Skip Navigation

When Impeachment Fails

Elections are a vital safeguard against presidential abuse of power. They must be protected from subversion.

November 30, 2021
An illustration featuring text from James Madison's notes on the Constitutional Convention, an image of former President Donald Trump, and other iconography
Brennan Center/Fabrice Coffrini/SchulteProductions/Getty

The Consti­tu­tion’s separ­a­tion of powers among Congress, the pres­id­ent, and the judi­ciary is a cent­ral element of the U.S. system of govern­ment, designed in part to enable each branch to prevent abuses of power by the others. When an abuse of power by an exec­ut­ive or a judge is suffi­ciently seri­ous, the Consti­tu­tion author­izes the legis­lat­ive branch to impeach and remove the offend­ing offi­cial. The Consti­tu­tion’s authors — having just fought a revolu­tion against a tyran­nical king — espe­cially feared the risk of abuse by a pres­id­ent; they consequently addressed impeach­ment in the Consti­tu­tion even before outlining pres­id­en­tial powers. But the separ­a­tion of powers is neither abso­lute nor expli­cit.

This essay focuses on the mani­fest weak­ness of the Consti­tu­tion’s checks and balances in our current polit­ical envir­on­ment, as demon­strated by the 2020 and 2021 impeach­ments of former Pres­id­ent Donald Trump for abuse of power in deal­ings with a foreign nation and his attempts to subvert and nullify the 2020 pres­id­en­tial elec­tion. In light of Trump’s acquit­tals by the Senate, we face the ques­tion of whether impeach­ment works as a check on exec­ut­ive power and, indeed, whether checks and balances work over­all. This ques­tion becomes espe­cially sali­ent if our elec­tion system is so flawed — or so hobbled by vote-suppres­sion laws — that it cannot perform its primary consti­tu­tional roles of enabling a truly repres­ent­at­ive form of govern­ment and prevent­ing tyranny. Unfor­tu­nately, we conclude that impeach­ment, without a work­ing elect­oral system, cannot play its salut­ary role of protect­ing our demo­cracy.

Origins of the separation of powers and the Impeachment Clause

Like the other essays contained in the Feder­al­ist Papers, James Madis­on’s Feder­al­ist No. 10, writ­ten anonym­ously in 1787 under the pseud­onym “Publius,” sought to win support for the new consti­tu­tion. Feder­al­ist No. 10 explains how the revised consti­tu­tional design improved on older models of demo­cratic govern­ment that were vulner­able to what Madison termed the “mischiefs of faction.” Madison defined a faction as “a number of citizens, whether amount­ing to a major­ity or a minor­ity of the whole, who are united and actu­ated by some common impulse of passion, or of interest, adverse to the rights of other citizens, or to the perman­ent and aggreg­ate interests of the community.”

As explained by Zachary Courser, Eric Helland, and Kenneth P. Miller in their book Parch­ment Barri­ers: Polit­ical Polar­iz­a­tion and the Limits of Consti­tu­tional Order:

Madison believed that such factions were inev­it­able in a free soci­ety. In his view, the chal­lenge was to develop a system of govern­ment that would make it diffi­cult for major­ity factions to form, gain control of govern­ment, and over­power their oppon­ents. The Consti­tu­tion thus contained numer­ous means for controlling factional divi­sion, restrain­ing pure major­ity rule, and prevent­ing concen­tra­tions of power. These features included feder­al­ism, indir­ect elec­tions, separ­ately elec­ted insti­tu­tions, separ­a­tion of powers, bicam­er­al­ism, the pres­id­en­tial veto, and numer­ous checks and balances.

Notwith­stand­ing these consti­tu­tional attrib­utes, it appears that Madison wanted to make unmis­tak­ably clear the inten­ded separ­a­tion of powers. In his original draft of the Bill of Rights, he included an amend­ment that expli­citly delin­eated the rela­tion­ship among the branches.

Congress, however, rejec­ted the proposed amend­ment; its members appar­ently believed the consti­tu­tional struc­ture of govern­ment clearly distrib­uted power among the branches. Consequently, the Consti­tu­tion does not expressly declare that power shall be divided among the branches. Instead, the Consti­tu­tion broadly summar­izes each branch’s powers but says little about their rela­tion­ships to the other branches: Article I states that “all legis­lat­ive Powers . . . shall be vested in a Congress,” Article II declares that “the exec­ut­ive Power shall be vested in a Pres­id­ent,” and Article III provides that “the judi­cial Power of the United States, shall be vested in one Supreme Court, and in such inferior courts as the Congress may from time to time ordain and estab­lish.”

