Asking for Help Without Outsourcing Agency

I used to avoid asking for help until I was already in trouble.

Then I over-corrected. I got better at asking—but I started asking in a way that quietly outsourced my agency. I would hand someone my mess and wait for them to turn it into a plan. I would treat their certainty as a substitute for my own responsibility. I would borrow their spine because mine was tired.

Both patterns are brittle.

There’s a third way: receive help while keeping ownership.

This is not a moral stance. It’s a design choice. It’s the difference between:

• Help that strengthens me (capacity increases)
• Help that replaces me (dependence increases)

Below is the method I use when I need support but don’t want to disappear inside it.

The Failure Mode: “Take My Life, Fix It”

If you’ve ever asked for help and then felt worse afterwards, it’s often because the help came with an invisible transfer of authority.

You asked for assistance, but what you implicitly asked for was:

• permission to stop thinking
• relief from consequence
• someone else’s identity to lean on (“tell me what a good person would do”)
• a clean answer that removes ambiguity (and therefore removes your contact with reality)

Sometimes we do need triage. Sometimes we are too flooded to self-govern well. But even then, the goal is stabilization—not abdication.

The aim is: supported autonomy.

The Three Roles of Help (Pick One)

Before you ask anyone for anything, decide which role you are inviting them into. Most confusion happens because you invite someone to witness, but they try to manage. Or you ask for strategy, but they only empathize. Or you want accountability, but they become a therapist.

I use three roles:

1) Witness (truth without fixing)

You want someone to hear you clearly and help you stay honest.

What you’re asking for:

• presence
• reflection
• questions
• naming patterns you can’t see

What you are not asking for:

• commands
• a plan
• evaluation of your worth

2) Advisor (options and tradeoffs)

You want someone to help you think through decisions.

What you’re asking for:

• options
• risks
• tradeoffs
• “if this, then that” reasoning

What you are not asking for:

• them deciding for you
• them doing the follow-through

3) Teammate (shared execution)

You want hands-on help executing a defined piece of work.

What you’re asking for:

• a shared task
• clear handoff points
• explicit definition of “done”

What you are not asking for:

• them owning the whole project
• them becoming responsible for your identity

Pick one role per ask. If you need more than one, sequence them.

The Ownership Principle: Decision Rights Stay With You

Here’s the line I now repeat to myself:

I can borrow someone’s eyes, but I can’t borrow their responsibility.

If you want help without outsourcing agency, keep decision rights.

That means:

• You decide the goal (even if it changes).
• You decide the constraints (money, time, values, relationships).
• You decide the next step (even if it’s small).
• You decide when the conversation is over.

The other person can influence you deeply—sometimes they should—but they do not become the author of your life.

When you hand over decision rights, you don’t become more supported. You become less real.

The Ask Template (One Minute to Clarity)

I try to ask for help in a single message that contains five pieces:

1. Context (two sentences, not your autobiography)
2. What I want (witness / advisor / teammate)
3. The decision I’m holding (or the deliverable I need)
4. Constraints (time, money, values, hard boundaries)
5. What would be helpful right now (a specific action)

Here are three scripts.

Script A: Witness

“Can I voice-dump for five minutes and have you reflect back what you hear? No fixing—just help me stay honest and name what I’m avoiding.”

Script B: Advisor

“I’m deciding between A and B. Can you help me list the tradeoffs and risks for each? I’ll make the call, but I want a clearer map.”

Script C: Teammate

“Could you take the first pass on X by Friday? I’ll review and integrate it. Definition of done is: (1) … (2) … (3) …”

The magic is not the words. It’s what the words prevent: the unconscious slide from support into surrender.

A Real Example: The “Please Fix My Calendar” Trap

I once sent a friend my entire weekly plan and asked, “Can you help me make this realistic?”

What I meant was: “Help me see the contradictions.”

What I communicated was: “Please become the governor of my life for an hour.”

So he did. He cut things. He re-ordered. He assigned time blocks. He gave me a beautifully structured plan that fit his nervous system.

I felt relief—and then shame.

Not because he did anything wrong. Because the plan didn’t belong to me. It wasn’t attached to my values, my hidden resistances, my actual appetite for difficulty that week. It was a foreign object. It could not metabolize inside my body.

Here’s what I ask for now instead:

“Can you look at this and tell me the top three conflicts you see? Where am I lying to myself about time and energy?”

Same friend. Same care. Different design.

The “Two Commitments” Rule (Boundaries Without Coldness)

When someone helps you, they often take on invisible commitments:

1. Task commitment (helping with the thing)
2. Emotional commitment (carrying your anxiety about the thing)

If you don’t manage this, you will either:

• drain good people, or
• train yourself to only ask for help from rescuers

So I make it explicit:

“I’m asking you for task help, not for you to carry my panic.”

Or:

“I’m asking you to witness me while I feel this. You don’t need to solve it.”

It’s not unfeeling. It’s respectful. It keeps the relationship clean.

Help From AI (Without Turning Yourself Into a Managed Object)

AI makes this easier and harder.

Easier because you can ask for:

• first drafts
• checklists
• reframes
• a second brain for options

Harder because it’s tempting to let the machine become the decider. To treat output as authority. To stop wrestling with your own values because the response sounds coherent.

My rule with AI is:

Use it to expand options, not to eliminate responsibility.

Good AI prompts keep you as the actor:

• “Ask me five clarifying questions before you suggest anything.”
• “Give me three plans with different tradeoffs, and tell me what each plan sacrifices.”
• “Reflect the values implied by what I wrote. What am I optimizing for without saying it?”
• “Turn this mess into a one-page brief I can share with a human helper.”

Bad AI prompts are pleas for replacement:

• “Tell me what to do.”
• “Make a plan and hold me accountable.”
• “Decide which path is correct.”

If you want to keep your agency, keep your author voice.

The Exit: Close the Loop Like an Adult

One more place we outsource agency is the ending.

We get help, we feel better, we vanish.

If you want to stay dignified—and keep help available in the future—close the loop:

• “Thank you. Here’s what I’m doing next.”
• “This helped because it clarified X.”
• “I’m not taking your suggestion, but it still helped me see Y.”

This is not politeness theatre. It’s part of the system. It teaches your relationships what kind of help you can receive without collapsing.

What This Looks Like In Practice

If you want a tiny checklist:

1. Name the role: witness / advisor / teammate.
2. Keep decision rights with you.
3. Ask for a specific action with clear constraints.
4. Don’t transfer your anxiety as an unspoken assignment.
5. Close the loop with what you’re doing next.

Agency doesn’t mean doing everything alone.

Agency means: even when I am supported, I remain present to my own life. I do not abdicate. I do not hide. I do not turn someone else’s certainty into my religion.

I can be helped—and still be the one steering.