About
An archive of decisions, not reputations.
HighAgency.uz is an editorial archive of people, cases and ideas about agency in Uzbekistan. We follow people through documented decisions, the constraints they were working against, and what they actually did.
It is a working archive rather than a finished reference. It currently holds 9 published cases and 21 people, and it says where it is thin.
02What we mean by high agency
Eight behaviours we look for in the record.
These are the editorial framework this archive files by. They are not a psychological taxonomy, they were not validated by anyone, and they are not the only reasonable way to cut the subject. They are the eight things we found ourselves pointing at repeatedly when reading what people had done.
Two of these definitions were narrowed after we checked them against our own cases and found they claimed more than the record showed.
Bias to Action
Moves before the conditions are settled.
Resourcefulness
Works with what is already there instead of waiting for the right conditions.
Self-Authorship
Writes their own brief rather than accepting the one they were handed.
Persistence
Stays with the same work across years rather than weeks.
Clear Thinking
Reads the situation as it is, including when that is inconvenient.
Disagreeability
Holds a position the room does not share, and carries the cost of it.
Community Building
Builds the thing that lets other people act.
Craft & Skill
Earns the outcome through depth of practice rather than positioning.
03How a case works
The case is the unit, not the person.
- 01Problem
- 02Constraint
- 03Decision
- 04Move
- 05Result
A case is one decision, taken under one constraint, by one or more people. It is the smallest thing this archive is willing to publish, because a decision without the constraint it answered does not show much.
- A person can have several cases.
- Co-founders share one case rather than having one each.
- A person can sit in the archive with no publishable case at all — 10 currently do.
04Evidence
Every link says how well we know it.
Each of the five links in a case carries its own state, shown on the page. Inference never quietly becomes fact: a link marked as our reasoning stays marked that way permanently, and no amount of time or confidence promotes it.
- Documented
Carried by a source outside the people involved.
- Inferred
Our reading of the record. No source states it directly.
- Not established
Nothing in the research addresses this yet.
Where a source stands
A claim only counts as documented when it is carried by someone outside the people involved. Two separate things decide that: who published an account, and whether that publisher is connected to its subject. An organisation writing about its own director is not independent evidence, however accurate it happens to be.
- Independent reporting
- A newsroom, journal, patent office or statistics body with no stake in the outcome. The only kind that can carry a claim on its own.
- Published by an involved organisation
- An organisation reporting on work it is part of. Usually accurate, never disinterested.
- Published by the subject
- The subject's own channel — a company page, a founder's post, a fundraising appeal. Cannot establish a claim here.
- Reference database
- Reference databases and encyclopaedias. Useful for corroborating a fact, never for establishing one.
- Awarding body
- A body that selected someone. Evidence of recognition, which is a different thing from evidence of action.
- Personal or social post
- A named third party posting without an editorial process — a personal blog, a channel, a professional network. It can carry first-hand testimony or back up a claim, but never establish one alone.
- Connected to the subject
- Marked separately from the above, because it cuts across all of them. A conference speaker profile is written by the speaker; an organisation’s newsroom writes about itself. Neither can carry a claim alone.
05What we don’t claim
The rules are easier to state as refusals.
We don’t rank people.
There is no order of merit anywhere on this site. People are listed alphabetically, and the archive holds no field that could sort them by worth.
We don’t calculate an agency score.
No number is attached to a person, and none could be. A trait exists only as an assertion against one moment in one case — it does not roll up into anything.
We don’t treat fame, titles or awards as evidence.
Recognition tells you what other people concluded. A world chess title and a completed company exit both sit in this archive without either one carrying a single published claim.
We don’t treat an impressive outcome as proof.
Results are what agency sometimes produces, not what it consists of. Without a documented constraint and a documented decision, an outcome is a fact about the world rather than about the person.
We don’t turn one case into a pattern.
A reading drawn from a single case is published as an observation and labelled one. Two cases sharing a shape is weak evidence; one case is a hypothesis with a name.
We don’t hide unresolved conflicts.
Where good sources disagree, both readings publish next to the claim they affect and the disputed figure stays out. We are not positioned to arbitrate between them.
We don’t invent missing constraints.
When the record does not reach the thing someone was working against, the page says so. An empty link is more useful than a plausible one.
We don’t present our interpretation as fact.
What a source says and what we make of it are stored separately and rendered differently. Our reading is always marked as ours.
06Where the archive is thin
The gaps are published, not patched.
Reporting is consistently good on what people decided and did, and consistently thin on what they were working against. That shapes what we can publish, and it means several cases carry a documented decision with an empty constraint above it. Those cases go out incomplete rather than filled in.
3
Cases without a documented constraint
4
Links that are our reasoning, not a source
1
Contradictions left unresolved in public
These figures are counted from the archive when this page is built, so they change as the work does. Where a case has a gap, the case page names it.
Field notes on the archive’s limits →07Sources
Every published case lists its sources, with who published each account and whether they are connected to the subject. The archive currently draws on 28 sources, of which 18 are independent of the people involved. We prefer those and hold cases back when they are missing.
We are not claiming objectivity, and this is not academic work. Selecting what to look at is an editorial act, and so is deciding what a decision means. What we can offer is that the factual half is checkable and the interpretive half is labelled.