How Is This Supposed to Work?: The Fundamental Ways America Gets Democracy Wrong and What to Do About It

6 Nov

I. What’s not wrong with American politics?

The constant calls of the permanent Washington establishment for centrist governance lead us to ask, ‘What is the Center, and what’s so great about it?’ Given centrism’s self-professed lack of ideology, there isn’t any real ‘there’ there to speak of in positive terms, so we instead most simply define centrism by what it’s supposedly not, neither ‘extremist’ nor ‘partisan’:

  • ‘Extremism’ comes in two forms. The first is simply a radical policy agenda, e.g. pushing for the invasion of Canada. The second form is willingness to push political norms and processes to the brink to enact a favored agenda, whatever that agenda may be. These two forms often coincide, but not always: pushers of unremarkable policy prescriptions may still be extremist if they simply refuse to abide losing.
  • ‘Partisanship’ simply refers to pursuing a political agenda of advantage—advantage of one’s own ideology, one’s own party, or one’s own self-interest—as opposed to pursuing a political agenda for the public interest.

Defined like this, it’s hard to see how anyone could object to centrism… but at the same time hard to see the usefulness of the term as anything but self-flattery, as in, ‘I’m centrist, you’re not.’  Virtually no one, short of the Unabomber, recognizes themselves as employing extremist means, and absolutely no one recognizes their own agenda as extremist in the sense of ‘not sensible’. Likewise, even when partisans recognize the unpopularity of their ideas, they almost always assume the accordance of their ideas with the true desires of the electorate: while the true democratic will reflects the interests of the electorate, the electorate, these argument go, may not always know what best serves those interests. It works out that very few political actors nakedly admit, even to themselves, the selfishness of their agenda.

But of course, what does it matter if extremists and partisans don’t recognize themselves? Don’t the labels still have meaning? Unfortunately, recent American political experience has shown that the answer is ‘no, not in practice’. Partisan, today, is used simply as a label for any policy disfavored by the centrist establishment, which is the centrist way of claiming popular legitimacy: ‘Everyone else is out for themselves; only we centrists represent the democratic will.’ Extremist has been made just as meaningless, as it’s an inherently relative term, and a political elite fixated on what it is not is easily gamed: the further extremists go in one direction, the further a centrist-über-alles ideology must move along with the extremists, and by reflexively defining itself as the mid-point of two extremes, centrism has actually enabled the extremism it claims to abhor.

‘This is a right-leaning country’

So why has the shift happened all in one direction? If tactical extremism works, why has only one side successfully exploited this loophole? Doesn’t this prove what Washington centrists and the extremists they enable always claim, that the rightward slant of American policy in the last thirty years reflects the natural right-of-center tilt of the American electorate? One possible counter-explanation is that the bulk of the liberal-minded American electorate lacks the same radical impulses as the conservative base, but this is a wildly difficult sort of claim to prove, let alone explain (even though I and most liberals suspect it’s true). In any case, a more important explanation lies not in the contrasting nature of the two sides but in the nature of the centrist establishment playing referee:

Consider that, in the same thirty-year period dominated by the agenda of the American right, social liberalism has quite steadily advanced, most notably on issues of sexuality. What allowed this to happen?

  • First, social change moves forward mainly by cultural force, not political force, and so, recognizing this, Washington centrists feel free to sacrifice the political agenda of their leftist social leanings in the confidence that the whole domain will pretty much take care of itself in time.
  • Second, establishment centrists tend to buy into the conservative narrative of their own social liberalism as an unfair conspiracy against the American right (rather than the product of the natural urban/rural cultural divide seen in virtually every country for the last three-hundred years). This leads centrist media figures to treat conservatism with kids’-gloves as a kind of self-hating apology for their own social liberalism.1
  • Third, social-issue lobbies on both sides pale in funding next to real monied interests, and few big pay checks stem from these social-issue lobbying efforts. In contrast, when Goldman Sachs lobbies congress on financial policy, there’s not only sizable payouts at stake for lobbying firms and members of congress but also far larger sums at stake in the very effects of the policies themselves.
  • Fourth, while the social policies favored by Washington centrists don’t really need establishment support in the long run, the economic and (to a lesser extent) foreign policy agenda favored by the center is susceptible to stalling and even rollback in the face of sustained left-of-center political victories. Consequently, the centrist establishment is much more concerned with protecting its economic and international agendas than its social agenda.
  • Fifth, even though most Washington centrists espouse a much less radical form of rightwing economic policy than the typical Republican congressman, centrists still feel most comfortable with a party that supports their basic economic narrative, even in a dangerously radical form.

