Krisang mentions, almost as a throwaway at the very end of the episode, a rule he's kept for five or six years: he doesn't give answers until someone asks him a question.1 Not a rhetorical flourish — an actual, sustained personal discipline. "I have followed this rule for myself almost for last five to six years now. It has made my life really simple and easy and peaceful because I don't try to go and give answers to people if they don't ask me questions."1
Sit with how strange that is as a stated goal. Most people who know things want to tell you what they know. Krisang has spent half a decade deliberately not doing that, and the payoff he names isn't better teaching outcomes — it's peace. Withholding the answer, in his account, isn't a technique for making the answer land harder. It's a way of living that happens to also produce better transmission as a side effect.
Here's the proof, live, in the room. At the close of the episode, Krisang tells his co-host directly: there are so many things he wants to tell the world, but he never does, until someone asks. "That's why you were able to ask the right questions and get out these answers."2 The podcast itself — the very recording the audience is listening to — is presented as an instance of the rule working. No question, no episode. The format is the proof of the claim, not just an illustration of it.
Krisang doesn't claim to have invented this. He names it as the default shape of transmission across an entire tradition: "There are so many scriptures — like it's one of the most fundamental ways of transferring knowledge in our civilization, which is question and answers. It is never lectures. We have never done lectures."3
The two examples he reaches for are specific and load-bearing. The Upaniṣads, broadly, framed as question-and-answer texts. And the Bhagavad Gītā — Arjuna's questions to Kṛṣṇa, not Kṛṣṇa's unprompted lecture to Arjuna.3 It's worth noticing what this framing does to a text most people encounter as a monologue. Read casually, the Gītā looks like Kṛṣṇa delivering a philosophy. Read as Krisang frames it, it's a text that only exists because Arjuna, paralyzed on a battlefield, asked something real. The teaching is downstream of a genuine question, not upstream of one waiting to be delivered.
This is the part that's easy to nod along with and hard to actually feel the weight of: in this model, the question isn't a courtesy that opens the door for the answer. It's the thing that makes the answer true in a way an unprompted answer can't be. An answer given before anyone asked is a guess about what someone needs. An answer given after a real question is a response to an actual, located gap — this person, this confusion, this specific place where their understanding runs out. Same words might come out either way. The second version is calibrated to something real; the first is calibrated to nothing but the speaker's own sense of what matters.
This is also, not incidentally, the discipline behind Pūrva-Pakṣa's insistence on restating the other person's position before offering your own — see Pūrva-Pakṣa as a Content Creator's Debate Discipline. Both rules refuse to let the speaker go first. Both make the other person's actual state — their position, their question — the precondition for anything you're allowed to say next.
You're building a course, or a newsletter, or any piece of teaching content, and the instinct is to start with what you know and organize it into a curriculum — module one, module two, module three, the shape of your own expertise laid out in the order it makes sense to you. Try the inversion instead. Collect the actual questions people are asking — comments, DMs, the confused replies to your last post — and build the curriculum as answers to those, in the order the questions actually came. The resulting structure will look messier than your own logical outline. It will also be teaching something someone was actually stuck on, instead of something you assumed they needed.
You're recording a conversation with a collaborator, the way this very episode was recorded. Resist the pull to prepare a list of talking points in advance. Krisang and his co-host describe scrapping their prepared questions the night before this exact recording and deciding to just talk and let the questions emerge live.4 Notice what changes in the footage when the questions are genuinely live rather than pre-loaded — the answers come from somewhere the speaker didn't already have queued up.
You catch yourself wanting to correct someone, unprompted — a comment on your post, a friend's take you disagree with. Before you respond, check: did they ask? If not, the five-to-six-year rule suggests holding it, not because you're wrong, but because an answer nobody asked for rarely lands as anything but ego. Wait for the question. If it never comes, maybe the answer wasn't as necessary as it felt.
There's a structural move worth naming directly: the episode this page is sourced from is not just about question-and-answer transmission. It performs it. The co-host's questions throughout are what pull Krisang's answers out — the framework only exists on the record because she asked. That's not a coincidence the interview happens to demonstrate; per Krisang's own closing remark, it's the entire mechanism working exactly as designed, with the audience watching it happen in real time rather than being told about it after the fact.
