About Psychology's Blind Spot

There is a problem with how psychology measures people. Not a problem that better tools will fix. A structural one.

I have been thinking about this for a while, and it got sharper recently when I went through a diagnostic process and the results did not quite fit. Not because the clinician was bad. She was not. But because the instruments have limits that nobody in the room talked about.

So I want to talk about them here.

Gödel First

In 1931, Kurt Gödel proved something uncomfortable about formal systems. Any system powerful enough to describe basic arithmetic will always contain true statements it cannot prove. You cannot build a system that is both complete and consistent. There will always be something true that sits outside what the system can see.

This is not a flaw in any particular system. It is a mathematical fact about all of them.

Then Penrose

Roger Penrose took this further in The Emperor’s New Mind. His argument was roughly this: humans can look at a Gödel statement and recognize it as true, even though no formal system can prove it. We just see it. That suggests something about human cognition that goes beyond pure computation. We are not just running an algorithm. Something else is happening.

Penrose is controversial. Many researchers disagree with him. But the question he raises is real: can a formal system ever fully capture what a human mind does?

Now Apply This to Psychology

Psychology tries to measure human experience with formal instruments. Questionnaires. Structured interviews. Standardized tests. IQ batteries.

These are formal systems. They have categories, scoring rules, cutoff values. They produce numbers.

And here is the issue: if Penrose has a point, then there are aspects of human cognition and experience that are, by definition, outside what any formal instrument can capture. Not hard to reach. Not requiring better tests. Simply outside the frame.

This is not an attack on psychology. Formal instruments do real work. They are better than nothing. But they have a boundary, and that boundary is structural, not practical.

The Edge Case Problem

The issue gets worse with edge cases. Every diagnostic instrument is calibrated on a population. The cutoff values, the norms, the expected patterns all come from studies on specific groups of people.

If you are not in that group, the instrument does not fail you because it is poorly designed. It fails you because it was never designed for you.

Twice exceptional people are a good example. High cognitive ability masking neurodivergent traits. The giftedness compensates for the ADHD. The ADHD suppresses the measured IQ. Both stay invisible. The instrument sees something in the middle and calls it normal.

Gödel would recognize this. The system produces a result that looks clean from inside the system. But the person sitting across the table knows something is off. That gap between the formal output and the lived reality is exactly the kind of thing Gödel described.

What This Actually Means

It does not mean diagnostic reports are worthless. It means they are partial by design.

The right question to ask after getting a diagnosis is not just what does this say but what can this kind of instrument not see.

That second question almost never gets asked. Not because clinicians are lazy. Because the system is not built to ask it. Asking about your own limits is not something formal systems do well. Gödel proved that too.

The Honest Position

I think psychology is a useful empirical discipline with real tools that do real things. I also think it is trying to measure something that partially escapes measurement by its nature. Human experience, especially at the edges of the distribution, keeps sliding out of the frame.

The instruments are not the problem. The assumption that the instruments are sufficient is the problem.

And the fix is not better instruments. It is honesty about what instruments can and cannot do.

That is not a radical position. It is just taking Gödel seriously.

Open BSD and Zen

OpenBSD and Zen

I’ve been using OpenBSD for a long time.

I’ve been sitting Zen for about two years, in the tradition of Seungsahn, with Ryōkan as a guide.

Nobody asked me to connect these two things. But they kept connecting themselves.

This is me trying to say why.

Less is not a compromise

OpenBSD ships without a lot of things other systems include by default.

That’s a choice. Every piece of code that isn’t there is a piece of code that can’t have a vulnerability.

The OpenBSD developers call this “correct by default”. I’d call it knowing what you actually need.

Zen works the same way.

Zazen, sitting meditation, is almost nothing. You sit. You follow your breath. You don’t try to achieve anything. You don’t add anything. Whatever arises, you don’t chase it.

What’s left when you stop adding is not emptiness. It’s clarity.

Ryōkan spent most of his life in a small hut on a mountain, owning almost nothing, writing poems about moon and rain and loneliness. He wasn’t poor in spirit. He was clear.

What is there I can give you? Take this handful of snow. – Ryōkan

I think about that poem when I look at OpenBSD’s base system. Take what’s here. It’s enough.

Correctness is not perfectionism

Every commit to OpenBSD goes through code review. Not as a process checkbox, as a genuine practice. Someone reads what you wrote. They question it. The code changes or it doesn’t go in.

This is not about perfection. Bugs exist in OpenBSD. Always have.

It’s about care. About not shipping something you haven’t looked at.

In Zen, this shows up as attention to the thing in front of you.

Not the thing you’re planning. Not the thing you finished an hour ago. This breath. This step. This line of code.

The enso, the Zen circle, brushed in one stroke, is never geometrically perfect. That’s not the point. The point is that it’s fully committed. The brush doesn’t hesitate.

A half-committed security decision is worse than no decision. A half-present sitting is still a sitting.

What breaks is information

OpenBSD’s malloc aborts the process when it detects memory corruption.

Not silently. Not gracefully. It stops.

That sounds harsh. It’s actually kind. A crash you can see is a bug you can fix. A crash that doesn’t happen is a vulnerability waiting to be found by someone else.

Zen does this too, differently.

When you sit long enough, what comes up is not peace. First it’s restlessness. Then it’s the things you’ve been avoiding thinking about. Then, maybe, something quieter.

The discomfort is not the problem. The discomfort is the practice.

Seungsahn used to say: don’t know. Not as an excuse, as a posture. Hold the question. Don’t rush to an answer that makes you comfortable.

OpenBSD’s default answer to uncertainty is: stop and look.

That’s a Zen answer.

A note on this comparison

I’m not saying the OpenBSD developers are Zen practitioners.

I’m not saying Zen is a system design philosophy.

What I’m saying is that I find the same posture in both: take less, commit fully, don’t hide what breaks.

That’s a way of being in the world. I’ve found it in a terminal and on a cushion. That’s enough for me.