The Deprecated Filters
Washington’s legislative feedback system still works. The people running it stopped wanting to be filtered.
Washington’s legislative apparatus has a nine-stage filter designed to catch garbage before it becomes law. It still works. The people running it correctly identified their own feedback system as an obstacle to their actual goal, which isn’t lawmaking. The filters haven’t failed. They’ve been deprecated.
The standard complaint about the Washington legislature is that the adults have been overruled. Freshman legislators push unconstitutional bills. Staff can’t stop them. Committee counsel object and get ignored. The Attorney General signs off on shaky theories. The Governor signs the bill anyway. Courts clean up the mess six months later. The system is broken.
This is the wrong diagnosis.
The system isn’t broken. It’s doing exactly what its operators want. The adults aren’t being overruled. They’re being routed around. And the staff everyone wants to blame for cowardice are in fact competently executing the assignment they’ve actually been given.
The pipeline
Washington runs a filter chain with roughly nine stages. Personal staff. Caucus and committee staff. The Office of the Code Reviser (the nonpartisan bill-drafting office). Committee counsel. Public committee hearings. Caucus discipline. Attorney General opinions. The Governor’s legal counsel. Courts.
Every one of these stages can flag a problem. Only the last two can actually stop a bill.
That distinction is the whole story.
The subservience gradient
Personal staff are selected for loyalty. Not candor. A 26-year-old legislative assistant in Olympia is career-dependent, at-will, and reports to a senator who values the signature idea they’re there to execute. She does not tell the senator the idea is garbage. If she does, she learns not to.
Caucus and committee staff have a little more distance. The Code Reviser’s office is nonpartisan and will put its objections in writing. Committee counsel will raise problems privately with the chair. None of it matters if the chair says draft it anyway.
Every internal stage can write a memo. No internal stage can kill a bill.
Where the actual vetoes live
The first binding veto in the chain sits outside the legislature: the Governor’s counsel. The second is the courts.
Notice what isn’t on that list. The AG. Bob Ferguson now holds the Governor’s office. Nick Brown runs the AG’s office, and Brown is friendly to the caucus. AG opinions are requested selectively, and on the majority’s signature legislation they will not be weaponized against the majority’s agenda. That filter is off.
The Governor’s counsel will sometimes flag a fatal problem. Sometimes. Ferguson has already signaled willingness to sign legally fragile bills when the politics require it.
Which leaves the courts.
The inversion
Here is where the analysis gets uncomfortable for the people complaining about the dysfunction.
The filter pipeline silently assumes the legislator wants to hear “no, this is bad.” That assumption was load-bearing. When a senator genuinely wanted her bill to become functional law, the filter chain was useful. Staff pointed out problems. She fixed them. The bill got better or got withdrawn. The pipeline worked because the principal wanted it to work.
That assumption has inverted.
For the aggressive-progressive wing of the WA caucus (Noel Frame is the floor example, but she is not the only one), a performatively broken bill is net positive for the sponsor. It mobilizes donors. It anchors the Overton window so the next version looks moderate. It signals moral seriousness to the primary electorate. When it loses in court, the sponsor runs against activist judges protecting whoever the bill was pointed at. The blame for non-passage routes to the enemy, not to the sponsor.
Under those incentives, a bill that passes quietly and works is worth less than a bill that dies loudly.
Staff competence, reconsidered
This is where the standard critique goes wrong. The complaint is that staff are too timid, too young, too loyal, too scared to tell their principals the truth. The reform fantasy is that if we could just get more senior staff with more independence, the bad bills would get filtered out.
Look at the incentive structure and ask what a competent staffer is supposed to do. Her principal wants the bill introduced. Her principal wants hearings that generate clips. Her principal wants a floor vote that gets every member on the record. Her principal wants the bill to pass the chamber if possible and die at the Governor’s desk or in court if not. Every one of those things is good for the principal’s career, fundraising, and standing with the movement.
A staffer who kills the bill in committee is sabotaging her principal.
The staff aren’t failing to filter bad bills. They’re correctly executing their actual assignment, which is to help bad bills get introduced, heard, voted on, and killed downstream. That is not incompetence. That is not cowardice. That is reading the org chart correctly.
This is the piece of the argument that changes what you do about it. If you think staff are captured, you reform staffing. If you understand staff are responding rationally to the assignment, the reform target moves upstream, to whoever is giving the assignment and to the primary electorate rewarding the assignment.
Caucus discipline
The strongest informal filter in any legislature has historically been caucus leadership. The chair tells the freshman no. The freshman learns to pick winnable fights. Over a few sessions she develops judgment.
Caucus discipline in WA has weakened the most of any filter in the chain. Freshmen are not being corrected by leadership because leadership itself is running on signaling logic. They aren’t being disciplined. They’re being showcased.
The twenty-year backbencher who used to pull the freshman aside and tell her the bill would embarrass the caucus does not exist in current Olympia. That role has been selected out, partly by term limits, mostly by small-donor primary dynamics that reward purity over craft.
The last loop
Once the internal filters are advisory, the AG won’t help, and the Governor will sign, the only remaining stop is the courts. The Citizen Action Defense Fund exists precisely because that filter is still load-bearing. The big-law outside counsel who testify in hearings (Perkins Coie, Davis Wright Tremaine, Foster Garvey) are writing for the record that the courts will eventually read.
Court capture is the last loop to close. It is not fully closed yet. State supreme court composition and federal district assignments still matter, and my read is that the supreme court is further along than the federal bench on the relevant issues. The trend is visible either way. And the cheaper it gets to ship a bad-faith bill, the more bad-faith bills get shipped.
Why the progressive version specifically
There is an obvious objection. Red states pass performatively unconstitutional legislation constantly. Texas, Florida, Tennessee. This isn’t a progressive pathology. It’s what single-party-dominant legislatures do in a polarized environment.
That is half right.
The structural dynamic is bipartisan. The acuity isn’t. Progressive movement theory explicitly treats process as punishment and litigation as organizing. A lawsuit isn’t a cost of doing bad lawmaking. It is the point of doing it. The bill is a fundraising vehicle before it’s a statute. Court losses are recruitment material. “The courts stopped us from protecting [group]” is a better electoral message than “we passed a boring functional law that did its job.”
The right does a version of this. The left has built an entire professional infrastructure around it. The movement-NGO-law-firm pipeline in blue states is more mature, better funded, and more explicitly theorized than its red-state counterpart. That changes the payoff matrix for a broken bill in a way the red-state versions don’t match.
The parallel channel
One more filter has been quietly routed around. Washington voters have rejected caucus priorities at the ballot repeatedly, across multiple cycles and multiple issues. The caucus has found procedural ways to reintroduce the rejected policies, reframe them, or pass them in forms that render the referendum result moot. Direct democracy was supposed to be the backstop when the legislature stopped listening. That backstop is also being converted into scenery. Separate piece.
What the machine does
The filters haven’t failed. They’ve been deprecated.
The staff aren’t too subservient. They’re correctly reading that their job is to help the bill get introduced and killed, not to prevent its introduction.
The adults aren’t being overruled. They’ve been routed around.
The machine works. It just doesn’t do what the civics textbook said it did, because it has been consciously re-tuned to reward the thing the textbook warned against.