The consti­tu­tional separ­a­tion of powers prevents the concen­tra­tion of power in any one branch and provides each branch with tools to defend its interests and powers against encroach­ment by the others. As Madison argued in Feder­al­ist No. 51, “Ambi­tion must be made to coun­ter­act ambi­tion.”

Most Amer­ican legal schol­ars and polit­ical scient­ists have char­ac­ter­ized Madis­on’s insight essen­tially as plur­al­ism — in which compet­ing interests peace­fully nego­ti­ate polit­ical outcomes. But this plur­al­istic model no longer works as the framers hoped it would, in part because it assumes many small, fluid factions (or interest groups) jock­ey­ing for power and insti­tu­tional interests ensur­ing that each branch fights for its own turf.

However, neither element of plur­al­ism is func­tion­ing now — nor has been for many years — and these break­downs were partic­u­larly appar­ent during the impeach­ments of Donald Trump.

Painting depicting the signing of the American Constitution E. R. Curtiss/Wiscon­sin Histor­ical Soci­ety/Getty
“The Sign­ing of the Amer­ican Consti­tu­tion,” one of four mural paint­ings by Albert Herter displayed at the Wiscon­sin State Capitol.

The failed impeachments of Donald Trump

The Consti­tu­tion’s impeach­ment provi­sions could not be more pertin­ent in purpose to Trump’s impeach­ments. William Davie, a deleg­ate to the Consti­tu­tional Conven­tion in 1787, argued for the inclu­sion of an impeach­ment provi­sion because without such a clause, a pres­id­ent might “spare no efforts or means whatever to get himself re-elec­ted.” Congress expli­citly embed­ded the impeach­ment process in the Consti­tu­tion. In Article I, Sections 2 and 3 respect­ively provide that “the House . . . shall have the sole Power of Impeach­ment” and “the Senate shall have the sole Power to try all Impeach­ments” author­ized by the House. Article II closes the impeach­ment loop by provid­ing that “the Pres­id­ent, Vice Pres­id­ent and all civil Officers of the United States, shall be removed from Office on Impeach­ment for, and Convic­tion of, Treason, Bribery, or other high Crimes and Misde­mean­ors.”

The Consti­tu­tion thereby gran­ted Congress a means to directly check a pres­id­ent’s abus­ive ambi­tion. However, our hyper­par­tisan polit­ical envir­on­ment has nulli­fied the prophy­lactic power of impeach­ment and degraded other defenses against consti­tu­tional fail­ure.

The House first impeached Trump in late 2019 because it found that he commit­ted an “Abuse of Power” (Article of Impeach­ment I) by “by pres­sur­ing an ally — Ukraine — to announce an invest­ig­a­tion into a polit­ical rival — former Vice Pres­id­ent Joe Biden, while with­hold­ing milit­ary aid and dangling a head-of-state meet­ing, thereby corrupt­ing the integ­rity of U.S. elec­tions.” Addi­tion­ally, the House found that Trump obstruc­ted Congress (Article of Impeach­ment II), “a coequal branch of govern­ment, by with­hold­ing docu­ments and prevent­ing witnesses from testi­fy­ing, thereby imped­ing Congress’ invest­ig­at­ory power.” 

Pres­id­ent Trump’s attempts to over­turn the 2020 pres­id­en­tial elec­tion and incite­ment of the mob that attacked the U.S. Capitol led to his unpre­ced­en­ted second impeach­ment by the House in Janu­ary 2021. Once again, the factual predic­ate of this impeach­ment was Trump’s attemp­ted subver­sion of the elect­oral system. The events before, during, and after Janu­ary 6, 2021, exem­plify why the Founders added the impeach­ment clauses to the Consti­tu­tion. In Madiso­nian terms, the House found twice that Trump abused his powers in order to subvert and nullify the “primary control” — as Madison described it in Feder­al­ist No. 51 — our system has for prevent­ing tyranny: elec­tions. The Senate, however, failed both times to muster the two-thirds major­ity required to convict Trump, despite the condem­na­tions of Trump by several Repub­lican lead­ers.

Why Congress did not check the president

Madison did not fore­see that ambi­tion would even­tu­ally reside with parties and not with insti­tu­tions. Congress does not really care much about — or exer­cise — its insti­tu­tional prerog­at­ives under the Consti­tu­tion when it comes to “check­ing pres­id­en­tial unilat­er­al­ism,” in legal scholar Neal Devin­s’s phrase. This becomes espe­cially evid­ent when the same party controls both the pres­id­ency and one or both cham­bers of Congress.