This last point requires some explanation. Despite protestations to the contrary, ‘identify politics’ is not the exclusive province of the left. Everyone has an identity, and one’s political identity is largely tied up in a grand narrative of what’s wrong with the world. Whom we choose as villains and heroes (or victims) in our narrative both explains and determines most of our policy positions. Whether you find corporations heroic or villainous determines your position on business regulations of all kinds. Whether you find the poor heroic or villainous determines your position on food stamps. Etc.2 Of course, our designations of hero and villain status must come from somewhere and so are presumably informed by consideration and experience. Once well-formed, however, opinions of hated/loved figures and hated/loved classes of people very rarely get reconsidered, regardless of any new evidence or argument. Unsurprisingly, no one ever casts themselves as villain in their own narrative, and in fact, most people strain to confer upon themselves hero status: the grand narrative of the world we tell ourselves is almost always a story of self-validation.

For establishment centrists, even those inclined towards policy wonkishness (a liberal proclivity), self-validation requires guarding against “excessive” indictments of the economic elite with whom they identify as successful elites themselves. ‘Sure, maybe business sometimes engages in excess or makes mistakes, but business can only go so wrong and only for so long; any theory positing broad dysfunction or wrong-doing simply can’t be true.’ Consider the response to the 2008 financial crisis and following economic decline. Despite broad, far-reaching evidence of catastrophic failure on the part of financial and political elites over many years (at least back to the late 90′s), the centrist narrative at best minimizes such explanations. The real villains, according to centrists, are the common people, who borrowed too much money, demanded too much in government entitlements, and didn’t do well enough in school to give us the work force we need to sustain reasonable employment levels. If only more Americans were like our super-awesome elites, we wouldn’t be having these problems. Though entirely unhelpful for fixing our current predicament (and in fact dangerously counterproductive), this assessment of the economy holds sway because it protects the establishment centrists who got us here from culpability.3 Given the choice between truly moderate economic policy and face-saving right-wing narratives, Washington centrists will go for self-validation every time.4

Reasonableness won’t save us

So if the establishment centrists we have don’t really act as the impartial moderates they claim to be, who or what could possibly save our politics? Well oddly enough, political reform is as easy as actually changing our political system. Insane, I know, but bear with me. Rather than just brow-beating voters and political actors into being nice to each other, actual reform would entail:

  • rethinking how we select office holders
  • rethinking the powers of our political offices

But if we’re going to rethink such basic rules, we should first re-examine the fundamental principles and requirements of a democracy.

II. What is democracy?

If democracy were simple, votes could settle all differences of opinion, but a little pondering quickly reveals the naivite of this grade-school-civics conception of democracy.

First, simply divining the will of the majority is not easy:

  • Questions to vote upon must somehow be selected from the domain of all possible questions.
  • Not all questions fit into an either-or mold, meaning the desires of the public may not be best gauged by a simple plurality vote, e.g. given three options, the one which carries a plurality may leave the majority stuck with their least-favored option. Such scenarios imply that our intuitions of political justice require not a simple notion of ‘majority will’ but rather a more complicated notion of ‘democratic will’.

Even more problematic is deciding when, exactly, the democratic will should not carry the day: every democracy restrains the democratic will in certain ways; the only question is how. Democracies typically enact such restraints for three purposes:

  • To safe-guard the democratic order from civil strife.
  • To safe-guard the democratic order against a slide into corruption and autocracy.
  • To safe-guard minority rights.

Freedom of speech, for example, is a restraint on the democratic will which is justified on all three grounds: as a protection of electoral competition, as a check against government power, and as a protection for religious and political minorities.

So you could say our conception of political justice is, in a way, strangely two-faced: we believe that the policy options favored by the most people should naturally carry the day—except where we think they shouldn’t. We deem certain matters of policy not subject to the will of the majority regardless of any kind of vote.

The problem, of course, is that we disagree about what exactly these special policies ought to be, so effectively, in trying to devise a system for settling our disputes, we end up stuck having to settle many of our most contentious issues. For example, many disputes in modern America hinge upon differing claims of minority rights, e.g. liberals often insist that broad rights of privacy and rights of equal protection ought cover abortion and same-sex marriage while conservatives claim that the minority right claims of taxpayers (especially the rich) and industry ought constrain progressive taxation and the very scope of government. These claims hold, respectively, that freedom to get an abortion or freedom from class-targeted taxation must never be violated by government no matter what the electorate wants.5

The third major complication of democracy lies in setting the bounds on the polity itself: we all agree that the people should rule themselves—that democracy is superior to autocracy—but we don’t have a settled method for deciding which people. Yes, we should decide things by vote, but who gets to vote?

The two principled answers to this question swing in polar opposite directions: the globalist quickly concludes that no one should be excluded from the polity (nor escape its reach), so obviously the only correct polity is a one-world government; the libertarian, believing that one man is an island unto himself and thereby only ever bound by contract, quickly concludes that the only legitimate polities are those freely entered into by all members.

