The Anxiety Was the Better Self

*Every SSRI-for-social-anxiety source shares one unexamined assumption — that the person beneath the anxiety is kinder than the anxious version. What if they're not?*

There's a Reddit thread where people on Lexapro compare notes on becoming less — less anxious, less inhibited, less careful. It sits inside a much larger literature, and that literature tells one consistent story about what's happening: the anxiety was a shell. Remove the shell and the real person emerges — warmer, more connected, more authentically themselves. Refresh Psychiatry, Talkiatry, Telapsychiatry, LifeStance — different wording, identical architecture. Anxiety reduction is framed as unalloyed good. The post-medication self is framed as the real one. Anxious self: costume. Medicated self: face.

Nobody in any of these sources considers the possibility that the anxiety *was* the better self. Not the whole self — but the part that noticed social cost, modulated expression, paid attention to impact. The system that ran a check before speaking. The friction that made someone pause before delivering a fixed judgment, not because they were cowardly but because they could feel the weight of what they were about to do. That's not a shell. That's a conscience with a working signal.

If the anxiety was the system keeping harsh patterns in check, then "authentic self-expression" isn't the opposite of cruelty — it's the vehicle for it.

And emotional blunting — the most commonly reported side effect across the SSRI literature, including that thread — compounds the problem in a specific, predictable way. It's not just that someone becomes more comfortable expressing fixed negative judgments. It's that they stop feeling the friction of doing so. The internal signal that used to say *that landed wrong* goes quiet. No course-correction because there's no signal to course-correct from. The person doesn't get harsher and feel bad about it. They get harsher and feel fine. The framework calls that confidence.

This isn't anti-psychiatry. SSRIs reduce anxiety; that's well-established and for many people that's unambiguously good. The question is whether our evaluation framework has a blind spot that makes a specific kind of harm invisible and relabels it as wellness. The tell is structural: if the framework can't even *name* disinhibition-that-looks-like-confidence as a possible outcome, it can't detect it when it happens. Every source celebrates increased comfort in social situations. None of them asks: comfortable doing what?

The real test isn't pharmacological. It's whether anyone around the person who just got "more confident" is allowed to name what they're seeing — or whether the increased comfort in fixed judgment makes that naming impossible too. A framework that treats anxiety reduction as self-evidently good has no language for the person saying *you changed and not for the better.* It will hear that as resistance to treatment. It will hear it as the anxiety talking. It has already decided what the real self looks like, and the real self is always calmer.