The Bucolic Green Pastures of Empiricism

My girlfriend tells me I should care more about current events. Here is another post that has nothing to do with the news. Sorry.

I struggled coming up with a good title for this post. I wish I had a time machine so I could go back to the 1920s, capture Ernest Hemingway, and chain him to the wall in my basement. Whenever I needed a title, I would walk down the stairs with an expensive bottle of whisky…

---

In the spirit of Jordan Rubin’s great substack, here’s a summary of the post for those who prefer a higher signal-to-noise ratio:

TL;DR

Some traditions feel more “objective” or “subjective” because of how their institutions structure their assumptions. Specifically, it is the amount of implicit assumptions in conflict which, on the margin, determines the feeling of objectivity or subjectivity.

I start with a simple view. Truth = what follows from accepted axioms. I then show why persistent disagreement often comes from conflicting base assumptions. I initially claim that high stakes explain why some traditions converge, but revise that after a strong antithesis using Rubin’s @antithesize coding skill. Stakes are now necessary but not sufficient.

The stronger model has three parts:

Operationalizability: Can claims be clearly defined and tested?

Verification infrastructure: Are there shared methods for settling disputes?

And of course, stakes: Are there incentives to build and maintain (1) and (2)?

Fields with high measurability like math, engineering, aviation, and medicine tend to converge more. Low-measurability but high-stakes fields like politics and ethics tend to polarize, splinter into camps, or lean on an authority which monopolizes force.

Main takeaway: if you want more objectivity in any tradition, build better measurement and verification systems. Also, the world can plausibly be explained without appealing to moral realism. Also, convergence in traditions, like a higher volume of 3-point attempts or the marvelization of cinema, can be explained well by this theory.

---

In 2026, it’s actually contrarian to not be a moral realist. (Moral realism means you think there are stance-independent moral truths) The moral realists who think they’re edgy are already behind the curve. Accomplished substacker and friend Xavier Lauriault recently wrote: “When engaging in field experimental philosophy, asking if there are moral truths that exist independently of our opinions, the answer is nearly always. “No.”” True, Xavier finds that most people are moral anti-realists. In this way, it makes the moral realists more contrarian. My enlightenment heroes rest peacefully in their graves. However, 62.07% of professional philosophers disagree with these university students! So in the ivory tower, moral realism is the conventional view! The horror.

Montesquieu, Newton, Voltaire, and David Hume in disbelief at the dearth of empiricism among contemporary philosophers.

In debating these hubristic moral realists, reminiscent of the anti-Copernican Catholics, I have generated my own meta-epistemological theory. My theory, if successful, will explain why certain traditions like film watching seem much more “subjective” than math or chess, what makes statements in morality, math, or aesthetics “true”, and how any tradition, if it so chooses, can become more objective.

As an outside observer of meta-ethics and epistemology, I acknowledge my argument’s weirdness. (Or as Nick said, “quite interesting, but probably wrong.”) But here it is nonetheless. I pray to my enlightenment gods to give me the strength to defeat these presumptuous realists. I will go over counter arguments after the sketch.

First let’s establish what truth means. In my experience this word becomes a problem quickly. Without a definition people get confused. But then again, whenever I ask for a definition, or try to come up with one, the conversation seems to trail on forever.

So I have come up with a very simple definition: X is true when X follows from one or multiple axioms. Truth is simply internal consistency. For example, 2 = 2. This statement is true because you assume the equality rule/axiom, which is that X = X. Likewise, if my coauthor Nick was going 90 mph in a 70, then the statement: “Nick was speeding” would be of course true by virtue of the legal rule + the stated fact. If the observation that he was going 90 mph is proven true, then that premise + the legal rule makes the statement true. Likewise for the definition of a triangle or the rules of the NBA.

Why did I choose this definition?

I’m glad you asked.

One might counter that you can evaluate assumptions as being true or false. If I assume I’m purple, someone could say that’s a false assumption! But why, because it doesn’t fit reality? So you’re leaning on perception. To trust perception, you lean on assumptions about perception. To trust those assumptions, you lean on… you see where I’m going. It seems there are stopping points where some claims are taken as basic.

So, since your epistemology is just assumptions (and claims which follow from assumptions) all the way down, what explains why some traditions feel more objective than others? There is an enormous diversity of ice cream flavours which could be argued over forever with seemingly no “true” answer. Yet, in math, there is remarkable convergence on truth. In soccer, there seems to be some. In music, some. In visual art, less. And so forth.

