A bicameral 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 Länder in Germany. Its seats are the consideration that bought their agreement. The U.S. 17th Amendment (1913) left the Senate’s seat allocation untouched but changed what those seats represent — from state governments to statewide electorates. That single swap of the represented unit is what voided the founding bargain, removed the states’ last leverage over federal power, and retroactively turned “equal representation” from the deal into an apparent defect. This is the doctrinal spine behind the 17th Amendment specimen.
Links: Government Formation and Structure (the federalism-decay and distributed-protocol material lives there; this page is the upper-chamber-specific argument it points to), Administrative Agencies (the downstream chapter — the 16th+17th built the administrative state, and Loper Bright/Slaughter are the re-accountability response), The Industrial Revolution and the Remaking of Political Order (the global story this is the American branch of — Wilson/Goodnow imported the administrative science that the 16th/17th enabled), Corruption, Aggregation vs. Sorting (federalism as embedded sorting), Registration and Scope Creep (the may-vs-must / revocable-vs-constitutionalized distinction), Originalism and the Limiting Document (the read-it-as-a-contract premise this argument runs on), The Gödel Governance Problem (the 17th as the meta-bypass / innocuous-threshold worked example), Press Freedom and Governance (the Hearst outrage-lever specimen), Opposing Forces
Specimens (evidence, dated):
A unicameral legislature represents a population. A bicameral legislature almost always exists because the union was assembled out of pre-existing power centers that would not have joined without a guaranteed seat. The second chamber is that guarantee. It is not a “revising chamber” bolted on for prudence; it is the standing consideration in the contract of union.
| Union | Pre-existing powers | Their chamber | Still represents the unit today? |
|---|---|---|---|
| England / Parliament | the landed nobility | House of Lords | historically yes (hereditary tie to the estate); attenuated by modern reform |
| USA / Congress | the several states | Senate | No — since the 17th Amendment (1913) |
| Germany / federal republic | the Länder governments | Bundesrat | Yes — members are delegates of state governments, voting in instructed blocs |
The English case looks like the odd one out (nobility, not states), and at the surface it is. But the principle is identical and operates one level up: the Lords would not have consented to Parliament without representation. You would have to unwind feudal history to make the units line up — states don’t exist in England — but the structural role is the same: the upper chamber is where the parties whose consent was required keep a permanent voice. Germany is the clean living proof that the U.S. design is not a quaint 1788 artifact — a modern industrial federation runs exactly this model today, upper house composed of state-government delegates.
The U.S. Senate’s equal apportionment (two per state regardless of population) is routinely treated today as an embarrassing anti-democratic residue. It was nothing of the kind at adoption. It was the consideration that bought small-state ratification — the Connecticut Compromise. Small states joined because they were promised equal standing as states. Read the Constitution as a contract (as Originalism and the Limiting Document insists you must), and equal Senate representation is a bargained-for term, not a bug.
That framing is what makes the 17th a serious charge rather than a policy complaint (§4).
The 17th Amendment did not touch the allocation of Senate seats (still two per state). It changed the represented unit: senators went from agents of state legislatures to agents of a statewide electorate. Jake the Lawyer’s mechanism (the specimen): it changed who a senator fears — the legislature before, the voters after — and thereby dissolved the states’ principal-agent hold on the one chamber that ratifies treaties, confirms officers, and writes the enabling statutes of the administrative state.
Everything downstream follows from that one substitution.
If equal representation was the consideration small states received for representation as states, then swapping the represented unit to people voids that consideration while keeping the signature. The 17th kept the form of the deal (two seats each) and removed its substance (those seats speak for the state government). This is a breach-of-bargain argument, not a “this was unwise” argument — and it is reachable only from the vault’s own contract premise. It is a stronger and narrower charge than the video’s general federalism complaint.
The modern objections — “Wyoming vs. California,” “abolish the Senate,” “land doesn’t vote” — are category errors against the original design that only became sayable after the unit was swapped.
So the 17th did not just weaken federalism; it manufactured the complaint against the Senate’s structure by changing the frame in which that structure is read. Compare the “land doesn’t vote” smuggle flagged in LaCorte — Urban-Rural Divide and the scope-not-the-franchise move in pool-kyla: in each, an objection that feels obvious is load-bearing only under an unstated reframing of what is being represented.
The 17th is the second of two amputations, and the vault already had both halves — unjoined until this page:
| Year | What the states lost | Hirschman term | Vault location |
|---|---|---|---|
| 1865 | secession foreclosed; union permanent | Exit | government-formation §The Exit Problem |
| 1868 | 14th Amendment — incorporation converts the exit-loss into ongoing upward authority transfer | (the bridge) | government-formation §Federalism Decay |
| 1913 | 17th removes the states’ principal-agent hold on the Senate | Voice | government-formation §Federalism Decay |
An actor with neither exit nor voice has no leverage of any kind — and that is the states’ post-1913 position. This is why “which single repeal fixes it” is malformed: the erosion is multi-level (Civil War → 14th → 16th → 17th → Wickard), and the 17th’s significance is that it removed voice after 1865 had already removed exit. 1865 removed the escape valve; 1913 removed the say; everything after is a ratchet with no release — which is the abstract “why does government always get bigger?” ratchet named in Gödel-governance and Opposing Forces, here given a dated, single-mechanism worked example.
The strongest-seeming objection to repeal: ~29 states had voluntarily adopted popular senator selection (the “Oregon system”) before 1913, so direct election was the popular will and repeal just fights it. This fails on a revocable-vs-constitutionalized distinction (developed in registration-and-scope-creep):
The two produce a superficially identical practice (a popular vote picks the senator) but are opposite in kind. So repeal does not ban popular election — a state could still run one by its own choice, exactly the Article II elector-appointment discretion. What repeal restores is state discretion over the manner, and the operative variable turns out to be constituency scope, not election directness: even a state-run popular election scopes the senator to the state, whereas the 17th scoped his incentives national (national money, national media, national party). General form: a delegation that can be withdrawn is a check; a delegation that is constitutionalized is a transfer.
The whole argument assumes federalism is worth protecting. It is, for two independent reasons — and centralization attacks both at once (see government-formation §Federalism Decay for the full treatment):
The “laboratories of democracy” defense and the “check on Washington” defense, usually treated as alternatives, are these two faces of one argument.