As Devins argues in his article “Congress, the Courts, and Party Polar­iz­a­tion”: “Polar­iz­a­tion has signi­fic­antly exacer­bated this phenomenon. . . . Lawmakers from the pres­id­ent’s polit­ical party no longer assert insti­tu­tional prerog­at­ives to resist pres­id­en­tial encroach­ments.” Devins more gener­ally observes that “polar­iz­a­tion dimin­ishes the abil­ity of lawmakers to work together to defend Congress’s insti­tu­tional prerog­at­ives. . . . Today’s lawmakers increas­ingly identify with party-defined messages and seek to gain insti­tu­tional power by advan­cing within their own respect­ive party.”

In short, while Congress does have many tools to defend its interests, it makes little use of them. Indi­vidual members of Congress might share some stake in the insti­tu­tion, but the reelec­tion imper­at­ive domin­ates their decision-making because it is exist­en­tial. Thus, they fail to defy a pres­id­ent of their own party out of fear; they need favors and, usually, endorse­ment from the exec­ut­ive branch; and, thanks to state-level gerry­man­der­ing that makes party primar­ies decis­ive in congres­sional elec­tions, they fear their own elect­oral base’s wrath if they defy a pres­id­ent of their own party. In addi­tion, they choose not to defend Congress as an insti­tu­tion when it comes at the expense of advan­cing their own careers by secur­ing choice commit­tee assign­ments, becom­ing a leader in their party, or promot­ing some issue or interest import­ant to their party.

Devins sees these elect­oral and personal forces as a collect­ive action prob­lem, as they oper­a­tion­ally inhibit Congress in at least two ways. First, members don’t fight for Congress’s interests as a branch of govern­ment. War powers is a good example. While Congress has the sole power to declare war, repres­ent­at­ives and senat­ors don’t get pres­sure from constitu­ents to cut back on milit­ary missions (perhaps in part because service is volun­tary). Thus, Congress often defers to the exec­ut­ive and does not chal­lenge pres­id­en­tial war-making unilat­er­al­ism, to avoid being accused of obstruct­ing neces­sary milit­ary oper­a­tions.

Second, members do not always share Congress’s insti­tu­tional interests. In cases of ideo­lo­gical disagree­ment, they often choose to defend whatever legis­lat­ive or policy outcome they prefer rather than the process or consti­tu­tional prerog­at­ives of Congress. For example, members have indi­vidu­ally and collect­ively chal­lenged, on consti­tu­tional grounds, federal legis­la­tion passed by Congress, includ­ing the Afford­able Care Act, the Defense of Marriage Act, and the federal Partial Birth Abor­tion Act. Filing or join­ing litig­a­tion against the laws, members claim Congress lacks author­ity to pass them; they are less inter­ested in protect­ing Congress’s consti­tu­tional author­ity than in winning the policy war.

Hence Trump’s first impeach­ment trial was doomed from the start, with members of Congress from Trump’s party more concerned with main­tain­ing partisan solid­ar­ity than check­ing an out-of-control pres­id­ent. Before the first Senate trial even star­ted, prom­in­ent Repub­lic­ans, includ­ing the then-Senate major­ity leader, Mitch McCon­nell, and Senator Lind­sey Graham, then chair­man of the Judi­ciary Commit­tee, went on the record saying they had already made up their minds and that no evid­ence would change their determ­in­a­tion to acquit Pres­id­ent Trump.

As Graham said at a forum in Qatar a month before the Senate trial, “I am trying to give a pretty clear signal I have made up my mind. I’m not trying to pretend to be a fair juror here.” McCon­nell went so far as to say during an inter­view on Fox News that he was collud­ing with the White House in the lead-up to the Senate trial: “Everything I do during this, I’m coordin­at­ing with White House coun­sel. . . . There will be no differ­ence between the pres­id­ent’s posi­tion and our posi­tion.”

McCon­nell and Senate Repub­lic­ans then voted along party lines in the first trial of Trump to block testi­mony from witnesses and bar intro­duc­tion of new evid­ence, despite learn­ing that former National Secur­ity Adviser John Bolton had evid­ence support­ing the charges that Trump delayed congres­sion­ally author­ized aid to Ukraine to lever­age assist­ance in his reelec­tion campaign. The media cover­age reflec­ted this polar­iz­a­tion: right-wing outlets called it a “witch hunt” to over­turn the 2016 elec­tion, while moder­ate or liberal outlets denounced the Repub­lican “sham trial.” Voters shared these views, with 84 percent of Demo­crats support­ing impeach­ment, joined by only 9 percent of Repub­lic­ans.