What explains these varying levels of subjectiveness is the degree to which the tradition has conflicting assumptions. Examples: Let’s start with math again. Once the tradition defines 2, 4, =, and the axioms of addition, the statement “2+2=4” is true. There is no room for disagreement because there are no conflicting axioms such as “the = sign means a + b = b.” Everyone agrees which assumptions are pertinent to that question.

Another example: Spoilers for Marty Supreme, but after watching the film and walking out of the theatre, my friend Rowdy and I had radically different takes on the ending. The ending, to remind you, was Marty seeing his baby and crying tears of joy, deciding to finally favour being a father over his career.

I thought this was bad because it felt like whiplash! The entire time, this guy hasn’t given a shit about his girlfriend nor the baby, he’s been a narcissist who, worse still, is obsessed with something (being an international table tennis star) which directly conflicts with his ability to be a dad. Suddenly, after beating one guy in a match that didn’t matter in the long run, and seeing his baby, he has given up the sport he’s been obsessed with his entire life? Given up his narcissistic personality? Given up his disregard for his girlfriend? It felt way too convenient.

Rowdy had the opposite take. It’s Marty realizing that no matter what he does in his career, he’ll never find genuine satisfaction or peace and looking for that in something potentially much more fulfilling. He felt that seeing a newborn could induce a massive psychological change, and felt that beating Endo was enough of a justification for him to be at least temporarily satisfied. Rowdy eventually brought me somewhat around, but I find the initial disagreement to be really interesting.

Nice shot

Rowdy might disagree with this but roughly here are the (very implicit) axioms I think are in conflict:

I assume a continuity of psychology is a necessary constraint for good character arcs. Big reversals should be seeded through earlier moments. (This is why I love A Song of Ice and Fire) So, if the movie spends most of its time showing obsession + disregard, the ending should resolve that obsession through consequences, not by a sudden moral pivot.

On the other hand, Rowdy’s axioms: Psychological realism is more important than strict narrative continuity. Also, Rowdy assumes that people do change abruptly, (in retrospect this might be because he has friends who moved to Austin and converted to Christianity overnight), especially with intense triggers, and fulfilling this assumption is part of what can make a movie good.

These assumptions are in conflict! Thus the disagreement that feels like it has no right answer. There are many such contradictory assumptions in the tradition which is cinema criticism, much more than math and sports. (Why do you shoot a 3 when you’re open? “Because I want to score.” Why? “Because it helps us win.” Why do you want to win? “Because winning is good.” Why? “Because it just is!” (No one in sports has a contradictory assumption on this basic point, and the statistical research in this tradition has generated a lot of convergence, thus, sports are becoming more truthful.)

In the tradition of politics, I often see debates go on and on about how interventionist the state should be. Side A might argue the state should actively redistribute and regulate to secure equality and welfare. Let’s go through their likely assumptions. I’ve doing this more rigorously now because without structure I get long-winded.

Layer 1

1. [normative] Equality and material security ought to be the ultimate political goals.

2. [empirical] State interventions can materially improve welfare.

3. [structural] Institutions can be constrained to avoid abuse/capture.

Layer 2

1. (for L1-1) [normative] Market outcomes are not morally decisive.

2. (for L1-2) [empirical] Market failures are large and correctable.

3. (for L1-3) [structural] Oversight mechanisms can work.

4. (for L1-4) [definitional] “Justice” includes distributive outcomes.

Side B might argue that the state should be limited and freedom and property rights should dominate.

Axioms:

Layer 1

1. [normative] Individual liberty/property are primary.

2. [empirical] State power tends to expand and be abused.

3. [structural] Markets and civil society allocate resources better.

4. [definitional] Justice is procedural, not distributive.

Layer 2

1. (for L1-2) [empirical] Policy errors are frequent and hard to reverse.

2. (for L1-3) [structural] Competition disciplines misuse of power.

3. (for L1-4) [definitional] Fair process yields legitimate outcomes.

I’m tempted to keep going but I think you get the point. When everyone in a tradition agrees on which assumptions are in force for any specific question, you get true statements. The less contradictory assumptions that are present in a tradition, the more “objective” it feels, and the less people debate or resign themselves from debate entirely (“it’s only my opinion!”).

Now you’ll probably be wondering, why do some traditions have almost zero axioms in conflict, while others have dozens? Why are some incredibly implicit and almost invisible in debate (visual art) while others are widely accessible (music theory)?

My preferred explanation is incentives. Some traditions have higher stakes for finding the truth. If U.S. mathematicians have contradictory assumptions, then the Apollo mission is crashing and burning, people are dying, and the country may lose the space race. If film bros have contradictory assumptions, Rowdy and I have something fun to talk about. Lower stakes incentivize less rigorous assumption organization in the tradition.

