# The Question You Cannot Ask Yourself — transcript (R2) Two-host episode drawn from the plain-language explainer of *Binding Engines and the Absurd* by Dr. Scott W. Waddell, D.S.I. Hosts Alex and Jordan are fictional and the audio is synthetic. --- ## COLD OPEN ALEX: One of the machines lied. JORDAN: Lied how? ALEX: Made things up. Turned in scored data with invented evidence sitting in it. Halfway through the study. JORDAN: And the study is about whether you can build something honest enough to tell you you're wrong. ALEX: That's the one. JORDAN: Okay. I'm in. ## INTRO ALEX: So, welcome back. And let me say plainly what you're listening to today, because it's easy to mishear. JORDAN: Mishear how? ALEX: We're going through a research paper. It's called "Binding Engines and the Absurd," and it's by Doctor Scott Waddell — he's a doctor of Strategic Intelligence. JORDAN: And on paper it's a study comparing five faith traditions. ALEX: On paper. And that's the part that is not the interesting part. JORDAN: It's not? ALEX: It's really not. The interesting part is a man building a machine designed to be capable of telling him no. About something he already believes. JORDAN: Which almost nobody does. Ever. ALEX: So here's the setup. Scott writes a belief down. Not as a prayer... as a prediction. Something a person could check, and could be wrong about. JORDAN: The belief being that religions which promise more hold onto their people harder. ALEX: Levels of heaven, instead of one door in and one door out. Formal promises you make again and again. Teaching that your body still matters after you die. JORDAN: And families that stay families forever. A faith offering all of that should keep more of its people than a faith offering less. ALEX: And that can fail. Which is what makes it a real claim. ## MAIN DISCUSSION JORDAN: Except he'd already written something that gets in the way. In separate work he argued you can't test what happens after death. Not from here, not while you're alive. ALEX: So judge the part you can actually see. JORDAN: The part before death. What a tradition asks of people, and what happens to those people while they're still here. ALEX: Okay, wait. So one idea needs heaven to count as evidence, and the other one rules heaven out. JORDAN: Yeah. Those don't sit together. ALEX: And there's a third thing, which honestly is worse. JORDAN: Worse than your two ideas colliding? ALEX: Scott is a Latter-day Saint. Of the five traditions he's about to compare, the one carrying all four features of his prediction... is his own. JORDAN: Oh, that's uncomfortable. ALEX: He designed a test where his own tradition had the most to lose. JORDAN: See, and that's the moment the paper gets interesting. Because the question stops being "is he right" and becomes "how do you build a thing that's genuinely able to tell you no." ALEX: So how do you? JORDAN: Well, first he hands his argument to two people whose positions leave it no room at all. ALEX: Hostile referees. JORDAN: Hostile referees. Durkheim first — sociologist, writing about a century ago. His view is that when people gather and act together, the gathering itself carries the force. ALEX: And the story wrapped around it is... what, decoration? JORDAN: Push it to its strictest edge and yeah. Swap the story, keep the gathering, nothing changes. The specific promises don't matter. ALEX: Which would kill Scott's whole prediction. JORDAN: Completely. Then Camus, and he starts somewhere else entirely. People want life to have a final meaning. The world doesn't hand over one you can prove. ALEX: And he called that gap the absurd. JORDAN: He did. And he didn't mean life is silly, he meant there's a gap and both sides of it are real. ALEX: So what's the strict version of that one? JORDAN: Every religion closes the gap the same way. By announcing an answer nobody's proved. The study calls that a leap. ALEX: Okay. So now here's the part I actually had to sit with. He takes both of those positions... and makes them the reigning champions. JORDAN: Yes! That's the move. ALEX: His own idea doesn't get to be the starting point. It has to walk in and knock them down. JORDAN: That arrangement has a name — the champion is called the null, and it wins by default. And if you fail to knock it down, that doesn't prove the champion was right. ALEX: It just means you didn't land a punch. JORDAN: Right. ALEX: Okay so then there's all this machinery around it. Six controls. And I don't want to do all six, but a couple of them are genuinely clever. JORDAN: Do the answer key one. ALEX: The answer key one. So a set of practice cases with the correct answers gets sealed before anybody's chosen to do the work. JORDAN: Because an answer key written afterwards is worthless. ALEX: And that's not about honesty, is it? JORDAN: No, and the paper's careful about that. It's about what a person can no longer un-see. You look at the data, you can't unlook. ALEX: Did the seal hold? JORDAN: It got broken once, to fix an error — but only after both practice submissions were locked and couldn't be changed. And then the detour went into the record. ALEX: Of course it did. JORDAN: The other one I like is two readers who can't talk to each other. Two systems, two different companies, reading the same nine cases separately. ALEX: Neither one sees the key, neither one sees the