Where to Look

Sources exist before you remember them.

This sounds obvious. It isn’t.

When we try to verify something, we look. But where we look isn’t neutral. It’s shaped by instinct, by habit, by the tools at hand, by what we’ve found before.

We don’t search the territory. We search our mental model of the territory.

And mental models have edges.

At those edges, we stop looking. Not because we’ve exhausted the sources—because we’ve exhausted what we thought to consider.

The shape of looking

Investigation feels like a process of elimination. Check here. Check there. Cross things off. But the list of “things to check” is already filtered before we begin.

Filtered by experience. Filtered by pattern. Filtered by what feels like a source and what doesn’t.

The most common failure isn’t missing the evidence. It’s missing the category of evidence.

You can be thorough within your model and still miss the answer entirely. Rigor inside a boundary isn’t the same as seeing the boundary.

Sources don’t announce themselves

Here’s what makes this hard: sources are quiet.

A database is obviously a source. A log file is obviously a source. But a conversation? A relationship? Your own past words, explaining what you were doing at the time you were doing it?

These are sources too. Often better sources—richer, more contextual, more revealing of intent.

But they don’t feel like sources. They feel like things that happened. Background, not evidence.

If you did something with someone, you probably told them about it. Your own narration, at the time, is evidence. Often the best evidence—because it includes reasoning, not just events.

The relational source

The sources we most often miss are relational.

We think of relationships as context, not content. As the setting in which things happen, not a record of what happened. But relationships accumulate evidence constantly—every exchange, every explanation, every decision made together.

This evidence isn’t indexed. It doesn’t show up in search results. It lives in the space between people, and it surfaces only when someone thinks to look there.

The question isn’t “do I have evidence?” It’s “what kind of evidence might exist that I’m not conceiving of?”

Verification and imagination

The “unverified by default” frame asks: what would it take to verify this?

But there’s a prior question: what sources could verify this?

And an even more prior question: what sources exist that I haven’t considered?

Verification fails at the edges of imagination, not the edges of evidence.

Before you search, ask what could exist.

We don’t fail to verify because proof doesn’t exist. We fail because we don’t conceive of where it might live.

A practice

I don’t think the answer is a checklist. “Always check X before Y.” That trades one rigidity for another.

The practice is a pause before the search. A moment to ask:

What kinds of sources could contain this?
Which ones am I instinctively reaching for?
Which ones am I not considering, and why?

The instincts aren’t wrong. They’re incomplete. The pause makes room for what they miss.

Sources exist before you remember them.

The work is remembering.