Stable governance doesn’t require consensus
Accountability matters more than agreement
History is full of reckless individuals, impulsive factions, and immoral majorities who have imbued in us an instinct to slow political systems down and force them to earn a higher threshold before acting. People want governance that is orderly rather than lurching from one extreme to another every few years.
This instinct to keep governance ‘stable’ is as old as political thought itself. Plato’s solution was philosopher-kings, rulers so constrained by their own wisdom that bad outcomes would be prevented before they could occur. Intellectuals since then have offered theories of stability through divinity, historicism, and might. Every era has looked at the problem of bad governance and asked how to prevent it. Today we tend to think the most-reliable means of preserving a stable government comes from consensus; the will of the people as a check on itself. If we can’t get the right leaders and keep them there, we contend, we must at least prevent bad ones from doing too much damage. And since we cannot hold referendums on each and every decision, we built procedural constraints that force majorities to share responsibility and earn broader support before they can govern.
Our goal is the same as Plato’s: prevent the government from doing the wrong things by structuring the system so that it cannot act without sufficient wisdom. We differ in that we think that wisdom comes from collective decision-making rather than inherent virtue. And yet, the underlying belief is the same: stability requires constraining power until it is wise enough to act.
I posit that this concept of stability is entirely backwards. Governments are most stable not when their actions are constrained by aggregated wisdom, but when they can be held accountable—blamed, criticized, and removed. True stability comes from the system’s ability to absorb movement and adjust course without collapsing.
Compromised consensus
We seek an assurance that bad governments can’t make too many mistakes; that minority interests won’t be steamrolled; that the system won’t swing so violently that ordinary people can’t keep up with it. Those are legitimate concerns. The main question though is whether consensus requirements actually deliver on its purported goal to protect, to stabilize, and to prevent error. I don’t think it does because that would depend entirely on being able to identify who failed and their ability to organize criticism around them. That requires clarity of responsibility.
Failed majorities are only removed when people hold negative opinions about the outcomes under their watch. But if they’re unable to tell who is to blame, it necessarily becomes much more difficult to tell who it is to criticize, and thus to remove. Procedural rules requiring broad agreement dilute responsibility before a single policy is ever implemented because they necessitate compromise. Compromised policies get a good reputation but it’s primarily due to our yearn for consensus. They’re amalgamations of no one’s idea of what will work. When it fails, no one learns anything, because no one ever agreed with it. The majority blames the minority for watering it down; the minority blames the majority for passing it at all. The underlying ideas of the policy that seemed good to some factions never get properly tested. The losing coalition can’t point to a responsible party and can’t remove those who failed because the failure belongs to everyone, and therefore to no one.
The instinct behind requiring broad agreement to act is, I think, ultimately about protecting those on the losing side. Legislative procedures like the filibuster is that instinct expressed in procedural form: making broad agreement the price of action, with the aim of protecting the interests of those not in control. Its defenders call this a safeguard for ‘the minority’. The problem is there is no stable, coherent minority. Minority status is dynamic at every level simultaneously: the individual mind all the way to the coalitions they compose.
Procedural consensus vs. fluid minds
Consensus only work as a source of stable wisdom if people’s opinions are stable. But since people change their minds continuously, consensus must be unstable too. Voters realign and shift priorities. Entire coalitions absorb new members and move with them. These changes happen constantly—between elections, during them, and in the long stretches of governance in between. The system is never still.
Procedural constraints that aim to protect against excesses by enforcing stability through shared wisdom rest on the assumption of a static system. Thus, governing rules like the filibuster does its job by privileging a model of legitimacy based on durable consensus. But in a fast-moving environment of shifting preferences, that requirement forces governments to lag behind reality by making action harder than the system would otherwise allow.
I think polarization adds an interesting layer to this. The common understanding is that polarization means both sides getting more extreme—more radical, more distant from other coalition members. I think that’s mistaken. What polarization actually means is that individuals are more consistently aligned with their party’s bundle of positions than they used to be. It doesn’t mean those positions are radical at all; only that fewer people are cross-pressured on as many issues as they once were. But consistent alignment is not the same as uniform alignment. A more polarized electorate still contains individuals whose views cut across party lines by existing inside both coalitions simultaneously.
And because minority status is fluid, the filibuster has no principled basis for what its protecting. It simply hands veto power to whoever happens to be in the minority on a particular vote—even though other members of that minority might support the policy, and members of the majority may hold legitimate criticisms. In our attempt to guard against potential mistakes by government, we have inadvertently diluted and stagnated a system that thrives on clear responsibility and the capacity for movement. This is because it blocks the mechanisms of change by acting as if failure comes from change. Which is true—failure does come from change. But so does something far more significant: progress.
The dynamics of dynamism
The capacity for progress—the ability to correct errors through continuous testing, criticizing, and correcting—already provides the protection people seek from stable systems. Yet it relies on mechanisms that broad-consensus requirements suppress.
Between coalitions, the constant threat of electoral and legislative reversal constrains majorities at all times—not just in November. Within coalitions, broad majorities already internalize minority interests through the very process of assembling themselves. It is the ongoing adjustments from individual ideological movements that enforces consideration of legitimate criticisms. Every election is an opportunity for a losing faction to reorganize, expand, and return.
Plato’s insight helps explain our instincts. We want to prevent mistakes. But if wisdom were sufficient for government, there’d be no need for protection or procedural constraints, because it would already make the right decisions. That’s what Plato wanted from philosopher-kings, what Europe sought in divine monarchs, what Americans still hope consensus can deliver.
However, a healthy democracy relies on errors. Protection, stability, order—all these desired assurances—are realized from the capacity to make mistakes, recognize them, and correct them without the system imploding. Making mistakes is how we learn. When we prevent actions from being clearly attributable to anyone, we prevent learning.
Consider a party commanding 80% of the electorate. Even at that extreme, decision-making through consensus defenders would still claim the remaining 20% needs procedural protection. At this point, it becomes indistinguishable from minority-rule. And this isn’t just theoretical. Take a bill to ban transgender athletes in sports, for example, polling at roughly 80–20. The moderate Democratic faction wouldn’t object to its passage, but the faction most likely to deploy the filibuster isn’t them. The result is an undemocratic outcome: the system cannot test the idea, assign responsibility, or correct course—despite broad agreement. In both cases, requiring broad agreement to do things privileges procedure over the actual policies and outcomes the government produces.
By the same token, when the majority governs well, the coalition expands further by absorbing additional interests. Therefore, the stability, protection, and order we seek from government comes from remaining a credible threat to the majority’s survival. And that relies on the system’s ability to move—the capacity to make mistakes, to assign blame, and to correct.