Neither of these principles work. The globalist principle is simply not viable in this century (and maybe not for many centuries or perhaps not ever), and the contractually-minded principle of libertarians seems to imply that I should be free to not only choose my own citizenship6 but also draw the polity as tightly around myself as I choose, e.g. declaring my half of my duplex an independent nation. If I consistently don’t like the way you vote against my preferences, why am I obligated to continue voting with you? Not only did I not ask to be born, I certainly didn’t ask that you be born in relatively close proximity, so what claims do you have on diluting my political say? If we buy into these notions of allowing people to freely pick-and-choose their own government to maximize their own preferences, the whole idea of government becomes unrecognizable if not meaningless.

So when it comes to settling bounds on the polity, the only viable known solution is practical, not principled: we simply accept as given the historical accidents that make up the national groups we’ve inherited and call it close enough. It’s not a satisfying answer—and it permanently leaves open the door for future conflicts—but it’s all we have.

III. What’s wrong with the status quo?

In some of these basic matters, I don’t think the American system could be realistically improved to the satisfaction of a stable majority of Americans. Sure, I have my preferences about individual and class rights and the bounds of the polity (i.e. who’s rightfully American and who’s not), but imposing my preferences would be just improving the world as I see it when the goal here is just to improve the political order. For example, though I’m mostly indifferent about the issue of gun control, I find the second amendment at best anachronistic and so would prefer omitting it from a revised constitution: I think gun control is a perfectly legitimate realm of public policy-making for elected officials to enact the will of the electorate, whether that means banning all firearms or removing all prohibitions. The exercise here, though, is to devise a better system of making policy in the face of our disagreements, not to resettle all of our most contentious debates. So because the questions of individual rights, class rights, and the bounds of the polity are delicate matters of long-standing dispute, we’ll leave those parts of the status quo basically untouched. We’ll even preserve some not-so-great ideas in the constitution, like federalism.

The only parts of the constitution we intend to meddle with are elections and the powers of elected officials, and we only do so for the purpose of creating a politics that delivers policy outcomes that better reflect the democratic will. To this end, we intend to create a politics that better guards against the corrupting influence of economic power, focuses political discourse on actual policy, and defuses ideological hysteria.

First, though, we must make a case for the inadequacy of the status quo in these regards. After all, if it ain’t broke, what’s to fix? So here’s what’s wrong:

‘We all want the same thing’

Perhaps the most ridiculous platitude of American politics is that ‘we all want the same thing.’ Nope! The whole point of democracy is that we have disagreements, and that’s why we have elections—to set policy in the face of those disagreements. No election and legislative process—no matter how cordial—can magically make the disagreements go away, and in every dispute, somebody loses in one way or another because that’s the only way it could possibly be.

What politicians and pundits really mean when they persist in using the ‘we all want the same thing’ bromide is that ‘we only disagree about means to the same ends’, that we all share core values. But this is only true in a uselessly broad sense: we all want physical and economic security, sure, but we have vastly different narratives explaining why the world lacks this security and how best to remedy this state of affairs; in particular, we disagree about who is responsible for this fallen state and who deserves relief.

Bargaining vs. alteration 

If we don’t all want the same thing, how can we get anything done by agreement? In theory, both sides meet halfway, getting only half of what they want. This bargaining between political adversaries, however, is unlike any other contentious bargaining scenario, such as business deals, legal settlements, plea bargains, or peace treaties. For adversarial parties to reach a compromise, both sides must get something out of the deal, whether a positive good or an avoidance of a negative (or some mix thereof). What, then, do two rivalrous political parties get out of compromising with each other? While the majority most prefers acting unilaterally whenever possible to maximize its position, the minority most prefers winning the next election. In theory, supermajority requirements and divided government force the two sides to compromise, but in practice, electoral incentives disincline the minority from giving the majority ‘victories’ of any kind, leading to perpetual inaction. As it stands, legislation of any import in the US can only possibly get passed when the planets align with one party controlling the presidency, the house, and the senate (by at least a 20-vote margin). Increasingly, these extraordinary circumstances are even needed to effectively conduct merely routine business, like passing a yearly budget.7

Democracies do not have to work this way, and in fact, the United States is unique among modern democracies in its reliance upon bargaining between political adversaries. The political systems of most other countries have a radical notion that the winners of elections should actually get to do things. Rather than bargaining, democracy can work upon a model of alteration: the winning party gets to enact the laws it wants while the losing party sits on the sidelines and complains, both sides all the while positioning themselves for the next election. The very idea may sound alarming: wouldn’t the winning party do all the crazy things it’s always wanted to do? Actual experience in other countries, however, shows that winning parties usually exercise their freedom of action with considerable caution in the interest of preserving their electoral viability. Enacting whatever laws you want regardless of your campaign promises and public opinion is a sure way to end up ousted from office and your policies overturned. The end effect is that simply having a different legislative process leads to different kinds of politicians getting elected.8

Underlying our system of bargaining and interlocking veto points is an implied premise: the worst thing in the world is your political opponents getting something they want. This fear Americans have of their rival party’s victories is actually not unreasonable, for in the American experience, any passed legislation becomes a virtually impossible-to-dislodge fait accomplit. However, the corollary of making new laws difficult to create is that old laws become equally difficult to replace or repeal. Making new laws easy to create would make their replacement or repeal just as easy.