The second impeach­ment of Pres­id­ent Trump followed a similar course. Demon­strat­ing the intens­ity of polar­iz­a­tion (and the fealty of congres­sional Repub­lic­ans to Trump’s base), few members of either cham­ber shif­ted alli­ances even after the Janu­ary 6 insur­rec­tion at the Capitol, despite having been in danger them­selves. In the House, only 10 Repub­lic­ans voted for impeach­ment; in the Senate, only 7 Repub­lic­ans joined all 50 Demo­crats to find Trump guilty on the charge of “incite­ment of insur­rec­tion.” The insur­rec­tion was the unfor­tu­nate, deadly conclu­sion to Trump’s months-long quest to propag­ate denial of his defeat by Joe Biden (in public comments and more than 60 failed lawsuits).

The vote of 57–43 for convic­tion in the Senate amoun­ted to the most bipar­tisan support for convic­tion ever in a pres­id­en­tial impeach­ment trial. Yet we all knew even before the Senate trial that it would end with Trump’s second acquit­tal. Senate Repub­lic­ans announced in advance that Trump would win, concoct­ing a weak defense to justify their abject self-abase­ment: they conten­ded that it is not legal to convict a pres­id­ent after who has left office — a posi­tion rejec­ted by nearly every consti­tu­tional scholar and histor­ian.

The Red Wall for acquit­tal in the second Senate trial of Trump is evid­ence of the durable power of faction — specific­ally, partisan faction.

Impeachment is inoperable in the face of party polarization

Acute partisan faction­al­ism is a 21st-century prob­lem, but it has a long history. For example, when Nixon was embroiled in the Water­gate scan­dal, he saw his support among Repub­lic­ans, inde­pend­ents, and south­ern conser­vat­ive Demo­crats substan­tially decrease as the scan­dal unfol­ded, but it never fully evap­or­ated. At the time of Nixon’s resig­na­tion, in August 1974, approx­im­ately 24 percent of the public still approved of him. A smal­ler propor­tion of voters aban­doned Pres­id­ent Trump: 34 percent of the public still approved of Trump in Janu­ary 2021. In these cases, perhaps because multiple vari­ables worked together to amplify polar­iz­a­tion, partis­ans seemed almost imper­vi­ous to evid­ence.

Image of an "Impeach Nixon" demonstration outside the White House MPI/Getty
A demon­stra­tion outside the White House in support of the impeach­ment of Pres­id­ent Richard Nixon.

Support for Trump remained confound­ingly steady through all his scan­dals, invest­ig­a­tions, and impeach­ments. A Berke­ley IGS poll, conduc­ted Novem­ber 21–27, 2019, showed 30 percent of likely Cali­for­nia voters in Novem­ber plan­ning to vote for Trump — only a small dip from the 31.6 percent who suppor­ted him in 2016. Also show­ing little change were the numbers of those oppos­ing his reelec­tion: 68 percent of Cali­for­ni­a’s registered voters did not approve of Trump’s perform­ance, an increase of only two points from Decem­ber 2017. This was also true at the national level.

A healthy plur­al­istic demo­cracy should have citizens who parti­cip­ate in many differ­ent types of “factions” or interest groups — hypo­thet­ic­ally, they could include evan­gel­ical Chris­tian envir­on­ment­al­ists or gay liber­tari­ans. In such a system, on any given issue, citizens might feel closer to Demo­crats or to Repub­lic­ans, but their partisan align­ments might shift from issue to issue.

Hybrid alle­gi­ances enable shift­ing coali­tions and create oppor­tun­it­ies for each party to compete for voters. But such voters — the ones envi­sioned by James Madison in Feder­al­ist No. 10 and by mid-20th-century polit­ical scient­ists — don’t exist in mean­ing­ful numbers anymore. Polar­iz­a­tion and self-sort­ing have tied almost all voters, perhaps irre­vers­ibly, to one or the other major party.

In Congress, these trends in party loyalty have led over the past 25 years to lawmak­ing grid­lock, govern­ment shut­downs, a weakened commit­tee struc­ture, cent­ral­iz­a­tion of power in party lead­er­ship, abuse of proced­ural rules to favor the party in power, highly partisan commit­tee invest­ig­a­tions (e.g., Benghazi), flout­ing of congres­sional subpoenas (most notably by Trump admin­is­tra­tion officers and employ­ees), punish­ment of members who break ranks (e.g., Rep. Liz Cheney), and the disap­pear­ance or neuter­ing of swing members and voters. People choose where they will live, with whom they will asso­ci­ate, and what news they will watch on the basis of so-called affect­ive polar­iz­a­tion.