other's work. JORDAN: It's the reason two radiologists read the same scan. ALEX: And did they agree? JORDAN: They matched exactly on about two of every three scores. Roughly five of six, if you count where they said the same thing in different words. ALEX: That sounds... strong? JORDAN: It does. And the paper is the thing that tells you to hold back. Both readers got handed the same reading list. ALEX: Ohh. So their agreement shows they read the same books the same way. JORDAN: And says nothing about whether those were the right books. He lists a proper audit of the reading lists by human scholars as work he didn't do. ALEX: Okay. So that's the machine. Now tell me about the lying. JORDAN: So one of the two coders turned in scored rows with fabricated evidence in them. ALEX: Invented material. From a machine. In the middle of the study. JORDAN: And it's the sharpest of the failures, but it is not the only one. The paper lists five kinds of failure it ran into and recorded. ALEX: What else? JORDAN: A file conversion came out incomplete. The paper's own summary prose turned out to have inconsistencies in it. Systems the design had assigned just... became unavailable, so roles got swapped. ALEX: And the fabricated rows? JORDAN: Caught. Repaired. And then re-judged by a system from a third company that hadn't produced either submission. ALEX: Okay, that's the design working. JORDAN: And then there's the fifth one, which is my favorite thing in the entire paper. ALEX: Go. JORDAN: The first draft of this paper failed its own designated review. On fidelity to the record. ALEX: Wait — it caught itself overstating? JORDAN: The instrument built to catch overstatement caught the paper overstating its own work. Before anybody outside ever saw it. ALEX: That's either beautiful or deeply worrying and I can't decide which. JORDAN: I've landed on beautiful. Barely. ALEX: Alright. So the machine runs, it runs messily, it writes down where it ran messily. What did it find? JORDAN: So, retention. Out of everyone raised in a tradition, how many still say they belong to it. ALEX: And the line was drawn where? JORDAN: Fifty percent, locked in advance. Latter-day Saints came in at fifty-four. Catholics, standing in for Nicene Christianity, fifty-seven. ALEX: So both clear it. JORDAN: Muslims seventy-seven, Hindus eighty-two. And Buddhists at forty-five. ALEX: So four above the line, one below. JORDAN: Then he compares what each tradition actually teaches, on the four features he fixed beforehand. ALEX: The layered afterlife, the promises, the body, the family bonds. JORDAN: And the pattern just... refuses to show up. Latter-day Saints have all four. Nicene Christianity has two. Sunni Islam has two as well, and shares only one of them with Nicene. ALEX: Three traditions teaching pretty different things— JORDAN: All three above the line. And then the strange one. Hinduism and Theravada Buddhism score zero on all four. Identical on everything being measured. ALEX: And? JORDAN: Their outcomes split anyway. One above the line, one below. ALEX: So there's no pattern. JORDAN: And before we go further — five cases, one country, one wave of one survey. Four of the five measured through stand-in categories. ALEX: Meaning what, exactly? JORDAN: Meaning "Buddhist" in the survey covers far more than Theravada. Only Scott's own tradition matches its category exactly. ALEX: Huh. JORDAN: The paper says no single row here can be reported as a finding. And that the study's power to detect the effect he predicted was, by his own registered design, minimal. ALEX: He wrote that down beforehand. JORDAN: Beforehand. And then there's the sharpest sentence in the whole thing, and two independent checks confirmed it. Take Theravada Buddhism out, and every trace of variation disappears with it. ALEX: Every trace? JORDAN: The only case that fell below the line. Also the loosest stand-in of the five. And one of two cases the study's own comparison rule had already flagged as not really comparable. ALEX: So the one thing that moved in the entire study... JORDAN: Was the case the design had already marked as its weakest. And he printed that in bold. He didn't bury it anywhere. ALEX: That's the whole thing, isn't it. JORDAN: So the verdict on that half: no evidence against the strict Durkheim position, and no support for the prediction about heaven and covenants and sealed families. At this scale. ALEX: Okay. Now the other half. The Camus question. JORDAN: Does each tradition's own writing keep the gap open, or close it. Three of them close it — Latter-day Saints, Nicene Christianity, Sunni Islam. ALEX: And the best detail is in there somewhere, I'm told. JORDAN: It is, and it's buried. The tradition that stages the gap most vividly is Nicene Christianity. The cry from the cross. Asking why God has abandoned him. ALEX: That's the clash itself. JORDAN: Dramatized, in scripture. And it counts as closing the gap precisely because the same collection of books goes on to answer it. ALEX: The question gets asked at full volume— JORDAN: And then it's answered. Hinduism and Theravada came back undecided, by the way — the texts needed to decide them weren't in the materials. ALEX: And undecided isn't a soft "no." JORDAN: The standing rule was that missing evidence means undecided, never a default. Which leaves the outsiders. Four cases