Consider national healthcare in the United States. Under our system of bargaining, the electorate has felt for decades that something must be done about healthcare, but it wasn’t until 2010 that the planets aligned for a legislative initiative even remotely substantial enough to plausibly address the problem. Part of what explained this impasse in the face of consistent voter displeasure was the lack of large enough majority support for any particular solution. But what would have likely happened under a system of governing by alteration? Wouldn’t ideologues have caused radical policy instability as governance changed hands between right and left? This is actually very unlikely, as the electorate is cautious by habit about major policy changes, so ideologues intent on erecting huge new initiatives or tearing down existing ones would simply not get elected. Large policy initiatives would rarely get adopted whole in the first place, policies instead starting out as small programs that expand or contract as their popularity fluctuates. The most likely scenario, then, is that some kind of modest national healthcare would have been enacted in the Truman years, and this initial design would have likely gotten tweaked (or even fully superseded) multiple times over the years, leaving us with healthcare policy more-or-less satisfactory to a stable majority of Americans, both left-of-center and right-of-center. Of course we’d have plenty of fights over healthcare, but the points of contention would be relatively inconsequential details. Unlike in our current screwy arrangement, this major problem wouldn’t have festered for decades only to get addressed in one massive all-or-nothing attempt. By allowing for easy iteration, governing by alteration can develop more stable, popular, and effective policies.

All of this hinges on the ability of the electorate to accurately appraise which policies achieve their aims and which parties support which policies. The more complicated the legislative process, the less clarity voters have of who is actually responsible for which policies, what effects those policies have, and even what policies are currently law9. By making elections and legislating into an 11-dimensional chess game, the common voter is disempowered simply by their inability to spectate the game of politics, let alone effectively act within it. This leaves the door open for professional lobbyists to insert themselves with outsized influence, for while the average voter can’t follow bills through congressional committees and win access to elected officials, paid professionals can. (It’s also this complicated political meta-game that gives the media endless excuses to waste time discussing things other than actual policy.)

Fortunately, simplicity is another area where a system of alteration has the advantage: while legislating by bargaining across (or even within) party lines makes for endlessly entertaining political prognosticating, it’s naturally slow, cumbersome, and fraught with intrigue, giving the insider track to monied interests and the status quo. Legislating by alteration puts the vast bulk of the decision making into the election process itself, where voters are paying the most attention and have the most direct influence.

Whom do we want to govern us?

American elected officials most often start their political careers by trading on the status of their occupation to run for an office commensurate with that position, and from there, they often trade up for higher office. For example, Barrack Obama ran for state senate on his biography as a law professor and community activist, later trading up for the U.S. Senate and then the presidency. The campaigns of the wealthy generally follow this same logic, but the advantage of a personal fortune to self-finance their campaigns often allows them to skip tenure in lower offices. Mitt Romney, for instance, traded a chunk of his fortune for a straight run at the Massachusetts governorship, which he since has tried to trade up for the presidency.

Then there’s the campaign process itself, which requires coordination, professional consultation, and serious advertising at all but the lowest levels, things which require money. For the non-wealthy, getting this money requires a patronage network, the development and maintenance of which constitutes a near full-time job throughout one’s political career, even before ever landing office.

In practice, then, it’s nearly impossible to elect anyone but “career politicians” despite the public’s consistently stated distrust of the motives of careerists, and it’s trust which concerns voters most about candidates above all else10. Will this candidate hold to their stated positions once in office? Will the candidate abuse their office for personal gain? Is the candidate just telling us what we want to hear? To many Americans, the lying of politicians is the primary dysfunction of our politics.

Well if elected officials can’t be trusted, why not simply take them out of the loop by adopting a direct democracy? This is the only way to totally bridge the gap between the policies for which the electorate votes and the policies which the electorate actually gets. Most voters, though, have neither the time nor interest to immerse themselves in policy details, so the only way a large-scale direct democracy could possibly work is if most voters make their decisions via signaling, voting on the advice of a trusted party. Even allowing for this, it’s still unclear how, in a direct democracy, you’d decide which legislation with which precise language gets put to a vote. There are very good reasons, then, to have a representative democracy, and besides, we needn’t eliminate the gap between voter preferences and enacted law, only minimize it. So the question to ask is, ‘Why is the gap between expressed voter policy preferences and delivered policy so great in our system?’

Policy gaps and broken promises

Actually, a few good arguments suggest that the gap is not nearly as large as most American voters perceive. For example, when liberals vote for Barrack Obama as president, his failure to enact across the board liberal policies probably mostly reflects the constraints imposed on his office by the existence of Congress. Who knows what agenda Obama would adopt if he could just follow his electoral mandate and his personal convictions? In our system of bargaining between divided government, it’s a moot question: congress members and even the president are all constrained by each other; while Obama was elected, the congress members who constrain him were just as much elected, and our system expects the true democratic will to lie somewhere in between.