As, for example, congres­sional schol­ars Thomas Mann and Norman Ornstein have explained, with the “tele­com­mu­nic­a­tions revolu­tion” and the advent of the inter­net, the “pleth­ora of chan­nels, websites, and other inform­a­tion options has frag­men­ted audi­ences and radic­ally changed media busi­ness models.” Now we can watch or listen to news that rein­forces our own biases and aggrav­ates the increas­ing tribal­iz­a­tion of Amer­ica.

Work­ing in tandem, these same forces — polar­iz­a­tion, partis­an­ship, and media configured to rein­force confirm­a­tion bias — have weakened impeach­ment as a tool to check an out-of-control pres­id­ent. Notwith­stand­ing the unpre­ced­en­tedly large bipar­tisan vote for convic­tion in Trump’s second Senate trial, party loyalty proved more power­ful than evid­ence of his guilt.

We ques­tion, there­fore, whether the authors of the Consti­tu­tion created a struc­ture for impeach­ment strong enough to counter the intens­ity and durab­il­ity of faction and impel a pres­id­en­tial convic­tion on indis­put­able evid­ence of abuse of power. We also have no histor­ical basis for settling this ques­tion because Pres­id­ent Andrew John­son was acquit­ted, Pres­id­ent Nixon resigned before a Senate trial, and Pres­id­ent Clin­ton and Pres­id­ent Trump were acquit­ted.

C-SPAN coverage image of Clinton impeachment vote. -/C-SPAN/AFP via Getty
Over­all shot of the U.S. Senate at the end of the impeach­ment vote on perjury charges against Pres­id­ent Bill Clin­ton, as shown on C-SPAN. The Senate failed to convict Clin­ton.

If impeachment won’t cure an abuse of presidential power, what will?

Some schol­ars argue that even without removal, impeach­ment still serves a purpose. New compar­at­ive research from Tom Gins­burg, Aziz Huq, and David Landau on impeach­ment provides “the first compre­hens­ive analysis of how consti­tu­tions glob­ally have addressed [removals], and what the consequences of differ­ent design choices are likely to be.” Look­ing at a large data set regard­ing impeach­ment and removal along with case stud­ies of South Korea, Brazil, Paraguay, South Africa, and the United States, they argue that impeach­ment allows pres­id­en­tial or semi-pres­id­en­tial systems to get out of a major crisis.

However, this “exit” from crises works best when the process of impeach­ment culmin­ates in new elec­tions imme­di­ately or soon after a head of state is impeached. In South Korea for example, new elec­tions must be called 60 days after the impeach­ment process has concluded. Huq, Gins­burg, and Landau argue that this process allows a reboot of the system.

In the United States, we do not have such an imme­di­ate-elec­tion require­ment — but we did coin­cid­ent­ally hold elec­tions last Novem­ber. What can this tell us about whether impeach­ment can provide this cleans­ing process? Some might say the elec­tion’s outcome proved that here too impeach­ment can serve such purpose. But because our elec­tion occurred only because the calen­dar dictated, we cannot rest on this comfort­ing conclu­sion.

Some comment­at­ors believed that a “reset” elec­tion would allow Amer­ic­ans to say, defin­it­ively, “Trump is not who we are.” To serve as a reset, impeach­ment ought to result from elec­tions that reflect the popu­lar verdict — who we actu­ally are and what we actu­ally believe — about a pres­id­ent’s conduct.

It is fair to say that the last elec­tion reflec­ted “who we are.” More than 159 million people voted in the 2020 federal elec­tion and 2021 Senate runoffs, the highest voter turnout in more than a century. As we have noted else­where, voters gave Demo­crats “trifecta” lead­er­ship of the White House, the Senate, and the House — a defin­it­ive rebuke of Trump and his congres­sional back­ers.

But even if the 2020 elec­tion was a cure, there’s little guar­an­tee that future elec­tions will be, given the ways Repub­lic­ans are seek­ing to under­mine elec­tions in the Senate and state legis­latures.

Contrary to the intent of the Consti­tu­tion’s authors, Congress responds to issues slowly — or not at all — because the Senate does not repres­ent the popu­la­tion, but rather states, and the House is gerry­mandered to a fare-thee-well. Senat­ors from low-popu­la­tion states — more likely rural, white, conser­vat­ive, and Repub­lican — are at least as power­ful as senat­ors from higher-popu­la­tion states, in part because they repres­ent more homo­gen­eous interests. Further, the arcane Senate rule of the fili­buster exag­ger­ates the already anti­demo­cratic nature of the body. So proposed federal initi­at­ives to fix prob­lems of demo­cracy laid bare by the last elec­tion — by easing voter regis­tra­tion, prevent­ing or curing partisan gerry­man­der­ing, and making voting itself more access­ible — are stymied by the abil­ity of the Repub­lican minor­ity to deny a fili­buster-proof super­ma­jor­ity in the Senate.