Krisang's textual examples (Upaniṣads as Q&A, the Gītā as Arjuna's questions to Kṛṣṇa) are broad, popular-register characterizations of primary texts, not close readings — the transcript is auto-generated and phonetically garbles the Sanskrit throughout, and no specific verses are cited. [PARAPHRASED], treat the textual claims as a practitioner's gloss rather than verified scholarship.
Open question: the five-to-six-year personal rule is presented as universally applicable ("I have followed this rule for myself"), but Krisang is also a public teacher actively producing unprompted content — videos, courses, this very episode's own marketing copy. There's an unexamined gap between "I don't answer unasked questions in person" and "I run a content business that proactively publishes answers to questions nobody in the audience has yet asked." The source doesn't reconcile this; whether the rule is about interpersonal conversation specifically, or about content production too, is left ambiguous.
Reading the Gita as a question-driven text and reading it as a delivered philosophical monologue are both live in how the text actually functions for different readers. A reader encountering the Gītā through a philosophy syllabus meets eighteen chapters of Kṛṣṇa's teaching, organized by topic — karma yoga, jñāna yoga, bhakti yoga — the shape of a systematic treatise. A reader encountering it the way Krisang frames it meets a terrified man on a battlefield who can't fight, asking a real question because he is actually stuck, and getting an answer that only makes sense in response to that stuckness. Neither reading is wrong. The tension is about which one you'd hand someone first. Hand them the treatise-shaped version and they learn a philosophy. Hand them the question-shaped version and they learn that philosophy this dense usually starts with someone who couldn't move.
Creative practice — Long Interview and Staged Writing Method. Gelb's documentary method runs on sustained, repeated questioning across a two-week shoot, and the deepest material doesn't surface on day one — it surfaces on day seven or eight, after enough rapport has accumulated that the subject volunteers something they hadn't planned to say. Krisang's five-to-six-year rule and Gelb's two-week shoot look, on the surface, like completely different timescales solving completely different problems. They're not. Both are naming the same underlying law: real disclosure is a function of readiness, and readiness can't be rushed by asking a better question sooner — it has to be waited for, or built toward, through duration and real engagement. Gelb's version operationalizes the wait as a production schedule (two weeks, multiple interviews, a vulnerability exchange where the director goes first). Krisang's version operationalizes it as a personal ethic (don't speak until asked, for years at a stretch). Put together, they suggest something neither states outright: "wait for the real question" isn't a communication tactic you can adopt for a single conversation and drop afterward. It's a stance that has to be sustained long enough for the other party to trust that the question, once asked, will actually be met — which is exactly why Krisang frames it as a five-year practice and Gelb's crews spend two weeks building the same trust from scratch with every new subject.
Creative practice — Self-Interview Protocol — Margin Questions. Karlsson's practice looks like the inversion of Krisang's rule: instead of waiting for someone else to ask, Karlsson manufactures the question himself, writing "what do you mean by that?" and "give me three examples" in the margins of his own notebook, weeks after the original claim was written. On the surface this contradicts the wait-for-the-question ethic — Karlsson isn't waiting for anyone. But look at what's actually load-bearing in both practices, and the contradiction dissolves into a shared mechanism: in both cases, an answer is only trustworthy when it's produced in response to a real question rather than delivered as an unprompted assertion. Krisang externalizes the questioner (another person, a co-host, a community member) because he's distrustful of his own sense of what needs saying. Karlsson internalizes the questioner because he's working alone and has no one else to supply the interrogation — so he builds a second self whose only job is to demand specifics and counter-examples from the first self. Neither practice trusts a claim that hasn't been pressure-tested by a question first; they differ only in where the question comes from. The handshake produces a genuinely useful distinction the source material for either page doesn't state on its own: dialogic transmission doesn't require a second person in the room. It requires a genuine question, asked by someone with standing to ask it — and when no one else is available, the discipline is to become that someone yourself, on a delay long enough that you're actually surprised by your own answer.
Sharpest implication. If knowledge only transmits well downstream of a real question, then most teaching content — the unprompted explainer, the "here's what you need to know" post, the lecture nobody asked for — is working against the grain of how transmission actually functions in the tradition Krisang is drawing from. The content might still be accurate. It's unlikely to be received the way an answer to an actual question would be, because nobody's actual gap in understanding is what triggered it.
Generative questions.