Assumptology: Making the Implicit Explicit

Richard Feynman once described a way of living that most people say they want, right up until uncertainty starts to itch:

“I think it is much more interesting to live not knowing than to have answers which might be wrong…I have approximate answers and possible beliefs and different degrees of uncertainty about different things, but I am not absolutely sure of anything…I don’t feel frightened not knowing things…”, Richard P. Feynman

That isn’t “anti-knowledge.” It’s sanity.

Because the real mess in public discourse usually isn’t ignorance. It’s false certainty, answers delivered with more confidence than the assumptions behind them can carry.

I’ve seen the same thing in tech and business too: one bad assumption gets baked into a system early, and then everything downstream quietly inherits it. Years later people argue about “results” without ever touching the premise that produced them.

That’s the layer I’m interested in.

I call it Assumptology: the habit of noticing hidden assumptions, dragging them into the open, and asking which ones are doing the real work.

The Move

When someone makes a strong claim, don’t only ask:

“Is this true?”

Ask:

• What has to be assumed for this to be true?
• Which assumptions are being slipped in as “obvious” or “common sense”?
• If I change one assumption, does the whole conclusion change?

You’d be surprised how many arguments don’t survive that test.

The Clearest Modern Demo: AI Answers

AI makes this easy to see because it speaks with calm, fluent confidence.

You ask a question. You get a confident paragraph back. It sounds like knowledge.

But the confidence is borrowed.

It rests on things you were never shown: what went into the training mix, what got filtered out, whether your prompt actually captured what you meant, whether “sounds right” is being mistaken for “is right,” and whether the model is just echoing the loudest consensus in its data.

Flip one assumption, “this reflects the full spectrum of knowledge” → “this reflects a time-stamped slice of mostly online text”, and the same answer suddenly feels less solid.

You don’t have to reject AI to see this. You just have to stop outsourcing your stopping point.

Why “Ask Better Questions” Isn’t Enough

You’ve heard the advice:

• “Ask better questions.”
• “Keep asking why.”
• “Start from first principles.”

All useful. Still incomplete.

Because every question you ask is already shaped by the assumptions you didn’t question.

If your background assumptions are off, your questions can be brilliant and still miss the leverage.

If you assume the problem is motivation, you ask: “How do I get more disciplined?” If you assume the problem is incentives, you ask: “What rewards this behaviour?” If you assume the problem is framing, you ask: “Who benefits from me seeing it this way?”

Same person. Same intelligence. Different path, because a different assumption got installed upstream.

Assumptology goes one level deeper than “ask better questions.” It asks:

What must I assume to even think this is the right question to ask?

Once you see that, your questions upgrade themselves.

Why This Matters Right Now

We’re drowning in certainty: headlines, hot takes, algorithm-fed outrage, “trust the experts,” “do your own research,” and now machines that can generate plausible answers on demand.

Certainty is persuasive. It’s also cheap.

Assumptology is basically a way of not getting hypnotized, not by media, not by institutions, not by your own side, not by an AI that writes like it knows.

(And no, this isn’t anti-science or anti-expertise. It’s just refusing to treat authority as automatic closure.)

What You’ll Get Here

Short essays applying Assumptology to real things:

• language that installs beliefs (“when” vs “if”)
• persuasion, marketing, propaganda
• media narratives and framing wars
• politics and culture
• AI outputs and “answers” from black boxes
• everyday decisions (buying appliances, watching TV, scrolling feeds)

The goal isn’t to tell you what to think.

It’s to make the invisible layer visible: what you were asked to assume before you reached your conclusion.

If that’s your kind of thinking, subscribe.