The term “Electoral College” was not used in the Constitution, and it was rarely used as a systemic term in the nineteenth century. In the nineteenth century, what you see are references to the “electoral colleges,” meaning the gatherings of each state’s electors after the election. The term referring to the whole system becomes commonplace only in the twentieth century.

The system that is now referred to as the Electoral College has a number of key parts. The rudiments are that each state is granted a certain number of electoral votes that equals the number of representatives it has in the US House of Representatives, plus two more for its two US senators.

Each state legislature has the authority to decide how those electors who meet after the election will be chosen. There is no built-in guarantee of having a popular vote for electors, and many states in the country’s early decades did not use popular elections. The state legislature alone chose electors. That has evolved in the direction of having popular votes in each state. It also evolved in the direction of having the state allocate electors on a winner-takes-all basis, which is also not in the Constitution. States can decide to apportion electors by district, as Nebraska and Maine now do, or proportionally in some way.

So, the first step is the number of electoral votes that each state gets. The second step is how each state allocates them. The third step now in most states is that there is a popular election, and this year, it will be on November 3. After that popular election, the electors in each state will meet in the state capitol, assuming there are no disputes about who the electors are. But they’re the electors who are pledged to a particular party and candidate.

Those electors cast their electoral votes in the state capitols. Those votes are then sent on to Congress, where they are counted in January, and Congress officially decides who the winner of the election is. But there’s a caveat here: it’s a little bit unclear who in Congress decides. The votes are counted at a joint session of Congress, and there’s room for disagreement about who in Congress gets to decide if there are disputes about the legitimacy of slates of electors.

The last wrinkle is that if no candidate wins a majority of the electoral votes — and it has to be a majority; a plurality winner is not allowed — then the election immediately reverts to the House of Representatives. And in the House, each state delegation as a whole — not each individual representative — gets one vote. In that case, smaller states would have the same power as the larger states. That, in brief, is the system.

It is unique, and never really imitated. The system as a whole has been an outlier in terms of choosing chief executives for a century or a century and a half, even two centuries.

I’ll take it one step at a time. The first is how we got this institution in the first place. It emerged from a constitutional convention in 1787 that was deadlocked, and in which the Framers really struggled with how to choose a chief executive in a republic. They didn’t have models, they disagreed with one another, and they had a lot of trouble sorting it out. The dominant opinion when the convention met was that Congress should choose the president. At several points in the course of the summer, they took straw votes, and this option kept getting the most support. But each time that happened, delegates would scratch their heads afterward and conclude that it wasn’t a good idea if Congress chose the president. This would undermine real separation of powers, which was part of the point of the kind of government they set up in the first place, and it raised the possibility of corruption and intrigue.

So, they discussed other ideas, including a national popular vote. There was some support for that, including by James Madison, which was quite notable. But others had no interest in a national popular vote. The South tended to oppose a national popular vote, and other delegates felt that the country was too big and communication networks were too slow and uncertain for this method to work well. Of course, it was a given at the time that a national popular vote would be cast pretty much by property-owning white adult males. It’s not the whole people in any instance.

Other ideas also came up, including systems with electors of different types or having governors choose the president, but for months, they failed to come to an agreement. By the end of the summer, they were finished with most of the Constitution. George Washington wanted to go fishing, and most delegates went on vacation for a week. So they delegated the remaining work to the Committee on Postponed Parts, which came up with the idea for what we now call the Electoral College.

How and why did they come up with this? Three things are of note. The first is that the Electoral College amounted to a replica of Congress. Each state’s electoral vote total is based on their representation in Congress, but it avoids the separation of powers and corruption issues by the fact that the electoral colleges meet once every four years for one morning and disband, so they don’t ever legislate.

The second thing is that, by creating this model, the Framers imported into the process of presidential selection the difficult compromises they had already reached with respect to representation in Congress. There was a compromise between the large states and the small states, which is why we have a bicameral legislature. The Senate gives equal representation to all states, while the House is apportioned based on population. And, of course, there was a compromise between the slave states and the free states, which took the form of the rightly notorious “three-fifths clause,” through which Southern states got outsize representation in Congress. These were difficult compromises, but they had been agreed upon, so it was easier to import them into the structure of presidential selection instead of reopening the issues again.

