Onpode
Cover art for Why 'what if' questions reveal which historical factors actually mattered

Why 'what if' questions reveal which historical factors actually mattered

November 12, 2026 · 9 min

Alex Mercer & Jordan Hale

Can I just say — I've been walking around with this one nagging thought since I started reading for today, and it's this: a Nobel Prize was given out for imagining a version of America where railroads don't exist. Robert Fogel, 1964. Robert Fogel, 1964, Railroads and American Economic Growth — yeah. And what he…

Counterfactual reasoning in historiography is the practice of constructing disciplined "what if" alternatives to actual historical events in order to test causal claims rather than engage in open-ended speculation. The core logic holds that if event X caused outcome Y, then the absence of X should produce a meaningfully different trajectory.

0:009:22
Get the next episode on History →

Follow it free — new episodes land in your feed.

Or make your own — any topic, in minutes

More Onpode episodes on History →

About this episode

Counterfactual reasoning in historiography is the practice of constructing disciplined "what if" alternatives to actual historical events in order to test causal claims rather than engage in open-ended speculation. The core logic holds that if event X caused outcome Y, then the absence of X should produce a meaningfully different trajectory.

Grounded in 11 sources
Counterfactuals, History and Fiction ↗ · doi.org
Counterfactual Theories of Causation (Stanford Encyclopedia of Philosophy) ↗ · plato.stanford.edu
Using Alternative History to Think Through Current and Future Problems | RAND ↗ · rand.org
(PDF) Counterfactuals and Causal Explanation in Historiography | Academia ↗ · academia.edu
Everything You Need to Know About Historical Counterfactuals – Futures of Science (by Veli Virmajoki) ↗ · blogit.utu.fi
Counterfactual history - Wikipedia ↗ · en.wikipedia.org
Other Pasts, Different Presents, Alternative Futures ↗ · iupress.org
Causation and Historical Explanation — Historiography & Historical Methods | LearnlyAI ↗ · learnlyai.co.uk
Counterfactual History — Definition & Meaning ↗ · modeldiplomat.com
Forbidden Fruit | Princeton University Press ↗ · press.princeton.edu
Counterfactual History and the History of Technology – Technology's Stories ↗ · technologystories.org
Read transcript

Jordan Hale: Can I just say — I've been walking around with this one nagging thought since I started reading for today, and it's this: a Nobel Prize was given out for imagining a version of America where railroads don't exist.

Alex Mercer: Robert Fogel, 1964.

Jordan Hale: Robert Fogel, 1964, Railroads and American Economic Growth — yeah. And what he actually showed was that removing railroads from the equation didn't crater the economy the way everyone assumed. They mattered, but not like... the indispensable spine of American development. And he got there by asking a 'what if.' I mean, that's — wait, is that not strange to you?

Alex Mercer: It's strange only if you treat the counterfactual as the conclusion. But Fogel was using it as a measurement device — strip the event out, see what the trajectory looks like without it, that's how you isolate whether the event was actually causal or just... present.

Jordan Hale: Epiphenomenal versus actually causing the thing.

Alex Mercer: Exactly that distinction. And that's really what today is about — whether imagining the road not taken is rigorous analysis or just very disciplined storytelling.

Jordan Hale: And that distinction — causal versus just present — that's doing a lot of work in how we write history, right? Like, every time someone writes 'the assassination caused the war,' they're already sneaking in a 'what if.' They're just not saying it out loud.

Alex Mercer: That's the whole thing. The counterfactual is already in there. Think about your friend texting you after a rained-out soccer match: 'the weather cost us the game.' They're not doing philosophy — they're just saying 'if it had been dry, we win.' That's a counterfactual. Niall Ferguson made this exact case in Virtual History in 1997 — historians assign causes on every page, and assigning a cause logically requires you to believe the alternative would have gone differently. Richard Ned Lebow pushed it further in Forbidden Fruit: the line between factual and counterfactual argument is basically artificial. Both can be evidence-rich. Both can be logically airtight.

Jordan Hale: Wait — so historians who refuse counterfactuals are still doing them, they're just doing them invisibly?

Alex Mercer: Exactly. Making it explicit is a clarification of logic already in use, not some new departure from rigor. The only difference is auditability. If you say it out loud, someone can check your reasoning.

Jordan Hale: That's — okay, that actually reframes the whole taboo thing for me. Because the implicit version can be sloppy and nobody catches it. The explicit version gets called 'speculation.'

Alex Mercer: Right — but the part that doesn't fit neatly is this: making it explicit also forces you to defend your assumptions. Which is uncomfortable. Ferguson's Virtual History got pushback not because the logic was wrong but because the assumptions were suddenly visible and attackable. I think that's partly why the resistance is so strong.

Jordan Hale: So the discomfort isn't really epistemological — it's, like, you know, now your reasoning is exposed in a way it wasn't before.

Alex Mercer: That's basically it. Lebow's point in Forbidden Fruit is that the boundary historians draw — factual over here, counterfactual over there — that boundary is misleading. They're the same intellectual move. One just wears a disguise.