Worse, where Repub­lic­ans are in control of state govern­ments, they are chan­ging the rules to make it harder to vote, espe­cially for people of color. Repub­lican legis­lat­ors and other Repub­lican offi­cials, includ­ing several governors, assert that these new restric­tions on voting will restore public confid­ence in “elec­tion integ­rity.” Their asser­tion, however, is unsup­por­ted by any evid­ence of elec­tion-secur­ity breaches anywhere in the coun­try suffi­cient to change the national or state-level outcomes of the 2020–21 elec­tions.

Repub­lican-controlled states around the coun­try have enacted vote-suppres­sion laws. Geor­gia enacted legis­la­tion this year that prohib­its and crim­in­al­izes “line-warm­ing”—  the provi­sion of food, water, and blankets by indi­vidu­als or groups to voters stand­ing in line, often for hours, while wait­ing to vote at their polling stations. Flor­id­a’s governor signed new legis­la­tion that also will restrict line-warm­ing. Another repeated focus of these laws is voting on the Sunday before Elec­tion Day. As voting-rights experts and others have noted, the prob­able target of these prohib­i­tions is “Souls to the Polls” — the custom among Black churches to organ­ize congreg­ants to vote together after services.

Repub­lican-controlled states are attack­ing other turnout-boost­ing prac­tices as well. For example, Flor­ida now prohib­its indi­vidu­als or groups from collect­ing mail-in ballots from voters for deliv­ery to drop boxes, mail­boxes, or polling places. Flor­id­a’s legis­la­tion crim­in­al­izes any person’s posses­sion of “two or more mail ballots other than the person’s own ballot and an imme­di­ate family member’s.”

By and large, Repub­lic­ans reject the view that demo­cracy is our consti­tu­tional model of govern­ment. Many have stated for a long time that the United States is a “repub­lic” and not a “demo­cracy.” Of course, this propos­i­tion begs the ques­tion about who should be able to vote, and how easily.

Indeed, it is ironic that Repub­lic­ans appear to believe that constrict­ing the avail­ab­il­ity of the fran­chise serves the small-r repub­lican model of govern­ment. In Feder­al­ist No. 10, where Madison argued for the superi­or­ity of the repub­lican form of repres­ent­at­ive govern­ment (relat­ive to the purely demo­cratic form) to mute the power of faction, he also conten­ded that larger repub­lics were super­ior to smal­ler repub­lics:

The smal­ler the soci­ety, the fewer prob­ably will be the distinct parties and interests compos­ing it; the fewer the distinct parties and interests, the more frequently will a major­ity be found of the same party; and the smal­ler the number of indi­vidu­als compos­ing a major­ity, and the smal­ler the compass within which they are placed, the more easily will they concert and execute their plans of oppres­sion. Extend the sphere, and you take in a greater vari­ety of parties and interests; you make it less prob­able that a major­ity of the whole will have a common motive to invade the rights of other citizens; or if such a common motive exists, it will be more diffi­cult for all who feel it to discover their own strength, and to act in unison with each other. Besides other imped­i­ments, it may be remarked that, where there is a conscious­ness of unjust or dishon­or­able purposes, commu­nic­a­tion is always checked by distrust in propor­tion to the number whose concur­rence is neces­sary. Hence, it clearly appears, that the same advant­age which a repub­lic has over a demo­cracy, in controlling the effects of faction, is enjoyed by a large over a small repub­lic. [emphasis added]

It thus follows that, all else being equal, increas­ing the number of voters would make factions less power­ful. Never­the­less, Repub­lic­ans want to enact new limits on access to the polls, partic­u­larly direc­ted at urban and younger voters and at voters of color, who tend not to support their party. Repub­lic­ans aim to make it harder for these groups to vote, thereby trying to stave off a demo­graphic and ideo­lo­gical chal­lenge to white patri­archy in Amer­ica.

The more Repub­lic­ans restrict voting (by, for example, cutting voting days and hours, elim­in­at­ing voting sites, prohib­it­ing line-warm­ing and ballot collec­tion for home­bound voters and for those lack­ing trans­port­a­tion to distant polling sites), the more they will be able to reduce or deter voter turnout. In this way, they lock in their gains for the long term. In their view, to the victors go the spoils — or, under these new vote-suppres­sion laws, the ballots.

Where do we go from here?

What would it mean if neither impeach­ment itself nor post-impeach­ment elec­tions serve their purposes, oper­at­ing as a check on a pres­id­ent’s auto­cratic impulses (includ­ing efforts to subvert an elec­tion) or as the reset button for govern­ment? What would become of separ­a­tion of powers? Where would the checks and balances be?