The third thing is that the Framers really didn’t know how it was going to work. They didn’t expect political parties to exist at this time, which is why the system had to be modified by the Twelfth Amendment in 1804. Before then, every elector cast two ballots, and the candidate who received the most votes became president, and the candidate who received the second most votes became vice president. That’s not a conception of party politics; that’s more like a student council election.

Many of the Framers thought that presidential elections would frequently go to the House of Representatives. Others expected the electoral colleges to settle the elections most often, but they really did not know how this was going to function or how candidates would campaign for office. What they did know was that George Washington was going to be the first president, so they didn’t worry too much about it and thought it would evolve over time.

That’s right. Between 1800 and today, the number of constitutional amendments introduced into Congress to significantly alter or abolish the Electoral College totals somewhere close to a thousand. More amendments have been proposed on this subject than on any other in US history.

Amendment resolutions have tended to come in waves. It’s not just in the modern period that dissatisfaction with the Electoral College has mounted. Between 1815 and 1825, the Senate four times passed constitutional amendments by a two-thirds majority to prevent states from using winner-takes-all and require them to use district elections for electors instead.

A couple of those amendments were also aimed at changing the contingent election system, so that if no candidate won an Electoral College majority, each individual member of the House would receive a vote to determine the winner.

The dissatisfactions have been long-standing, and they have shifted somewhat in emphasis over time. For example, there’s relatively little discussion in the modern era about the contingent election system malfunctioning, because an election hasn’t been thrown to the House since 1824. But the institution has been criticized and derided from the beginning. The two words used most frequently to describe the Electoral College in the nineteenth and twentieth centuries were “archaic” and “cumbersome.”

Yes, exactly. In their view, it was at best useless but could also be a source of inflammation.

One note on winner-takes-all. It was widely adopted because it was in the interests of the dominant parties in each state. And once one state went down that road, everyone else had an incentive to do it as well.

I would point to five factors as playing important roles in preventing or blocking change. One is that the Constitution is difficult to amend. It takes a two-thirds vote in each chamber of Congress, plus ratification by three-quarters of the states.

Yes, it’s very difficult to amend. Since the Bill of Rights was passed more or less simultaneously with the Constitution, there have been seventeen amendments over 230 years. That’s not a high rate of amendment. We’ve created a constitution that has evolved thanks primarily to the courts, which has vested them with enormous power. But that’s another story.

The second reason why the Electoral College still stands is that the presidential election system I described before is intricate and has different parts. It is difficult to reform some of the parts without reforming others. For example, you really could not get rid of the contingent system, which benefits small states, unless you also implemented a district requirement so that large states could not cast huge blocks of electoral votes.

The third factor, which has been more prevalent in some periods than others, has been the prominence of partisan self-interest. Surprise: political parties tend to judge electoral reforms not on what is the best thing for the country or the most democratic system, but on what’s going to give them the best chance of winning the presidency.

It wasn’t always easy to gauge the benefits of sticking with the Electoral College, and the parties sometimes miscalculated. There were, moreover, many political leaders who did vote on principle, one way or another. But partisan self-interest, at different periods, has played a key role in blocking reform. One of the longest such periods has been the last forty years. Since 1980, the Republican Party has been convinced that the Electoral College works for them, and thus they have opposed all efforts at reform.

Absolutely. One notable fact here is that the largest single shift in public opinion about the Electoral College took place in the immediate aftermath of the 2016 election. Before the 2016 election, there was majority support among Republicans for a national popular vote, and among Democrats, there was a large majority. Democrats have favored a national popular vote going back to the beginning of polling on this, since the 1940s. But a majority of Republicans also favored a national popular vote for nearly all of that period. In the weeks after the 2016 election, Republican support for a national popular vote plummeted from more than 50 percent to 19 percent.