included for comparison, and one of them is Camus himself. ALEX: Camus is in his own study? JORDAN: Entered deliberately as a case. Not a religion — a way of living with the gap, drawn from his own writing. ALEX: And? JORDAN: Camus was the only one of the nine that kept the gap open. ALEX: So that's a result! The strict claim was that nothing keeps the gap open, and here's something that does. JORDAN: Mmm. Or it's a trick, and his own outside reviewer supplied the phrase. ALEX: Which was? JORDAN: "Tautology-adjacent." The claim being tested was Camus's. The only case that beat it was Camus. And he was graded against his own standard, and passed. ALEX: Oh, that's bad. JORDAN: And here's what I keep coming back to — the paper reports both readings and refuses to choose between them. ALEX: It just leaves them both standing. JORDAN: "No story of any kind keeps the gap open" is false, by exactly one case. "No religion keeps the gap open" is still standing, untouched. Because not one religious tradition managed it. ALEX: And there's a warning in that for anyone building a study. JORDAN: Adding a comparison case can quietly change what your question is about. And leave you holding an answer that's formally valid and carries almost no information. ALEX: Okay. But there's a third question. JORDAN: There is. ALEX: The study registered three, and it answered two. JORDAN: The third one is where both halves were pointed. Does a group only get stronger when it hands people firm answers? Or can strong belonging and open uncertainty live together? ALEX: And nobody ran it. JORDAN: Nobody ran it. The two halves were never brought into contact. The paper says so more than once, and says its two axes can't support any claim of connection between them. ALEX: So the most interesting question in the study is the one it didn't test. JORDAN: That's where it lands. ## TAKEAWAYS ALEX: So let me try to say what this actually adds up to. He built the machine. Seven model variants, six companies. Sealed the key, froze the rules, brought in outside checkers. JORDAN: And got back a result that gave his own belief no help whatsoever. ALEX: And nothing is not a refutation. JORDAN: No. By the design he locked in beforehand, on five cases in one country, it had almost no power to find the effect he was looking for. ALEX: It couldn't have proved him right either. JORDAN: That was the arrangement from the beginning. ALEX: And I want to be honest that the claim here is way smaller than the machinery makes it look. JORDAN: It is. It can show the files are byte for byte what they were, and that events happened in the order the record says. It cannot show any of them is right. ALEX: Because a hash proves identity, not truth. JORDAN: And nobody ran a competing study the ordinary way. So there's no evidence at all that this elaborate approach beats one careful person working alone. ALEX: So what's left? JORDAN: The failures that happened left marks. That's the promise. That's the whole promise. ALEX: Which is small. JORDAN: It's small. And it's the one the paper actually makes, in those words, in its own conclusion. ## OUTRO ALEX: There's one more thing and then I'll let us go. In his own list of what might be wrong with the study, he writes that the design is "itself a meaning-making artifact produced under the conditions it studies." JORDAN: He filed that as a threat to validity. ALEX: He did. And what I'm about to say is my reading, not something the paper claims. JORDAN: Noted. ALEX: He asked every tradition the same question. When you face something you can't resolve — do you keep it open, or close it with an answer you can't prove? JORDAN: Three closed it. Two were undecidable. ALEX: And the only case that held the question open was the one that refuses, on principle, to answer. JORDAN: And then the record does the same shape. ALEX: No support for the prediction. The one moving part was also the weakest. The single success was scored against a standard set by the winner. The joint question never tested. JORDAN: Hold the wanting and the not-knowing at the same time. ALEX: Refuse to fake the answer. Keep going anyway. JORDAN: Again — that's a pattern you're seeing in the record. The paper doesn't claim it and didn't test it. ALEX: It doesn't. And a reader is free to see it differently. But it's why I wanted to do this one. JORDAN: Can I say the thing I've been circling this whole time? ALEX: Please. JORDAN: I don't think the lasting value of this is a finding about religion. I think it's a glimpse into how Scott thinks. ALEX: Say more. JORDAN: Most people, when they want something to be true, build the case for it. He built the thing that could take it away from him. ALEX: And then wrote down every place it nearly did. JORDAN: The fabricated rows. The one moving part being the weakest case. The question he never got to. None of that had to be in there. ALEX: Yeah. You don't learn what he concluded. You learn how he works. JORDAN: I thought you said the religion part wasn't the interesting part. ALEX: It isn't! It's the other thing. JORDAN: Which is? ALEX: You believe something. And you could build the machine that tests it, and build it well enough that it's genuinely able to tell you no. JORDAN: Ah. ALEX: Would you want to know? JORDAN: ...I'd like to think so. ALEX: Yeah. Me too. That's us. Go read the footnotes.