO.

Our current primary system shifted control from party insiders to voters. Now, in a primary with several credible contenders, a candidate can “win” with 35 percent of the vote. This allows polarizing candidates to win the nomination even if many party members find them objectionable. (In 2016, Donald Trump won many primaries with less than 40 percent of the vote.)

How can we restore some of the party-wide consensus the convention system required? The parties can use ranked-choice voting, which allows voters to rank candidates in order of preference. This rewards candidates with broad appeal to a party’s voters, even if they have fewer passionate supporters. In this system, a candidate intensely popular with 35 percent of the party’s voters but intensely disliked by much of the rest would not prevail. A candidate who is the first choice of only 35 percent but the second choice of another 50 percent would do better. Ranked-choice voting reduces the prospects of factional party candidates. Presidents with a broad base of support can institute major reforms, as Teddy Roosevelt, Franklin Delano Roosevelt, Lyndon Johnson and Ronald Reagan demonstrated.

Reform the party primaries

Many incumbents take more extreme positions than they might otherwise endorse because they worry about a primary challenge.

One way to help defang that threat is to eliminate “sore-loser” laws. These laws, which exist in some form in 47 states, bar candidates who have lost in a party primary from running in the general election as an independent or third-party candidate. Thus, if a more moderate candidate loses in a primary to a more extreme one, that person is shut out from the general election — even if he or she would likely beat the (sometimes extreme) winners of the party primaries. One study finds that sore-loser laws favor more ideological candidates: Democratic candidates in states with the law are nearly six points more liberal and Republicans nearly nine-to-10 points more conservative than in states without these laws.

Though Alaska has a sore-loser law, Senator Lisa Murkowski’s 2010 re-election is still instructive. That year, as an incumbent, she lost the Republican primary to a conservative candidate endorsed by the Tea Party and Sarah Palin. But the state permitted an exception to the sore-loser law for write-in candidates, and Ms. Murkowski, running as a write-in Republican candidate, won the general election.

If sore-loser laws are eliminated, that reform should be combined with ranked-choice voting in the general election. That would ensure that in a multicandidate general election, the winner would reflect a broad consensus. Other ideas for restructuring primaries to minimize the existence of factional candidates include one adopted by Alaska voters in November: The top four candidates in a single primary move on to the general election, where the winner is chosen through ranked-choice voting.

Reform gerrymandering

Many reformers agree on the need to take redistricting out of the hands of partisan state legislatures and give it to a commission. In several recent state ballot initiatives, voters have endorsed this change. But that still raises a question: What constitutes a fair map?

Redistricting reform should have as a goal the creation of competitive election districts. Competitive districts pressure candidates from both the left and the right, which creates incentives to appeal to the political center. They also encourage more moderate candidates to run in the first place, because they know they have a greater prospect of winning than in a district whose seat is safe for the other party.

In safe seat districts, as long as a candidate survives the primary, that person is assured of winning the general election — which means primary candidates don’t have to move toward the center.

The sources of centrism in the House or Senate frequently come from politicians in swing districts or states. In the recent House impeachment, for example, the percentage of Republicans elected with 57 percent of the vote or less who voted for impeachment was more than double that of Republicans elected with more than 57 percent of the vote. Similarly, it was Democrats holding competitive seats who resisted the initial impeachment of President Trump, until news broke of his call with Ukraine.

Not every district can be made competitive. But in 2018, maps that emphasized competitiveness could have produced at least 242 highly competitive districts, although only 72 races actually were competitive. The more senators and representatives who face competitive pressures in their general elections, the larger the forces of compromise and negotiation will be in Congress.

The goal of creating competitive districts should not take a back seat to approaches that focus on whether the partisan outcomes match vote shares in a particular map. In these approaches, the closer a plan comes to matching the number of seats one party gets to its statewide share of the vote, the fairer that map is deemed to be. So, if 55 percent of the statewide vote goes to Democrats, then Democrats should have roughly 55 percent of the seats in the state Legislature and the U.S. House delegation from the state. The problem comes when a fair partisan map produces candidates, in getting to that 55 percent overall, who are all elected from seats so safe for one party, they never have to compete for voters in the center.

If we want to reduce extremist forces in our politics, candidates should have to appeal to a diverse set of interests and voters in competitive districts as much as possible.

Reform campaign-finance reform

The way campaigns are financed also has major effects on the types of candidates who run and win.

Campaign-finance efforts are now rightly focused on “leveling up” campaign dollars — by providing public funds to candidates — rather than trying to “level down” by imposing caps on election spending. That shift is partly a result of Supreme Court doctrine, but also of the difficulties of narrowing the number of channels through which money can flow to candidates.

But publicly financed elections can take at least two different basic forms, and the form taken can have significant ramifications for whether the forces of extremism are further accentuated or limited.

In the traditional form of public financing, which is used in around 11 states that have public financing, the government provides grants of campaign funds to the qualified candidates.

In the other form — which has taken up much of the reform energy in recent years — the government provides matching funds for small donations. This based on a matching-funds program that has existed in New York City for a number of years.

The campaign-finance reform proposal that House Democrats passed after the 2018 midterms, which is now a focus of the Democratic agenda, would include a small-donor matching program. The legislation would provide $6 in public funds to candidates for every dollar they raise in small donations (those of $200 or less), up to a certain level.

But there is a risk that making public funding proportional to small donations will accelerate polarization and extremism even further. Research suggests small donors are more ideologically extreme than average citizens and donate to ideologically more extreme candidates. In his campaigns, Mr. Trump raised a higher percentage of his contributions from small donors than any major-party presidential nominee in history.

Numerous studies have shown that in general, individual donors (large and small) are the most ideological source of money in politics. Traditional public financing is far more neutral in the types of candidates who benefit.

In debating campaign-finance reform, we must focus not just on the values of participation or equality but also on the overall effects different approaches to reform are likely to have on political extremism or moderation.

Jan. 6 provided a painful demonstration of the dangerous currents gathering in American political culture. Every proposed election reform must now be measured against this reality to make sure political reform furthers American democracy.

Richard H. Pildes is a professor at New York University’s School of Law and an author of the casebook “The Law of Democracy: Legal Structure of the Political Process.”

The Times is committed to publishing a diversity of letters to the editor. We’d like to hear what you think about this or any of our articles. Here are some tips. And here’s our email: letters@nytimes.com.

Follow The New York Times Opinion section on Facebook, Twitter (@NYTopinion) and Instagram.