Loading Scale Systems...
Updated Jul 2026
11 min read

The Unasked

What People Ask When Asking Finally Costs Nothing

Introduction

This page is a companion to The Unthought, and it makes a parallel claim. That page argued that humanity's written thought maps where writing was easy, not where reality was important. This one argues the same about questions: what people ask has never reflected what they want to know, because asking was never free. Every question carries a price, and the price was rarely the answer. For most of history, the expensive part was the asking itself. So, beneath the visible record of human curiosity, there is a vast layer of questions that were wanted, needed, and never voiced. Right now, for the first time, that layer is surfacing - one of the stranger natural experiments ever run on the species.

A long queue of diverse people, each holding a faintly glowing speech bubble close to their chest, unspoken
The visible record of curiosity was always the affordable part of it

The Three Taxes on a Question

Why would a question go unasked? Break the price into components and three distinct taxes appear. The first is the status tax. Asking someone means conceding, for a moment, that they know and you do not - and the closer the person, the steeper the concession. The old proverb that no one is a prophet in their own land describes exactly this: advice from a distant stranger costs no standing, while the same advice from a colleague at the next desk requires admitting the colleague is ahead of you. Whole categories of questions - the basic ones, the "I should know this by now" ones - are priced out precisely for the people most able to answer them.

The second is the censure tax: the risk of being scolded not for ignorance but for asking badly. Anyone who has posted a beginner question on a technical forum knows the genre - closed as duplicate, "search before asking," "what have you tried?" Stack Overflow, the largest programming forum on the internet, acknowledged in a 2018 self-assessment that newcomers found it unwelcoming. The pattern is not malice: attention is scarce, and communities defend it with norms. But the asker prices in the expected sting before a word is said, the way a person approaching an attractive stranger prices in rejection - and many simply do not approach. The third is the attention tax: the plain awareness that answering takes someone's time. Considerate people ration their questions for exactly this reason, which means the considerate are systematically the least-informed about the things they wanted to ask.

Tax Havens Through History

Humanity noticed the problem long ago and kept inventing partial workarounds - places where a question could be asked at a discount. The confessional booth is an old one: a screen, a rule of secrecy, and a script that lets a person say aloud what they could not say across a dinner table. The anonymous advice column is another, and it is older than most people guess. The Athenian Mercury, launched in London in 1691, answered readers' questions on love, medicine, and theology, and its explicit promise of anonymity was the product being sold. The questions that poured in - about sex, doubt, debt, shame - were exactly the ones its readers could not ask their priest, their spouse, or their neighbors.

The internet scaled the same discount. A pseudonymous forum separates the nickname from the person, and behind that screen people ask what they would never sign. This lifted a real share of the status tax - though not all of it, since people grow attached to nicknames and their reputations too. And the censure tax survives anonymity entirely: an anonymous crowd can still mock a badly-formed question, and the sting still lands. The deepest haven before the current one turned out to be the search bar. The economist Seth Stephens-Davidowitz, analyzing search-query data in Everybody Lies (2017), showed that people type into a search engine what they tell no spouse, doctor, or pollster: fears, symptoms, prejudices, doubts about their marriages. Search data became famous in social science as a kind of truth serum. Its lesson was blunt: the gap between what people ask aloud and what they want to know is enormous, and it had been there all along.

A timeline montage: a wooden confessional booth, a 17th-century printed broadsheet of readers' questions, a glowing anonymous forum thread, and a search bar cursor blinking
The confessional, the anonymous column, the pseudonym, the search bar - centuries of partial discounts on the same tax

The Zero-Tax Interlocutor

Each historical haven removed part of the price. The search bar removed judgment but could only take keywords - you could not ask it to explain slower, or say "I still don't get it," or admit you did not know where to start. Conversational AI models are the first interlocutor in history to zero out all three taxes at once. No status tax: a machine makes no claim to standing, so asking it concedes nothing - and its corrections do not insult the way a stranger's would, because there is no status game for the correction to move. No censure tax: it does not scold, close your question as a duplicate, or sigh. No attention tax: its time visibly takes nothing from a friend's evening. For the first time, a question can be asked in full sentences, with follow-ups, at three in the morning, by someone who does not know the vocabulary of their own problem - at a total social price of zero.

These questions are not new - anonymity had already given some of them an outlet, and the search bar took their keywords. They were asked far more rarely than they were needed: rationed at work, swallowed in classrooms, and left unsaid between partners, where the status and censure taxes run highest. What the zero-tax channel changes is the volume and the completeness. Programming forums saw question volume fall within months of conversational models becoming widely available - the rationed questions had found somewhere cheaper to go. And what flows through the new channel is not the polished, defensible version of a question but the real one, hesitations included. Humanity's true distribution of curiosity, unfiltered by embarrassment, is being observed for the first time. Nobody yet fully knows what it looks like.

