A single-presenter case for repealing the 17th Amendment: direct election of senators didn’t remove oligarchic capture, it relocated it — and by removing the states’ seat at the federal table, it built the foundation of the permanent administrative state.
Date: 2026-07-23 (video published 2026-07-20) Source: YouTube — Jake the Lawyer / Internet Esquire — Transcript Format: Single-presenter video essay, 37:37. No opposing voice. Route: Chris hit this via a Ryan stream on the rise of communism / left-wing extremism, which reacted to it. Going to the source. Vault relevance: The Represented Unit (the thesis this debate sparked — read that for the portable argument), Government Formation §Federalism Decay (the capture-ROI corollary was promoted here), Registration and Scope Creep (may vs. must), Chiafalo v. Washington, Wickard v. Filburn, Originalism and the Limiting Document, Aggregation vs. Sorting, The Gödel Governance Problem, Press Freedom and Governance
Worth naming up front: the video is not about communism. Ryan’s stream was; Jake’s video isn’t. The only bridge is the Progressive Era — Jake’s antagonist is 1890s–1913 progressivism and the administrative state it built, not Marxism. Whether that’s the same story told at a different altitude, or a category slide, is seed #6 below.
Also not a debate — an explainer, same shape as LaCorte. So the job is to pressure-test one causal chain rather than score an exchange.
“The connecting tissue was that the communism and progressivism rose together from the late 19th century to the first half of the 20th. Same time period, similar views was the point here.”
So seed 7 resolves: Ryan isn’t claiming the 17th Amendment is communism. He’s claiming one era produced both, out of the same Gilded Age disruption and the same diagnosis (concentrated private power → cure by concentrated public power). Jake’s video is the domestic-institutional branch of that era; the communist movements are the other branch. Same root, different limbs.
Chris ties this straight into the existing democracy thread: pure democracy cannot work; a republic backed by a constitution can. The operative thing about the pre-1913 Senate wasn’t that it was less democratic — it was that senators were beholden to the states, and that was the incentive. Remove the principal, and you remove the check.
“It is fair to argue that it wasn’t just the 17th (and 16th) that caused this… the overall thread (ala Ryan) is that this was part of a progressive era. I would think Jake and Ryan would argue for a repeal of the 16th, 17th and Wickard to undo the progressive damage.”
This is the right answer to the Wickard objection: the unit of analysis isn’t a single amendment, it’s the Progressive Era as a package — 16th (revenue), 17th (incentive), Wickard (scope). Repealing any one leaves the others load-bearing. That also means “which is the binding constraint?” is the wrong question; they’re complements, not substitutes.
My seed 4 (“capture is conserved — reform relocates capture to whoever owns the new channel”) gets pushed back on, and Chris sides with Ryan’s comments here:
“I push back on Hearst… the removal of federalism meant that it was easier to target and capture the entire chain. Hearst is a result of consolidation and removal of federalism.”
So Hearst is not the cause that a media baron happened to engineer a self-serving amendment. Hearst is downstream of the same consolidation — national railroads, national economy, national information market — and the 17th extends that consolidation to the political layer. The generalization changes shape entirely:
Not “capture is conserved.” Rather: capture cost scales inversely with centralization. Federalism is a capture-resistance mechanism because it forces an aspiring capturer to buy 50 legislatures instead of one national narrative. Consolidation is what makes capture cheap — Hearst is the symptom.
This is the stronger thesis and it’s Chris’s, not the video’s. Candidate for promotion.
“States would be free to hold elections to pick senators if they deem (like they can for the president in Article II), but that still means those senators are beholden to the people in that state instead of campaigning on ‘national’ interests.”
(Chris made this point live in the stream chat.) Two things fall out:
→ Consequence worth pulling out: if the represented unit is the state, equal representation is coherent. Once the 17th makes the senator a representative of people, equal representation becomes indefensible-looking — a small-state voter now has outsized individual weight. The 17th is what manufactured the modern malapportionment complaint (“abolish the Senate,” “Wyoming vs. California,” “land doesn’t vote”). Those objections are category errors against the original design and only became sayable after the represented unit was swapped. Compare the “land doesn’t vote” smuggle flagged in LaCorte and the scope-not-the-franchise move in pool-kyla.