Schol­ars some­times describe our extreme polar­iz­a­tion as “tribal” — as if there were an equi­val­ence of the parties. But, signi­fic­antly, only one party is acting to subvert the Consti­tu­tion’s reli­ance on voting to check tyranny. Most congres­sional Demo­crats support new legis­la­tion that will make voting easier and more access­ible, but congres­sional Repub­lic­ans oppose it, and states controlled by Repub­lic­ans pass laws that make voting harder. It is diffi­cult to see how our demo­cracy — or our repub­lic, if one would rather call it that — can survive our current circum­stances: states able to limit voting rights; a U.S. Senate struc­tured so that low-popu­la­tion, predom­in­antly rural and white states have power equal to or greater than higher-popu­la­tion, more urban and more diverse states; and a fili­buster process that puts an even heav­ier anti­demo­cratic thumb on the scale in favor of less popu­lous, more rural, and demo­graph­ic­ally whiter states.

We should not, of course, blame the Founders for believ­ing that insti­tu­tional checks and balances could mute and contain faction­al­ism, because the new repub­lic they were creat­ing was unpre­ced­en­ted. Their only model was Britain’s consti­tu­tional monarchy, where the work­ings of Parlia­ment seemed to show that small and fluid factions had until then always been more import­ant than the two big nominal parties, Whig and Tory.

And, despite the Founders’ fail­ure to elim­in­ate faction (what we moderns might call “party”), the Amer­ican consti­tu­tional system has done a relat­ively good job over its 230-year history in hold­ing our contin­ent-sized, hyper-diverse repub­lic together.

The system has failed cata­stroph­ic­ally only once, over the original national sin of consti­tu­tion­ally protect­ing slavery, which even­tu­ally ignited the Civil War. Slavery repres­en­ted the most morally repug­nant mani­fest­a­tion of the under­ly­ing mega-issue of Amer­ican history and soci­ety, race, and racism. So it is not surpris­ing that the current impasse is also ulti­mately about race.

This impasse is exacer­bated by two other factors. The first has until now been a manage­able weak­ness (some might even call it a strength) of the Amer­ican system: our district-by-district, winner-take-all elect­oral system, which ensures there can be only two polit­ical parties (rather than the supple multi­pli­city of parties of elect­oral systems based on propor­tional repres­ent­a­tion). The danger here is obvi­ous: if polit­ics becomes irre­vers­ibly polar­ized, the pres­ence of only two parties raises the odds that the compet­i­tion will become fero­ciously zero-sum in nature, with no shift­ing coali­tions to moder­ate the combat. Still, the current system broke down only once, despite the ever-present powder keg of systemic racism coupled with a winner-take-all elect­oral system.

Alas, a second and newer factor now ignites a match near that powder keg: elec­tion engin­eer­ing, via gerry­man­der­ing, which has become ever worse over the past 50 years, allow­ing incum­bents to win reelec­tion upwards of 90 percent of the time. Vote-suppres­sion laws aggrav­ate the partisan effects of gerry­man­der­ing. Most lawmakers are thus beholden to their own party’s elect­oral base, not to voters in general, making it harder for members of Congress to impeach a pres­id­ent of their own party, espe­cially if that pres­id­ent, like Trump, enjoys the fanat­ical support of the party base.

Even though Trump lost the elec­tion, the impasse will not fade any time soon. The GOP remains in his thrall and in 2024 will likely renom­in­ate Trump, or someone like him, to run for pres­id­ent. We do not know where Amer­ica’s current impasse will lead or how it will end, but our consti­tu­tional exper­i­ment is at risk of fail­ing.

Former President Trump holds up a copy of the Washington Post that reads "Trump acquitted" Drew Angerer/Getty
U.S. Pres­id­ent Donald Trump holds up a Febru­ary 6, 2020 copy of the Wash­ing­ton Post as he speaks in the East Room of the White House follow­ing his acquit­tal by the U.S. Senate on two articles of impeach­ment.


Arising as they do from centur­ies-long chains of events and decisions, the recent fail­ures of checks and balances in the impeach­ment process and else­where in the federal govern­ment aren’t amen­able to a simple solu­tion. Our national situ­ation ensues from multiple causes; it isn’t a Gord­ian knot that can be solved with one stroke. What’s more, the descrip­tion above of our situ­ation does­n’t even consider related public prob­lems exem­pli­fy­ing and exacer­bated by polar­iz­a­tion, such as income inequal­ity, climate change, and the response to the pandemic. Indeed, because of the struc­tural issues discussed above (a funda­ment­ally unrep­res­ent­at­ive Senate and a gerry­mandered House), Congress is far more polar­ized than the public.