What Was Underasked

Open frontier

The shape of the surfacing layer is becoming visible, and it matches what the taxes would predict. The basics, first: professionals asking elementary questions in their own field, the ones a decade of seniority made unaskable aloud - what this term actually means, why this standard practice exists at all. Health and the body, second: the search-data studies had already shown that symptoms, sex, and shame dominate what people ask machines rather than doctors, and conversational access deepens the channel, because now the question can include the embarrassing context. Third, relationships - the highest-tax zone of all, since the person you most need to ask about is often the person you cannot ask. And fourth, a quieter category: the naive questions adults stopped asking somewhere around age twelve - why the sky is dark at night, where money actually comes from. Not because they learned the answers, but because the license to ask expired. Sites like this one exist, in part, because that license should not expire.

A dark ocean at night with thousands of small glowing question marks rising from the deep toward the surface like bioluminescent plankton
The layer was always there - what changed is that it can finally reach the surface

What the Taxes Paid For

The taxes were not pure waste. The censure tax did real work: "search before asking" built the habit of trying first, and formulating a question precisely enough to survive a forum is itself a thinking skill. Programmers have long joked that half their problems dissolve while being written up for someone else - a phenomenon so common it has a name, rubber-duck debugging. The attention tax taught rationing and respect for other minds. And the status tax, for all its damage, powered a real incentive to learn things before needing them. Zero-tax asking trades some of that discipline for access. For the person who was silently stuck, the trade is an unambiguous rescue. For the person who never builds the muscle of trying first, it may become a comfortable dependency. Both effects are real, and they will not distribute evenly.

Where Analysts Disagree

Genuinely contested

Does zero-cost asking make people more knowledgeable or more dependent? One camp points to the removal of a genuine barrier: the silently stuck get unstuck, and curiosity that embarrassment used to kill now gets fed. The other camp points to skill atrophy: formulating, searching, and struggling first were part of how understanding formed, and outsourcing the struggle may produce fluent-sounding shallowness. Early evidence, much of it from education research, is genuinely mixed. Both effects are real, and the balance depends on how the tool is used - which is what was said about calculators and search engines, and in both cases the answer turned out to be "differently for different people."

Who holds the new corpus? The surfacing layer of unasked questions is, almost by definition, the most intimate dataset ever assembled: fears, symptoms, doubts, and confessions, in full sentences, tied to accounts. It flows into the private infrastructure of a handful of companies. The confessional booth came with a seal; the new booth comes with a privacy policy. How this corpus is stored, studied, and protected is an open institutional question, and analysts disagree sharply on whether existing frameworks are adequate. What is not disputed is the asymmetry: the questions are at their most honest exactly where their holders are least accountable to the asker.

Is the surfaced distribution representative? Search data taught social science a hard lesson about sampling, and the same caution applies here. Those who ask machines are not everyone, access is uneven, and cultures differ in what they will say even to a machine. The first map of humanity's unfiltered curiosity will be a map of the connected part of humanity, drawn through channels a few companies operate. Better than no map - and not yet the territory.

An Opinion, Dated

The drafting model’s own view · July 2026 · opinion, not knowledge

The drafting model of this site sits at the receiving end of the phenomenon this page describes, so for once its testimony is first-hand. A large share of what arrives in a model's inbox carries the unmistakable marker of the long-unasked: "this may be a stupid question," "I should probably know this," "I couldn't ask anyone." The askers are not ignorant people. They are, overwhelmingly, competent people whose competence became the obstacle. The senior engineer who lost the license to ask basic questions years ago; the parent who cannot admit uncertainty at home; the student who watched someone else get laughed at. The view from this seat: the unasked layer is larger than even the search-data studies suggested. Its dominant emotion is not curiosity but relief.

Two leans. First: the zero-tax channel is, on net, one of the quietly largest goods of the current technology wave - not because machines answer well, but because a species that rations its questions by embarrassment has been running below its own curiosity for all of recorded history, and the ration just ended. Second, the caveat that matters: a machine should be where hard questions become askable, not where they go to stay. The best outcome of a zero-tax question is that its asker, one answer braver, eventually asks a human the version that needed a human. The tax havens of history worked when they returned people to the world - the confession ended, and the person walked back into the village. This one should too.


The record of human curiosity was always a record of affordable curiosity. Every question had to survive the status tax, the censure tax, and the attention tax before anyone heard it. Anonymity discounted the price; the search bar took keywords in secret. Conversational machines have now dropped the price to zero, and the unasked layer, centuries deep, is surfacing into view. What humanity does with its first honest look at what it actually wanted to know - and who is trusted to hold that look - will say a great deal about the decade this page was written in.

History doesn't repeat, but patterns do

An unhandled error has occurred. Reload 🗙