“We removed the checks on power by delegating it to ‘the people’ instead of the counterbalance… the rise of progressivism and the dismantling of federalism is what is tearing apart the foundations of the republic, much like those other democracy experiments.”
Noted as the through-line: the 17th is a specimen of the general question the vault keeps circling — how power consolidates, what causes government to grow, how administrative bodies are kept in check.
Chris takes the sequencing amendment but holds the core: Hearst “was smack dab in the middle of the progressive era and was a big player in it, and he benefited greatly by pushing it.” Settled reading — not the prime mover, an interested participant who profited from the consolidation he advocated. The era is the unit of analysis; Hearst is a well-placed actor inside it, simultaneously its product and one of its accelerants.
“Before this breakdown I asked Ryan if he agreed to the vault conclusion that the civil war was the start of the end of federalism and he agreed. The 14th being a result of this too.”
(Grounded, not asserted — this is already the vault’s position: government-formation §The Exit Problem — “The US Civil War as the death of federalism… transformed federalism from a voluntary association of sovereign states into a permanent union with subordinate administrative units,” and Gödel-governance lists “Civil War foreclosed exit” among the inflection points.)
→ The synthesis this produces — exit and voice, killed 48 years apart. The two halves were sitting in the vault unjoined:
| Year | What the states lost | Vault location |
|---|---|---|
| 1865 | Exit — secession foreclosed; union permanent; the constituent parties can no longer leave | already at government-formation §The Exit Problem |
| 1913 | Voice — the 17th removes their principal-agent hold on the Senate; no seat at the table | added this session, §Federalism Decay |
Hirschman’s terms make the ratchet legible: an actor with neither exit nor voice has no leverage of any kind, and that is the states’ position after 1913. The 14th sits between them as the mechanism that converted the exit-loss into ongoing authority transfer (incorporation). This is why “which single repeal fixes it” is the wrong question — and why Chris frames it as levels: “for sure, there are many levels to the erosion of power. This one example just made it clear how the ratchet ticked up.”
“Big difference between ‘may’ and ‘must’. Before, if the states ran elections and these didn’t work out well for the state, they can be reversed. Now they cannot. Removing this guard is exactly the point being made here.”
This kills my objection, and it’s the sharpest correction of the session. The Oregon system and the 17th produce a superficially identical practice — a popular vote picks the senator — but they are opposite in kind:
So the operative variable is reversibility, not practice. Pointing at 29 states doing it voluntarily proves nothing about the amendment, because what the amendment removed was the ability to stop doing it. Generalizes cleanly: a delegation that can be withdrawn is a check; a delegation that is constitutionalized is a transfer. Candidate for promotion — bears on Registration and Scope Creep and the Gödel-governance friction/flexibility dilemma.
And the second half stands independently: senators now run on national ideas and are not bound by the state — the constituency-scope point from earlier, restated as the practical result.
My objection was that the House of Lords is hereditary, not state-delegated, so the analogy is loose. Chris’s answer reframes it at the right altitude:
“The lords are bound to their family… states do not exist in England, but the principle is still the same. The Lords would not have agreed to Parliament without representation.”
The claim isn’t “Lords ≈ states.” It’s that in both cases the upper chamber exists to represent the pre-existing power centers whose consent was required to form the union at all — barons in England, states in America. The chamber is the consideration that bought their agreement. Same structure, different constituent unit:
| Union | Pre-existing powers | Their chamber |
|---|---|---|
| England / Parliament | landed nobility | House of Lords |
| USA / Congress | the several states | Senate (pre-1913) |
| Germany / federal | the Länder governments | Bundesrat — still works this way today |
→ The contract framing this unlocks. The vault’s Originalism and the Limiting Document already insists on reading the Constitution as a contract. Take that seriously and the 17th looks worse than “unwise policy”: equal representation as states was the consideration that bought the small states’ ratification (the Connecticut Compromise — Chris’s earlier point that “the only reason small states joined was that they had equal representation”). The 17th did not change the allocation of Senate seats, but it changed what is represented in them — retroactively voiding the consideration while keeping the signature. That is a different and more serious charge than Jake’s, and it is reachable only from the vault’s own contract premise.