Unfortunately, the voters themselves forget that this is actually how our system is supposed to work. Voters tend to reflexively blame incumbents of all parties and offices—but the president especially—for the morass of the status quo without making any distinction between incumbents’ favored policies, the actually enacted policies, and the outcomes which result. Consequently, the balance between left and right in today’s politics has devolved into an incoherent, reflexive disaffection that has broken the connection between successful or disastrous policy and electoral success or failure.

Closing the gap between the policies voters choose and the policies they actually get would mitigate this disconnect, but not perfectly: while voters would suffer much less confusion over who enacted what policies, the causal link between policies and outcomes would remain debatable: did these tax cuts raise employment? Did these drug laws reduce crime? Etc. So the accountability for elected officials would not become perfect, but at least questions of policy effects focus the debate on actual policy rather than the political meta-game, and hopefully, a more direct connection between elections and policies would motivate more voters to change their positions when their favored policies get enacted but fail to deliver. (Of course, the power of ego and ideological investment being what they are, mea culpa‘s would probably remain uncommon, even in the face of dramatic evidence.11)

Which brings us back to trust: we could empower politicians to actually enact the policies they run on, but how could we possibly trust them to stick to their professed agendas? This concern is not unfounded, as the American experience is that elected officials constantly go back on their campaign promises. But why? If a politician gets elected running on an agenda, why not just stick to that agenda?

Well to a large extent, it’s the bargaining process of divided government that causes politicians to break their word and that creates the gaps between voter expectations and enacted policies. Partly these gaps reflect the true democratic will, for as we just mentioned, one president or one congress member isn’t supposed to get everything which the people who voted for them wanted. Other times, these gaps result from the conflicting desires of those who voted for a candidate, e.g. while some Obama voters may favor a policy, other Obama voters may oppose that very same policy. Such gaps are troublesome but at least democratically legitimate.

Another source of these gaps, however, is the patronage networks upon which elected officials rely, not just for obtaining and keeping office, but also for getting anything done once in office. To win party nominations, win office, and then wield influence once elected, politicians rely upon their patron donors and interest groups to curry favor with other politicians (both from the other party and from within their own party). Without this support, no office holder—whether the president or a backbencher in congress—can influence legislation much at all. Interest group lobbies effectively operate as the glue that holds political coalitions together and allows the bargaining process to make policy.

Some would argue this is all just representative democracy at work, but when politicians, including even the president, can’t deliver the policies voters elect them for (let alone deliver the hoped for outcomes), voters unsurprisingly get disillusioned by politics. The sum effect is a system that operates under enormous biases towards the status quo, which not coincidentally favors incumbent economic interests. Every few elections, voters convince themselves they’ve found the candidate who’s different, the ‘authentic’ candidate who will ‘work hard’ and ‘fight’ for them in office, and when the elected politician inevitably disappoints, the voters convince themselves that the politician simply mustn’t have been the ‘real deal’. If voters identify any systemic failure at all, they at best vaguely attribute it to the ‘undue influence of money’. Foolishly, we hardly ever consider the possibility that the very structure of our government and elections are at fault.

IV. How should this thing work?

So here is how we could structure our political offices and elections around a model of alteration rather than bargaining. (If some parts strike you as odd, keep in mind I give a more detailed defense afterwards, and some mechanisms I only explain in full in an addendum.) 

The presidency

We needn’t move to a parliamentary-style full integration of legislature and executive, but the two should operate more independently rather than block each other at every turn. This requires simply three tweaks:

  • The president should no longer have a veto over Congress.
  • The president should no longer require congressional approval for administrative appointments.
  • The president and congress should make judicial appointments by adding names to a hat and drawing.

This judicial appointment scheme would work by allotting each president and congress a number of names they may add to a list of nominees from which judges are then selected at random. This would allow both branches to give their input without the gridlock and charade of appointment hearings.


While the presidency sans veto can go otherwise untouched, Congress should get totally restructured. First, Congress should become unicameral with just one member for each state. Even more radical, all seats should always be held by members of the same party, and each party should stand for national election to Congress as a 50-candidate slate, winning either everything or nothing.

Because the members would all get elected as part of the same slate and so presumably mostly agree with each other, they needn‘t legislate by congressional vote. Rather, within the year-long congressional term, three congress members would get selected by random to each serve four-month terms as Speaker. The Speaker alone would wield unilateral power to enact any legislation, subject only to veto by a supermajority of the congress (for now, let’s say a two-thirds supermajority). Legislation put forward by a Speaker would go through a process of public disclosure and debate but then pass at the end of the Speaker’s term.

