It almost always happens in the same order. Someone realizes he was justifying himself too much, reads up on it, and corrects in the opposite direction: now he answers curtly, explains nothing and returns every jab with a harder one.
The result is worse than the starting point, because now there's something that wasn't there before: intent.
Why harshness reads as insecurity
Hitting back requires having felt the blow. Whoever returns a jab with a harder one is reporting that the first one landed and that he needed to make up for it, and that's exactly what justifying communicates, just in a different tone.
Payback and indifference are easy to tell apart: someone who truly wasn't fazed doesn't need to score a point.
Where the line is
It's clearer in pairs. Same situation, two answers, and the difference isn't firmness but whether there's a victim.
Alexis, 28Contéstalo túThe four signs of overcorrection
- Your answers leave someone looking bad, not just close the topic.
- You answer curtly even when there was no jab to return.
- You refuse to share normal information on principle, not because it's irrelevant.
- You're focused on not seeming needy, which is another way of being focused on how you're seen.
The fourth is the hardest to see from the inside, and it's the one that sums up all the others.
How to correct it
By lowering the volume, not changing the content. Almost every harsh answer works if you remove the damage and keep the edge: you say the same thing without anyone being humiliated.
And there's a home test that works. If after saying the line you're watching her face to see if it worked, that answer was for you, not for the conversation.
Why the simulator punishes both
The scoring scale drops the same for neediness and for cruelty. An insult or a heavy put-down falls into the low band for the same reason as begging: both prove the other person's line had an effect.
Where overcorrection comes from
Almost always from a half-reading. Someone reads that justifying is bad and concludes that the opposite of justifying is attacking. The opposite of justifying is not justifying, which looks nothing like attacking.
The other source is more uncomfortable: carrying the accumulated resentment of all the times he did justify himself. There, harshness isn't a misapplied technique, it's an unpaid bill that comes out against someone who didn't run it up, and it shows in the mismatch between the jab and the response.
What calm confidence sounds like
It has three traits you can check in any line before saying it.
- Nobody looks bad. There can be an edge, irony and even some arrogance, but there's no victim at the end of the sentence.
- It doesn't need to work. An answer that needs the other person to laugh is an answer asking for approval in a different costume.
- It can hold up in a low voice. If the line only works said with tension, the content wasn't holding anything up.
The third is the most useful of all. Try saying your answer in a low, relaxed voice: if it sounds like a calm joke, it works. If it sounds like a threat, it was payback.
When you went too far by accident
If you overdid it, own it in four words and move on: "too far, sorry. Go on." That costs no frame. What does cost it is insisting you didn't go too far when you both know you did.
Frequently asked questions
So I have to be nice all the time?
You don't have to be anything in particular. Forced niceness to be liked scores as badly as forced harshness to avoid seeming soft, because both are calculated while watching the other person.
What if she genuinely crossed a line?
Then you set a boundary, which is something else: say it once, calmly and without punishment, and if nothing changes, you leave. A boundary isn't trying to hurt or win the exchange, so it looks nothing like payback.
Now train it until it comes naturally
The AI will throw it at you without warning, inside a normal conversation, and score what you answer. You start with status and money tests.
Train for free2 free sessions a day · no card