The Capitol Audit began with a scorecard grading Congress’s leadership, then turned to a single office, the parliamentarian. This entry turns to the institution itself: the Senate, graded the same way the Grievance Audit graded the republic against 1776 and the Risk of Rights graded the Bill of Rights against 2026 — one constitutional design feature at a time, quoted, checked against the current record, and scored.
The method
Article I, Section 3 and Section 5, plus Article II, Section 2 and the Seventeenth Amendment, lay out what the Senate is supposed to be and do: how it’s composed, how long members serve, what it can block or confirm, and what it can do to its own members. Ten of those design features are scored below, each on the same 0–100 scale used across the Capitol Audit: 85+ is functioning as designed with no real erosion, 65–84 is solid with real but limited strain, 45–64 is mixed — meaningful erosion alongside real function, 25–44 is a design feature substantially bypassed in practice, under 25 is functionally hollowed out.
Equal state representation
Article I, Section 3"The Senate of the United States shall be composed of two Senators from each State." Two per state, full stop — no population weighting, and Article V specifically forbids amending this without a state's consent, making it the single most entrenched clause in the Constitution. It is functioning exactly as the framers designed it as part of the Great Compromise. What's graded here isn't a deviation from design — it's the growing real-world strain the design was never built to absorb: California's roughly 39.4 million residents and Wyoming's roughly 589,000 both get two senators, a disparity of roughly 67 to 1. The clause works precisely as written. Whether "as written" still fits a country this size is a question the Constitution itself doesn't let voters answer.
Staggered six-year terms
Article I, Section 3Senators serve six years, divided into three classes so roughly a third of the chamber is up for election every two years. The design goal was continuity — a body that never fully turns over at once, insulated from any single election's mood. That is exactly how it still functions two centuries later: no controversy, no erosion, no workaround. It's the cleanest-functioning piece of Senate design on this list.
Qualifications for Senators
Article I, Section 3Age 30, nine years a citizen, an inhabitant of the state at the time of election — three requirements, unchanged since ratification, still applied without exception or workaround. No modern controversy over eligibility rises to the level of the design failing; the clause simply works, quietly, the way most durable design does.
Direct election of Senators
Seventeenth Amendment (1913)The original design let state legislatures pick senators; the Seventeenth Amendment, ratified April 8, 1913, replaced that with direct popular election. This is the one entry on this list that is itself a correction — the framers' original mechanism produced enough vacancies, deadlocks, and corruption scandals in state legislatures that the country amended it out. Since 1913 it has functioned without serious incident or reversal effort. A design change that actually worked.
Rules of proceedings — the filibuster
Article I, Section 5"Each House may determine the Rules of its Proceedings" — that's the entire textual basis for the modern filibuster. The Constitution names specific supermajority requirements in only a handful of places — treaties, expulsion, veto overrides, impeachment conviction — and implies simple majority for ordinary business everywhere else; the 60-vote cloture threshold for ordinary legislation is a Senate rule, not a constitutional one, and it has been used at a scale the framers never contemplated. Cloture motions were filed 201 times in the 115th Congress (2017–2018), climbing to 328 in the 116th and 336 in the 117th, before dipping to 266 in the 118th — a level of routine supermajority gatekeeping unrecognizable against most of the chamber's first 180 years. Both parties have chipped at it with "nuclear option" carve-outs rather than a floor debate on the rule itself: Democrats in 2013 for most nominations, Republicans in 2017 for Supreme Court nominations, and Republicans again in September 2025 for en bloc sub-cabinet confirmations. The rule is real, self-imposed, and now edited by whichever party has the votes to edit it — the opposite of a stable constitutional design.
Advice and consent — nominations
Article II, Section 2The Constitution gives the Senate a check on executive and judicial appointments; it says nothing about a home-state senator's veto. "Blue-slip" practice — letting a nominee's home-state senators block a Judiciary Committee vote — is a 20th-century committee custom, not a constitutional rule, and it's actively eroding: in 2025, Sens. Cory Booker and Andy Kim withheld blue slips to block a U.S. Attorney nomination, while Judiciary Chairman Chuck Grassley has publicly resisted pressure from the Trump White House to abandon the practice for blue-state nominees entirely. Meanwhile the September 2025 rules change allowing en bloc, simple-majority confirmation of large batches of sub-cabinet nominees speeds up exactly the kind of individualized "advice" the clause was written to require. The constitutional check still exists; the custom that shaped how it's actually exercised is being renegotiated in real time by whoever holds the majority.
Advice and consent — treaties
Article II, Section 2"He shall have Power, by and with the Advice and Consent of the Senate, to make Treaties, provided two thirds of the Senators present concur" — one of the Constitution's few explicit supermajority requirements, and the one most thoroughly routed around. Since 1789, the United States has concluded roughly 1,834 formal treaties against 16,704 executive agreements that never touch the Senate floor at all; in the post-WWII era, treaties have made up less than 7% of U.S. international agreements. The two-thirds check hasn't been repealed. It's just been made optional by relabeling almost everything an executive agreement instead.