The Speaker is meant to seek consensus of the members, a scheme that works fine as long as party members mostly agree about their agenda (as they normally should, given the election process). Presumably voters don’t like incumbent parties that squabble, so congress members would have a strong incentive to negotiate their disagreements informally in private.

For emergency purposes, legislation could also get passed at any point in the congressional term with a supermajority vote.


To give smaller states a disproportional say as they have in today’s Senate and electoral college, votes from smaller states could simply get tallied with a bonus modifier, e.g. one vote from the smallest state would count X times as much as one vote from the largest (the precise value of X depending upon a yet to be specified formula; also, the modifier for the presidential vote count arguably should differ from that for the congressional vote count).

Congress and the president would serve only one-year terms, the president serving up to four consecutive terms and a maximum of eight terms over his or her lifetime. In their first re-elections, the incumbent congress and president would be given small bonus modifiers, i.e. the incumbent placing a close second in the election would still win. This special allowance would mitigate overly erratic turnover while still allowing the electorate the opportunity to turn out a strongly disliked incumbent.

The congress and president would both be elected by IRV (instant-runoff vote) from a field of 7 candidates (including the incumbent standing for re-election). The selection of the 7 congressional caucuses and 7 presidential candidates (aside from the incumbents) would be determined in two preceding primaries held simultaneously with the previous two elections.

In the first of these primaries, forty or fifty caucus leaders get nationally elected by STV (single transferrable vote) from an open field of potentially thousands of candidates. These caucus leaders then select congressional candidates from each state. The second primary would then whittle the field of congressional caucuses down to 7, again by STV.

An individual congressional candidate may simultaneously stand for election in more than one caucus, and in fact, while a congressional candidate chosen by a caucus leader could of course refuse to run, if they consent to run on any one ticket, they must then consent to run on any other ticket. (This would allow lesser known caucus leaders to remain competitive with ideologically similar competitors by poaching recognizable names.)

Caucus leaders would be allowed to withdraw from the race before the second primary to consolidate support for another candidate with a similar agenda. (Of course, in an STV election, withdrawal is a less useful tactical maneuver.)

Before the actual election, each caucus would vote amongst themselves to determine a party agenda to run on and select a presidential candidate. The caucus can eject and replace one of its candidates by a majority vote. The replacement candidate is then selected by the caucus leader. (The caucus leader actually participates in all votes and is given more than one vote—let’s say for now 15 votes.) The caucus leader of the winning party, though not a member of congress, would retain this special role with the caucus as long as it remains in office.


To deter abuse of office, congress could enact bills of ostracism to charge and sentence current or former presidents or members of congress. These charges and sentences would require approval by supermajority vote of the electorate in the next election. Congress could also issue bills of pardon or commutation without need for voter approval. What constitutes abuse of office should be left up to the discretion of congress: it could be something as simple as governing in a manner egregiously inconsistent with the platform upon which you were elected, or even just bad budgeting decisions. The sentences prescribed by the bills may also vary widely as fits the crime in the eyes of Congress and the electorate (short of the death penalty). The punishment for most abuses should be monetary fines or short terms (to be served under house arrest or in a specially-built prison for just these offenders—think Elba Island, not San Quentin).

V. How is this better?

I won’t say I’m married to every particular above, and in fact I entertain the possibility that a totally different design might function better. Still, I believe that what I laid out stands up against some obvious criticisms:

Isn’t this one-party rule?

No, this is one-party-at-a-time rule, and the ruling party cannot make constitutional changes without approval of ¾‘ths of the states. Besides, the ruling party in Congress is not necessarily the ruling party in the White House, and this proposed system discourages long tenures in power.

Why the three-phase election?

Given an unbounded set of candidates, it’s difficult for the electorate to express a coherent choice, and likewise getting lost in the crowd makes it difficult for candidates to calibrate their campaign agenda to suit a sizeable plurality of the electorate. Our current primary system narrows down the choices in a way that gives party insiders the most say by the simple fact that most rank-and-file voters can’t name plausible party nominees off the top of their head (especially for down-ticket offices). Our primaries also tend to lock the broader electorate into narrow choices too early, e.g. while Democratic voters may prefer one Democrat, the broader electorate may favor a Democrat who fails to win the Democratic nomination.

Why narrow it down to 7 candidates?

The actual election will get the most participation and therefore the least informed voters. For the sake of casual observers of politics, seven is not too many, not too few, and allows for an equal number of left and right candidates (not counting the 7th candidate, the incumbent).12

Why narrow it down to 40 candidates in the first primary?

The first primary relies upon the most informed voters making a choice among a potentially very large set of nominees, something which most voters just won’t care much to do. Most people simply don’t have political preferences that are all that particular.

But why not just skip ahead to picking seven? Going directly to seven would overly favor established names and overly compress the full election cycle. Forty candidates is a small enough number to be manageable but large enough to allow for diversity among several ideological poles.

Voters go into the booth to express a preference for what they want now, so why should they simultaneously express a preference about what they want one and two years down the line?

