Quick Summary
- Three years into a PhD, I realised I was climbing the wrong mountain; halfway up is exactly where quitting gets expensive.
- I dropped out. The backlash lasted months, and two years later someone told me I was "braver than that"; an accusation of cowardice dressed up as a compliment.
- Her whole argument was the sunk cost fallacy in a smart suit: it's just 250 pages, it's only 2-3 more years, think of the money.
"You are braver than that."
That's what someone I knew said to me, two years after I walked away from my PhD. And if you're wondering whether that was a compliment, so was I, for about half a second.
It wasn't. It was an accusation of cowardice, wrapped carefully enough that I couldn't object to it.
Over the years, I forgot about it but as I was reading something recently, nine years later, I suddenly remembered that sentence again.
The wrong mountain
I'd been at the PhD for three years. Trying, more accurately.
My guide never guided. The topic was useless. The work was going to change nothing for anybody, and I knew exactly what it was for: a line on a file that would get me promoted.
Somewhere in year three, it became clear I was climbing the wrong mountain.
Here's the problem with realising that halfway up: the easy choice is to keep climbing. You're already there. The summit is closer than the base.
Going back down means undoing three years in full view of everyone who watched you start, and then beginning again from the bottom of something else.
So I dithered on the edge for months.
But I couldn't reconcile myself with where the path was heading. I could see the future at the end of it clearly enough to know I'd hate it. Not dislike it. Hate it.
So I chose to climb down.
The backlash lasted months. Slander. The endless well-meaning attempts to make me see sense. Reputation damage, quiet and not so quiet.
And these people weren't at a distance; they were in my face, daily, and there was no polite way to ask them to step back.
I stuck to my decision anyway. It remains one of the hardest things I've done, though I've since made much harder choices with less noise around them.

"It's just 250 pages"
Two years later, that conversation.
She was making what she thought was the reasonable case. It went something like this:
"In the long run you can do whatever you like. It's just 250 pages and a degree. It's a comfortable arrangement."
"My mind couldn't accept it. So I left. I moved my focus to something I thought was more worthwhile."
"You are braver than that. It was just 250 pages. You could have ground it out over five or six years."
"It's been two years since I left. I'm not going back to anything that makes me a slave to the system."
"It's more money. Why would you not? You'll lose ten lakhs. Probably more."
"We all want different things from life."
"Your revolt is sweet today. It may not be tomorrow. A kitchen pan, a vacuum cleaner, quality, lifestyle. It all costs money."
And then, near the end: "Okay, let it be. It beats me."
"Precisely," I said. "That's exactly why this conversation is futile."
Now read that exchange again. Notice what the argument actually is.
Sunk cost dressed in a smart suit
Every line of it is the sunk cost fallacy.
You've already done three years. It's only 250 pages. It's only three or four more years. Look at the money you're walking away from.
Here's the thing about a sunk cost: it is, by definition, unrecoverable. The three years were spent. Nothing I did next was going to bring them back.
Finishing wouldn't refund them; quitting wouldn't waste them any further. They were simply gone, and they belonged in no column of the decision at all.
But that's not how the mind treats them. The mind treats past investment as a debt the future owes you, and it will keep you climbing a mountain purely because of the altitude you've already gained.
| What you're weighing | Sunk-cost thinking (the trap) | Future-value thinking (the reality) |
|---|---|---|
| Time | I've already put three years in. I can't let them go to waste. | The three years are gone either way. The real question is whether I want to spend three more. |
| Identity | If I quit, I'm someone who couldn't finish. | I tested something seriously, learned it was wrong, and acted on what I learned. |
| Effort | All that work drops to zero the moment I stop. | The output may be shelved, but the capability I built stays with me permanently. |
| Cost | Leaving means losing the money and the credential. | Staying has a price too, and it's every year of the better thing I'm not doing instead. |
That last row is the one people miss. Sunk cost gets all the attention; opportunity cost quietly does the real damage. Every month on the wrong mountain isn't neutral. It's a month of the right one, unclimbed.
The ‘Clean Slate Question’
If you're standing where I was standing, there's one question that cuts through all of it, and I'd offer it to any founder sitting on a product line, a market, a client, or a role they can't quite justify.
Imagine you woke up today with everything you currently have. Your skills, your savings, your network, your reputation.
All of it intact. The only difference is that you had never started this particular thing.
Now someone walks up and offers it to you, exactly as it stands today, on today's terms.
Would you say yes?
If the answer is a clear no, then you're not staying because it's worth doing. You're staying because you already started.
That's the whole answer, and it's usually available to you long before you're willing to act on it.
Sunk cost asks what you've already paid. The only useful question asks what you're about to.
The part nobody can see
There's a second lesson in that conversation, and it took me longer to understand.
There were things about my situation at the time that I wasn't free to explain to anyone. Her picture of my life was missing information that would have ended the argument in a sentence, and I couldn't give her that sentence.
So I said nothing and let her conclude what she liked.
Recently, I was reading Ryan Holiday's Courage Is Calling and hit a chapter arguing that you should never question another person's courage.
He quotes James Baldwin, who sat at his father's funeral and heard the preacher say a line Baldwin never forgot: "Thou knowest this man's fall; but thou knowest not his wrassling."
You see the fall. You never see the wrestling.
That's the whole problem with judging a decision from outside it. You get the visible half: the leaving, the pivot, the shutdown, the thing that didn't work.
You almost never get the constraints, the private arithmetic, the risks someone was quietly managing, the people they were protecting.
Which is why I stopped labelling people; not out of politeness, but out of accuracy. I don't have the data.
Neither does anyone standing outside my choices. That cuts in every direction, and the only scrutiny that's reliable is the kind you point at yourself.
The practical version of this, for anyone building something: you don't owe the world a full explanation of your reasoning, and you'll exhaust yourself trying to supply one.
Some people have already decided what you are. Nothing you say will move that, and the energy you spend appealing the verdict is energy that isn't going into the work.
What actually compounded
Nine years on, here's the part that surprises me.
The decision itself matters far less than I thought it would. Walking away took one afternoon of courage.
What mattered was every day afterwards where I didn't quietly walk back; didn't take the safe promotion, didn't collect the title, didn't slip back into the arrangement when it would have been easy and nobody would have said a word.

That's the thing nobody tells you about a hard choice. Making it is the cheap part. Holding it for a decade is the actual price, and holding it is also the entire return.
Because conviction behaves exactly like capital. It compounds, and compounding has one non-negotiable condition: you can't keep interrupting it.
Charlie Munger put it plainly; the first rule of compounding is to never interrupt it unnecessarily.
Every reversal resets the clock. Every re-entry into the thing you left means starting the accumulation again from zero, only older.
So the two ideas in this post are mirror images of each other, and both are really about time.
Sunk cost is the past pretending it has a claim on your future. Compounding is the future quietly rewarding you for leaving it alone.
One traps you on the wrong mountain. The other only pays out if you stay on the right one long enough for the years to stack.
Which means the real work was never the leaving. It was the nine years of not going back.
She was right about one thing, by the way. My revolt did look sweet at 28.
It looks a lot better at 36.