Even still, you can see things like the NBA and the movie industry converging on the truth. The truth in 2026 is unfortunately, a high volume of threes and the Marvelizationof cinema. Both of these states of affairs resulted from some assumptions conflicting and then being resolved. (Expected value of threes is higher than many twos, vs threes are lower quality and elite specialists should be treated differently than average players), and (shared universes and event spectacle increase lifetime revenue + general risk aversion vs artistic value not being reducible to revenue or popularity and the assumption that audiences can and should be shaped toward inventive art.) Unfortunately for me, to maximize team winning and company profits, the assumptions I personally enjoyed more were combed out of the traditions for the most part, and they continue to become more objective over time.

Counterargument

Bentham’s Bulldog, perhaps: “So, here are two cruxes of your argument: If low‑stakes fields routinely develop strong, stable measurement regimes, your argument collapses, and if high‑stakes fields show no greater investment in verification infrastructure, your argument collapses. Well, for the former, how about chess, and for the latter, how about macroeconomic policy or presidential campaigns?”

Me: “Well then, it seems stakes are necessary but not sufficient. On average, my argument is probably correct, but it’s not hard to find clear examples where, even when the cost of being wrong is high, there’s much lower truth convergence than other, lower-stakes traditions. Well Bentham, I’m not wedded to this theory, I’ll change it on a dime! An amendment is in order.”

New theory:

Policy debates and moral controversies feel more subjective (don’t converge) than structural engineering, not because people care less, but because the object of inquiry is less operationalizable, and the infrastructure for verification is weaker. A more complete theory has three parts:

1. Operationalizability: Can many parts of the tradition have testable criteria for truth? Mathematics and physics score high here, film criticism scores low.

2. Verification infrastructure: Are there shared practices that settle disputes (proof systems, controlled experiments, standardized tests,

adjudication bodies)? Interestingly wine tasting scores higher than ice cream tasting, simply because there are more adjudication bodies.

3. And finally stakes: Do incentives push traditions to build and maintain the first two parts?

I love this theory now! Thanks Bentham. (I could have just started with this, but I find developing a narrative enjoyable. Perhaps it would be better for everyone’s time if I devoted myself to being a novelist rather than a theorist.) This helps explain the counterargument above: recreational mathematics (low stakes) converges reliably, while parts of public policy (high stakes) remain permanently contested. The first is operationalizable with strong verification; the second is not! The 4 possible examples:

1. High measurability + high stakes: strong convergence (aviation safety, formal engineering).

2. High measurability + low stakes: marginally lower convergence without pressure (puzzles, chess).

3. Low measurability + high stakes: polarization and use of authority (policy, ethics).

4. Low measurability + low stakes: pluralism and taste (film, food & fashion).

I find traditions that find themselves in 3. the most interesting. They seem to play out a few different ways. Often you’ll just see people divide themselves into separate groups because they can’t agree on base assumptions. This seems to happen quite a bit in religion and politics. Sometimes it just seems an authority comes in to substitute for the empirical disagreement. Lineage or charisma seems to be defining features here. This authority (state or church usually) will use orthodoxy policing to suppress rival interpretations within the tradition. And then there’s the fun one where neither side decides to use force, but the stakes are still red hot, so they’re stuck in perpetual disagreement over assumptions! I’m thinking of various schools of thought in ethics here, for example, consequentialists and deontologists. Parties in democratic regimes with stable succession systems also fit.

Of course, all traditions were at one point just as bad off as ice cream tasting is today. Medicine has made great strides towards objectivity. Many centuries ago you would have had debates about medicine as you would ice cream! Or perhaps the debates were stopped by some vague authority-based judgment that said you had to bleed yourself when you got sick. Of course, the stakes (death) are high in this tradition, and germ theory and randomized trials and the scientific method and a salad of other beautiful, replicable measurement techniques have brought this tradition up from the filthy slums of politics and marginally less filthy slums of ethics, out and into the bucolic green pastures of empiricism. Similar things can be said of seafaring and aviation and a whole host of other traditions.

The takeaway here is to become more of an empiricist, not less. Any tradition you admire for knowing the truth was once terribly incorrect. The only solution is to embrace enlightenment style evidence-based inquiry. Also, if my theory is correct, we don’t need to appeal to any of the armchair arguments moral realists are fond of to explain the seeming objectivity and subjectivity of different fields of conversation.

P.S. I’m still searching for an objective theory of title making. Help!