True, this is certainly odd. The motivation here is mainly practical: a truly consequential election should be held every year, but we need some preliminary round(s) to narrow the choices down, and holding three elections within a year would be burdensome for voters and wouldn’t provide the candidates sufficient time to coalesce around a campaign agenda. Besides, the media and a segment of the electorate always like to look ahead a few years, and having unknowns appear upon the scene and take office within months would be too chaotic and disconcerting.

Why term limits? Don’t we need experienced legislators?

The need for experienced legislators is a good reason to oppose term limits under our current regime. In the system I propose, however, there aren’t really any obstacles to navigate to pass legislation, and consequently, there isn’t any real need for what we currently think of as legislative skill: all the legislative bargaining would happen entirely within very ideologically coherent parties, and the large bulk of that bargaining would happen before the election is even held. Given this difference, and given the greater power afforded Congress, I feel comfortable imposing term limits as an extra precaution against any party or individual capturing too much power for too long.

Isn’t this voting for parties over people?

If a political agenda is coherent, it should be able to stand independently of any one person’s personality. Besides, obsession with personalities and “character” is one of the most toxic elements of our current politics. While some amount of concern with the personal biographies of political figures is probably inevitable (and arguably even desirable), the system I propose should greatly deemphasize the role of such concerns: the shorter election cycle and existence of many more candidates (fifty per slate) should make it impractical for the media to focus on any one candidate’s biography.

Why ostracism? Won’t it lead to a cycle of political reprisals?

The prospect of getting turned out in the next election is not always sufficient accountability for high-office holders, especially when it comes to policies that can’t easily be reversed, e.g. starting a war or spending a lot of money. The mere threat of prosecution doesn’t work for such cases because the crimes are at root political in nature, namely thwarting the will of the electorate, which is inherently a political determination, not something to be proven in court. If the electorate wants to see former presidents or congress members punished on a case-by-case basis, I don’t see why the electorate shouldn’t get its way. Sure, some politicians might end up unfairly treated, but with great power comes great responsibility. No one’s forcing these people to wield the highest power in the land.

Now, the real concern here is that an outgoing regime, while rightly fearing rollback of its policy agenda, shouldn’t fear personal reprisals against its members. As we’ve seen in many other countries, such fear contributes to political instability and the eventual dissolution of the whole democratic process. But consider that, in such examples, unjust reprisals are generally delivered by corruption of the legal process, so it’s not the lack of a formal punishment mechanism that protects against reprisals. What really allows elected governments in unstable democracies to abuse their predecessors is general corruption of the whole political system. If the U.S. were to succumb to the pervasive corruption endemic in developing countries, we’d have bigger problems than abuse of a formal mechanism for punishing former office holders.

Doesn’t removing the presidential veto effectively put Congress in charge of the executive branch? 

In theory, yes, because Congress could micro-manage the administration by simply passing ad hoc legislation to override the president’s judgment. In practice, though, Congress would most likely refrain from doing this: first, though elected independently, Congress and the president are elected every year at the same time, so situations like the mid-term election backlashes that produce divided government would not arise very often; second, even when the president and Congress aren’t from the same party, chances are they won’t be from polar opposite ends of the political spectrum; third, the electorate would likely find such moves distasteful and so punish the Congress in the next election (after all, the president was elected in the last election and so has just as much democratic legitimacy as Congress, not to mention better pull in the media as a single, consistent figure compared to a Speaker who wasn’t personally elected and only serves a single four-month term). Besides, even if a degree of Congressional micro-management becomes the norm, I don’t see how this would be terribly harmful. (In the checks and balances between the three branches, the constraints of the executive and legislature upon each other aren’t nearly as essential as the constraints of the judiciary upon both.) 

If the 50 congress members all mostly agree with each other, why not have them vote to pass legislation by simple majority?

Having members formally vote puts us back into the mode of bargaining, this time between state interests. The resulting horse trading would quite likely lead to the corruption of the national interest.

Giving the final say to one randomly appointed member only subject to supermajority veto puts the focus on governing by consensus (within the caucus, that is). No single member can act as a spoiler, and the only path to influence is acting as a team player pushing the consensus forward. The Speaker, of course, is the weak point in this regard, except Speakers are totally exposed, as they can’t hide their parochial-interest pushing in a committee process no one notices. And besides, congress members are elected as a national group rather than by region, so they can’t be credibly threatened by home district interest groups.

Why not make just small changes, leaving our current structure mostly in tact?

The less sweeping version of these reforms would make all the same balance of power changes between the president and congress (removing the presidential veto, removing senatorial approval of appointees, and instituting joint appointment of judges by lottery) while finding some simple tweak to ensure (or at least encourage) single-party-at-a-time control of both houses of Congress without radically tweaking congressional structure or elections. While such a simple tweak would fix the problem of cross-party bargaining, it would preserve intra-party bargaining among single-member district representatives, representatives elected in the same electoral process easily captured by elites.