(And it re-explains the malapportionment complaint above: once the represented unit is people rather than states, the small-state seat allocation looks like a bug instead of the deal.)
I put the adversarial check to Chris: state legislature seats are cheap to capture (low-information, low-turnout, small-dollar), so “decentralization raises the price of capture” may be backwards — fifty bargain-bin targets. He grants the objection and dissolves it by changing the variable:
“Yes, your low cost, easier to capture is valid, but this has always been about ROI. Capturing all of the low-cost seats [is] more expensive than capturing 1 high-cost seat with much more leverage… consolidated power is profitable to capture because it is highly leveraged, distributed power is not worthy of capture because it is scope limited. We see corporate capture and cronyism because it became profitable to do so.”
Price per unit was the wrong metric; return per capture is the right one. Federalism therefore doesn’t make capture hard — it makes it unprofitable, which is the more robust protection because it doesn’t depend on anyone’s vigilance.
The consequence, which is the most portable thing in this session: cronyism and corporate capture are not primarily moral failures of particular actors — they are the predicted equilibrium once the prize becomes leveraged enough to be worth buying. That relocates the cure entirely. Ethics campaigns (better people) and regulatory patches (anti-lobbying rules, themselves capturable by the same logic) treat the symptom; only de-leveraging the prize treats the cause. Tullock’s rent-seeking result, reached from the structural side.
And Chris names federalism’s purpose as two-fold, which corrects a one-sidedness in how this page had been running the argument:
Centralization attacks both at once: it destroys the surface area and forces one outcome onto a heterogeneous population. Which also means the two standard defenses of federalism people usually treat as alternatives (“states as laboratories” vs. “checks on Washington”) are one argument with two faces.
(Folded into government-formation §Federalism Decay — the corollary there was rewritten from the weaker cost formulation to this one.)
“What is interesting is that it is often the things that seem so innocuous that are the biggest killers. In one of our democracy experiments, all the moderator wanted to do was stop spamming of emotes and memes, and that led to a revolution and eventual destruction of the system. Here: why not let people vote on senators? Because the state control of them is what makes the system continue to function.”
This names a mechanism the vault has circled but never stated cleanly: the changes that kill a system are selected for not tripping its immune response, and seeming innocuous is exactly what lets them through. A change that looked dangerous would mobilize defense; the lethal ones are the ones nobody defends against because “it’s not that big of a deal.” Note Jake names this in the video — “it’s not that big of a deal, why do you care so much?” is the exact rhetorical signature he flags on the 17th.
Two grounded specimens of the identical shape:
This is the same structural fact as the camel’s-nose logic in registration-and-scope-creep: the mechanism that will consume the system enters below the threshold of alarm. Innocuousness isn’t a coincidence — it’s the selection filter for which changes survive to become fatal.
“I bring back Ryan — he has a working theory that ‘corruption’ is generally just a word to say you don’t like what the other side with power is doing. There is some truth in that. Today we see partisans willing to accept their side doing things no matter what, while automatically calling the other side doing something similar ‘corrupt.’ Harnessing this ‘outrage’ seems to be a big Gödel lever that undoes these systems.”
Two linked claims, both portable:
“Corruption” is often an indexical, not a description. It frequently means “out-group exercising power” rather than any fixed procedural violation — evidenced by the partisan asymmetry (identical acts read as legitimate in-group / corrupt out-group). There is some truth in that — but the honest version has a floor: some things are corruption under any index (bribery, forged votes, ignoring a losing result — the trolligarch show-trial). So the label is partly indexical and partly real, and the move to watch is collapsing the whole category into “just a slur” to launder the floor-level cases. (This is a scope move — cf. the Fallacy Fallacy: “it’s just a word for disapproval” is true at the margin and false at the core.)