Sole power to try impeachments
Article I, Section 3"The Senate shall have the sole Power to try all Impeachments," with a two-thirds conviction threshold — a deliberately high bar, and the Senate has never lowered it or found a procedural workaround. The design has held structurally through four presidential impeachment trials (Andrew Johnson, Clinton, and Trump twice) without a single conviction, in every case falling along near-total party lines. That's not a broken mechanism — the two-thirds requirement is doing exactly the counter-majoritarian work it was built to do. What drags the grade down is that every modern trial has functioned as a foregone partisan head-count rather than the deliberative proceeding the framers described, which is a real erosion of the clause's spirit even where its letter holds.
Expulsion of members
Article I, Section 5A two-thirds vote can expel a sitting senator — deliberately hard to reach, and used only 15 times in the chamber's history: William Blount in 1797 for conspiring to help Britain seize Spanish territory, and 14 more during the Civil War for supporting the Confederacy. Nobody has been expelled since 1862. That's not the clause failing; the 2024 case of Sen. Bob Menendez shows it working indirectly — facing a bribery conviction and Democratic colleagues publicly organizing to expel him, he resigned rather than face the vote he was expected to lose. The formal power sits almost entirely unused because the threat of it still functions.
Quorum and journal-keeping
Article I, Section 5A majority constitutes a quorum to do business, and the Senate must keep and publish a journal of its proceedings. Both requirements function as basic housekeeping today without major controversy — quorum calls are sometimes used as a procedural stalling tactic rather than a genuine head-count, but no sourced case from the current Congress shows this rising to a real breakdown of the requirement itself. The most mechanical piece of Article I on this list is also the one working closest to as-written.
What the pattern shows
Separate the ten entries into what’s about who the Senate is and what’s about what the Senate does, and a clear line appears. Composition, terms, qualifications, and the Seventeenth Amendment — the mechanics of who sits in the chamber — all score in the high 80s and 90s. They are the oldest, most literal parts of Article I, and they haven’t moved.
Everything about how the Senate exercises its actual constitutional powers scores worse, and the pattern isn’t partisan — it’s structural. The filibuster (35) isn’t in the Constitution at all; it’s the Senate’s own rule, expanded by both parties’ nuclear options until routine legislation needs a supermajority the framers never wrote. The treaty power (28) hasn’t been repealed, just quietly bypassed by relabeling nine agreements out of ten as something that skips the Senate floor entirely. Nominations (55) run through a home-state veto custom that both parties invoke when it suits them and resist when it doesn’t. Even impeachment (65), the cleanest of the big constitutional levers, now runs on a predictable party-line headcount instead of the individualized judgment Article I describes.
The last word
A 68 is neither an indictment nor a clean bill of health — it’s what you get when a two-century-old design still works exactly as intended in its bones and has been substantially renegotiated in its actual exercise of power. The Senate that votes today is still the Senate the Great Compromise built: two members per state, six-year staggered terms, the same eligibility bar since 1789. It is not the same Senate when it comes to what those members actually do with the seat once they’re in it. Treaties became executive agreements. A supermajority rule nobody voted into the Constitution now governs most legislation. Impeachment trials became headcounts.
None of that required amending the Constitution. It only required each majority, in turn, deciding the existing rules were inconvenient — and finding that changing how the Senate uses its power is quieter, and easier, than changing what the document actually says.
Ubi non est lex, non est transgressio — where there is no law, there is no transgression. Nothing catalogued here breaks a law. It’s a chamber quietly relocating where its own power actually lives, one rule change and one executive agreement at a time.
Sources & methodology:
- Article I, Sections 3, 5 and Article II, Section 2 — U.S. Constitution
- Filibuster and cloture rule history: CRS RL30360, “Filibusters and Cloture in the Senate”; cloture-motion counts by Congress: Senate.gov cloture tables
- 2013 and 2017 nuclear option precedents: NBC News; CRS R44819
- September 2025 en bloc confirmation rules change: CNN, Sept. 11, 2025
- Blue-slip practice and 2025 disputes: CRS R44975; Courthouse News, Dec. 2025; Iowa Capital Dispatch, July 31, 2025
- Treaties vs. executive agreements: CRS RL32528, “International Law and Agreements: Their Effect upon U.S. Law”
- Expulsion history and the Blount case: Senate.gov, “Expulsion and Censure”; Menendez resignation: CNN, July 23, 2024
- State population figures: U.S. Census Bureau, 2025 state population estimates
- Seventeenth Amendment ratification: National Archives, Milestone Documents
- Quorum procedure: CRS 96-452, “Voting and Quorum Procedures in the Senate”
- Committee seniority norms: CRS R46806; Axios, July 22, 2026
- The grades are the journal’s editorial judgment; every underlying factual claim above is independently sourced as cited.