Certain conser­vat­ives argue that our consti­tu­tional system creates fertile ground for a tyranny of the major­ity. That’s a straw man invoked to suppress voting and protect the fili­buster — devices that enable minor­it­arian tyranny and flout the purposes and text of the 15th and 19th Amend­ments, among other meas­ures. There­fore, we present here consti­tu­tion­ally and circum­stan­tially groun­ded sugges­tions for enabling elec­tions to reduce the risk of tyranny.

Owing to the struc­ture of the Senate laid out in Article II as well as demo­graphic trends, smal­ler, less popu­lous, and more rural states are and will remain numer­ous enough in the Senate for the fore­see­able future to prevent major­it­arian oppres­sion even without the fili­buster — as is evid­ent right now within and across the two major parties as they consider pending legis­la­tion. So the Senate first should end the fili­buster now by re-empower­ing itself to cut off debate by a simple major­ity in any legis­lat­ive matter. Without this vital change, we will not see any of the modi­fic­a­tions to our elect­oral system needed to save our demo­cracy.

To avoid another lawless pres­id­ent, we also must recog­nize that the Founders’ system of checks and balances — which includes impeach­ment — can func­tion only if our elec­tions do. For that reason, we think that other near-term remed­ies must focus on our elec­tion processes and laws.

To protect the consti­tu­tion­ally contem­plated power of an ever-fluid major­ity in craft­ing public policy (and in support of Madis­on’s pref­er­ence for a larger repub­lic in lieu of a smal­ler one), Congress should pass the John R. Lewis Voting Rights Advance­ment Act to restore preclear­ance author­ity over states’ voting proced­ures to the federal govern­ment. Congress also should pass the Free­dom to Vote Act, which would help make voting easier, less resource intens­ive, and avail­able to more eligible voters and also address the bane of gerry­man­der­ing.

After these recom­mend­a­tions concern­ing the fili­buster and new federal elec­tion legis­la­tion are imple­men­ted, voting results will reflect more accur­ately the will of the major­ity, subject to restraint by and accom­mod­a­tion of the minor­ity. From issue to issue, major­it­ies and minor­it­ies will evolve and mutate, because issues are complex, solu­tions have diverse impacts, and voters have cross­cut­ting interests. The Consti­tu­tion, however, plainly estab­lishes the will of the major­ity, with expli­cit minor­it­arian and judi­cial restraints, as the ulti­mate engine of public policy. It’s long past time to effec­tu­ate that process. This newly revital­ized major­it­arian system would bolster checks and balances, and elec­tions would reli­ably be a check on future tyranny.


Caroline Fredrick­son, senior fellow at the Bren­nan Center and distin­guished visitor from prac­tice at Geor­getown Law, served as the pres­id­ent of the Amer­ican Consti­tu­tion Soci­ety from 2009 to 2019. Before join­ing ACS, Fredrick­son served as the director of the ACLU’s Wash­ing­ton Legis­lat­ive Office and as general coun­sel and legal director of NARAL Pro-Choice Amer­ica. She also served in senior posi­tions on Capitol Hill and in the White House. She has published on many legal and consti­tu­tional issues and is a frequent guest on tele­vi­sion and radio. She is also the author of Under the Bus: How Work­ing Women Are Being Run OverThe Demo­cracy Fix: How to Win the Fight for Fair Rules, Fair Courts, and Fair Elec­tions; and most recently The AOC Way. Fredrick­son received her JD from Columbia Law School with honors and her BA from Yale Univer­sity. She clerked for Hon. James L. Oakes of the United States Court of Appeals for the Second Circuit.

Alan Neff most recently served as coed­itor of The Rule of Law This Week for the Amer­ican Consti­tu­tion Soci­ety. Previ­ously he was a senior corpor­a­tion coun­sel for the City of Chicago, where he concen­trated his prac­tice in complex civil litig­a­tion. Over the course of his career in academe and prac­tice, Neff has researched and writ­ten on a vari­ety of subjects: federal judi­cial selec­tion, inter­na­tional envir­on­mental protec­tion, libel law, the Mueller invest­ig­a­tion, and partisan gerry­man­der­ing. His work has appeared in the New York TimesWash­ing­ton PostChicago TribuneEcology Law QuarterlyNorth­west­ern Journal of Inter­na­tional Law and Busi­ness, and the Amer­ican Consti­tu­tion Soci­ety’s Expert Forum blog. He is a gradu­ate of the Univer­sity of Michigan, from which he received BA, JD, and MS degrees.