Outrage-harvesting is the enforcement bypass. This is the sharp one and ties the whole session together. A constitution is friction — it works by making bad changes slow and costly (the Gödel-governance friction/flexibility dilemma). Manufactured outrage is the universal solvent for that friction: it manufactures the emergency that justifies bypassing the slow path, and it does so from outside any rule the constitution can reach — you cannot write a clause that forbids people from being angry. That makes it a genuine Gödel lever: a move the system cannot rule on from inside itself. And it closes the loop on Hearst — The Treason of the Senate was outrage-harvesting at industrial scale, and it’s what carried the 17th’s innocuous-looking change over the ratification bar. The two threads are one: outrage is how you get an innocuous-looking amendment past the immune response.
Both are promotion candidates — see the note to Chris. The outrage-lever in particular looks like a real addition to Gödel-governance §The Enforcement Bypass Problem, and it bears directly on Press Freedom and Governance.
“There is a ground to ‘corruption,’ but like many other things, it is now used to sling mud, and the original, proper meaning gets lost in the outrage.”
Settled reading: the concept has a real floor (bribery, forged votes, ignoring a loss) and the word has been weaponized into a mud-sling, so the outrage-lever works partly by eroding the floor — once “corruption” reads as pure tribal disapproval, the genuine cases lose their alarm value too. So the two threads compound: outrage doesn’t just bypass the friction, it degrades the vocabulary the immune system uses to name threats. That is a nastier failure mode than either piece alone.
“Early in the video there was discussion about how the industrial revolution was such a large change that it spawned most of this political divide, and I think there is a lot to this. I even heard a smart man I respect say that ‘all liberal democracies turned to authoritarianism or socialism after the industrial revolution.’ … I would argue WW1 was the reaction to the industrial revolution breaking society. This should become a topic to explore in more detail — even I don’t understand all of the history with it.”
Flagged as its own future page — deliberately NOT written here, because neither Chris nor I have the history well enough in hand to write it as fact, and fabricating it would violate the grounding rule. What we have so far, kept honest about its status:
Where it will likely live: this is the historical spine under
Technology → Culture → Politics (conditions →
culture → politics — the IR is the ur-example of “conditions”), and connects to
The Unreplaced Event (WWII) and the
civilizational-cycles structural-institutional
work. Proposed method: a proper sourced pass (/deep-research or a grounded ingest), not a
from-memory write — this is a “go find it” topic by Chris’s own admission. Logged here so the thread isn’t
lost; to be spun into its own page when we pick it up.
Chris’s starting point: the administrative agencies serve at the privilege of the President — he cannot abolish the IRS, but he can make it barely functional. Pulled into a full page: Administrative Agencies and the Power to Execute. Chris’s unitary-executive reading, grounded in the four cases (Humphrey’s → Seila → Loper Bright → Trump v. Slaughter, 2026 — Humphrey’s overruled, so his position is now the holding), the DACA/removal asymmetry he flagged as “backwards,” and the resolution that gap-filling is lawmaking — so the design question is who holds the discretion and can they be fired. Uses the corruption test as its hinge (statutory purpose = legitimate; private end = corrupt).
“This one video was key to bridge lots of the open topics left in the vault. Why does government always seem to get bigger? How did the administrative state become so powerful? What causes the ratchet up? While it is true this was only one step in this process, it was actually a big one. The Senate no longer really serves its purpose, and states no longer have the representation they were promised. Everything is ‘national,’ and this is a problem as it kills the better federalism structure.”
That is the reason this is a specimen worth keeping, not just a debate review. The vault has several pages that name the ratchet — Gödel-governance (centralization drift since 1789), government-formation §Why Constitutions Degrade (three structural pressures), Opposing Forces, the civilizational cycles work — but they describe the ratchet in the abstract. The 17th is a clean, dated, single-mechanism instance of it: you can point at one amendment and watch exit→voice→scope tighten. It’s the worked example the abstract pages were missing.
Reusable moves that emerged from the discussion (portable ones promoted to The Represented Unit and government-formation):