Seven years ago at the hearing for then-Supreme-Court-nominee Brett Kavanaugh, then-Senator Ben Sasse offered a then-widely-lauded explanation of the politicization of judicial hearings. Congress, he said, had steadily ceded authority to the executive branch. In his telling, agencies were quietly writing laws and courts publicly reviewing them, so political attention naturally migrated courtward.

It's a bit much to blame the nomination circus on bureaucratic lawsuits. Most of the Supreme Court cases that get attention in campaign season – abortion, presidential immunity, gender-affirming care, flag burning, etc – are big constitutional issues, not red tape. If we genuinely want less political courts, we should address our big constitutional questions with citizen conventions and get nominations out of campaigns altogether with citizen appointment boards.

But the phenomenon of congressional abdication that Sasse described merits a closer look. Why has Congress given up its power? Why should legislation ever be written by the executive branch and refined by the judicial branch? The only branch that should be legislating is stepping aside.

Agencies govern much of our lives. They regulate taxes, financial securities, product safety, environmental protection, public health, housing, transportation, social security, law enforcement, and on and on and on. In all cases, Congress sets bounds on what the agencies can do, but the agencies are left to work out the details, in which there is abundant devil. Why does public deliberation on such important details only happen within our least democratic branch and only after someone sues? Why doesn't Congress handle such things?

That last one's easy to answer: they can't. It's just too much. Congress is too small and its attention too limited to tackle every concern of the sprawling federal government. Even if we clawed back the massive amounts of time congresspeople spend on fundraising, campaigning, and other distractions, there would not be time for them to read and consider agency technicalities. Congresspeople can't even feasibly read or even understand all the laws they "write". No wonder they outsource.

There's very little way around it: the US government is too much for the US Congress.

Even if not to blame for politicizing courts, the narrative of power-hungry bureaucrats has had political legs in recent years. "Deep state" stories warn of red tape turned blue. Government, they, say, has been quietly usurped by nerds. The image of shadowy deskmen micromanaging our lives serves to justify the use of executive power to run rampant through government agencies, firing at will, often in the name of "accountability".

But will further executive takeover really resolve the problem? Will it yield better agency regulations? Surely what is too complex for 435 representatives and 100 senators is too complex for one president.

Far too complex. And too influential. An executive branch with scant checks and balances being given extensive legislative authority puts much of the power of two branches of government under one man's thumb.

Arguably, we have fallen into this authoritarian trap in part because the executive branch can do something Congress can't: scale. The executive branch has departments. Those departments have agencies, each of which has multiple echelons of leadership. The judicial branch, too, scales somewhat via appellate and district courts. By contrast, the size of the legislative branch, congressional aides notwithstanding, is largely frozen.

The legislative branch can't grow to reflect the scope of its duties. Yet.

My favored solution will surprise no one following this essay series. Mini-publics again. Members of Congress don't have the bandwidth for agency regulations, and nor does the American public at large, but mini-publics could. Randomly selected deliberators could regularly review agency policies and approve them or require changes. The legislative branch could scale dramatically if it embraced the power of sortition (selection by lot). And they could begin by assigning mini-publics tasks that are clearly legislative and yet have been jettisoned from their proper branch.

Agency minutiae don't make for the most obvious mini-public subject matter; they often set their gaze on sexier things. But don't picture random citizens sitting down with blank paper and drafting FEMA policies or nuclear energy regulations or EPA emissions rules; that wouldn't happen. The deliberators would learn about key issues, hear testimony from various interest groups, invite proposals, weigh solutions to contentious problems, and, where appropriate, demand changes.

The obvious objection that random citizens don't know much about the inner workings of bureaucracies meets the obvious rejoinder that congresspeople don't either. Pretty much nobody does but the bureaucrats. But citizens could comprise focused, representative audiences to learn about such things and weigh issues of importance. They would ask probing questions. They would almost certainly push for agency regulations to be written in clear language, a massive gain on its own. They could make bureaucrats genuinely democratically accountable. They could make the deep state shallow.

And they could ensure that government policymaking emerges from actual deliberation, securely nested in the legislative branch where it belongs.

Listening to current debate on agency policy, you could almost forget that legislation is supposed to fall to legislators. A recent Supreme Court case (Loper Bright Enterprises v. Raimondo) controversially ended a many-decades-long precedent of judicial-branch deference to executive-branch policymaking. Pundits argue about whether and how far legislative power should mosey from the executive to the judicial branch. But neither branch should be legislating. Neither.

In their book Legislature by Lot, John Gastil and Erik Olin Wright propose a new democratic pattern: a bicameral legislature in which one chamber's members are elected and the other's are chosen at random. (The Sortition Foundation has a related campaign to replace the House of Lords with a House of Citizens.)

Their book is lovely; it offers a thoughtful proposal and rationale, clear but not overly specified. It considers concerns and a few design variants and invites a rich array of scholars to weigh in, not all of them favorably. It explores many key democratic concepts from fresh perspectives, challenging my thinking in multiple realms. If you're game for academic writing, I recommend the read.

I do not, however, recommend their bicameral solution, especially not for the US federal government (their solution is proposed more generally). Reasons why and proposals for alternate solutions will trickle out in future essays, but one key concern relevant here is this: the rule-by-legislative-chamber design that dominates in electoral democracies is empirically not scaling well. One or two chambers don't have the bandwidth to deliberate deeply about all the issues that face a nation today.

We need to go beyond simple one-to-one translations from the old electoral model to a new sortition plan. Simple sortition chamber solutions are too copy-paste. Add an "s" to election, drop a "bal" from ballots, call it a government.

Instead we should explore the new design possibilities sortition affords. We should reorganize the legislative branch so that the same handful of deliberators aren't making all the decisions about everything. Or worse, realizing they don't have time to deliberate about everything and letting many important decisions pass with no deliberative attention at all.

Small-government and large-government advocates alike can agree that a government's legislative branch should be capable of deliberating about any rule it passes. No honest player supports ill-considered law. Even if the government were to shrink considerably, a right-sized collection of deliberative bodies would involve a lot more folks.

Today we hand all federal legislative authority to a few hundred people (who in turn hand much of that authority away). You could fit all of our national legislators in a small concert venue. Roughly 1 person for every 75,000 citizens gets a say in our laws. 1 person for a large football stadium. And given the uneven power with the legislature and the poor quality of deliberation, fewer really shape policy.

Mini-publics would allow the legislative branch to subdivide and scale. Instead of two groups of hundreds serving for years, we might see dozens of groups of dozens of citizens serving for months. When you combine the numerous bodies and the faster cycling of terms, you might rapidly see a hundred-fold increase in the number of deliberators shaping federal law. So the proportions would flip: you'd need a football stadium for all the legislators, and the number of citizens per legislator would fit in a small concert venue. In short order, pretty much all of us would personally know someone who'd sat in an assembly molding federal law.

Unlike a bicameral legislature with a sortition chamber, which would require a constitutional amendment, Congress could begin implementing department- and agency-specific mini-public solutions today. They have delegated their authority outside of their branch; they could re-delegate within it. I would certainly support such experiments. In fact, if mini-publics were responsible for agency oversight, I would suggest Congress pass even more authority to mini-publics than they do to agencies now.

In the long term, we'll have to alter the Constitution to make permanent the deep changes we need, especially if we wish to move core legislative authority away from elected bodies. I do wish for that and will dig much deeper into ideas for reorganizing the legislative branch soon.

But even those who find such ideas too radical and like elected congressional chambers can recognize the basic democratic need of having deliberative bodies duly consider all the nation's laws. With the status quo, that's simply impossible.