VI. How do we get from here to there?

Any proposed constitutional amendment has a long row to hoe, let alone one making radical structural changes, so a direct campaign for constitutional reform is almost certainly doomed. Instead, reformers should focus first upon adopting similar reforms at the state level, as the constitutional amendment processes of the states are less onerous to navigate, and voters regard government at the state level as much less sacred than at the federal level. Californian state government, in particular, seems a likely target for broad reform, as it reeks of dysfunction. Once new structures of government have been proven effective at the state level, voters will find similar federal reforms much more appealing.

What’s in it for centrists?

Centrists should be the most enthusiastic for these reforms, as centrists would be the true winners: imagine a government composed entirely of the candidates who most successfully ran to the center on every issue, all without previously going through a party-pleasing primary. Of course, this ‘centrism’ wouldn’t make Washington centrists entirely happy, as it would produce neither right-of-center economic policy nor left-of-center social policy but rather center-center economic and social policy. This would be true, by-definition “centrism”: government that enacts the agenda favored by the clearest majority of Americans.

What’s in it for liberals and conservatives?

Conservatives are the hard case. In a few areas, this system might deliver more conservative policy, especially in the short run, but it would forever take certain goals off the table, especially culture war issues. Politicians with strong anti-abortion positions, for example, would likely disappear from office as long as a solid 60+ percent of the electorate opposes outlawing abortion. Tax hikes on the middle class wouldn’t get any easier to enact into law, but tax hikes on the rich would. In the end, it comes down to a dare: do conservatives really believe that conservative policies are really more effective and more popular than liberal policies?

As for what liberals get, I simply believe a political system that’s more democratically responsive and allows us to focus on policy instead of process will, in the long run, lead to better, more liberal policy.

  1. This deference, of course, doesn’t placate conservatives, who have gone on to proliferate their own propaganda networks. []
  2. In fact, contra George Lakoff, narratives of heroes and villains best explain the American political divide: the fundamental disagreements between left and right are not about what is right and wrong but who is right and wrong. []
  3. Why non-elites buy into this narrative is a whole other question. []
  4. When we eventually crawl out of the ditch to relative prosperity, the elites who drove us into the ditch and spun the tires will claim victory and credit. ‘Hooray! This unspeakably awful situation we caused petered out after several years. We must have made the right moves.’ Also see Iraq (surge). []
  5. Interestingly, Americans habitually disguise disputes over minority rights as competing constitutional interpretations, in effect using fights over constitutional text as proxy for our deeper disagreements; we instinctually constrain our arguments of process and rights to a document mostly written over two hundred years ago. Unfortunately, removing these competing claims from the normal political process doesn’t resolve our deep disagreements but rather merely defers them. []
  6. Contractually-minded theorists, like Locke, justify holding non-founding citizens to the social contract by arguing that such citizens, having missed out on the chance to say yay or nay to the original formation of the government, would have acceded because government is preferable to the state of nature. This doesn’t answer, however, why a citizen is thereby bound to any particular government. Just because I happen to be born in Italy, shouldn’t I get to choose to be Canadian? []
  7. In fact, bipartisanship in the US has only ever really worked in the anomalous Post-War period, during which the ideological coherence of the two parties was muddled by the cross-cutting issue of race. For a few decades, the two parties contained many members of mixed positions, but once white Southerners realigned from being Democrats to Republicans, virtually every elected Democrat since has stood to the left of every elected Republican on core ideological issues. []
  8. Dangers of an unconstrained governing party do remain, of course: some policies are difficult or impossible to truly undo, like starting wars and massive spending- or tax cut-sprees, so a recklessly suicidal governing party could burden the government with damaging commitments even if that means getting turned out in the next election. Later, I’ll discuss how to curb this behavior.

    Also, the UK is a cautionary example of governing by alteration, as evidenced by the situation with Blair and the Iraq War. The Labour-majority electorate couldn’t find a way to oust Blair and his war policies without handing the reigns over to the Conservatives. A better electoral system—like instant runoff voting—should have fixed this. []

  9. Witness the current confusion over the 2010 Health Care laws. []
  10. Well, trust is the primary concern in primaries at least, where most of the real decision making gets done in American politics. The key swing voters seem to just reflexively vote out incumbents when dissatisfied with the economy. []
  11. See, for example, ‘Bush wasn’t a real conservative’. []
  12. Be clear that nothing guarantees an equal number of right and left candidates in the final election, though this is the most likely situation in most years. []

Trackbacks and Pingbacks

  1. Centrists: America’s most dangerous extremists » brian will . net - November 6, 2011

    [...] revised and expanded upon this post here.  Posted by Brian Will at 3:07 [...]

  2. Elections should matter » brian will . net - December 11, 2011

    [...] that my recent essay post is so long, I thought it a good idea to distill it into a workable political message. In other [...]

Leave a Reply