On asking better questions of the deck

A man named Bem booked me for a thirty-minute reading in January with a question I get a version of nearly every week. "Will my ex come back?" He typed it into the booking form in those exact words. He'd been waiting nine months. He'd done what he called "all the work." He wanted to know if she was coming.

I read the question on the form, sighed in a way that probably wasn't very professional, and replied: "I can't read that question, but I can read three others related to it, would you like me to send them and you can pick?"

He picked one. The session was good. We didn't talk about whether she was coming back. We talked about what nine months of waiting had done to him, which was the question worth asking. He emailed me three weeks later to say he'd stopped waiting. He didn't tell me whether she came back. I suspect he no longer knew.

I want to write today about tarot questions, and how the way you ask shapes the reading more than the cards you pull. This is the most important practical skill in a reading and the one most beginners, and a depressing number of experienced readers, skip past entirely.

The three kinds of questions

Across roughly fifteen hundred readings I've done in the last five years, I've come to sort tarot questions into three rough categories. Closed, narrow, and open. Each produces a different kind of reading. Each is appropriate in different circumstances. One of them is the one I push my clients toward almost always.

A closed question asks the cards for a yes-or-no, a will-or-won't, a prediction about a fixed event. "Will I get the job?" "Will they call?" "Is she cheating?" "Will I be married by thirty-five?" The answer the client is hoping for is a binary, and the spread, however many cards it contains, is collapsed in their head into that binary.

A narrow question asks the cards to choose between known options. "Should I take the job at company A or the one at company B?" "Should I move to Berlin or stay in London?" "Should I text him or wait?" The answer space has been pre-defined by the client. The cards are being asked to pick.

An open question asks the cards to illuminate a situation without specifying what kind of answer is wanted. "What should I be paying attention to in this transition?" "What's the dynamic between me and my sister actually about?" "What do I need to know about how I'm relating to this decision?" The answer space is, by design, undetermined. The reading goes where it goes.

I do not think these categories are objective. They're working tools. I have colleagues who would carve the space differently. But the categories track something real about how questions shape sessions, and they've held up across a lot of sessions for me.

Why closed questions get the thinnest readings

Closed questions force the spread into a single bit of information. The most elaborate ten-card Celtic Cross, asked of a yes-or-no question, gets collapsed by the client into yes or no the moment the session ends. The other nine cards become decoration. The client did not pay for nuance. The client paid for a binary. They'll get a binary, even if you read the cards in twelve dimensions.

There's a craft argument and an ethics argument against closed questions, and I think both matter.

The craft argument: cards are not designed for binaries. They're designed for situations. A card like the Two of Pentacles, in answer to "will I get the job," cannot honestly become a yes or a no. It's a card about juggling, about flux, about ongoing balance. You can torture it into a yes or a no, but you've left most of its meaning on the table to do that. The reading was poorer than it had to be because the question was the wrong shape for what was being asked.

The ethics argument: predictions about fixed future events, especially involving other people's behaviour, are unreliable in a specific way. They suggest a determinism the client does not need, particularly when they're already in distress. Bem asking whether his ex was coming back was, in practical terms, asking me to take responsibility for another adult's choices on a timeline I had no access to. The honest answer was: I cannot read for that, and even if I could, you should not orient your life around the answer.

I push clients away from closed questions almost always. There are exceptions. Sometimes a yes-or-no is genuinely what's needed, the way a coin flip is sometimes genuinely what's needed, to surface a reaction. But the exceptions are exceptions. The default should not be the binary.

Why narrow questions feel useful but aren't

Narrow questions look more sophisticated than closed ones. The client has at least identified options. They're not asking for fate. They're asking for guidance between two known paths.

The problem is that the narrow framing has, almost always, pre-decided the shape of the answer in a way that excludes the most interesting third options. "Should I take the job at company A or the one at company B" assumes those are the two paths. The cards may want to say: don't take either. Or: take A but renegotiate before you start. Or: the real question isn't which job, it's what you're running from. The narrow framing makes those answers harder for the client to hear, because the client has already constructed the choice space.

I'll let a narrow question through more readily than a closed one. It at least admits there's a decision in motion. But I usually re-frame it, gently, before I lay any cards. "You're asking me to choose between A and B. Can I instead read on what would make this decision good either way, and then you decide?"

Sometimes the client says no, they really do want A or B. Fine. We do that reading. More often the client says, oh, that's actually what I wanted, and we get to a richer session because we widened the answer space.

Why open questions produce the best readings

Open tarot questions get the richest readings, every time. Here's the framework I give my clients before sessions.

An open question identifies a situation, names what you want help with, and leaves the answer space undetermined. The grammar of an open question often starts with "what" or "how," rarely with "will" or "should." Some examples:

• "What should I be paying attention to as I move into this next phase?"
• "How am I currently relating to this decision, and what's that costing me?"
• "What's underneath my reluctance to do this thing I keep saying I want to do?"
• "What do I need to see about this relationship that I've been missing?"
• "What would shift if I stopped waiting?"

Notice what these have in common. They identify a real situation. They name a real curiosity. They do not demand a particular shape of answer. The reading can go anywhere within the situation, and the client has agreed in advance that wherever it goes will be useful.

When Bem changed his closed question into an open one, the question we ended up with was: "What's nine months of waiting actually done to me?" The reading that came out of that was about him. About the way the wait had become its own form of identity. About the cost of orienting his life around a return that may or may not arrive. The cards I pulled don't matter for this essay, but they pulled clean, and Bem cried twice, and at the end he said "I didn't know that was what I was asking but that was what I was asking." That's an open question doing its work.

The craft case for not predicting other people

I want to make a craft case for something usually framed as an ethics one. Tarot does not predict other people's behaviour well. This is, I'd argue, not a moral position. It's a description of what the tool is good at.

The deck is, in my reading, a mirror for the person sitting in front of it. It surfaces patterns the person is already in. It illuminates how they're holding a situation. It does not, in any reliable way, tell you what someone else is going to do next week. People have free will, or near enough to it, that no symbolic system gets to call their choices in advance with any consistency. Readers who claim otherwise have made a craft mistake, regardless of whether they've also made an ethics one.

When a client asks me to read for whether someone else will do something, I tell them this directly. I say: the deck doesn't read well for other people's decisions. It reads well for yours. Would you like to ask the question in a form the tool can actually do work with?

Reframing "will she come back" as "what am I asking when I ask if she'll come back" gets you a reading. Reframing "is he cheating" as "what do I already know about this relationship that I'm asking the cards to confirm" gets you a reading. Reframing "will I get the job" as "what would it serve me to know about this hiring process before the decision lands" gets you a reading. The client retains agency. The cards do what they're good at. Everyone leaves better-served.

A short reframing exercise

If you're new to reading, or new to having your readings be useful, try this. Take a question you've been wanting to ask the cards. Write it down. Sort it into closed, narrow, or open.

If it's closed, reframe it. "Will X happen" becomes "what should I know about X." "Will they call" becomes "what's keeping me oriented around the possibility of their call."

If it's narrow, widen it. "Should I do A or B" becomes "what would make either decision a good one." "Should I stay or leave" becomes "what is staying costing me, what is leaving costing me, and what's underneath the choice."

If it's already open, sharpen it. The best open tarot questions name a real situation and a real curiosity. "What do I need to know" is fine. "What do I need to know about why I've been avoiding finishing this project for three months" is better.

Then ask the new question. Lay your spread. You'll find the reading is more useful, almost without exception. Not because the cards changed. Because you finally gave them something worth working with.