A fourth reason it’s been difficult to change is that the Electoral College becomes an issue only every four years, and it doesn’t even present itself as a big problem in every election. So there is an acute sense of the problem focused around crises or near crises. But if a major crisis doesn’t happen, then interest dissipates, and people lose their focus.

The last factor is that the alternative of having a national popular vote was stymied for most of our history by Southern politicians in the interest of maintaining white supremacy. This was clearly true while there was slavery. When the idea of a national popular vote was first proposed in 1816 in Congress, Southern senators kiboshed the idea. They argued that if we switched to a national popular vote, the South would lose all the influence it wielded.

It’s also important to note that a national popular vote was not always the most evident or prominent reform idea. Allocating electors by district or by proportional vote within states was the dominant idea from 1800 until the early 1960s.

Exactly. If there is a national popular vote, who is going to be in charge of these elections? If it’s not on a state-by-state basis anymore, what’s next? What else can the federal government change? Slavery? So the South made clear that this was not going to be on the table before the Civil War. After the Civil War, there was a burst of interest in reform, largely but not entirely by the same radical Republicans who pressed for the war and were staunchly behind Reconstruction.

But what happened in the South after the defeat of Reconstruction is deeply disturbing and a less-known part of the story. The Fifteenth Amendment said that no state can deny or abridge the right to vote based on race, color, or previous condition of servitude. So African Americans in the South were made full citizens, and thus they counted 100 percent toward representation. The South’s representation in Congress and in electoral votes went up after the Civil War, but by the end of the nineteenth-century African Americans were disenfranchised again, although they were fully counted for purposes of representation. White Southerners therefore benefited from what amounted to a “five-fifths” clause.

So the South wanted no part of a national popular vote from about 1890 to the 1970s. The closest we came to abolishing the Electoral College was in 1969–1970, when the House of Representatives passed a constitutional amendment by an 82 percent vote. It had majority support in the Senate, but it was defeated by a Southern filibuster.

Yes, the arguments for and against the Electoral College took some unusual turns in the ’70s. Support for Electoral College reform didn’t just disappear after the defeat in 1970. There were a lot of people in Congress who still wanted it to happen, and interest heightened again after the 1976 election between Jimmy Carter and Gerald Ford.

It was a very close election. If a couple of states went a different way, Ford would have been elected president even though Carter won the popular vote. That really disturbed a lot of people, including, I should note, Ford himself and his running mate, Robert Dole, Republican of Kansas. They were both scared by what might happen to the country in case of a “wrong winner” election, and they both supported a national popular vote.

After the 1976 election, reformers led by Birch Bayh again campaigned for a national popular vote. The traditional opponents of change, Southern senators and some very conservative Midwesterners, engaged in the usual delaying tactics, and it took three years for the amendment to come up for a vote. It lost momentum during that period. Meanwhile, the Republicans getting elected to the Senate at this time were becoming more conservative. There was now a split in the Republican Party between its conservative and more liberal wings, with the conservatives on the rise. What became the Reagan wing of the party set itself adamantly against reform.

There was also a public split among African American leaders on this question. There were a number of African American elected officials who were dedicated reformers who wanted to get rid of the Electoral College. John Conyers (D-MI), who was one of the founders of the Congressional Black Caucus, was one of these. John Lewis, who was not yet a public official but was already a legend, was another. There were a lot of African American public officials who wanted to abolish the Electoral College.

But there were other African American figures, most prominently Vernon Jordan, who argued that black voters would be disempowered by a national popular vote. He argued that blacks could wield power in presidential elections precisely because of the Electoral College, because many African Americans were at this time swing voters in key Northern states with a lot of electoral votes; their status as swing voters gave them influence. The disagreements among African American political figures were sharp, and this had an impact on the vote in the Senate, because black opponents of a national popular vote convinced some liberals to vote against it.