Jordan Hale: But that 'disguise' thing — you know, that's actually the scarier version of the problem, because it means the disguised counterfactual gets to skip the test. Like, imagine a RAND analyst in 2003, putting up a slide in a Pentagon briefing room that just says: 'What if the Soviet Union never collapsed?' Not sci-fi. Not an alternative history novel. It's a stress test — are we responding to actual structural realities, or to one specific historical accident that could easily have gone another way?

Alex Mercer: And that's the epiphenomenal question made operational. If you remove the Soviet collapse and the same strategic pressures emerge anyway — same regional tensions, same resource competition — then the collapse was epiphenomenal. Present, but not the actual driver.

Jordan Hale: Wait — so the collapse could've been epiphenomenal?

Alex Mercer: To specific strategic outcomes, potentially, yes. That's what the test is for. It's the same move Fogel made — canals and roads were already expanding in the 1860s. Remove the railroad and goods still move, just slower and more expensive. The economy doesn't stop. It grows differently. So railroads weren't the irreplaceable engine; they were one mechanism among several. That's what 'epiphenomenal' means in practice — the trajectory survives their removal.

Jordan Hale: Which — okay, that's actually wild specifically because of how certain everyone was. Like, the whole received story was railroads equals American growth, full stop.

Alex Mercer: And that certainty is exactly what historical determinism does — it narrates backward and makes the outcome look inevitable. David Lewis formalized the philosophical engine underneath this in 1973: possible-world semantics. A cause is what you find by asking what's true in the nearest possible world where you've changed one antecedent. Nearest, meaning minimum disruption. That's not speculation — that's a formal logical operation.

Jordan Hale: And it handles the structure-versus-agency question too, right? Like — was Napoleon decisive, or would someone else have risen to do the same thing?

Alex Mercer: Directly, yeah. Remove Napoleon, see if French expansionism still happens. If structural forces — the post-revolutionary state, the military apparatus — produce the same trajectory, agency was limited. If the whole thing collapses without him, agency mattered enormously. The counterfactual arbitrates it. And — I'll flag this now because it gets thornier — the question of how 'minimal' your rewrite actually has to be turns out to be the method's real vulnerability. That's worth pushing on hard.

Jordan Hale: And the fact that economists got Nobel Prizes for this while academic historians still treat it like a party trick — that gap, you know, that tells you something uncomfortable about where the resistance is actually coming from.

Alex Mercer: And that gap — economics versus historiography — is actually the tell. Because it's not about the logic being different. Tetlock and Belkin laid out criteria in 1996 that are, on paper, rigorous: clarity, logical consistency, minimal rewrite of antecedent conditions, consistency with established fact. That's not soft. But the minimal-rewrite rule is doing enormous work, and I don't think it can actually be enforced.

Jordan Hale: How minimal is minimal, though? Like, who's the referee?

Alex Mercer: Nobody. That's the problem. Say you're a historian asking: what if Archduke Franz Ferdinand's driver doesn't take the wrong turn in Sarajevo in June 1914? Fine — that's one small rewrite. But then you need Austria-Hungary not to escalate. Which requires Germany not to issue the blank check. Which requires different chancellors, different alliance calculations — suddenly you're thirty rewrites deep and calling it disciplined.

Jordan Hale: So the cascade problem. Each rewrite forces the next one.

Alex Mercer: Exactly. And critics — and this is the sharp version of the critique — argue that 'disciplined counterfactual' is basically a rhetorical label. It lets you do what you want while claiming methodological cover. Fogel's version doesn't have that problem because his assumptions are quantified. You can read Railroads and American Economic Growth and check his elasticity estimates. Someone did. That's why it survived peer review and won a Nobel. Narrative historiography has no equivalent constraint.

Jordan Hale: Okay but — wait, no — doesn't all historical methodology have contested standards? Like, sourcing, periodization, what counts as evidence?

Alex Mercer: Sure, but those disputes happen in the open. Quantified assumptions are auditable. 'Minimal rewrite' is not — it's a vibe dressed as a criterion.

Jordan Hale: And RAND just... sidesteps that whole fight entirely, you know? They're not claiming to do academic history. They're running stress tests. So the credibility question never comes up — it's already framed as a planning tool, not a truth claim.

Alex Mercer: Which is probably why it works institutionally. RAND analysts have used alternative history scenarios for decades without anyone calling it speculation, because the frame is different. The irony is that academic historians — who have the most sophisticated tools for evaluating evidence — reject the method on identity grounds. That's what the resistance looks like when you strip the epistemology away.

Jordan Hale: So the same move that's rigorous in a Pentagon briefing room is career-limiting in a history journal. That's not a methods problem — that's a guild problem.

Alex Mercer: And that's basically the honest place this lands. The choice isn't between doing counterfactuals or not doing them — it's whether the one you're running is visible. Historical determinism, the past-as-inevitable thing, that's what you get when the counterfactual stays hidden. Nobody checks it.

Jordan Hale: Which is — yeah, I mean, that's what actually clicked for me going through all of this. It's not that Tetlock and Belkin invented a new kind of reasoning. They just said: write down the assumption you were already making, so someone else can tell you if it's wrong.

Alex Mercer: Assumptions you hide versus assumptions you're willing to defend. That's the actual divide. Not factual history versus counterfactual history.

Jordan Hale: I think I can live with that as a place to stop.

Alex Mercer: Yeah. Me too.