No U.S. political leader in recent memory has stirred quite as much concern about the health of the U.S. political system as Donald Trump. Of course, his administration may eventually prove to be just a one-off episode in the United States’ history, a fluky, non-replicable byproduct of populist backlash and the Electoral College. But it is also possible that Trump’s ascent signifies something more fundamental than that, a harbinger of impending party realignment or a major shift in the way politics is conducted. In the words of Buffalo Springfield, “There’s something happening here. What it is ain’t exactly clear.”

Count me as one who is sure that there are deeper issues at play but unsure as to how they will actually play out in the long run. The biggest surprise to me is not Donald Trump the man. He revealed himself to us many years ago. I was more astonished that the Republican Party would acquiesce so meekly to policies that blatantly contradict conservative principles of fiscal restraint, free trade, and residual Cold War opposition to Russia.

That they have done so raises lots of TBD questions. Has the Republican Party actually changed its policy orientation, or is it temporarily going along with the President in order to avoid internal strife? To what extent is the weakness of the Republican Congressional leaders a reflection of the strength of the underlying ideological tensions in the party’s coalition? And are the Democrats headed in a similar direction, splitting between progressive and moderate wings and united only in opposition to all things Republican?

Assuming for the moment that the current political situation actually reflects something deeper about our current politics and is not just the temporary chaos caused by one unusual political figure, where might this lead us? I think that it could go in three different directions.

The first is the usual story of American politics in transition: Party coalitions are shifting as they have in the past, but the American political party system will retain its same duopolistic form and function. The education, racial, and gender divide sharpens over time with women and college grads continuing to migrate towards the Democrats while Republicans solidify their appeal to rural and exurban whites.

Donald Trump gets political credit for spotting the opportunity to garner the support of those who feel left behind by automation, free trade and competition from immigrant labor. But to turn this into a permanent realignment, the Republicans will eventually need to figure out how to deliver some tangible returns to the downwardly mobile segment of Trump’s base constituency.

In this scenario, there is no radical break from two-party politics. All the coalitional movement takes place within the existing party structures.

Efforts to create a third party will most likely be quickly extinguished by the single member simple plurality electoral system and the deeply embedded accumulation of other rules that favor the current duopoly (e.g. anti-fusion laws, high ballot access thresholds, sore loser prohibitions, etc.).  The only real hope for a third party to gain a foothold would be if the white nationalist base could become regionally concentrated and therefore dominant in some section of the country such as the interior mountain or plains states. But even then, the odds are against any party that bases its appeal on a population that is dwindling due to prevailing demographic trends.

A second direction our party system could take is a more serious departure from politics as usual based on changes in the way we communicate and organize politically. Perhaps President Trump’s populism is not just about particular economic and racial resentments, but represents a fundamental shift in the way we conduct politics.

There is nothing new about a populist leader whipping up fervor in the base to consolidate political power and pursue a particular policy agenda, but in the past it required capturing, owning or suppressing the traditional media. Now the traditional media are fading and are increasingly replaced by social media and the internet. Capturing the entire media space is unnecessarily inefficient. It is more expedient to target messages through social media to a winning coalition of supporters. Big data and the internet enable a much greater capacity to communicate directly with the blocs of voters you need support from. Voters can be identified and mobilized much more easily than ever before. Governing coalitions could thus arise and fade more quickly behind emerging political entrepreneurs. The party duopoly would no longer be necessary and would be replaced by a more fluid and responsive politics.

This may seem implausible at first glance, but it is consistent with certain modern trends. In recent decades, referendums and initiatives have been on the rise in mature democracies. The UK made its momentous decision to withdraw from the European Union based on a vote of the people. The Five Star Party in Italy has emerged out of nowhere to become a governing party by means of an internet platform. Perhaps traditional institutionalized parties will fade and be replaced by direct democracy hybrids.

Unlike the first scenario, this second one involves not just movement across coalitional boundaries, but a radical transformation in the form and function of the party system as a whole. While there is some plausibility to this scenario, there are many reasons to think that it could fail to fully develop. Parties arose because it was necessary to coordinate voters and office holders in order to win elections and govern successfully. Time will tell whether the Five Star party can govern and hold itself together. The notion that “the people” have the time, energy, motivation and knowledge to govern effectively has to date proven illusory.

I am more inclined to think that U.S. politics is headed down a third path where we retain the duopoly form and the essential intermediating function of political parties, but the party as organization moves into the largely unregulated internet space. The history of U.S. political reform is that political activity gravitates into the areas of least legal resistance. This is no clearer example of that principle than campaign finance reform. We imposed stricter restrictions on campaign donations after Watergate, and it eventually gave rise to PACs, independent spending, and now Super PACs. We passed disclosure regulations, and big money found safer ground in nonprofit 501c4s. We tried to offset private campaign money with public subsidies, but the restrictions proved too burdensome, and presidential candidates now avoid the public finance system entirely.

In this third scenario, the Democratic and Republican parties are still dominant and favored in many ways by state and federal laws. But the political parties continue the present trend of morphing into networks of party affiliated groups that spend “independently” on behalf of candidates. Outside groups and social media figures with large followings enforce party discipline rather than Congressional leaders.

This might work effectively as a strategy for forging a winning electoral coalition, but will it lead to effective government? The latter requires aggregating separate interests into some collective consensus. Log-rolling is the politically easiest way to aggregate, but it is not always possible and can lead to disjointed, inconsistent, and excessively costly policies in the end. Bargaining to compromise is better, but harder to do unless there are strong pressures to participate and make concessions. If we are headed toward a politics dominated by a loose coalition of affiliated groups, then where will the centripetal forces that can bind the party coalition in office together come from?

The rhetoric of populism focuses on the swamp of inside players and elected officials. The reality of achieving effective governance may increasingly reside with outside groups. Navigating that civil society morass might prove even more difficult than working in the DC swamp.