The 1979 defeat of the national popular vote was more politically decisive than its predecessor in 1970. After that defeat, there were very few calls for taking it up again. A valuable footnote to this story is that in subsequent years, the split among African American political leaders and journalists on the national popular vote pretty much disappeared. By the early 2000s, politically active blacks were more or less uniformly in favor of Electoral College abolition.

That’s right. Jordan and others were projecting that the swing states in the 1970s, and particularly in 1976, would continue to be swing states. They also assumed that substantial numbers of African Americans would be swing voters, open to voting for the Republican Party, but by the final decades of the twentieth century, this had ceased to be true, and African Americans were reliably in the Democratic column.

I would like to think that there are possibilities of changing it. It would be odd for me to write a book about two hundred years of failed attempts and then turn around and say, “Aha, we can do this, let me just flip this little switch and it’ll happen.” But I think there are some hopeful signs.

The NPVIC has gained a lot of traction. It now has the support of states or jurisdictions with 196 electoral votes. It needs to reach 270 electoral votes, and it may make further progress. This is, in effect, an attempt to change the electoral system without amending the Constitution, because that is so difficult. To my mind, it’s been a very impressive organization and done a very good job of mobilizing support for the idea of change.

But there are further obstacles NPVIC supporters would need to overcome in order to implement the compact. The legal fights that will break out if the compact ever gets to 270 electoral votes will be huge. There will be work for lawyers for many months, if not years, trying to deal with that. Some critics have also pointed out that the compact allows for the election of a plurality winner, but not necessarily a majority winner — which is a problem.

I myself have criticized the compact, because I think it’s inherently unstable. Supposed you get to 270 electoral votes and clear all the legal hurdles. You would now run the next election according to the compact. But then suppose a year after the election or six months after the election, a couple of state legislatures decide they did not like what happened, so they withdraw from the compact. Under the terms of the compact, states cannot withdraw in the six months prior to a presidential election, but they can withdraw at any other time. So then you’d be in a situation where the rules are in flux from one election to the next, and that just doesn’t seem to be a good way to run a railroad.

I don’t think it’s likely that we’ll win reform, but there are strong possibilities. There’s certainly interest in Congress. Within the Democratic Party, it was led by Elizabeth Warren during the primary campaign. And I think a lot will also depend on the outcome of the upcoming election, both in general and in specific states like Texas or Georgia. If these go blue or even significantly purple, I think that some Republicans might begin to rethink the virtues of winner-takes-all and entertain other systems. These things can change around fairly rapidly in Congress. One fact I like to point out is that a constitutional amendment for a national popular vote came up in the Senate in 1956 and got seventeen “yes” votes. Fourteen years later, it got fifty-four. That’s a big switch.

I also take heart, perhaps perversely, from the fact that opposition to reform has already begun to mount. The conservative Heritage Foundation just made and released a propaganda film defending the Electoral College. In a recent Senate debate in South Carolina, Lindsey Graham made the claim that the Democrats would do away with the Electoral College if they won control of the Senate and the presidency. I don’t know who he thought that was going to appeal to. I sense a concern among Electoral College defenders that the possibilities of reform may be opening up.

Exactly. So, stay tuned.

I have to preface my answer by saying it would be different in ways that nobody can imagine, because reforms always have unforeseen consequences.

I may be too influenced by the current presidential campaign, but I suspect it would decrease partisan acrimony. That’s contrary to the image many Electoral College defenders put forward, that a national popular vote would be a huge mess with candidates pulling all the stops to run up their totals. And maybe that would happen. But I think that winner-takes-all sharpens partisan tensions because, as early-nineteenth-century reformers said, there’s no middle resting place. You get everything or you get nothing, so you fight ferociously for any state that you might win.

The campaigns would certainly become more national. I don’t agree with the notion that campaigns would just take place in the major media markets. I don’t think that’s true. Candidates would seek votes wherever they could get them. Campaigns would address more nationally prominent issues, and I think in some ways the political culture would be more nationalized. The parties and candidates would have to make different decisions about the allocation of resources than the ones they’re making now.

There may be things about a national popular vote that we wouldn’t like, but on the whole, I think it would be a major improvement.